Augmented Inverse Probability Weighting Research Articles

Impact of farm mechanization on crop productivity and economic efficiency in central and southern Oromia, Ethiopia

IntroductionFarm mechanization has multi-dimensional impacts on agricultural production systems like economic efficiency and productivity, thereby improving the quality of life in the farming community by reducing work drudgeries. However, these impacts were not studied empirically in Ethiopia. Hence, this research was initiated to estimate the level of economic efficiency, and productivity of wheat and barley, and the impact of farm mechanization on economic efficiency and productivity of wheat and barley.MethodsThe analysis was done for 232 and 257 wheat and barley producer farmers respectively who are selected from the Arsi and West Arsi zones. The stochastic frontier model was used to estimate economic efficiency while augmented inverse probability weighted (AIPW) was used to estimate the impact.Results and discussionBased on the result, it is apparent that farm mechanization has a significant positive impact on wheat productivity while the percentage change in average treatment effect for the barley was not statistically significant. Farm mechanization also affects wheat and barley economic efficiency positively. Hence, we recommended the wider use of farm mechanization to improve economic efficiency and productivity. Therefore, policy design should focus on ways to avail farm machinery easily like establishing farm mechanization service centers and facilitating credit services for mechanization service renders.

Machine learning in causal inference for epidemiology.

In causal inference, parametric models are usually employed to address causal questions estimating the effect of interest. However, parametric models rely on the correct model specification assumption that, if not met, leads to biased effect estimates. Correct model specification is challenging, especially in high-dimensional settings. Incorporating Machine Learning (ML) into causal analyses may reduce the bias arising from model misspecification, since ML methods do not require the specification of a functional form of the relationship between variables. However, when ML predictions are directly plugged in a predefined formula of the effect of interest, there is the risk of introducing a "plug-in bias" in the effect measure. To overcome this problem and to achieve useful asymptotic properties, new estimators that combine the predictive potential of ML and the ability of traditional statistical methods to make inference about population parameters have been proposed. For epidemiologists interested in taking advantage of ML for causal inference investigations, we provide an overview of three estimators that represent the current state-of-art, namely Targeted Maximum Likelihood Estimation (TMLE), Augmented Inverse Probability Weighting (AIPW) and Double/Debiased Machine Learning (DML).

Mental health impacts of spousal caregiving intensity in the US.

In the US, spouses are a major source of informal care for older adults and, therefore, key to the US national strategy to provide long-term care to the growing population of older adults. Understanding the mental health impacts of spousal caregiving is therefore critically important. Existing studies on the topic have often been limited by methodological limitations, and most overlook the role of caregiving intensity. In this study, we assess the impact of providing different intensities of caregiving to a spouse on mental health outcomes using data from the Health and Retirement Study. We address the endogeneity of the decision to provide different caregiving intensities using an augmented inverse probability weighted (AIPW) estimator adapted to handle multivalued treatments. We check the robustness of our estimates to the AIPW's identifying assumptions by re-estimating the model using a dynamic fixed effects estimator. We find strong evidence that becoming a caregiver is associated with worsened mental health outcomes for women, while the evidence for men is weaker. When disaggregating by caregiving intensity, we find that transitions into high-intensity caregiving lead to large increases in depression symptoms and the probability of exhibiting major depression for both men and women. The overall mental health impact of transiting into caregiving is driven almost exclusively by the negative impact of high-intensity caregiving, as transitions into low- and moderate-intensity caregiving are not associated with worsening mental health in most specifications. This study provides timely insights that can inform the targeting of long-term care policies and programs aimed at supporting family caregivers of older adults.

