Estimating Causal Effects in Mediation Analysis using Propensity Scores

Intergenerational Support and Second-Child Fertility Intention in the Chinese Sandwich Generation: The Parallel Mediation Model of Double Burnout

In 2013, the Chinese government implemented a two-child policy to increase the country’s fertility rate. However, the persistently low rates necessitated other measures to boost fertility. This study empirically investigated the association between intergenerational support and second-child fertility intention in the Chinese sandwich generation and demonstrated the mediating role of parental burnout and burnout in caring for grandparents. Survey data collected at Time 1, Time 2, and Time 3 included 2939 participants from different regions of China. Before analyzing the data, coarsened exact matching and propensity score matching was conducted to reduce sampling bias. Regression analysis results indicated that intergenerational support has a significant total positive effect on second-child fertility intention. Furthermore, mediation path analysis revealed that parental burnout and burnout in caring for grandparents play significant but opposite directional mediating roles in the association between intergenerational support and second-child fertility intention. Sensitivity analysis using different calipers yielded similar results. These results indicated that second-child fertility intention can be increased among the Chinese sandwich generation with intergenerational support, by mitigating parental burnout. However, intergenerational support did not alleviate burnout in caring for grandparents in the sandwich generation; therefore, formal older adult care policies are required to help the sandwich generation experience lower burnout, while receiving intergenerational support.

Instrumental variable-based high-dimensional mediation analysis with unmeasured confounders for survival data in the observational epigenetic study

Background: High dimensional mediation analysis is frequently conducted to explore the role of epigenetic modifiers between exposure and health outcome. However, the issue of high dimensional mediation analysis with unmeasured confounders for survival analysis in observational study has not been well solved. Methods: In this study, we proposed an instrumental variable based approach for high dimensional mediation analysis with unmeasured confounders in survival analysis for epigenetic study. We used the Sobel's test, the Joint test, and the Bootstrap method to test the mediation effect. A comprehensive simulation study was conducted to decide the best test strategy. An empirical study based on DNA methylation data of lung cancer patients was conducted to illustrate the performance of the proposed method. Results: Simulation study suggested that the proposed method performed well in the identifying mediating factors. The estimation of the mediation effect by the proposed approach is also reliable with less bias compared with the classical approach. In the empirical study, we identified two DNA methylation signatures including cg21926276 and cg26387355 with a mediation effect of 0.226 (95%CI: 0.108-0.344) and 0.158 (95%CI: 0.065-0.251) between smoking and lung cancer using the proposed approach. Conclusion: The proposed method obtained good performance in simulation and empirical studies, it could be an effective statistical tool for high dimensional mediation analysis.

Effect of Health Shocks on Poverty Status in South Korea: Exploring the Mechanism of Medical Impoverishment

BACKGROUND

South Korea has the highest out-of-pocket burden for medical expenses among the Organisation for Economic Co-operation and Development (OECD) member countries and has no formal sickness benefit system, along with United States and Switzerland, greatly increasing the risk of poverty due to a sudden illness.

METHODS

We identify the causal effect of health shocks on poverty status and explore the mechanisms of medical impoverishment by analyzing longitudinal data from 13 670 households that participated in the representative Korean Welfare Panel Study (KOWEPS) from 2007 to 2016. In this study, we define a health shock as a case in which no household members were hospitalized in the previous year, but together they had more than 30 days of hospitalization in this year. The propensity score matching method was combined with a mediation analysis in this work.

RESULTS

The proportion of households in absolute poverty increased by 4.6-8.0 percentage points among households that experienced a health shock compared with matched controls. The selection effects due to health shock were estimated to be 5.6-8.2 percentage points. On average, a sudden hospitalization reduces annual non-medical expenditures and equivalized disposable income by just over 3.2 million KRW (2500 USD) and 1.2 million KRW (1000 USD), respectively. Health shock induces impoverishment after one year through both the medical expense and work capacity pathways, which explain 12.8% and 12.8% of the total effect, respectively. However, when we decompose the mediation effect of a health shock on poverty status after two years, we find that a health shock leads to poverty mainly through labor force nonparticipation (9.9%).

CONCLUSION

Income stabilizing scheme to protect households that experience a health shock should be introduced as a policy alternative to confront the issue of medical impoverishment.

Linear high-dimensional mediation models adjusting for confounders using propensity score method

High-dimensional mediation analysis has been developed to study whether epigenetic phenotype in a high-dimensional data form would mediate the causal pathway of exposure to disease. However, most existing models are designed based on the assumption that there are no confounders between the exposure, the mediators, and the outcome. In practice, this assumption may not be feasible since high-dimensional mediation analysis (HIMA) tends to be observational where a randomized controlled trial (RCT) cannot be conducted for some economic or ethical reasons. Thus, to deal with the confounders in HIMA cases, we proposed three propensity score-related approaches named PSR (propensity score regression), PSW (propensity score weighting), and PSU (propensity score union) to adjust for the confounder bias in HIMA, and compared them with the traditional covariate regression method. The procedures mainly include four parts: calculating the propensity score, sure independence screening, MCP (minimax concave penalty) variable selection, and joint-significance testing. Simulation results show that the PSU model is the most recommended. Applying our models to the TCGA lung cancer dataset, we find that smoking may lead to lung disease through the mediation effect of some specific DNA-methylation sites, including site Cg24480765 in gene RP11-347H15.2 and site Cg22051776 in gene KLF3.

