Faculty Bibliography

In observational studies with a survival outcome, treatment initiation may be time dependent, which is likely to be affected by both time-invariant and time-varying covariates. In situations where the treatment is necessary for the study population, all or most subjects may be exposed to the treatment sooner or later. In this scenario, the causal effect of interest is the delay in treatment reception. A simple comparison of those receiving treatment early vs. those receiving treatment late might not be appropriate, as the timing of the treatment reception is not randomized. Extending Lu's matching design with time-varying covariates, we propose a propensity score matching strategy to estimate the treatment delay effect. The goal is to balance the covariate distribution between on-time treatment and delayed treatment groups at each time point using risk set matching. Our simulation study shows that, in the presence of treatment delay effects, the matching-based analyses clearly outperform the conventional regression analysis using the naive Cox proportional hazards model. We apply this method to study the treatment delay effect of 17 alpha-hydroxyprogesterone caproate (17P) for patients with recurrent preterm birth.

OBJECTIVES/OBJECTIVE:We evaluated the effectiveness of Moms2B, a community-based group pregnancy and parenting program, in an effort to assess whether the program improved pregnancy and infant outcomes. METHODS:We conducted a retrospective matched exposure cohort study comparing women exposed to the Moms2B program during pregnancy (two or more prenatal visits) who delivered a singleton live birth or stillbirth (≥ 20 weeks gestation) from 2011-2017 to a closely matched group of women not exposed to the program. Primary outcomes were preterm birth and low birth weight. Propensity score methods were used to provide strong control for confounders. RESULTS:The final analytic file comprised 675 exposed pregnancies and a propensity score-matched group of 1336 unexposed pregnancies. Most of the women were non-Hispanic Black. We found evidence of better outcomes among pregnancies exposed to Moms2B versus unexposed pregnancies, particularly for the primary outcome of low birth weight [9.45% versus 12.00%, respectively, risk difference (RD) = -2.55, 95% confidence interval (CI) = (-5.44, 0.34)]. Point estimates for all adverse pregnancy outcomes uniformly favored exposure to Moms2B. CONCLUSIONS FOR PRACTICE/UNASSIGNED:Our findings suggest that participation in the Moms2B program improves pregnancy and infant outcomes. The program offers an innovative group model of pregnancy and parenting support for women, especially in non-Hispanic Black women with high-risk pregnancies.

Randomized controlled trials are the gold standard of clinical research for comparing therapies in well-defined groups of participants.¹ Randomization avoids confounding due to unmeasured variables or to treatment selection and enables a causal interpretation of the estimated treatment effect. It has long been recognized, however, that standard explanatory clinical trials are slow, costly, and subject to participant selection. To preserve the strengths of randomized trials while mitigating their weaknesses, pragmatic randomized clinical trials emerged; these trials aim to facilitate decision-making rather than explicate a mechanism of action and enroll a diverse set of participants using existing structures and data sources.².

BACKGROUND:Short interpregnancy interval has been associated with increased risk of preterm delivery; recent studies employing within-sibship designs suggest that this risk may be exaggerated. There are unresolved issues regarding properties of this design. OBJECTIVES/OBJECTIVE:To compare directly the results, for short intervals, of between-person and within-sibship analyses when applied to the same target population. METHODS:Cross-sectional data are from the National Survey of Family Growth, a statistically representative survey of women and men in the USA, 2006-2015. Participants provided a complete pregnancy history including outcome, duration and ending date, enabling calculation of interval. Conventional analysis employed log-linear regression, controlling survey design, early life events, demographic variables, pregnancy intendedness, breastfeeding of the previous birth and obstetric history. Within-sibship analyses, utilising conditional log-linear regression, controlled the same variables, except those remaining static within each participant. RESULTS:Among participants with at least three live- or stillbirths, the percentage of pregnancies in each interval, and the percent of deliveries that were preterm following that interval were 9.2%, 14.6% for <6, and 14.7%, 15.4% for 6-11, versus 12.2%, 14.7% for 18-23 months. Among participants with at least three live- or stillborn infants, those in the within-sibship analysis had a higher risk profile than comparably parous, ineligible participants. In a between-participant analysis, among those included in within-sibship models, the adjusted risk ratios (vs 18-23 months) for preterm delivery for intervals <6 and 6-11 months were 0.74 (95% CI 0.63, 0.88) and 0.85 (95% CI 0.74, 0.98). The corresponding risk ratios were 0.56 (95% CI 0.14, 2.30) and 0.49 (95% CI 0.13, 1.80) for those ineligible for the within-sibship models. CONCLUSIONS:When comparable analyses were employed, the association between interval and preterm delivery was similar between participants included in the within-sibship analysis and those ineligible for the within-sibship analysis, but differed from those in the full cohort, perhaps due to different target populations.