When does adjusting covariate under randomization help? A comparative study on current practices

PurposeWe aim to thoroughly compare past and current methods that leverage baseline covariate information to estimate the average treatment effect (ATE) using data from of randomized clinical trials (RCTs). We especially focus on their performance, efficiency gain, and power.MethodsWe compared 6 different methods using extensive Monte-Carlo simulation studies: the unadjusted estimator, i.e., analysis of variance (ANOVA), the analysis of covariance (ANCOVA), the analysis of heterogeneous covariance (ANHECOVA), the inverse probability weighting (IPW), the augmented inverse probability weighting (AIPW), and the overlap weighting (OW) as well as the augmented overlap weighting (AOW) estimators. The performance of these methods is assessed using the relative bias (RB), the root mean square error (RMSE), the model-based standard error (SE) estimation, the coverage probability (CP), and the statistical power.ResultsEven with a well-executed randomization, adjusting for baseline covariates by an appropriate method can be a good practice. When the outcome model(s) used in a covariate-adjusted method is closer to the correctly specified model(s), the efficiency and power gained can be substantial. We also found that most covariate-adjusted methods can suffer from the high-dimensional curse, i.e., when the number of covariates is relatively high compared to the sample size, they can have poor performance (along with lower efficiency) in estimating ATE. Among the different methods we compared, the OW performs the best overall with smaller RMSEs and smaller model-based SEs, which also result in higher power when the true effect is non-zero. Furthermore, the OW is more robust when dealing with the high-dimensional issue.ConclusionTo effectively use covariate adjustment methods, understanding their nature is important for practical investigators. Our study shows that outcome model misspecification and high-dimension are two main burdens in a covariate adjustment method to gain higher efficiency and power. When these factors are appropriately considered, e.g., performing some variable selections if the data dimension is high before adjusting covariate, these methods are expected to be useful.

Augmented inverse probability weighted estimation for conditional treatment effect

The Augmented Inverse Probability Weighted (AIPW) estimation has been extensively studied in various scenarios across diverse research areas. This investigation explores the asymptotic properties of the AIPW estimator for the conditional average treatment effect. Under various combinations of the parametric, semiparametric, and nonparametric structure in the nuisance propensity score and outcome regression models, we discuss the asymptotic bias and compare the asymptotic variances among the corresponding estimators. The study covers the asymptotic properties with no model misspecified; with either propensity score or outcome regressions locally/globally misspecified; and with all models locally/globally misspecified. To provide a deeper insight into the nature of these estimators and out of curiosity, we reveal the phenomenon that the asymptotic variances, with model-misspecification, could sometimes be even smaller than those with all models correctly specified. We also conduct a numerical study to validate the theoretical results.

Pooled analysis of the MANTICO2 and MONET randomized controlled trials comparing drug efficacy for early treatment of COVID-19 during Omicron waves

