Lall R, Socha P*, Harper S, Ofili D, Hetherington E*. “Inequalities in Preterm Birth among Immigrants to Canada by Race and Time since Immigration: A Population-Based Repeated Cross-Sectional Study.” International Journal of Gynecology and Obstetrics. 2026;173(1):537–539
Socha PM*, Hutcheon JA, Strumpf EC, Liauw J, Srour M, Ting JY, Skoll MA, Harper S. “Antenatal Corticosteroids and Risk of Cerebral Palsy: A Regression Discontinuity Study.” The Journal of Pediatrics. 2026;290:114960
Objective To use a natural experiment to investigate the effect of antenatal corticosteroids on the risk of cerebral palsy. Study design We included singleton livebirths with a maternal admission for delivery from 31 + 0 through 36 + 6 weeks of gestation, in British Columbia, Canada, between 2000 and 2015. Guidelines recommended antenatal corticosteroids through 33 + 6 weeks, and we estimated the effect of the corresponding sharp drop in the proportion treated at 34 + 0 weeks on the risk of a composite of death before age 2 or cerebral palsy. We defined cerebral palsy using diagnostic codes in hospital and physician-billing records before age 5 years and corrected for misclassification using external estimates of the sensitivity and specificity. We used logistic regression to estimate marginal effects at 34 + 0 weeks. Results There were 20 009 children in our study sample. The crude and misclassification-corrected risks of cerebral palsy were 6.2 and 5.6 per 1000, respectively. The risk of death before age 2 or cerebral palsy declined with increasing gestational age at maternal admission for delivery, but we found no convincing evidence of an abrupt change just before vs just after 34 + 0 weeks (risk ratio: 0.98, 95\% confidence interval: 0.50 to 1.98). Results were similar using a composite outcome of in-hospital newborn death or cerebral palsy, and using cerebral palsy alone. Conclusions We did not find evidence that the lower likelihood of being treated with antenatal corticosteroid at 34 + 0 weeks affected the risk of cerebral palsy, but the estimates were imprecise and compatible with benefits or harms. © 2025 The Author(s).
Socha P*, Oskoui M, Hutcheon J, Harper S. “A Multivariable Model for Improving the Identification of Cerebral Palsy Cases in Administrative Health Data.” Annals of Epidemiology. 2026;114:26–31
Purpose To improve the identification of cerebral palsy cases in administrative health data. Methods We included all children in a population-based cerebral palsy registry in Quebec, Canada, born from 1999 through 2002, and a sample of children without cerebral palsy. Population-based hospitalization and physician billing records through 2012 were obtained for all children. We used logistic regression to model the probability of cerebral palsy, using International Classification of Diseases codes for related diseases. We reported receiver operating characteristic (ROC) and precision-recall (PR) curves, and compared the accuracy to that of existing algorithms. We also reported the accuracy of cerebral palsy codes by age, data source, and gestational age at birth. Results The area under the ROC and PR curves of our model were 0.98 (95 \% CI: 0.97–0.99) and 0.73 (95 \% CI: 0.63–0.79), respectively. Cut-offs with a similar specificity to existing algorithms yielded sensitivities that were 1–14 \%age-points higher. The sensitivity of cerebral palsy codes was higher (and the specificity was lower) with longer follow-up times since birth, when using both hospitalization and billing records, and among children born preterm. Conclusions Our model improved identification of cerebral palsy cases in administrative data, but residual misclassification remained. © 2026 The Authors.
Alcantara M, Harper SB, Shapiro GD, Bushnik T, Kaufman JS, Vang Z, Mashford-Pringle A, Yang S. “Perinatal Health Inequalities between Canadian-born and Foreign-Born Women in Canada: A Decomposition Analysis.” BMJ Public Health. 2025;3(2):e001231
Background Foreign-born mothers are generally believed to experience better perinatal outcomes than Canadian-born mothers, despite relatively lower socioeconomic status. However, the magnitude and direction of inequalities by nativity status vary across outcomes. Little is known about factors contributing to the health inequalities by nativity status across different perinatal outcomes. Thus, we aim to examine the direction and magnitude of inequalities by nativity status across perinatal outcomes and estimate the contributions of select individual-level characteristics to the inequalities in Canada. Methods Using 132 639 singleton births from the 2016 Canadian Birth-Census Cohort, we estimated the risk of preterm birth (PTB), small-for-gestational-age (SGA) and large-for-gestational-age (LGA) birth, stillbirth, and infant and neonatal death by maternal nativity status. We estimated the contribution of maternal race, maternal and paternal education, paternal nativity status and employment, family income and homeownership, as well as maternal age, marital status, activity limitations and parity to inequalities specific to each outcome, using Kitagawa’s decomposition method. Results Compared with Canadian-born mothers, foreign-born mothers experienced higher rates of all outcomes examined (eg, 627 (95\% CI 608, 646) PTBs per 10\,000 live births among foreign-born mothers vs 580 (568, 592) among Canadian-born mothers), except for LGA births (677 (648, 706) per 10\,000 for foreign-born vs 1006 (959, 1054) for Canadian-born mothers). Non-White maternal race explained the largest proportion of the observed differences for non-fatal outcomes, while the highest income quartile explained the most for the differences in fatal outcomes. Conclusion Foreign-born women fared worse than Canadian-born women for all adverse perinatal outcomes examined apart from LGA births. Our results highlight differential contributions of determinants to perinatal health inequalities by maternal nativity status across outcomes.
Bannon OS, Been JV, Harper S, Laverty AA, Millett C, family = Lenthe g=FJ, Filippidis FT, Radó MK. “Cigarette Taxation and Socioeconomic Inequalities in Under-5 Mortality across 94 Low-Income and Middle-Income Countries: A Longitudinal Ecological Study.” The Lancet Public Health. 2025;10(5):e380-e390
Background: Although increasing cigarette taxes is known to improve child survival, there are few data on their effect on socioeconomic inequalities in child mortality. We investigated the association between cigarette taxation and socioeconomic inequalities in mortality in children younger than 5 years (hereafter referred to as under-5 mortality) in low-income and middle-income countries (LMICs). Methods: This was a longitudinal ecological study. We linked country-level annual data on 94 LMICs, as defined by the World Bank, and annual data on under-5 mortality by wealth quintile from the UN Inter-agency Group for Child Mortality Estimation from 2008 to 2020. We used fixed-effect panel regression models to assess the association of cigarette taxes with absolute and relative inequalities in under-5 mortality by wealth quintile. Findings: Increasing total cigarette tax by 10-percentage-points was associated with reduced under-5 mortality rates in all wealth quintiles. Raising total cigarette tax from 0·0–24·9\% to 25·0–74·9\% and 75·0\% or more of their total retail value was associated with 3·8\% (95\% CI 0·2 to 7·3) and 7·6\% (1·4 to 13·4) decreases in absolute inequality in under-5 mortality, respectively. This finding was mainly attributable to specific tax, which was associated with a 1·4\% (0·3 to 2·6) reduction in absolute inequality for each 10-percentage-point increase. We estimated that raising total cigarette taxes to 75·0\% or more in all 94 LMICs could have averted 281 017 (196 916 to 362 301) under-5 deaths in 2021. Interpretation: High cigarette taxes are associated with a large decrease in absolute inequality in child mortality in LMICs. These findings support raising cigarette taxes to the WHO-recommended 75\% or more of the retail value to protect the poorest children. Funding: Swedish Research Council for Health, Working Life, and Welfare; Stiftelsen Riksbankens Jubileumsfond; European Union's Horizon 2020 Research and Innovation; and UK National Institute for Health and Care Research. © 2025 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license
Chen D, Momen NC, Ejlskov L, Bødkergaard K, Werenberg Dreier J, Sørensen HT, Laustsen LM, Harper S, Hakulinen C, McGrath JJ, Plana-Ripoll O. “Socioeconomic Inequalities in Mortality Associated with Mental Disorders: A Population-Based Cohort Study.” World Psychiatry. 2025;24(1):92–102
Mental disorders are associated with elevated mortality rates and reduced life expectancy. However, it is unclear whether these associations differ by socioeconomic position (SEP). The aim of this study was to explore comprehensively the role of individual-level SEP in the associations between specific types of mental disorders and mortality (due to all causes, and to natural or external causes), presenting both relative and absolute measures. This was a cohort study including all residents in Denmark on January 1, 2000, following them up until December 31, 2020. Information on mental disorders, SEP (income percentile, categorized into low, {$<$}20\%; medium, 20-79\%; and high, ≥80\%), and mortality was obtained from nationwide registers. We computed the average reduction in life expectancy for those with mental disorders, relative and absolute differences in mortality rates, and proportional attributable fractions. Subgroup analyses by sex and age groups were performed. Overall, 5,316,626 individuals (2,689,749 females and 2,626,877 males) were followed up for 95.2 million person-years. People with mental disorders had a shorter average life expectancy than the general population regardless of SEP (70.9-77.0 vs. 77.2-85.1 years, depending on income percentile). Among individuals with a mental disorder, the subgroup in the top 3\% of the income distribution had the longest average life expectancy (77.0 years), and this estimate was lower than the shortest life expectancy in the general Danish population (77.2 years for individuals in the bottom 6\% income distribution). The mortality rate differences were larger in the low-income than the high-income group (19.6 vs. 13.3 per 1,000 person-years). For natural causes of death, a socioeconomic gradient for differences in life expectancy and mortality rates was observed across most diagnoses, both sexes, and all age groups. For external causes, no such gradient was observed. In the low-SEP group, 10.1\% of all deaths and 23.7\% of those related to external causes were attributable to mental disorders, compared with 3.5\% and 8.7\% in the high-SEP group. Thus, our data indicate that people with mental disorders have a shorter life expectancy even than people with the lowest SEP in the general population. The socioeconomic gradients in mortality rates due to natural causes highlight a greater need for coordinated care of physical diseases in people with mental disorders and low SEP.
Hutcheon JA, Harper S, Cordingley MC, Liauw J, Skoll MA, Socha PM*, Srour M, Ting JY, Strumpf EC. “Antenatal Corticosteroid Administration and Childhood Respiratory Morbidity: A Regression Discontinuity Study.” BJOG: An International Journal of Obstetrics & Gynaecology. 2025;133(2):1471-0528.18252
ABSTRACT Objective To determine if routine administration of antenatal corticosteroids affects the risk of infant lower respiratory tract infection and/or childhood asthma. Design Linked population‐based cohort analysed using a regression discontinuity design, which better controls for confounding than standard observational studies. Setting British Columbia, Canada. Population Singleton pregnancies with a maternal admission for delivery between 31\,+\,0 and 36\,+\,6\,weeks' gestation from 2000 to 2016, with follow‐up to 2020. Methods We estimated if risks of childhood respiratory outcomes differed between pregnancies admitted just before the Canadian recommended clinical cut‐off for antenatal corticosteroid administration of 34\,+\,0\,weeks gestation (i.e., with higher probability of exposure to antenatal corticosteroids; ‘exposed’) than those admitted just after this cut‐off (i.e., with lower probability of exposure; ‘unexposed’) using log binomial regression (infant lower respiratory infection hospitalisation) and pooled log binomial regression (asthma). Main Outcome Measures Infant lower respiratory tract infection hospitalisation, inpatient or outpatient asthma diagnosis at 1–18\,years. Results In our cohort of 21\,965 children, 412 (1.9\%) infants were hospitalised with a lower respiratory tract infection and 2287 (10.4\%) were diagnosed with asthma. Routine administration of antenatal corticosteroids was not associated with infant lower respiratory tract infection (risk ratio\, = \,0.95 [95\% CI: 0.61, 1.37], risk difference\, = \,−0.15 excess cases per 100 [95\% CI: −1.30, 0.99]) or childhood asthma (rate ratio\, = \,1.08 [95\% CI: 0.88, 1.24]~5.49 excess cases per 100\,by age 13 years [95\% CI: −1.78, 14.39]). Conclusions We found no evidence that routine administration of antenatal corticosteroids affects the risk of later childhood respiratory illnesses.
Li X, Brehmer C, Hirst K, Sternbach T, Yuan W*, Zhang X, Barrington-Leigh C, Baumgartner J, Harper S, Robinson B, Shen G, Tao S, Zhang Y, Carter E. “Multi-Year Evaluation of a Clean Heating Policy on Residents’ Air Pollution Exposures in Beijing, China.” Environmental Science & Technology. 2025;59(45):24347–24358
China’s Clean Heating Policy (CHP), aimed at shifting households from coal to electricity for space heating, represents a major residential energy transition initiative. We evaluated its multi-year impacts on outdoor, indoor, and personal exposures to PM2.5and black carbon (BC) across 50 villages and 1,236 households in rural Beijing. Using a difference-in-differences (DiD) design, we observed a substantial (31 μg/m3) reduction in winter (3 month) indoor PM2.5(95\% CI: −53, −9), but an increase in 24-h indoor BC by 2.6 (0.4, 4.7) μg/m3. CHP-driven reductions in personal exposures were limited, emphasizing the limitations of using single 24-h measurements to estimate “usual” exposure. Outdoor air quality improved in all villages, with no difference between treated versus untreated villages. Exposure-energy trade-off analysis showed that untreated households achieved similar personal PM2.5reductions at lower cost, with smaller coal use reductions and less electricity expenditures. The CHP significantly reduced seasonal indoor PM2.5, but continued burning of biomass, which was accessible at no cost, limited air quality improvements and may have contributed to the observed increase in 24-h indoor BC. This illustrates how behavioral choices, economic feasibility, and selection of exposure metrics influence the measured impact of household energy transitions. © 2025 American Chemical Society
Ma J, Zhu X, Li X, Zhang X, Zhao F, Zhao A, Xue K, Song Q, Wang L, Zhang X, Carter E, Harper S, Robinson B, Baumgartner J, Zhang Y. “Investigating Indoor VOC Complexities in Rural Beijing: Environmental, Resident Behavioral, and Household Energy-Use Paradigms.” Building and Environment. 2025;279
Chronic exposure to volatile organic compounds (VOCs) poses significant health risks, especially in rural households with prolonged winter occupancy and diverse emission sources. This study investigates the key influencing factors of indoor VOCs complexities in rural Beijing through an integrated analysis of environmental conditions, behavioral patterns, and energy consumption frameworks. Analyzing 320 air samples from 192 randomly selected households, we reveal that indoor VOC concentrations were significantly higher than outdoor levels, with C3-C4 alkanes/alkenes constituting the dominant fraction (62.1 \%). A critical finding identifies liquefied petroleum gas (LPG) cooking as a significant contributor to indoor VOC burdens. Contrary to conventional assumptions, heating fuel type (coal, biomass, or clean energy) exhibited negligible influence on VOC levels, attributable to spatial configurations of combustion infrastructure: 90 \% of coal-heated households utilized radiator systems that isolated stoves from living spaces, while biomass units were predominantly outdoor-operated, mitigating direct indoor emissions. Among all variables, CO2 concentration demonstrated the strongest statistical association with VOCs, showing significant correlations across all seven VOC categories. Elevated aromatic hydrocarbons and oxygenated VOCs (OVOCs) in smoking households underscore tobacco smoke as a persistent indoor pollutant source. These findings challenge energy transition narratives by emphasizing that spatial design of heating systems, rather than fuel type alone, critically governs indoor air quality. © 2025
Richardson RA*, Harper S, Keyes KK, Crowe CL, Calvo E. “Contributors to Age Inequalities in Loneliness among Older Adults: A Decomposition Analysis of 29 Countries.” Aging & Mental Health. 2025;0(0):1–9
Loneliness is highly prevalent and can have severe health consequences. While generally assumed to increase with age, some evidence suggests the relationship between age and loneliness may vary across country. In this study, we investigate the contribution of demographic and health factors to age-related inequalities in loneliness both within and across countries. We used population-based cross-sectional data from 64,324 older adults (age range: 50–90 years) across 29 countries. Loneliness was measured with the 3 item UCLA loneliness scale. We quantified the magnitude of age inequalities in loneliness using concentration indices, and we estimated the contribution of demographic and health factors to age inequalities in loneliness using a decomposition approach. Loneliness was generally more concentrated among the oldest adults in the sample, although in the US and the Netherlands it was more concentrated among younger adults. Top contributors to age inequalities in loneliness were being unmarried and not working; however, the amount that factors contributed to inequalities differed markedly by country. Age inequalities in loneliness, and contributors to these inequalities, vary substantially across countries, suggesting that loneliness is not an inevitable consequence of age but may instead be shaped by environments within countries (e.g. social cohesion).
Yuan W*, Li X, Brehmer C, Sternbach T, Zhang X, Carter E, Zhang Y, Shen G, Tao S, Baumgartner J, Harper S. “Effects of Outdoor and Household Air Pollution on Hand Grip Strength in a Longitudinal Study of Rural Beijing Adults.” International Journal of Environmental Research and Public Health. 2025;22(8):1283
Background: Outdoor and household PM2.5 are established risk factors for chronic disease and early mortality. In China, high levels of outdoor PM2.5 and solid fuel use for cooking and heating, especially in winter, pose large health risks to the country’s aging population. Hand grip strength is a validated biomarker of functional aging and strong predictor of disability and mortality in older adults. We investigated the effects of wintertime household and outdoor PM2.5 on maximum grip strength in a rural cohort in Beijing. Methods: We analyzed data from 877 adults (mean age: 62 y) residing in 50 rural villages over three winter seasons (2018–2019, 2019–2020, and 2021–2022). Outdoor PM2.5 was continuously measured in all villages, and household (indoor) PM2.5 was monitored for at least two months in a randomly selected \textasciitilde 30\% subsample of homes. Missing data were handled using multiple imputation. We applied multivariable mixed effects regression models to estimate within- and between-individual effects of PM2.5 on grip strength, adjusting for demographic, behavioral, and health-related covariates. Results: Wintertime household and outdoor PM2.5 concentrations ranged from 3 to 431 μg/m3 (mean = 80 μg/m3) and 8 to 100 μg/m3 (mean = 49 μg/m3), respectively. The effect of a 10 μg/m3 within-individual increase in household and outdoor PM2.5 on maximum grip strength was 0.06 kg (95\%CI: −0.01, 0.12 kg) and 1.51 kg (95\%CI: 1.35, 1.68 kg), respectively. The household PM2.5 effect attenuated after adjusting for outdoor PM2.5, while outdoor PM2.5 effects remained robust across sensitivity analyses. We found little evidence of between-individual effects. Conclusions: We did not find strong evidence of an adverse effect of household PM2.5 on grip strength. The unexpected positive effects of outdoor PM2.5 on grip strength may reflect transient physiological changes following short-term exposure. However, these findings should not be interpreted as evidence of protective effects of air pollution on aging. Rather, they highlight the complexity of air pollution’s health impacts and the value of longitudinal data in capturing time-sensitive effects. Further research is needed to better understand these patterns and their implications in high-exposure settings.
Al-Soneidar W*, Harper S, Coutlée F, Gheit T, Tommasino M, Nicolau B. “Prevalence of Alpha, Beta, and Gamma Human Papillomaviruses in Patients with Head and Neck Cancer and Noncancer Controls and Relation to Behavioral Factors.” The Journal of Infectious Diseases. 2024;229(4):1088–1096
Background. Human papillomaviruses (HPVs) cause head and neck cancer (HNC), which is increasing in incidence in developed countries. We investigated the prevalence of alpha (α), beta (β), and gamma (γ) HPVs among HNC cases and controls, and their relationship with sociodemographic, behavioral, and oral health factors. Methods. We obtained oral rinse and brush samples from incident HNC cases (n = 369) and hospital-based controls (n = 439) and tumor samples for a subsample of cases (n = 121). We genotyped samples using polymerase chain reaction with PGMY09–PGMY11 primers and linear array for α-HPV and type-specific multiplex genotyping assay for β-HPV and γ-HPV. Sociodemographic and behavioral data were obtained from interviews. Results. The prevalence of α-, β-, and γ-HPV among controls was 14\%, 56\%, and 24\%, respectively, whereas prevalence among cases was 42\%, 50\%, and 33\%, respectively. Prevalence of α- and γ-HPV, but not β-HPV, increased with increase in sexual activity, smoking, and drinking habits. No HPV genus was associated with oral health. Tumor samples included HPV genotypes exclusively from the α-genus, mostly HPV-16, in 80\% of cases. Conclusions. The distribution of α- and γ-HPV, but not β-HPV, seems to vary based on sociodemographic and behavioral characteristics. We did not observe the presence of cutaneous HPV in tumor tissues. © The Author(s) 2023. Published by Oxford University Press on behalf of Infectious Diseases Society of America. All rights reserved.
Burgos-Ochoa L, Bertens L, Boderie N, Gravesteijn B, Obermann-Borst S, Rosman A, P.R.E.P.A.R.E.-consortium. “Impact of COVID-19 Mitigation Measures on Perinatal Outcomes in the Netherlands.” Public Health. 2024;236:322–327
Objective: Investigate the acute impact of COVID-19 mitigation measures implemented in March 2020 on a comprehensive range of perinatal outcomes. Study design: National registry-based quasi-experimental study. Methods: We obtained data from the Dutch Perinatal Registry (2010-2020) which was linked to multiple population registries containing sociodemographic variables. A difference-in-discontinuity approach was used to examine the impact of COVID-19 mitigation measures on various perinatal outcomes. We investigated preterm birth incidence across onset types, alongside other perinatal outcomes including low birth weight, small-for-gestational-age, NICU admission, low-APGAR-score, perinatal mortality, neonatal death, and stillbirths. Results: The analysis of the national-level dataset revealed a consistent pattern of reduced preterm births after the enactment of COVID-19 mitigation measures on March 9, 2020 (OR = 0.80, 95\% CI 0.68-0.96). A drop in spontaneous preterm births post-implementation was observed (OR = 0.80, 95\% CI 0.62-0.98), whereas no change was observed for iatrogenic births. Regarding stillbirths (OR = 0.95, 95\% CI 0.46-1.95) our analysis did not find compelling evidence of substantial changes. For the remaining outcomes, no discernible shifts were observed. Conclusions: Our findings confirm the reduction in preterm births following COVID-19 mitigation measures in the Netherlands. No discernible changes were observed for other outcomes, including stillbirths. Our results challenge previous concerns of a potential increase in stillbirths contributing to the drop in preterm births, suggesting alternative mechanisms.
Gravesteijn B, Boderie N, Akker T, Bertens L, Bloemenkamp K, Burgos Ochoa L, P.R.E.P.A.R.E.-consortium. “Effect of COVID-19 Lockdown on Maternity Care and Maternal Outcome in the Netherlands: A National Quasi-Experimental Study.” Public Health. 2024;235:15–25
Objectives: The COVID-19 pandemic and associated lockdowns disrupted health care worldwide. High-income countries observed a decrease in preterm births during lockdowns, but maternal pregnancy–related outcomes were also likely affected. This study investigates the effect of the first COVID-19 lockdown (March–June 2020) on provision of maternity care and maternal pregnancy–related outcomes in the Netherlands. Study design: National quasi-experimental study. Methods: Multiple linked national registries were used, and all births from a gestational age of 24+0 weeks in 2010–2020 were included. In births starting in midwife-led primary care, we assessed the effect of lockdown on provision of care. In the general pregnant population, the impact on characteristics of labour and maternal morbidity was assessed. A difference-in-regression-discontinuity design was used to derive causal estimates for the year 2020. Results: A total of 1,039,728 births were included. During the lockdown, births to women who started labour in midwife-led primary care (49\%) more often ended at home (27\% pre-lockdown, +10\% [95\% confidence interval: +7\%, +13\%]). A small decrease was seen in referrals towards obstetrician-led care during labour (46\%, −3\% [−5\%,−0\%]). In the overall group, no significant change was seen in induction of labour (27\%, +1\% [−1\%, +3\%]). We found no significant changes in the incidence of emergency caesarean section (9\%, −1\% [−2\%, +0\%]), obstetric anal sphincter injury (2\%, +0\% [−0\%, +1\%]), episiotomy (21\%, −0\% [−2\%, +1\%]), or post-partum haemorrhage: {$>$}1000 ml (6\%, −0\% [−1\%, +1\%]). Conclusions: During the first COVID-19 lockdown in the Netherlands, a substantial increase in homebirths was seen. There was no evidence for changed available maternal outcomes, suggesting that a maternity care system with a strong midwife-led primary care system may flexibly and safely adapt to external disruptions. © 2024 The Author(s)
Hetherington E*, Darling E, Harper S, Nguyen F, Schummers L, Norman W. “Inequalities in Access to Prenatal Care during the COVID-19 Pandemic: Analysis of a Population-Based Cohort.” Paediatric and Perinatal Epidemiology. 2024;38(4):291–301
Background: Before the COVID-19 pandemic, access to prenatal care was lower among some socio-demographic groups. This pandemic caused disruptions to routine preventative care, which could have increased inequalities. Objectives: To investigate if the COVID-19 pandemic increased inequalities in access to prenatal care among those who are younger, live in rural areas, have a lower socio-economic situation (SES) and are recent immigrants. Methods: We used linked administrative datasets from ICES to identify a population-based cohort of 455,245 deliveries in Ontario from January 2018 to December 2021. Our outcomes were first-trimester prenatal visits, first-trimester ultrasound and adequacy of prenatal care. We used joinpoint analysis to examine outcome time trends and identify trend change points. We stratified analyses by age, rural residence, SES and recent immigration, and examined risk differences (RD) with 95\% confidence intervals (CI) between groups at the beginning and end of the study period. Results: For all outcomes, we noted disruptions to care beginning in March or April 2020 and returning to previous trends by November 2020. Inequalities were stable across groups, except recent immigrants. In July 2017, 65.0\% and 69.8\% of recent immigrants and non-immigrants, respectively, received ultrasounds in the first trimester (RD −4.8\%, 95\% CI −8.0, −1.5). By October 2020, this had increased to 75.4\%, with no difference with non-immigrants (RD 0.4\%, 95\% CI −2.4, 3.2). Adequacy of prenatal care showed more intensive care as of November 2020, reflecting a higher number of visits. Conclusions: We found no evidence that inequalities between socio-economic groups that existed prior to the pandemic worsened after March 2020. The pandemic may be associated with increased access to care for recent immigrants. The introduction of virtual visits may have resulted in a higher number of prenatal care visits. © 2024 The Authors. Paediatric and Perinatal Epidemiology published by John Wiley \& Sons Ltd.
Lee M, Chang J, Deng Q, Hu P, Bixby H, Harper S, Liu J. “Effects of a Coal to Clean Heating Policy on Acute Myocardial Infarction in Beijing: A Difference-in-Differences Analysis.” The Lancet. Planetary Health. 2024;8(11):924–932
Background: In 2015, the Chinese Government launched the coal to clean heating policy (CHP), designed to improve air quality and health in China. The CHP banned household coal burning and provided subsidies for clean electric or gas-powered heating for millions of peri-urban and rural households. We aimed to investigate whether the CHP affected the incidence of acute myocardial infarction in Beijing townships. Methods: In this quasi-experimental study, we obtained township data on acute myocardial infarction hospital admissions and deaths, exposure to the CHP (yes vs no), and a range of covariates for periods before (Jan 1, 2013, to Dec 31, 2014) and after the CHP began (Jan 1, 2016, to Dec 31, 2017; and Jan 1, 2018, to Dec 31, 2019). The policy was gradually rolled out across villages, and townships in our study were considered exposed to the policy in periods when more than 50\% of their villages were assigned into the CHP. We estimated the effect of the CHP on township incidence of acute myocardial infarction for all adults (aged ≥35 years) and separately for sex and older adults (aged ≥65 years) using a difference-in-differences approach that accommodates the progressive roll-out of the policy. Findings: Of 307 townships in Beijing, we excluded 156 (51\%) urban townships where most villages had central heating and were thus ineligible for the CHP. Of the 151 peri-urban and rural Beijing townships considered eligible for the CHP, 75 (50\%) townships were exposed to the CHP by the end of 2017 and 92 (61\%) by the end of 2019. We estimated an overall reduction of 6·6\% (95\% CI –12·3 to –0·8) in the incidence of acute myocardial infarction from before to after roll-out of the CHP in exposed townships relative to those not exposed to the policy, with some evidence of larger effects in women (–11·7\% [–19·0 to –4·1\%]), older adults (–10·7\% [–17·4 to –3·6\%]), and in townships exposed for longer (–3·5\% [–9·5 to 2·8\%] after {$<$}2 years and –9·7\% [–18·3 to –0·5\%] after 2–4 years). Interpretation: Our results provide among the first empirical evidence of possible cardiovascular benefits from a household clean energy policy, and support efforts to implement and assess such policies in China and globally. Funding: Wellcome Trust, the Canadian Institutes for Health Research, and the National Natural Science Foundation of China. © 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license
Nandi A, Agarwal P, Chandrashekar A, Maloney S, Richardson R, Thakur L, Harper S. “Access to Affordable Daycare and Women’s Mental Health in Rajasthan, India: Evidence from a Cluster-Randomised Social Intervention.” Journal of Global Health. 2024;14:04063
Background Women in India are often responsible for unpaid household work, family caregiving, and paid work, which can contribute to poorer mental health. The provision of childcare has the potential to improve women’s mental health, but evidence on the effects of providing access to daycare is limited. Methods We designed a cluster-randomised trial and used data from a sample of 2858 mothers with age-eligible children from 160 village hamlets in rural Rajasthan, India, to evaluate the impact of providing access to a community-based daycare programme on social and emotional aspects of women’s mental health. We conducted a baseline survey in early 2016, randomised hamlets to intervention or control groups approximately six months later, and delivered the final post-intervention survey approximately two years thereafter. Results Treatment assignment increased the probability that a respondent used a daycare over the two-year follow-up by 40.9 percentage points. Providing randomised access to a daycare resulted in 0.2 (95\% confidence interval (CI)\, = \,−0.1, 0.4) fewer symptoms of mental distress, representing a 9.5\% decline compared to the baseline mean of 2.1 symptoms, as well as a 3.7 (95\% CI\, = \,−0.8, 8.3) percentage point increase in the proportion of women who reported feeling very happy, equivalent to an 11.0\% increase relative to the baseline mean of 33.6\%. Among social indicators, treatment assignment was associated with a 5.6 (95\% CI\, = \,−1.2, 12.4) percentage point increase in membership in an association, a relative increase of 43.4\% compared to the baseline mean of 12.9\%. The intervention did not have an appreciable impact on measures of life satisfaction or trust in institutions. Two-stage least squares instrumental variable analyses showed that daycare use decreased mental distress by 0.4 (95\% CI\, = \,−0.1, 0.8) symptoms, increased the proportion of women who were very happy by 9.4 (95\% CI\, = \,0.0, 17.6) percentage points, and increased membership in an organisation by 15.9 (95\% CI\, = \,8.4, 23.7) percentage points. Conclusions The provision of affordable, community-based daycare was associated with substantial uptake and showed potential for improving mothers’ mental health in a rural context where most women were not employed in the formal labour force. Registration ISRCTN clinical trial registry (ISRCTN45369145), registered on 16 May 2016; American Economic Association’s registry for randomised controlled trials (AEARCTR-0000774), registered on 15 July 2015.
Socha P*, Harper S, Hutcheon J. “Methods of Confounder Selection in Obstetrics and Gynaecology Studies: An Overview of Recent Practice.” BJOG: An International Journal of Obstetrics and Gynaecology. 2024;131(10):1430–1431
Socha P*, Harper S, Strumpf E, Murphy K, Hutcheon J. “Antenatal Corticosteroids and Newborn Respiratory Outcomes in Twins: A Regression Discontinuity Study.” BJOG: An International Journal of Obstetrics and Gynaecology. 2024;131(8):1064–1071
Objective: To estimate the effect of antenatal corticosteroids on newborn respiratory morbidity in twins. Design: Regression discontinuity applied to population-based birth registry data. Setting: British Columbia, Canada, 2008–2018. Population: Twin pregnancies admitted for birth between 31+0 and 36+6 weeks of gestation. Methods: During our study period, Canadian clinical practice guidelines recommended antenatal corticosteroid administration for imminent preterm birth up to 33+6 weeks. We used a logistic model to compare the predicted risks of our outcomes among pregnancies admitted for birth immediately before this clinical cut-point (higher probability of exposure to antenatal corticosteroids) versus immediately after it (lower probability). Main outcome measures: Our primary outcome was a composite of newborn respiratory distress or in-hospital death. Our secondary outcome was a composite of newborn respiratory intervention or in-hospital death. Results: Among 2524 pregnancies (5035 liveborn twins), 47\% of admissions before 34+0 weeks of gestation were exposed to antenatal corticosteroids but only 4.2\% of admissions after this cut-point were exposed. The risk of newborn respiratory distress or in-hospital mortality increased abruptly at 34+0 weeks, corresponding to a protective effect of treatment (risk ratio [RR] 0.69, 95\% CI 0.53–0.90; risk difference [RD]~−12 cases per 100 births, 95\% CI −20 to −4.1). There was no clear evidence for or against an effect on newborn respiratory intervention or in-hospital death (RR 0.89, 95\% CI 0.70–1.13; RD −4.2 per 100, 95\% CI −13 to +4.2). Conclusions: Our findings provide evidence for the effectiveness of antenatal corticosteroids in preventing adverse newborn respiratory outcomes in twins. © 2024 The Authors. BJOG: An International Journal of Obstetrics and Gynaecology published by John Wiley \& Sons Ltd.
Anato J, Ma H, Hamilton M, Xia Y, Harper S, Buckeridge D, Maheu-Giroux M. “Impact of a Vaccine Passport on First-Dose SARS-CoV-2 Vaccine Coverage by Age and Area-Level Social Determinants of Health in the Canadian Provinces of Quebec and Ontario: An Interrupted Time Series Analysis.” CMAJ Open. 2023;11(5):995–1005
Harper S, Nandi A. “Empirical Challenges in Defining Treatments and Time in the Evaluation of Gun Laws.” Epidemiology. 2023;34(6):793–795
Rigorous policy analysis is hard and fraught with many empirical challenges, which are often compounded for issues that are politically polarizing, such as gun policy in the United States. In this issue, Sharkey and Kang1 analyze over two decades worth of changes in US state-level gun policies and argue that they have been extremely successful in reducing gun deaths. Sharkey and Kang’s1 analysis has a number of commendable dimensions. They motivate their analysis with an important and timely policy question, which is whether changes in gun policies contributed to the decline in gun-related deaths during the period from 1991 to 2016. Although their primary focus is on estimates generated using a pooled outcome (all gun deaths) and a cumulative index of all state changes to gun laws, they helpfully provide separate estimates for several different classes of gun laws and examine their impact across different subsets of gun-related and nongun-related outcomes. They also conduct several sensitivity analyses (using an alternative database of laws, fixed effects, and instrumental variable analyses) to bolster the credibility of their causal claims, as well as providing the source code that was used to generate their figures and tables. These positive aspects notwithstanding, the preferred model specification and analytic framework that Sharkey and Kang1 use to generate their evidence also raise some important methodologic questions. © 2023 Lippincott Williams and Wilkins. All rights reserved.