The Effects of Charitable Giving on Life Satisfaction of Older Korean Adults: The Moderating Role of Relationship Satisfaction and Social Trust

Objectives: This study investigates the effects of charitable giving on the life satisfaction of older Korean adults, examining the moderating role of relationship satisfaction and social trust, as the indicators for social capital. Methods: Nationally representative sample of individuals aged 65 to100 ( N = 8,359) from the 2019 Social Survey was used for the analyses. Results: The results from Coarsened Exact Matching and Structural Equation Modeling show that charitable giving positively affects older Korean adults’ life satisfaction. Moreover, the results of moderation analyses suggest that the effects of charitable giving on life satisfaction are stronger for those with lower levels of relationship satisfaction and social trust. Discussion: The results imply that social capital, such as relationship satisfaction or social trust, needs to be taken into consideration when exploring the effects of charitable giving in later life.

Assessing Potential Outcomes Mediation in HIV Interventions

Knowledge of causal processes through mediation analysis can help improve the effectiveness and reduce costs of public health programs, like HIV prevention and treatment interventions. Advancements in mediation using the potential outcomes framework provide a method for estimating the causal effect of interventions on outcomes via a mediating variable. The purpose of this paper is to provide practical information about mediation and the potential outcomes framework that can enhance data analysis and causal inference for intervention studies. Causal mediation effects are defined and then estimated using data from an HIV intervention randomized trial among people who inject drugs (PWID) in Ukraine. Results from a potential outcomes mediation analysis show that the intervention had a total causal effect on incident HIV infection such that participants in the experimental group were 36% less likely to become infected during the 12-month study than those in the control arm, but that neither self-efficacy nor network communication mediated this effect. Because neither putative mediator was significant, measurement and confounding issues should be investigated to rule out these mediators. Other putative mediators, such as injection frequency, route of administration, or HIV knowledge can be considered. Future research is underway to examine additional, multiple mediators explaining efficacy of the current intervention and sensitivity to confounding effects.

High-Dimensional Mediation Analysis With Confounders in Survival Models

Mediation analysis is a common statistical method for investigating the mechanism of environmental exposures on health outcomes. Previous studies have extended mediation models with a single mediator to high-dimensional mediators selection. It is often assumed that there are no confounders that influence the relations among the exposure, mediator, and outcome. This is not realistic for the observational studies. To accommodate the potential confounders, we propose a concise and efficient high-dimensional mediation analysis procedure using the propensity score for adjustment. Results from simulation studies demonstrate the proposed procedure has good performance in mediator selection and effect estimation compared with methods that ignore all confounders. Of note, as the sample size increases, the performance of variable selection and mediation effect estimation is as well as the results shown in the method which include all confounders as covariates in the mediation model. By applying this procedure to a TCGA lung cancer data set, we find that lung cancer patients who had serious smoking history have increased the risk of death via the methylation markers cg21926276 and cg20707991 with significant hazard ratios of 1.2093 (95% CI: 1.2019-1.2167) and 1.1388 (95% CI: 1.1339-1.1438), respectively.

Rural-to-urban migration, discrimination experience, and health in China: Evidence from propensity score analysis

This research examines how rural-to-urban migration influences health through discrimination experience in China after considering migration selection bias. We conducted propensity score matching (PSM) to obtain a matched group of rural residents and rural-to-urban migrants with a similar probability of migrating from rural to urban areas using data from the 2014 China Family Panel Studies (CFPS). Regression and mediation analyses were performed after PSM. The results of regression analysis after PSM indicated that rural-to-urban migrants reported more discrimination experience than rural residents, and those of mediation analysis revealed discrimination experience to exert negative indirect effects on the associations between rural-to-urban migration and three measures of health: self-reported health, psychological distress, and physical discomfort. Sensitivity analysis using different calipers yielded similar results. Relevant policies and practices are required to respond to the unfair treatment and discrimination experienced by this migrant population.

Adult body height as a mediator between early-life conditions and socio-economic status: the case of the Dutch Potato Famine, 1846-1847

Adult body height appears to be a relatively accurate summary variable of early-life exposures' influence on health, and may be a useful indicator of health in populations where more traditional health-related indicators are lacking. In particular, previous studies have shown a strong, positive relationship between environmental conditions in early life (particularly nutritional availability and the disease environment) and adult height. Research has also demonstrated positive associations between height and socioeconomic status. We therefore hypothesize that height mediates the relationship between early-life conditions and later-life socio-economic outcomes. We also hypothesize that the period of exposure in early life matters, and that conditions during pregnancy or the first years of life and/or the years during puberty have the largest effects on height and socio-economic status. To test these relationships, we use a sample of 1817 Dutch military conscripts who were exposed during early life to the Dutch Potato Famine (1846-1847). We conduct mediation analyses using structural equation modelling, and test seven different time periods in early-life. We use potato prices and real wages to proxy early-life environmental conditions, and occupational status (using the HISCAM scale) to proxy socioeconomic status. We find no evidence of mediation, partial or full, in any models. However, there are significant relationships between potato prices in adolescence, height and socio-economic status. To determine causality in these relationships, further research is needed.

