A Quarter Century of Multilevel Models in Social Epidemiology
A Short Story in Three Acts
2023-06-14
- Multilevel methods developed in 1980s in demography (Entwistle), statistics (Wong/Mason), education (Bryk/Raudenbush).
- Critical need for theory.
The development of multi-level approaches in epidemiologic research may facilitate research which elucidates the independent and joint effects of individual and environmental factors on health behaviors and health outcomes.
Epidemiology has lost its way
Social context and ‘population perspective’ has been forgotten.
Multilevel health determinants
Social context a crucial element of conceptual models for ‘social determinants of health’
Why multilevel social epidemiology?
Place-based comparisons of health are revealing (Villermé, Farr, Graunt, Snow, DuBois, many others)
Communities inherently reflect social dynamics.
Host-Agent-Environment (physical and social).
“Population perspective”, contra biomedical individualism.
Early influential studies in social epidemiology
Neighborhood ‘effects’ on violence, mortality
- Focus on mutual adjustment
- Clustering addressed as nuisance
Extended to CVD, low birthweight, other outcomes
- Random effects implementation
- Exploration of multi-level EMM
- Strong theory, field measurements, sophisticated models, potential mechanisms linked to violent crime.
Multilevel analyses showed that a measure of collective efficacy yields a high between-neighborhood reliability and is negatively associated with variations in violence, when individual-level characteristics, measurement error, and prior violence are controlled. Associations of concentrated disadvantage and residential instability with violence are largely mediated by collective efficacy.
Focus on ‘simultaneous’ effects:
By incorporating multiple levels of determination in the study of individual outcomes, multilevel analysis allows for the effects of macro- and micro-level variables as well as their interactions
Potential:
Multilevel analysis is one way to begin to restore a population or societal dimension to epidemiologic research
The ‘Big Idea’:
The big idea is that what matters in determining mortality and health in a society is less the overall wealth of that society and more how evenly wealth is distributed.
- Inequality = contextual, but how?
Traditional measures of association such as odds ratios thus provide an incomplete epidemiological basis for decision making in public health interventions.
Large-scale ‘multilevel’ RCT
~4600 families in high poverty randomized to housing vouchers.
Generated large differences in exposure to high-poverty neighborhoods.
5-year follow-up (2003):
- No impacts on economic self-sufficiency of mothers.
- Other outcomes mixed, some positive, some negative.
Many limitations.
Income inequality: not so bad for health?
Evidence for the income inequality/health link was “slowly dissipating”
Multilevel studies inconsistent in US.
Weak evidence from Europe and Asia.
Individual-level controls matter.
Fixed effects: No. 
Random effects: Yes! 
Neighborhood evidence to 2007
- 86 multilevel papers on neighborhoods
- 80% cross-sectional designs
- Inconsistencies within and across studies.
Neighborhood effects at 20 years
“it is not clear how much we are learning, or whether such lessons are improving population health…experimental evidence of neighborhood effects is mixed, and observational studies too often report mere correlations, side-stepping critical effect identification issues. Since epidemiologists have long known that disadvantaged environments are not healthy, the utility of studies that do not face the difficult methodological challenges is questionable”
Merging of multilevel and causal inference
Greater focus on credible study designs.
- Cluster RCTs
- Quasi-experiments
Utilizing longitudinal data to focus on changes in exposure
Weighting methods to deal with observables and post-exposure covariates
Extensions to mediation
All fit within the scope of multilevel design and analysis
Methods development and clarification
Defining assumptions for causal effects of contextual exposures
Accounting time-varying exposures and confounding in a multilevel context
Conditional vs. marginal effects
Healthy discussion of MTO design / results
Observational data as a neighborhood experiment
Time-varying covariates controlled using IPTW, exposure effects estimated using MSMs.
Can replicate MTO findings.
Found significant lagged effect of living in concentrated disadvantage compared with advantage at wave 1
Recent review of ‘causal analyses’ of neighborhood effects.
Much more mixed.
Evidence of selection and confounding.
Lots of heterogeneity.
Stronger evidence for children than adults.
What about community RCTs?
