Helped into Harm: Mediation of a Housing Voucher Intervention on Mental Health and Substance Use in Boys

Practical causal mediation analysis: extending nonparametric estimators to accommodate multiple mediators and multiple intermediate confounders

Mediation analysis is appealing for its ability to improve understanding of the mechanistic drivers of causal effects, but real-world data complexities challenge its successful implementation, including (i) the existence of post-exposure variables that also affect mediators and outcomes (thus, confounding the mediator-outcome relationship), that may also be (ii) multivariate, and (iii) the existence of multivariate mediators. All three challenges are present in the mediation analysis we consider here, where our goal is to estimate the indirect effects of receiving a Section 8 housing voucher as a young child on the risk of developing a psychiatric mood disorder in adolescence that operate through mediators related to neighborhood poverty, the school environment, and instability of the neighborhood and school environments, considered together and separately. Interventional direct and indirect effects (IDE/IIE) accommodate post-exposure variables that confound the mediator-outcome relationship, but currently, no readily implementable nonparametric estimator for IDE/IIE exists that allows for both multivariate mediators and multivariate post-exposure intermediate confounders. The absence of such an IDE/IIE estimator that can easily accommodate both multivariate mediators and post-exposure confounders represents a significant limitation for real-world analyses, because when considering each mediator subgroup separately, the remaining mediator subgroups (or a subset of them) become post-exposure intermediate confounders. We address this gap by extending a recently developed nonparametric estimator for the IDE/IIE to allow for easy incorporation of multivariate mediators and multivariate post-exposure confounders simultaneously. We apply the proposed estimation approach to our analysis, including walking through a strategy to account for other, possibly co-occurring intermediate variables when considering each mediator subgroup separately.

The social determinants of mental health and disorder: evidence, prevention and recommendations

People exposed to more unfavourable social circumstances are more vulnerable to poor mental health over their life course, in ways that are often determined by structural factors which generate and perpetuate intergenerational cycles of disadvantage and poor health. Addressing these challenges is an imperative matter of social justice. In this paper we provide a roadmap to address the social determinants that cause mental ill health. Relying as far as possible on high-quality evidence, we first map out the literature that supports a causal link between social determinants and later mental health outcomes. Given the breadth of this topic, we focus on the most pervasive social determinants across the life course, and those that are common across major mental disorders. We draw primarily on the available evidence from the Global North, acknowledging that other global contexts will face both similar and unique sets of social determinants that will require equitable attention. Much of our evidence focuses on mental health in groups who are marginalized, and thus often exposed to a multitude of intersecting social risk factors. These groups include refugees, asylum seekers and displaced persons, as well as ethnoracial minoritized groups; lesbian, gay, bisexual, transgender and queer (LGBTQ+) groups; and those living in poverty. We then introduce a preventive framework for conceptualizing the link between social determinants and mental health and disorder, which can guide much needed primary prevention strategies capable of reducing inequalities and improving population mental health. Following this, we provide a review of the evidence concerning candidate preventive strategies to intervene on social determinants of mental health. These interventions fall broadly within the scope of universal, selected and indicated primary prevention strategies, but we also briefly review important secondary and tertiary strategies to promote recovery in those with existing mental disorders. Finally, we provide seven key recommendations, framed around social justice, which constitute a roadmap for action in research, policy and public health. Adoption of these recommendations would provide an opportunity to advance efforts to intervene on modifiable social determinants that affect population mental health.

Using instrumental variables to address unmeasured confounding in causal mediation analysis

Mediation analysis is a strategy for understanding the mechanisms by which interventions affect later outcomes. However, unobserved confounding concerns may be compounded in mediation analyses, as there may be unobserved exposure-outcome, exposure-mediator, and mediator-outcome confounders. Instrumental variables (IVs) are a popular identification strategy in the presence of unobserved confounding. However, in contrast to the rich literature on the use of IV methods to identify and estimate a total effect of a non-randomized exposure, there has been almost no research into using IV as an identification strategy to identify mediational indirect effects. In response, we define and nonparametrically identify novel estimands-double complier interventional direct and indirect effects-when 2, possibly related, IVs are available, one for the exposure and another for the mediator. We propose nonparametric, robust, efficient estimators for these effects and apply them to a housing voucher experiment.

Eviction as a community health exposure

Evidence suggests that being evicted harms health. Largely ignored in the existing literature is the possibility that evictions exert community-level health effects, affecting evicted individuals' social networks and shaping broader community conditions. In this narrative review, we summarize evidence and lay out a theoretical model for eviction as a community health exposure, mediated through four paths: 1) shifting ecologies of infectious disease and health behaviors, 2) disruption of neighborhood social cohesion, 3) strain on social networks, and 4) increasing salience of eviction risk. We describe methods for parsing eviction's individual and contextual effects and discuss implications for causal inference. We conclude by addressing eviction's potentially multilevel consequences for policy advocacy and cost-benefit analyses.

Social Deprivation and Population Density Trajectories Before and After Psychotic Disorder Diagnosis

Importance

People with psychosis are more likely to be born and live in densely populated and socioeconomically deprived environments, but it is unclear whether these associations are a cause or consequence of disorder.

Objective

To investigate whether trajectories of exposure to deprivation and population density before and after diagnosis are associated with psychotic disorders or nonpsychotic bipolar disorder.