Evaluating a Method to Estimate Mediation Effects With Discrete-Time Survival Outcomes

The utility of evaluating mediation effects spans across research domains. The model facilitates investigation of underlying mechanisms of event timing and, as such, has the potential to help strengthen etiological research and inform intervention work that incorporates the evaluation of mediating variables. In order for the analyses to be maximally useful however, it is critical to employ methodology appropriate for the data under investigation. The purpose of this paper is to evaluate a regression-based approach to estimating mediation effects with discrete-time survival outcomes. We empirically evaluate the performance of the discrete-time survival mediation model in a statistical simulation study, and demonstrate that results are functionally equivalent to estimates garnered from a potential-outcomes framework. Simulation results indicate that parameter estimates of mediation in the model were statistically accurate and precise across the range of examined conditions. Type 1 error rates were also tolerable in the conditions studied. Adequate power to detect effects in the model, with binary X and continuous M variables, required effect sizes of the mediation paths to be medium or large. Possible extensions of the model are also considered.

Marital Status as a Partial Mediator of the Associations Between Young Adult Substance Use and Subsequent Substance Use Disorder: Application of Causal Inference Methods

OBJECTIVE

Young adult substance use is linked with the risk of substance use disorders (SUDs) later in adulthood. Marriage may be part of this pathway because of both selection effects (early substance use reducing marriage) and socialization effects (marriage reducing later substance use and disorder). We examine whether marital status mediates the association between young adult substance use and subsequent SUDs, using causal mediation methods to strengthen inferences.

METHOD

Using panel data from high school seniors in 1990-1998, we examined whether the effects of two exposures (level of alcohol/marijuana use at age 19/20) on the outcomes (alcohol use disorder [AUD]/ marijuana use disorder [MUD], nondisordered use, or abstinence at age 35) were mediated by marital status at age 29/30. Propensity score weights adjusted for potential confounding regarding both the exposures and the mediator.

RESULTS

Moderate and heavy alcohol/marijuana use at age 19/20 were associated with higher odds of AUD/MUD and lower odds of abstinence, each relative to nondisordered use, at age 35. The association between heavy alcohol use at age 19/20 and subsequent AUD was partially mediated by being unmarried at age 29/30; the associations between moderate and heavy marijuana use at age 19/20 and subsequent marijuana abstinence were partially mediated by being unmarried at age 29/30.

CONCLUSIONS

Both selection and socialization effects related to marriage explain the perpetuation of substance use behaviors across adulthood. Selection effects on marriage seem to occur at different thresholds for young adult alcohol and marijuana use.

Marital Status as a Partial Mediator of the Associations Between Young Adult Substance Use and Subsequent Substance Use Disorder: Application of Causal Inference Methods

OBJECTIVE

Young adult substance use is linked with the risk of substance use disorders (SUDs) later in adulthood. Marriage may be part of this pathway because of both selection effects (early substance use reducing marriage) and socialization effects (marriage reducing later substance use and disorder). We examine whether marital status mediates the association between young adult substance use and subsequent SUDs, using causal mediation methods to strengthen inferences.

METHOD

Using panel data from high school seniors in 1990-1998, we examined whether the effects of two exposures (level of alcohol/marijuana use at age 19/20) on the outcomes (alcohol use disorder [AUD]/ marijuana use disorder [MUD], nondisordered use, or abstinence at age 35) were mediated by marital status at age 29/30. Propensity score weights adjusted for potential confounding regarding both the exposures and the mediator.

RESULTS

Moderate and heavy alcohol/marijuana use at age 19/20 were associated with higher odds of AUD/MUD and lower odds of abstinence, each relative to nondisordered use, at age 35. The association between heavy alcohol use at age 19/20 and subsequent AUD was partially mediated by being unmarried at age 29/30; the associations between moderate and heavy marijuana use at age 19/20 and subsequent marijuana abstinence were partially mediated by being unmarried at age 29/30.

CONCLUSIONS

Both selection and socialization effects related to marriage explain the perpetuation of substance use behaviors across adulthood. Selection effects on marriage seem to occur at different thresholds for young adult alcohol and marijuana use.

Age variations in cohort differences in the United States: Older adults report fewer constraints nowadays than those 18 years ago, but mastery beliefs are diminished among younger adults

Life Span psychological and life course sociological perspectives have long acknowledged that individual functioning is shaped by historical and sociocultural contexts. Secular increases favoring later-born cohorts are widely documented for fluid cognitive performance and well-being (among older adults). However, little is known about secular trends in other key resources of psychosocial function such as perceptions of control and whether historical changes have occurred in young, middle-aged, and older adults alike. To examine these questions, we compared data from two independent national samples of the Midlife in the United States survey obtained 18 years apart (1995/96 vs. 2013/14) and identified case-matched cohorts (per cohort, n = 2,223, aged = 23-75 years) based on age and gender. We additionally examined the role of economic resources for cohort differences in perceived mastery and constraints. Results revealed that older adults in later-born cohorts reported perceiving fewer constraints than did matched controls 18 years ago, with such positive secular trends being particularly pronounced among women. In contrast, younger adults reported perceiving more constraints in later-born cohorts than those 18 years ago and also reported perceiving lower mastery. We conclude from our national U.S. sample that secular trends generalize to central psychosocial resources across adulthood, such as perceptions of control, but are not unanimously positive. We discuss possible underlying mechanisms and practical implications. (PsycINFO Database Record

