Marginal Structural Model Approach Research Articles

Opioid Coprescription Through Risk Mitigation Guidance and Opioid Agonist Treatment Receipt

At the onset of the COVID-19 pandemic, the government of British Columbia, Canada, released clinical guidance to support physicians and nurse practitioners in prescribing pharmaceutical alternatives to the toxic drug supply. These alternatives included opioids and other medications under the risk mitigation guidance (RMG), a limited form of prescribed safer supply, designed to reduce the risk of SARS-CoV-2 infection and harms associated with illicit drug use. Many clinicians chose to coprescribe opioid medications under RMG alongside opioid agonist treatment (OAT). To examine whether prescription of hydromorphone tablets or sustained-release oral morphine (opioid RMG) and OAT coprescription compared with OAT alone is associated with subsequent OAT receipt. This population-based, retrospective cohort study was conducted from March 27, 2020, to August 31, 2021, included individuals from 10 linked health administrative databases from British Columbia, Canada. Individuals who were receiving OAT at opioid RMG initiation and individuals who were receiving OAT and eligible but unexposed to opioid RMG were propensity score matched at opioid RMG initiation on sociodemographic and clinical variables. Data were analyzed between January 2023 and February 2024. Opioid RMG receipt (≥4 days, 1-3 days, or 0 days of opioid RMG dispensed) in a given week. The main outcome was OAT receipt, defined as at least 1 dispensed dose of OAT in the subsequent week. A marginal structural modeling approach was used to control for potential time-varying confounding. A total of 4636 individuals (2955 [64%] male; median age, 38 [31-47] years after matching) were receiving OAT at the time of first opioid RMG dispensation (2281 receiving ongoing OAT and 2352 initiating RMG and OAT concurrently). Opioid RMG receipt of 1 to 3 days in a given week increased the probability of OAT receipt by 27% in the subsequent week (adjusted risk ratio, 1.27; 95% CI, 1.25-1.30), whereas receipt of opioid RMG for 4 days or more resulted in a 46% increase in the probability of OAT receipt in the subsequent week (adjusted risk ratio, 1.46; 95% CI, 1.43-1.49) compared with those not receiving opioid RMG. The biological gradient was robust to different exposure classifications, and the association was stronger among those initiating opioid RMG and OAT concurrently. This cohort study, which acknowledged the intermittent use of both medications, demonstrated that individuals who were coprescribed opioid RMG had higher adjusted probability of continued OAT receipt or reengagement compared with those not receiving opioid RMG.

Chronic Exposure to Community Violence and Criminal Behavior: A Marginal Structural Modeling Approach

Chronic Exposure to Community Violence and Criminal Behavior: A Marginal Structural Modeling Approach

Marginal structural models with latent class growth analysis of treatment trajectories: Statins for primary prevention among older adults.

Latent class growth analysis is increasingly proposed as a solution to summarize the observed longitudinal treatment into a few distinct groups. When latent class growth analysis is combined with standard approaches like Cox proportional hazards models, confounding bias is not properly addressed because of time-varying covariates that have a double role of confounders and mediators. We propose to use latent class growth analysis to classify individuals into a few latent classes based on their medication adherence pattern, then choose a working marginal structural model that relates the outcome to these groups. The parameter of interest is defined as a projection of the true marginal structural model onto the chosen working model. Simulation studies are used to illustrate our approach and compare it with unadjusted, baseline covariates adjusted, time-varying covariates adjusted, and inverse probability of trajectory groups weighted adjusted models. Our proposed approach yielded estimators with little or no bias and appropriate coverage of confidence intervals in these simulations. We applied our latent class growth analysis and marginal structural model approach to a database comprising information on 52,790 individuals from the province of Quebec, Canada, aged more than 65 and who were statin initiators to estimate the effect of statin-usage trajectories on a first cardiovascular event.

Effect of cytomegalovirus infection on post-transplant hospitalization days among children undergoing allogeneic hematopoietic cell transplantation: A marginal structural model approach.