Hawkins S, Harper S, Baum C, Kaufman J. “Associations between State-Level Changes in Reproductive Health Services and Indicators of Severe Maternal Morbidity.” JAMA Pediatrics. 2023;177(1):93–95
Hetherington E*, Harper S, Davidson R, Festo C, Lampkin N, Mtenga S, Teixeira C, Vincent I, Nandi A. “Impact Evaluation of the TAMANI Project to Improve Maternal and Child Health in Tanzania.” J Epidemiol Community Health. 2023;77(6):410–416
Background The Tabora Maternal and Newborn Health Initiative project was a multicomponent intervention to improve maternal and newborn health in the Tabora region of Tanzania. Components included training healthcare providers and community health workers, infrastructure upgrades, and improvements to health management. This study aimed to examine the impact of trainings on four key outcomes: skilled birth attendance, antenatal care, respectful maternity care and patient–provider communication. Methods Trainings were delivered sequentially at four time points between 2018 and 2019 in eight districts (two districts at a time). Cross-sectional surveys were administered to a random sample of households in all districts at baseline and after each training wave. Due to practical necessities, the original stepped wedge cluster randomised design of the evaluation was altered mid-programme. Therefore, a difference-in-differences for multiple groups in multiple periods was adopted to compare outcomes in treated districts to not yet treated districts. Risk differences were estimated for the overall average treatment effect on the treated and group/time dynamic effects. Results Respondents reported 3895 deliveries over the course of the study. The intervention was associated with a 12.9 percentage point increase in skilled birth attendance (95\% CI 0.4 to 25.4), which began to increase 4 months after the end of training in each district. There was little evidence of impact on antenatal care visits, respectful treatment during delivery and patient–provider communication. Conclusion Interventions to train local healthcare workers in basic and comprehensive emergency obstetric and newborn care increased skilled birth attendance but had limited impact on other pregnancy-related outcomes.
Socha P*, Harper S, Hutcheon JA. “Subgroup Effects Should Be Examined Using Both Relative and Absolute Effect Measures.” Australian and New Zealand Journal of Obstetrics and Gynaecology. 2023;63(3):469–472
Treatment effects can be measured on the relative scale (eg, risk ratios, odds ratios) or the absolute scale (eg, risk differences). If the baseline risk of an outcome is different between subgroups, the effect of the treatment will differ between subgroups on at least one scale (relative, absolute, or both). We illustrate this using two examples from the literature where only relative effects were estimated, but conclusions about subgroup differences would likely have changed had absolute effects also been considered. To identify all meaningful subgroup differences, researchers and clinicians should compare~effects on the relative and absolute scale.
Al-Soneidar WA*, Harper S, Madathil SA, Schlecht NF, Nicolau B. “Do Cutaneous Human Papillomavirus Genotypes Affect Head and Neck Cancer? Evidence and Bias-Correction from a Case-Control Study.” Cancer Epidemiology. 2022;79:102205
Background Three genera of human papillomavirus (HPV) infect the oral cavity and oropharynx— alpha (α), beta (β) and gamma (γ). While α-HPV infection is an established risk factor for head and neck cancers (HNC), the role of other genera remains unclear. We aimed to estimate the effect of α-, β-, γ-HPV on HNC using a hospital-based case-control study. Methods We recruited incident HNC cases (396) and controls (439), frequency-matched by age and sex from four main referral hospitals in Montreal, Canada. We collected information on sociodemographic and behavior characteristics using in-person interviews, and tested rinse, brush and tumor specimens for HPV genotypes. We estimated adjusted odds ratios (aOR) and 95\% confidence intervals (CI) for the effect of HPV on HNC using logistic regression, adjusting for confounding. We conducted probabilistic bias analysis to account for potential exposure misclassification, selection bias, and residual confounding. Results α-HPV genus had a strong effect on HNC, particularly HPV16 (aOR = 22.6; 95\% CI: 10.8, 47.2). β-HPV was more common among controls (aOR = 0.80; 95\% 0.57, 1.11). After adjustment for HPV16, we found weaker evidence for γ-HPV (aOR = 1.29; 95\% CI: 0.80, 2.08). Combined bias analyses for HPV16 increased the strength of the point estimate, but added imprecision (aOR = 54.2, 95\%~CI: 10.7, 385.9). Conclusions α-HPV, especially HPV16, appears to increase the risk for HNC, while there is little evidence for an effect of β- or γ-HPV. β-HPV may have a preventive effect, while γ-HPV may increase the risk of HNC, although to a lesser extent than that of α-HPV. Results for cutaneous HPV were imprecise and less conclusive. Due to possible epidemiologic biases, the true relation between HPV and HNC could be underestimated in the literature. Further improvement in current methods and more studies of the biologic mechanisms of the three genera in HNC development are warranted.
Al-Soneidar WA*, Harper S, Alli BY, Nicolau B. “Interaction of HPV16 and Cutaneous HPV in Head and Neck Cancer.” Cancers. 2022;14(21):5197
Objectives: Human papillomavirus 16 (HPV16) is an established risk factor for Head and Neck Cancer (HNC). Recent reports have shown that genotypes from the beta (β) and gamma (γ) genera, also known as cutaneous HPV, can be found in the oral cavity, but their role is largely unidentified. We investigated the interaction between oral HPV16 and cutaneous HPV in HNC. Methods: We use data on incident HNC cases (n = 384) and frequency-matched hospital-based controls (n = 423) from the HeNCe Life study in Montreal, Canada. Participants were tested for alpha HPV and cutaneous genera using oral mouth rinse and brush samples. We used unconditional logistic regression to obtain adjusted odds ratios (aOR) and 95\% confidence interval (CI) as a measure of the effect between HPV and HNC and assessed the interaction between HPV genotypes on the multiplicative and additive scales. Results: Prevalence of HPV infection was higher among cases (73\%) than controls (63.4\%), with cases more likely to be coinfected with more than a single genotype, 52.9\% vs. 43.5\%, respectively. Infection with HPV16 alone had a strong effect on HNC risk aOR = 18.2 [6.2, 53.2], while infection with any cutaneous HPV, but not HPV16, appeared to have the opposite effect aOR = 0.8 [0.6, 1.1]. The observed effect of joint exposure to HPV16 and any cutaneous HPV (aOR = 20.4 [8.3, 50.1]) was stronger than the expected effect based on an assumption of independent exposures but was measured with considerable imprecision. While the point estimate suggests a positive interaction between HPV16 and cutaneous HPV, results were imprecise with relative excess risk due to interaction (RERI) = 2.4 [−23.3, 28.2]. Conclusion: There could be biologic interaction between HPV16 and genotypes from cutaneous genera, which warrants further investigation. Although cutaneous HPVs are not usually found in tumor tissues, they are cofactors that could interact with HPV16 in the oral cavity and thus strengthen the latter’s carcinogenic effect.
Capurro DA*, Harper S. “Socioeconomic Inequalities in Health Care Utilization in Paraguay: Description of Trends from 1999 to 2018.” Journal of Health Services Research & Policy. 2022;27(3):13558196221079160
ObjectiveParaguay?s health care system is characterized by segmented provision and low public spending, with limited coverage and asymmetries in terms of access and quality of care. The present study provides national estimates of income-related inequality in health care utilization and trends in the country over the past two decades.MethodsUsing data from the Paraguayan Permanent Household Survey, we estimated socioeconomic inequality in health care use during the period 1999?2018. We used poverty-to-income ratio as the socioeconomic stratifier and defined health care use as having reported a health problem and subsequent health care use in the last 90 days before interview. Inequality was summarized by rank- and level-based versions of the Concentration Index for binary outcomes.ResultsInequalities affecting those with lower incomes were present in all years assessed, although the magnitude of these inequalities declined over time. Inequality as expressed by the rank-based index decreased from 0.209 (95\%CI 0.164; 0.253) in 1999 to 0.032 (95\%CI -0.010; 0.075) in 2018. The level-based index decreased from 0.076 (95\%CI -0.029; 0.182) in 1999 to 0.024 (0.002; 0.045) in 2018. Trends in both indices were generally stable from 1999 to 2009, with a noticeable decrease in 2010. The sharpest decreases relative to the 1999 baseline were observed in the period 2010?2018, reflecting changes in health care use and income distribution. Stratification by area, sex and older people suggest similar trends within subgroups.ConclusionsDecreases in inequality coincide temporally with increments in public health expenditure, removal of user fees in public health care facilities and the expansion of conditional cash-transfer programmes. Future research should disentangle the role of each of these policies in explaining the trends described.
Carabali M, Harper S, Lima Neto AS, family = Santos de Sousa g=G, Caprara A, Restrepo BN, Kaufman JS. “Decomposition of Socioeconomic Inequalities in Arboviral Diseases in Brazil and Colombia (2007–2017).” Transactions of The Royal Society of Tropical Medicine and Hygiene. 2022;116(8):trac004
We used surveillance data from Brazil and Colombia during 2007–2017 to assess the presence of socioeconomic inequalities on dengue, chikungunya and Zika at the neighborhood level in two Latin American cities.To quantify the inequality, we estimated and decomposed the relative concentration index of inequality (RCI) accounting for the spatiotemporal distribution of the diseases.There were 281 426 arboviral cases notified in Fortaleza, Brazil, and 40 889 in Medellin, Colombia. The RCI indicated greater concentration of dengue cases among people living in low socioeconomic settings in both sites. The RCIs for chikungunya in Fortaleza covered the line of equality during their introduction in 2014, while the RCIs for Zika and chikungunya in Medellin indicated the presence of a small inequality. The RCI decomposition showed that year of notification and age were the main contributors to this inequality. In Medellin, the RCI decomposition showed that age and access to waste management accounted for 75.5\%, 72.2\% and 54.5\% of the overall inequality towards the poor for dengue, chikungunya and Zika, respectively.Our study presents estimates of the socioeconomic inequality of arboviruses and its decomposition in two Latin American cities. We corroborate the concentration of arboviral diseases in low socioeconomic neighborhoods and identify that year of occurrence, age, presence of healthcare facilities and waste management are key determinants of the heterogenous distribution of endemic arboviruses across the socioeconomic spectrum.
Clark S, Harper S, Weber B. “Growing Up in Rural America.” RSF: The Russell Sage Foundation Journal of the Social Sciences. 2022;8(4):1–47
This article examines the context of growing up in rural America and how rural roots shape life chances. The distinctive physical, social, and cultural attributes of rural areas can exacerbate many of the challenges of childhood poverty. Yet rural children have better access to public childcare services and perform as well as urban children on standardized tests. Life trajectories diverge most sharply when rural youths decide whether to leave their home communities. Those who stay typically face limited opportunities for higher education and well-paid, stable employment, whereas those who leave fare remarkably well with respect to their educational, economic, and health outcomes. In sum, growing up in rural America offers distinctive advantages and disadvantages, yet the benefits may accrue primarily to those who leave.
Hutcheon JA, Strumpf EC, Liauw J, Skoll MA, Socha P*, Srour M, Ting JY, Harper S. “Antenatal Corticosteroid Administration and Attention-Deficit/Hyperactivity Disorder in Childhood: A Regression Discontinuity Study.” CMAJ. 2022;194(7):E235-E241
Background: Antenatal corticosteroids reduce respiratory morbidity in preterm infants, but their use during late preterm gestation (34–36 weeks) is limited because their safety for longer-term child neurodevelopment is unclear. We sought to determine if fetuses with higher probability of exposure to antenatal corticosteroids had increased rates of prescriptions for attention-deficit/hyperactivity disorder (ADHD) medication in childhood, using a quasiexperimental design that better controls for confounding than existing observational studies. Methods: We identified 16 358 children whose birthing parents were admitted for delivery between 31 + 0 (31 weeks, 0 days) and 36 + 6 weeks’ gestation in 2000–2013, using a perinatal data registry from British Columbia, Canada, and linked their records with population-based child ADHD medication data (2000–2018). We used a regression discontinuity design to capitalize on the fact that pregnancies presenting for delivery immediately before and immediately after the clinical cut-off for antenatal corticosteroid administration of 34 + 0 weeks’ gestation have very different levels of exposure to corticosteroids, but are otherwise similar with respect to confounders. Results: Over a median follow-up period of 9 years, 892 (5.5\%) children had 1 or more dispensations of ADHD medication. Children whose birthing parents were admitted for delivery just before the corticosteroid clinical cut-off of 34 + 0 weeks’ gestation did not appear to be more likely to be prescribed ADHD medication than those admitted just after the cut-off (rate ratio 1.1, 95\% confidence interval [CI] 0.8 to 1.6; 1.3 excess cases per 100 children, 95\% CI −2.5 to 5.7). Interpretation: We found little evidence that children with higher probability of exposure to antenatal corticosteroids have higher rates of ADHD prescriptions in childhood, supporting the safety of antenatal corticosteroids for this neurodevelopmental outcome.
Li X, Baumgartner J, Harper S, Zhang X, Sternbach T, Barrington-Leigh C, Brehmer C, Robinson B, Shen G, Zhang Y, Tao S, Carter E. “Field Measurements of Indoor and Community Air Quality in Rural Beijing before, during, and after the COVID-19 Lockdown.” Indoor Air. 2022;32(8):e13095
The coronavirus (COVID-19) lockdown in China is thought to have reduced air pollution emissions due to reduced human mobility and economic activities. Few studies have assessed the impacts of COVID-19 on community and indoor air quality in environments with diverse socioeconomic and household energy use patterns. The main goal of this study was to evaluate whether indoor and community air pollution differed before, during, and after the COVID-19 lockdown in homes with different energy use patterns. Using calibrated real-time PM2.5 sensors, we measured indoor and community air quality in 147 homes from 30 villages in Beijing over 4 months including periods before, during, and after the COVID-19 lockdown. Community pollution was higher during the lockdown (61 ± 47 μg/m3) compared with before (45 ± 35 μg/m3, p {$<$} 0.001) and after (47 ± 37 μg/m3, p {$<$} 0.001) the lockdown. However, we did not observe significantly increased indoor PM2.5 during the COVID-19 lockdown. Indoor-generated PM2.5 in homes using clean energy for heating without smokers was the lowest compared with those using solid fuel with/without smokers, implying air pollutant emissions are reduced in homes using clean energy. Indoor air quality may not have been impacted by the COVID-19 lockdown in rural settings in China and appeared to be more impacted by the household energy choice and indoor smoking than the COVID-19 lockdown. As clean energy transitions occurred in rural households in northern China, our work highlights the importance of understanding multiple possible indoor sources to interpret the impacts of interventions, intended or otherwise.
Li X, Baumgartner J, Barrington-Leigh C, Harper S, Robinson B, Shen G, Sternbach T, Tao S, Zhang X, Zhang Y, Carter E. “Socioeconomic and Demographic Associations with Wintertime Air Pollution Exposures at Household, Community, and District Scales in Rural Beijing, China.” Environmental Science & Technology. 2022;56(12):8308–8318
The Chinese government implemented a national household energy transition program that replaced residential coal heating stoves with electricity-powered heat pumps for space heating in northern China. As part of a baseline assessment of the program, this study investigated variability in personal air pollution exposures within villages and between villages and evaluated exposure patterns by sociodemographic factors. We randomly recruited 446 participants in 50 villages in four districts in rural Beijing and measured 24 h personal exposures to fine particulate matter (PM2.5) and black carbon (BC). The geometric mean personal exposure to PM2.5 and BC was 72 and 2.5 μg/m3, respectively. The variability in PM2.5 and BC exposures was greater within villages than between villages. Study participants who used traditional stoves as their dominant source of space heating were exposed to the highest levels of PM2.5 and BC. Wealthier households tended to burn more coal for space heating, whereas less wealthy households used more biomass. PM2.5 and BC exposures were almost uniformly distributed by socioeconomic status. Future work that combines these results with PM2.5 chemical composition analysis will shed light on whether air pollution source contributors (e.g., industrial, traffic, and household solid fuel burning) follow similar distributions.
Nandi A, Charters TJ, Quamruzzaman A, Strumpf EC, Kaufman JS, Heymann J, Mukherji A, Harper S. “Health Care Services Use, Stillbirth, and Neonatal and Infant Survival Following Implementation of the Maternal Health Voucher Scheme in Bangladesh: A Difference-in-Differences Analysis of Bangladesh Demographic and Health Survey Data, 2000 to 2016.” PLOS Medicine. 2022;19(8):e1004022
Background Starting in 2006 to 2007, the Government of Bangladesh implemented the Maternal Health Voucher Scheme (MHVS). This program provides pregnant women with vouchers that can be exchanged for health services from eligible public and private sector providers. In this study, we examined whether access to the MHVS was associated with maternal health services utilization, stillbirth, and neonatal and infant mortality. Methods and findings We used information on pregnancies and live births between 2000 to 2016 reported by women 15 to 49 years of age surveyed as part of the Bangladesh Demographic and Health Surveys. Our analytic sample included 23,275 pregnancies lasting at least 7 months for analyses of stillbirth and between 15,125 and 21,668 live births for analyses of health services use, neonatal, and infant mortality. With respect to live births occurring prior to the introduction of the MHVS, 31.3\%, 14.1\%, and 18.0\% of women, respectively, reported receiving at least 3 antenatal care visits, delivering in a health institution, and having a skilled birth attendant at delivery. Rates of neonatal and infant mortality during this period were 40 and 63 per 1,000 live births, respectively, and there were 32 stillbirths per 1,000 pregnancies lasting at least 7 months. We applied a difference-in-differences design to estimate the effect of providing subdistrict-level access to the MHVS program, with inverse probability of treatment weights to address selection into the program. The introduction of the MHVS program was associated with a lagged improvement in the probability of delivering in a health facility, one of the primary targets of the program, although associations with other health services were less evident. After 6 years of access to the MHVS, the probabilities of reporting at least 3 antenatal care visits, delivering in a health facility, and having a skilled birth attendant present increased by 3.0 [95\% confidence interval (95\% CI) = −4.8, 10.7], 6.5 (95\% CI = −0.6, 13.6), and 5.8 (95\% CI = −1.8, 13.3) percentage points, respectively. We did not observe evidence consistent with the program improving health outcomes, with probabilities of stillbirth, neonatal mortality, and infant mortality decreasing by 0.7 (95\% CI = −1.3, 2.6), 0.8 (95\% CI = −1.7, 3.4), and 1.3 (95\% CI = −2.5, 5.1) percentage points, respectively, after 6 years of access to the MHVS. The sample size was insufficient to detect smaller associations with adequate precision. Additionally, we cannot rule out the possibility of measurement error, although it was likely nondifferential by treatment group, or unmeasured confounding by concomitant interventions that were implemented differentially in treated and control areas. Conclusions In this study, we found that the introduction of the MHVS was positively associated with the probability of delivering in a health facility, but despite a longer period of follow-up than most extant evaluations, we did not observe attendant reductions in stillbirth, neonatal mortality, or infant mortality. Further work and engagement with stakeholders is needed to assess if the MHVS has affected the quality of care and health inequalities and whether the design and eligibility of the program should be modified to improve maternal and neonatal health outcomes.
Sternbach TJ, Harper S, Li X, Zhang X, Carter E, Zhang Y, Shen G, Fan Z, Zhao L, Tao S, Baumgartner J. “Effects of Indoor and Outdoor Temperatures on Blood Pressure and Central Hemodynamics in a Wintertime Longitudinal Study of Chinese Adults.” Journal of Hypertension. 2022;40(10):1950–1959
Objectives:~ We aimed to estimate the effects of indoor and outdoor temperature on wintertime blood pressure (BP) among peri-urban Beijing adults. Methods:~ We enrolled 1279 adults (ages: 40–89 years) and conducted measurements in two winter campaigns in 2018–2019 and 2019–2020. Study staff traveled to participant homes to administer a questionnaire and measure brachial and central BP. Indoor temperature was measured in the 5 min prior to BP measurement. Outdoor temperature was estimated from regional meteorological stations. We used multivariable mixed-effects regression models to estimate the within-individual and between-individual effects of indoor and outdoor temperatures on BP. Results:~ Indoor and outdoor temperatures ranged from 0.0 to 28 °C and −14.3 to 6.4 °C, respectively. In adjusted models, a 1 °C increase in indoor temperature was associated with decreased SBP [−0.4 mmHg, 95\% confidence interval (CI): −0.7 to −0.1 (between-individual; brachial and central BP); −0.5 mmHg, 95\% CI: −0.8 to −0.2 (within-individual, brachial BP); −0.4 mmHg, 95\% CI: −0.7 to −0.2 (within-individual, central BP)], DBP [−0.2 mmHg, 95\% CI:−0.4 to −0.03 (between-individual); −0.3 mmHg, 95\% CI: −0.5 to −0.04 (within-individual)], and within-individual pulse pressure [−0.2 mmHg, 95\% CI: −0.4 to −0.04 (central); −0.3 mmHg, 95\% CI: −0.4 to −0.1 (brachial)]. Between-individual SBP estimates were larger among participants with hypertension. There was no evidence of an effect of outdoor temperature on BP. Conclusion:~ Our results support previous findings of inverse associations between indoor temperature and BP but contrast with prior evidence of an inverse relationship with outdoor temperature. Wintertime home heating may be a population-wide intervention strategy for high BP and cardiovascular disease in China.
Windle SB, Socha P*, Nazif-Munoz JI, Harper S, Nandi A. “The Impact of Cannabis Decriminalization and Legalization on Road Safety Outcomes: A Systematic Review.” American Journal of Preventive Medicine. 2022;63(6):1037–1052
Introduction: There is substantial debate concerning the impact of cannabis decriminalization and legalization on road safety outcomes. Methods: Seven databases were systematically searched: Embase, MEDLINE, and PsycINFO through Ovid as well as Web of Science Core Collection, SafetyLit, Criminal Justice Database (ProQuest), and Transport Research International Documentation (from inception to June 16, 2021). Eligible primary studies examined group-level cannabis decriminalization or legalization and a road safety outcome in any population. Results: A total of 65 reports of 64 observational studies were eligible, including 39 that applied a quasi-experimental design. Studies examined recreational cannabis legalization (n = 50), medical cannabis legalization (n = 22), and cannabis decriminalization (n = 5). All studies except 1 used data from the U.S. or Canada. Studies found mixed impacts of legalization on attitudes, beliefs, and self-reported driving under the influence. Medical legalization, recreational legalization, and decriminalization were associated with increases in positive cannabis tests among drivers. Few studies examined impacts on alcohol or other drug use, although findings suggested a decrease in positive alcohol tests among drivers associated with medical legalization. Medical legalization was associated with reductions in fatal motor-vehicle collisions, whereas recreational legalization was conversely associated with increases in fatal collisions. Discussion: Increased cannabis positivity may reflect changes in cannabis use; however, it does not in itself indicate increased impaired driving. Subgroups impacted by medical and recreational legalization, respectively, likely explain opposing findings for fatal collisions. More research is needed concerning cannabis decriminalization; the impacts of decriminalization and legalization on nonfatal injuries, alcohol and other drugs; and the mechanisms by which legalization impacts road safety outcomes. © 2022 American Journal of Preventive Medicine
Carabali M, Harper S, Lima Neto AS, family = Santos de Sousa g=G, Caprara A, Restrepo BN, Kaufman JS. “Spatiotemporal Distribution and Socioeconomic Disparities of Dengue, Chikungunya and Zika in Two Latin American Cities from 2007 to 2017.” Tropical Medicine & International Health. 2021;26(3):301–315
Objective To assess the presence, pattern and magnitude of socioeconomic inequalities on dengue, chikungunya and Zika in Latin America, accounting for their spatiotemporal distribution. Methods Using longitudinal surveillance data (reported arboviruses) from Fortaleza, Brazil and Medellin, Colombia (2007–2017), we fit Bayesian hierarchical models with structured random effects to estimate: (i) spatiotemporally adjusted incidence rates; (ii) Relative Concentration Index and Absolute Concentration Index of inequality; (iii) temporal trends in RCIs; and (iv) socioeconomic-specific estimates of disease distribution. The spatial analysis was conducted at the neighbourhood level (urban settings). The socioeconomic measures were the median monthly household income (MMHI) for Brazil and the Socio-Economic Strata index (SES) in Colombia. Results There were 281 426 notified arboviral cases in Fortaleza and 40 887 in Medellin. We observed greater concentration of dengue among residents of low socioeconomic neighbourhoods in both cities: Relative Concentration Index = −0.12 (95\% CI = −0.13, −0.10) in Fortaleza and Relative Concentration Index = −0.04 (95\% CI = −0.05, −0.03) in Medellin. The magnitude of inequalities varied over time across sites and was larger during outbreaks. We identified a non-monotonic association between disease rates and socioeconomic measures, especially for chikungunya, that changed over time. The Relative Concentration Index and Absolute Concentration Index showed few if any inequalities for Zika. The socioeconomic-specific model showed increased disease rates at MMHI below US\$400 in Brazil and at SES-index below level four, in Colombia. Conclusions We provide robust quantitative estimates of socioeconomic inequalities in arboviruses for two Latin American cities. Our findings could inform policymaking by identifying spatial hotspots for arboviruses and targeting strategies to decrease disparities at the local level.
Graham E, Watson T, Deschênes S, Filion K, Henderson M, Harper S, Rosella L, Schmitz N. “Depression-Related Weight Change and Incident Diabetes in a Community Sample.” Scientific Reports. 2021;11(1)
This cohort study aimed to compare the incidence of type 2 diabetes in adults with depression-related weight gain, depression-related weight loss, depression with no weight change, and no depression. The~study sample included 59,315 community-dwelling adults in Ontario, Canada. Depression-related weight change in the past 12~months was measured using the Composite International Diagnostic Interview—Short Form. Participants were followed for up to 20~years using administrative health data. Cox proportional hazards models compared the incidence of type 2 diabetes in adults with depression-related weight change and in adults with no depression. Adults with depression-related weight gain had an increased risk of type 2 diabetes compared to adults no depression (HR 1.70, 95\% CI 1.32–2.20), adults with depression-related weight loss (HR 1.62, 95\% CI 1.09–2.42), and adults with depression with no weight change (HR 1.39, 95\% CI 1.03–1.86). Adults with depression with no weight change also had an increased risk of type 2 diabetes compared to those with no depression (HR 1.23, 95\% CI 1.04–1.45). Associations were stronger among women and persisted after adjusting for attained overweight and obesity. Identifying symptoms of weight change in depression may aid in identifying adults at higher risk of type 2 diabetes and in developing tailored prevention strategies. © 2021, The Author(s).
Harper S, Riddell CA*, King NB. “Declining Life Expectancy in the United States: Missing the Trees for the Forest.” Annual Review of Public Health. 2021;42(1):381–403
In recent years, life expectancy in the United States has stagnated, followed by three consecutive years of decline. The decline is small in absolute terms but is unprecedented and has generated considerable research interest and theorizing about potential causes. Recent trends show that the decline has affected nearly all race/ethnic and gender groups, and the proximate causes of the decline are increases in opioid overdose deaths, suicide, homicide, and Alzheimer's disease. A slowdown in the long-term decline in mortality from cardiovascular diseases has also prevented life expectancy from improving further. Although a popular explanation for the decline is the cumulative decline in living standards across generations, recent trends suggest that distinct mechanisms for specific causes of death are more plausible explanations. Interventions to stem the increase in overdose deaths, reduce access to mechanisms that contribute to violent deaths, and decrease cardiovascular risk over the life course are urgently needed to improve mortality in the United States.
Hutcheon JA, Harper S. “If It Sounds Too Good to Be True, It Probably Is: Conducting within-Woman Comparison Studies with Only One Exposure Observation per Woman.” Paediatric and Perinatal Epidemiology. 2021;35(4):447–449
Hutcheon JA, Harper S. “No, It Is Not Too Good to Be True: Response by Hutcheon and Harper.” Paediatric and Perinatal Epidemiology. 2021;35(6):781–782
Jahagirdar D, Dimitris M, Strumpf E, Kaufman JS, Harper S, Heymann J, Atabay E, Vincent I, Nandi A. “Balancing Work and Care: The Effect of Paid Adult Medical Leave Policies on Employment in Europe.” Journal of Social Policy. 2021;50(3):552–568
Increasing caregiving needs for family members has created pressure on prime-age workers. Combined with the ageing population, the demand for care related to illness and disability by relatives mean more of the workforce may have to consider caring needs (Bauer and Sousa-Poza, 2015). ‘Informal caregivers’ provide care generally without payment (Yoo et al. , 2004). In contrast to formal care, informal caregivers usually have a close relationship with the recipient: for example, siblings and adult children. Informal caregiving is considered a desirable option to meet support needs from several perspectives; these caregivers may be preferred by recipients relative to formal arrangements especially during severe acute illnesses. Caregivers may also feel a personal sense of responsibility to look after loved ones rather than defer to strangers (Fine, 2012) though this may depend on the individual’s needs and the available alternatives. Although men are starting to play an important role due to shifting social gender roles, the vast majority of informal caregivers are women who increasingly attempt to juggle caring with labour force participation (Carmichael et al. , 2008).
Arsenault C*, Harper S, Nandi A. “Effect of Vaccination on Children’s Learning Achievements: Findings from the India Human Development Survey.” J Epidemiol Community Health. 2020;74(10):778–784
Background Beyond the prevention of illness and death, vaccination may provide additional benefits such as improved educational outcomes. However, there is currently little evidence on this question. Our objective was to estimate the effect of childhood vaccination on learning achievements among primary school children in India. Methods We used cohort data from the India Human Development Survey. Vaccination status and confounders were measured among children who were at least 12 months old at baseline in 2004–2005. In 2011–2012, the same children completed basic reading, writing and math tests. We estimated the effect of full vaccination during childhood on learning achievements using inverse probability of treatment-weighted logistic regression models and results reported on the risk difference scale. The propensity score included 33 potential community-, household-, mother- and child-level confounders as well as state fixed effects. Results Among the 4877 children included in our analysis, 54\% were fully vaccinated at baseline, and 54\% could read by the age of 8–11 years. The estimated effect of full vaccination on learning achievements ranged from 4 to 6 percentage points, representing relative increases ranging from 6\% to 12\%. Bias analysis suggested that our observed effects could be explained by unmeasured confounding, but only in the case of strong associations with the treatment and outcome. Conclusion These results support the hypothesis that vaccination has lasting effects on children’s learning achievements. Further work is needed to confirm findings and elucidate the potential mechanisms linking vaccines to educational outcomes.
Goldstein N, Hamra G, Harper S. “Are Descriptions of Methods Alone Sufficient for Study Reproducibility? An Example from the Cardiovascular Literature.” Epidemiology. 2020;31(2):184–188
Hawkins S, Ghiani M, Harper S, Baum C, Kaufman J. “Impact of State-Level Changes on Maternal Mortality: A Population-Based, Quasi-Experimental Study.” American Journal of Preventive Medicine. 2020;58(2):165–174
Introduction: Recent increases in maternal mortality and persistent disparities have led to speculation about why the U.S. has higher rates than most high-income countries. The aim was to examine the impact of changes in state-level factors plausibly linked to maternal mortality on overall rates and by race/ethnicity. Methods: This quasi-experimental, population-based, difference-in-differences study used 2007–2015 National Vital Statistics System microdata mortality files from 38 states and DC. The primary exposures were 5 state-level sexual and reproductive health indicators and 6 health and economic conditions. Maternal mortality rate was defined as number of deaths of women while pregnant or within 42 days of termination of pregnancy per 100,000 live births. A difference-in-differences zero-inflated negative binomial regression model was estimated using the race/ethnicity-age-state-year population as the denominator and adjusting for race/ethnicity, age, state, and year. Data were analyzed in 2017–2018. Results: There were 4,767 deaths among women up to age 44 years, resulting in a maternal mortality rate of 17.9. Reducing the proportion of Planned Parenthood clinics by 20\% from the state-year mean increased the maternal mortality rate by 8\% (incidence rate ratio, 1.08; 95\% CI = 1.04, 1.12). States that enacted legislation to restrict abortions based on gestational age increased the maternal mortality rate by 38\% (incidence rate ratio, 1.38; 95\% CI = 1.03, 1.84). Planned Parenthood clinic closures negatively impacted all women, increasing mortality by 6\%–15\% across racial/ethnic groups, whereas gestational limits primarily increased mortality among white women. Conclusions: Recent fiscal and legislative changes reducing women's access to family planning and reproductive health services have contributed to rising maternal mortality rates. © 2019 American Journal of Preventive Medicine
Hutcheon J, Harper S, Liauw J, Skoll M, Srour M, Strumpf E. “Antenatal Corticosteroid Administration and Early School Age Child Development: A Regression Discontinuity Study in British Columbia, Canada.” PLoS Medicine. 2020;17(12)
Background There are growing concerns that antenatal corticosteroid administration may harm children's neurodevelopment. We investigated the safety of antenatal corticosteroid administration practices for children's overall developmental health (skills and behaviors) at early school age. Methods and findings We linked population health and education databases from British Columbia (BC), Canada to identify a cohort of births admitted to hospital between 31 weeks, 0 days gestation (31+0 weeks), and 36+6 weeks, 2000 to 2013, with routine early school age child development testing. We used a regression discontinuity design to compare outcomes of infants admitted just before and just after the clinical threshold for corticosteroid administration of 34+0 weeks. We estimated the median difference in the overall Early Development Instrument (EDI) score and EDI subdomain scores, as well as risk differences (RDs) for special needs designation and developmental vulnerability ({$<$}10th percentile on 2 or more subdomains). The cohort included 5,562 births admitted between 31+0 and 36+6 weeks, with a median EDI score of 40/50. We found no evidence that antenatal corticosteroid administration practices were linked with altered child development at early school age: median EDI score difference of -0.5 [95\% CI: -2.2 to 1.7] (p = 0.65), RD per 100 births for special needs designation -0.5 [-4.2 to 3.1] (p = 0.96) and for developmental vulnerability of 3.9 [95\% CI: -2.2 to 10.0] (p = 0.24). A limitation of our study is that the regression discontinuity design estimates the effect of antenatal corticosteroid administration at the gestational age of the discontinuity, 34 + 0 weeks, so our results may become less generalisable as gestational age moves further away from this point. Conclusions Our study did not find that that antenatal corticosteroid administration practices were associated with child development at early school age. Our findings may be useful for supporting clinical counseling about antenatal corticosteroids administration at late preterm gestation, when the balance of harms and benefits is less clear. © 2020 Public Library of Science. All rights reserved.
