Objective: To assess the impact of rural‐urban residence on children with obstructive sleep‐disordered breathing (SDB) who were candidates for tonsillectomy with or without adenoidectomy (TA).Study Design: Retrospective cohort study.Setting: Tertiary children’s hospital.Methods: A cohort of otherwise healthy children aged 2 to 18 years with a diagnosis of obstructive SDB between April 2016 and December 2018 who were recommended TA were included. Rural‐urban designation was defined by ZIP code approximation of rural‐urban commuting area codes. The main outcome was association of rurality with time to TA and loss to follow‐up using Cox and logistic regression analyses.Results: In total, 213 patients were included (mean age 6 ± 2.9 years, 117 [55%] male, 69 [32%] rural dwelling). Rural‐dwelling children were more often insured by Medicaid than private insurance (P <. 001) and had a median driving distance of 74.8 vs 16.8 miles (P <. 001) compared to urban‐dwelling patients. The majority (94.9%) eventually underwent recommended TA once evaluated by an otolaryngologist. Multivariable logistic regression analysis did not reveal any significant predictors for loss to follow‐up in receiving TA. Cox regression analysis that adjusted for age, sex, insurance, and race showed that rural‐dwelling patients had a 30% reduction in receipt of TA over time as compared to urban‐dwelling patients (hazard ratio, 0.7; 95% CI, 0.50‐0.99).Conclusion: Rural‐dwelling patients experienced longer wait times and driving distance to TA. This study suggests that rurality should be considered a potential barrier to surgical intervention and highlights the need to further investigate geographic access as an important determinant of care in pediatric SDB.
21 May 2025 in Journal of the Royal Statistical Society Series A: Statistics in Society
Cardiometabolic risk factors (CRFs) during pregnancy are early indicators of maternal diseases, such as stroke and type 2 diabetes. The total number of CRFs typically takes the form of binomial counts that exhibit overdispersion and zero inflation due to correlations among the underlying CRFs. Motivated by an examination of spatiotemporal trends in five CRFs among pregnant women in the U.S. state of South Carolina during the COVID-19 pandemic, we developed a zero-inflated beta-binomial model within a spatiotemporal framework. This model combines a point mass at zero to account for zero inflation and a beta-binomial distribution to model the remaining CRF counts. Given the notable racial disparities in CRFs during pregnancy that vary across the state over time, we incorporate a spatially varying coefficient model to explore the complex relationships between CRFs and geographic and temporal disparities among non-Hispanic White and non-Hispanic Black women. For posterior inference, we developed an efficient hybrid Markov Chain Monte Carlo algorithm that relies on easily sampled Gibbs and Metropolis–Hastings steps. Our analysis of CRFs in South Carolina reveals that certain counties, such as Chesterfield and Clarendon, exhibit gaps in racial health disparities, making them prime candidates for community-level interventions aimed at reducing these disparities.
Preterm delivery (PTD) complications are a major cause of childhood morbidity and mortality. We aimed to assess trends in PTD and small for gestational age (SGA) and whether trends varied between race–ethnic groups in South Carolina (SC). We utilized 2015–2021 SC vital records linked to hospitalization and emergency department records. PTD was defined as clinically estimated gestation less than (<) 37 weeks (wks.) with subgroup analyses of PTD < 34 wks. and < 28 wks. SGA was defined as infants weighing below the 10th percentile for gestational age. This retrospective study included 338,532 (243,010 before the COVID-19 pandemic and 95,522 during the pandemic) live singleton births of gestational age ≥ 20 wks. born to 260,276 mothers in SC. Generalized estimating equations and a change-point during the first quarter of 2020 helped to assess trends. In unadjusted analyses, pre-pandemic PTD showed an increasing trend that continued during the pandemic (relative risk (RR) = 1.04, 95% CI: 1.02–1.06). PTD < 34 wks. rose during the pandemic (RR = 1.07, 95% CI: 1.02–1.12) with a significant change in the slope. Trends in SGA varied by race and ethnicity, increasing only in Hispanics (RR = 1.02, 95% CI: 1.00–1.04) before the pandemic. Our study reveals an increasing prevalence of PTD and a rise in PTD < 34 wks. during the pandemic, as well as an increasing prevalence of SGA in Hispanics during the study period.
7 October 2024 in Statistical Methods in Medical Research
In a recent 12-week smoking cessation trial, varenicline tartrate failed to show significant improvements in enhancing end-of-treatment abstinence when compared with placebo among adolescents and young adults. The original analysis aimed to assess the average effect across the entire population using timeline followback methods, which typically involve overdispersed binomial counts. We instead propose to investigate treatment effect heterogeneity among latent classes of participants using a Bayesian beta-binomial piecewise linear growth mixture model specifically designed to address longitudinal overdispersed binomial responses. Within each class, we fit a piecewise linear beta-binomial mixed model with random changepoints for each study group to detect critical windows of treatment efficacy. Using this model, we can cluster subjects who share similar characteristics, estimate the class-specific mean abstinence trends for each study group, and quantify the treatment effect over time within each class. Our analysis identified two classes of subjects: one comprising high-abstinent individuals, typically young adults and light smokers, in which varenicline led to improved abstinence; and another comprising low-abstinent individuals for whom varenicline showed no discernible effect. These findings highlight the importance of tailoring varenicline to specific participant subgroups, thereby advancing precision medicine in smoking cessation studies.
This paper identifies the causal effect of fund flows on liquidation risk using a policy‑based instrumental variable. Flows are endogenous: stronger funds attract capital and are less likely to fail, so conventional hazard estimates confound funding effects with unobserved quality. To address this, we construct an instrument that interacts a dated policy shock, the CFTC’s 2012 rescission of removing exemption for commodity pool operators, with a fund’s predetermined regulatory exposure and contractual flexibility. The policy differentially affected derivatives‑intensive funds and generated policy‑timed variation in monthly flows, with stronger effects for funds whose contractual terms permit rapid capital adjustment. We implement a two‑stage estimation: a first‑stage panel regression for flows, followed by a second‑stage discrete‑time hazard model for liquidation risk. The CFTC‑based instrument is strong (first‑stage F‑statistic > 10), and placebo lead tests reveal no anticipatory effects. The second‑stage estimates show that instrumented flows significantly reduce the hazard of liquidation. Results are robust to alternative measures of exposure and flexibility. The naïve hazard model that ignores endogeneity yields a negative coefficient but biased inference, understating the protective effect of flows. These results provide causal evidence that exogenous flow shocks materially affect hedge fund survival, identified through a regulatory instrument. This identification corrects the bias inherent in conventional estimates, revealing that naïve models understate the protective role of stable funding. The implications are that investors should evaluate contractual liquidity and regulatory exposure alongside performance, while regulators should recognise that policy changes can generate transitional fragility through the flow channel.