Cytomegalovirus (CMV) commonly reactivates after allogeneic hematopoietic cell transplant (HCT), potentially leading to CMV disease and significant morbidity and mortality. To reduce morbidity and mortality, many centers conduct weekly CMV blood polymerase chain reaction (PCR) surveillance testing with subsequent initiation of antiviral therapy upon CMV DNAemia detection. However, the impact of CMV DNAemia on subsequent hospitalization risk has not been assessed using models accounting for the time-varying nature of the exposure, outcome, and confounders. All allogeneic HCTs at the Children's Hospital of Philadelphia from January 2004-April 2017 were considered for inclusion. Patients were monitored with CMV surveillance via PCR testing for up to 105 days after HCT receipt. We estimated the association between CMV DNAemia and rate of hospitalization using marginal structural models (MSM). There were 343 allogeneic HCT episodes in 330 with CMV surveillance; median age was 9.0 (range: 0.1-26.2) and 46.5% were female. And 24.1% of HCT patients had at least one positive CMV blood PCR during the follow-up period. Median time to CMV DNAemia detection was 19 days (range: 4-97). The MSM estimated the incidence rate ratios for an association of CMV DNAemia with hospitalization to be 1.24, (95% confidence interval: 1.04-1.47). CMV DNAemia was associated with an increased hospitalization in the post-HCT period. The MSM accounted for time-varying nature of the outcome, exposure and confounders. The findings support prevention of CMV DNAemia in this population. We recommend further investigation into the effectiveness and safety of prophylaxis versus pre-emptive CMV prevention approaches.

The impact of modifiable risk factors in the association between socioeconomic status and sudden cardiac death in a prospective cohort study: equal access to healthcare, unequal outcome.

Low socioeconomic status is associated with all-cause mortality and cardiac risk factors. Furthermore, sudden cardiac death (SCD) is among the leading causes of death in the general population, and identification of high-risk subgroups is needed. The aim of this study was to investigate the association between income and education level and incidence of SCD, and to calculate the impact of modifiable mediating risk factors. Participants in the Copenhagen City Heart Study were followed from 1993-2016. SCD was identified using high quality death certificates, autopsy reports, discharge summaries, and national registry data. Hazard ratios were calculated using Cox proportional hazards regression and adjusted cumulative incidences were predicted using cause-specific Cox models. Mediation analyses were performed using a marginal structural model approach. During 24 years of follow-up, 10,006 participated, whereof 5,514 died during the study period with 822 SCDs. Compared to long education, persons with elementary school level education had an SCD incidence rate ratio (IRR) of 2.48 (CI-95% 1.86-3.31), and low income was likewise associated with an SCD IRR of 2.34 (CI-95% 1.85-2.96) compared to high income. In the association between education and SCD, the combined mediating effect of smoking, physical activity, and BMI accounted for about 20% of the risk differences. We observed an inverse association between both income and education and the risk of SCD, which was only in part explained by common cardiac risk factors, implying that further research into competing causes of SCD is needed and stresses the importance of targeted preventive measures.

A tutorial on a marginal structural modeling approach to mediation analysis in occupational health research: Investigating education, employment quality, and mortality.

Life expectancy inequities between more- and less-educated groups have grown by 1 to 2 years over the last several decades in the United States. Simultaneously, employment conditions for many workers have deteriorated. Researchers hypothesize that these adverse conditions mediate educational inequities in mortality. However, methodological barriers have impeded research on the role of employment conditions and other hazards as mediating factors in health inequities. Indeed, traditional mediation analysis methods are often biased in occupational health settings, including in those with exposure-mediator interactions and mediator-outcome confounders that are caused by exposure. In this paper, we outline-and provide code for-a marginal structural modeling (MSM) approach for estimating total effects and controlled direct effects originally proposed elsewhere, which can be applied to common mediation analysis settings in occupational health research. As an example, we apply our approach to assess the extent to which disparities in employment quality (EQ)-a multidimensional construct characterizing the terms and conditions of the worker-employer relationship-explained educational inequities in mortality in a 1999-2015 US Panel Study of Income Dynamics sample of workers with mortality follow-up through 2017. Under certain strong assumptions described in the text, our estimates suggest that over 70% of the educational inequity in mortality would have been eliminated if EQ had been at the 80th percentile (100th = best) across exposure groups.

Moderate Alcohol Consumption and Depression: A Marginal Structural Model Approach Promoting Causal Inference.

