Optimal model averaging for partially linear models with missing response variables and error-prone covariates.

We consider the problem of optimal model averaging for partially linear models when the responses are missing at random and some covariates are measured with error. A novel weight choice criterion based on the Mallows-type criterion is proposed for the weight vector to be used in the model averaging. The resulting model averaging estimator for the partially linear models is shown to be asymptotically optimal under some regularity conditions in terms of achieving the smallest possible squared loss. In addition, the existence of a local minimizing weight vector and its convergence rate to the risk-based optimal weight vector are established. Simulation studies suggest that the proposed model averaging method generally outperforms existing methods. As an illustration, the proposed method is applied to analyze an HIV-CD4 dataset.

Matched case-control data with a misclassified exposure: what can be done with instrumental variables?

Matched case-control studies are used for finding the association between a disease and an exposure after controlling the effect of important confounding variables. It is a known fact that the disease-exposure association parameter estimators are biased when the exposure is misclassified, and a matched case-control study is of no exception. Any bias correction method relies on validation data that contain the true exposure and the misclassified exposure value, and in turn the validation data help to estimate the misclassification probabilities. The question is what we can do when there are no validation data and no prior knowledge on the misclassification probabilities, but some instrumental variables are observed. To answer this unexplored and unanswered question, we propose two methods of reducing the exposure misclassification bias in the analysis of a matched case-control data when instrumental variables are measured for each subject of the study. The significance of these approaches is that the proposed methods are designed to work without any validation data that often are not available when the true exposure is impossible or too costly to measure. A simulation study explores different types of instrumental variable scenarios and investigates when the proposed methods work, and how much bias can be reduced. For the purpose of illustration, we apply the methods to a nested case-control data sampled from the 1989 US birth registry.

Analysis of linear transformation models with covariate measurement error and interval censoring.

Among several semiparametric models, the Cox proportional hazard model is widely used to assess the association between covariates and the time-to-event when the observed time-to-event is interval-censored. Often, covariates are measured with error. To handle this covariate uncertainty in the Cox proportional hazard model with the interval-censored data, flexible approaches have been proposed. To fill a gap and broaden the scope of statistical applications to analyze time-to-event data with different models, in this paper, a general approach is proposed for fitting the semiparametric linear transformation model to interval-censored data when a covariate is measured with error. The semiparametric linear transformation model is a broad class of models that includes the proportional hazard model and the proportional odds model as special cases. The proposed method relies on a set of estimating equations to estimate the regression parameters and the infinite-dimensional parameter. For handling interval censoring and covariate measurement error, a flexible imputation technique is used. Finite sample performance of the proposed method is judged via simulation studies. Finally, the suggested method is applied to analyze a real data set from an AIDS clinical trial.

Confidence intervals for the difference in the success rates of two treatments in the analysis of correlated binary responses.

In clinical studies, we often compare the success rates of two treatment groups where post-treatment responses of subjects within clusters are usually correlated. To estimate the difference between the success rates, interval estimation procedures that do not account for this intraclass correlation are likely inappropriate. To address this issue, we propose three interval procedures by direct extensions of recently proposed methods for independent binary data based on the concepts of design effect and effective sample size used in sample surveys. Each of them is then evaluated with four competing variance estimates. We also extend three existing methods recommended for complex survey data using different weighting schemes required for those three existing methods. An extensive simulation study is conducted for the purposes of evaluating and comparing the performance of the proposed methods in terms of coverage and expected width. The interval estimation procedures are illustrated using three examples in clinical and social science studies. Our analytic arguments and numerical studies suggest that the methods proposed in this work may be useful in clustered data analyses.

Confidence intervals for the common intraclass correlation in the analysis of clustered binary responses.