Historical trends in modifiable indicators of cardiovascular health and self-rated health among older adults: Cohort differences over 20 years between the Berlin Aging Study (BASE) and the Berlin Aging Study II (BASE-II)

Background

The last decades have seen great advances in the understanding, treatment, and prevention of cardiovascular disease (CVD). Although mortality rates due to CVD have declined significantly in the last decades, the burden of CVD is still high, particularly in older adults. This raises the question whether contemporary populations of older adults are experiencing better or worse objective as well as subjective health than earlier-born cohorts. The aim of this study was to examine differences in modifiable indicators of cardiovascular health (CVH), comparing data obtained 20 years apart in the Berlin Aging Study (BASE, 1990–93) and the Berlin Aging Study II (BASE-II, 2009–2014).

Methods

Serial cross-sectional analysis of 242 propensity-score-matched participants of BASE (born 1907–1922) and BASE-II (born 1925–1942). Body mass index (BMI), blood pressure, total cholesterol, glycated hemoglobin (HbA1c), diet, smoking and physical activity were operationalized according to the “Life’s simple 7“(LS7) criteria of the American Heart Association.

Results

121 matched pairs were identified based on age, sex, and education. In the later-born BASE-II sample, the mean LS7 score was significantly higher than in the earlier-born sample (7.8±1.8 vs. 6.4±2.1, p<0.001), indicating better CVH. In detail, diet, physical activity, smoking, cholesterol, and HbA1c were more favorable, whereas blood pressure was significantly higher in individuals from the later-born cohort. BMI did not differ significantly between the two matched samples. Notably, despite better CVH, later-born individuals (BASE-II) reported lower self-rated health, presumably because of higher health expectations.

Conclusions

Overall, cardiovascular health was significantly better in the later-born cohort, but several notable exceptions exist.

SAS® Macros for Computing Causal Mediated Effects in Two- and Three-Wave Longitudinal Models

Mediation analysis is a statistical technique for investigating the extent to which a mediating variable transmits the effect of an independent variable to a dependent variable. Because it is used in many fields, there have been rapid developments in statistical mediation. The most cutting-edge statistical mediation analysis focuses on the causal interpretation of mediated effects. Causal inference is particularly challenging in mediation analysis because of the difficulty of randomizing subjects to levels of the mediator. The focus of this paper is on updating three existing SAS® macros (%TWOWAVEMED, %TWOWAVEMONTECARLO, and %TWOWAVEPOSTPOWER, presented at SAS® Global Forum 2017) in two important ways. First, the macros are updated to incorporate new cutting-edge methods for estimating longitudinal mediated effects from the Potential Outcomes Framework for causal inference. The two new methods are inverse-propensity weighting, an application of propensity scores, and sequential G-estimation. The causal inference methods are revolutionary because they frame the estimation of mediated effects in terms of differences in potential outcomes, which align more naturally with how researchers think about causal inference. Second, the macros are updated to estimate mediated effects across three waves of data. The combination of these new causal inference methods and three waves of data enable researchers to test how causal mediated effects develop and maintain over time.

Research design issues for evaluating complex multicomponent interventions in neighborhoods and communities

Major advances in population health will not occur unless we translate existing knowledge into effective multicomponent interventions, implement and maintain these in communities, and develop rigorous translational research and evaluation methods to ensure continual improvement and sustainability. We discuss challenges and offer approaches to evaluation that are key for translational research stages 3 to 5 to advance optimized adoption, implementation, and maintenance of effective and replicable multicomponent strategies. The major challenges we discuss concern (a) multiple contexts of evaluation/research, (b) complexity of packages of interventions, and (c) phases of evaluation/research questions. We suggest multiple alternative research designs that maintain rigor but accommodate these challenges and highlight the need for measurement systems. Longitudinal data collection and a standardized continuous measurement system are fundamental to the evaluation and refinement of complex multicomponent interventions. To be useful to T3-T5 translational research efforts in neighborhoods and communities, such a system would include assessments of the reach, implementation, effects on immediate outcomes, and effects of the comprehensive intervention package on more distal health outcomes.

The Effect of Light Rail Transit on Physical Activity: Design and Methods of the Travel-Related Activity in Neighborhoods Study

Background

Use of mass transit has been proposed as a way to incorporate regular physical activity into daily life because transit use typically requires additional travel to access and depart the stop or station. If this additional travel is active, a small but potentially important amount of physical activity can be achieved daily. Although prior research has shown that transit use is associated with physical activity, important questions remain unanswered. Utilizing a major expansion of the Houston, TX, USA light-rail system as a natural experiment, the Houston Travel-Related Activity in Neighborhoods (TRAIN) Study was developed to address these unanswered questions.

Purpose

The purpose of the TRAIN Study is to determine if the development of light-rail lines in Houston, TX, USA will prospectively affect both transit use and physical activity over 4 years. We also aim to understand how contextual effects (i.e., moderators or interaction effects), such as the neighborhood built environment and socioeconomic factors, affect the primary relations under study.