Moderate alcohol consumption is associated with decreased risk for depression, but it remains unclear whether this is a causal relationship or a methodological artifact. To compare the effects of consistent abstinence and occasional, moderate, and above-guideline alcohol consumption throughout early to middle adulthood on depression at age 50, the authors conducted a secondary analysis of the National Longitudinal Survey of Youth 1979 cohort and employed a marginal structural model (MSM) approach. Baseline was set at 1994, when individuals were ages 29-37. The MSM incorporated measurements of alcohol consumption in 1994, 2002, and 2006, baseline and time-varying covariates, and repeated measurements with the Center for Epidemiologic Studies Depression Scale-Short Form (CES-D-SF). A total of 5,667 eligible participants provided valid data at baseline, 3,593 of whom provided valid outcome data. The authors used all observed data to predict CES-D-SF means and rates of probable depression for hypothetical trajectories of consistent alcohol consumption. The results approximated J-curve relationships. Specifically, both consistent occasional and consistent moderate drinkers were predicted to have reduced CES-D-SF scores and rates of probable depression at age 50 compared with consistent abstainers (CES-D-SF scores: b=-0.84, 95% CI=-1.47, -0.11; probable depression: odds ratio=0.58, 95% CI=0.36, 0.88 for consistent occasional drinkers vs. abstainers; CES-D-SF scores: b=-1.08, 95% CI=-1.88, -0.20; probable depression: odds ratio=0.59, 95% CI=0.26, 1.13 for consistent moderate drinkers vs. consistent abstainers). Consistent above-guideline drinkers were predicted to have slightly increased risk compared with consistent abstainers, but this was not significant. In sex-stratified analyses, results were similar for females and males. This study contributes preliminary evidence that associations between moderate alcohol consumption and reduced risk for depression may reflect genuine causal effects. Further research using diverse methodologies that promote causal inference is required.

Longitudinal associations between time-varying insomnia symptoms and all-cause health care services utilization among middle-aged and older adults in the United States.

To examine longitudinal associations between time-varying insomnia symptoms (difficulty initiating sleep, difficulty maintaining sleep, early-morning awakenings, and nonrestorative sleep) and all-cause health care services utilization (HSU), including overnight hospital stays, nursing home stays, and home health care services among middle-aged and older adults. The Health and Retirement Study (HRS), a nationwide, population-representative survey of primarily middle-aged and older adults in the United States. This study is an analysis of prospective data from the HRS for a cohort of 13,168 adults (aged ≥50 years; females=57.7%). Study participants were followed for 16 years. This study focuses on the associations between time-varying insomnia symptoms, both cumulatively and independently, and repeated HSUs. A marginal structural modeling approach was used to capture time-varying biological, psycho-cognitive, and behavioral health factors, and to adjust for selection bias such as differential loss to follow-up. Generalized estimating equations were employed to compute average marginal effects and their 95% confidence intervals. We extracted longitudinal data from 2002 through 2018 waves of the HRS. Experiencing higher numbers of insomnia symptoms on a cumulative scale was associated with higher probabilities of HSU. For instance, the likelihood of overnight hospital stays for individuals reporting one symptom increased from 4.7 percentage points on average (95% CI: 3.7-5.6, p < 0.001), to 13.9 percentage points (95% CI: 10.3-17.5, p < 0.001) for those reporting four symptoms, relative to individuals experiencing no insomnia symptoms. Further, experiencing each of difficulty initiating and maintaining sleep, and nonrestorative sleep, as standalone symptoms, was associated with a higher likelihood of HSU when compared to those not experiencing the symptoms. The results demonstrate the potential consequences and adverse impacts of insomnia symptoms on HSU among middle-aged and older adults. Future investigations should focus on the underlying causes and health systems pathways linking insomnia symptoms to HSU.

Examining the impacts of neighborhood poverty on bodyweight across the BMI distribution: a quantile and MSM modeling approach

Given that the relationships between higher BMI and adverse health outcomes are nonconstant and most pronounced at either ends of the BMI distribution, we assess the association between neighborhood poverty and BMI at multiple points along the BMI distribution. Using data from the 1999 to 2015 Panel Study of Income Dynamics of Black and White adults in the United States, we estimate quantile regression models while jointly applying a marginal structural modeling approach to account for time-varying individual-level factors that may be simultaneously mediators as well as confounders. Neighborhood poverty was not found to be associated with bodyweight at any point along the BMI distribution for Black or White males. However, high neighborhood poverty, compared to low neighborhood poverty, predicted increases in bodyweight for Black females at the lower end of the BMI distribution and for White females at the higher end of the BMI distribution. No association was found between neighborhood poverty and BMI at the mean. Results identify the most vulnerable subgroups, suggesting that White females at the higher end of the BMI distribution as well as Black females at the lower end of the BMI distribution are particularly sensitive to obesogenic environments.

Long-term weight changes and risk of rheumatoid arthritis among women in a prospective cohort: a marginal structural model approach.