In cluster randomized trials, it is often of interest to estimate the common intraclass correlation at the design stage for sample size and power calculations, which are greatly affected by the value of a common intraclass correlation. In this article, we construct confidence intervals (CIs) for the common intraclass correlation coefficient of several treatment groups. We consider the profile likelihood (PL)-based approach using the beta-binomial models and the approach based on the concept of generalized pivots using the ANOVA estimator and its asymptotic variance. We compare both approaches with a number of large sample procedures as well as both parametric and nonparametric bootstrap procedures in terms of coverage and expected CI length through a simulation study, and illustrate the methodology with two examples from biomedical fields. The results support the use of the PL-based CI as it holds the preassigned confidence level very well and overall gives a very competitive length.

Parameter estimation methods for gene circuit modeling from time-series mRNA data: a comparative study.

Parameter estimation is a challenging computational problem in the reverse engineering of biological systems. Because advances in biotechnology have facilitated wide availability of time-series gene expression data, systematic parameter estimation of gene circuit models from such time-series mRNA data has become an important method for quantitatively dissecting the regulation of gene expression. By focusing on the modeling of gene circuits, we examine here the performance of three types of state-of-the-art parameter estimation methods: population-based methods, online methods and model-decomposition-based methods. Our results show that certain population-based methods are able to generate high-quality parameter solutions. The performance of these methods, however, is heavily dependent on the size of the parameter search space, and their computational requirements substantially increase as the size of the search space increases. In comparison, online methods and model decomposition-based methods are computationally faster alternatives and are less dependent on the size of the search space. Among other things, our results show that a hybrid approach that augments computationally fast methods with local search as a subsequent refinement procedure can substantially increase the quality of their parameter estimates to the level on par with the best solution obtained from the population-based methods while maintaining high computational speed. These suggest that such hybrid methods can be a promising alternative to the more commonly used population-based methods for parameter estimation of gene circuit models when limited prior knowledge about the underlying regulatory mechanisms makes the size of the parameter search space vastly large.

Using Walk Score™ and Neighborhood Perceptions to Assess Walking Among Middle-Aged and Older Adults.

We aimed to determine the relationship between neighborhood characteristics (walkability, cohesion/safety) and recommended activity levels among community-dwelling middle-aged and older adults. Subjective and objective data on 394 individuals aged ≥50 years were used to assess the likelihood of walking ≥150 min/week. Environmental factors associated with a greater likelihood of any walking ≥150 min/week included living in a neighborhood with high perception of cohesion/safety versus low, living in walkable areas versus car-dependent, and living in an area with a low-moderate median income versus the lowest. Middle-aged and older adults were more likely to walk ≥150 min/week in a walkable, perceived safe/cohesive neighborhood. Identifying neighborhood factors associated with promoting walking among this population can enable stakeholders (e.g., researchers, planners, and policy makers) to direct interventions focusing on the built environment.

The direct integral method for confidence intervals for the ratio of two location parameters.

In a relative risk analysis of colorectal caner on nutrition intake scores across genders, we show that, surprisingly, when comparing the relative risks for men and women based on the index of a weighted sum of various nutrition scores, the problem reduces to forming a confidence interval for the ratio of two (asymptotically) normal random variables. The latter is an old problem, with a substantial literature. However, our simulation results suggest that existing methods often either give inaccurate coverage probabilities or have a positive probability to produce confidence intervals with infinite length. Motivated by such a problem, we develop a new methodology which we call the Direct Integral Method for Ratios (DIMER), which, unlike the other methods, is based directly on the distribution of the ratio. In simulations, we compare this method to many others. These simulations show that, generally, DIMER more closely achieves the nominal confidence level, and in those cases that the other methods achieve the nominal levels, DIMER has comparable confidence interval lengths. The methodology is then applied to a real data set, and with follow up simulations.

Children's active commuting to school: an interplay of self-efficacy, social economic disadvantage, and environmental characteristics.