Public health interventions are often designed to target communities defined either geographically (e.g. cities, counties) or socially (e.g. schools or workplaces). The group randomized trial (GRT) is regarded as the gold standard for evaluating these interventions. However, community leaders may object to randomization as some groups may be denied a potentially beneficial intervention. Under a regression discontinuity design (RDD), individuals may be assigned to treatment based on the levels of a pretest measure, thereby allowing those most in need of the treatment to receive it. In this article, we consider analysis, power, and sample size issues in applying the RDD and related cutoff designs in community-based intervention studies. We examine the power of these designs as a function of intraclass correlation, number of groups, and number of members per group and compare results to the traditional GRT.

BACKGROUND:While the clinical trials and statistical methodology literature on sample size re-estimation (SSRE) is robust, evaluation of SSRE procedures following the completion of a clinical trial has been sparsely reported. In blinded sample size re-estimation, only nuisance parameters are re-estimated, and the blinding of the current trial treatment effect is preserved. Blinded re-estimation procedures are well-accepted by regulatory agencies and funders. We review our experience of sample size re-estimation in a large international, National Institutes of Health funded clinical trial for adjuvant breast cancer treatment, and evaluate our blinded sample size re-estimation procedure for this time-to-event trial. We evaluated the SSRE procedure by examining assumptions made during the re-estimation process, estimates resulting from re-estimation, and the impact on final trial results with and without the addition of participants, following sample size re-estimation. METHODS:We compared the control group failure probabilities estimated at the time of SSRE to estimates used in the original planning, to the final un-blinded control group failure probability estimates for those included in the SSRE procedure (SSRE cohort), and to the final total control group failure probability estimates. The impact of re-estimation on the final comparison between randomized treatment groups is evaluated for those in the originally planned cohort (n = 340) and for the combination of those recruited in the originally planned cohort and those added after re-estimation (n = 509). RESULTS:Very little difference is observed between the originally planned cohort and all randomized patients in the control group failure probabilities over time or in the overall hazard ratio estimating treatment effect (originally planned cohort HR 1.25 (0.86, 1.79); all randomized cohort HR 1.24 95% CI (0.91, 1.68)). At the time of blinded SSRE, the estimated control group failure probabilities at 3 years (0.24) and 5 years (0.40) were similar to those for the SSRE cohort once un-blinded (3 years, 0.22 (0.16, 0.30); 5 years, 0.33 (0.26, 0.41)). CONCLUSIONS:We found that our re-estimation procedure performed reasonably well in estimating the control group failure probabilities at the time of re-estimation. Particularly for time-to-event outcomes, pre-planned blinded SSRE procedures may be the best option to aid in maintaining power. TRIAL REGISTRATION/BACKGROUND:ClinicalTrials.gov, NCT00201851 . Registered on 9 September 2005. Retrospectively registered.