Methods

The TRAIN Study is a longitudinal cohort design, in which participants are recruited at baseline from a 3-mile buffer around each of the three new lines and measured annually four times. Recruitment is accomplished via telephone contact, ads in newspapers and advertising circulars, and targeted community outreach. Data are collected via mail and include questionnaire-assessed factors, such as perceived neighborhood characteristics, attitudes about transportation, demographics, and reported physical activity; a travel diary; and accelerometry. Additionally, field-based neighborhood audits are conducted to capture micro-scale environmental features. To assess macro-scale environmental characteristics, we utilize GIS mapping and spatial analyses. Statistical analyses will be conducted using latent growth curve modeling and discrete choice models, with a focus on identifying moderating factors (i.e., statistical interaction effects). Selection bias will be controlled via propensity score analysis.

Conclusion

The TRAIN study is a unique opportunity to study how a multi-billion dollar investment in mass transit can simultaneously affect transportation needs and physical activity behavior. This comprehensive evaluation will provide needed evidence for policy makers, and can inform health impact assessments of future transportation projects around the world.

Maintaining Perceived Control with Unemployment Facilitates Future Adjustment

Unemployment is a major challenge to individuals' development. An important personal resource to ameliorate the negative impact of unemployment may be perceived control, a general-purpose belief system. Little is known, however, about how perceived control itself changes with the experience of unemployment and what the antecedents, correlates, and consequences of such change in perceived control are in different ages. We use data from the German Socio-Economic Panel Study (N = 413 who experienced unemployment and N = 413 case-matched controls; time period of data collection: 1994 - 1996) to examine whether perceived control changes with unemployment, explore the role of socio-demographic, psychosocial and health factors in moderating such change, and investigate whether levels of perceived control prior to unemployment and unemployment-related change in perceived control predict unemployment-related outcomes up to five years following. Results indicated that, on average, perceived control remained relatively stable with unemployment, and that younger and older workers did not differ in this regard. However, there were sizeable individual differences in change in perceived control, with women and those with fewer years of education experiencing greater unemployment-related declines in perceived control. Lower levels of perceived control prior to unemployment and steeper unemployment-related decrements in perceived control were each associated with a higher risk of remaining unemployed in the 12 months immediately following unemployment. Steeper unemployment-related declines in perceived control also predicted lower life satisfaction up to five years following. We discuss possible pathways by which perceived control may facilitate adjustment to unemployment, consider the role of perceived control for better understanding the dynamics of unemployment, and suggest routes for further more process-oriented inquiry.

On the (In)Validity of Tests of Simple Mediation: Threats and Solutions

Mediation analysis is a popular framework for identifying underlying mechanisms in social psychology. In the context of simple mediation, we review and discuss the implications of three facets of mediation analysis: (a) conceptualization of the relations between the variables, (b) statistical approaches, and (c) relevant elements of design. We also highlight the issue of equivalent models that are inherent in simple mediation. The extent to which results are meaningful stem directly from choices regarding these three facets of mediation analysis. We conclude by discussing how mediation analysis can be better applied to examine causal processes, highlight the limits of simple mediation, and make recommendations for better practice.

Gender Disparities in Lipid-Lowering Therapy in Cardiovascular Disease: Insights from a Managed Care Population

BACKGROUND

Numerous studies have documented the strong inverse relationship between low-density lipoprotein cholesterol (LDL-C) levels and atherosclerotic cardiovascular disease (ASCVD). However, women are less likely to be screened for hypercholesterolemia, receive lipid-lowering therapy (LLT), and achieve optimal LDL-C levels.

MATERIALS AND METHODS

Data were extracted from a U.S. administrative claims database between January 2008 and December 2012 for patients with established ASCVD. The earliest date of valid LDL-C value was defined as the index date. Patients were followed for ±12 months from the index date and were stratified by gender, by baseline LDL-C level, and whether they were initially treated with a LLT then propensity score matched by gender using demographic and clinical characteristics. Both descriptive statistics and logistic regression models were used to explore the association of gender with the frequency of LDL-C monitoring, LLT treatment initiation in initially untreated patients, and prescribing patterns in initially treated patients.

RESULTS

A total of 76,414 subjects with established ASCVD were identified; 42% of the sample was women. In the unmatched cohort, 50.3% of men and 32.0% of women were prescribed a preindex statin (p < 0.0001). Among matched patients (n = 51,764), women initially treated with LLT were significantly less likely to receive a prescription for a higher potency LLT. Even among those with LDL-C levels above 160 mg/dL, women were more likely to discontinue LLT, odds ratio (95% confidence interval) 1.8 (1.2-2.3). Female gender and older age were significant predictors of discontinuation, and the potency of the index medication was the strongest predictor of dose titration. Initially untreated women were less likely to initiate LLT treatment than men, irrespective of index LDL-C levels (p < 0.0001).

CONCLUSIONS

The observed disparities further reinforce the need for targeted efforts to reduce the gender gap for secondary prevention in women at high risk of cardiovascular disease.

Cohort Differences in Psychosocial Function over 20 Years: Current Older Adults Feel Less Lonely and Less Dependent on External Circumstances

BACKGROUND

Lifespan psychological and life course sociological perspectives indicate that individual development is shaped by social and historical circumstances. Increases in fluid cognitive performance over the last century are well documented and researchers have begun examining historical trends in personality and subjective well-being in old age. Relatively less is known about secular changes in other key components of psychosocial function among older adults.