BACKGROUND: Active commuting to school (ACS) can promote children's physical activity and may help prevent childhood obesity. Previous researchers in various disciplines, e.g., health, urban planning, and transportation, have identified various predictors of ACS. However, little research has been carried out into investigating the effect of self-efficacy on ACS. The purpose of this study is to investigate the roles of children's and parents' self-efficacy in children's ACS, controlling for sociodemographic and objective environmental characteristics. METHODS: This study is part of the Texas Childhood Obesity Prevention Policy Evaluation (T-COPPE) project, which includes data from 857 parent/child pairs from 74 schools who lived within two miles of school in Texas. Measures included children's usual modes of commuting to school, participants' sociodemographics, perceived self-efficacy toward ACS, sources of children's self-efficacy, school settings, and objective environmental constraints. Multilevel structural equation modeling (SEM) was employed to test the hypothesized pathways using Mplus 7.0. RESULTS: Around 18% of the children were active commuters. Two sources of children's self-efficacy were identified, i.e., emotional states (ß = 0.36, p < 0.001) and social modeling (ß = 0.28, p < 0.01). Compared with children's self-efficacy (ß = 0.16, p < 0.001), parents' self-efficacy (ß = 0.63, p < 0.001) had a stronger influence on children's ACS. Participants' social economic disadvantage (ß = 0.40, p < 0.001), environmental constraints (ß = -0.49, p < 0.001), and school setting (ß = -0.17, p = 0.029) all had statistically significant direct effects on children's ACS. CONCLUSIONS: Future initiatives should consider both parents' and children's self-efficacy in developing strategies for promoting children's ACS. Social disadvantage and environmental constraints also need to be addressed for effective interventions. The work reported here provides support for the continuing exploration of the role of self-efficacy in children's ACS.

A framework for scalable parameter estimation of gene circuit models using structural information.

MOTIVATION: Systematic and scalable parameter estimation is a key to construct complex gene regulatory models and to ultimately facilitate an integrative systems biology approach to quantitatively understand the molecular mechanisms underpinning gene regulation. RESULTS: Here, we report a novel framework for efficient and scalable parameter estimation that focuses specifically on modeling of gene circuits. Exploiting the structure commonly found in gene circuit models, this framework decomposes a system of coupled rate equations into individual ones and efficiently integrates them separately to reconstruct the mean time evolution of the gene products. The accuracy of the parameter estimates is refined by iteratively increasing the accuracy of numerical integration using the model structure. As a case study, we applied our framework to four gene circuit models with complex dynamics based on three synthetic datasets and one time series microarray data set. We compared our framework to three state-of-the-art parameter estimation methods and found that our approach consistently generated higher quality parameter solutions efficiently. Although many general-purpose parameter estimation methods have been applied for modeling of gene circuits, our results suggest that the use of more tailored approaches to use domain-specific information may be a key to reverse engineering of complex biological systems. AVAILABILITY: http://sfb.kaust.edu.sa/Pages/Software.aspx. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Semiparametric approach for non-monotone missing covariates in a parametric regression model.

Missing covariate data often arise in biomedical studies, and analysis of such data that ignores subjects with incomplete information may lead to inefficient and possibly biased estimates. A great deal of attention has been paid to handling a single missing covariate or a monotone pattern of missing data when the missingness mechanism is missing at random. In this article, we propose a semiparametric method for handling non-monotone patterns of missing data. The proposed method relies on the assumption that the missingness mechanism of a variable does not depend on the missing variable itself but may depend on the other missing variables. This mechanism is somewhat less general than the completely non-ignorable mechanism but is sometimes more flexible than the missing at random mechanism where the missingness mechansim is allowed to depend only on the completely observed variables. The proposed approach is robust to misspecification of the distribution of the missing covariates, and the proposed mechanism helps to nullify (or reduce) the problems due to non-identifiability that result from the non-ignorable missingness mechanism. The asymptotic properties of the proposed estimator are derived. Finite sample performance is assessed through simulation studies. Finally, for the purpose of illustration we analyze an endometrial cancer dataset and a hip fracture dataset.

Perceived barriers to children's active commuting to school: a systematic review of empirical, methodological and theoretical evidence.