BackgroundThe clinical effectiveness of early therapies for mild-to-moderate COVID-19, comparing antivirals and monoclonal antibodies (mAbs) during the Omicron era, has not been conclusively assessed through a post-approval comparative trial. We present a pooled analysis of two randomized clinical trials conducted during Omicron waves. Methods The MANTICO2/MONET trial is a pooled analysis of two multicentric, independent, phase-4, three-arm, superiority, randomized, open-label trials. Nonhospitalized patients with early mild-to-moderate COVID-19 (≤5 days after symptoms’ onset) and at least one risk factor for disease progression were randomized 1:1:1 to receive 500mg of intravenous sotrovimab (SOT) or 600mg of intramuscular tixagevimab/cilgavimab (TGM/CGM) or oral 5-days course of nirmatrelvir/ritonavir (NMV/r) 300/100mg BID. Primary outcome was COVID-19-related hospitalization or death within 29 days after randomization. Fisher’s exact test for pooled data and incidence of failure was reported as overall and by arm with respective 95% CI.Pairwise comparisons across the arms were conducted using unadjusted exact logistic regression. An analysis by means of a doubly robust marginal model using augmented inverse probability weighting (AIPW) was also conducted to estimate the potential outcomes (Pom) in each treatment group and their difference by the average treatment effect (ATE). Analysis of symptom persistence within 30 days after randomization was performed using a 2-level hierarchical mixed-effects logistic model with a random intercept at the patient’s level. Point estimates and 95% confidence intervals were adjusted for age and sex and calculated using ANOVA-like methods for the mixedeffects logistic model. These trials are registred with the European Clinical Trials Database, EudraCT2021-002612-31 (MANTICO2) and EudraCT2021-004188-28 (MONET) and ClinicalTrials.gov, NCT05321394 (MANTICO2) Findings Between March 2022 and February 2023, 991 patients (SOT=332, TGM/CGM =327, NMV/r=332) were enrolled in 15 Italian centersThe overall mean age was 66 years; 482 participants (48.80%) were male and 856 vaccinated with at least a primary course ( 86%). Among the 8/991 hospitalizations observed, one resulted in death. The overall estimate of failure was 0.81% (95%CI; 0.35-1.58%). The odds ratio (OR) for the primary outcome in the NMV/r arm compared to the TGM/CGM and SOT arms was 8.41 (95% CI 1.21 to infinity; p=0.015) and 2.42 (95% CI 0.19 to infinity; p=0.499), respectively. No significant difference was observed between SOT and TGM/CGM (OR 0.32; 95% CI 0.032-1.83; p=0.174). Results were similar when we applied the marginal weighted model accounting for potential residual confounding bias. There was no evidence for a difference in the prevalence of symptoms between treatment groups, except for cough, which was higher in the SOT group compared to the other two groups at the 21-day follow-up (P=0.039) and a higher prevalence of nausea at the 7-day follow-up in the NMV/r group compared to the mAbs group (p=0.036). Interpretation NMV/r was superior to TGM/CGM in reducing hospital admission or death in clinical vulnerable patients with SARS-CoV-2 infection treated within 5 days of symptoms’ onset. No significant difference in symptom prevalence over time across the arms was found. Funding: The trials were funded by the Italian Agency of Drugs (AIFA) in 2021.

How Does CEO Duality Influence ESG Scores in Hospitality and Tourism Companies? Confounding Roles of Governance Mechanisms and Financial Indicators

Previous studies have yielded inconsistent results about the impact of CEO duality on corporate performance in the hospitality and tourism (H&T) industry. To further delve into this relationship, we investigated the causal relationship between CEO duality and environmental, social, and governance (ESG) performance under various board characteristics and financial indicators. The data from the Thomson Reuters Eikon database were evaluated using a machine learning technique that included targeted maximum likelihood estimation (TMLE), augmented inverse probability weighting (AIPW), and neural network analysis, all of which are doubly robust estimators with cross-fitting. The findings suggest that CEO duality negatively impacts environmental pillar scores but not other outcomes (i.e., governance and social pillar scores). Among the governance practices and financial indicators, policy executive compensation performance, policy executive compensation ESG performance, and return on invested capital (ROIC) have positive relations with total ESG scores. The results have important ramifications for helping H&T companies develop effective boards of directors and governance systems, as well as achieve targeted ESG performance objectives.

Spinal fusion surgery use among adults with low back pain enrolled in a digital musculoskeletal program: an observational study

ObjectivesTo compare 12-month spinal fusion surgery rates in the setting of low back pain among digital musculoskeletal (MSK) program participants versus a comparison cohort who only received usual care.Study DesignRetrospective cohort study with propensity score matched comparison cohort using commercial medical claims data representing over 100 million commercially insured lives.MethodsAll study subjects experienced low back pain between January 2020 and December 2021. Digital MSK participants enrolled in the digital MSK low back program between January 2020 and December 2021. Non-participants had low back pain related physical therapy (PT) between January 2020 and December 2021. Digital MSK participants were matched to non-participants with similar demographics, comorbidities and baseline MSK-related medical care use. Spinal fusion surgery rates at 12 months post participation were compared.ResultsCompared to non-participants, digital MSK participants had lower rates of spinal fusion surgery in the post-period (0.7% versus 1.6%; p < 0.001). Additionally, in the augmented inverse probability weighting (AIPW) model, digital MSK participants were found to have decreased odds of undergoing spinal fusion surgery (adjusted odds ratio: 0.64, 95% CI: 0.51–0.81).ConclusionsThis study provides evidence that participation in a digital MSK program is associated with a lower rate of spinal fusion surgery.