Johri M, Chandra D, Kone K, Sylvestre M, Mathur A, Harper S, Nandi A. “Social and Behavior Change Communication Interventions Delivered Face-to-Face and by a Mobile Phone to Strengthen Vaccination Uptake and Improve Child Health in Rural India: Randomized Pilot Study.” JMIR mHealth and uHealth. 2020;8(9)
Background: In resource-poor settings, lack of awareness and low demand for services constitute important barriers to expanding the coverage of effective interventions. In India, childhood immunization is a priority health strategy with suboptimal uptake. Objective: To assess study feasibility and key implementation outcomes for the Tika Vaani model, a new approach to educate and empower beneficiaries to improve immunization and child health. Methods: A cluster-randomized pilot trial with a 1:1 allocation ratio was conducted in rural Uttar Pradesh, India, from January to September 2018. Villages were randomly assigned to either the intervention or control group. In each participating village, surveyors conducted a complete enumeration to identify eligible households and requested participation before randomization. Interventions were designed through formative research using a social marketing approach and delivered over 3 months using strategies adapted to disadvantaged populations: (1) mobile health (mHealth): entertaining educational audio capsules (edutainment) and voice immunization reminders via mobile phone and (2) face-to-face: community mobilization activities, including 3 small group meetings offered to each participant. The control group received usual services. The main outcomes were prespecified criteria for feasibility of the main study (recruitment, randomization, retention, contamination, and adoption). Secondary endpoints tested equity of coverage and changes in intermediate outcomes. Statistical methods included descriptive statistics to assess feasibility, penalized logistic regression and ordered logistic regression to assess coverage, and generalized estimating equation models to assess changes in intermediate outcomes. Results: All villages consented to participate. Gaps in administrative data hampered recruitment; 14.0\% (79/565) of recorded households were nonresident. Only 1.4\% (8/565) of households did not consent. A total of 387 households (184 intervention and 203 control) with children aged 0 to 12 months in 26 villages (13 intervention and 13 control) were included and randomized. The end line survey occurred during the flood season; 17.6\% (68/387) of the households were absent. Contamination was less than 1\%. Participation in one or more interventions was 94.0\% (173/184), 78.3\% (144/184) for the face-to-face strategy, and 67.4\% (124/184) for the mHealth strategy. Determinants including place of residence, mobile phone access, education, and female empowerment shaped intervention use; factors operated differently for face-to-face and mHealth strategies. For 11 of 13 intermediate outcomes, regression results showed significantly higher basic health knowledge among the intervention group, supporting hypothesized causal mechanisms. Conclusions: A future trial of a new intervention model is feasible. The interventions could strengthen the delivery of immunization and universal primary health care. Social and behavior change communication via mobile phones proved viable and contributed to standardization and scalability. Face-to-face interactions remain necessary to achieve equity and reach, suggesting the need for ongoing health system strengthening to accompany the introduction of communication technologies. © Mira Johri, Dinesh Chandra, Karna Georges Kone, Marie-Pierre Sylvestre, Alok K Mathur, Sam Harper, Arijit Nandi.
Nandi A, Agarwal P, Chandrashekar A, Harper S. “Access to Affordable Daycare and Women’s Economic Opportunities: Evidence from a Cluster-Randomised Intervention in India.” Journal of Development Effectiveness. 2020;12(3):219–239
We used data from a cluster-randomized trial in rural Rajasthan, India to evaluate the impact of providing access to a community-based daycare program on women’s economic outcomes two years later. The sample included 2858 mothers with age-eligible children. Providing access to daycare led 43\% of households to utilize them. The intervention reduced time on childcare by 16.0 minutes/day (95\%CI = -10.6, 42.5) and increased the probabilities that women were paid in cash and spent time during the prior day on paid work by 2.3 (95\%CI = 0.0, 4.5) and 2.6 (95\%CI = 0.9, 4.4) percentage points. Other indicators of labor force participation and income were unaffected.
Nunes A, Harper S, Hernandez K. “The Price Isn’t Right: Autonomous Vehicles, Public Health, and Social Justice.” American Journal of Public Health. 2020;110(6):796–797
Richardson R, Nandi A, Jaswal S, Harper S. “The Effect of Intimate Partner Violence on Women’s Mental Distress: A Prospective Cohort Study of 3010 Rural Indian Women.” Social Psychiatry and Psychiatric Epidemiology. 2020;55(1):71–79
Purpose: Intimate partner violence (IPV) encompasses physical, sexual, and psychological abuse, as well as controlling behavior. Most research focuses on physical and sexual abuse, and other aspects of IPV are rarely investigated. We estimated the effect of these neglected aspects of IPV on women’s mental distress. Methods: We used data from 3010 women living in rural tribal communities in Rajasthan, India. Women completed baseline interviews and were re-interviewed approximately 1.5~years later. We measured IPV with questions adopted from the Demographic and Health Survey’s Domestic Violence Module, which asked seven questions about physical abuse, three questions about psychological abuse, and five questions about partner controlling behavior. Mental distress was measured with the 12-item General Health Questionnaire (score range 0–12). We used Poisson regression models to estimate the relation between changes in IPV and mental distress, accounting for time-fixed characteristics of individuals using individual fixed effects. Results: Women reported an average of 2.1 distress symptoms during baseline interviews. In models that controlled for time-varying confounding (e.g., wealth, other types of abuse), experiencing psychological abuse was associated with an increase of 0.65 distress symptoms (95\% CI 0.32, 0.98), and experiencing controlling behavior was associated with an increase of 0.31 distress symptoms (95\% CI 0.18, 0.44). However, experiencing physical abuse was not associated with an increase in distress symptoms (mean difference = −~0.15, 95\% CI −~0.45, 0.15). Conclusions: Psychological abuse and controlling behavior may be important drivers of the relation between IPV and women’s mental health. © 2019, Springer-Verlag GmbH Germany, part of Springer Nature.
Richardson R, Harper S, Weichenthal S, Nandi A, Mishra V, Jha P. “Extremes in Water Availability and Suicide: Evidence from a Nationally Representative Sample of Rural Indian Adults.” Environmental Research. 2020;190
Background: Extremes in water availability, either exceptionally wet or dry conditions, can damage crops and may detrimentally affect the livelihood and well-being of people engaged in agriculture. We estimated the effect of water availability on suicide in rural India, a context where the majority of households are dependent upon agriculture. Methods: We used data from a nationally representative sample of 8.5 million people who were monitored for causes of death from 2001 to 2013. Water availability was measured with high-resolution precipitation and temperature data (i.e., the Standardized Precipitation Evapotranspiration Index). We used a fixed effects approach that modeled changes in water availability within districts (n = 569) over time (n = 13 years) to estimate the impact on suicide deaths. We restricted our analysis to rural areas and to deaths occurring during the growing season (June–March) among adults aged 15 or older, and controlled for sex, age, region, and year. We used Poisson regression with standard errors clustered at the district level and total deaths as the offset. Results: There were 9456 suicides and 249,786 total deaths in our study population between 2001 and 2013. Compared to normal growing seasons, the percent of deaths due to suicide increased by 18.7\% during extremely wet growing seasons (95\% CI: 6.2, 31.2) and by 3.6\% during extremely dry growing seasons (95\% CI: −17.9, 25.0). We found that effects varied by age. Conclusions: We found extremes is water availability associated with an increase in suicide. Abnormally wet growing conditions may play an important, yet overlooked, role in suicide among rural Indian adults. © 2020 Elsevier Inc.
Riddell C, Kaufman J, Torres J, Harper S. “Using Change in a Seat Belt Law to Study Racially-Biased Policing in South Carolina.” Preventive Medicine. 2020;130
Seat belt laws have increased seat belt use in the US and contributed to reduced fatalities and injuries. However, these policies provide the potential for increased discrimination. The objective of this study is to determine whether a change in seat belt use enforcement led to a differential change in the number of stops, arrests, and searches to White, Black and Hispanic drivers in one US state. We used data on 1,091,424 traffic stops conducted by state troopers in South Carolina in 2005 and 2006 to examine how the change from secondary to primary enforcement of seat belt use in December 2005 affected the number of stops, arrests, and searches to White, Black, and Hispanic drivers using quasi-Poisson and logistic regressions. We found that the policy led to a 50\% increase in the number of non-speeding stops for White drivers, and that this increase was 5\% larger among Black drivers [RR (95\% CI) = 1.05 (1.00, 1.10)], but not larger among Hispanic drivers [1.00 (0.93, 1.08)]. The policy decreased arrests and searches among non-speeding stops, with larger decreases for Black vs. White drivers [RR searches = 0.86 (0.81, 0.91) and RR arrests = 0.90 (0.85, 0.96)]. For Hispanic drivers, effects of the policy change were also found among stops for speeding, which failed the falsification test and suggested that other changes likely affected this group. These findings may support the hypothesis of differential enforcement of seat belt policy in South Carolina for Black and White drivers. © 2019
Ahn J, Harper S, Yu M, Feuer E, Liu B. “Improved Monte Carlo Methods for Estimating Confidence Intervals for Eleven Commonly Used Health Disparity Measures.” PLoS ONE. 2019;14(7)
Health disparities are commonplace and of broad interest to policy makers, but are also challenging to measure and communicate. The Health Disparity Calculator software (HD*Calc, v1.2.4) offers Monte Carlo simulation (MCS)-based confidence interval (CI) estimation of eleven disparity measures. The MCS approach provides accurate CI estimation, except when data are scarce (e.g., rare cancers). To address sparse data challenges to CI estimation, we propose two solutions: 1) employing the gamma distribution in the MCS and 2) utilizing a zero-inflated Poisson estimate for Poisson sampling in simulation experiments. We evaluate each solution through simulation studies using female breast, female brain, lung, and cervical cancer data from the Surveillance, Epidemiology, and End Results (SEER) program. We compare the coverage probabilities (CPs) of eleven health disparity measures based on simulated datasets. The truncated normal distribution implemented in the MCS with the standard Poisson samples (the default setting of HD∗Calc) leads to lessthan- optimal coverage probabilities ({$<$}95\%). When both the gamma distribution and the estimated mean from the zero-inflated Poisson are used for the MCS, the coverage probabilities are close to the nominal level of 95\%. Simulation studies also demonstrate that collapsing age categories for better CI estimation is not a pragmatic solution. This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Ahrens K, Hutcheon J, Ananth C, Basso O, Briss P, Ferré C, Frederiksen B, Harper S, Hernández-Díaz S, Hirai A, Kirby R, Klebanoff M, Lindberg L, Mumford S, Nelson H, Platt R, Rossen L, Stuebe A, Thoma M, Vladutiu C, Moskosky S. “Report of the Office of Population Affairs’ Expert Work Group Meeting on Short Birth Spacing and Adverse Pregnancy Outcomes: Methodological Quality of Existing Studies and Future Directions for Research.” Paediatric and Perinatal Epidemiology. 2019;33(1):O5-O14
Background: The World Health Organization (WHO) recommends that women wait at least 24 months after a livebirth before attempting a subsequent pregnancy to reduce the risk of adverse maternal, perinatal, and infant health outcomes. However, the applicability of the WHO recommendations for women in the United States is unclear, as breast feeding, nutrition, maternal age at first birth, and total fertility rate differs substantially between the United States and the low- and middle-resource countries upon which most of the evidence is based. Methods: To inform guideline development for birth spacing specific to women in the United States, the Office of Population Affairs (OPA) convened an expert work group meeting in Washington, DC, on 14-15 September 2017 among reproductive, perinatal, paediatric, social, and public health epidemiologists; obstetrician-gynaecologists; biostatisticians; and experts in evidence synthesis related to women's health. Results: Presentations and discussion topics included the methodological quality of existing studies, evaluation of the evidence for causal effects of short interpregnancy intervals on adverse perinatal and maternal health outcomes, good practices for future research, and identification of research gaps and priorities for future work. Conclusions: This report provides an overview of the presentations, discussions, and conclusions from the expert work group meeting. © 2018 The Authors. Paediatric and Perinatal Epidemiology Published by John Wiley \& Sons Ltd
Austin N, Harper S. “Quantifying the Impact of Targeted Regulation of Abortion Provider Laws on US Abortion Rates: A Multi-State Assessment.” Contraception. 2019;100(5):374–379
Objectives: In this paper, we estimate the average effect of two common TRAP (targeted regulation of abortion providers) laws on abortion rates using a novel longitudinal database of state-level policy shifts. Study design: We merged several sources of policy, abortion, and sociodemographic data from 1991–2014. We used a difference-in-differences design to control for time-fixed state-level characteristics and common factors affecting abortion trends across all states, as well as measured time-varying state-level factors that may impact TRAP enforcement and abortion rates. We used generalized linear models with cluster-robust standard errors to obtain our estimates. Results: Enforcement of ambulatory surgical center (ASC) laws reduced the abortion rate by 1.25 abortions per 1000 women aged 15–44 (95\% CI: −3.39,.89), and admitting privilege laws increased the abortion rate by.57 abortions per 1000 women aged 15–44 (95\% CI: −.68, 1.83), but neither effect was statistically distinguishable from zero. Our findings were robust to the inclusion of covariates and various sensitivity analyses. Conclusion: Our results suggest that ASC and admitting privilege laws did not, on average, lead to a meaningful change in abortion rates. Implications: US abortion rates are currently at record lows, but our findings suggest that TRAP laws are not a meaningful driver of this trend. However, this does not mean that these laws are without consequence in a particular state (or a given year). Researchers should assess the average long-run impact of TRAP laws on other outcomes in the future. © 2019 Elsevier Inc.
Austin N, Harper S. “Constructing a Longitudinal Database of Targeted Regulation of Abortion Providers Laws.” Health Services Research. 2019;54(5):1084–1089
Objective: To build a longitudinal state-level database on targeted regulation of abortion providers (TRAP) laws. Data Sources: Primary sources included state websites, Lexis Nexis Quicklaw, and WestlawNext. We used a range of secondary sources to pinpoint policy timing. Study Design: This was a state-level review of TRAP shifts from 1973 to present. Data Collection: We captured data on TRAP policy activity and timing, focusing specifically on ambulatory surgical center (ASC) laws, admitting privilege requirements, and transfer agreements. Principal Findings: Twenty-five states had ever enacted an ASC, admitting privilege, or transfer agreement law. Many currently face legal challenges. Conclusions: Targeted regulation of abortion providers laws are favored by many states as a way to regulate abortion provision. These data can be used to better understand the impact of these laws. © Health Research and Educational Trust
Hamadani F, Razek T, Massinga E, Gupta S, Muataco M, Muripiha P, Maguni C, Muripa V, Percina I, Costa A, Yohannan P, Bracco D, Wong E, Harper S, Deckelbaum D, Neves O. “Trauma Surveillance and Registry Development in Mozambique: Results of a 1-Year Study and the First Phase of National Implementation.” World Journal of Surgery. 2019;43(7):1628–1635
Background: Mozambique has had no policy-driven trauma system and no hospital-based trauma registries, and injury was not a public health priority. In other low-income countries, trauma system implementation and trauma registries have helped to reduce mortality from injury by up to 35\%. In 2014, we introduced a trauma registry in four hospitals in Maputo serving 18,000 patients yearly. The project has since expanded nationally. This study summarizes the challenges, results, and lessons learned from this large national undertaking. Methods: Between October 2014–September 2015, we implemented a trauma registry at four hospitals in Maputo. In October 2015, the project began to be expanded nationally. Physicians and allied health professionals at each hospital were trained to implement the registry, and each identified and trained data collectors. We conducted semi-structured interviews with the key stakeholders of this project to identify the challenges, results, and creative solutions implemented for the success of this project. Results: Most participants identified the importance of having a trauma registry and its usefulness in identifying gaps in trauma care. The registry identified that less than 5\% of injured patients arrived by ambulance, which served as evidence for the need for a prehospital system, which the Ministry of Health had already begun implementing. Participants also highlighted how the registry has allowed for a structured clinical approach to patients, ensuring that severely injured patients are identified early. Challenges reported included the high rates of missing data, the difficulty in establishing a streamlined flow of trauma patients within each hospital, and the bureaucratic challenges faced when attempting to improve capacity for trauma care at each hospital by introducing a trauma bay and new technologies. Participants identified the need to improve data completeness, to disseminate the results of the project nationally and internationally, to improve inter-divisional cooperation, and to continue educating health providers on the importance of registries. Participants also identified political instabilities in the region as a potential source of challenge in expanding the project nationally; they also identified the lack of uniform resource allocation and low personnel in many areas, especially rural, as a major burden that would need to be overcome. Conclusion: Introduction of a trauma registry system in Mozambique is feasible and necessary. Initial findings provide insight into the nature of traumas seen in Maputo hospitals, but also underscore future challenges, especially in minimizing missing data, utilizing data to develop evidence-based trauma prevention policies, and ensuring the sustainability of these efforts by ensuring continued governmental support, education, and resource allocation. Many of these measures are being undertaken. © 2019, Société Internationale de Chirurgie.
Hamra G, Goldstein N, Harper S. “Resource Sharing to Improve Research Quality.” Journal of the American Heart Association. 2019;8(15)
Harper S. “A Future for Observational Epidemiology: Clarity, Credibility, Transparency.” American journal of epidemiology. 2019;188(5):840–845
Observational studies are ambiguous, difficult, and necessary for epidemiology. Presently, there are concerns that the evidence produced by most observational studies in epidemiology is not credible and contributes to research waste. I argue that observational epidemiology could be improved by focusing greater attention on 1) defining questions that make clear whether the inferential goal is descriptive or causal; 2) greater utilization of quantitative bias analysis and alternative research designs that aim to decrease the strength of assumptions needed to estimate causal effects; and 3) promoting, experimenting with, and perhaps institutionalizing both reproducible research standards and replication studies to evaluate the fragility of study findings in epidemiology. Greater clarity, credibility, and transparency in observational epidemiology will help to provide reliable evidence that can serve as a basis for making decisions about clinical or population-health interventions.
Harper S, Palayew A. “The Annual Cannabis Holiday and Fatal Traffic Crashes.” Injury Prevention. 2019;25(5):433–437
Cannabis use has been linked to impaired driving and fatal accidents. Prior evidence suggests the potential for population-wide effects of the annual cannabis celebration on April 20th (a € 4/20'), but evidence to date is limited. Methods We used data from the Fatal Analysis Reporting System for the years 1975-2016 to estimate the impact of a € 4/20' on drivers involved in fatal traffic crashes occurring between 16:20 and 23:59 hours in the USA. We compared the effects of 4/20 with those for other major holidays, and evaluated whether the impact of a € 4/20' had changed in recent years. Results Between 1992 and 2016, a € 4/20' was associated with an increase in the number of drivers involved in fatal crashes (IRR 1.12, 95\% CI 0.97 to 1.28) relative to control days 1 week before and after, but not when compared with control days 1 and 2 weeks before and after (IRR 1.05, 95\% CI 0.92 to 1.28) or all other days of the year (IRR 0.98, 95\% CI 0.88 to 1.10). Across all years we found little evidence to distinguish excess drivers involved in fatal crashes on 4/20 from routine daily variations. Conclusions There is little evidence to suggest population-wide effects of the annual cannabis holiday on the number of drivers involved in fatal traffic crashes. © Author(s) (or their employer(s)) 2019. No commercial re-use. See rights and permissions. Published by BMJ.
Harper S. “Comment on the Equity Impact of Women’s Community Groups on Inequalities in Neonatal Mortality.” International Journal of Epidemiology. 2019;48(1):182–185
Harper S. “Would Stronger Seat Belt Laws Reduce Motor Vehicle Crash Deaths?: A Semi-Bayesian Analysis.” Epidemiology (Cambridge, Mass.). 2019;30(3):380–387
BACKGROUND: For policy questions where substantial empirical background information exists, conventional frequentist policy analysis is hard to justify. Bayesian analysis quantitatively incorporates prior knowledge, but is not often used in applied policy analysis. METHODS: We combined 2000-2016 data from the Fatal Analysis Reporting System with priors based on past empirical studies and policy documents to study the impact of mandatory seat belt laws on traffic fatalities. We used a Bayesian data augmentation approach to combine information from prior studies with difference-in-differences analyses of recent law changes to provide updated evidence on the impact that upgrading to primary enforcement of seat belt laws has on fatalities. RESULTS: After incorporating the evidence from past studies, we find limited evidence to support the hypothesis that recent policy upgrades affect fatality rates. We estimate that upgrading to primary enforcement reduced fatality rates by 0.37 deaths per billion vehicle miles traveled (95\% posterior interval -0.90, 0.16), or a rate ratio of 0.96 (95\% posterior interval 0.91, 1.02), and increased the proportion of decedents reported as wearing seat belts by 7 percentage points (95\% posterior interval 5, 8), or a risk ratio of 1.18 (95\% posterior interval 1.13, 1.24). CONCLUSIONS: Bayesian methods can provide credible estimates of future policy impacts, especially for policy questions that occur in dynamic environments, such as traffic safety.
Hutcheon J, Moskosky S, Ananth C, Basso O, Briss P, Ferré C, Frederiksen B, Harper S, Hernández-Díaz S, Hirai A, Kirby R, Klebanoff M, Lindberg L, Mumford S, Nelson H, Platt R, Rossen L, Stuebe A, Thoma M, Vladutiu C, Ahrens K. “Good Practices for the Design, Analysis, and Interpretation of Observational Studies on Birth Spacing and Perinatal Health Outcomes.” Paediatric and Perinatal Epidemiology. 2019;33(1):O15-O24
Background: Meta-analyses of observational studies have shown that women with a shorter interpregnancy interval (the time from delivery to start of a subsequent pregnancy) are more likely to experience adverse pregnancy outcomes, such as preterm delivery or small for gestational age birth, than women who space their births further apart. However, the studies used to inform these estimates have methodological shortcomings. Methods: In this commentary, we summarise the discussions of an expert workgroup describing good practices for the design, analysis, and interpretation of observational studies of interpregnancy interval and adverse perinatal health outcomes. Results: We argue that inferences drawn from research in this field will be improved by careful attention to elements such as: (a) refining the research question to clarify whether the goal is to estimate a causal effect vs describe patterns of association; (b) using directed acyclic graphs to represent potential causal networks and guide the analytic plan of studies seeking to estimate causal effects; (c) assessing how miscarriages and pregnancy terminations may have influenced interpregnancy interval classifications; (d) specifying how key factors such as previous pregnancy loss, pregnancy intention, and maternal socio-economic position will be considered; and (e) examining if the association between interpregnancy interval and perinatal outcome differs by factors such as maternal age. Conclusion: This commentary outlines the discussions of this recent expert workgroup, and describes several suggested principles for study design and analysis that could mitigate many potential sources of bias. © 2018 The Authors. Paediatric and Perinatal Epidemiology Published by John Wiley \& Sons Ltd.
Hutcheon J, Harper S. “Invited Commentary: Promise and Pitfalls of the Sibling Comparison Design in Studies of Optimal Birth Spacing.” American Journal of Epidemiology. 2019;188(1):17–21
Numerous observational studies have shown that infants born after short interpregnancy intervals (the interval between birth and subsequent conception) are more likely to experience adverse perinatal outcomes than infants born following longer intervals. Yet it remains controversial whether the link between short interpregnancy interval and adverse outcomes is causal or is confounded by factors such as low socioeconomic position, inadequate access to health care, and unintended pregnancy. Sibling comparison studies, which use a woman as her own control by comparing exposure and outcome status of her different pregnancies (i.e., comparing sibling offspring), have gained popularity as a strategy to reduce confounding by these difficult-to-measure factors that are nevertheless relatively stable within women. A variant of this approach, used by Regan et al. (Am J Epidemiol. 2019;188(1):9-16) and reported in this issue of the Journal, is a maternally matched design based on a single interpregnancy interval per woman. Using real and simulated data, we highlight underappreciated shortcomings of these designs that could limit the validity of study findings. In particular, we illustrate how the single-interval variant appears to derive estimates from comparisons between different mothers, not within mothers. Future studies of optimal birth spacing using sibling comparison designs should examine in detail the potential consequences of these methodological limitations. © The Author(s) 2018. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Kaufman J, Riddell C, Harper S. “Black and White Differences in Life Expectancy in 4 US States, 1969-2013.” Public Health Reports. 2019;134(6):634–642
Objectives: Racial differences in mortality in the United States have narrowed and vary by time and place. The objectives of our study were to (1) examine the gap in life expectancy between white and black persons (hereinafter, racial gap in life expectancy) in 4 states (California, Georgia, Illinois, and New York) and (2) estimate trends in the contribution of major causes of death (CODs) to the racial gap in life expectancy by age group. Methods: We extracted data on the number of deaths and population sizes for 1969-2013 by state, sex, race, age group, and 6 major CODs. We used a Bayesian time-series model to smooth and impute mortality rates and decomposition methods to estimate trends in sex- and age-specific contributions of CODs to the racial gap in life expectancy. Results: The racial gap in life expectancy at birth decreased in all 4 states, especially among men in New York (from 8.8 to 1.1 years) and women in Georgia (from 8.0 to 1.7 years). Although few deaths occurred among persons aged 1-39, racial differences in mortality at these ages (mostly from injuries and infant mortality) contributed to the racial gap in life expectancy, especially among men in California (1.0 year of the 4.3-year difference in 2013) and Illinois (1.9 years of the 6.7-year difference in 2013). Cardiovascular deaths contributed most to the racial gap in life expectancy for adults aged 40-64, but contributions decreased among women aged 40-64, especially in Georgia (from 2.8 to 0.5 years). The contribution of cancer deaths to inequality increased in California and Illinois, whereas New York had the greatest reductions in inequality attributable to cancer deaths (from 0.6 to 0.2 years among men and from 0.2 to 0 years among women). Conclusions: Future research should identify policy innovations and economic changes at the state level to better understand New York’s success, which may help other states emulate its performance. © 2019, Association of Schools and Programs of Public Health.
Richardson R, Schmitz N, Harper S, Nandi A. “Development of a Tool to Measure Women’s Agency in India.” Journal of Human Development and Capabilities. 2019;20(1):26–53
Ensuring and expanding women’s agency is an essential component of efforts to promote the rights and well-being of women. However, inadequate measurement hampers monitoring and research into achieving this goal. In this study, we developed a theory-based measurement tool of women’s agency. We developed a conceptual model of agency through a review of the literature, and then used this model to identify potential indicators of agency. These indicators were asked as part of a population-based household survey that was completed between July and November 2016 by 3042 women in rural Rajasthan, India. We tested the construct validity of the hypothesized measurement model using confirmatory factor analysis. We identified a conceptual model of agency, composed of 23 indicators, which measured the domains Household Decision-Making, Freedom of Movement, Participation in the Community, and Attitudes and Perceptions. This conceptual model fit the study data well (CFI = 0.974, TLI = 0.970, RMSEA = 0.031). Our results have implications for measurement efforts in a number of settings, and our tool can be used to measure women’s agency in rural India. © 2018, © 2018 Human Development and Capability Association.
Richardson R, Harper S, Bates L, Nandi A. “The Effect of Agency on Women’s Mental Distress: A Prospective Cohort Study from Rural Rajasthan, India.” Social Science and Medicine. 2019;233:47–56
Agency, the ability to identify goals and then act upon them, is a core component of women's empowerment and has important implications for the rights and well-being of women and girls. However, inadequate measurement of agency impedes empirical investigation, and few studies have investigated the relation between agency and health. Using a theory-based measure of women's agency, we investigated the longitudinal association between agency and mental distress among women living in rural Rajasthan, India. Women completed baseline interviews between June and October 2016 and follow-up interviews between June and November 2017 (n = 2859). We measured mental distress with the Hindi version of the 12 item General Health Questionnaire, which asked women 12 questions about symptoms of mental distress (score range: 0–12). We measured agency using a measurement model which was composed of 23 indicators tapping into four domains of agency and validated in a prior research study. We modeled the relation between women's agency and mental distress using Poisson regression and an individual-level fixed effects approach to account for time-fixed characteristics of individuals. In models that controlled for time-varying confounding (e.g., household wealth, number of sons), a one standard deviation increase in agency was associated with a reduction of 0.21 distress symptoms (95\% CI: -0.32, −0.09), which corresponds to a 7\% reduction (95\% CI: 3\%, 11\%) relative to the mean. We found that specific domains of agency varied in their association with mental distress; namely, an increase in women's agency regarding her attitudes about gender norms corresponded to a reduction in mental distress, whereas an increase in women's agency regarding speaking up in public corresponded to an increase in mental distress. Our research demonstrates that agency may be a determinant of mental health and that comprehensive measurement can reveal nuanced relationships. © 2019 Elsevier Ltd
Sreeramareddy C, Harper S. “Trends in Educational and Wealth Inequalities in Adult Tobacco Use in Nepal 2001-2016: Secondary Data Analyses of Four Demographic and Health Surveys.” BMJ Open. 2019;9(9)
Objective To measure trends in socioeconomic inequalities tobacco use in Nepal. Setting Adults interviewed during house-to-house surveys. Participants Women (15-45 years) and men (15-49 years) surveyed in four Nepal Demographic and Health Surveys done in 2001, 2006, 2011 and 2016. Outcome measure Current tobacco use (in any form). Results The prevalence of tobacco use for men declined from 66\% in 2001 to 55\% in 2016, and declined from 29\% to 8.4\% among women. Across both education and wealth quintiles for both men and women, the prevalence of tobacco use generally declines with increasing education or wealth. We found persistently larger absolute inequalities by education than by wealth among men. Among women we also found larger educational than wealth-related gradients, but both declined over time. For men, the Slope Index of Inequality (SII) for education was larger than for wealth (44\% vs 26\% in 2001) and changed very little over time. For women, the SII for both education and wealth were similar in magnitude to men, but decreased substantially between 2001 and 2016 (from 44\% to 16\% for education; from 37\% to 16\% for wealth). Women had a larger relative index of inequality than men for both education (6.5 vs 2.0 in 2001) and wealth (4.8 vs 1.5 in 2001), and relative inequality increased between 2001 and 2016 for women (from 6.5 to 16.0 for education; from 4.8 to 12.0 for wealth). Conclusion Increasing relative inequalities indicates suboptimal reduction in tobacco use among the vulnerable groups suggesting that they should be targeted to improve tobacco control. © 2019 Author(s).
Ahn J, Harper S, Yu M, Feuer E, Liu B, Luta G. “Variance Estimation and Confidence Intervals for 11 Commonly Used Health Disparity Measures.” JCO Clinical Cancer Informatics. 2018;2018(2):1–19
There is increased interest in eliminating health disparities in the United States and worldwide. Broadly defined, health disparities refer to preventable inequalities in health status, such as cancer to ethnicity, socioeconomic status, gender, education, environment, and geographic locations. To make informed health policy decisions, it is essential to precisely measure the magnitude of disparities and assess trends over time. The Health Disparities Calculator (HDz.ast;Calc) is free statistical software that calculates 11 commonly used measures of health disparities and provides corresponding 95\% CIs for the 11 measures using either an analytic method or a Monte Carlo simulation-based method; however, the derivation of SEs and coverage properties of the CIs have not been formally evaluated. We used simulation studies to assess the coverage properties of these CIs. We have also conducted bias analyses for measures implemented in HDz.ast;Calc using age-adjusted cancer incidence rates from national, state, and county level SEER data. The results of these analyses indicate that HDz.ast;Calc should be used with caution to construct CIs for some health disparity measures when the proportion of zero event counts is greater than 25\%. © 2018 American Society of Clinical Oncology.
Austin N, Harper S. “Assessing the Impact of TRAP Laws on Abortion and Women’s Health in the USA: A Systematic Review.” BMJ Sexual and Reproductive Health. 2018;44(2):128–134
Introduction Targeted Regulation of Abortion Providers (TRAP) laws impose extensive and sometimes costly requirements on abortion providers and facilities, potentially leading to barriers to care. Understanding the impact of these laws is important given their prevalence in the USA, but no review to date has summarised the available evidence. We conducted a systematic review of literature on TRAP laws and their impact on abortion trends and women's health. Methods We searched MEDLINE, PubMed and EconLit for original, quantitative studies where the exposure was at least one TRAP policy and the outcome was abortion and/or any women's physical or mental health outcome. Results Six articles met our inclusion criteria. The most common outcome was population-level abortion trends; studies also assessed the effect of TRAP laws on gestational age at presentation and measures of self-perceived burden. While certain TRAP laws (eg, admitting privilege requirements) appeared to have an effect on abortion outcomes, the impact of other laws-or combinations of laws-was unclear, due in part to heterogeneity between studies with respect to study design, geography, and exposure definition. Conclusions TRAP laws may have an impact on the experience of obtaining an abortion in the USA. However, our review revealed a paucity of empirical research on their population and individual-level impact, as well as some disagreement about the effect of different TRAP laws on subsequent abortion outcomes. Future research should prioritise the specific TRAP laws that may have a uniquely strong effect on state-level abortion rates and other outcomes. © 2018 Article author(s) (or their employer(s) unless otherwise stated in the text of the article). All rights reserved.
Banack HR, Harper S, Kaufman JS. “Accounting for Selection Bias in Studies of Acute Cardiac Events.” Canadian Journal of Cardiology. 2018;34(6):709–716
Background: In cardiovascular research, pre-hospital mortality represents an important potential source of selection bias. Inverse probability of censoring weights are a method to account for this source of bias. The objective of this article is to examine and correct for the influence of selection bias due to pre-hospital mortality on the relationship between cardiovascular risk factors and all-cause mortality after an acute cardiac event.