OBJECTIVE

In the present study, we examined cohort differences in key components of psychosocial function, including subjective age, control beliefs, and perceived social integration, as indicated by loneliness and availability of very close others.

METHODS

We compared data obtained 20 years apart in the Berlin Aging Study (in 1990-1993) and the Berlin Aging Study II (in 2013-2014) and identified case-matched cohort groups based on age, gender, cohort-normed education, and marital or partner status (n = 153 in each cohort, mean age = 75 years). In follow-up analyses, we controlled for having lived in former East versus West Germany, physical diseases, cohort-normed household income, cognitive performance, and the presence of a religious affiliation.

RESULTS

Consistently across analyses, we found that, relative to the earlier-born BASE cohort (year of birth: mean = 1916; SD = 3.38 years; range = 1901-1922), participants in the BASE-II sample (year of birth: mean = 1939; SD = 3.22 years; range = 1925-1949) reported lower levels of external control beliefs (d = -1.01) and loneliness (d = -0.63). Cohorts did not differ in subjective age, availability of very close others, and internal control beliefs.

CONCLUSION

Taken together, our findings suggest that some aspects of psychosocial function of older adults have improved across the two recent decades. We discuss the possible role of sociocultural factors that might have led to the observed set of cohort differences.

Evaluating the impact of implementation factors on family-based prevention programming: methods for strengthening causal inference

Despite growing recognition of the important role implementation plays in successful prevention efforts, relatively little work has sought to demonstrate a causal relationship between implementation factors and participant outcomes. In turn, failure to explore the implementation-to-outcome link limits our understanding of the mechanisms essential to successful programming. This gap is partially due to the inability of current methodological procedures within prevention science to account for the multitude of confounders responsible for variation in implementation factors (i.e., selection bias). The current paper illustrates how propensity and marginal structural models can be used to improve causal inferences involving implementation factors not easily randomized (e.g., participant attendance). We first present analytic steps for simultaneously evaluating the impact of multiple implementation factors on prevention program outcome. Then, we demonstrate this approach for evaluating the impact of enrollment and attendance in a family program, over and above the impact of a school-based program, within PROSPER, a large-scale real-world prevention trial. Findings illustrate the capacity of this approach to successfully account for confounders that influence enrollment and attendance, thereby more accurately representing true causal relations. For instance, after accounting for selection bias, we observed a 5% reduction in the prevalence of 11th grade underage drinking for those who chose to receive a family program and school program compared to those who received only the school program. Further, we detected a 7% reduction in underage drinking for those with high attendance in the family program.

Historical improvements in well-being do not hold in late life: Birth- and death-year cohorts in the United States and Germany

One key objective of life span research is to examine how individual development is shaped by the historical time people live in. Secular trends favoring later-born cohorts on fluid cognitive abilities have been widely documented, but findings are mixed for well-being. It remains an open question whether secular increases in well-being seen in earlier phases of life also manifest in the last years of life. To examine this possibility, we made use of longitudinal data obtained from the mid-1980s until the late 2000s in 2 large national samples in the United States (Health and Retirement Study [HRS]) and Germany (German Socio-Economic Panel [SOEP]). We operationally defined historical time from 2 complementary perspectives: birth-year cohorts based on the years in which people were born (earlier: 1930s vs. later: 1940s) and death-year cohorts based on the years in which people died (earlier: 1990s vs. later: 2000s). To control for relevant covariates, we used case-matched groups based on age (at death) and education and covaried for gender, health, and number of observations. Results from both countries revealed that well-being in old age was indeed developing at higher levels among later-born cohorts. However, for later-deceased cohorts, no evidence for secular increases in well-being was found. To the contrary, later-dying SOEP participants reported lower levels of well-being at age 75 and 2 years prior to death and experienced steeper late-life declines. Our results suggest that secular increases in well-being observed in old age do not manifest in late life, where "manufactured" survival may be exacerbating age- and mortality-related declines.

Gang membership and substance use: guilt as a gendered causal pathway

OBJECTIVES

We examine whether anticipated guilt for substance use is a gendered mechanism underlying the noted enhancement effect of gang membership on illegal drug use. We also demonstrate a method for making stronger causal inferences when assessing mediation in the presence of moderation and time-varying confounding.

METHODS

We estimate a series of inverse propensity weighted models to obtain unbiased estimates of mediation in the presence of confounding of the exposure (i.e., gang membership) and mediator (i.e., anticipated guilt) using three waves of data from a multi-site panel study of a law-related education program for youth (N=1,113).

RESULTS

The onset of gang membership significantly decreased anticipated substance use guilt among both male and female respondents. This reduction was significantly associated with increased frequency of substance use only for female respondents, however, suggesting that gender moderates the mechanism through which gang membership influences substance use.

CONCLUSIONS

Criminologists are often concerned with identifying causal pathways for antisocial and/or delinquent behavior, but confounders of the exposure, mediator, and outcome often interfere with efforts to assess mediation. Many new approaches have been proposed for strengthening causal inference for mediation effects. After controlling for confounding using inverse propensity weighting, our results suggest that interventions aimed at reducing substance use by current and former female gang members should focus on the normative aspects of these behaviors.