Analyzing Left-Truncated Samples with the Cox Model in the Presence of Missing Covariates

AbstractDelayed enrollment of subjects into a time-to-event study may result in a sample with biased outcome and covariate distributions. Additionally, missing covariate data may arise in these studies when information is difficult to collect due to patient burden or high testing costs. Some common missing data strategies, such as multiple imputation (MI) and augmented inverse probability weighting (AIPW), involve modeling the distribution of the missing covariate, which may be inaccurate with a left-truncated sample. Through simulation studies, we explore the performance of these methods in estimating Cox regression parameters under a variety of truncation and missing data scenarios. We find that MI and AIPW may be approximately unbiased when truncation is very low. Otherwise, biased estimation of the covariate distribution can cause MI to perform poorly. Similarly, AIPW may rely more heavily on correct estimation of the probability of having non-missing covariates in some settings. We apply these approaches to a Parkinson’s disease dementia biomarker study. In analyzing left-truncated data with missing covariates, careful consideration of the data characteristics and method assumptions is needed to obtain valid results.

Impact of the COVID-19 vaccinations on economic outcomes and mental health services in the U.S

The development of vaccines has long been viewed as a critical public health tool to raise immunity against the COVID-19 virus and to reopen economic activities globally. Against this background, this study examines the impact of COVID-19 vaccinations on economic outcomes and mental health services in the U.S. We employed the Household Pulse Survey covering 549, 950 and 57, 892 vaccinated and unvaccinated individuals, respectively. For the causal analysis, we control the differences between the vaccinated and unvaccinated groups using propensity score matching (PSM) and augmented inverse probability weighting (AIPW) methods. Economic outcomes include whether individuals work on-site, telework or work from home, in-store shopping, job loss, or food sufficient. We also consider whether individuals experience mental health distress and receive mental health services as a potential outcome. The empirical results show that the probability of working on-site, in-store shopping, food sufficiency, and using mental health services are significantly higher among the recipients of the COVID-19 vaccines than non-recipients. We also find that the likelihood of teleworking and job loss decreased significantly among the recipients of the COVID-19 vaccine compared to non-recipients. The magnitude of the estimated impact across race and ethnic groups is mixed, which shows that vaccination effects are heterogeneous across the races/ethnic groups in the sample.

The impact of family residence structure on adolescents' non-cognitive abilities: evidence from China.

The family residence structure serves as a crucial pathway through which the family environment influences adolescents' development. Drawing on nationally representative data, this study employs multiple linear regression models and propensity score matching to examine the impact of various family residence structures on adolescents' non-cognitive abilities. Causal identification is achieved through propensity score matching, while robustness is assessed using methods such as augmented inverse probability weighting and placebo tests. Heterogeneity analysis is conducted based on gender and household registration, aiming to explore the mechanisms by which family residence structure affects adolescents' non-cognitive abilities. The findings indicate that compared to two-parent co-residence households, three-generation co-residence families have significantly positive effects on emotional stability, conscientiousness, and agreeableness among adolescents. In contrast, skip-generation coresidence families exhibit significant negative effects on emotional stability and agreeableness in adolescents. Further investigation into the underlying mechanisms reveals that parental involvement and family socioeconomic status within three-generation co-residence families positively influence adolescents' non-cognitive abilities. This study highlights the importance of considering grandparents' role in adolescent growth and advocates for policy recommendations focusing on enhancing non-cognitive abilities in adolescents from skip-generation co-residence families.

