A Tutorial for Propensity Score Weighting for Moderation Analysis With Categorical Variables: An Application Examining Smoking Disparities Among Sexual Minority Adults.

OBJECTIVE: To provide step-by-step guidance and STATA and R code for using propensity score (PS) weighting to estimate moderation effects with categorical variables. RESEARCH DESIGN: Tutorial illustrating the key steps for estimating and testing moderation using observational data. Steps include: (1) examining covariate overlap across treatment groups within levels of the moderator; (2) estimating the PS weights; (3) evaluating whether PS weights improved covariate balance; (4) estimating moderated treatment effects; and (5) assessing the sensitivity of findings to unobserved confounding. Our illustrative case study uses data from 41,832 adults from the 2019 National Survey on Drug Use and Health to examine if gender moderates the association between sexual minority status (eg, lesbian, gay, or bisexual [LGB] identity) and adult smoking prevalence. RESULTS: For our case study, there were no noted concerns about covariate overlap, and we were able to successfully estimate the PS weights within each level of the moderator. Moreover, balance criteria indicated that PS weights successfully achieved covariate balance for both moderator groups. PS-weighted results indicated there was significant evidence of moderation for the case study, and sensitivity analyses demonstrated that results were highly robust for one level of the moderator but not the other. CONCLUSIONS: When conducting moderation analyses, covariate imbalances across levels of the moderator can cause biased estimates. As demonstrated in this tutorial, PS weighting within each level of the moderator can improve the estimated moderation effects by minimizing bias from imbalance within the moderator subgroups.

Altered regulation of cardiac ankyrin repeat protein in heart failure.

BACKGROUND: Left ventricular assist devices (LVADs) have revolutionized and improved the care of the sickest heart failure (HF) patients, and it is imperative that they receive appropriate ventricular unloading. Assessing this critical parameter with current methodologies (labs, imaging) is usually suboptimal in this patient population. Hence it is imperative to elucidate the molecular underpinnings involved in ventricular unloading. We have previously identified the cytoskeletal protein ßII spectrin as an essential nodal protein involved in post-translational targeting and ßII spectrin protein levels are significantly altered in multiple forms of human and animal HF. We therefore hypothesized that the ßII spectrin pathway would play a critical role in LVAD remodeling. METHODS: Human heart failure samples were obtained from patients undergoing heart transplantation. Wild type (WT) mice and our previously validated ßII spectrin conditional knock out (ßII cKO) mice were used for animal experiments. Transaortic constriction (TAC) was performed on WT mice. Protein expression was assessed via immunoblots, and protein interactions were assessed with co-immunoprecipitation. Transcriptome analysis was performed using isolated whole hearts from control adult WT mice (n = 3) compared to ßII cKO spectrin mice (n = 3). RESULTS: We report that hearts from mice selectively lacking ßII spectrin expression in cardiomyocytes displayed altered transcriptional regulation of cardiac ankyrin repeat protein (CARP). Notably, CARP protein expression is increased after TAC. Additionally, our findings illustrate that prior to LVAD support, CARP levels are elevated in HF patients compared to normal healthy controls. Further, for the first time in a LVAD population, we show that elevated CARP levels in HF patients return to normal following LVAD support. CONCLUSION: Our findings illustrate that CARP is a dynamic molecule that responds to reduced afterload and stress, and has the potential to serve as a prognostic biomarker to assess for an adequate response to LVAD therapy.