Faculty Bibliography

BACKGROUND:Living in neighborhoods with higher levels of walkability has been associated with a reduced risk of obesity and higher levels of physical activity. Obesity has been linked to increased risk of 13 cancers in women. However, long-term prospective studies of neighborhood walkability and risk for obesity-related cancer are scarce. OBJECTIVES:We evaluated the association between long-term average neighborhood walkability and obesity-related cancer risk in women. METHODS:The New York University Women's Health Study (NYUWHS) is a prospective cohort with 14,274 women recruited between 1985 and 1991 in New York City and followed over nearly three decades. We geocoded residential addresses for each participant throughout follow-up and calculated an average annual measure of neighborhood walkability across years of follow-up using data on population density and accessibility to destinations associated with geocoded residential addresses. We used ICD-9 codes to characterize first primary obesity-related cancers and employed Cox proportional hazards models to assess the association between average neighborhood walkability and risk of overall and site-specific obesity-related cancers. RESULTS: DISCUSSION:Our study highlights a potential protective role of neighborhood walkability in preventing obesity-related cancers in women. https://doi.org/10.1289/EHP11538.

Studying time-dependent exposure mixtures has gained increasing attentions in environmental health research. When a scalar outcome is of interest, distributed lag (DL) models have been employed to characterize the exposures effects distributed over time on the mean of final outcome. However, there is a methodological gap on investigating time-dependent exposure mixtures with different quantiles of outcome. In this article, we introduce semiparametric partial-linear single-index (PLSI) DL quantile regression, which can describe the DL effects of time-dependent exposure mixtures on different quantiles of outcome and identify susceptible periods of exposures. We consider two time-dependent exposure settings: discrete and functional, when exposures are measured in a small number of time points and at dense time grids, respectively. Spline techniques are used to approximate the nonparametric DL function and single-index link function, and a profile estimation algorithm is proposed. Through extensive simulations, we demonstrate the performance and value of our proposed models and inference procedures. We further apply the proposed methods to study the effects of maternal exposures to ambient air pollutants of fine particulate and nitrogen dioxide on birth weight in New York University Children's Health and Environment Study (NYU CHES). This article is protected by copyright. All rights reserved.

Detecting and characterizing subgroups with differential effects of a binary treatment has been widely studied and led to improvements in patient outcomes and population risk management. Under the setting of a continuous treatment, however, such investigations remain scarce. We propose a semiparametric change-plane model and consequently a doubly robust test statistic for assessing the existence of two subgroups with differential treatment effects under a continuous treatment. The proposed testing procedure is valid when either the baseline function for the covariate effects or the generalized propensity score function for the continuous treatment is correctly specified. The asymptotic distributions of the test statistic under the null and local alternative hypotheses are established. When the null hypothesis of no subgroup is rejected, the change-plane parameters that define the subgroups can be estimated. This paper provides a unified framework of the change-plane method to handle various types of outcomes, including the exponential family of distributions and time-to-event outcomes. Additional extensions with nonparametric estimation approaches are also provided. We evaluate the performance of our proposed methods through extensive simulation studies under various scenarios. An application to the Health Effects of Arsenic Longitudinal Study with a continuous environmental exposure of arsenic is presented. This article is protected by copyright. All rights reserved.

STUDY OBJECTIVE:To study associations between nighttime sleep characteristics and time to pregnancy. METHODS:Pregnant people age ≥18 years and<18 weeks' gestation were recruited from 3 New York University Grossman School of Medicine affiliated hospitals in Manhattan and Brooklyn (n = 1428) and enrolled into the New York University Children's Health and Environment Study. Participants in the first trimester of pregnancy were asked to recall their time to pregnancy and their sleep characteristics in the 3 months before conception. RESULTS:Participants who reported sleeping<7 hours per night tended to have shorter time to pregnancy than those who slept 7-9 hours per night (adjusted fecundability odds ratio = 1.16, 95% confidence interval: 0.94, 1.41). Participants with a sleep midpoint of 4 AM or later tended to have longer time to pregnancy compared with those with earlier sleep midpoints (before 4 AM) (adjusted fecundability odds ratio = 0.88, 95% confidence interval: 0.74, 1.04). When stratified by sleep midpoint, sleeping<7 hours was significantly associated with shorter time to pregnancy only among those whose sleep midpoint was before 4 AM (adjusted fecundability odds ratio = 1.33, 95% confidence interval: 1.07, 1.67). CONCLUSIONS:The association of sleep duration with time to pregnancy was modified by chronotype, suggesting that both biological and behavioral aspects of sleep may influence fecundability.