Clinicians Who Practice Primarily in Nursing Homes and the Quality of End-of-Life Care Among Residents

Clinician specialization in the care of nursing home (NH) residents or patients in skilled nursing facilities (SNFs) has become increasingly common. It is not known whether clinicians focused on NH care, often referred to as SNFists (ie, physicians, nurse practitioners, and physician assistants concentrating their practice in the NH or SNF setting), are associated with a reduced likelihood of burdensome transitions in the last 90 days of life for residents, which are a marker of poor-quality end-of-life (EOL) care. To quantify the association between receipt of care from an SNFist and quality of EOL care for NH residents. This cohort study analyzed Medicare fee-for-service claims for a nationally representative 20% sample of beneficiaries to examine burdensome transitions among NH decedents at the EOL from January 1, 2013, through December 31, 2019. Statistical analyses were conducted from December 2022 to June 2023. Receipt of care from an SNFist, defined as physicians and advanced practitioners who provided 80% or more of their evaluation and management visits in NHs annually. This study used augmented inverse probability weighting in analyses of Medicare fee-for-service claims for a nationally representative 20% sample of beneficiaries. Main outcomes included 4 measures of burdensome transitions: (1) hospital transfer in the last 3 days of life; (2) lack of continuity in NHs after hospitalization in the last 90 days of life; (3) multiple hospitalizations in the last 90 days of life for any reason or any hospitalization for pneumonia, urinary tract infection, dehydration, or sepsis; and (4) any hospitalization in the last 90 days of life for an ambulatory care-sensitive condition. Of the 2 091 954 NH decedents studied (mean [SD] age, 85.4 [8.5] years; 1 470 724 women [70.3%]), 953 722 (45.6%) received care from SNFists and 1 138 232 (54.4%) received care from non-SNFists; 422 575 of all decedents (20.2%) experienced a burdensome transition at the EOL. Receipt of care by an SNFist was associated with a reduced risk of (1) hospital transfer in the last 3 days of life (-1.6% [95% CI, -2.5% to -0.8%]), (2) lack of continuity in NHs after hospitalization (-4.8% [95% CI, -6.7% to -3.0%]), and (3) decedents experiencing multiple hospitalizations for any reason or any hospitalization for pneumonia, urinary tract infection, dehydration, or sepsis (-5.8% [95% CI, -10.1% to -1.7%]). There was not a statistically significant association with the risk of hospitalization for an ambulatory care-sensitive condition in the last 90 days of life (0.0% [95% CI, -14.7% to 131.7%]). This study suggests that SNFists may be an important resource to improve the quality of EOL care for NH residents.

Quantifying bias due to missing data in quality of life surveys of advanced-stage cancer patients.

Many studies on cancer patients investigate the impact of treatment on health-related quality of life (QoL). Typically, QoL is measured longitudinally, at baseline and at predefined timepoints thereafter. The question is whether, at a given timepoint, patients who return their questionnaire (available cases, AC) have a different QoL than those who do not return their questionnaire (non-AC). We employed augmented inverse probability weighting (AIPW) to estimate the average QoL of non-AC in two studies on advanced-stage cancer patients. The AIPW estimator assumed data to be missing at random (MAR) and used machine learning (ML)-based methods to estimate answering probabilities of individuals at given timepoints as well as their reported QoL, as a function of auxiliary variables. These auxiliary variables were selected by medical oncologists based on domain expertise. We aggregated results both by timepoint and by time until death and compared AIPW estimates to the AC averages. Additionally, we used a pattern mixture model (PMM) to check sensitivity of our AIPW estimates against violation of the MAR assumption. Our study included 1927 patients with advanced pancreatic and 797 patients with advanced breast cancer. The AIPW estimate for average QoL of non-AC was below the average QoL of AC when aggregated by timepoint. The difference vanished when aggregated by time until death. PMM estimates were below AIPW estimates. Our results indicate that non-AC have a lower average QoL than AC. However, estimates for QoL of non-AC are subject to unverifiable assumptions about the missingness mechanism.

Association of Stressful Life Events With Oral Health Among Japanese Workers.