Mah S, Sanmartin C, Harper S, Ross N. “Childbirth-Related Hospital Burden by Socioeconomic Status in a Universal Health Care Setting.” International Journal of Population Data Science. 2018;3(1)
Introduction Hospital utilization varies across socioeconomic and demographic strata in Canada, a country with a universal health care system. Rates of adverse birth outcomes are known to differ among women of high and low socioeconomic status (SES), but less is known of the excess hospital burden related to SES over the course of childbirth across Canadian provinces. Objective To examine length of stay and risk of hospitalization surrounding delivery, relative to women’s sociodemographic characteristics. Methods A population-based record linkage between the Canadian Community Health Survey (CCHS) years 2005-2011 and the Discharge Abstract Database (DAD) allowed the tracking of hospital utilization for linked survey respondents between 2005 and 2011. Hourly length of stay for delivery, risk of readmission, and risk of admission prior to delivery was modeled by socio-demographic factors, controlling for other clinical and individual-level characteristics. Results There were 21,914 complete delivery records from 15,458 female CCHS respondents who agreed to link and share their information. Average length of stay (for both vaginal and Caesarian deliveries) dropped over the study period from 67.86 hours in 2005 to 59.37 hours in 2011. In multivariate analyses, women with the lowest income had on average, two-hour longer stays for vaginal delivery as compared to high-income women (IRR 1.04, 95\% CI 1.00-1.08) and higher risk of admission prior to delivery (OR 1.43, CI 1.13-1.81). Low-income women, Aboriginal women and women living in rural areas were also at elevated risk for longer hospital stays and for hospital admission prior to delivery. There was no consistent socioeconomic patterning of hospital burden for Caesarian deliveries. Conclusion The length of hospital stays for childbirth has declined in Canada. Length of stay remains modestly longer, and risk of hospitalization in the perinatal period higher, for low income women, Aboriginal women and rural women. The absence of egregious income-related differences in hospital burden related to childbirth is reassuring for the equity goals of the Canadian health care system. The persistence of marginally longer, and in turn, costlier visits for low-income and Aboriginal women before and during delivery is, however, suggestive that resources targeted to the prenatal period might be highly cost-effective if they achieve population-wide reductions in length of stay and hospitalization in the perinatal period. © The Authors. Open Access under CC BY-NC-ND 4.0 (https://creativecommons.org/licenses/by-nc-nd/4.0/deed.en)
McLinden T, Moodie E, Harper S, Hamelin A, Anema A, Aibibula W, Klein M, Cox J. “Injection Drug Use, Food Insecurity, and HIV-HCV Co-Infection: A Longitudinal Cohort Analysis.” AIDS Care - Psychological and Socio-Medical Aspects of AIDS/HIV. 2018;30(10):1322–1328
Injection drug use (IDU) and food insecurity (FI) are highly prevalent among individuals living with HIV-hepatitis C virus (HCV) co-infection. We quantified the association between IDU and FI among co-infected individuals using biannual data from the Canadian Co-infection Cohort (N = 608, 2012–2015). IDU (in the past six months) and IDU frequency (non-weekly/weekly in the past month) were self-reported. FI (in the past six months) and FI severity (marginal FI, moderate FI, and severe FI) were measured using the Household Food Security Survey Module. Generalized estimating equations were used to estimate risk ratios (RR) quantifying the associations between IDU, IDU frequency, and FI with Poisson regression. The associations between IDU, IDU frequency, and FI severity were quantified by relative-risk ratios (RRR) estimated with multinomial regression. At the first time-point in the analytical sample, 54\% of participants experienced FI in the past six months, 31\% engaged in IDU in the six months preceding the FI measure, and 24\% injected drugs in the past month. After adjustment for confounding, IDU in the past six months (RR = 1.15, 95\% confidence interval [CI] = 1.04–1.28) as well as non-weekly (RR = 1.15, 95\% CI = 1.02–1.29) and weekly IDU (RR = 1.21, 95\% CI = 1.07–1.37) in the past month are associated with FI. Weekly IDU in the past month is also strongly associated with severe FI (RRR = 2.68, 95\% CI = 1.47–4.91). Our findings indicate that there is an association between IDU and FI, particularly weekly IDU and severe FI. This suggests that reductions in IDU may mitigate FI, especially severe FI, in this vulnerable subset of the HIV-positive population. © 2018, © 2018 Informa UK Limited, trading as Taylor \& Francis Group.
McLinden T, Moodie E, Hamelin A, Harper S, Rossi C, Walmsley S, Rourke S, Cooper C, Klein M, Cox J. “Methadone Treatment, Severe Food Insecurity, and HIV-HCV Co-Infection: A Propensity Score Matching Analysis.” Drug and Alcohol Dependence. 2018;185:374–380
Background: Severe food insecurity (FI) is common among individuals living with HIV-hepatitis C virus (HCV) co-infection. We hypothesize that the injection of opioids is partly responsible for the association between injection drug use and severe FI. Therefore, this analysis examines whether methadone maintenance treatment for opioid dependence is associated with a lower risk of severe FI. Methods: We used biannual data from the Canadian Co-infection Cohort (N = 608, 2012–2015). Methadone treatment (exposure) was self-reported and severe FI (outcome) was measured using the Household Food Security Survey Module. To quantify the association between methadone treatment and severe FI, we estimated an average treatment effect on the treated (marginal risk difference [RD]) using propensity score matching. Results: Among participants, 25\% experienced severe FI in the six months preceding the first time-point in the analytical sample and 5\% concurrently reported receiving methadone treatment. Injection of opioids in the six months preceding the treatment and outcome measurements was much higher among those who received methadone treatment (39\% vs. 12\%). Among the treated participants, 97\% had injected opioids in their lifetimes. After propensity score matching, the average risk of experiencing severe FI was 12.3 percentage-points lower among those receiving methadone treatment, compared to those who were not receiving treatment (marginal RD = −0.123, 95\% CI = −0.230, −0.015). Conclusions: After adjustment for socioeconomic, sociodemographic, behavioural, and clinical confounders, methadone treatment was associated with a lower risk of severe FI. This finding suggests that methadone treatment may mitigate severe FI in this vulnerable subset of the HIV-positive population. © 2018 Elsevier B.V.
Mejia G, Elani H, Harper S, Murray Thomson W, Ju X, Kawachi I, Kaufman J, Jamieson L. “Socioeconomic Status, Oral Health and Dental Disease in Australia, Canada, New Zealand and the United States.” BMC Oral Health. 2018;18(1)
Background: Socioeconomic inequalities are associated with oral health status, either subjectively (self-rated oral health) or objectively (clinically-diagnosed dental diseases). The aim of this study is to compare the magnitude of socioeconomic inequality in oral health and dental disease among adults in Australia, Canada, New Zealand and the United States (US). Methods: Nationally-representative survey examination data were used to calculate adjusted absolute differences (AD) in prevalence of untreated decay and fair/poor self-rated oral health (SROH) in income and education. We pooled age- and gender-adjusted inequality estimates using random effects meta-analysis. Results: New Zealand demonstrated the highest adjusted estimate for untreated decay; the US showed the highest adjusted prevalence of fair/poor SROH. The meta-analysis showed little heterogeneity across countries for the prevalence of decayed teeth; the pooled ADs were 19.7 (95\% CI = 16.7-22.7) and 12.0 (95\% CI = 8.4-15.7) between highest and lowest education and income groups, respectively. There was heterogeneity in the mean number of decayed teeth and in fair/poor SROH. New Zealand had the widest inequality in decay (education AD = 0.8; 95\% CI = 0.4-1.2; income AD = 1.0; 95\% CI = 0.5-1.5) and the US the widest inequality in fair/poor SROH (education AD = 40.4; 95\% CI = 35.2-45.5; income AD = 20.5; 95\% CI = 13.0-27.9). Conclusions: The differences in estimates, and variation in the magnitude of inequality, suggest the need for further examining socio-cultural and contextual determinants of oral health and dental disease in both the included and other countries. © 2018 The Author(s).
Nandi A, Jahagirdar D, Dimitris M, Labrecque J, Strumpf E, Kaufman J, Vincent I, Atabay E, Harper S, Earle A, Heymann S. “The Impact of Parental and Medical Leave Policies on Socioeconomic and Health Outcomes in OECD Countries: A Systematic Review of the Empirical Literature.” Milbank Quarterly. 2018;96(3):434–471
Policy Points: Historically, reforms that have increased the duration of job-protected paid parental leave have improved women's economic outcomes. By targeting the period around childbirth, access to paid parental leave also appears to reduce rates of infant mortality, with breastfeeding representing one potential mechanism. The provision of more generous paid leave entitlements in countries that offer unpaid or short durations of paid leave could help families strike a balance between the competing demands of earning income and attending to personal and family well-being. Context: Policies legislating paid leave from work for new parents, and to attend to individual and family illness, are common across Organisation for Economic Co-operation and Development (OECD) countries. However, there exists no comprehensive review of their potential impacts on economic, social, and health outcomes. Methods: We conducted a systematic review of the peer-reviewed literature on paid leave and socioeconomic and health outcomes. We reviewed 5,538 abstracts and selected 85 published papers on the impact of parental leave policies, 22 papers on the impact of medical leave policies, and 2 papers that evaluated both types of policies. We synthesized the main findings through a narrative description; a meta-analysis was precluded by heterogeneity in policy attributes, policy changes, outcomes, and study designs. Findings: We were able to draw several conclusions about the impact of parental leave policies. First, extensions in the duration of paid parental leave to between 6 and 12 months were accompanied by attendant increases in leave-taking and longer durations of leave. Second, there was little evidence that extending the duration of paid leave had negative employment or economic consequences. Third, unpaid leave does not appear to confer the same benefits as paid leave. Fourth, from a population health perspective, increases in paid parental leave were consistently associated with better infant and child health, particularly in terms of lower mortality rates. Fifth, paid paternal leave policies of adequate length and generosity have induced fathers to take additional time off from work following the birth of a child. How medical leave policies for personal or family illness influence health has not been widely studied. Conclusions: There is substantial quasi-experimental evidence to support expansions in the duration of job-protected paid parental leave as an instrument for supporting women's labor force participation, safeguarding women's incomes and earnings, and improving child survival. This has implications, in particular, for countries that offer shorter durations of job-protected paid leave or lack a national paid leave entitlement altogether. © 2018 Milbank Memorial Fund
Richardson R, Harper S, Schmitz N, Nandi A. “The Effect of Affordable Daycare on Women’s Mental Health: Evidence from a Cluster Randomized Trial in Rural India.” Social Science and Medicine. 2018;217:32–41
Access to affordable daycare might improve population mental health. However, evidence is sparse and restricted to middle- and high-income country settings. We conducted a cluster-randomized controlled trial in one low-income setting, rural Rajasthan, India. Communities lacking daycare facilities were identified (n = 160) and randomly selected for assistance in setting up a community-based daycare program (n = 80) or not (n = 80). Women eligible for the daycare program living in these communities completed structured interviews before the intervention (participation rate = 89\%) and approximately one year after rollout of the intervention (participation rate = 96\%), resulting in a final analytic sample of 3041. Mental distress was measured with the Hindi version of the 12-item General Health Questionnaire (score range: 0–12). We modeled the relation between access to daycare and number of mental distress symptoms (GHQ-12 score) with negative binomial regression using an intention-to-treat approach, which groups women according to if they lived in communities randomized to affordable daycare. We also evaluated the effect of access to daycare on secondary outcomes that may be related to mental distress, including women's work burden, agency, and intimate partner violence (IPV). We found that access to daycare resulted in modest reductions in symptoms of mental distress (mean difference = 0.21, 95\% CI: −0.43, 0.02). We found some evidence that daycare reduced IPV, but virtually no change in women's work burden or agency. Our results provide some indication that access to affordable daycare might be one policy lever to improve population mental health. © 2018 Elsevier Ltd
Riddell CA*, Morrison KT, Kaufman JS, Harper S. “Trends in the Contribution of Major Causes of Death to the Black-White Life Expectancy Gap by US State.” Health & Place. 2018;52:85–100
Life expectancy has increased in the United States over many decades. The difference in life expectancy between black and white Americans has also decreased, but some states have made much more progress towards racial equality than others. This paper describes the pattern of contributions of six major causes of death to the black-white life expectancy gap within US states and the District of Columbia between 1969 and 2013, and identifies states diverging from the overall pattern. Across multiple causes, the District of Columbia, Illinois, Wisconsin, and Michigan had the highest contributions to black-white inequality, while New York, Massachusetts, and Rhode Island had the lowest contributions and have either achieved or are the closest to achieving black-white equality in life expectancy.
Riddell C, Harper S, Cerda M, Kaufman J. “Comparison of Rates of Firearm and Nonfirearm Homicide and Suicide in Black and White Non-Hispanic Men, by u.S. State.” Annals of Internal Medicine. 2018;168(10):712–720
Background: The extent to which differences in homicide and suicide rates in black versus white men vary by U.S. state is unknown. Objective: To compare the rates of firearm and nonfirearm homicide and suicide in black and white non-Hispanic men by U.S. state and to examine whether these deaths are associated with state prevalence of gun ownership. Design: Surveillance study. Setting: 50 states and the District of Columbia, 2008 to 2016. Cause-of-death data were abstracted by using the Centers for Disease Control and Prevention's WONDER (Wide-ranging Online Data for Epidemiologic Research) database. Participants: Non-Hispanic black and non-Hispanic white males, all ages. Measurements: Absolute rates of and rate differences in firearm and nonfirearm homicide and suicide in black and white men. Results: During the 9-year study period, 84 113 homicides and 251 772 suicides occurred. Black–white differences in rates of firearm homicide and suicide varied widely across states. Relative to white men, black men had between 9 and 57 additional firearm homicides per 100 000 per year, with black men in Missouri, Michigan, Illinois, Indiana, and Pennsylvania having more than 40 additional firearm homicides per 100 000 per year. White men had between 2 fewer and 16 more firearm suicides per 100 000 per year, with the largest inequalities observed in southern and western states and the smallest in the District of Columbia and densely populated northeastern states. Limitations: Some homicides and suicides may have been misclassified as deaths due to unintentional injury. Survey data on state household gun ownership were collected in 2004 and may have shifted during the past decade. Conclusion: The large state-to-state variation in firearm homicide and suicide rates, as well as the racial inequalities in these numbers, highlights states where policies may be most beneficial in reducing homicide and suicide deaths and the racial disparities in their rates. © 2018 American College of Physicians.
Sreeramareddy C, Harper S, Ernstsen L. “Educational and Wealth Inequalities in Tobacco Use among Men and Women in 54 Low-Income and Middle-Income Countries.” Tobacco Control. 2018;27(1):26–34
Background Socioeconomic differentials of tobacco smoking in high-income countries are well described. However, studies to support health policies and place monitoring systems to tackle socioeconomic inequalities in smoking and smokeless tobacco use common in low- and-middle-income countries (LMICs) are seldom reported. We aimed to describe, sex-wise, educational and wealth-related inequalities in tobacco use in LMICs. Methods We analysed Demographic and Health Survey data on tobacco use collected from large nationally representative samples of men and women in 54 LMICs. We estimated the weighted prevalence of any current tobacco use (including smokeless tobacco) in each country for 4 educational groups and 4 wealth groups. We calculated absolute and relative measures of inequality, that is, the slope index of inequality (SII) and relative index of inequality (RII), which take into account the distribution of prevalence across all education and wealth groups and account for population size. We also calculated the aggregate SII and RII for low-income (LIC), lower-middle-income (lMIC) and upper-middle-income (uMIC) countries as per World Bank classification. Findings Male tobacco use was highest in Bangladesh (70.3\%) and lowest in Sao Tome (7.4\%), whereas female tobacco use was highest in Madagascar (21\%) and lowest in Tajikistan (0.22\%). Among men, educational inequalities varied widely between countries, but aggregate RII and SII showed an inverse trend by country wealth groups. RII was 3.61 (95\% CI 2.83 to 4.61) in LICs, 1.99 (95\% CI 1.66 to 2.38) in lMIC and 1.82 (95\% CI 1.24 to 2.67) in uMIC. Wealth inequalities among men varied less between countries, but RII and SII showed an inverse pattern where RII was 2.43 (95\% CI 2.05 to 2.88) in LICs, 1.84 (95\% CI 1.54 to 2.21) in lMICs and 1.67 (95\% CI 1.15 to 2.42) in uMICs. For educational inequalities among women, the RII varied much more than SII varied between the countries, and the aggregate RII was 14.49 (95\% CI 8.87 to 23.68) in LICs, 3.05 (95\% CI 1.44 to 6.47) in lMIC and 1.58 (95\% CI 0.33 to 7.56) in uMIC. Wealth inequalities among women showed a pattern similar to that of men: the RII was 5.88 (95\% CI 3.91 to 8.85) in LICs, 1.76 (95\% CI 0.80 to 3.85) in lMIC and 0.39 (95\% CI 0.09 to 1.64) in uMIC. In contrast to men, among women, the SII was pro-rich (higher smoking among the more advantaged) in 13 of the 52 countries (7 of 23 lMIC and 5 of 7 uMIC). Interpretation Our results confirm that socioeconomic inequalities tobacco use exist in LMIC, varied widely between the countries and were much wider in the lowest income countries. These findings are important for better understanding and tackling of socioeconomic inequalities in health in LMIC.
Arsenault C*, Harper S, Nandi A, Mendoza Rodríguez J, Hansen P, Johri M. “Monitoring Equity in Vaccination Coverage: A Systematic Analysis of Demographic and Health Surveys from 45 Gavi-supported Countries.” Vaccine. 2017;35(6):951–959
Objectives (1) To conduct a systematic analysis of inequalities in childhood vaccination coverage in Gavi-supported countries; (2) to comparatively assess alternative measurement approaches and how they may affect cross-country comparisons of the level of inequalities. Methods Using the most recent Demographic and Health Surveys (2005–2014) in 45 Gavi-supported countries, we measured inequalities in vaccination coverage across seven dimensions of social stratification and of vulnerability to poor health outcomes. We quantified inequalities using pairwise comparisons (risk differences and ratios) and whole spectrum measures (slope and relative indices of inequality). To contrast measurement approaches, we pooled the estimates using random-effects meta-analyses, ranked countries by the magnitude of inequality and compared agreement in country ranks. Results At the aggregate level, maternal education, multidimensional poverty, and wealth index poverty were the dimensions associated with the largest inequalities. In 36 out of 45 countries, inequalities were substantial, with a difference in coverage of 10 percentage points or more between the top and bottom of at least one of these social dimensions. Important inequalities by child sex, child malnutrition and urban/rural residence were also found in a smaller set of countries. The magnitude of inequality and ranking of countries differed across dimension and depending on the measure used. Pairwise comparisons could not be estimated in certain countries. The slope and relative indices of inequality were estimated in all countries and produced more stable country rankings, and should thus facilitate more reliable international comparisons. Conclusions Inequalities in vaccination coverage persist in a large majority of Gavi-supported countries. Inequalities should be monitored across multiple dimensions of vulnerability. Using whole spectrum measures to quantify inequality across multiple ordered social groups has important advantages. We illustrate these findings using an equity dashboard designed to support decision-making in the Sustainable Development Goals period. © 2017 Elsevier Ltd
Arsenault C*, Johri M, Nandi A, Mendoza Rodríguez J, Hansen P, Harper S. “Country-Level Predictors of Vaccination Coverage and Inequalities in Gavi-supported Countries.” Vaccine. 2017;35(18):2479–2488
Background Important inequalities in childhood vaccination coverage persist between countries and population groups. Understanding why some countries achieve higher and more equitable levels of coverage is crucial to redress these inequalities. In this study, we explored the country-level determinants of (1) coverage of the third dose of diphtheria-tetanus-pertussis- (DTP3) containing vaccine and (2) within-country inequalities in DTP3 coverage in 45 countries supported by Gavi, the Vaccine Alliance. Methods We used data from the most recent Demographic and Health Surveys (DHS) conducted between 2005 and 2014. We measured national DTP3 coverage and the slope index of inequality in DTP3 coverage with respect to household wealth, maternal education, and multidimensional poverty. We collated data on country health systems, health financing, governance and geographic and sociocultural contexts from published sources. We used meta-regressions to assess the relationship between these country-level factors and variations in DTP3 coverage and inequalities. To validate our findings, we repeated these analyses for coverage with measles-containing vaccine (MCV). Results We found considerable heterogeneity in DTP3 coverage and in the magnitude of inequalities across countries. Results for MCV were consistent with those from DTP3. Political stability, gender equality and smaller land surface were important predictors of higher and more equitable levels of DTP3 coverage. Inequalities in DTP3 coverage were also lower in countries receiving more external resources for health, with lower rates of out-of-pocket spending and with higher national coverage. Greater government spending on heath and lower linguistic fractionalization were also consistent with better vaccination outcomes. Conclusion Improving vaccination coverage and reducing inequalities requires that policies and programs address critical social determinants of health including geographic and social exclusion, gender inequality and the availability of financial protection for health. Further research should investigate the mechanisms contributing to these associations. © 2017 Elsevier Ltd
Arsenault C*, Harper S, Nandi A, Mendoza Rodríguez J, Hansen P, Johri M. “An Equity Dashboard to Monitor Vaccination Coverage.” Bulletin of the World Health Organization. 2017;95(2):128–134
Equity monitoring is a priority for Gavi, the Vaccine Alliance, and for those implementing The 2030 agenda for sustainable development. For its new phase of operations, Gavi reassessed its approach to monitoring equity in vaccination coverage. To help inform this effort, we made a systematic analysis of inequalities in vaccination coverage across 45 Gavi-supported countries and compared results from different measurement approaches. Based on our findings, we formulated recommendations for Gavi’s equity monitoring approach. The approach involved defining the vulnerable populations, choosing appropriate measures to quantify inequalities, and defining equity benchmarks that reflect the ambitions of the sustainable development agenda. In this article, we explain the rationale for the recommendations and for the development of an improved equity monitoring tool. Gavi’s previous approach to measuring equity was the difference in vaccination coverage between a country’s richest and poorest wealth quintiles. In addition to the wealth index, we recommend monitoring other dimensions of vulnerability (maternal education, place of residence, child sex and the multidimensional poverty index). For dimensions with multiple subgroups, measures of inequality that consider information on all subgroups should be used. We also recommend that both absolute and relative measures of inequality be tracked over time. Finally, we propose that equity benchmarks target complete elimination of inequalities. To facilitate equity monitoring, we recommend the use of a data display tool – the equity dashboard – to support decisionmaking in the sustainable development period. We highlight its key advantages using data from Côte d’Ivoire and Haiti. © 2017, World Health Organization. All rights reserved.
Carter R, Lévesque J, Harper S, Quesnel-Vallée A. “Measuring the Effect of Family Medicine Group Enrolment on Avoidable Visits to Emergency Departments by Patients with Diabetes in Quebec, Canada.” Journal of Evaluation in Clinical Practice. 2017;23(2):369–376
The Family Medicine Group (FMG) model of primary care in Quebec, Canada, was driven by the voluntary implementation of family physicians. Our main objective was to measure the effect of FMG enrolment on avoidable use of the emergency department (ED) by diabetic patients. We also sought to determine if effects differed according to whether patients were infrequent or frequent users of the ED and according to high- versus low-regional levels of enrolment. We used data from provincial health administrative databases to identify the diabetic patient population over the age of 20~years for each fiscal year between 2003-2004 and 2011-2012. We used fixed effects and marginal structural models to estimate the effect of enrolment in FMGs on avoidable use of the ED. Our results indicated that for every 10-percentage point increase in the population enrolled with an FMG in the year prior to an event, there was a 3\% reduction in avoidable visits to the ED made by an individual (RR~ = ~0.97; 95\% CI~ = ~0.95, 0.99). We found a significant reduction among diabetic patients who had at most 1 visit to the ED per year (RR~ = ~0.97; 95\% CI~ = ~0.95, 0.99) and nonsignificant effects among more frequent users. Within low-enrolment regions, a 10-percentage point increase in enrolment in FMG practices at t~−~1 led to an 18\% decrease in the number of avoidable ED visits (RR~ = ~0.82; 95\% CI~ = ~0.78, 0.87). The effect disappeared when the analyses were restricted to the high-enrolment regions (RR~ = ~1.00; 95\% CI~ = ~0.92, 1.09). The design and implementation of the incentive to promote team-based practice may not have borne much influence on early adopters who may have been overrepresented by physicians from high-performing practices before the introduction of the reform. © 2016 John Wiley \& Sons, Ltd.
Elani H, Harper S, Thomson W, Espinoza I, Mejia G, Ju X, Jamieson L, Kawachi I, Kaufman J. “Social Inequalities in Tooth Loss: A Multinational Comparison.” Community Dentistry and Oral Epidemiology. 2017;45(3):266–274
Objectives: To conduct cross-national comparison of education-based inequalities in tooth loss across Australia, Canada, Chile, New Zealand and the United States. Methods: We used nationally representative data from Australia's National Survey of Adult Oral Health; Canadian Health Measures Survey; Chile's First National Health Survey Ministry of Health; US National Health and Nutrition Examination Survey; and the New Zealand Oral Health Survey. We examined the prevalence of edentulism, the proportion of individuals having {$<$}21 teeth and the mean number of teeth present. We used education as a measure of socioeconomic position and measured absolute and relative inequalities. We used random-effects meta-analysis to summarize inequality estimates. Results: The USA showed the widest absolute and relative inequality in edentulism prevalence, whereas Chile demonstrated the largest absolute and relative social inequality gradient for the mean number of teeth present. Australia had the narrowest absolute and relative inequality gap for proportion of individuals having {$<$}21 teeth. Pooled estimates showed substantial heterogeneity for both absolute and relative inequality measures. Conclusions: There is a considerable variation in the magnitude of inequalities in tooth loss across the countries included in this analysis. © 2017 John Wiley \& Sons A/S. Published by John Wiley \& Sons Ltd
Harper S, Kaufman J, Cooper R. “Declining US Life Expectancy.” Epidemiology. 2017;28(6):E54-E56
Harper S, Bruckner T. “Did the Great Recession Increase Suicides in the USA? Evidence from an Interrupted Time-Series Analysis.” Annals of Epidemiology. 2017;27(7):409-414.e6
Purpose Research suggests that the Great Recession of 2007–2009 led to nearly 5000 excess suicides in the United States. However, prior work has not accounted for seasonal patterning and unique suicide trends by age and gender. Methods We calculated monthly suicide rates from 1999 to 2013 for men and women aged 15 and above. Suicide rates before the Great Recession were used to predict the rate during and after the Great Recession. Death rates for each age-gender group were modeled using Poisson regression with robust variance, accounting for seasonal and nonlinear suicide trajectories. Results There were 56,658 suicide deaths during the Great Recession. Age- and gender-specific suicide trends before the recession demonstrated clear seasonal and nonlinear trajectories. Our models predicted 57,140 expected suicide deaths, leading to 482 fewer observed than expected suicides (95\% confidence interval −2079, 943). Conclusions We found little evidence to suggest that the Great Recession interrupted existing trajectories of suicide rates. Suicide rates were already increasing before the Great Recession for middle-aged men and women. Future studies estimating the impact of recessions on suicide should account for the diverse and unique suicide trajectories of different social groups. © 2017 Elsevier Inc.
Harper S, Strumpf E. “Primary Enforcement of Mandatory Seat Belt Laws and Motor Vehicle Crash Deaths.” American Journal of Preventive Medicine. 2017;53(2):176–183
Introduction Policies that allow directly citing motorists for seat belt non-use (primary enforcement) have been shown to reduce motor vehicle crash deaths relative to secondary enforcement, but the evidence base is dated and does not account for recent improvements in vehicle designs and road safety. The purpose of this study was to test whether recent upgrades to primary enforcement still reduce motor vehicle crash deaths. Methods In 2016, researchers used motor vehicle crash death data from the Fatal Analysis Reporting System for 2000–2014 and calculated rates using both person- and exposure-based denominators. Researchers used a difference-in-differences design to estimate the effect of primary enforcement on death rates, and estimated negative binomial regression models, controlling for age, substance use involvement, fixed state characteristics, secular trends, state median household income, and other state-level traffic safety policies. Results Models adjusted only for crash characteristics and state-level covariates models showed a protective effect of primary enforcement (rate ratio, 0.88, 95\% CI = 0.77, 0.98; rate difference, –1.47 deaths per 100,000 population, 95\% CI = –2.75, –0.19). After adjustment for fixed state characteristics and secular trends, there was no evidence of an effect of upgrading from secondary to primary enforcement in the whole population (rate ratio, 0.98, 95\% CI = 0.92, 1.04; rate difference, –0.22, 95\% CI = –0.90, 0.46) or for any age group. Conclusions Upgrading to primary enforcement no longer appears protective for motor vehicle crash death rates. © 2017 American Journal of Preventive Medicine
Hutcheon J, Riddell C, Strumpf E, Lee L, Harper S. “Safety of Labour and Delivery Following Closures of Obstetric Services in Small Community Hospitals.” CMAJ. 2017;189(11):E431-E436
Background: In recent decades, many smaller hospitals in British Columbia, Canada, have stopped providing planned obstetric services. We examined the effect of these service closures on the labour and delivery outcomes of pregnant women living in affected communities. Methods: We used maternal postal codes to identify delivery records (1998-2014) of women residing in a community affected by service closure. The records were obtained from the British Columbia Perinatal Data Registry. We examined the effect of the closures using a within-communities fixed-effects framework and included similar-sized communities without service closures to control for underlying time trends. The primary outcome was a previously published composite measure of labour and delivery safety, the Adverse Outcome Index, which includes adverse events such as birth injury and unanticipated operative procedures, and includes weights for severity of adverse events. Secondary outcomes included maternal or newborn transfer, and use of obstetric interventions. Results: We found little evidence that closure of planned obstetric services affected the risk of composite adverse maternal-newborn outcome (-0.4 excess adverse events per 100 deliveries, 95\% confidence interval [CI] -2.0 to 1.1), or most other secondary outcomes. The severity of composite outcome events decreased following the closures (rate ratio 0.58, 95\% CI 0.36 to 0.89). Closures were associated with increases in use of epidural analgesia (3.4 excess events per 100 deliveries, 95\% CI 0.4 to 6.3) and length of antepartum stay (0.6 h, 95\% CI 0.1 to 1.0 h). Interpretation: Closure of planned obstetric services in low-volume hospitals was not associated with an increase or decrease in frequency of adverse events during labour and delivery. © 2017 Joule Inc.
Hutcheon J, Harper S, Strumpf E. “Oregon’s Hard-Stop Policy Limiting Elective Early-Term Deliveries: Limitations of before-and-after Studies in Evaluating Obstetric Policies.” Obstetrics and Gynecology. 2017;129(4):753–754
Jahagirdar D, Harper S, Heymann J, Swaminathan H, Mukherji A, Nandi A. “The Effect of Paid Maternity Leave on Early Childhood Growth in Low-Income and Middle-Income Countries.” BMJ global health. 2017;2(3):e000294
Background: Despite recent improvements, low height-for-age, a key indicator of inadequate child nutrition, is an ongoing public health issue in low-income and middle-income countries. Paid maternity leave has the potential to improve child nutrition, but few studies have estimated its impact. Methods: We used data from 583 227 children younger than 5 years in 37 countries surveyed as part of the Demographic and Health Surveys (2000-2014) to compare the change in children's height-for-age z score in five countries that increased their legislated duration of paid maternity leave (Uganda, Zambia, Zimbabwe, Bangladesh and Lesotho) relative to 32 other countries that did not. A quasiexperimental difference-in-difference design involving a linear regression of height-for-age z score on the number of weeks of legislated paid maternity leave was used. We included fixed effects for country and birth year to control for, respectively, fixed country characteristics and shared trends in height-for-age, and adjusted for time-varying covariates such as gross domestic product per capita and the female labour force participation rate. Results: The mean height-for-age z scores in the pretreatment period were -1.91 (SD = 1.44) and -1.47 (SD = 1.57) in countries that did and did not change their policies, respectively. The scores increased in treated and control countries over time. A 1-month increase in legislated paid maternity leave was associated with a decrease of 0.08(95\% CI -0.20 to 0.04) in child height-for-age z score. Sensitivity analyses did not support a robust association between paid maternity leave policies and height-for-age z score. Conclusion: We found little evidence that recent changes in legislated paid maternity leave have been sufficient to affect child height-for-age z scores. The relatively short durations of leave, the potential for low coverage and the strong increasing trend in children's growth may explain our findings. Future studies considering longer durations or combined interventions may reveal further insight to support policy.
Manivong P*, Harper S, Strumpf E. “The Contribution of Excise Cigarette Taxes on the Decline in Youth Smoking in Canada during the Time of the Federal Tobacco Control Strategy (2002-2012).” Canadian journal of public health = Revue canadienne de sante publique. 2017;108(2):e117-e123
OBJECTIVES: To evaluate the impact of changes in cigarette taxes on smoking for youths aged 15-18 in Canada during the time of the Federal Tobacco Control Strategy (FTCS). METHODS: We used a difference-in-differences framework and leveraged the variation in cigarette taxes across Canada and over time. We used regression models with province and year fixed effects, and individual-level and provincial-level covariates on 2002-2012 data from the Canadian Tobacco Use Monitoring Survey. RESULTS: Tax increases generally did not affect smoking outcomes. Each increase of CAD 1.00 (adjusted to year 2000 dollars) in excise cigarette taxes per package of 20 was associated with a 0.2 percentage point (95\% CI: -1.8; 2.2) change in smoking prevalence, and a change of 0.3 in mean cigarettes smoked in the past week (95\% CI: -1.2; 1.8). CONCLUSION: From 2002 to 2012, smoking prevalence and mean smoking frequency were in steady decline among youths in Canada. This decline, however, was evident even among provinces with stable or decreasing cigarette tax levels. Tobacco taxes have mostly increased since the 1980s, and so, tax levels were already quite high by the launch of the FTCS. Province fixed effects and common temporal changes accounted for 83.7\% of the variation in smoking prevalence. We derived similar results for smoking frequency. The cumulative tax increase during our study period was at least 1.00 for only three provinces. Thus, our findings suggest that factors driving down tobacco use among youths in all provinces appear to outweigh any impact of small tax increases at already high tax levels.