Active commuting to school (ACS) may increase children's daily physical activity and help them maintain a healthy weight. Previous studies have identified various perceived barriers related to children's ACS. However, it is not clear whether and how these studies were methodologically sound and theoretically grounded. The purpose of this review was to critically assess the current literature on perceived barriers to children's ACS and provide recommendations for future studies. Empirically based literature on perceived barriers to ACS was systematically searched from six databases. A methodological quality scale (MQS) and a theory utilization quality scale (TQS) were created based on previously established instruments and tailored for the current review. Among the 39 studies that met the inclusion criteria, 19 (48.7%) reported statistically significant perceived barriers to child's ACS. The methodological and theory utilization qualities of reviewed studies varied, with MQS scores ranging between 7 and 20 (Mean =12.95, SD =2.95) and TQS scores from 1 to 7 (Mean =3.62, SD =1.74). A detailed appraisal of the literature suggests several empirical, methodological, and theoretical recommendations for future studies on perceived barriers to ACS. Empirically, increasing the diversity of study regions and samples should be a high priority, particularly in Asian and European countries, and among rural residents; more prospective and interventions studies are needed to determine the causal mechanism liking the perceived factors and ACS; future researchers should include policy-related barriers into their inquiries. Methodologically, the conceptualization of ACS should be standardized or at least well rationalized in future studies to ensure the comparability of results; researchers' awareness need to be increased for improving the methodological rigor of studies, especially in regard to appropriate statistical analysis techniques, control variable estimation, multicollinearity testing, and reliability and validity reporting. Theoretically, future researchers need to first ground their investigations in theoretical foundations; efforts should be devoted to make sure theories are used thoroughly and correctly; important theoretical constructs, in particular, need to be conceptualized and operationalized appropriately to ensure accurate measurement. By reviewing what has been achieved, this review offered insights for more sophisticated ACS studies in the future.

Identifying the Heterogeneity in the Association between Workforce Diversity and Retention in Opioid Treatment among Black clients.

Background: This study investigates the impact of workforce diversity, specifically staff identified as Black/African American, on retention in opioid use disorder (OUD) treatment, aiming to enhance patient outcomes. Employing a novel machine learning technique known as 'causal forest,' we explore heterogeneous treatment effects on retention. Methods: We relied on four waves of the National Drug Abuse Treatment System Survey (NDATSS), a nationally representative longitudinal dataset of treatment programs. We analyzed OUD program data from the years 2000, 2005, 2014 and 2017 (n = 627). Employing the 'causal forest' method, we analyzed the heterogeneity in the relationship between workforce diversity and retention in OUD treatment. Interviews with program directors and clinical supervisors provided the data for this study. Results: The results reveal diversity-related variations in the association with retention across 61 out of 627 OUD treatment programs (less than 10%). These programs, associated with positive impacts of workforce diversity, were more likely private-for-profit, newer, had lower percentages of Black and Latino clients, lower staff-to-client ratios, higher proportions of staff with graduate degrees, and lower percentages of unemployed clients. Conclusions: While workforce diversity is crucial, our findings underscore that it alone is insufficient for improving retention in addiction health services research. Programs with characteristics typically linked to positive outcomes are better positioned to maximize the benefits of a diverse workforce in client retention. This research has implications for policy and program design, guiding decisions on resource allocation and workforce diversity to enhance retention rates among Black clients with OUDs.

Predicting and responding to change: Perceived environmental uncertainty among substance use disorder treatment programs.