Psychological stress can cause various mental and physical health problems. The previous results on stress and oral health are inconsistent, possibly because of the narrow stress measurements. We aimed to examine the association between a broader range of stressful life events and oral health among workers. This cross-sectional study analyzed anonymous individual data from a national survey in Japan. Data on stressful life events, oral health problems which are one or more of tooth pain, gum swelling/bleeding, and difficulty chewing, and covariates were obtained using a self-reported questionnaire. Covariates used included gender, age group, and disease under treatment. Logistic regression analysis was used to estimate the association between stressful life events and oral health problems. We then estimated the causal treatment effects of stress using the augmented inverse-probability weighting (AIPW) method. Among the 274,881 subjects, 152,850 men (55.6%) and 122,031 women (44.4%) with a mean age of 47.0 (standard deviation, 14.4) years, 4.0% reported oral health problems, with a prevalence of 2.1% among those without any stress. The prevalence increased with stress score, reaching 15.4% for those with the maximum stress score. The adjusted odds ratio of this group compared to those without any stress was 9.2 (95% confidence interval [CI], 8.2-10.3). The estimated prevalence of oral health problems by the AIPW analysis was 2.2% (95% CI, 2.1-2.3%) for those without any stress and 14.4% (95% CI, 12.1-16.7%) for those with the maximum stress scores. There was a clear dose-response association between stressful life events and oral health problems.

Occupational feminization, gender, and wages: evidence from Thailand

This study examines the relationship between occupational feminization and wages in Thailand using the Thailand Labor Force Survey data (LFS) (2008-2017). Employing the Augmented Inverse Probability Weighting (AIPW) and the Inverse Probability Weighted Regression Adjustment (IPWRA) double robust matching techniques, this study investigates the effects of occupational feminization and the interaction effect of occupational feminization and gender on wages. The findings indicate that occupational feminization decreased wages in Thailand, and the negative effect weakened over time. While the gender wage gap persisted in feminized occupations over the years, there was a reduction in the gender wage gap in non-feminized occupations. The wage gap between females in feminized occupations and males in non-feminized occupations decreased over time, even though it remained larger than the overall gender wage gap. Male workers continued to take advantage and ride the “glass escalator” in both feminized and non-feminized occupations.

Sampling Importance Resampling Algorithm with Nonignorable Missing Response Variable Based on Smoothed Quantile Regression

The presence of nonignorable missing response variables often leads to complex conditional distribution patterns that cannot be effectively captured through mean regression. In contrast, quantile regression offers valuable insights into the conditional distribution. Consequently, this article places emphasis on the quantile regression approach to address nonrandom missing data. Taking inspiration from fractional imputation, this paper proposes a novel smoothed quantile regression estimation equation based on a sampling importance resampling (SIR) algorithm instead of nonparametric kernel regression methods. Additionally, we present an augmented inverse probability weighting (AIPW) smoothed quantile regression estimation equation to reduce the influence of potential misspecification in a working model. The consistency and asymptotic normality of the empirical likelihood estimators corresponding to the above estimating equations are proven under the assumption of a correctly specified parameter working model. Furthermore, we demonstrate that the AIPW estimation equation converges to an IPW estimation equation when a parameter working model is misspecified, thus illustrating the robustness of the AIPW estimation approach. Through numerical simulations, we examine the finite sample properties of the proposed method when the working models are both correctly specified and misspecified. Furthermore, we apply the proposed method to analyze HIV—CD4 data, thereby exploring variations in treatment effects and the influence of other covariates across different quantiles.

Estimation in optimal treatment regimes based on mean residual lifetimes with right-censored data.

An optimal individualized treatment regime (ITR) is a decision rule in allocating the best treatment to each patient and, hence, maximizing overall benefits. In this paper, we propose a novel framework based on nonparametric inverse probability weighting (IPW) and augmented inverse probability weighting (AIPW) estimators of the value function when the data are subject to right censoring. In contrast to most existing approaches that are designed to maximize the expected survival time under a binary treatment framework, the proposed method targets maximizing the mean residual lifetime of patients. Specifically, the proposed IPW method searches the optimal ITR by maximizing an estimator for the overall population outcome directly, without specifying the regression model for the conditional mean residual lifetime, whereas the AIPW method integrates the model information of the mean residual lifetime to improve the robustness. Furthermore, to overcome the computational difficulty in a nonsmooth value estimator, smoothed IPW and AIPW estimators are constructed. In theory, we establish the asymptotic properties of the proposed method under suitable regularity conditions. The empirical performances of the proposed IPW and AIPW estimators are evaluated using simulation studies and are further illustrated with an application to the real-world data set from the Acquired Immunodeficiency Syndrome Clinical Trial Group Protocol 175 (ACTG175).