BACKGROUND:Air pollution is a health risk in pregnant women and children. Despite the importance of refined exposure assessment, the characterisation of personalised air pollution exposure remains a challenge in paediatric and perinatal epidemiology. OBJECTIVE:We used portable personal air monitors to characterise personalised exposure to air pollutants in pregnant women. METHODS:), and volatile organic compounds (average use = 14 days). Data were stored in real-time on a secure database via synchronisation with a smartphone application. Of 497 women who agreed to use air monitors, 273 women (55%) were successful in using air monitors for longer than a day. For these participants, we identified daily patterns of exposure to air pollutants using functional principal component analysis (3827 days of air monitoring). RESULTS:had higher daily variations compared to PM. CONCLUSIONS:Small wearables are useful for the measurement of personalised air pollution exposure in birth cohorts and identify daily patterns that cannot be captured otherwise. Successful participation, however, depends on certain individual characteristics. Future studies should consider strategies in design and analysis to account for selective participation.

STUDY QUESTION:How did the first two coronavirus disease 2019 (COVID-19) waves affect fertility rates in the USA? SUMMARY ANSWER:States differed widely in how their fertility rates changed following the COVID-19 outbreak and these changes were influenced more by state-level economic, racial, political, and social factors than by COVID-19 wave severity. WHAT IS KNOWN ALREADY:The outbreak of the COVID-19 pandemic contributed to already declining fertility rates in the USA, but not equally across states. Identifying drivers of differential changes in fertility rates can help explain variations in demographic shifts across states in the USA and motivate policies that support families in general, not only during crises. STUDY DESIGN, SIZE, DURATION:This is an ecological study using state-level data from 50 US states and the District of Columbia (n = 51). The study period extends from 2020 to 2021 with historical data from 2016 to 2019. We identified Wave 1 as the first apex for each state after February 2020 and Wave 2 as the second apex, during Fall/Winter 2020-2021. PARTICIPANTS/MATERIALS, SETTING, METHODS:State-level COVID-19 wave severity, defined as case acceleration during each 3-month COVID-19 wave (cases/100 000 population/month), was derived from 7-day weekly moving average COVID-19 case rates from the US Centers for Disease Control and Prevention (CDC). State-level fertility rate changes (change in average monthly fertility rate/100 000 women of reproductive age (WRA)/year) were derived from the CDC Bureau of Vital Statistics and from 2020 US Census and University of Virginia 2021 population estimates 9 months after each COVID-19 wave. We performed univariate analyses to describe national and state-level fertility rate changes following each wave, and simple and multivariable linear regression analyses to assess the relation of COVID-19 wave severity and other state-level characteristics with fertility rate changes. MAIN RESULTS AND THE ROLE OF CHANCE:Nationwide, fertility dropped by 17.5 births/month/100 000 WRA/year following Wave 1 and 9.2 births/month/100 000 WRA/year following Wave 2. The declines following Wave 1 were largest among majority-Democrat, more non-White states where people practiced greater social distancing. Greater COVID-19 wave severity was associated with steeper fertility rate decline post-Wave 1 in simple regression, but the association was attenuated when adjusted for other covariates. Adjusting for the economic impact of the pandemic (hypothesized mediator) also attenuated the effect. There was no relation between COVID-19 wave severity and fertility rate change following Wave 2. LIMITATIONS, REASONS FOR CAUTION:Our study harnesses state-level data so individual-level conclusions cannot be inferred. There may be residual confounding in our multivariable regression and we were underpowered to detect some effects. WIDER IMPLICATIONS OF THE FINDINGS:The COVID-19 pandemic initially impacted the national fertility rate but, overall, the fertility rate rebounded to the pre-pandemic level following Wave 2. Consistent with prior literature, COVID-19 wave severity did not appear to predict fertility rate change. Economic, racial, political, and social factors influenced state-specific fertility rates during the pandemic more than the severity of the outbreak alone. Future studies in other countries should also consider whether these factors account for internal heterogeneity when examining the impact of the COVID-19 pandemic and other crises on fertility. STUDY FUNDING/COMPETING INTEREST(S):L.G.K. received funding from the National Institute of Environmental Health Sciences (R00ES030403), M.C. from the National Science Foundation Graduate Research Fellowship Program (20-A0-00-1005789), and M.L. and E.S. from the National Institute of Environmental Health Sciences (R01ES032808). None of the authors have competing interests. TRIAL REGISTRATION NUMBER:N/A.