Mediation of smoking consumption on the association of perception of smoking risks with successful spontaneous smoking cessation

BACKGROUND

The perception of smoking risks has positive and smoking consumption has negative effects on successful smoking cessation, but no information is now available for what is the role of smoking consumption on the relationship between perception of smoking risks and successful smoking cessation.

PURPOSE

To assess the mediation of smoking amount on the association of perception of smoking risks with successful spontaneous smoking cessation.

METHOD

A community-based case-control study was conducted with 294 adult spontaneous successful smoking quitters who have been quitting smoking consecutively for more than 2 years as the cases, and 347 adult failed spontaneous smoking quitters who quitted but relapsed or have been quitting smoking continuously equal to or less than 2 years as the controls. The smoking amount was evaluated by pack-year that was categorized into tertiles and scored inversely. The score of perception of smoking risks was also categorized into tertiles. Propensity score as the covariate in the regression model was used to adjust the potential confounding. The total effect was decomposed into direct effect and indirect (mediating) effect by using logistic regression based on the KHB method proposed by Karlson, Holm, and Breen. The proportion of mediating effect among the total effect is calculated as the indirect effect divided by the total effect.

RESULTS

After adjusting age, profession, education, marital status, and reasons for quitting smoking by using propensity score, the small mediating effect of smoking amount on the association of perception of smoking risks with successful spontaneous smoking cessation was observed, and the mediating effect of smoking amount was 23.75% among the total effect. The dose-response analysis showed that the mediating effect of smoking amount in the groups with middle and high score of perception of smoking risks were 10.15 and 37.84% among their total effect, respectively.

CONCLUSION

This study reveals that smoking amount has only small mediating effect among the total effect of perception of smoking risks on the successful spontaneous smoking cessation; enhancing the smokers' perception of smoking risks could promote their attempt to reduce smoking, so as to increase the probability of successful smoking cessation.

Advances in mediation analysis: a survey and synthesis of new developments

Mediation processes are fundamental to many classic and emerging theoretical paradigms within psychology. Innovative methods continue to be developed to address the diverse needs of researchers studying such indirect effects. This review provides a survey and synthesis of four areas of active methodological research: (a) mediation analysis for longitudinal data, (b) causal inference for indirect effects, (c) mediation analysis for discrete and nonnormal variables, and (d) mediation assessment in multilevel designs. The aim of this review is to aid in the dissemination of developments in these four areas and suggest directions for future research.

Statistical approaches for enhancing causal interpretation of the M to Y relation in mediation analysis

Statistical mediation methods provide valuable information about underlying mediating psychological processes, but the ability to infer that the mediator variable causes the outcome variable is more complex than widely known. Researchers have recently emphasized how violating assumptions about confounder bias severely limits causal inference of the mediator to dependent variable relation. Our article describes and addresses these limitations by drawing on new statistical developments in causal mediation analysis. We first review the assumptions underlying causal inference and discuss three ways to examine the effects of confounder bias when assumptions are violated. We then describe four approaches to address the influence of confounding variables and enhance causal inference, including comprehensive structural equation models, instrumental variable methods, principal stratification, and inverse probability weighting. Our goal is to further the adoption of statistical methods to enhance causal inference in mediation studies.

Incarceration and Post-release Health Behavior

This study investigates the link between incarceration and health behavior among a sample of young adults from the National Longitudinal Study of Adolescent Health (N = 1,670). The association is analyzed using propensity score methods and a strategic comparison group: respondents who have been convicted of crimes, but not incarcerated. Findings suggest that former inmates consume more fast food and have a higher likelihood of smoking than do similarly situated peers. These associations operate partly through increased financial strife and decreased social standing. Given the role of health behavior in predicting future health outcomes, poor health behavior may be a salient force driving health and mortality risk among the formerly incarcerated population.

Improving Our Ability to Evaluate Underlying Mechanisms of Behavioral Onset and Other Event Occurrence Outcomes: A Discrete-Time Survival Mediation Model

The purpose of this article is to introduce and describe a statistical model that researchers can use to evaluate underlying mechanisms of behavioral onset and other event occurrence outcomes. Specifically, the article develops a framework for estimating mediation effects with outcomes measured in discrete-time epochs by integrating the statistical mediation model with discrete-time survival analysis. The methodology has the potential to help strengthen health research by targeting prevention and intervention work more effectively as well as by improving our understanding of discretized periods of risk. The model is applied to an existing longitudinal data set to demonstrate its use, and programming code is provided to facilitate its implementation.

Sensitivity plots for confounder bias in the single mediator model

BACKGROUND

Causal inference continues to be a critical aspect of evaluation research. Recent research in causal inference for statistical mediation has focused on addressing the sequential ignorability assumption; specifically, that there is no confounding between the mediator and the outcome variable.

OBJECTIVES

This article compares and contrasts three different methods for assessing sensitivity to confounding and describes the graphical depiction of these methods.

DESIGN

Two types of data were used to fully examine the plots for sensitivity analysis. The first type was generated data from a single mediator model with a confounder influencing both the mediator and the outcome variable. The second was data from an actual intervention study. With both types of data, situations are examined where confounding has a large effect and a small effect.