To examine the association of long-term weight change with RA risk in a large prospective cohort study. The Nurses' Health Study II started in 1989 (baseline); after exclusions, we studied 108 505 women 25-42 years old without RA. Incident RA was reported by participants and confirmed by medical record review. Body weight was reported biennially through 2015. We investigated two time-varying exposures: weight changes from baseline and from age 18; change was divided into five categories. We used a marginal structural model approach to account for time-varying weight change and covariates. Over 2 583 266 person-years, with a median follow-up time of 25.3 years, 541 women developed RA. Compared with women with stable weight from baseline, weight change was significantly associated with increased RA risk [weight gain 2-<10 kg: RR = 1.98 (95% CI 1.38, 2.85); 10-<20 kg: RR = 3.28 (95% CI 2.20, 4.89); ≥20 kg: RR = 3.81 (95% CI 2.39, 6.07); and weight loss >2 kg: RR = 2.05 (95% CI 1.28, 3.28)]. Weight gain of 10 kg or more from age 18 compared with stable weight was also associated with increased RA risk [10-< 20 kg: RR = 2.12 (95% CI 1.37, 3.27), ≥20 kg: RR = 2.31 (95% CI 1.50, 3.56)]. Consistent findings were observed for seropositive and seronegative RA. Long-term weight gain was strongly associated with increased RA risk in women, with weight gain of ≥20 kg associated with more than a three-fold increased RA risk. Maintenance of healthy weight may be a strategy to prevent or delay RA.

Association Between Enrollment in Maryland's Behavioral Health Homes and Use of Outpatient Mental Health Services.

This study examined the association between enrollment in Maryland's behavioral health home (BHH) program and use of outpatient mental health services among people with serious mental illness. The study sample, drawn from Maryland Medicaid administrative claims data from 2012 through 2017, included 12,232 individuals ages 21-64 with a serious mental illness who were enrolled in a BHH (N=3,319) or never enrolled (N=8,913). A marginal structural modeling approach with inverse probability of treatment weighting was used to examine the association between BHH enrollment and outpatient mental health utilization, measured as the number of visits per 3-month period. BHH enrollment was associated with 0.8 additional outpatient mental health visits per 3-month period, or approximately three additional visits annually. A specialty mental health system-based health home model with a primary goal of improving access to general medical care was associated with increased use of outpatient mental health services.

Does Free-Market Reform Induce Protest? Selection, Post-Treatment Bias, and Depoliticization

AbstractAcross myriad literatures, it is widely held that expanding economic grievances induce violence, protest, or other forms of backlash. In Latin America, where economic liberalization deepened the downturn of the ‘lost decade’ of the 1980s (and 1990s), reform has been tightly associated with protest and mobilization. At the same time, liberal economic reforms have proven to be remarkably durable, even where long-promised benefits are hard to discern. This article makes the case that economically liberal reforms, despite inducing or deepening severe and sustained economic downturns, have actually undermined political protest. Previous work confirming the conventional wisdom foundered on two main methodological problems. First, selection into economic reform was a consequence of the very economic pain and macroeconomic imbalances it also served to induce. Secondly, because of this, these key (macro)economic characteristics are both pre- and post-treatment. Utilizing a marginal structural model approach to assess the impact of economic liberalization on protest outcomes net of this selection process, and the prior history of treatment, the study finds that painful reform reduces political protest even as it heightens grievances. This depoliticizing dynamic helps to explain the surprising durability of liberal reforms in Latin America.

Impact of Ethiopia's productive safety net program on household food security and child nutrition: A marginal structural modeling approach.

Safety nets are expanding in African countries as a policy instrument to alleviate poverty and food insecurity. Whether safety nets have improved household food security and child diet and nutrition in sub-Saharan Africa has not been well documented. This paper takes the case of Ethiopia's Productive Safety Net Program (PSNP) and provides evidence of the impact of safety nets on household food security and child nutritional outcomes. Prior studies provide inconclusive evidence as to whether PSNP has improved household food security and child nutrition. These studies used analytical approaches that correct for selection bias but have overlooked the effect of time-varying confounders that might have resulted in biased estimation. Given that household food security status is both the criteria for participation and one of the desirable outcomes of the program, estimating the causal impact of PSNP on household food security and child nutrition is prone to endogeneity due to selection bias and time-varying confounders. Therefore, the objectives of this paper are (1) to examine the impacts of PSNP on household food security, child meal frequency, child diet diversity, and child anthropometry using marginal structural modeling approach that takes into account both selection bias and time-varying confounders and (2) to shed some light on policy and programmatic implications. Results show that PSNP has not improved household food insecurity, child dietary diversity, and child anthropometry despite its positive impact on child meal frequency. Household participation in PSNP brought a 0.308 unit gain on child meal frequency. Given the consequence of food insecurity and child undernutrition on physical and mental development, intergenerational cycle of poverty, and human capital formation, the program would benefit if it is tailored to nutrition-specific and nutrition-sensitive interventions.