INTRODUCTION: Substance use disorder (SUD) treatment programs offering addiction health services (AHS) must be prepared to adapt to change in their operating environment. These environmental uncertainties may have implications for service delivery, and ultimately patient outcomes. To adapt to a multitude of environmental uncertainties, treatment programs must be prepared to predict and respond to change. Yet, research on treatment programs preparedness for change is sparse. We examined reported difficulties in predicting and responding to changes in the AHS system, and factors associated with these outcomes. METHODS: Cross-sectional surveys of SUD treatment programs in the United States in 2014 and 2017. We used linear and ordered logistic regression to examine associations between key independent variables (e.g., program, staff, and client characteristics) and four outcomes, (1) reported difficulties in predicting change, (2) predicting effect of change on organization, (3) responding to change, and (4) predicting changes to make to respond to environmental uncertainties. Data were collected through telephone surveys. RESULTS: The proportion of SUD treatment programs reporting difficulty predicting and responding to changes in the AHS system decreased from 2014 to 2017. However, a considerable proportion still reported difficulty in 2017. We identified that different organizational characteristics are associated with their reported ability to predict or respond to environmental uncertainty. Findings show that predicting change is significantly associated with program characteristics only, while predicting effect of change on organizations is associated with program and staff characteristics. Deciding how to respond to change is associated with program, staff, and client characteristics, while predicting changes to make to respond is associated with staff characteristics only. CONCLUSIONS: Although treatment programs reported decreased difficulty predicting and responding to changes, our findings identify program characteristics and attributes that could better position programs with the foresight to more effectively predict and respond to uncertainties. Given resource constraints at multiple levels in treatment programs, this knowledge might help identify and optimize aspects of programs to intervene upon to enhance their adaptability to change. These efforts may positively influences processes or care delivery, and ultimately translate into improvements in patient outcomes.

Examining the impacts of public transit on healthy aging through a natural experiment: study protocols and lessons learned from the Active El Paso project.

This paper describes protocols and experiences from a seven-year natural-experiment study in El Paso, Texas, a border city of predominantly Latino/Hispanic population. The study focuses on how Bus Rapid Transit (BRT) impacts physical activity and thus plays a role in alleviating obesity and related chronic diseases that impact healthy aging. Our protocols describe a longitudinal and case-comparison study, which compared residents exposed to new BRT stations with those who were not. This paper also introduces lessons and experiences to overcome the following challenges: delays in the BRT opening (the main intervention), the COVID-19 pandemic, methodological challenges, participant recruitment and retention, and predatory survey takers. Our transdisciplinary approach was pivotal in addressing these challenges. We also proposed and tested multi-level intervention strategies to reduce modifiable barriers to transit use. Our most important takeaway for researchers, practitioners, and policy makers is the importance of being flexible and ready to adapt to new circumstances. Future natural-experiment researchers need to become more versatile in an increasingly volatile and uncertain world.

Workforce Diversity and disparities in wait time and retention among opioid treatment programs.

BACKGROUND: Workforce diversity is a key strategy to improve treatment engagement among members of racial and ethnic minority groups. In this study, we seek to determine whether workforce diversity plays a role in reducing racial and ethnic differences in wait time to treatment entry and retention in different types of opioid use disorder treatment programs. METHODS: We conducted comparative and predictive analysis in a subsample of outpatient opioid treatment programs (OTPs), who completed access and retention survey questions in four waves of the National Drug Abuse Treatment System Survey (162 OTPs in 2000, 173 OTPs in 2005, 282 OTPs in 2014, and 300 OTPs in 2017). We sought to assess the associations between workforce diversity on wait time and retention, accounting for the role of Medicaid expansion and the moderating role of program ownership type (i.e., public, non-profit, for-profit) among OTPs located across the United States. RESULTS: We found significant differences in wait time to treatment entry and retention in treatment across waves. Average number of waiting days decreased in 2014 and 2017; post Medicaid expansion per the Affordable Care Act, while retention rates varied across years. Key findings show that programs with high diversity, measured by higher percent of African American staff and a higher percent of African American clients, were associated with longer wait times to enter treatment, compared to low diversity programs. Programs with higher percent of Latino staff and a higher percent of Latino clients were associated with lower retention in treatment compared with low diversity programs. However, program ownership type (public, non-profit and for-profit) played a moderating role. Public programs with higher percent of African American staff were associated with lower wait time, while non-profit programs with higher percent of Latino staff were related to higher retention. CONCLUSIONS: Findings show decreases in wait time over the years with significant variation in retention during the same period. Concordance in high workforce and client diversity was associated with higher wait time and lower retention. But these relations inverted (low wait time and high retention) in public and non-profit programs with high staff diversity. Findings have implications for building resources and service capacity among OTPs that serve a higher proportion of minority clients.

Bias reduction in conditional logistic regression.