McLinden T, Moodie E, Hamelin A, Harper S, Walmsley S, Paradis G, Aibibula W, Klein M, Cox J. “Injection Drug Use, Unemployment, and Severe Food Insecurity Among HIV-HCV Co-Infected Individuals: A Mediation Analysis.” AIDS and Behavior. 2017;21(12):3496–3505
Severe food insecurity (FI), which indicates reduced food intake, is common among HIV-hepatitis C virus (HCV) co-infected individuals. Given the importance of unemployment as a proximal risk factor for FI, this mediation analysis examines a potential mechanism through which injection drug use (IDU) is associated with severe FI. We used biannual data from the Canadian Co-infection Cohort (N~ = ~429 with 3 study visits, 2012–2015). IDU in the past 6~months (exposure) and current unemployment (mediator) were self-reported. Severe FI in the following 6~months (outcome) was measured using the Household Food Security Survey Module. An overall association and a controlled direct effect were estimated using marginal structural models. Among participants, 32\% engaged in IDU, 78\% were unemployed, and 29\% experienced severe FI. After adjustment for confounding and addressing censoring through weighting, the overall association (through all potential pathways) between IDU and severe FI was: risk ratio (RR)~ = ~1.69 (95\% confidence interval [CI]~ = ~1.15–2.48). The controlled direct effect (the association through all potential pathways except that of unemployment) was: RR~ = ~1.65 (95\% CI~ = ~1.08–2.53). We found evidence of an overall association between IDU and severe FI and estimated a controlled direct effect that is suggestive of pathways from IDU to severe FI that are not mediated by unemployment. Specifically, an overall association and a controlled direct effect that are similar in magnitude suggests that the potential impact of IDU on unemployment is not the primary mechanism through which IDU is associated with severe FI. Therefore, while further research is required to understand the mechanisms linking IDU and severe FI, the strong overall association suggests that reductions in IDU may mitigate severe FI in this vulnerable subset of the HIV-positive population. © 2017, Springer Science+Business Media, LLC.
Richardson R, Nandi A, Jaswal S, Harper S. “Are Work Demands Associated with Mental Distress? Evidence from Women in Rural India.” Social Psychiatry and Psychiatric Epidemiology. 2017;52(12):1501–1511
Purpose: High work demands might be a determinant of poor mental health among women in low- and middle-income countries, especially in rural settings where women experience greater amounts of labor-intensive unpaid work. Research originating from such settings is lacking. Methods: We estimated the cross-sectional association between work demands and mental distress among 3177 women living in 160 predominantly tribal communities in southern Rajasthan, India. A structured questionnaire captured the number of minutes women spent on various activities in the last 24~h, and we used this information to measure women’s work demands, including the total work amount, nature of work (e.g., housework), and type of work (e.g., cooking). Mental distress was measured with the Hindi version of the 12-item General Health Questionnaire. We used negative binomial regression models to estimate the association between work demands (amount, nature, and type) and mental distress. Results: On average, women spent more than 9.5~h a day on work activities. The most time, intensive work activity was caring for children, the elderly, or disabled (149~min). In adjusted models, we found a U-shaped association between work amount and mental distress. High amounts of housework were associated with higher distress, whereas paid work and farmwork amount were not. Certain types of housework, including collecting water and cleaning, were associated with increased distress scores. Conclusions: We found an association between aspects of work demands and mental distress. Research in other contexts where women perform high amounts of unpaid work, particularly within the home or farm, is warranted. © 2017, Springer-Verlag GmbH Germany.
Riddell C, Harper S, Kaufman J. “Trends in Differences in US Mortality Rates between Black and White Infants.” JAMA Pediatrics. 2017;171(9):911–913
Strumpf E, Charters T, Harper S, Nandi A. “Did the Great Recession Affect Mortality Rates in the Metropolitan United States? Effects on Mortality by Age, Gender and Cause of Death.” Social Science and Medicine. 2017;189:11–16
Objectives Mortality rates generally decline during economic recessions in high-income countries, however gaps remain in our understanding of the underlying mechanisms. This study estimates the impacts of increases in unemployment rates on both all-cause and cause-specific mortality across U.S. metropolitan regions during the Great Recession. Methods We estimate the effects of economic conditions during the recent and severe recessionary period on mortality, including differences by age and gender subgroups, using fixed effects regression models. We identify a plausibly causal effect by isolating the impacts of within-metropolitan area changes in unemployment rates and controlling for common temporal trends. We aggregated vital statistics, population, and unemployment data at the area-month-year-age-gender-race level, yielding 527,040 observations across 366 metropolitan areas, 2005–2010. Results We estimate that a one percentage point increase in the metropolitan area unemployment rate was associated with a decrease in all-cause mortality of 3.95 deaths per 100,000 person years (95\%CI −6.80 to −1.10), or 0.5\%. Estimated reductions in cardiovascular disease mortality contributed 60\% of the overall effect and were more pronounced among women. Motor vehicle accident mortality declined with unemployment increases, especially for men and those under age 65, as did legal intervention and homicide mortality, particularly for men and adults ages 25–64. We find suggestive evidence that increases in metropolitan area unemployment increased accidental drug poisoning deaths for both men and women ages 25–64. Conclusions Our finding that all-cause mortality decreased during the Great Recession is consistent with previous studies. Some categories of cause-specific mortality, notably cardiovascular disease, also follow this pattern, and are more pronounced for certain gender and age groups. Our study also suggests that the recent recession contributed to the growth in deaths from overdoses of prescription drugs in working-age adults in metropolitan areas. Additional research investigating the mechanisms underlying the health consequences of macroeconomic conditions is warranted. © 2017 Elsevier Ltd
Austin N, Harper S, Kaufman J, Hamra G. “Challenges in Reproducing Results from Publicly Available Data: An Example of Sexual Orientation and Cardiovascular Disease Risk.” Journal of Epidemiology and Community Health. 2016;70(8):807–812
Background Replication is a vital part of the research process and has recently received considerable attention. Analyses using publicly available data should, if adequately described, be reproducible without assistance from the original investigators. Using data from the US National Health and Nutrition Examination Survey (NHANES), a recent study reported a statistically significant difference in cardiovascular disease risk comparing subgroups of sexual minority men. We attempted to reproduce these findings and assessed whether the results were robust to alternative analytic strategies and assumptions. Methods We used the exclusion criteria and coding strategy described in the original paper to construct our analytical data set. Sampling weights were constructed in accordance with NHANES analytical guidelines. We estimated crude and covariate-adjusted associations between sexual orientation and vascular age using the regression models specified in the original report. We also conducted a series of sensitivity analyses to improve on the original findings. Results Our replication attempt was partially successful: We replicated the general trends reported in the original analysis, but not identical effect estimates. Importantly, we identified a potential misapplication of the Framingham Risk Score; correcting for this increased the probability that the reported null hypothesis test was a type I error. Conclusions This paper supports the recent calls for greater transparency and improved reporting in research. Even with a publicly available and well-documented data source, we were unable to exactly replicate another study's original findings. Our sensitivity analyses revealed key issues in the original analysis and demonstrate the scientific importance of research replication.
Austin N, Harper S, Strumpf E. “Does Segregation Lead to Lower Birth Weight?.” Epidemiology. 2016;27(5):682–689
Background: Racial residential segregation in the United States has been linked to racial differences in birth outcomes, with studies reporting associations between segregation and birth weight. However, this relationship is likely confounded, and many individual and neighborhood-level covariates included in previous models are likely mediators, potentially obscuring any causal impact of segregation on birth weight. Methods: We compiled a record of non-Hispanic black and white singleton births to US-born/resident mothers in 2000, linked to segregation indices at the metropolitan statistical area (MSA) level in the non-Southern US. Segregation was measured via the dissimilarity index. The outcomes were individual-level birth weight and the metropolitan statistical area-level black/white gap in birth weight. We instrumented for segregation using the railroad division index. We compared race-stratified ordinary least squares models to two-stage least squares models, with cluster robust standard errors. Results: We estimated a 1.2 g decrease in black birth weight for every one-percentage point increase in segregation (95\% confidence interval [CI]: -1.9, -0.50) via ordinary least squares but a 2.8 g decrease (95\% CI: -6.0, 0.48) using two-stage least squares. For white infants, our ordinary least squares estimate was 0.53 (95\% CI: -0.23, 1.3), and our two-stage least squares estimate was in the opposite direction (-0.68, 95\% CI: -3.5, 2.1). Conclusions: Ordinary least squares estimates may understate the effect of segregation on birth weight in blacks. Evidence from instrumental variable models was consistent with a causal impact of segregation on black birth outcomes, but estimates were imprecise and may be affected by weak instrument bias. © 2016 Wolters Kluwer Health, Inc.
Charters T, Harper S, Strumpf E, Subramanian S, Arcaya M, Nandi A. “The Effect of Metropolitan-Area Mortgage Delinquency on Health Behaviors, Access to Health Services, and Self-Rated Health in the United States, 2003-2010.” Social Science and Medicine. 2016;161:74–82
The recent housing crisis offers the opportunity to understand the effects of unique indicators of macroeconomic conditions on health. We linked data on the proportion of mortgage borrowers per US metropolitan-area who were at least 90 days delinquent on their payments with individual-level outcomes from a representative sample of 1,021,341 adults surveyed through the Behavioral Risk Factor Surveillance System (BRFSS) between 2003 and 2010. We estimated the effects of metropolitan-area mortgage delinquency on individual health behaviors, medical coverage, and health status, as well as whether effects varied by race/ethnicity. Results showed that increases in the metropolitan-area delinquency rate resulted in decreases in heavy alcohol consumption and increases in exercise and health insurance coverage. However, the delinquency rate was also associated with increases in smoking and obesity in some population groups, suggesting the housing crisis may have induced stress-related behavioral change. Overall, the effects of metropolitan-area mortgage delinquency on population health were relatively modest. © 2016 Elsevier Ltd.
Gray A, Richer F, Harper S. “Individual- and Community-Level Determinants of Inuit Youth Mental Wellness.” Canadian Journal of Public Health. 2016;107(3):e251-e257
OBJECTIVES: Following the onset of intensive colonial intervention and rapid social change in the lives of Inuit people, youth in Nunavik have experienced high rates of mental health problems and suicide. Inuit people describe a broad range of contextual influences on mental wellness based on lived experience, but most epidemiological studies have focused on individual risk factors and pathologies. This study aimed to assess the influence of multiple determinants of mental wellness among Inuit youth in Nunavik, including culturally meaningful activities, housing and community social characteristics. METHODS: Mental wellness was measured in the form of two primary outcomes: self-esteem and suicidal ideation. Using cross-sectional data from the 2004 Nunavik Inuit Health Survey and multilevel regression modelling, we estimated associations between these two outcomes and various independent individual- and community-level explanatory factors among Inuit youth. All variables were selected to reflect Inuit perspectives on determinants of mental wellness. The study design and interpretation of results were validated with Inuit community representatives. RESULTS: Pride in Inuit identity, traditional activities, community-level social support and community-level socio-economic status were found to be protective. Barriers to participating in traditional activities, household crowding and high community rates of violence were risk factors. CONCLUSION: These findings support Inuit perspectives, expand the scope of epidemiological analysis of Inuit mental wellness and reinforce the need for locally informed, community-wide approaches to mental wellness promotion for Inuit youth. © 2016 Canadian Public Health Association or its licensor.
Hajna S, Ross N, Joseph L, Harper S, Dasgupta K. “Neighbourhood Walkability and Daily Steps in Adults with Type 2 Diabetes.” PLoS ONE. 2016;11(3)
Introduction: There is evidence that greater neighbourhood walkability (i.e., neighbourhoods with more amenities and well-connected streets) is associated with higher levels of total walking in Europe and in Asia, but it remains unclear if this association holds in the Canadian context and in chronic disease populations. We examined the relationships of different walkability measures to biosensor-assessed total walking (i.e., steps/day) in adults with type 2 diabetes living in Montreal (QC, Canada). Materials and Methods: Participants (60.5±10.4 years; 48.1\% women) were recruited through McGill University-affiliated clinics (June 2006 to May 2008). Steps/day were assessed once per season for one year with pedometers. Neighbourhood walkability was evaluated through participant reports, in-field audits, Geographic Information Systems (GIS)-derived measures, and the Walk Score®. Relationships between walkability and daily steps were estimated using Bayesian longitudinal hierarchical linear regression models (n = 131). Results: Participants who reported living in the most compared to the least walkable neighbourhoods completed 1345 more steps/day (95\% Credible Interval: 718,1976; Quartiles 4 versus 1). Those living in the most compared to the least walkable neighbourhoods (based on GIS-derived walkability) completed 606 more steps per day (95\% CrI: 8,1203). No statistically significant associations with steps were observed for audit-assessed walkability or the Walk Score®. Conclusions: Adults with type 2 diabetes who perceived their neighbourhoods as more walkable accumulated more daily steps. This suggests that knowledge of local neighborhood features that enhance walking is a meaningful predictor of higher levels of walking and an important component of neighbourhood walkability. © 2016 Hajna et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Hutcheon J, Strumpf E, Harper S, Giesbrecht E. “Authors’ Reply Re: Maternal and Neonatal Outcomes after Implementation of a Hospital Policy to Limit Low-Risk Planned Caesarean Deliveries before 39 Weeks of Gestation: An Interrupted Time-Series Analysis.” BJOG: An International Journal of Obstetrics and Gynaecology. 2016;123(6):1035
Jamieson L, Elani H, Mejia G, Ju X, Kawachi I, Harper S, Thomson W, Kaufman J. “Inequalities in Indigenous Oral Health: Findings from Australia, New Zealand, and Canada.” Journal of Dental Research. 2016;95(12):1375–1380
The objective was to compare absolute differences in the prevalence of Indigenous-related inequalities in dental disease experience and self-rated oral health in Australia, Canada, and New Zealand. Data were sourced from national oral health surveys in Australia (2004 to 2006), Canada (2007 to 2009), and New Zealand (2009). Participants were aged ≥18 y. The authors measured age- and sex-adjusted inequalities by estimating absolute prevalence differences and their corresponding 95\% confidence intervals (95\% CIs). Clinical measures included the prevalence of untreated decayed teeth, missing teeth, and filled teeth; self-reported measures included the prevalence of "fair" or "poor" self-rated oral health. The overall pattern of Indigenous disadvantage was similar across all countries. The summary estimates for the adjusted prevalence differences were as follows: 16.5 (95\% CI: 11.1 to 21.9) for decayed teeth (all countries combined), 18.2 (95\% CI: 12.5 to 24.0) for missing teeth, 0.8 (95\% CI: -1.9 to 3.5) for filled teeth, and 17.5 (95\% CI: 11.3 to 23.6) for fair/poor self-rated oral health. The I2 estimates were small for each outcome: 0.0\% for decayed, missing, and filled teeth and 11.6\% for fair/poor self-rated oral health. Irrespective of country, when compared with their non-Indigenous counterparts, Indigenous persons had more untreated dental caries and missing teeth, fewer teeth that had been restored (with the exception of Canada), and a higher proportion reporting fair/poor self-rated oral health. There were no discernible differences among the 3 countries. © International \& American Associations for Dental Research.
King N, Strumpf E, Harper S. “Has the Increase in Disability Insurance Participation Contributed to Increased Opioid-Related Mortality?.” Annals of Internal Medicine. 2016;165(10):729–730
Kopec J, Sayre E, Fines P, Flanagan W, Nadeau C, Okhmatovskaia A, Wolfson M, Buckeridge D, Dahaghin S, Harper S, Heath A, Hennessy D, Liu R, Manuel D, Oderkirk J, Rahman M, Sharif B, Smith B, Tuna M. “Effects of Reductions in Body Mass Index on the Future Osteoarthritis Burden in Canada: A Population-Based Microsimulation Study.” Arthritis Care and Research. 2016;68(8):1098–1105
Objective: Osteoarthritis (OA) is the most common joint disease and a major cause of disability. Incidence and prevalence of OA are expected to increase due to population aging and increased levels of obesity. The purpose of this study was to project the effect of hypothetical interventions that change the distribution of body mass index (BMI) on OA burden in Canada. Methods: We used a microsimulation computer model of OA based on the Population Health Model platform. The model used demographic predictions for Canada and population data from an administrative database in British Columbia and national Canadian surveys. Results: Under the base-case scenario, between 2010 and 2030, OA prevalence is expected to increase from 11.5\% to 15.6\% in men and 16.3\% to 21.1\% in women. In scenarios assuming, on average, a 0.3-, 0.5-, or 1-unit drop in BMI per year, OA prevalence in 2030 would reach 14.9\%, 14.6\%, and 14.2\% in men and 20.3\%, 19.7\%, and 18.5\%, in women, respectively. Under these scenarios, the proportion of new cases prevented would be 9.5\%, 13.2\%, and 16.7\%, respectively, in men, and 9.1\%, 15.2\%, and 25.0\% in women. Targeting only those people ages ≥50 years for weight reduction would achieve approximately 70\% of the impact of a full population strategy. Targeting only the obese (BMI ≥30) would likely result in a larger benefit for men than women. Conclusion: Due to the aging of the population, OA will remain a major and growing health issue in Canada over the next 2 decades, regardless of the course of the obesity epidemic. © 2016, American College of Rheumatology
Majid M, Mendoza Rodríguez J, Harper S, Frank J, Nandi A. “Do Minimum Wages Improve Early Life Health? Evidence from Developing Countries.” Social Science and Medicine. 2016;158:105–113
The impact of legislated minimum wages on the early-life health of children living in low and middle-income countries has not been examined. For our analyses, we used data from the Demographic and Household Surveys (DHS) from 57 countries conducted between 1999 and 2013. Our analyses focus on height-for-age z scores (HAZ) for children under 5 years of age who were surveyed as part of the DHS. To identify the causal effect of minimum wages, we utilized plausibly exogenous variation in the legislated minimum wages during each child's year of birth, the identifying assumption being that mothers do not time their births around changes in the minimum wage. As a sensitivity exercise, we also made within family comparisons (mother fixed effect models). Our final analysis on 49 countries reveal that a 1\% increase in minimum wages was associated with 0.1\% (95\% CI = -0.2, 0) decrease in HAZ scores. Adverse effects of an increase in the minimum wage were observed among girls and for children of fathers who were less than 35 years old, mothers aged 20-29, parents who were married, parents who were less educated, and parents involved in manual work. We also explored heterogeneity by region and GDP per capita at baseline (1999). Adverse effects were concentrated in lower-income countries and were most pronounced in South Asia. By contrast, increases in the minimum wage improved children's HAZ in Latin America, and among children of parents working in a skilled sector. Our findings are inconsistent with the hypothesis that increases in the minimum wage unconditionally improve child health in lower-income countries, and highlight heterogeneity in the impact of minimum wages around the globe. Future work should involve country and occupation specific studies which can explore not only different outcomes such as infant mortality rates, but also explore the role of parental investments in shaping these effects. © 2016 Elsevier Ltd.
McKinnon B, Harper S, Kaufman J. “Do Socioeconomic Inequalities in Neonatal Mortality Reflect Inequalities in Coverage of Maternal Health Services? Evidence from 48 Low- and Middle-Income Countries.” Maternal and Child Health Journal. 2016;20(2):434–446
Objectives: To examine socioeconomic and health system determinants of wealth-related inequalities in neonatal mortality rates (NMR) across 48 low- and middle-income countries. Methods: We used data from Demographic and Health Surveys conducted between 2006 and 2012. Absolute and relative inequalities for NMR and coverage of antenatal care, facility-based delivery, and Caesarean delivery were measured using the Slope Index of Inequality and Relative Index of Inequality, respectively. Meta-regression was used to assess whether variation in the magnitude of NMR inequalities was associated with inequalities in coverage of maternal health services, and whether country-level economic and health system factors were associated with mean NMR and socioeconomic inequality in NMR. Results: Of the three maternal health service indicators examined, the magnitude of socioeconomic inequality in NMR was most strongly related to inequalities in antenatal care. NMR inequality was greatest in countries with higher out-of-pocket health expenditures, more doctors per capita, and a higher adolescent fertility rate. Determinants of lower mean NMR (e.g., higher government health expenditures and a greater number of nurses/midwives per capita) differed from factors associated with lower NMR inequality. Conclusions: Reducing the financial burden of maternal health services and achieving universal coverage of antenatal care may contribute to a reduction in socioeconomic differences in NMR. Further investigation of the mechanisms contributing to these cross-national associations seems warranted. © 2015, Springer Science+Business Media New York.
Nandi A, Hajizadeh M*, Harper S, Koski A, Strumpf EC, Heymann J. “Increased Duration of Paid Maternity Leave Lowers Infant Mortality in Low- and Middle-Income Countries: A Quasi-Experimental Study.” PLoS medicine. 2016;13(3):e1001985
BACKGROUND: Maternity leave reduces neonatal and infant mortality rates in high-income countries. However, the impact of maternity leave on infant health has not been rigorously evaluated in low- and middle-income countries (LMICs). In this study, we utilized a difference-in-differences approach to evaluate whether paid maternity leave policies affect infant mortality in LMICs. METHODS AND FINDINGS: We used birth history data collected via the Demographic and Health Surveys to assemble a panel of approximately 300,000 live births in 20 countries from 2000 to 2008; these observational data were merged with longitudinal information on the duration of paid maternity leave provided by each country. We estimated the effect of an increase in maternity leave in the prior year on the probability of infant (¡1 y), neonatal (¡28 d), and post-neonatal (between 28 d and 1 y after birth) mortality. Fixed effects for country and year were included to control for, respectively, unobserved time-invariant confounders that varied across countries and temporal trends in mortality that were shared across countries. Average rates of infant, neonatal, and post-neonatal mortality over the study period were 55.2, 30.7, and 23.0 per 1,000 live births, respectively. Each additional month of paid maternity was associated with 7.9 fewer infant deaths per 1,000 live births (95\% CI 3.7, 12.0), reflecting a 13\% relative reduction. Reductions in infant mortality associated with increases in the duration of paid maternity leave were concentrated in the post-neonatal period. Estimates were robust to adjustment for individual, household, and country-level characteristics, although there may be residual confounding by unmeasured time-varying confounders, such as coincident policy changes. CONCLUSIONS: More generous paid maternity leave policies represent a potential instrument for facilitating early-life interventions and reducing infant mortality in LMICs and warrant further discussion in the post-2015 sustainable development agenda. From a policy planning perspective, further work is needed to elucidate the mechanisms that explain the benefits of paid maternity leave for infant mortality.
Nandi A, Maloney S, Agarwal P, Chandrashekar A, Harper S. “The Effect of an Affordable Daycare Program on Health and Economic Well-Being in Rajasthan, India: Protocol for a Cluster-Randomized Impact Evaluation Study.” BMC Public Health. 2016;16(1)
Background: The provision of affordable and reliable daycare services is a potentially important policy lever for empowering Indian women. Access to daycare might reduce barriers to labor force entry and generate economic opportunities for women, improve education for girls caring for younger siblings, and promote nutrition and learning among children. However, empirical evidence concerning the effects of daycare programs in low-and-middle-income countries is scarce. This cluster-randomized trial will estimate the effect of a community-based daycare program on health and economic well-being over the life-course among women and children living in rural Rajasthan, India. Methods: This three-year study takes place in rural communities from five blocks in the Udaipur District of rural Rajasthan. The intervention is the introduction of a full-time, affordable, community-based daycare program. At baseline, 3177 mothers with age eligible children living in 160 village hamlets were surveyed. After the baseline, these hamlets were randomized to the intervention or control groups and respondents will be interviewed on two more occasions. Primary social and economic outcomes include women's economic status and economic opportunity, women's empowerment, and children's educational attainment. Primary health outcomes include women's mental health, as well as children's nutritional status. Discussion: This interdisciplinary research initiative will provide rigorous evidence concerning the effects of daycare in lower-income settings. In doing so it will address an important research gap and has the potential to inform policies for improving the daycare system in India in ways that promote health and economic well-being. Trial registration: (1) The ISRCTN clinical trial registry (ISRCTN45369145), http://www.isrctn.com/ISRCTN45369145, registered on May 16, 2016 and (2) The American Economic Association's registry for randomized controlled trials (AEARCTR-0000774), http://www.socialscienceregistry.org/trials/774, registered on July 15, 2015 © 2016 Nandi et al.
Hajizadeh M*, Heymann J, Strumpf E, Harper S, Nandi A. “Paid Maternity Leave and Childhood Vaccination Uptake: Longitudinal Evidence from 20 Low-and-Middle-Income Countries.” Social Science and Medicine. 2015;140:104–117
The availability of maternity leave might remove barriers to improved vaccination coverage by increasing the likelihood that parents are available to bring a child to the clinic for immunizations. Using information from 20 low-and-middle-income countries (LMICs) we estimated the effect of paid maternity leave policies on childhood vaccination uptake. We used birth history data collected via Demographic and Health Surveys (DHS) to assemble a multilevel panel of 258,769 live births in 20 countries from 2001 to 2008; these data were merged with longitudinal information on the number of full-time equivalent (FTE) weeks of paid maternity leave guaranteed by each country. We used Logistic regression models that included country and year fixed effects to estimate the impact of increases in FTE paid maternity leave policies in the prior year on the receipt of the following vaccines: Bacillus Calmette-Guérin (BCG) commonly given at birth, diphtheria, tetanus, and pertussis (DTP, 3 doses) commonly given in clinic visits and Polio (3 doses) given in clinic visits or as part of campaigns. We found that extending the duration of paid maternity leave had a positive effect on immunization rates for all three doses of the DTP vaccine; each additional FTE week of paid maternity leave increased DTP1, 2 and 3 coverage by 1.38 (95\% CI = 1.18, 1.57), 1.62 (CI = 1.34, 1.91) and 2.17 (CI = 1.76, 2.58) percentage points, respectively. Estimates were robust to adjustment for birth characteristics, household-level covariates, attendance of skilled health personnel at birth and time-varying country-level covariates. We found no evidence for an effect of maternity leave on the probability of receiving vaccinations for BCG or Polio after adjustment for the above-mentioned covariates. Our findings were consistent with the hypothesis that more generous paid leave policies have the potential to improve DTP immunization coverage. Further work is needed to understand the health effects of paid leave policies in LMICs. © 2015 Elsevier Ltd.
Hajna S, Ross N, Joseph L, Harper S, Dasgupta K. “Neighbourhood Walkability, Daily Steps and Utilitarian Walking in Canadian Adults.” BMJ Open. 2015;5(11)
Objectives: To estimate the associations of neighbourhood walkability (based on Geographic Information System (GIS)-derived measures of street connectivity, land use mix, and population density and the Walk Score) with self-reported utilitarian walking and accelerometer-assessed daily steps in Canadian adults. Design: A cross-sectional analysis of data collected as part of the Canadian Health Measures Survey (2007-2009). Setting: Home neighbourhoods (500 m polygonal street network buffers around the centroid of the participant's postal code) located in Atlantic Canada, Québec, Ontario, the Prairies and British Columbia. Participants: 5605 individuals participated in the survey. 3727 adults (≥18 years) completed a computer-assisted interview and attended a mobile clinic assessment. Analyses were based on those who had complete exposure, outcome and covariate data (n = 2949). Main exposure measures: GIS-derived walkability (based on land use mix, street connectivity and population density); Walk Score. Main outcome measures: Self-reported utilitarian walking; accelerometer-assessed daily steps. Results: No important relationship was observed between neighbourhood walkability and daily steps. Participants who reported more utilitarian walking, however, accumulated more steps ({$<$}1 h/week: 6613 steps/day, 95\% CI 6251 to 6975; 1 to 5 h/week: 6768 steps/day, 95\% CI 6420 to 7117; ≥6 h/week: 7391 steps/day, 95\% CI 6972 to 7811). There was a positive graded association between walkability and odds of walking ≥1 h/week for utilitarian purposes (eg, Q4 vs Q1 of GIS-derived walkability: OR = 1.66, 95\% CI 1.31 to 2.11; Q3 vs Q1: OR = 1.41, 95\% CI 1.14 to 1.76; Q2 vs Q1: OR = 1.13, 95\% CI 0.91 to 1.39) independent of age, sex, body mass index, married/common law status, annual household income, having children in the household, immigrant status, mood disorder, perceived health, ever smoker and season. Conclusions: Contrary to expectations, living in more walkable Canadian neighbourhoods was not associated with more total walking. Utilitarian walking and daily steps were, however, correlated and walkability demonstrated a positive graded relationship with utilitarian walking.
Harper S. “Invited Commentary: A-P-C … It’s Easy as 1-2-3!.” American Journal of Epidemiology. 2015;182(4):313–317
Investigations of age, period, and cohort effects are difficult because the 3 factors are linearly dependent. In a novel application, Kramer et al. (Am J Epidemiol. 2015;182(4):302-312) have used graphical analysis and statistical models to estimate the impact that age, period, and cohort effects have had on trends in black-white inequalities in heart disease mortality. Using a constrained regression approach (with the first 2 periods' effects constrained to zero), Kramer et al. find evidence that age and cohort effects figure more prominently than do period effects in contributing to relative black-white mortality differences, and they argue that early-life exposures should be given greater consideration for mitigation of racial differences in heart disease. In this invited commentary, I argue that the utility of age-period-cohort models for understanding health inequalities depends on the plausibility of the assumptions used to break the link between the 3 factors. Based on the existing age-period-cohort literature, alternative assumptions seem likely to produce substantially different results. I also argue that interpretations of the impacts of age, period, and cohort effects on racial inequalities in heart disease mortality may depend on whether inequalities are assessed on the absolute scale or the relative scale. © The Author 2015. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved.
Harper S, Charters T, Strumpf E, Galea S, Nandi A. “Economic Downturns and Suicide Mortality in the USA, 1980-2010: Observational Study.” International Journal of Epidemiology. 2015;44(3):956–966
Background: Several studies have suggested strong associations between economic downturns and suicide mortality, but are at risk of bias due to unmeasured confounding. The rationale for our study was to provide more robust evidence by using a quasi-experimental design.Methods: We analysed 955561 suicides occurring in the USA from 1980 to 2010 and used a broad index of economic activity in each US state to measure economic conditions. We used a quasi-experimental, fixed-effects design and we also assessed whether the effects were heterogeneous by demographic group and during periods of official recession.Results: After accounting for secular trends, seasonality and unmeasured fixed characteristics of states, we found that an economic downturn similar in magnitude to the 2007 Great Recession increased suicide mortality by 0.14 deaths per 100 000 population [95\% confidence interval (CI) 0.00, 0.28] or around 350 deaths. Effects were stronger for men (0.28, 95\% CI 0.07, 0.49) than women and for those with less than 12 years of education (1.22 95\% CI 0.83, 1.60) compared with more than 12 years of education. The overall effect did not differ for recessionary (0.11, 95\% CI -0.02, 0.25) vs non-recessionary periods (0.15, 95\% CI 0.01, 0.29). The main study limitation is the potential for misclassified death certificates and we cannot definitively rule out unmeasured confounding.Conclusions: We found limited evidence of a strong, population-wide detrimental effect of economic downturns on suicide mortality. The overall effect hides considerable heterogeneity by gender, socioeconomic position and time period. © The Author 2015; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association.
Harper S, Charters T, Strumpf E. “Trends in Socioeconomic Inequalities in Motor Vehicle Accident Deaths in the United States, 1995-2010.” American Journal of Epidemiology. 2015;182(7):606–614
Motor vehicle accident (MVA) mortality has been declining overall, but little is known about trends by socioeconomic position. We examined trends in education-related inequalities in US MVA death rates from 1995 to 2010. We used mortality data from the National Center for Health Statistics and population estimates from the Current Population Survey, and we calculated vehicle- and person-miles traveled using data from the National Household Travel Survey. We used negative binomial regression to estimate crude and age-, sex-, and race-adjusted mortality rates among adults aged 25 years or more. We found larger mortality decreases among the more highly educated and some evidence of mortality increases among the least educated. Adjusted death rates were 15.3 per 100,000 population (95\% confidence interval (CI): 10.7, 19.9) higher at the bottom of the education distribution than at the top of the education distribution in 1995, increasing to 17.9 per 100,000 population (95\% CI: 14.8, 21.0) by 2010. In relative terms, adjusted death rates were 2.4 (95\% CI: 1.7, 3.0) times higher at the bottom of the education distribution than at the top in 1995, increasing to 4.3 times higher (95\% CI: 3.4, 5.3) by 2010. Inequality increases were larger in terms of vehicle-miles traveled. Although overall MVA death rates declined during this period, socioeconomic differences in MVA mortality have persisted or worsened over time. © The Author 2015. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health.
Hutcheon J, Harper S, Strumpf E, Lee L, Marquette G. “Using Inter-Institutional Practice Variation to Understand the Risks and Benefits of Routine Labour Induction at 41+0 Weeks.” BJOG: An International Journal of Obstetrics and Gynaecology. 2015;122(7):973–981
Objective To evaluate the risks and benefits of routine labour induction at 41+0 weeks' gestation for mother and newborn. Design Population-based retrospective cohort study of inter-institutional variation in labour induction practices for women at or beyond 41+0 weeks' gestation. Population Women in British Columbia, Canada, who remained pregnant ≥41+0 weeks and delivered at one of the province's 42 hospitals with {$>$}50 annual deliveries, 2008-2012 (n = 14 627). Methods The proportion of women remaining pregnant a week or more past the expected delivery date who were induced at 41+0 or 41+1 weeks' gestation for an indication of 'post-dates' was calculated for each institution. We used instrumental variable analysis (using the institutional rate of labour induction at 41+0 weeks as the instrument) to estimate the effect of labour induction on maternal and neonatal health outcomes. Main outcome measures Caesarean delivery, instrumental delivery, post-partum haemorrhage, 3rd or 4th degree lacerations, macrosomia, neonatal intensive care unit admission, and 5-minute Apgar score {$<$}7. Results Institutional rates of labour induction at 41+0 weeks ranged from 14.3 to 46\%. Institutions with higher (≥30\%) and average (20-29.9\%) induction rates did not have significantly different rates of caesarean delivery, instrumental delivery, or other maternal or neonatal outcomes than institutions with lower induction rates ({$<$}20\%). Instrumental variable analyses also demonstrated no significantly increased (or decreased) risk of caesarean delivery (0.69 excess cases per 100 pregnancies [95\% CI -10.1, 11.5]), instrumental delivery (8.9 per 100 [95\% CI -2.3, 20.2]), or other maternal or neonatal outcomes in women who were induced (versus not induced). Conclusions Within the current range of clinical practice, there was no evidence that differential use of routine induction at 41+0 weeks affected maternal or neonatal health outcomes. © 2014 Royal College of Obstetricians and Gynaecologists.