Reproducible translation of transcriptomics data has been hampered by the ubiquitous presence of batch effects. Statistical methods for managing batch effects were initially developed in the setting of sample group comparison and later borrowed for other settings such as survival outcome prediction. The most notable such method is ComBat, which adjusts for batches by including it as a covariate alongside sample groups in a linear regression. In survival prediction, however, ComBat is used without definable groups for survival outcome and is done sequentially with survival regression for a potentially batch-confounded outcome. To address these issues, we propose a new method called BATch MitigAtion via stratificatioN (BatMan). It adjusts batches as strata in survival regression and uses variable selection methods such as the regularized regression to handle high dimensionality. We assess the performance of BatMan in comparison with ComBat, each used either alone or in conjunction with data normalization, in a resampling-based simulation study under various levels of predictive signal strength and patterns of batch-outcome association. Our simulations show that (1) BatMan outperforms ComBat in nearly all scenarios when there are batch effects in the data and (2) their performance can be worsened by the addition of data normalization. We further evaluate them using microRNA data for ovarian cancer from the Cancer Genome Atlas and find that BatMan outforms ComBat while the addition of data normalization worsens the prediction. Our study thus shows the advantage of BatMan and raises caution about the use of data normalization in the context of developing survival prediction models. The BatMan method and the simulation tool for performance assessment are implemented in R and publicly available at LXQin/PRECISION.survival-GitHub.

BACKGROUND:Sub-cohort sampling designs such as a case-cohort study play a key role in studying biomarker-disease associations due to their cost effectiveness. Time-to-event outcome is often the focus in cohort studies, and the research goal is to assess the association between the event risk and risk factors. In this paper, we propose a novel goodness-of-fit two-phase sampling design for time-to-event outcomes when some covariates (e.g., biomarkers) can only be measured on a subgroup of study subjects. METHODS:Assuming that an external model, which can be the well-established risk models such as the Gail model for breast cancer, Gleason score for prostate cancer, and Framingham risk models for heart diseases, or built from preliminary data, is available to relate the outcome and complete covariates, we propose to oversample subjects with worse goodness-of-fit (GOF) based on an external survival model and time-to-event. With the cases and controls sampled using the GOF two-phase design, the inverse sampling probability weighting method is used to estimate the log hazard ratio of both incomplete and complete covariates. We conducted extensive simulations to evaluate the efficiency gain of our proposed GOF two-phase sampling designs over case-cohort study designs. RESULTS:Through extensive simulations based on a dataset from the New York University Women's Health Study, we showed that the proposed GOF two-phase sampling designs were unbiased and generally had higher efficiency compared to the standard case-cohort study designs. CONCLUSION:In cohort studies with rare outcomes, an important design question is how to select informative subjects to reduce sampling costs while maintaining statistical efficiency. Our proposed goodness-of-fit two-phase design provides efficient alternatives to standard case-cohort designs for assessing the association between time-to-event outcome and risk factors. This method is conveniently implemented in standard software.