We employ a general bias preventive approach developed by Firth (Biometrika 1993; 80:27-38) to reduce the bias of an estimator of the log-odds ratio parameter in a matched case-control study by solving a modified score equation. We also propose a method to calculate the standard error of the resultant estimator. A closed-form expression for the estimator of the log-odds ratio parameter is derived in the case of a dichotomous exposure variable. Finite sample properties of the estimator are investigated via a simulation study. Finally, we apply the method to analyze a matched case-control data from a low birthweight study.

Fear of Outdoor Falling Among Community-Dwelling Middle-Aged and Older Adults: The Role of Neighborhood Environments.

Background and Objectives: Fear of falling is a substantial barrier to walking and has been associated with increased fall risks. This study examines neighborhood environmental risk factors related to fear of outdoor falling in middle-aged and older adults. Research Design and Methods: A total of 394 participants aged 50 years or older living independently in the community were recruited between 2013 and 2014 from an integrated health care network serving Central Texas. Fear of outdoor falling and perceived neighborhood environmental variables were assessed using self-reported questionnaires. Logistic regression identified perceived neighborhood environmental variables associated with fear of outdoor falling. Results: Sixty-nine (17.9%) of 385 participants reported having a fear of outdoor falling. Compared to those who did not report a fear of outdoor falling, those who reported having a fear of outdoor falling were more likely to be adults aged 65 years or older (odds ratio [OR] = 2.974, 95% confidence interval [CI] = 1.247-7.094), be female (OR = 4.423, 95% CI = 1.830-10.689), have difficulty with walking for a quarter of a mile (OR = 2.761, 95% CI = 1.124-6.782), and have had a fall in the past year (OR = 4.720, 95% CI = 1.472-15.137). Among the neighborhood environmental characteristics examined, low traffic speed on streets (OR = 0.420, 95% CI = 0.188-0.935), drainage ditches (OR = 2.383, 95% CI = 1.136-5.000), and broken sidewalks (OR = 3.800, 95% CI = 1.742-8.288) were associated with the odds of having a fear of outdoor falling. Discussion and Implications: In addition to the individual factors, findings from this study suggest the importance of addressing the environmental risk factors in identifying and reducing fear of outdoor falling among middle-aged and older adults.

Multiphoton laser-scanning microscopy and spatial analysis of dehydroergosterol distributions on plasma membrane of living cells.

Multiphoton laser-scanning microscopy (MPLSM) imaging in combination with advanced image analysis techniques provides unique opportunities to visualize the arrangement of cholesterol in the plasma membrane (PM) of living cells. MPLSM makes possible the use of a naturally occurring sterol, dehydroergosterol (DHE), for observing sterol-enriched areas of the PM. Pure DHE has properties similar to cholesterol as observed in model and cellular membranes but with a conjugated double-bond system that fluoresces at ultraviolet wavelengths. MPLSM enables the excitation of DHE at infrared wavelengths that many laser-scanning microscopy systems are able to transmit effectively and that are less harmful to the cell. Thus, with the incorporation of DHE into living cells and the advent of MPLSM, real-time images of the cellular distribution of DHE can be obtained. In juxtaposition, notably the application of newly advanced techniques in image analysis, aids not only the identification and segmentation of sterol-rich regions of the PM of cells, but also the elucidation of the statistical nature of the observed patterns. In studies involving murine L-cell (Larpt-+K-) fibroblasts, DHE is shown to exhibit strong cluster patterns within the PM.

Levels and correlates of exercise in a border mexican american population.

OBJECTIVE: To examine personal and environmental correlates of exercise among Mexican Americans living in the Texas-Mexico border region. METHODS: The study was based on data from a community health assessment conducted in 2 counties at the Texas-Mexico border region. A random-digit-dialed community survey was used in this cross-sectional study (n=933). RESULTS: A majority of border Mexican Americans (52%) did not exercise at all. Gender, age, and self-rated health were statistically significant correlates to exercise. CONCLUSIONS: Understanding personal and environmental factors that influence physical activity and exercise in this minority population is critical for the development of culturally sensitive health interventions.