%0 Journal Article %J Statistics in Medicine %D 2021 %T Accounting for selection bias due to death in estimating the effect of wealth shock on cognition for the Health and Retirement Study. %A Yaoyuan V. Tan %A Carol A C Flannagan %A Lindsay R Pool %A Michael R. Elliott %K Bayesian additive regression trees %K Causal inference %K Missing data %X

The Health and Retirement Study (HRS) is a longitudinal study of U.S. adults enrolled at age 50 and older. We were interested in investigating the effect of a sudden large decline in wealth on the cognitive ability of subjects measured using a dataset provided composite score. However, our analysis was complicated by the lack of randomization, time-dependent confounding, and a substantial fraction of the sample and population will die during follow-up leading to some of our outcomes being censored. The common method to handle this type of problem is marginal structural models (MSM). Although MSM produces valid estimates, this may not be the most appropriate method to reflect a useful real-world situation because MSM upweights subjects who are more likely to die to obtain a hypothetical population that over time, resembles that would have been obtained in the absence of death. A more refined and practical framework, principal stratification (PS), would be to restrict analysis to the strata of the population that would survive regardless of negative wealth shock experience. In this work, we propose a new algorithm for the estimation of the treatment effect under PS by imputing the counterfactual survival status and outcomes. Simulation studies suggest that our algorithm works well in various scenarios. We found no evidence that a negative wealth shock experience would affect the cognitive score of HRS subjects.

%B Statistics in Medicine %V 40 %P 2613-2625 %G eng %N 11 %R 10.1002/sim.8921 %0 Report %D 2018 %T Accounting for selection bias due to death in estimating the effect of wealth shock on cognition for the Health and Retirement Study %A Yaoyuan V. Tan %A Carol A C Flannagan %A Lindsay R Pool %A Michael R. Elliott %K Cognition & Reasoning %K Survey Methodology %K Wealth Shocks %X The Health and Retirement Study is a longitudinal study of US adults enrolled at age 50 and older. We were interested in investigating the effect of a sudden large decline in wealth on the cognitive score of subjects. Our analysis was complicated by the lack of randomization, confounding by indication, and a substantial fraction of the sample and population will die during follow-up leading to some of our outcomes being censored. Common methods to handle these problems for example marginal structural models, may not be appropriate because it upweights subjects who are more likely to die to obtain a population that over time resembles that would have been obtained in the absence of death. We propose a refined approach by comparing the treatment effect among subjects who would survive under both sets of treatment regimes being considered. We do so by viewing this as a large missing data problem and impute the survival status and outcomes of the counterfactual. To improve the robustness of our imputation, we used a modified version of the penalized spline of propensity methods in treatment comparisons approach. We found that our proposed method worked well in various simulation scenarios and our data analysis. %B Statistics > Applications %I arXiv.org %C Ithaca %8 12/2018 %G eng %U https://arxiv.org/abs/1812.08855 %0 Journal Article %J JAMA %D 2018 %T Association of a Negative Wealth Shock With All-Cause Mortality in Middle-aged and Older Adults in the United States. %A Lindsay R Pool %A Sarah A. Burgard %A Belinda L Needham %A Michael R. Elliott %A Kenneth M. Langa %A Carlos F. Mendes de Leon %K Mortality %K NDI %K Wealth Shocks %X

Importance: A sudden loss of wealth-a negative wealth shock-may lead to a significant mental health toll and also leave fewer monetary resources for health-related expenses. With limited years remaining to regain lost wealth in older age, the health consequences of these negative wealth shocks may be long-lasting.

Objective: To determine whether a negative wealth shock was associated with all-cause mortality during 20 years of follow-up.

Design, Setting, and Participants: The Health and Retirement Study, a nationally representative prospective cohort study of US adults aged 51 through 61 years at study entry. The study population included 8714 adults, first assessed for a negative wealth shock in 1994 and followed biennially through 2014 (the most recent year of available data).

Exposures: Experiencing a negative wealth shock, defined as a loss of 75% or more of total net worth over a 2-year period, or asset poverty, defined as 0 or negative total net worth at study entry.

Main Outcomes and Measures: Mortality data were collected from the National Death Index and postmortem interviews with family members. Marginal structural survival methods were used to account for the potential bias due to changes in health status that may both trigger negative wealth shocks and act as the mechanism through which negative wealth shocks lead to increased mortality.

Results: There were 8714 participants in the study sample (mean [SD] age at study entry, 55 [3.2] years; 53% women), 2430 experienced a negative wealth shock during follow-up, 749 had asset poverty at baseline, and 5535 had continuously positive wealth without shock. A total of 2823 deaths occurred during 80 683 person-years of follow-up. There were 30.6 vs 64.9 deaths per 1000 person-years for those with continuously positive wealth vs negative wealth shock (adjusted hazard ratio [HR], 1.50; 95% CI, 1.36-1.67). There were 73.4 deaths per 1000 person-years for those with asset poverty at baseline (adjusted HR, 1.67; 95% CI, 1.44-1.94; compared with continuously positive wealth).

Conclusions and Relevance: Among US adults aged 51 years and older, loss of wealth over 2 years was associated with an increased risk of all-cause mortality. Further research is needed to better understand the possible mechanisms for this association and determine whether there is potential value for targeted interventions.

%B JAMA %V 319 %P 1341-1350 %G eng %N 13 %1 http://www.ncbi.nlm.nih.gov/pubmed/29614178?dopt=Abstract %R 10.1001/jama.2018.2055