Associations of Endometriosis and Hormone Therapy With Risk of Hyperlipidemia.

Using claims data from the universal health insurance program of Taiwan, we conducted a retrospective cohort study to investigate whether endometriosis and hormone therapy are associated with the risk of developing hyperlipidemia. We selected 9,155 women aged 20-55 years with endometriosis diagnosed during the period 2000-2013 and 212,641 women without endometriosis with a median follow-up time of 7 years. Among patients with endometriosis, 86% of cases were identified on the basis of diagnosis codes with an ultrasound claim, and 14% were defined by diagnostic laparoscopy or surgical treatments. In a Cox proportional hazards model, the adjusted hazard ratio was 1.30 (95% confidence interval (CI): 1.19, 1.41) for all women, 1.04 (95% CI: 0.81, 1.32) for women under 35 years of age, 1.17 (95% CI: 1.03, 1.32) for women aged 35-44 years, and 1.34 (95% CI: 1.18, 1.52) for women aged 45-54 years. Hysterectomy and/or bilateral oophorectomy accounted for 46.9% of the association between endometriosis and hyperlipidemia, and hormone therapy accounted for 21.6%. Among women with endometriosis, the marginal structural model approach adjusting for time-varying hysterectomy/bilateral oophorectomy showed no association between use of hormone medications and risk of hyperlipidemia. We concluded that women with endometriosis are at increased risk of hyperlipidemia; use of hormone therapy by these women was not independently associated with the development of hyperlipidemia.

Investigating the causal effect of socioeconomic status on quality of care under a universal health insurance system - a marginal structural model approach

BackgroundSocial disparities in healthcare persist in the US despite the expansion of Medicaid under the Affordable Care Act. We investigated the causal impact of socioeconomic status on the quality of care in a setting with minimal confounding bias from race, insurance type, and access to care.MethodsWe designed a retrospective population-based study with a random 25% sample of adult Taiwan population enrolled in Taiwan’s National Health Insurance system from 2000 to 2016. Patient’s income levels were categorized into low-income group (<25th percentile) and high-income group (≥25th percentile). We used marginal structural modeling analysis to calculate the odds of hospital admissions for 11 ambulatory care sensitive conditions identified by the Agency for Healthcare Research and Quality and the odds of having an Elixhauser comorbidity index greater than zero for low-income patients.ResultsAmong 2,844,334 patients, those in lower-income group had 1.28 greater odds (95% CI 1.24–1.33) of experiencing preventable hospitalizations, and 1.04 greater odds (95% CI 1.03–1.05) of having a comorbid condition in comparison to high-income group.ConclusionsIncome was shown to be a causal factor in a patient’s health and a determinant of the quality of care received even with equitable access to care under a universal health insurance system. Policies focusing on addressing income as an important upstream causal determinant of health to provide support to patients in lower socioeconomic status will be effective in improving health outcomes for this vulnerable social stratum.

A marginal structural model approach to analyse work-related injuries: an example using data from the health and retirement study

BackgroundBiases may exist in the limited longitudinal data focusing on work-related injuries among the ageing workforce. Standard statistical techniques may not provide valid estimates when the data are time-varying and...

Investigating the Impact of Early Valve Surgery on Survival in Staphylococcus aureus Infective Endocarditis Using a Marginal Structural Model Approach: Results of a Large, Prospectively Evaluated Cohort