An efficient doubly-robust imputation framework for longitudinal dropout, with an application to an Alzheimer’s clinical trial

We develop a novel doubly-robust (DR) imputation framework for longitudinal studies with monotone dropout, motivated by the informative dropout that is common in FDA-regulated trials for Alzheimer's disease. In this approach, the missing data are first imputed using a doubly-robust augmented inverse probability weighting (AIPW) estimator, then the imputed completed data are substituted into a full-data estimating equation, and the estimate is obtained using standard software. The imputed completed data may be inspected and compared to the observed data, and standard model diagnostics are available. The same imputed completed data can be used for several different estimands, such as subgroup analyses in a clinical trial, allowing for reduced computation and increased consistency across analyses. We present two specific DR imputation estimators, AIPW-I and AIPW-S, study their theoretical properties, and investigate their performance by simulation. AIPW-S has substantially reduced computational burden compared to many other DR estimators, at the cost of some loss of efficiency and the requirement of stronger assumptions. Simulation studies support the theoretical properties and good performance of the DR imputation framework. Importantly, we demonstrate their ability to address time-varying covariates, such as a time by treatment interaction. We illustrate using data from a large randomized Phase III trial investigating the effect of donepezil in Alzheimer's disease, from the Alzheimer's Disease Cooperative Study (ADCS) group.

Bias-corrected and doubly robust inference for the three-level longitudinal cluster-randomized trials with missing continuous outcomes and small number of clusters: Simulation study and application to a study for adults with serious mental illnesses

Longitudinal cluster-randomized designs have been popular tools for comparative effective research in clinical trials. The methodologies for the three-level hierarchical design with longitudinal outcomes need to be better understood under more pragmatic settings; that is, with a small number of clusters, heterogeneous cluster sizes, and missing outcomes. Generalized estimating equations (GEEs) have been frequently used when the distribution of data and the correlation model are unknown. Standard GEEs lead to bias and an inflated type I error rate due to the small number of available clinics and non-completely random missing data in longitudinal outcomes. We evaluate the performance of inverse probability weighted (IPW) estimating equations, with and without augmentation, for two types of missing data in continuous outcomes and individual-level treatment allocation mechanisms combined with two bias-corrected variance estimators. Our intensive simulation results suggest that the proposed augmented IPW method with bias-corrected variance estimation successfully prevents the inflation of false positive findings and improves efficiency when the number of clinics is small, with moderate to severe missing outcomes. Our findings are expected to aid researchers in choosing appropriate analysis methods for three-level longitudinal cluster-randomized designs. The proposed approaches were applied to analyze data from a longitudinal cluster-randomized clinical trial involving adults with serious mental illnesses.

Labeling effects of initial juvenile justice system processing decision on youth interpersonal ties*

AbstractThe juvenile justice system can process youth in myriad ways. Youth who are formally processed, relative to being informally processed, may experience more public and harsh sanctions that label youth more negatively as “deviant.” Drawing on labeling theory, the current study evaluates the relative effect of formal justice system processing on the interpersonal dynamics of youth peer networks. Using data from the Crossroads Study, a multisite longitudinal sample of first‐time adolescent offenders, the current study applies augmented inverse probability weighting and generalized mixed‐effects models to estimate the effects of formal processing on friendship selection processes of homophily and withdrawal and considers whether these effects vary by race and ethnicity. Consistent with expectations of homophily, formally processed youth acquire more new deviant peers and fewer nondeviant peers during the 3 years after their initial processing decision compared with informally processed youth. The findings suggest no differences exist across processing types in withdrawal from friends. These effects were consistent across racial and ethnic groups. Ultimately, this study explores the dynamic interpersonal mechanisms associated with labeling theory and offers additional insight into the negative effects of formal processing.