Hutcheon J, Strumpf E, Harper S, Giesbrecht E. “Maternal and Neonatal Outcomes after Implementation of a Hospital Policy to Limit Low-Risk Planned Caesarean Deliveries before 39 Weeks of Gestation: An Interrupted Time-Series Analysis.” BJOG: An International Journal of Obstetrics and Gynaecology. 2015;122(9):1200–1206
Objective To evaluate the extent to which implementing a hospital policy to limit planned caesarean deliveries before 39 weeks of gestation improved neonatal health, maternal health, and healthcare costs. Design Retrospective cohort study. Setting British Columbia Women's Hospital, Vancouver, Canada, in the period 2005-2012. Population Women with a low-risk planned repeat caesarean delivery. Methods An interrupted time series design was used to evaluate the policy to limit planned caesarean deliveries before 39 weeks of gestation, introduced on 1 April 2008. Main outcome measures Composite adverse neonatal health outcome (respiratory morbidity, 5-minute Apgar score of {$<$}7, neonatal intensive care unit admission, mortality), postpartum haemorrhage, obstetrical wound infection, out-of-hour deliveries, length of stay, and healthcare costs. Results Between 2005 and 2008, 60\% (1204/2021) of low-risk planned caesarean deliveries were performed before 39 weeks of gestation. After the introduction of the policy, the proportion of planned caesareans dropped by 20 percentage points (adjusted risk difference of 20 fewer cases per 100 deliveries; 95\% CI -25.8, -14.3) to 41\% (1033/2518). The policy had no detectable impact on adverse neonatal outcomes (2.2 excess cases per 100; 95\% CI -0.4, 4.8), maternal complications, or healthcare costs, but increased the risk of out-of-hours delivery from 16.2 to 21.1\% (adjusted risk difference 6.3 per 100; 95\% CI 1.6, 10.9). Conclusions We found little evidence that a hospital policy to limit planned caesareans before 39 weeks of gestation reduced adverse neonatal outcomes. Hospital administrators intending to introduce such policies should anticipate, and plan for, modest increases in out-of-hours and emergency-timing. Tweetable abstract Implementing a policy to limit planned caesareans before 39 weeks of gestation did not improve newborn health. © 2015 Royal College of Obstetricians and Gynaecologists.
McKinnon B, Harper S, Kaufman JS, Bergevin Y. “Removing User Fees for Facility-Based Delivery Services: A Difference-in-Differences Evaluation from Ten Sub-Saharan African Countries.” Health policy and planning. 2015;30(4):432–41
BACKGROUND: Several countries in sub-Saharan Africa have recently adopted policies that remove user fees for facility-based delivery services. There is little rigorous evidence of the impact of these policies on utilization of delivery services and no evaluations have examined effects on neonatal mortality rates (NMR). In this article, we estimate the causal effect of removing user fees on the proportion of births delivered in facilities, the proportion of births delivered by Caesarean section, and NMR. METHODS: We used data from Demographic and Health Surveys conducted in 10 African countries between 1997 and 2012. Kenya, Ghana and Senegal adopted policies removing user fees for facility-based deliveries between 2003 and 2007, while seven other countries not changing user fee policies were used as controls. We used a difference-in-differences (DD) regression approach to control for secular trends in the outcomes that are common across countries and for time invariant differences between countries. RESULTS: According to covariate-adjusted DD models, the policy change was consistent with an increase of 3.1 facility-based deliveries per 100 live births (95\% confidence interval (CI): 0.9, 5.2) and an estimated reduction of 2.9 neonatal deaths per 1000 births (95\% CI: -6.8, 1.0). In relative terms, this corresponds to a 5\% increase in facility deliveries and a 9\% reduction in NMR. There was no evidence of an increase in Caesarean deliveries. We examined lead and lag-time effects, finding evidence that facility deliveries continued to increase following fee removal. CONCLUSIONS: Our findings suggest removing user fees increased facility-based deliveries and possibly contributed to a reduction in NMR. Evidence from this evaluation may be useful to governments weighing the potential benefits of removing user fees.
McKinnon B, Harper S, Kaufman JS. “Who Benefits from Removing User Fees for Facility-Based Delivery Services? Evidence on Socioeconomic Differences from Ghana, Senegal and Sierra Leone.” Social science & medicine (1982). 2015;135:117–23
Coverage of skilled delivery care has been increasing across most low-income countries; however, it remains far from universal and is very unequally distributed according to socioeconomic position. In an effort to increase coverage of skilled delivery care and reduce socioeconomic inequalities, governments of several countries in sub-Saharan Africa have recently adopted policies that remove user fees for facility-based delivery services. There is little rigorous evidence of the impact of these policies and few studies have examined effects on socioeconomic inequalities. This study investigates the impact of recent delivery fee exemption policies in Ghana, Senegal, and Sierra Leone on socioeconomic differences in the use of facility-based delivery services. Using Demographic and Health Survey data from nine sub-Saharan African countries, we evaluated the user fee policy changes using a difference-in-differences approach that accounts for underlying common secular trends and time invariant differences among countries, and allows for differential effects of the policy by socioeconomic position. Removing user fees was consistent with meaningful increases in facility deliveries across all categories of household wealth and maternal education. We found little evidence of differential effects of removing user fees across quartiles of household wealth, with increases of 5.4 facility deliveries per hundred live births (95\% CI: 2.1, 8.8) among women in the poorest quartile and 6.8 per hundred live births (95\% CI: 4.0, 9.7) for women in the richest quartile. However, our results suggest that educated women benefited more from removing user fees compared to women with no education. For women with at least some secondary education, the estimated effect was 8.6 facility deliveries per hundred live births (95\% CI: 5.4, 11.9), but only 4.6 per hundred live births (95\% CI: 2.2, 7.0) for women with no education (heterogeneity p-value = 0.04). Thus, while removing fees at the point of service increased facility deliveries across the socioeconomic gradient, it did not reduce inequalities defined by household wealth and may have contributed to a widening of educational inequalities. These findings emphasize the need for concerted efforts to address financial and other barriers that contribute to large and persistent socioeconomic inequalities in delivery care.
Richardson R, Charters T, King N, Harper S. “Trends in Educational Inequalities in Drug Poisoning Mortality: United States, 1994-2010.” American Journal of Public Health. 2015;105(9):1859–1865
Objectives. We estimated trends in drug poisoning death rates by educational attainment and investigated educational inequalities in drug poisoning mortality by race, gender, and region. Methods. We linked drug poisoning death counts from the National Vital Statistics System to population denominators from the Current Population Survey to estimate drug poisoning rates by gender, race, region, and educational attainment (less than high school degree, high school degree, some college, college degree) from 1994 to 2010. Results. There were 372 485 drug poisoning deaths. Education-related inequalities increased during the study among all demographic groups and varied by region. Absolute increases in educational inequalities were higher among Whites than Blacks and men than women. The age-adjusted rate difference between White men with less than a high school degree increased from 8.7 per 100 000 in 1994 to 27.4 in 2010 (change = 18.7). Among Black men, the corresponding increases were 11.7 and 18.3, respectively (change = 6.6). Conclusions. We found strong educational patterning in drug poisoning rates, chiefly by region and race. Rates are highest and increasing the fastest among groups with less education.
Auger N, Feuillet P, Martel S, Lo E, Barry A, Harper S. “Mortality Inequality in Populations with Equal Life Expectancy: Arriaga’s Decomposition Method in SAS, Stata, and Excel.” Annals of Epidemiology. 2014;24(8):575-580.e1
Purpose: Life expectancy is used to measure population health, but large differences in mortality can be masked even when there is no life expectancy gap. We demonstrate how Arriaga's decomposition method can be used to assess inequality in mortality between populations with near equal life expectancy. Methods: We calculated life expectancy at birth for Quebec and the rest of Canada from 2005 to 2009 using life tables and partitioned the gap between both populations into age and cause-specific components using Arriaga's method. Results: The life expectancy gap between Quebec and Canada was negligible ({$<$}0.1years). Decomposition of the gap showed that higher lung cancer mortality in Quebec was offset by cardiovascular mortality in the rest of Canada, resulting in identical life expectancy in both groups. Lung cancer in Quebec had a greater impact at early ages, whereas cardiovascular mortality in Canada had a greater impact at older ages. Conclusions: Despite the absence of a gap, we demonstrate using decomposition analyses how lung cancer at early ages lowered life expectancy in Quebec, whereas cardiovascular causes at older ages lowered life expectancy in Canada. We provide SAS/Stata code and an Excel spreadsheeet to facilitate application of Arriaga's method to other settings. © 2014 Elsevier Inc.
Frenz P, Delgado I, Kaufman J, Harper S. “Achieving Effective Universal Health Coverage with Equity: Evidence from Chile.” Health Policy and Planning. 2014;29(6):717–731
Chile's 'health guarantees' approach to providing universal and equitable coverage for quality healthcare in a dual public-private health system has generated global interest. The programme, called AUGE, defines legally enforceable rights to explicit healthcare benefits for priority health conditions, which incrementally covered 56 problems representing 75\% of the disease burden between 2005 and 2009. It was accompanied by other health reform measures to increase public financing and public sector planning to secure the guarantees nationwide, as well as the state's stewardship role. We analysed data from household surveys conducted before and after the AUGE reform to estimate changes in levels of unmet health need, defined as the lack of a healthcare visit for a health problem occurring in the last 30 days, by age, sex, income, education, health insurance, residence and ethnicity; fitting logistic regression models and using predictive margins. The overall prevalence of unmet health need was much lower in 2009 (17.6\%, 95\% CI: 16.5\%, 18.6\%) than in 2000 (30.0\%, 95\% CI: 28.3\%, 31.7\%). Differences by income and education extremes and rural-urban residence disappeared. In 2009, people who had been in treatment for a condition covered by AUGE in the past year had a lower adjusted prevalence of unmet need for their recent problem (11.7\%, 95\% CI: 10.5\%, 13.2\%) than who had not (21.0\%, 95\% CI: 19.6\%, 22.4\%). Despite limitations including cross-sectional and self-reported data, our findings suggest that the Chilean health system has become more equitable and responsive to need. While these changes cannot be directly attributed to AUGE, they were coincident with the AUGE reforms. However, healthcare equity concerns are still present, relating to quality of care, health system barriers and differential access for health conditions that are not covered by AUGE. © The Author 2013; all rights reserved.
Harper S, MacLehose RF, Kaufman JS. “Trends In The Black-White Life Expectancy Gap Among US States, 1990–2009.” Health Affairs. 2014;33(8):1375–1382
Nationwide differences in US life expectancy trends for blacks and whites may mask considerable differences by state that are relevant to policies aimed at reducing health inequalities. We calculated annual state-specific life expectancies for blacks and whites from 1990 to 2009 using age-specific mortality counts and census-based denominators. Nationally, the black-white difference in life expectancy at birth shrank during the period by 2.7 years for males (from 8.1 to 5.4 years) and by 1.7 years for females (from 5.5 to 3.8 years). We found considerable variation across states in both the magnitude of the life expectancy gap (approximately fifteen years) and the change during the past two decades (about six years). Decomposition analysis showed that New York made the most profound contribution to reducing the gap, but less favorable trends in a number of states, notably California and Texas, kept the gap from shrinking further. Large state variations in the pace of change in the racial gap in life expectancy suggest that state-specific determinants merit further investigation. Life. © 2014 by Project HOPE - The People-to-People Health Foundation.
Harper S, King N, Meersman S, Reichman M, Breen N, Lynch J. “Implicit Value Judgments in the Measurement of Health Inequalities.” Revista Panamericana de Salud Publica/Pan American Journal of Public Health. 2014;35(4):293–304
Context. Quantitative estimates of the magnitude, direction, and rate of change of health inequalities play a crucial role in creating and assessing policies aimed at eliminating the disproportionate burden of disease in disadvantaged populations. It is generally assumed that the measurement of health inequalities is a value-neutral process, providing objective data that are then interpreted using normative judgments about whether a particular distribution of health is just, fair, or socially acceptable. Methods. We discuss five examples in which normative judgments play a role in the measurement process itself, through either the selection of one measurement strategy to the exclusion of others or the selection of the type, significance, or weight assigned to the variables being measured. Findings. Overall, we find that many commonly used measures of inequality are value laden and that the normative judgments implicit in these measures have important consequences for interpreting and responding to health inequalities. Conclusions. Because values implicit in the generation of health inequality measures may lead to radically different interpretations of the same underlying data,we urge researchers to explicitly consider and transparently discuss the normative judgments underlying their measures. We also urge policymakers and other consumers of health inequalities data to pay close attention to the measures on which they base their assessments of current and future health policies.
Harper S, Strumpf E, Burris S, Smith G, Lynch J. “The Effect of Mandatory Seat Belt Laws on Seat Belt Use by Socioeconomic Position.” Journal of Policy Analysis and Management. 2014;33(1):141–161
We investigated the differential effect of mandatory seat belt laws on seat belt use among socioeconomic subgroups. We identified the differential effect of legislation across higher versus lower education individuals using a difference-in-differences model based on state variations in the timing of the passage of laws. We find strong effects of mandatory seat belt laws for all education groups, but the effect is stronger for those with fewer years of education. In addition, we find that the differential effect by education is larger for mandatory seat belt laws with primary rather than secondary enforcement. Our results imply that existing socioeconomic differences in seat belt use would be further mitigated if all states upgraded to primary enforcement. © 2013 by the Association for Public Policy Analysis and Management.
King N, Fraser V, Boikos C, Richardson R, Harper S. “Determinants of Increased Opioid-Related Mortality in the United States and Canada, 1990-2013: A Systematic Review.” American Journal of Public Health. 2014;104(8):e32-e42
We review evidence of determinants contributing to increased opioid-related mortality in the United States and Canada between 1990 and 2013. We identified 17 determinants of opioid-relatedmortality andmortality increases that we classified into 3 categories: prescriber behavior, user behavior and characteristics, and environmental and systemic determinants. These determinants operate independently but interact in complex ways that vary according to geography and population,making generalization from single studies inadvisable. Researchers in this area face significant methodological difficulties; most of the studies in our review were ecological or observational and lacked control groups or adjustment for confounding factors; thus, causal inferences are difficult. Preventing additional opioid-related mortality will likely require interventions that address multiple determinants and are tailored to specific locations and populations.
Manuel D, Ho T, Harper S, Anderson G, Lynch J, Rosella L. “Modelling Preventive Effectiveness to Estimate the Equity Tipping Point: At What Coverage Can Individual Preventive Interventions Reduce Socioeconomic Disparities in Diabetes Risk?.” Chronic Diseases and Injuries in Canada. 2014;34(2–3):94–102
Introduction: Most individual preventive therapies potentially narrow or widen health disparities depending on the difference in community effectiveness across socioeconomic position (SEP). The equity tipping point-defined as the point at which health disparities become larger-can be calculated by varying components of community effectiveness such as baseline risk of disease, intervention coverage and/ or intervention efficacy across SEP. Methods: We used a simple modelling approach to estimate the community effectiveness of diabetes prevention across SEP in Canada under different scenarios of intervention coverage. Results: Five-year baseline diabetes risk differed between the lowest and highest income groups by 1.76\%. Assuming complete coverage across all income groups, the difference was reduced to 0.90\% (144 000 cases prevented) with lifestyle interventions and 1.24\% (88 100 cases prevented) with pharmacotherapy. The equity tipping point was estimated to be a coverage difference of 30\% for preventive interventions (100\% and 70\% coverage among the highest and lowest income earners, respectively). Conclusion: Disparities in diabetes risk could be measurably reduced if existing interventions were equally adopted across SEP. However, disparities in coverage could lead to increased inequity in risk. Simple modelling approaches can be used to examine the community effectiveness of individual preventive interventions and their potential to reduce (or increase) disparities. The equity tipping point can be used as a critical threshold for disparities analyses.
McKinnon B, Harper S, Kaufman JS, Abdullah M. “Distance to Emergency Obstetric Services and Early Neonatal Mortality in Ethiopia.” Tropical Medicine & International Health. 2014;19(7):780–790
Objectives To assess the effect of distance to emergency obstetric and newborn care (EmONC) services on early neonatal mortality in rural Ethiopia and examine whether proximity to services contributes to socio-economic inequalities in early neonatal mortality. Methods We linked data from the 2011 Ethiopian Demographic and Health Survey with facility data from the 2008 Ethiopian National EmONC Needs Assessment based on geographical coordinates collected in both surveys. Health facilities were classified based on the performance of nine EmONC signal functions (e.g. neonatal resuscitation, Caesarean section). We used multivariable logistic regression to assess the relationship between distance to services and early neonatal mortality. A decomposition approach was used to quantify the relative contributions of distance to EmONC services and other determinants to overall and socio-economic inequality in early neonatal mortality. Results In general, closer proximity to EmONC services and higher level of care were associated with lower early neonatal mortality. Living more than 80 km from the nearest comprehensive EmONC facility able to perform all nine signal functions compared to living within 10 km was associated with an increase of 14.4 early neonatal deaths per 1000 live births (95\% CI: 0.1, 28.7). Closer proximity to a substandard EmONC facility compared with no facility was not associated with lower early neonatal mortality. Distance to EmONC services was an important determinant of early neonatal mortality, although it did not make a significant contribution to explaining socio-economic inequality. Conclusions Our results suggest that recent initiatives by the Ethiopian government to improve geographical access to EmONC services have the potential to reduce early neonatal mortality but may not affect inequalities.
McKinnon B, Harper S, Kaufman J, Bergevin Y. “Socioeconomic Inequality in Neonatal Mortality in Countries of Low and Middle Income: A Multicountry Analysis.” The Lancet Global Health. 2014;2(3):e165-e173
Background: Neonatal mortality rates (NMRs) in countries of low and middle income have been only slowly decreasing; coverage of essential maternal and newborn health services needs to increase, particularly for disadvantaged populations. Our aim was to produce comparable estimates of changes in socioeconomic inequalities in NMR in the past two decades across these countries. Methods: We used data from Demographic and Health Surveys (DHS) for countries in which a survey was done in 2008 or later and one about 10 years previously. We measured absolute inequalities with the slope index of inequality and relative inequalities with the relative index of inequality. We used an asset-based wealth index and maternal education as measures of socioeconomic position and summarised inequality estimates for all included countries with random-effects meta-analysis. Findings: 24 low-income and middle-income countries were eligible for inclusion. In most countries, absolute and relative wealth-related and educational inequalities in NMR decreased between survey 1 and survey 2. In five countries (Cameroon, Nigeria, Malawi, Mozambique, and Uganda), the difference in NMR between the top and bottom of the wealth distribution was reduced by more than two neonatal deaths per 1000 livebirths per year. By contrast, wealth-related inequality increased by more than 1·5 neonatal deaths per 1000 livebirths per year in Ethiopia and Cambodia. Patterns of change in absolute and relative educational inequalities in NMR were similar to those of wealth-related NMR inequalities, although the size of educational inequalities tended to be slightly larger. Interpretation: Socioeconomic inequality in NMR seems to have decreased in the past two decades in most countries of low and middle income. However, a substantial survival advantage remains for babies born into wealthier households with a high educational level, which should be considered in global efforts to further reduce NMR. Funding: Canadian Institutes of Health Research. © 2014 McKinnon et al.
Smith B, Smith P, Harper S, Manuel D, Mustard C. “Reducing Social Inequalities in Health: The Role of Simulation Modelling in Chronic Disease Epidemiology to Evaluate the Impact of Population Health Interventions.” Journal of Epidemiology and Community Health. 2014;68(4):384–389
Reducing health inequalities has become a major public health priority internationally. However, how best to achieve this goal is not well understood. Population health intervention research has the potential to address some of this knowledge gap. This review argues that simulation studies can produce unique evidence to build the population health intervention research evidence base on reducing social inequalities in health. To this effect, the advantages of using simulation models over other population health intervention research methods are discussed. Key questions regarding the potential challenges of developing simulation models to investigate population health intervention research on reducing social inequalities in health and the types of population health intervention research questions that can be answered using this methodology are reviewed. We use the example of social inequalities in coronary heart disease to illustrate how simulation models can elucidate the effectiveness of a number of 'what-if' counterfactual population health interventions on reducing social inequalities in coronary heart disease. Simulation models are a flexible, cost-effective, evidencebased research method with the capacity to inform public health policy-makers regarding the implementation of population health interventions to reduce social inequalities in health.
Auger N, Harper S, Barry A. “Diverging Socioeconomic Inequality in Life Expectancy of Francophones and Anglophones in Montréal, Québec: Tobacco to Blame?.” Journal of Public Health (Germany). 2013;21(4):317–324
Aim: We evaluated the ages and causes of death contributing to life expectancy gaps between economically advantaged and disadvantaged Francophones and Anglophones of Montréal, a Canadian metropolitan centre. Subject and Methods: We partitioned the life expectancy gap at birth between socioeconomically disadvantaged and advantaged Francophones and Anglophones of Montréal (Québec) into age and cause of death components for two periods (1989-1993, 2002-2006). Changes in the contributions of causes over time were evaluated. Results: Life expectancy was lower for disadvantaged Francophones and Anglophones by 5 years in men and 1.6 years in women compared with advantaged individuals. Over time, the socioeconomic gap widened for Francophones (men 0.3 years, women 2.8 years), due to smaller reductions in mortality from tobacco-related causes (cardiovascular, cancer, respiratory) in disadvantaged than in advantaged Francophones, especially after age ≥65 years (except lung cancer mortality that increased, particularly in disadvantaged women). The socioeconomic gap narrowed, however, for Anglophones (men 1.0 year, women 0.6 years), due to greater reductions in cardiovascular mortality in disadvantaged than advantaged Anglophones. Conclusion: Socioeconomic inequalities in life expectancy decreased for Anglophones but increased for Francophones in Montréal due to underlying trends in tobacco-related mortality. Despite strong tobacco control laws in Canada, socioeconomic inequality in tobacco-related mortality is widening for Francophones in Montréal. © 2013 Springer-Verlag Berlin Heidelberg.
Harper S, King N. “Commentary: Best Practice for What?.” Milbank Quarterly. 2013;91(1):205–209
Harper S, King N, Young M. “Impact of Selective Evidence Presentation on Judgments of Health Inequality Trends: An Experimental Study.” PLoS ONE. 2013;8(5)
Reducing health inequalities is a key objective for many governments and public health organizations. Whether inequalities are measured on the absolute (difference) or relative (ratio) scale can have a significant impact on judgments about whether health inequalities are increasing or decreasing, but both of these measures are not often presented in empirical studies. In this study we investigated the impact of selective presentation of health inequality measures on judgments of health inequality trends among 40 university undergraduates. We randomized participants to see either a difference or ratio measure of health inequality alongside raw mortality rates in 5 different scenarios. At baseline there were no differences between treatment groups in assessments of inequality trends, but selective exposure to the same raw data augmented with ratio versus difference inequality graphs altered participants' assessments of inequality change. When absolute inequality decreased and relative inequality increased, exposure to ratio measures increased the probability of concluding that inequality had increased from 32.5\% to 70\%, but exposure to difference measures did not (35\% vs. 25\%). Selective exposure to ratio versus difference inequality graphs thus increased the difference between groups in concluding that inequality had increased from 2.5\% (95\% CI -9.5\% to 14.5\%) to 45\% (95\% CI 29.4 to 60.6). A similar pattern was evident for other scenarios where absolute and relative inequality trends gave conflicting results. In cases where measures of absolute and relative inequality both increased or both decreased, we did not find any evidence that assignment to ratio vs. difference graphs had an impact on assessments of inequality change. Selective reporting of measures of health inequality has the potential to create biased judgments of progress in ameliorating health inequalities. © 2013 Harper et al.
Harper S, Ruder E, Roman H, Geggel A, Nweke O, Payne-Sturges D, Levy J. “Using Inequality Measures to Incorporate Environmental Justice into Regulatory Analyses.” International Journal of Environmental Research and Public Health. 2013;10(9):4039–4059
Formally evaluating how specific policy measures influence environmental justice is challenging, especially in the context of regulatory analyses in which quantitative comparisons are the norm. However, there is a large literature on developing and applying quantitative measures of health inequality in other settings, and these measures may be applicable to environmental regulatory analyses. In this paper, we provide information to assist policy decision makers in determining the viability of using measures of health inequality in the context of environmental regulatory analyses. We conclude that quantification of the distribution of inequalities in health outcomes across social groups of concern, considering both within-group and between-group comparisons, would be consistent with both the structure of regulatory analysis and the core definition of environmental justice. Appropriate application of inequality indicators requires thorough characterization of the baseline distribution of exposures and risks, leveraging data generally available within regulatory analyses. Multiple inequality indicators may be applicable to regulatory analyses, and the choice among indicators should be based on explicit value judgments regarding the dimensions of environmental justice of greatest interest. © 2013 by the authors; licensee MDPI, Basel, Switzerland.
Kaufman J, Harper S. “Health Equity: Utopian and Scientific.” Preventive Medicine. 2013;57(6):739–740
King N, Harper S, Young M. “Who Cares about Health Inequalities? Cross-country Evidence from the World Health Survey.” Health Policy and Planning. 2013;28(5):558–571
Reduction of health inequalities within and between countries is a global health priority, but little is known about the determinants of popular support for this goal. We used data from the World Health Survey to assess individual preferences for prioritizing reductions in health and health care inequalities. We used descriptive tables and regression analysis to study the determinants of preferences for reducing health inequalities as the primary health system goal. Determinants included individual socio-demographic characteristics (age, sex, urban residence, education, marital status, household income, self-rated health, health care use, satisfaction with health care system) and country-level characteristics [gross domestic product (GDP) per capita, disability-free life expectancy, equality in child mortality, income inequality, health and public health expenditures]. We used logistic regression to assess the likelihood that individuals ranked minimizing inequalities first, and rank-ordered logistic regression to compare the ranking of other priorities against minimizing health inequalities. Individuals tended to prioritize health system goals related to overall improvement (improving population health and health care responsiveness) over those related to equality and fairness (minimizing inequalities in health and responsiveness, and promoting fairness of financial contribution). Individuals in countries with higher GDP per capita, life expectancy, and equality in child mortality were more likely to prioritize minimizing health inequalities. © The Author 2012; all rights reserved.
Maika A, Mittinty MN, Brinkman S, Harper S, Satriawan E, Lynch JW. “Changes in Socioeconomic Inequality in Indonesian Children’s Cognitive Function from 2000 to 2007: A Decomposition Analysis.” PLOS ONE. 2013;8(10):e78809
Background Measuring social inequalities in health is common; however, research examining inequalities in child cognitive function is more limited. We investigated household expenditure-related inequality in children’s cognitive function in Indonesia in 2000 and 2007, the contributors to inequality in both time periods, and changes in the contributors to cognitive function inequalities between the periods. Methods Data from the 2000 and 2007 round of the Indonesian Family Life Survey (IFLS) were used. Study participants were children aged 7–14 years (n = 6179 and n = 6680 in 2000 and 2007, respectively). The relative concentration index (RCI) was used to measure the magnitude of inequality. Contribution of various contributors to inequality was estimated by decomposing the concentration index in 2000 and 2007. Oaxaca-type decomposition was used to estimate changes in contributors to inequality between 2000 and 2007. Results Expenditure inequality decreased by 45\% from an RCI = 0.29 (95\% CI 0.22 to 0.36) in 2000 to 0.16 (95\% CI 0.13 to 0.20) in 2007 but the burden of poorer cognitive function was higher among the disadvantaged in both years. The largest contributors to inequality in child cognitive function were inequalities in per capita expenditure, use of improved sanitation and maternal high school attendance. Changes in maternal high school participation (27\%), use of improved sanitation (25\%) and per capita expenditures (18\%) were largely responsible for the decreasing inequality in children’s cognitive function between 2000 and 2007. Conclusions Government policy to increase basic education coverage for women along with economic growth may have influenced gains in children’s cognitive function and reductions in inequalities in Indonesia.
McKinnon B, Harper S, Moore S. “The Relationship of Living Arrangements and Depressive Symptoms among Older Adults in Sub-Saharan Africa.” BMC Public Health. 2013;13(1)
Background: Older adults in sub-Saharan Africa are increasingly facing the twin challenges of reduced support from their adult children and taking on new roles caring for orphans and vulnerable children. How these changes affect the mental health of older adults is largely unknown. Methods. We use data from the 2002-2003 World Health Surveys for 15 countries in sub-Saharan Africa to examine whether older adults who may be lacking adequate support through living alone or in skipped-generation households are at an increased risk of depressive symptoms compared to those living with at least one working-age adult. Using meta-regression, we also examine whether heterogeneity across countries in the prevalence of depressive symptoms or in the association between living arrangements and depressive symptoms is associated with HIV/AIDS prevalence and national economic status. Results: The pooled prevalence of depressive symptoms among older adults was 9.2\%. Older adults living alone had a 2.3\% point higher predicted prevalence of depressive symptoms compared to individuals living with at least one working-age adult (95\% confidence interval: 0.2\%, 4.4\%). None of the country characteristics examined explained heterogeneity across countries in the relationship between living arrangements and depressive symptoms. However, there was some evidence suggesting a positive association between depressive symptom prevalence and the severity of a country's HIV/AIDS epidemic. Conclusion: As depressive symptoms are known to be predictive of poor quality of life and increased mortality, it is important to address how health and social policies can be put in place to mitigate the potentially detrimental effects of solitary living on the mental health of older persons in sub-Saharan Africa. © 2013 McKinnon et al.; licensee BioMed Central Ltd.
Nandi A, Charters T, Strumpf E, Heymann J, Harper S. “Economic Conditions and Health Behaviours during the ‘Great Recession’.” Journal of Epidemiology and Community Health. 2013;67(12):1038–1046
Background The adoption of healthier behaviours has been hypothesised as a mechanism to explain empirical findings of population health improvements during some economic downturns. Methods We estimated the effect of the local unemployment rate on health behaviours using pooled annual surveys from the 2003-2010 Behavioral Risk Factor Surveillance Surveys, as well as population-based telephone surveys of the US adult general population. Analyses were based on approximately 1 million respondents aged 25 years or older living in 90 Metropolitan Statistical Areas and Metropolitan Divisions (MMSAs). The primary exposure was the quarterly MMSA-specific unemployment rate. Outcomes included alcohol consumption, smoking status, attempts to quit smoking, body mass index, overweight/obesity and pastmonth physical activity or exercise. Results The average unemployment rate across MMSAs increased from a low of 4.5\% in 2007 to a high of 9.3\% in 2010. In multivariable models accounting for individual-level sociodemographic characteristics and MMSA and quarter fixed effects, a one percentage-point increase in the unemployment rate was associated with 0.15 (95\% CI -0.31 to 0.01) fewer drinks consumed in the past month and a 0.14 (95\% CI -0.28 to 0.00) percentage-point decrease in the prevalence of pastmonth heavy drinking; these effects were driven primarily by men. Changes in the unemployment rate were not consistently associated with other health behaviours. Although individual-level unemployment status was associated with higher levels of alcohol consumption, smoking and obesity, the MMSA-level effects of the recession were largely invariant across employment groups. Conclusions Our results do not support the hypothesis that health behaviours mediate the effects of local-area economic conditions on mortality.
Platt R, Harper S. “Survey Data with Sampling Weights: Is There a “Best” Approach?.” Environmental Research. 2013;120:143–144
Speybroeck N, Van Malderen C, Harper S, Müller B, Devleesschauwer B. “Simulation Models for Socioeconomic Inequalities in Health: A Systematic Review.” International Journal of Environmental Research and Public Health. 2013;10(11):5750–5780
Background: The emergence and evolution of socioeconomic inequalities in health involves multiple factors interacting with each other at different levels. Simulation models are suitable for studying such complex and dynamic systems and have the ability to test the impact of policy interventions in silico. Objective: To explore how simulation models were used in the field of socioeconomic inequalities in health. Methods: An electronic search of studies assessing socioeconomic inequalities in health using a simulation model was conducted. Characteristics of the simulation models were extracted and distinct simulation approaches were identified. As an illustration, a simple agent-based model of the emergence of socioeconomic differences in alcohol abuse was developed. Results: We found 61 studies published between 1989 and 2013. Ten different simulation approaches were identified. The agent-based model illustration showed that multilevel, reciprocal and indirect effects of social determinants on health can be modeled flexibly. Discussion and Conclusions: Based on the review, we discuss the utility of using simulation models for studying health inequalities, and refer to good modeling practices for developing such models. The review and the simulation model example suggest that the use of simulation models may enhance the understanding and debate about existing and new socioeconomic inequalities of health frameworks. © 2013 by the authors; licensee MDPI, Basel, Switzerland.
Young M, King N, Harper S, Humphreys K. “The Influence of Popular Media on Perceptions of Personal and Population Risk in Possible Disease Outbreaks.” Health, Risk and Society. 2013;15(1):103–114
Infectious disease outbreaks are uncertain and potentially risky events that often attract significant media attention. Previous research has shown that, regardless of their objective severity, diseases receiving greater coverage in the media are considered to be more serious and more representative of a disease than those receiving less coverage. This study assesses the role of media coverage in estimations of population risk (measured as perceived incidence among a specific population within a 1-year time period) and personal risk (measured as perceived personal likelihood of infection). Diseases with higher media coverage were considered more serious and more representative of a disease, and estimated to have lower incidence, than diseases less frequently found in the media. No difference in estimates of personal risk was found. A significant correlation between estimates of population and personal risk was found for diseases infrequently reported in the media. A weaker correlation between estimates of population and personal risk was found for diseases frequently reported in the media. The correlation remained unchanged when participants were exposed to additional information, including symptoms, mortality and estimates of prevalence. © 2013 Copyright Taylor and Francis Group, LLC.
Auger N, Park A, Harper S, Daniel M, Roncarolo F, Platt R. “Educational Inequalities in Preterm and Term Small-for-Gestational-Age Birth Over Time.” Annals of Epidemiology. 2012;22(3):160–167
Purpose: Time trends in educational inequalities in small-for-gestational-age (SGA) birth are important to evaluate for policy, especially at preterm gestational ages when morbidity and mortality are typically greater. We evaluated educational inequalities in preterm and term SGA birth over time, accounting for potential bias at preterm gestational ages. Methods: Data included 2,204,056 singleton live births from 25 to 43 gestational weeks, 1981 to 2007. We estimated prevalence ratios (PR) and percent prevalence differences (PPD) of preterm and term SGA birth for a continuous education score, accounting for maternal characteristics. Sensitivity analyses included correction for misclassification of preterm SGA status, and use of fetuses-at-risk denominators in regression models. Results: Although prevalence of SGA birth decreased over time, relative educational inequalities (PRs) persisted for preterm and term cases. PPDs decreased slightly, but more for term than preterm SGA birth. Sensitivity analyses indicated that PRs for education were stronger for preterm than term SGA birth. PPDs were larger for term SGA birth in the first period, but greater for preterm SGA birth in the last period. Conclusions: Relative educational inequalities in SGA birth persisted over time. The difference in prevalence between the least and most educated mothers is currently greater for preterm than for term SGA birth. © 2012 Elsevier Inc.