SUBJECTS

The nonsimulated data were from a large intervention study to decrease intentions to use steroids among high school football players. We demonstrate one situation where confounding is likely and another situation where confounding is unlikely.

CONCLUSIONS

We discuss how these methods could be implemented in future mediation studies as well as the limitations and future directions for these methods.

Some methodological considerations in theory-based health behavior research

As this special issue shows, much research in social and personality psychology is directly relevant to health psychology. In this brief commentary, we discuss three topics in research methodology that may be of interest to investigators involved in health-related psychological research. The first topic is statistical analysis of mediated and moderated effects. The second is measurement of latent constructs. The third is the Multiphase Optimization Strategy, a framework for translation of innovations from social and personality psychology into behavioral interventions.

Cohorts based on decade of death: no evidence for secular trends favoring later cohorts in cognitive aging and terminal decline in the AHEAD study

Studies of birth-year cohorts examined over the same age range often report secular trends favoring later-born cohorts, who are cognitively fitter and show less steep cognitive declines than earlier-born cohorts. However, there is initial evidence that those advantages of later-born cohorts do not carry into the last years of life, suggesting that pervasive mortality-related processes minimize differences that were apparent earlier in life. Elaborating this work from an alternative perspective on cohort differences, we compared rates of cognitive aging and terminal decline in episodic memory between cohorts based on the year participants had died, earlier (between 1993 and 1999) or later in historical time (between 2000 and 2010). Specifically, we compared trajectories of cognitive decline in 2 death-year cohorts of participants in the Asset and Health Dynamics Among the Oldest Old study that were matched on age at death and education and controlled for a variety of additional covariates. Results revealed little evidence of secular trends favoring later cohorts. To the contrary, the cohort that died in the 2000s showed a less favorable trajectory of age-related memory decline than the cohort that died in the 1990s. In examinations of change in relation to time to death, the cohort dying in the 2000s experienced even steeper terminal declines than the cohort dying in the 1990s. We suggest that secular increases in "manufacturing" survival may exacerbate age- and mortality-related cognitive declines among the oldest old.

Assessing mediation using marginal structural models in the presence of confounding and moderation

This article presents marginal structural models with inverse propensity weighting (IPW) for assessing mediation. Generally, individuals are not randomly assigned to levels of the mediator. Therefore, confounders of the mediator and outcome may exist that limit causal inferences, a goal of mediation analysis. Either regression adjustment or IPW can be used to take confounding into account, but IPW has several advantages. Regression adjustment of even one confounder of the mediator and outcome that has been influenced by treatment results in biased estimates of the direct effect (i.e., the effect of treatment on the outcome that does not go through the mediator). One advantage of IPW is that it can properly adjust for this type of confounding, assuming there are no unmeasured confounders. Further, we illustrate that IPW estimation provides unbiased estimates of all effects when there is a baseline moderator variable that interacts with the treatment, when there is a baseline moderator variable that interacts with the mediator, and when the treatment interacts with the mediator. IPW estimation also provides unbiased estimates of all effects in the presence of nonrandomized treatments. In addition, for testing mediation we propose a test of the null hypothesis of no mediation. Finally, we illustrate this approach with an empirical data set in which the mediator is continuous, as is often the case in psychological research.

Causal mediation of a human immunodeficiency virus preventive intervention

BACKGROUND

Assessing mediation is important because most interventions are designed to affect an intermediate variable or mediator; this mediator, in turn, is hypothesized to affect outcome behaviors. Although there may be randomization to the intervention, randomization to levels of the mediator is not generally possible. Therefore, drawing causal inferences about the effect of the mediator on the outcome is not straightforward.

OBJECTIVES

The aim of this study was to introduce an approach to causal mediation analysis using the potential outcomes framework for causal inference and then discuss this approach in terms of the scientific questions addressed and the assumptions needed for identifying and estimating the effects.

METHODS

The approach is illustrated using data from the Criminal Justice Drug Abuse Treatment Studies: Reducing Risky Relationships-HIV intervention implemented with 243 incarcerated women re-entering the community. The intervention was designed to affect various mediators at 30 days postintervention, including risky relationship thoughts, beliefs, and attitudes, which were then hypothesized to affect engagement in risky sexual behaviors, such as unprotected sex, at 90 days postintervention.

RESULTS

Using propensity score weighting, we found the intervention resulted in a significant decrease in risky relationship thoughts (-0.529, p = .03) and risky relationship thoughts resulted in an increase in unprotected sex (0.447, p < .001). However, the direct effect of the intervention on unprotected sex was not significant (0.388, p = .479).

DISCUSSION

By reducing bias, propensity score models improve the accuracy of statistical analysis of interventions with mediators and allow researchers to determine not only whether their intervention works but also how it works.

Guidelines for the Investigation of Mediating Variables in Business Research

Business theories often specify the mediating mechanisms by which a predictor variable affects an outcome variable. In the last 30 years, investigations of mediating processes have become more widespread with corresponding developments in statistical methods to conduct these tests. The purpose of this article is to provide guidelines for mediation studies by focusing on decisions made prior to the research study that affect the clarity of conclusions from a mediation study, the statistical models for mediation analysis, and methods to improve interpretation of mediation results after the research study. Throughout this article, the importance of a program of experimental and observational research for investigating mediating mechanisms is emphasized.