Design, Setting, and Participants

This nested case-control study included all individuals born in Sweden between January 1, 1982, and December 31, 2001, diagnosed for the first time with an International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) psychotic disorder or nonpsychotic bipolar disorder between their 15th birthday and cohort exit (December 31, 2016). One sex- and birth year-matched control participant per case was selected. Data analysis was performed from July 2021 to June 2023.

Exposures

The main exposures were quintiles of neighborhood-level deprivation and population density each year from birth to age 14 years and from first diagnosis until cohort exit.

Main Outcomes and Measures

The main outcomes were the odds of a serious mental illness outcome associated with trajectories of deprivation and population density, before and after diagnosis in cases. Group-based trajectory modeling was used to derive trajectories of each exposure in each period. Logistic regression was used to examine associations with outcomes.

Results

A total of 53 458 individuals (median [IQR] age at diagnosis in case patients, 23.2 [15.0-34.8] years; 30 746 [57.5%] female), including 26 729 case patients and 26 729 control participants, were studied. From birth to early adolescence, gradients were observed in exposure to deprivation and population density trajectories during upbringing and psychotic disorder, with those in the most vs least deprived (adjusted odds ratio [AOR], 1.17; 95% CI, 1.08-1.28) and most vs least densely populated (AOR, 1.49; 95% CI, 1.34-1.66) trajectories at greatest risk. A strong upward mobility trajectory to less deprived neighborhoods was associated with similar risk to living in the least deprived trajectory (AOR, 1.01; 95% CI, 0.91-1.12). Only 543 case patients (2.0%) drifted into more deprived areas after diagnosis; people with psychotic disorder were more likely to belong to this trajectory (AOR, 1.38; 95% CI, 1.16-1.65) or remain in the most deprived trajectory (AOR, 1.36; 95% CI, 1.24-1.48) relative to controls. Patterns were similar for nonpsychotic bipolar disorder and deprivation but weaker for population density.

Conclusions and Relevance

In this case-control study, greater exposure to deprivation during upbringing was associated with increased risk of serious mental illness, but upward mobility mitigated this association. People with serious mental illness disproportionately remained living in more deprived areas after diagnosis, highlighting issues of social immobility. Prevention and treatment should be proportionately located in deprived areas according to need.

Efficient and flexible estimation of natural direct and indirect effects under intermediate confounding and monotonicity constraints

Natural direct and indirect effects are mediational estimands that decompose the average treatment effect and describe how outcomes would be affected by contrasting levels of a treatment through changes induced in mediator values (in the case of the indirect effect) or not through induced changes in the mediator values (in the case of the direct effect). Natural direct and indirect effects are not generally point-identified in the presence of a treatment-induced confounder; however, they may be identified if one is willing to assume monotonicity between the treatment and the treatment-induced confounder. We argue that this assumption may be reasonable in the relatively common encouragement-design trial setting, where the intervention is randomized treatment assignment and the treatment-induced confounder is whether or not treatment was actually taken/adhered to. We develop efficiency theory for the natural direct and indirect effects under this monotonicity assumption, and use it to propose a nonparametric, multiply robust estimator. We demonstrate the finite sample properties of this estimator using a simulation study, and apply it to data from the Moving to Opportunity Study to estimate the natural direct and indirect effects of being randomly assigned to receive a Section 8 housing voucher-the most common form of federal housing assistance-on risk developing any mood or externalizing disorder among adolescent boys, possibly operating through various school and community characteristics.

When the Ends do not Justify the Means: Learning Who is Predicted to Have Harmful Indirect Effects

There is a growing literature on finding rules by which to assign treatment based on an individual's characteristics such that a desired outcome under the intervention is maximized. A related goal entails identifying a subpopulation of individuals predicted to have a harmful indirect effect (the effect of treatment on an outcome through mediators), perhaps even in the presence of a predicted beneficial total treatment effect. In some cases, the implications of a likely harmful indirect effect may outweigh an anticipated beneficial total treatment effect, and would motivate further discussion of whether to treat identified individuals. We build on the mediation and optimal treatment rule literatures to propose a method of identifying a subgroup for which the treatment effect through the mediator is expected to be harmful. Our approach is nonparametric, incorporates post-treatment confounders of the mediator-outcome relationship, and does not make restrictions on the distribution of baseline covariates, mediating variables, or outcomes. We apply the proposed approach to identify a subgroup of boys in the MTO housing voucher experiment who are predicted to have a harmful indirect effect of housing voucher receipt on subsequent psychiatric disorder incidence through aspects of their school and neighborhood environments.

The long shadow of residential racial segregation: Associations between childhood residential segregation trajectories and young adult health among Black US Americans

Residential racial segregation is a key manifestation of anti-Black structural racism, thought to be a fundamental cause of poor health; evidence has shown that it yields neighborhood disinvestment, institutional discrimination, and targeting of unhealthy products like tobacco and alcohol. Yet research on the long-term impacts of childhood exposure to residential racial segregation is limited. Here, we analyzed data on 1823 Black participants in the Panel Study of Income Dynamics, estimating associations between childhood segregation trajectories and young adult health. Black young adults who consistently lived in high-segregation neighborhoods throughout childhood experienced unhealthier smoking and drinking behaviors and higher odds of obesity compared to other trajectory groups, including children who moved into or out of high-segregation neighborhoods. Results were robust to controls for neighborhood and family poverty. Findings underscore that for Black children who grow up in segregated neighborhoods, the roots of structurally-determined health inequities are established early in life.