Auger N, Delézire P, Harper S, Platt R. “Maternal Education and Stillbirth: Estimating Gestational-Age-Specific and Cause-Specific Associations.” Epidemiology. 2012;23(2):247–254
Background: Associations between risk factors and perinatal outcomes may be biased at preterm gestational ages, if preterm delivery behaves as an effect modifier due to other unmeasured factors in the causal pathway. We evaluated whether fetuses-at-risk denominators could be used in regression models instead of conventional denominators to obtain less biased estimates of the association between maternal education and stillbirth at preterm gestational intervals. Methods: Data included 2,143,134 live-born and 8946 stillborn singletons from 1981 through 2006 in Québec, Canada. Odds ratios and 95\% confidence intervals were estimated for the relationship between education and stillbirth according to cause of fetal death, adjusting for maternal age, marital status, home language, parity, and period. We examined associations for 4 gestational intervals ({$<$}28, 28-31, 32-36, and ≥37 completed weeks), using both conventional denominators (ie, preterm live births) and fetuses-at-risk denominators. Results: Stillbirth rates were greater for mothers with fewer years of education at all gestational intervals. Using conventional denominators, low education (relative to high education) was more strongly associated with term than preterm stillbirth and was apparently protective at {$<$}28 weeks. Using fetuses-at-risk denominators, low education was more strongly associated with preterm stillbirth than term stillbirth, even at {$<$}28 weeks. Low education was most strongly associated with diabetic-related stillbirth at ≥28 weeks (odds ratio = 5.04) relative to high education. Conclusions: Low education is associated with stillbirth throughout gestation, especially diabetic-related stillbirth. Use of fetuses-at-risk denominators in regression models can avoid potentially biased estimates obtained with conventional denominators at preterm gestational ages. Copyright © 2012 by Lippincott Williams \& Wilkins.
Auger N, Park A, Harper S. “Francophone and Anglophone Perinatal Health: Temporal and Regional Inequalities in a Canadian Setting, 1981-2008.” International Journal of Public Health. 2012;57(6):925–934
Objectives We evaluated temporal and regional inequalities in adverse birth outcomes between Anglophones and Francophones of a Canadian province. Methods Odds ratios and rate differences in preterm birth (PTB, {$<$}37 gestational weeks) and small-for-gestationalage (SGA) birth were computed for Anglophones relative to Francophones for singleton live births in Québec from 1981 to 2008 (N = 2,292,237), adjusting for maternal characteristics. Trends over time and residential region were evaluated. Results Rates of PTB and SGA birth overall were lower for Anglophones relative to Francophones, but temporal and regional trends varied by outcome. Although PTB rates increased over time, inequalities between Francophones and Anglophones were relatively stable. In contrast, inequalities in SGA birth narrowed over time as Francophone rates declined more than Anglophones. Inequalities in SGA birth favored Anglophones overall, but the gap gradually reversed in Montréal (the largest metropolitan center) to currently favor Francophones. Conclusions PTB and SGA birth rates favored Anglophones over Francophones. The linguistic gap was generally stable over time for PTB, but narrowed or reversed for SGA birth. Language may be used to capture inequalities in perinatal health in countries where different linguistic groups predominate. © 2012 Swiss School of Public Health.
Auger N, Harper S, Barry A, Trempe N, Daniel M. “Life Expectancy Gap between the Francophone Majority and Anglophone Minority of a Canadian Population.” European Journal of Epidemiology. 2012;27(1):27–38
Language is an important determinant of health, but analyses of linguistic inequalities in mortality are scant, especially for Canadian linguistic groups with European roots. We evaluated the life expectancy gap between the Francophone majority and Anglophone minority of Québec, Canada, both over time and across major provincial areas. Arriaga's method was used to estimate the age and cause of death groups contributing to changes in the life expectancy gap at birth between 1989-1993 and 2002-2006, and to evaluate patterns across major provincial areas (metropolitan Montréal, other metropolitan centres, and small cities/rural areas). Life expectancy at birth was greater for Anglophones, but the gap decreased over time by 1.3 years (52\% decline) in men and 0.9 years (47\% decline) in women, due to relatively sharper reductions in Francophone mortality from several causes, except lung cancer which countered reductions in women. The life expectancy gap in 2002-2006 was widest in other metropolitan centres (men 5.1 years, women 3.2 years), narrowest in small cities/rural areas (men 0.8 years, women 0.7 years), and tobacco-related causes were the main contributors. Only young Anglophones {$<$}40 years in small cities/rural areas had mortality higher than Francophones, resulting in a narrower gap in these areas. Differentials in life expectancy favouring Anglophones decreased over time, but varied across areas of Québec. Tobacco-related causes accounted for the majority of the current life expectancy gap. © 2011 Springer Science+Business Media B.V.
Banack H, Harper S, Kaufman J. “Coronary Heart Disease Risk Factors and Mortality [2].” JAMA. 2012;307(11):1137–1138
Elani H, Harper S, Allison P, Bedos C, Kaufman J. “Socio-Economic Inequalities and Oral Health in Canada and the United States.” Journal of Dental Research. 2012;91(9):865–870
This paper describes and compares the magnitude of socio-economic inequalities in oral health among adults in Canada and the US over the past 35 years. We analyzed data from nationally representative examination surveys in Canada and the US: Nutrition Canada National Survey (1970-1972, N = 11,546), Canadian Health Measures Survey (2007-2009, N = 3,508), The First National Health and Nutrition Examination Survey (1971-1974, N = 13,131), and National Health and Nutrition Examination Survey (2007-2008, N = 5,707). Oral health outcomes examined were prevalence of edentulism, proportion of individuals having at least 1 untreated decayed tooth, and proportion of individuals having at least 1 filled tooth. Sociodemographic indicators included in our analysis were place of birth, education, and income. Data were age-adjusted, and survey weights were used to account for the complex survey design in making population inferences. Our findings demonstrate that oral health outcomes have improved for adults in both countries. In the 1970s, Canada had a higher prevalence of edentulism and dental decay and lower prevalence of filled teeth. This was also combined with a more pronounced social inequality gradient among place of birth, education, and income groups. Over time, both countries demonstrated a decline in absolute socio-economic inequalities in oral health. © 2012 International \& American Associations for Dental Research.
Harper S, Strumpf EC, Kaufman JS. “Do Medical Marijuana Laws Increase Marijuana Use? Replication Study and Extension.” Annals of epidemiology. 2012;22(3):207–12
PURPOSE: To replicate a prior study that found greater adolescent marijuana use in states that have passed medical marijuana laws (MMLs), and extend this analysis by accounting for confounding by unmeasured state characteristics and measurement error. METHODS: We obtained state-level estimates of marijuana use from the 2002 through 2009 National Survey on Drug Use and Health. We used 2-sample t-tests and random-effects regression to replicate previous results. We used difference-in-differences regression models to estimate the causal effect of MMLs on marijuana use, and simulations to account for measurement error. RESULTS: We replicated previously published results showing higher marijuana use in states with MMLs. Difference-in-differences estimates suggested that passing MMLs decreased past-month use among adolescents by 0.53 percentage points (95\% confidence interval [CI], 0.03-1.02) and had no discernible effect on the perceived riskiness of monthly use. Models incorporating measurement error in the state estimates of marijuana use yielded little evidence that passing MMLs affects marijuana use. CONCLUSIONS: Accounting for confounding by unmeasured state characteristics and measurement error had an important effect on estimates of the impact of MMLs on marijuana use. We find limited evidence of causal effects of MMLs on measures of reported marijuana use.
Harper S, Strumpf EC. “Commentary: Social Epidemiology Questionable Answers and Answerable Questions.” Epidemiology. 2012;23(6):795–798
Harper S, McKinnon B. “Global Socioeconomic Inequalities in Tobacco Use: Internationally Comparable Estimates from the World Health Surveys..” Cancer causes & control : CCC. 2012;23 Suppl 1:11–25
To produce internationally comparable estimates of socioeconomic differences in tobacco exposure within low and middle-income countries. We used data from 50 countries that participated in the World Health Surveys in 2002-2003. We measured two aspects of smoking: current smoking prevalence and accumulated pack-years of smoking. We used an asset-based approach to estimate permanent income. We measured absolute inequalities, separately by gender, across the entire socioeconomic distribution by using the concentration index and summarized the results and explored heterogeneity by meta-analysis. The overall prevalence of current smoking was highest in Southeast Asia, the Western Pacific, and Europe, and lowest in Africa. Pack-years among current male smokers were highest in Europe. Wealthier men were generally less likely to be current smokers in all regions. However, there was substantial heterogeneity within each region, and in some countries (Georgia, Mexico, Mauritania) current smoking was greater among the more advantaged. Among currently smoking men socioeconomic differences for pack-years of smoking were generally much weaker than for smoking prevalence. Among women the concentration index in current smoking was largest and favored the poor in Europe (1.4, 95\% CI 0.8, 2.1) but favored the rich in Southeast Asia and the Western Pacific. National income was generally not associated with the magnitude of socioeconomic gradients. In low and middle-income countries there is substantial between and within-region heterogeneity in socioeconomic inequality in tobacco exposure that is not explained by national income. Our results imply that the relationship between socioeconomic position and smoking in poorer countries is dynamic and may not reflect the historical pattern in wealthier countries.
Harper S, Rushani D, Kaufman J. “Trends in the Black-White Life Expectancy Gap, 2003-2008.” JAMA. 2012;307(21):2257–2259
Harper S, Kaufman J, King N. “Increased Risk of Coronary Heart Disease in Female Smokers.” The Lancet. 2012;379(9818):801–802
Hosseinpoor A, Bergen N, Mendis S, Harper S, Verdes E, Kunst A, Chatterji S. “Socioeconomic Inequality in the Prevalence of Noncommunicable Diseases in Low- and Middle-Income Countries: Results from the World Health Survey.” BMC Public Health. 2012;12(1)
Background: Noncommunicable diseases are an increasing health concern worldwide, but particularly in low- and middle-income countries. This study quantified and compared education- and wealth-based inequalities in the prevalence of five noncommunicable diseases (angina, arthritis, asthma, depression and diabetes) and comorbidity in low- and middle-income country groups. Methods. Using 2002-04 World Health Survey data from 41 low- and middle-income countries, the prevalence estimates of angina, arthritis, asthma, depression, diabetes and comorbidity in adults aged 18years or above are presented for wealth quintiles and five education levels, by sex and country income group. Symptom-based classification was used to determine angina, arthritis, asthma and depression rates, and diabetes diagnoses were self-reported. Socioeconomic inequalities according to wealth and education were measured absolutely, using the slope index of inequality, and relatively, using the relative index of inequality. Results: Wealth and education inequalities were more pronounced in the low-income country group than the middle-income country group. Both wealth and education were inversely associated with angina, arthritis, asthma, depression and comorbidity prevalence, with strongest inequalities reported for angina, asthma and comorbidity. Diabetes prevalence was positively associated with wealth and, to a lesser extent, education. Adjustments for confounding variables tended to decrease the magnitude of the inequality. Conclusions: Noncommunicable diseases are not necessarily diseases of the wealthy, and showed unequal distribution across socioeconomic groups in low- and middle-income country groups. Disaggregated research is warranted to assess the impact of individual noncommunicable diseases according to socioeconomic indicators. © 2012 Hosseinpoor et al.; licensee BioMed Central Ltd.
Hosseinpoor A, Harper S, Lee J, Lynch J, Mathers C, Abou-Zahr C. “International Shortfall Inequality in Life Expectancy in Women and in Men, 1950-2010.” Bulletin of the World Health Organization. 2012;90(8):588–594
Objective To assess international shortfall inequality in life expectancy at birth among women and men and the influence of geography and country income group. Methods The authors used estimates of life expectancy at birth, by sex, for 12 five-year periods between 1950-1955 and 2005-2010 and estimates of population for the midpoints of each period from the World population prospects, 2008 revision. Shortfall inequality was defined as the weighted average of the deviations of each country's average life expectancy by sex from the highest attained life expectancy by sex for each period. Findings International shortfall inequalities in life expectancy among men and among women decreased between 1950 and 1975 but stagnated thereafter. International shortfall inequality in life expectancy has been higher in women than in men, ranging from 1.9 to 2.9 years. Women in low-income countries have the biggest shortfall, currently at around 26.7 years. Conclusion International shortfall inequality is higher among women than men primarily because women in low-income and lower-middle-income country groups show larger differences in life expectancy than men. Further investigation is needed to determine the pathways causing these inequalities.
Hosseinpoor A, Bergen N, Kunst A, Harper S, Guthold R, Rekve D, family = Espaignet g=E, Naidoo N, Chatterji S. “Socioeconomic Inequalities in Risk Factors for Non Communicable Diseases in Low-Income and Middle-Income Countries: Results from the World Health Survey..” BMC public health. 2012;12:912
Monitoring inequalities in non communicable disease risk factor prevalence can help to inform and target effective interventions. The prevalence of current daily smoking, low fruit and vegetable consumption, physical inactivity, and heavy episodic alcohol drinking were quantified and compared across wealth and education levels in low- and middle-income country groups. This study included self-reported data from 232,056 adult participants in 48 countries, derived from the 2002-2004 World Health Survey. Data were stratified by sex and low- or middle-income country status. The main outcome measurements were risk factor prevalence rates reported by wealth quintile and five levels of educational attainment. Socioeconomic inequalities were measured using the slope index of inequality, reflecting differences in prevalence rates, and the relative index of inequality, reflecting the prevalence ratio between the two extremes of wealth or education accounting for the entire distribution. Data were adjusted for confounding factors: sex, age, marital status, area of residence, and country of residence. Smoking and low fruit and vegetable consumption were significantly higher among lower socioeconomic groups. The highest wealth-related absolute inequality was seen in smoking among men of low- income country group (slope index of inequality 23.0 percentage points; 95\% confidence interval 19.6, 26.4). The slope index of inequality for low fruit and vegetable consumption across the entire distribution of education was around 8 percentage points in both sexes and both country income groups. Physical inactivity was less prevalent in populations of low socioeconomic status, especially in low-income countries (relative index of inequality: (men) 0.46, 95\% confidence interval 0.33, 0.64; (women) 0.52, 95\% confidence interval 0.42, 0.65). Mixed patterns were found for heavy drinking. Disaggregated analysis of the prevalence of non-communicable disease risk factors demonstrated different patterns and varying degrees of socioeconomic inequalities across low- and middle-income settings. Interventions should aim to reach and achieve sustained benefits for high-risk populations.
Kaufman J, Harper S. “Deficiency of the Odds Ratio for Common Outcomes.” American Journal of Psychiatry. 2012;169(10):1118
King N, Harper S, Young M. “Use of Relative and Absolute Effect Measures in Reporting Health Inequalities: Structured Review.” BMJ (Online). 2012;345(7878)
Objective: To examine the frequency of reporting of absolute and relative effect measures in health inequalities research. Design: Structured review of selected general medical and public health journals. Data sources: 344 articles published during 2009 in American Journal of Epidemiology, American Journal of Public Health, BMJ, Epidemiology, International Journal of Epidemiology, JAMA, Journal of Epidemiology and Community Health, The Lancet, The New England Journal of Medicine, and Social Science and Medicine. Main outcome measures: Frequency and proportion of studies reporting absolute effect measures, relative effect measures, or both in abstract and full text; availability of absolute risks in studies reporting only relative effect measures. Results: 40\% (138/344) of articles reported a measure of effect in the abstract; among these, 88\% (122/138) reported only a relative measure, 9\% (13/138) reported only an absolute measure, and 2\% (3/138) reported both. 75\% (258/344) of all articles reported only relative measures in the full text; among these, 46\% (119/258) contained no information on absolute baseline risks that would facilitate calculation of absolute effect measures. 18\% (61/344) of all articles reported only absolute measures in the full text, and 7\% (25/344) reported both absolute and relative measures. These results were consistent across journals, exposures, and outcomes. Conclusions: Health inequalities are most commonly reported using only relative measures of effect, which may influence readers' judgments of the magnitude, direction, significance, and implications of reported health inequalities.
Speybroeck N, Harper S, De Savigny D, Victora C. “Inequalities of Health Indicators for Policy Makers: Six Hints.” International Journal of Public Health. 2012;57(5):855–858
Speybroeck N, Harper S, De Savigny D, Victora C. “Erratum to: Inequalities of Health Indicators for Policy Makers: Six Hints (International Journal of Public Health (2012) DOI 10.1007/S00038-012-0386-5).” International Journal of Public Health. 2012;57(5):859–860
Toporowski A, Harper S, Fuhrer R, Buffler P, Detels R, Krieger N, Franco E. “Burden of Disease, Health Indicators and Challenges for Epidemiology in North America.” International Journal of Epidemiology. 2012;41(2):540–556
Background: Commissioned by the International Epidemiological Association, this article is part of a series on burden of disease, health indicators and the challenges faced by epidemiologists in bringing their discoveries to provide equitable benefit to the populations in their regions and globally. This report covers the health status and epidemiological capacity in the North American region (USA and Canada). Methods: We assessed data from country-specific sources to identify health priorities and areas of greatest need for modifiable risk factors. We examined inequalities in health as a function of social deprivation. We also reviewed information on epidemiological capacity building and scientific contributions by epidemiologists in the region. Findings: The USA and Canada enjoy technologically advanced healthcare systems that, in principle, prioritize preventive services. Both countries experience a life expectancy at birth that is higher than the global mean. Health indicator measures are consistently worse in the USA than in Canada for many outcomes, although typically by only marginal amounts. Socio-economic and racial/ethnic disparities in indicators exist for many diseases and risk factors in the USA. To a lesser extent, these social inequalities also exist in Canada, particularly among the Aboriginal populations. Epidemiology is a well-established discipline in the region, with many degree-granting schools, societies and job opportunities in the public and private sectors. North American epidemiologists have made important contributions in disease control and prevention and provide nearly a third of the global scientific output via published papers. Conclusions: Critical challenges for North American epidemiologists include social determinants of disease distribution and the underlying inequalities in access to and benefit from preventive services and healthcare, particularly in the USA. The gains in life expectancy also underscore the need for research on health promotion and prevention of disease and disability in older adults. The diversity in epidemiological subspecialties poses new challenges in training and accreditation and has occurred in parallel with a decrease in research funding. Published by Oxford University Press on behalf of the International Epidemiological Association © The Author 2012; all rights reserved.
Yang S, Khang Y, Chun H, Harper S, Lynch J. “The Changing Gender Differences in Life Expectancy in Korea 1970-2005.” Social Science and Medicine. 2012;75(7):1280–1287
Women live much longer than men in Korea, with remarkable gains in life expectancy at birth for the past decades. The gender differential has steadily increased over time, reaching a peak of more than 8 years in 1980s, and decreased thereafter to 6.7 years in 2005. Studies to investigate the pattern and contributing factors to changes in the life expectancy gender gap have been mostly from Western countries, and there has been no such study in Asian countries, except in Japan. We therefore aimed to examine age- and cause-specific contributions to the changing gender differentials in life expectancy in Korea, in particular the decline of the gap, using a decomposition method. Between 1970 and 1979 when the gender gap in life expectancy widened, faster mortality decline among women in ages 20-44 explained 66\% of the total increase in the gender gap, which would be due to substantial improvements in reproductive health among women and excess male mortality in occupational injuries and transport accidents. Although greater survival advantage among elderly women over 70 contributed to further increase in the gender gap, the contributions from younger ages with the ages 15-64 contributing the most (-2 years) resulted in the overall reduction of the gender gap which began in 1992 and continued to 2005. Among causes of death, liver diseases (-0.5 years, 38\% of the total decline), transport accidents (-0.4 years, 31\%), hypertensive diseases (-0.3 years, 19\%), stroke (-0.1 years, 11\%), and tuberculosis (-0.1 years) contributed the most to the overall 1.4 years reduction in the gender gap. However, changes in mortality from lung cancer (+0.3 years), suicide (+0.3 years), chronic lower respiratory diseases (+0.2 years), and ischemic heart diseases (+0.1 years) contributed to widening the gap during the same period. In sum, while smoking-related causes of death have contributed most to the narrowing gap in most other industrialized countries, these causes contributed toward increasing the gender gap in Korea. Instead, liver disease, hypertension-related diseases, and transport accidents were major contributing causes of death to the narrowing of gender differentials in life expectancy in Korea. © 2012 Elsevier Ltd.
Adam-Smith J, Harper S, Auger N. “Causes of Widening Life Expectancy Inequalities in Québec, Canada, 1989-2004.” Canadian Journal of Public Health. 2011;102(5):375–381
Objectives: Inequalities in life expectancy between the most and least materially deprived areas in the province of Quebec, Canada are increasing, but the reasons for this trend are unclear. An analysis of which causes of death, in which age groups, are implicated in this trend is necessary to understand it and provide clear targets for intervention. Methods: We analyzed Québec mortality data for three 4-year periods from 1989-2004. Differences in life expectancy at birth between the most- and least-deprived areas in each period and over time were decomposed by age and cause of death for men and women using the Arriaga method. Results: Life expectancy increased for all areas, but the increase was smaller in deprived areas. Cancer, heart disease and unintentional injuries accounted for the largest shares of inequality in each period. Among women, the widening life expectancy gap was largely due to increased lung cancer mortality. Among men, mortality from HIV was an important cause of widening inequality. Increasing inequality in both sexes was offset by a decrease in mortality from unintentional injuries in deprived areas (narrowing the gap between the most- and least-deprived areas). The largest share of inequality among women was due to deaths in the 65-and-over age group. Among men, the largest share shifted over time from middle-age to the 65-and-over age group. Conclusion: The widening life expectancy gap between advantaged and disadvantaged areas in Québec is driven by relatively few causes of death with well-known risk factors and strategies for prevention. © Canadian Public Health Association, 2011. All rights reserved.
Agha G, Murabito J, Lynch J, Abrahamowicz M, Harper S, Loucks E. “Relation of Socioeconomic Position with Ankle-Brachial Index.” American Journal of Cardiology. 2011;108(11):1651–1657
Potential upstream determinants of coronary heart disease (CHD) include life-course socioeconomic position (e.g., childhood socioeconomic circumstances, own education and occupation); however, several plausible biological mechanisms by which socioeconomic position (SEP) may influence CHD are poorly understood. Several CHD risk factors appear to be more strongly associated with SEP in women than in men; little is known as to whether any CHD risk factors may be more strongly associated with SEP in men. Objectives were to evaluate whether cumulative life-course SEP is associated with a measurement of subclinical atherosclerosis, the anklebrachial index (ABI), in men and women. This study was a prospective analysis of 1,454 participants from the Framingham Heart Study Offspring Cohort (mean age 57 years, 53.8\% women). Cumulative SEP was calculated by summing tertile scores for father's education, own education, and own occupation. ABI was dichotomized as low (≤1.1) and normal ({$>$}1.1 to 1.4). After adjustment for age and CHD risk factors cumulative life-course SEP was associated with low ABI in men (odds ratio [OR] 2.04, 95\% confidence interval [CI] 1.22 to 3.42, for low vs high cumulative SEP score) but not in women (OR 0.86, 95\% CI 0.56 to 1.33). Associations with low ABI in men were substantially driven by their own education (OR 4.13, 95\% CI 1.86 to 9.16, for lower vs higher than high school education). In conclusion, cumulative life-course SEP was associated with low ABI in men but not in women. © 2011 Elsevier Inc. All rights reserved.
Auger N, Roncarolo F, Harper S. “Increasing Educational Inequality in Preterm Birth in Québec, Canada, 1981–2006.” J Epidemiol Community Health. 2011;65(12):1091–1096
Background Few studies have evaluated the relationship between preterm birth (PTB) and maternal education over time. We sought to determine whether educational inequalities in PTB have increased in Québec, Canada. Methods The authors analysed 2 124 909 singleton live births from 1981 to 2006, and computed the Relative Index of Inequality (RII) and Slope Index of Inequality (SII) with 95\% CIs for the relationship between maternal education and extreme, very or moderate PTB (≤27, 28–31, and 32–36 completed weeks of gestation, respectively) for five periods (1981–1985, 1986–1990, 1991–1995, 1996–2000, 2001–2006), adjusting for maternal age, marital status, birthplace, language spoken at home, parity and infant sex. Results Average rates of extreme and moderate PTB increased over time but decreased for very PTB. A statistically significant increase in the RII over time was present for extreme and moderate PTB. The adjusted RII for extreme PTB increased from 1.58 (95\% CI 1.24 to 2.01) in 1981–1985 to 3.11 (95\% CI 2.54 to 3.81) in 2001–2006. For moderate PTB, the corresponding RIIs were 1.53 (95\% CI 1.44 to 1.61) and 1.91 (95\% CI 1.81 to 2.01). Absolute differences in the PTB proportion between the least and most educated mothers increased from 1981 to 2006 for extreme (adjusted SII 0.11\% vs 0.28\%) and moderate PTB (adjusted SII 1.67\% vs 3.11\%). Absolute differences in the proportion very PTB did not increase. Conclusions Relative and absolute educational inequalities in extreme and moderate PTB have increased over time in Québec. Relative increases were largest for extreme PTB, and absolute increases were largest for moderate PTB.
Auger N, Gamache P, Adam-Smith J, Harper S. “Relative and Absolute Disparities in Preterm Birth Related to Neighborhood Education.” Annals of Epidemiology. 2011;21(7):481–488
Purpose: To evaluate relative and absolute relationships between preterm birth (PTB) and neighborhood education over time. Methods: Live births in Québec, Canada, were obtained for 1990-1995, 1996-2000, and 2001-2006. Mean maternal education and proportion of females with no high school diploma were expressed as continuous cumulative rank scores for 10,923 neighborhoods. We estimated the relative hazard of PTB ({$<$}37 gestational weeks) for neighborhood education in each period by using Cox proportional hazards regression, accounting for individual education, age, marital status, birthplace, language, parity, infant sex, rurality, neighborhood income, and area clustering. Adjusted absolute differences in the prevalence of PTB between the most and least educated neighborhoods were calculated. Results: PTB prevalence (6.1\% overall) was greater in less-educated neighborhoods. Although PTB proportions increased over time in all neighborhoods, the increase was proportionately greater for less-educated areas. Hazards of PTB for neighborhood education were proportional over gestation. Depending on the indicator of neighborhood education and period, adjusted hazards of PTB were 10\%-37\% greater for the least relative to most educated neighborhoods, and prevalence percentage differences ranged from 0.6\% to 1.9\%. Associations persisted over time. Conclusions: Relative and absolute neighborhood educational inequalities in PTB, independent of individual education, were present and persistent over time. © 2011 Elsevier Inc.
Harper S, Lynch J, Smith G. “Social Determinants and the Decline of Cardiovascular Diseases: Understanding the Links.” Annual Review of Public Health. 2011;32:39–69
This article reviews the historical declines in cardiovascular mortality and provides an overview of the contribution of social and economic factors to disease change. We document the magnitude of declines in cardiovascular diseases and the major role of changes in conventional risk factors, and we review the contributions of social determinants to changes in disease rates. We conclude by arguing that understanding patterns and trends of social inequalities in cardiovascular disease and its risk factors requires consideration of the specific intersections of health and social exposures acting across the life course in different settings, in both time and place. © 2011 by Annual Reviews. rights reserved.
Kaufman J, Harper S, King N. “A More Complete Picture of Higher Cardiovascular Disease Prevalence among Blacks Compared to Whites.” American Journal of Medicine. 2011;124(5):e5-e6
Larose A, Moore S, Harper S, Lynch J. “Global Income-Related Inequalities in HIV Testing.” Journal of Public Health. 2011;33(3):345–352
Voluntary counseling and testing (VCT) is an important prevention initiative in reducing HIV/AIDS transmission. Despite current global prevention efforts, many low- and middle-income countries continue reporting low VCT levels. Little is known about the association of within- and between-country socioeconomic inequalities and VCT. Based on the 'inverse equity hypothesis,' this study examines the degree to which low socioeconomic groups in developing countries are disadvantaged in VCT. MethodsUsing recently released data from the 2002 to 2003 World Health Survey (WHS) for 106 705 individuals in 49 countries, this study used multilevel logistic regression to examine the association of individual- and national-level factors with VCT, and whether national economic development moderated the association between individual income and VCT. Individual income was based on country-specific income quintiles. National economic development was based on national gross domestic product per capita (GDP/c). Effect modification was evaluated with the likelihood ratio test (G 2). Individuals eligible for the VCT question of the WHS were adults between the ages of 1849 years; women who had given birth in the last 2 years were excluded from this question. ResultsVCT was more likely among higher income quintiles and in countries with higher GDP/c. GDP/c moderated the association between individual income and VCT whereby relative income differences in VCT were greater in countries with lower GDP/c (G 2 9.21; P 0.002). Individual socio-demographic characteristics were also associated with the likelihood of a person having VCT. ConclusionsRelative socioeconomic inequalities in VCT coverage appear to decline when higher SES groups reach a certain level of coverage. These findings suggest that changes to international VCT programs may be necessary to moderate the relative VCT differences between high- and low-income individuals in lower GDP/c nations. © The Author 2011, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved.
McKinnon B, Harper S, Moore S. “Decomposing Income-Related Inequality in Cervical Screening in 67 Countries.” International Journal of Public Health. 2011;56(2):139–152
OBJECTIVES: The development of successful policies to reduce income-related inequalities in cervical cancer screening rates requires an understanding of the reasons why low-income women are less likely to be screened. We sought to identify important determinants contributing to inequality in cervical screening rates. METHODS: We analyzed data from 92,541 women aged 25-64 years, who participated in the World Health Survey in 2002-2003. Income-related inequality in Pap screening was measured using the concentration index (CI). Using a decomposition method for the CI, we quantified the contribution to inequality of age, education level, marital status, urbanicity and recent health-care need. RESULTS: There was substantial heterogeneity in the contributions of different determinants to inequality among countries. Education generally made the largest contribution (median = 15\%, interquartile range [IQR] = 23\%), although this varied widely even within regions (e.g., 5\% in Austria, 28\% in Hungary). The contribution of rural residence was greatest in African countries (median = 10\%, IQR = 13\%); however, there was again substantial within-region variation (e.g., 26\% in Zambia, 2\% in Kenya). CONCLUSIONS: Considerable heterogeneity in the contributions of screening determinants among countries suggests interventions to reduce screening inequalities may require country-specific approaches.
Sadana R, Harper S. “Data Systems Linking Social Determinants of Health with Health Outcomes: Advancing Public Goods to Support Research and Evidence-Based Policy and Programs.” Public Health Reports. 2011;126:6–13
Smith B, Lynch J, Fox C, Harper S, Abrahamowicz M, Almeida N, Loucks E. “Life-Course Socioeconomic Position and Type 2 Diabetes Mellitus the Framingham Offspring Study.” American Journal of Epidemiology. 2011;173(4):438–447
Evidence is lacking on whether the duration and timing of low socioeconomic position (SEP) across a person's life course may be associated with incidence of type 2 diabetes mellitus (T2D). The authors' objectives were to investigate associations between cumulative SEP and the incidence of T2D in the Framingham Offspring Study (n = 1,893; 52\% women; mean baseline age = 34 years). Pooled logistic regression analyses demonstrated that age-adjusted cumulative SEP was associated with T2D in women (for low vs. high cumulative SEP, odds ratio (OR) = 1.92, 95\% confidence interval (CI): 1.08, 3.42). Age-adjusted analyses for young-adulthood SEP (7.85 for ≤12 vs. {$>$}16 years of education, OR = 2.84, 95\% CI: 1.03), active professional life SEP (for laborer vs. professional/executive/supervisory/technical occupations, OR = 2.40, 95\% CI: 1.05, 5.47), and social-mobility frameworks (for declining life-course SEP, OR = 2.99, 95\% CI: 1.39, 6.44; for stable low vs. stable high life-course SEP, OR = 1.85, 95\% CI: 1.02, 3.35) all demonstrated associations between low SEP and T2D incidence in women. No association was observed between childhood SEP and T2D in women for father's education (some high school or less vs. any postsecondary education, OR = 1.26, 95\% CI: 0.72, 2.22). In men, there was little evidence of associations between life-course SEP and T2D incidence. These findings suggest that cumulative SEP is inversely associated with incidence of T2D in women, and that this association may be primarily due to the women's educational levels and occupations. © 2011 The Author.
Harper S, King NB, Meersman SC, Reichman ME, Breen N, Lynch J. “Implicit Value Judgments in the Measurement of Health Inequalities.” The Milbank Quarterly. 2010;88(1):4–29
CONTEXT: Quantitative estimates of the magnitude, direction, and rate of change of health inequalities play a crucial role in creating and assessing policies aimed at eliminating the disproportionate burden of disease in disadvantaged populations. It is generally assumed that the measurement of health inequalities is a value-neutral process, providing objective data that are then interpreted using normative judgments about whether a particular distribution of health is just, fair, or socially acceptable. METHODS: We discuss five examples in which normative judgments play a role in the measurement process itself, through either the selection of one measurement strategy to the exclusion of others or the selection of the type, significance, or weight assigned to the variables being measured. FINDINGS: Overall, we find that many commonly used measures of inequality are value laden and that the normative judgments implicit in these measures have important consequences for interpreting and responding to health inequalities. CONCLUSIONS: Because values implicit in the generation of health inequality measures may lead to radically different interpretations of the same underlying data, we urge researchers to explicitly consider and transparently discuss the normative judgments underlying their measures. We also urge policymakers and other consumers of health inequalities data to pay close attention to the measures on which they base their assessments of current and future health policies.