The use of propensity score methods to adjust for selection bias in observational studies has become increasingly popular in public health and medical research. A substantial portion of studies using propensity score adjustment treat the propensity score as a conventional regression predictor. Through a Monte Carlo simulation study, Austin and colleagues. investigated the bias associated with treatment effect estimation when the propensity score is used as a covariate in nonlinear regression models, such as logistic regression and Cox proportional hazards models. We show that the bias exists even in a linear regression model when the estimated propensity score is used and derive the explicit form of the bias. We also conduct an extensive simulation study to compare the performance of such covariate adjustment with propensity score stratification, propensity score matching, inverse probability of treatment weighted method, and nonparametric functional estimation using splines. The simulation scenarios are designed to reflect real data analysis practice. Instead of specifying a known parametric propensity score model, we generate the data by considering various degrees of overlap of the covariate distributions between treated and control groups. Propensity score matching excels when the treated group is contained within a larger control pool, while the model-based adjustment may have an edge when treated and control groups do not have too much overlap. Overall, adjusting for the propensity score through stratification or matching followed by regression or using splines, appears to be a good practical strategy.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Traumatic brain injury (TBI) mechanisms are often grouped together in research. Differences in acute and long-term outcomes across mechanisms of injury (MOIs) remain unclear, partly because of confounding by age. Modeling MOI-specific effects can inform clinical triage and prognostication. We examined the relationship between motor vehicle accidents (MVAs) vs falls, the 2 most common MOIs, and acute and 1-year post-injury outcomes, after rigorous control of demographic and preinjury personal factors. METHODS:Data were analyzed from individuals with moderate-to-severe TBI requiring inpatient rehabilitation from the TBI Model Systems National Database, a multicenter prospective longitudinal cohort study. The analytic sample was restricted to individuals aged 16-79 years with an MOI due to MVA or fall occurring between April 2010 and January 2023. We used inverse probability of treatment weighting, based on propensity scores, to adjust for 14 demographic and preinjury personal characteristics and estimate the causal effect of MOI on acute and 1-year outcomes after TBI. Acute hospital and rehabilitation outcomes included the following: Glasgow Coma Scale (GCS), sedation, intubation, post-traumatic amnesia duration, time to follow commands (TFC), length of hospital stay (LOS), and Functional Independence Measure (FIM) cognitive and motor scores. One-year outcomes included the following: Disability Rating Scale and Participation Assessment with Recombined Tools Objective. RESULTS:= 0.014). At 1 year after injury, disability levels and community participation did not differ. DISCUSSION/CONCLUSIONS:MVA-related TBI was associated with worse acute outcomes. However, by 1 year after injury, disability level and community participation do not differ. This work highlights novel findings in short-term and long-term outcomes after falls and MVAs, the leading TBI causes, which are not explained by confounders such as age. Findings may not generalize beyond patients receiving inpatient rehabilitation for TBI.

Electronic health records (EHRs) present opportunities to study uterine fibroids uterine fibroids and endometriosis within diverse populations. When using EHR data, it is important to validate outcome classification via diagnosis codes. We performed a validation study of three approaches (1: ICD-10 code alone, 2: ICD-10 code + diagnostic procedure, and 3: ICD-10 code + all diagnostic information) to identify incident uterine fibroids and endometriosis patients among n=750 NYU Langone Health 2016-2023. Chart review was used to determine the true diagnosis status. When using a binary classification system (incident vs. non-incident patient), Approaches 2 and 3 had higher positive predictive values (PPVs) for uterine fibroids (0.86 and 0.87 vs. 0.78) and for endometriosis (0.70 and 0.73 vs. 0.66), but Approach 1 outperformed the other two in negative predictive values (NPVs) for both outcomes. When using a three-level classification system (incident vs. prevalent vs. disease free patients), PPV for prevalent patients was low for all approaches, while PPV/NPV of disease-free patients was generally above 0.8. Using ICD-10 codes alone yielded higher NPVs but resulted in lower PPVs compared with the other approaches. Continued validation of uterine fibroids/endometriosis EHR studies is warranted to increase research into these understudied gynecologic conditions.