The impact of valve surgery on outcomes of Staphylococcus aureus infective endocarditis (SAIE) remains controversial. We tested the hypothesis that early valve surgery (EVS) improves survival by using a novel approach that allows for inclusion of major confounders in a time-dependent way. EVS was defined as valve surgery within 60 days. Univariable and multivariable Cox regression analyses were performed. To account for treatment selection bias, we additionally used a weighted Cox model (marginal structural model) that accounts for time-dynamic imbalances between treatment groups. To address survivor bias, EVS was included as a time-dependent variable. Follow-up of patients was 1 year. Two hundred and three patients were included in the analysis; 50 underwent EVS. All-cause mortality at day 30 was 26%. In the conventional multivariable Cox regression model, the effect of EVS on the death hazard was 0.85 (95% confidence interval [CI], .47-1.52). Using the weighted Cox model, the death hazard rate (HR) of EVS was 0.71 (95% CI, .34-1.49). In subgroup analyses, no survival benefit was observed in patients with septic shock (HR, 0.80 [CI, .26-2.46]), in NVIE (HR, 0.76 [CI, .33-1.71]) or PVIE (HR, 1.02 [CI, .29-3.54]), or in patients with EVS within 14 days (HR, 0.97 [CI, .46-2.07]). Using both a conventional Cox regression model and a weighted Cox model, we did not find a survival benefit for patients who underwent EVS in our cohort. Until results of randomized controlled trials are available, EVS in SAIE should be based on individualized decisions of an experienced multidisciplinary team. German Clinical Trials registry (DRKS00005045).

Causal Model of the Association of Social Support With Antepartum Depression: A Marginal Structural Modeling Approach.

We used marginal structural models to evaluate associations of social support with antepartum depression in late pregnancy, if everyone had had high social support both before pregnancy and during early pregnancy, compared with having low social support at one of the 2 time points or low social support at both time points. In 2012-2014, pregnant Peruvian women (n = 3,336) were recruited into a prospective cohort study (at a mean gestational age of 9 weeks). A follow-up interview (n = 2,279) was conducted (at 26-28 weeks of gestation). Number of available support providers and satisfaction with social support were measured using Sarason Social Support Questionnaire-6. Depression was measured using the Edinburgh Postnatal Depression Scale. Low number of support providers at both time points was associated with increased risk of depression (odds ratio = 1.62, 95% confidence interval: 1.12, 2.34). The association for low satisfaction at both time points was marginally significant (odds ratio = 1.41, 95% confidence interval: 0.99, 1.99). Depression risk was not significantly higher for women who reported high social support at one of the 2 time points. Our study reinforces the importance of assessing social support before and during pregnancy and underscores the need for future interventions targeted at increasing the number of support providers to prevent antepartum depression.

Severe COPD Exacerbation Risk and Long-Acting Bronchodilator Treatments: Comparison of Three Observational Data Analysis Methods.

ObjectiveResults from three observational methods for assessing effectiveness of long-acting bronchodilator therapies for reducing severe exacerbations of chronic obstructive pulmonary disease (COPD) were compared: intent-to-treat (ITT), as protocol (AP), and an as-treated analysis that utilized a marginal structural model (MSM) incorporating time-varying covariates related to treatment adherence and moderate exacerbations.Study Design and SettingSevere exacerbation risk was assessed over a 2-year period using claims data for patients aged ≥40 years who initiated long-acting muscarinic antagonist (LAMA), inhaled corticosteroid/long-acting beta-agonist (ICS/LABA), or triple therapy (LAMA + ICS/LABA).ResultsA total of 5475 COPD patients met inclusion criteria. Six months post-initiation, 53.5 % of patients discontinued using any therapy. The ITT analysis found an increased severe exacerbation risk for triple therapy treatment (hazard ratio [HR] 1.24; 95 % confidence interval [CI] 1.00–1.53). No increased risk was found in the AP (HR 1.00; 95 % CI 0.73–1.36), or MSM analyses (HR 1.11; 95 % CI 0.68–1.81). The MSM highlighted important associations among post-index events.ConclusionNeglecting to adjust for treatment discontinuation may produce biased risk estimates. The MSM approach is a promising tool to compare chronic disease management by illuminating relationships between treatment decisions, adherence, patient choices, and outcomes.Electronic supplementary materialThe online version of this article (doi:10.1007/s40801-015-0025-6) contains supplementary material, which is available to authorized users.

Marginal Structural Models: An Application to Incarceration and Marriage During Young Adulthood

Advanced methods for panel data analysis are commonly used in research on family life and relationships, but the fundamental issue of simultaneous time‐dependent confounding and mediation has received little attention. In this article the authors introduce inverse‐probability‐weighted estimation of marginal structural models, an approach to causal analysis that (unlike conventional regression modeling) appropriately adjusts for confounding variables on the causal pathway linking the treatment with the outcome. They discuss the need for marginal structural models in social science research and describe their estimation in detail. Substantively, the authors contribute to the ongoing debate on the effects of incarceration on marriage by applying a marginal structural model approach to panel data from the National Longitudinal Survey of Youth 1997 (N = 4,781). In line with the increasing evidence on the collateral consequences of contact with the criminal justice system, the authors find that incarceration is associated with reduced chances of entering marriage.