Harper S. “Editorial: Inequalities in Cancer Survival and the NHS Cancer Plan: Evidence of Progress.” British Journal of Cancer. 2010;103(4):437–438
King N, Harper S, Meersman S, Reichman M, Breen N, Lynch J. “We’ll Take the Red Pill: A Reply to Asada.” Milbank Quarterly. 2010;88(4):623–627
King N, Kaufman J, Harper S. “Relative Measures Alone Tell Only Part of the Story.” American Journal of Public Health. 2010;100(11):2014–2015
Konings P, Harper S, Lynch J, Hosseinpoor A, Berkvens D, Lorant V, Geckova A, Speybroeck N. “Analysis of Socioeconomic Health Inequalities Using the Concentration Index.” International Journal of Public Health. 2010;55(1):71–74
Kopec J, Finès P, Manuel D, Buckeridge D, Flanagan W, Oderkirk J, Abrahamowicz M, Harper S, Sharif B, Okhmatovskaia A, Sayre E, Rahman M, Wolfson M. “Validation of Population-Based Disease Simulation Models: A Review of Concepts and Methods.” BMC Public Health. 2010;10
Background: Computer simulation models are used increasingly to support public health research and policy, but questions about their quality persist. The purpose of this article is to review the principles and methods for validation of population-based disease simulation models. Methods. We developed a comprehensive framework for validating population-based chronic disease simulation models and used this framework in a review of published model validation guidelines. Based on the review, we formulated a set of recommendations for gathering evidence of model credibility. Results. Evidence of model credibility derives from examining: 1) the process of model development, 2) the performance of a model, and 3) the quality of decisions based on the model. Many important issues in model validation are insufficiently addressed by current guidelines. These issues include a detailed evaluation of different data sources, graphical representation of models, computer programming, model calibration, between-model comparisons, sensitivity analysis, and predictive validity. The role of external data in model validation depends on the purpose of the model (e.g., decision analysis versus prediction). More research is needed on the methods of comparing the quality of decisions based on different models. Conclusion. As the role of simulation modeling in population health is increasing and models are becoming more complex, there is a need for further improvements in model validation methodology and common standards for evaluating model credibility. © 2010 Kopec et al; licensee BioMed Central Ltd.
McTavish S, Moore S, Harper S, Lynch J. “National Female Literacy, Individual Socio-Economic Status, and Maternal Health Care Use in Sub-Saharan Africa.” Social Science and Medicine. 2010;71(11):1958–1963
The United Nations Millennium Development Goals have identified improving women's access to maternal health care as a key target in reducing maternal mortality in sub-Saharan Africa (sSA). Although individual factors such as income and urban residence can affect maternal health care use, little is known about national-level factors associated with use. Yet, such knowledge may highlight the importance of global and national policies in improving use. This study examines the importance of national female literacy on women's maternal health care use in continental sSA. Data that come from the 2002-2003 World Health Survey. Multilevel logistic regression was used to examine the association between national female literacy and individual's non-use of maternal health care, while adjusting for individual-level factors and national economic development. Analyses also assessed effect modification of the association between income and non-use by female literacy. Effect modification was evaluated with the likelihood ratio test (G2). We found that within countries, individual age, education, urban residence and household income were associated with lack of maternal health care. National female literacy modified the association of household income with lack of maternal health care use. The strength of the association between income and lack of maternal health care was weaker in countries with higher female literacy. We conclude therefore that higher national levels of female literacy may reduce income-related inequalities in use through a range of possible mechanisms, including women's increased labour participation and higher status in society. National policies that are able to address female literacy and women's status in sub-Saharan Africa may help reduce income-related inequalities in maternal health care use. © 2010 Elsevier Ltd.
Moore S, Hall J, Harper S, Lynch J. “Global and National Socioeconomic Disparities in Obesity, Overweight, and Underweight Status.” Journal of Obesity. 2010;2010
Objective. To examine the association between socioeconomic factors and weight status across 53 countries. Methods. Data are cross-sectional and from the long version of the World Health Survey (WHS). There were 172,625 WHS participants who provided self-reported height and weight measures and sociodemographic information. The International Classification of adult weight status was used to classify participants by body mass index (BMI): (1) underweight ({$<$}18.5), (2) normal weight (18.5-24.9), (3) overweight (25.0-29.9), and (4) obese ({$>$}30.0). Multinomial regression was used in the analyses. Results. Globally, 6.7\% was underweight, 25.7\% overweight, and 8.9\% obese. Underweight status was least (5.8\%) and obesity (9.3\%) most prevalent in the richest quintile. There was variability between countries, with a tendency for lower-income quintiles to be at increased risk for underweight and reduced risk for obesity. Conclusion. International policies may require flexibility in addressing cross-national differences in the socio-economic covariates of BMI status. Copyright © 2010 Spencer Moore et al.
Speybroeck N, Konings P, Lynch J, Harper S, Berkvens D, Lorant V, Geckova A, Hosseinpoor AR. “Decomposing Socioeconomic Health Inequalities.” International Journal of Public Health. 2010;55(4):347–351
Yang S, Khang Y, Harper S, Smith G, Leon D, Lynch J. “Understanding the Rapid Increase in Life Expectancy in South Korea.” American Journal of Public Health. 2010;100(5):896–903
Objectives. We assessed life expectancy increases in the past several decades in South Korea by age and specific causes of death. Methods. We applied Arriaga's decomposition method to life table data (1970-2005) and mortality statistics (1983-2005) to estimate age- and cause-specific contributions to changes in life expectancy. Results. Reductions in infant mortality made the largest age-group contribution to the life expectancy increase. Reductions in cardiovascular diseases (particularly stroke and hypertensive diseases) contributed most to longer life expectancy between 1983 and 2005 (30\% in males and 28\% in females). Lower rates of stomach cancer, liver disease, tuberculosis, and external-cause mortality accounted for 30\% of the male and 20\% of the female increase in longevity. However, higher mortality from ischemic heart disease, lung and bronchial cancer, colorectal cancer, breast cancer, diabetes, and suicide offset gains by 10\% in both genders. Conclusions. Rapid increases in life expectancy in South Korea were mostly achieved by reductions in infant mortality and in diseases related to infections and blood pressure.
Alvarado B, Harper S, Platt R, Smith G, Lynch J. “Would Achieving Healthy People 2010’s Targets Reduce Both Population Levels and Social Disparities in Heart Disease?.” Circulation: Cardiovascular Quality and Outcomes. 2009;2(6):598–606
Background-The US Healthy People 2010 (HP2010) agenda set targets for major risk factors for coronary heart disease (CHD). However, the potential impact of achieving those risk factor reductions on both population levels and social disparities in CHD has not been quantified. Methods and Results-Data on 10-year risk of CHD (from the First National Health and Nutrition Examination Epidemiological Follow-Up study 1971 to 1982), prevalence of major CHD risk factors (from the National Health and Nutrition Examination Survey 2003 to 2004), and HP2010 targets for CHD risk factors (reduction of smoking rate to 12\%, hypertension to 14\%, high cholesterol levels to 17\%, diabetes to 2.5\%, and obesity to 15\%) were used to estimate effects of different scenarios on population levels and social disparities in CHD. Over a 10-year period, the largest relative reductions in population levels of CHD (20.0\% in men; 23.9\% in women) would be achieved if all social groups met the HP2010 targets. CHD disparities would be most reduced if the less educated (absolute disparities reduced by 66.1\% in men; 56.3\% in women) and the low income group (absolute disparities reduced by 93.7\% in men; 94.3\% in women) achieved the targets before the most advantaged. These reductions are larger than those expected if targets were achieved overall for the population but relative social group differences in risk factors remained, or under leveling-up approaches in which the least advantaged achieved the current levels of risk factors of the most advantaged. Conclusions-Interventions to reduce CHD risk factors to HP2010 targets that focus on all social groups would produce the best overall scenario for both population levels and disparities in CHD. © 2009 American Heart Association, Inc.
Hall J, Moore S, Harper S, Lynch J. “Global Variability in Fruit and Vegetable Consumption.” American Journal of Preventive Medicine. 2009;36(5):402-409.e5
Background: Low fruit and vegetable consumption is an important risk factor for chronic diseases, but for many (mainly developing) countries, no prevalence data have ever been published. This study presents data on the prevalence of low fruit and vegetable intake for 52 countries and for various sociodemographic groups and settings across these countries. Methods: Data from 196,373 adult participants from 52 countries taking part in the World Health Survey (2002-2003) were analyzed in the summer of 2008. Low fruit and vegetable consumption was defined according to the WHO guidelines of a minimum of five servings of fruits and/or vegetables daily. Results: Low fruit and vegetable consumption prevalence ranged from 36.6\% (Ghana) to 99.2\% (Pakistan) for men and from 38.0\% (Ghana) to 99.3\% (Pakistan) for women. Significant differences in the likelihood of low fruit and vegetable intake between men and women were found in 15 countries. The prevalence of low fruit and vegetable consumption tended to increase with age and decrease with income. Although urbanicity was not associated overall with low fruit and vegetable consumption, urban and rural differences were significant for 11 countries. Conclusions: Overall, 77.6\% of men and 78.4\% of women from the 52 mainly low- and middle-income countries consumed less than the minimum recommended five daily servings of fruits and vegetables. Baseline global information on low fruit and vegetable consumption obtained in this study can help policymakers worldwide establish interventions for addressing the global chronic disease epidemic. © 2009 American Journal of Preventive Medicine.
Harper S, Lynch J, Meersman S, Breen N, Davis W, Reichman M. “Trends in Area-Socioeconomic and Race-Ethnic Disparities in Breast Cancer Incidence, Stage at Diagnosis, Screening, Mortality, and Survival among Women Ages 50 Years and over (1987-2005).” Cancer Epidemiology Biomarkers and Prevention. 2009;18(1):121–131
Background: Breast cancer is the most commonly diagnosed cancer and the second leading cause of cancer death among women in the United States and varies systematically by race-ethnicity and socioeconomic status. Previous research has often focused on disparities between particular groups, but few studies have summarized disparities across multiple subgroups defined by race-ethnic and socioeconomic position. Methods: Data on breast cancer incidence, stage, mortality, and 5-year cause-specific probability of death (100 - survival) were obtained from the Surveillance, Epidemiology, and End Results program and data on mammography screening from the National Health Interview Survey from 1987 to 2005. We used four area-socioeconomic groups based on the percentage of poverty in the county of residence ({$<$}10, 10-15, 15-20, +20\%) and five race-ethnic groups (White, Black, Asian, American Indian, and Hispanic). We used summary measures of disparity based on both rate differences and rate ratios. Results: From 1987 to 2004, area-socioeconomic disparities declined by 20\% to 30\% for incidence, stage at diagnosis, and 5-year cause-specific probability of death, and by roughly 100\% for mortality, whether measured on the absolute or relative scale. In contrast, relative area-socioeconomic disparities in mammography use increased by 161\%. Absolute race-ethnic disparities declined across all outcomes, with the largest reduction for mammography (56\% decline). Relative race-ethnic disparities for mortality and 5-year cause-specific probability of death increased by 24\% and 17\%, respectively. Conclusions: Our analysis suggests progress towards race-ethnic and area-socioeconomic disparity goals for breast cancer, especially when measured on the absolute scale. However, greater progress is needed to address increasing relative socioeconomic disparities in mammography and race-ethnic disparities in mortality and 5-year cause-specific probability of death. Copyright © 2009 American Association for Cancer Research.
Harper S. “Commentary: Trends in Indigenous Inequalities in Mortality in New Zealand.” International Journal of Epidemiology. 2009;38(6):1722–1724
Khang Y, Lynch J, Yang S, Harper S, Yun S, Jung-Choi K, Kim H. “The Contribution of Material, Psychosocial, and Behavioral Factors in Explaining Educational and Occupational Mortality Inequalities in a Nationally Representative Sample of South Koreans: Relative and Absolute Perspectives.” Social Science and Medicine. 2009;68(5):858–866
The contributions of material, psychosocial, and behavioral factors in explaining socioeconomic inequalities in health have been explored in many Western studies. Most prior investigations have looked at relative abilities to explain such inequalities. In addition, little research focuses on Asian countries, despite the fact that the prevalence and socioeconomic distribution of risk factors for mortality are different there. This study examined relative and absolute abilities of material, psychosocial, and behavioral pathways to explain educational and occupational inequalities in mortality in a nationally representative sample from South Korea. The 1998 and 2001 National Health and Nutrition Examination Survey data were pooled and linked to national mortality data. Of 8366 men and women over 30 years of age, 310 died between 1999 and 2005. Nine pathway variables were examined: three material factors (income, health insurance, and car ownership status), three psychosocial factors (depression, stress, and marital status), and three behavioral factors (smoking, alcohol consumption, and physical exercise). The relative risk and relative index of inequality were used as measures of relative inequality, and risk differences and the slope index of inequality were used as measures of absolute inequality. Material factors explained a total of 29.0\% of the excess in relative risk for education and 50.0\% of the excess in relative risk for occupational class. Material factors explained 78.6\% of the excess in absolute mortality difference for education and 41.1\% for occupational class. Psychosocial factors for both education and occupational class had a relative and absolute explanatory power of less than 15\%. Behavioral factors showed a relative explanatory power of about 15\%, but absolute explanatory power reached 84.0\% for education and 105.4\% for occupational class. However, the number of deaths used to calculate the absolute explanatory power was small. Results of this study suggest that absolute socioeconomic mortality inequalities could be substantially reduced if behavioral risk factors were reduced in the whole population. © 2008 Elsevier Ltd. All rights reserved.
Harper S, Lynch J, Meersman S, Breen N, Davis W, Reichman M. “An Overview of Methods for Monitoring Social Disparities in Cancer with an Example Using Trends in Lung Cancer Incidence by Area-Socioeconomic Position and Race-Ethnicity, 1992-2004.” American Journal of Epidemiology. 2008;167(8):889–899
The authors provide an overview of methods for summarizing social disparities in health using the example of lung cancer. They apply four measures of relative disparity and three measures of absolute disparity to trends in US lung cancer incidence by area-socioeconomic position and race-ethnicity from 1992 to 2004. Among females, measures of absolute and relative disparity suggested that area-socioeconomic and race-ethnic disparities increased over these 12 years but differed widely with respect to the magnitude of the change. Among males, the authors found substantial disagreement among summary measures of relative disparity with respect to the magnitude and the direction of change in disparities. Among area-socioeconomic groups, the index of disparity increased by 47\% and the relative concentration index decreased by 116\%, while for race-ethnicity the index of disparity increased by 36\% and the Theil index increased by 13\%. The choice of a summary measure of disparity may affect the interpretation of changes in health disparities. Important issues to consider are the reference point from which differences are measured, whether to measure disparity on the absolute or relative scale, and whether to weight disparity measures by population size. A suite of indicators is needed to provide a clear picture of health disparity change. © The Author 2008. Published by the Johns Hopkins Bloomberg School of Public Health. All rights reserved.
Harper S, Lynch J, Meersman S, Breen N, Davis W, Reichman M. “Harper et al. Respond to “Measuring Social Disparities in Health”.” American Journal of Epidemiology. 2008;167(8):905–907
Harper S, Lynch J, Meersman S, Breen N, Davis W, Reichman M. “The Authors Reply.” American Journal of Epidemiology. 2008;168(10):1216
Cho H, Khang Y, Yang S, Harper S, Lynch J. “Socioeconomic Differentials in Cause-Specific Mortality among South Korean Adolescents.” International Journal of Epidemiology. 2007;36(1):50–57
Background: There is inconsistent evidence regarding the presence of a socioeconomic differential in adolescent all-cause and cause-specific mortality. This study examines possible socioeconomic mortality differentials in Korean adolescents. Method: A total of 330-321 boys and 311-830 girls aged 10-19, who are health insurance beneficiaries for civil servants and private school teachers of Korean Health Insurance Cooperation, were followed for 9 years (1995-2003). Parental income information was linked to national death certificate data. Results: For boys, all-cause mortality showed a graded inverse relationship with income level in both 10-14 year olds (RR = 1.64, 95\% CI: 1.40-1.91) and 15-19 year olds (RR = 1.68, 95\% CI: 1.40-1.91). The major contributor was mortality differentials from external causes, with differentials of transport accident death the most important. Mortality from circulatory disease was higher in the lowest income groups in 15-19 year olds (RR = 2.21, 95\% CI: 1.09-4.50). A significant socioeconomic gradient of non-external cause mortality was found in 15-19 year olds. For girls, socioeconomic differentials were less evident than boys. The all-cause mortality gradient for girls was smaller than for boys and only significant between the lowest and the highest tertile in both 10-14 year olds and 15-19 year olds (RR = 1.33, 95\% CI: 1.02-1.72, RR = 1.38, 95\% CI: 1.11-1.72, respectively). There were significant socioeconomic mortality differentials in all external causes and transport accidents and a marginally significant difference in suicide mortality for 10-19 year olds. Mortality from non-external causes showed no social gradient in girls. Conclusions: Socioeconomic differentials in all-cause mortality were observed in adolescents, even in early youth. This pattern might also apply to mortality from non-external causes, especially cardiovascular disease in 15-19 year old males. © 2007 Published by Oxford University Press on behalf of the International Epidemiological Association © The Author 2006; all rights reserved.
Harper S, Lynch J, Burris S, Davey Smith G. “Trends in the Black-White Life Expectancy Gap in the United States, 1983-2003.” JAMA. 2007;297(11):1224
Context: Since the early 1980s, the black-white gap in life expectancy at birth increased sharply and subsequently declined, but the causes of these changes have not been investigated. Objective: To determine the contribution of specific age groups and causes of death contributing to the changes in the black-white life expectancy gap from 1983-2003. Design and Setting: US vital statistics data from the US National Vital Statistics System, maintained by the National Center for Health Statistics. Standard life table techniques were used to decompose the change in the black-white life expectancy gap by combining absolute changes in age-specific mortality with relative changes in the distribution of causes of death. Main Outcome Measure: The gap in life expectancy at birth between blacks and whites. Results: Among females, the black-white life expectancy gap increased 0.5 years in the period 1983-1993, primarily due to increased mortality from human immunodeficiency virus (HIV) (0.4 years) and slower declines in heart disease (0.1 years), which were somewhat offset by relative improvements in stroke (-0.1 years). The gap among males increased by 2 years in the period 1983-1993, principally because of adverse changes in HIV (1.1 years), homicide (0.5 years), and heart disease (0.3 years). Between 1993 and 2003, the female gap decreased by 1 year (from 5.59 to 4.54 years). Half of the total narrowing of the gap among females was due to relative mortality improvement among blacks in heart disease (-0.2 years), homicide (-0.2 years), and unintentional injuries (-0.1 years). The decline in the life expectancy gap was larger among males, declining by 25\% (from 8.44 to 6.33 years). Nearly all of the 2.1-year decline among males was due to relative mortality improvement among blacks at ages 15 to 49 years (-2.0 years). Three causes of death accounted for 71\% of the narrowing of the gap among males (homicide [-0.6 years], HIV [-0.6 years], and unintentional injuries [-0.3 years]), and lack of improvement in heart disease at older ages kept the gap from narrowing further. Conclusions: After widening during the late 1980s, the black-white life expectancy gap has declined because of relative mortality improvements in homicide, HIV, unintentional injuries, and, among females, heart disease. Further narrowing of the gap will require concerted efforts in public health and health care to address the major causes of the remaining gap from cardiovascular diseases, homicide, HIV, and infant mortality. ©2007 American Medical Association. All rights reserved.
Harper S, Lynch J. “Commentary: Using Innovative Inequality Measures in Epidemiology.” International Journal of Epidemiology. 2007;36(4):926–928
Harper S, Lynch J. “Trends in Socioeconomic Inequalities in Adult Health Behaviors among U.S. States, 1990-2004.” Public Health Reports. 2007;122(2):177–189
Objective. The purpose of this study was to measure state trends in educational inequalities in smoking, binge alcohol use, physical inactivity, obesity, and seatbelt use. Methods. The authors calculated the Relative Concentration Index of educational inequality for five health behaviors on adults from all 50 states and the District of Columbia using data from 1990 to 2004 in the Behavioral Risk Factor Surveillance System (n = 2,118,562). Linear regression was used to measure changes and trends in the Relative Concentration Index of health inequality across education groups in each state. Results. Except for binge alcohol use, poorer health behaviors were concentrated among the less educated. The largest educational inequalities were for physical inactivity. From 1990 to 2004, significant increases in relative educational inequalities occurred in 40 states for smoking and 31 states for physical inactivity. For binge alcohol use, 27 states showed significant declining inequality trends, but educational inequalities reversed direction and binge alcohol use is now more prevalent among the less educated in 19 states. Significant decreases in educational inequalities occurred in 36 states for obesity and 24 states for seat belt use. Changes in educational inequalities across the different health behaviors were not associated, except for a modest correlation between changes in inequality in smoking and binge alcohol use (r = 0.40; p = 0.004). Similarly, there was little association between changes in the population prevalence of health behaviors and changes in educational inequality in health behaviors, with substantial heterogeneity among states. Conclusions. State trends in relative educational inequality among health behaviors were mixed, increasing for smoking and physical inactivity and decreasing for obesity and seat belt use. The factors influencing relative inequality trends may differ from those affecting overall prevalence trends. ©2007 Association of Schools of Public Health.
Harper S, Lynch J. “Highly Active Antiretroviral Therapy and Socioeconomic Inequalities in AIDS Mortality in Spain.” European Journal of Public Health. 2007;17(2):231
Harper S. “Did Clean Water Reduce Black-White Mortality Inequalities in the United States? Water, Race, and Disease..” International journal of epidemiology. 2007;36(1):248–257
Harper S. “Commentary: What Explains Widening Geographic Differences in Life Expectancy in New Zealand?.” International Journal of Epidemiology. 2006;35(3):604–606
Jackson R, Lynch J, Harper S. “Preventing Coronary Heart Disease.” British Medical Journal. 2006;332(7542):617–618
Kelleher C, Lynch J, Daly L, Harper S, Fitz-simon N, Bimpeh Y, Daly E, Ulmer H. “The “Americanisation” of Migrants: Evidence for the Contribution of Ethnicity, Social Deprivation, Lifestyle and Life-Course Processes to the Mid-20th Century Coronary Heart Disease Epidemic in the US.” Social Science and Medicine. 2006;63(2):465–484
We investigated the contribution of the large-scale immigration of White Europeans into the US between 1850 and 1930 to the timing and extent of the epidemic pattern of heart disease between 1900 and 1980. The analyses are based on data collected through the United States Federal Census from 1850 to the present. The hardcopy historical record confirms that census reports themselves and related monographs were concerned from 1850 with excessive mortality from heart disease of immigrants, particularly of Northern European origin and initially at least, their first-generation native-born children. Our analysis of the electronic database indicates a strong relationship between the percentage of US population foreign born and native born of foreign parentage and age adjusted mortality from heart disease. We identified a lag of 50 years giving the maximum linear correlation coefficient for men ( r2 = 0.92), and for women a shorter lag of 38 years and an earlier decline in Coronary Heart Disease (CHD) rates ( r2 = 0.96). Both the rise and fall of the CHD epidemic over an 80-year period correspond closely to the rise and fall of the foreign population in previous years. For the foreign born only, age adjusted negative binomial general estimated equation (GEE) models calculate the relative risk of dying of heart disease per 10\% increase in proportion foreign born. There is an independent influence for men until 1930 and for women throughout the period from 1910 onwards. We conclude there is an impact of immigration on the pattern of the epidemic, mediated through a combination of factors, such as accumulated life-course susceptibility, deprived socio-economic conditions upon arrival, and the enthusiastic uptake of behaviours related to the classic risk factors of smoking, high saturated fat and salt diet. Our analysis provides a more contextualised understanding of the scale and timing of the epidemic of CHD in the US. © 2006 Elsevier Ltd. All rights reserved.
Lynch J, Smith G, Harper S, Bainbridge K. “Explaining the Social Gradient in Coronary Heart Disease: Comparing Relative and Absolute Risk Approaches.” Journal of Epidemiology and Community Health. 2006;60(5):436–441
Study objectives: There are contradictory perspectives on the importance of conventional coronary heart disease (CHD) risk factors in explaining population levels and social gradients in CHD. This study examined the contribution of conventional CHD risk factors (smoking, hypertension, dyslipidaemia, and diabetes) to explaining population levels and to absolute and relative social inequalities in CHD. This was investigated in an entire population and by creating a low risk sub-population with no smoking, dyslipidaemia, diabetes, and hypertension to simulate what would happen to relative and social inequalities in CHD if conventional risk factors were removed. Design, setting, and participants: Population based study of 2682 eastern Finnish men aged 42, 48, 54, 60 at baseline with 10.5 years average follow up of fatal (ICD9 codes 410-414) and non-fatal (MONICA criteria) CHD events. Main results: In the whole population, 94.6\% of events occurred among men exposed to at least one conventional risk factor, with a PAR of 68\%. Adjustment for conventional risk factors reduced relative social inequality by 24\%. However, in a low risk population free from conventional risk factors, absolute social inequality reduced by 72\%. Conclusions: Conventional risk factors explain the majority of absolute social inequality in CHD because conventional risk factors explain the vast majority of CHD cases in the population. However, the role of conventional risk factors in explaining relative social inequality was modest. This apparent paradox may arise in populations where inequalities in conventional risk factors between social groups are low, relative to the high levels of conventional risk factors within every social group. If the concern is to reduce the overall population health burden of CHD and the disproportionate population health burden associated with the social inequalities in CHD, then reducing conventional risk factors will do the job.
Huynh M, Parker J, Harper S, Pamuk E, Schoendorf K. “Contextual Effect of Income Inequality on Birth Outcomes.” International Journal of Epidemiology. 2005;34(4):888–895
Background: Though associations between income inequality and birth outcome have been suggested, mechanisms underlying this relationship are not known. In this analysis, we examined the relationship between income inequality and preterm birth (PTB) and post-neonatal mortality (PNM) to explore two potential mechanisms - the proposed psychosocial stress and neo-material pathways. Methods: Data on singleton births from 1998 to 2000 were obtained from the CDC's National Center for Health Statistics' Linked Birth and Infant Death files. The Gini Index was utilized to measure income inequality and was divided into tertiles representing high, medium, and low county-level inequality. To determine the association between the birth outcomes and county income inequality and to account for clustering within counties, we employed generalized estimating equation (GEE) modelling. Results: PTB increased from 8.3\% in counties with low income inequality to 10.0\% in counties with high inequality. The Gini Index remained modestly associated with PTB after adjusting for individual level variables and mean county-level per capita income within the total population (AOR: 1.06; 95\% CI 1.03-1.09) as well as within most of the racial/ethnic groups. PNM increased from 1.15 deaths per 1000 live births in low inequality counties to 1.32 in high-inequality counties. However, after adjustment, income inequality was only associated with PNM within the non-Hispanic black population (AOR: 1.20; 95\% CI 1.03-1.39). Conclusions: These findings may provide some support for the association between income inequality and PTB. Further research is required to elucidate the biological mechanisms of income inequality.
Lynch J, Harper S, Kaplan G, Smith G. “Associations between Income Inequality and Mortality among US States: The Importance of Time Period and Source of Income Data.” American Journal of Public Health. 2005;95(8):1424–1430
Objectives. We used census data to examine associations between income inequality and mortality among US states for each decade from 1949 to 1999 and tax return income data to estimate associations for 1989. Methods. Cross-sectional correlation analyses were used to assess income inequality-mortality relationships. Results. Census income analyses revealed little association between income inequality and mortality for 1949, 1959, or 1969. An association emerged for 1979 and strengthened for 1989 but weakened for 1999. When income inequality was based on tax return data, associations were weaker for both 1989 and 1999. Conclusions. The strong association between income inequality and mortality observed among US states for 1989 was not observed for other periods from 1949 through 1999. In addition, when tax return rather than census data were used, the association was weaker for 1989 and 1999. The potential for distal social determinants of population health (e.g., income inequality) to affect mortality is contingent on how such determinants influence levels of proximal risk factors and the time lags between exposure to those risk factors and effects on specific health outcomes.
Kelleher C, Lynch J, Harper S, Tay J, Nolan G. “Hurling Alone? How Social Capital Failed to Save the Irish from Cardiovascular Disease in the United States.” American Journal of Public Health. 2004;94(12):2162–2169
Objectives. We performed a historical review of cardiovascular risk profiles of Irish immigrants to the United States, 1850-1970, in regard to lifestyle, socioeconomic circumstances, and social capital. Methods. We analyzed US Census data from 1850-1970, area-based social and epidemiological data from Boston, data from Ireland's National Nutrition Surveillance Centre, and literature on Irish migration. Results. The Irish were consistently at increased risk of cardiovascular diseases, a risk that related initially to material deprivation, across the life course of at least 2 generations. Conclusions. The principal difference between the Irish and other disadvantaged immigrant groups, such as the Italians, was dietary habits influenced by experiences during the Irish famine. Although there was a psychosocial component to the disadvantage and discrimination they experienced as an ethnic group, the Irish also exhibited strong community networks and support structures that might have been expected to counteract discrimination's negative effects. However, the Irish's high levels of social capital were not protective for cardiovascular disease.
Lynch J, Smith GD, Harper S, Hillemeier M, Ross N, Kaplan GA, Wolfson M. “Is Income Inequality a Determinant of Population Health? Part 1. A Systematic Review.” The Milbank Quarterly. 2004;82(1):5–99
This article reviews 98 aggregate and multilevel studies examining the associations between income inequality and health. Overall, there seems to be little support for the idea that income inequality is a major, generalizable determinant of population health differences within or between rich countries. Income inequality may, however, directly influence some health outcomes, such as homicide in some contexts. The strongest evidence for direct health effects is among states in the United States, but even that is somewhat mixed. Despite little support for a direct effect of income inequality on health per se, reducing income inequality by raising the incomes of the most disadvantaged will improve their health, help reduce health inequalities, and generally improve population health.
Lynch J, Smith G, Harper S, Hillemeier M. “Is Income Inequality a Determinant of Population Health? Part 2. U.S. National and Regional Trends in Income Inequality and Age- and Cause-Specific Mortality.” Milbank Quarterly. 2004;82(2):355–400
This article describes U.S. income inequality and 100-year national and 30-year regional trends in age- and cause-specific mortality. There is little congruence between national trends in income inequality and age- or cause-specific mortality except perhaps for suicide and homicide. The variable trends in some causes of mortality may be associated regionally with income inequality. However, between 1978 and 2000 those regions experiencing the largest increases in income inequality had the largest declines in mortality (r = 0.81, p {$<$} 0.001). Understanding the social determinants of population health requires appreciating how broad indicators of social and economic conditions are related, at different times and places, to the levels and social distribution of major risk factors for particular health outcomes.
Lynch J, Harper S, Smith G, Ross N, Wolfson M, Dunn J. “US Regional and National Cause-Specific Mortality and Trends in Income Inequality: Descriptive Findings.” Demographic Research. 2004;10:183–228
We examined the concordance of income inequality trends with 30-year US regional trends in cause-specific mortality and 100-year trends in heart disease and infant mortality. The evidence suggests that any links between income inequality and population health trends is likely to be complex. The descriptive findings here imply that income inequality would have to be linked and de-linked across different time periods, with different exposures to generate the observed heterogeneous regional and national levels and trends in different causes of death. © 2004 Max-Planck-Gesellschaft.
Hillemeier M, Lynch J, Harper S, Raghunathan T, Kaplan G. “Relative or Absolute Standards for Child Poverty: A State-Level Analysis of Infant and Child Mortality.” American Journal of Public Health. 2003;93(4):652–657
Objectives. The purpose of the present study was to compare the associations of state-referenced and federal poverty measures with states' infant and child mortality rates. Methods. Compressed mortality and Current Population Survey data were used to examine relationships between mortality and (1) state-referenced poverty (percentage of children below half the state median income) and (2) percentage of children below the federal poverty line. Results. State-referenced poverty was not associated with mortality among infants or children, whereas poverty as defined by national standards was strongly related to mortality. Conclusions. Infant and child mortality is more closely tied to families' capacity for meeting basic needs than to relative position within a state's economic hierarchy.
Hillemezer M, Lynch J, Harper S, Casper M. “Measuring Contextual Characteristics for Community Health.” Health Services Research. 2003;38:1645–1717
Objective. To conceptualize and measure community contextual influences on population health and health disparities. Data Sources. We use traditional and nontraditional secondary sources of data comprising a comprehensive array of community characteristics. Study Design. Using a consultative process, we identify 12 overarching dimensions of contextual characteristics that may affect community health, as well as specific subcomponents relating to each dimension. Data Collection. An extensive geocoded library of data indicators relating to each dimension and subcomponent for metropolitan areas in the United States is assembled. Principal Findings. We describe the development of community contextual health profiles, present the rationale supporting each of the profile dimensions, and provide examples of relevant data sources. Conclusions. Our conceptual framework for community contextual characteristics, including a specified set of dimensions and components, can provide practical ways to monitor health-related aspects of the economic, social, and physical environments in which people live. We suggest several guiding principles useful for understanding how aspects of contextual characteristics can affect health and health disparities.
Lynch J, Harper S, Smith G. “Commentary: Plugging Leaks and Repelling Boarders - Where to next for the SS Income Inequality?.” International Journal of Epidemiology. 2003;32(6):1029–1036
Harper S, Lynch J, Hsu W, Everson S, Hillemeier M, Raghunathan T, Salonen J, Kaplan G. “Life Course Socioeconomic Conditions and Adult Psychosocial Functioning.” International Journal of Epidemiology. 2002;31(2):395–403
Background. Various psychosocial factors have been linked to adult physical health and are also associated with socioeconomic position in adulthood. We evaluated the effect of socioeconomic conditions over the life course on measures of psychosocial functioning in adulthood. Methods. Life course socioeconomic position was assessed by retrospective recall of parents' education and occupation when the respondent was age 10, and the respondents' education, occupation, and income in 2585 men from eastern Finland aged 42, 48, 54, and 60 years. Measures of psychosocial functioning were derived from scales measuring cynical hostility, hopelessness, and depressive symptoms. Results. Men with both parents who had less than a primary school education or who both had unskilled manual jobs had higher age-adjusted levels of cynical hostility, hopelessness, and depressive symptoms in adulthood. Mutually adjusted analyses showed that parents' education and the respondents' education, occupation, and income all had statistically independent effects on adult levels of cynical hostility and hopelessness. For instance, men for whom neither parent had completed primary education had a 0.15 standard deviation (P = 0.006) higher cynical hostility score, and a 0.20 standard deviation (P = 0.00018) higher hopelessness score, after adjustment for education, occupation and income. In contrast, depressive symptoms in adulthood were only associated with the respondent's occupation and income. Conclusions. Childhood socioeconomic position was associated with adult psychosocial functioning, but these effects were specific to some aspects of adult psychosocial functioning-cynical hostility and hopelessness, but not depressive symptoms. Adult occupation and income were associated with all measures of psychosocial functioning. In addition to the impact of adult socioeconomic position, some aspects of poor psychosocial functioning in adulthood may also have socioeconomic roots early in life.