Joel A. Dubin’s research while affiliated with University of Waterloo and other places

Purpose To estimate six-month prevalence of child mental health service contacts and quantify associations between child health status and mental health service contacts, including number of types of contacts. Methods Data come from 6,242 children aged 4–17 years in the Ontario Child Health Study. A list of chronic conditions developed by Statistics Canada measured physical illness. The Emotional Behavioural Scales assessed mental illness. Child health status was categorized as healthy, physical illness only, mental illness only, and multimorbid (≥ 1 physical and ≥ 1 mental illness). Mental health service contact was aggregated to general medicine, urgent medicine, specialized mental health, school-based, alternative, and any contact (≥ 1 of the aforementioned contacts). Regression models quantified associations between health status and type of mental health contact, including number of types of contacts. Results Weighted prevalence estimates showed 261,739 (21.4%) children had mental health-related service contact, with school-based services being the most common contact amongst all children, regardless of health status. Children with multimorbidity had higher odds for every mental health contact than healthy controls (OR range: 4.00-6.70). A dose-response was observed, such that the number of contacts increased from physical illness only (OR = 1.49, CI: 1.10–1.99) to mental illness only (OR = 3.39, CI: 2.59–4.44) to multimorbidity (OR = 4.13, CI: 2.78–6.15). Conclusion Over one-fifth of children had mental health-related service contact and contacts were highest among children with multimorbidity. Types of mental health contacts for children with multimorbidity are diverse, with further research needed to elucidate the barriers and facilitators of mental health use.

In this article, we consider the imputation of missing responses in a longitudinal dataset via matrix completion. We propose a fixed‐effect, longitudinal, low‐rank model that incorporates both subject‐specific and time‐specific covariates. To solve the optimization problem, a two‐step optimization algorithm is proposed, which provides good statistical properties for the estimation of the fixed effects and the low‐rank term. In a theoretical investigation, the non‐asymptotic error bounds on the fixed effects and low‐rank term are presented. We illustrate the finite‐sample performance of the proposed algorithm via simulation studies, and apply our method to a power plant SO emissions dataset in which the monthly recorded amounts of emissions data on monitors are subject to missingness.

Background Peritoneal dialysis (PD) is being promoted because it is cost-effective and has equivalent outcomes to facility-based hemodialysis (HD). Determining PD eligibility is critical but subjective, with high variability among renal programs. This study aimed to establish a predictive model for PD eligibility among individuals who started treatment with HD. A secondary objective was to identify predictors of PD eligibility and determine if eligible patients went on to receive PD. Methods This retrospective cohort study included individuals starting HD at multiple hospitals in Alberta, Canada, as part of the START program between 1 October 2016 and 31 March 2018. Twenty-seven predictors, including patient characteristics, laboratory values, and comorbidities, were considered in logistic regression modeling. The outcome variable was PD eligibility, as determined by a standardized interdisciplinary assessment. The model selection was based on the Akaike information criterion. The confusion matrix was used for each model to compare the predicted versus observed eligibility. The final model was calibrated and presented. Results Among the 598 participants, 391 (65.4%) were considered eligible for PD. The logistic regression model achieved a modest performance in discriminating patients who were eligible for PD, with a high sensitivity of 91.3%, an accuracy of 0.68 (95% CI, 0.65–0.72), and an area under the receiver operating characteristic curve ranging from 0.69 to 0.71. Age (OR = 0.98; 95% CI, 0.97–0.99), body mass index (OR = 0.95; 95% CI, 0.93–0.97), starting dialysis in intensive care unit (OR = 0.53; 95% CI, 0.31–0.92), and polycystic kidney disease (OR = 0.37; 95% CI, 0.13–0.99) were statistically significant factors associated with a lower likelihood of being considered eligible for PD. Out of the 391 eligible PD patients, 87 (22.3%) received PD treatment within 6 months of starting HD. Conclusions The majority of patients starting HD were considered eligible for PD. Our model exhibits a high level of sensitivity and could serve as a valuable tool for screening potential candidates following the commencement of HD.

Identifying interventions that are optimally tailored to each individual is of significant interest in various fields, in particular precision medicine. Dynamic treatment regimes (DTRs) employ sequences of decision rules that utilize individual patient information to recommend treatments. However, the assumption that an individual's treatment does not impact the outcomes of others, known as the no interference assumption, is often challenged in practical settings. For example, in infectious disease studies, the vaccine status of individuals in close proximity can influence the likelihood of infection. Imposing this assumption when it, in fact, does not hold, may lead to biased results and impact the validity of the resulting DTR optimization. We extend the estimation method of dynamic weighted ordinary least squares (dWOLS), a doubly robust and easily implemented approach for estimating optimal DTRs, to incorporate the presence of interference within dyads (i.e., pairs of individuals). We formalize an appropriate outcome model and describe the estimation of an optimal decision rule in the dyadic‐network context. Through comprehensive simulations and analysis of the Population Assessment of Tobacco and Health (PATH) data, we demonstrate the improved performance of the proposed joint optimization strategy compared to the current state‐of‐the‐art conditional optimization methods in estimating the optimal treatment assignments when within‐dyad interference exists.

Objective This epidemiological study estimated the lifetime prevalence of chronic physical illness (i.e., an illness that lasted or was expected to last ≥6 months) and 6-month prevalence of mental disorder and multimorbidity (i.e., ≥1 physical illness and ≥1 mental disorder) in youth. Associations between physical illness and mental disorder were quantified, including the number of illnesses. Secondary objectives examined factors associated with mental disorder, after controlling for physical illness. Methods Data come from 10,303 youth aged 4–17 years in the 2014 Ontario Child Health Study (OCHS). Physical illness was measured using a list of chronic conditions developed by Statistics Canada. Mental disorders were measured using the OCHS Emotional Behavioural Scales. The Health Utility Index Mark III assessed overall functional health. Results Weighted prevalence estimates showed 550,090 (27.8%) youth had physical illness, 291,986 (14.8%) had mental disorder, and 108,435 (5.4%) had multimorbidity. Physical illness was not associated with mental disorder. However, youth with 2 physical illnesses, as compared to no physical illnesses, had increased odds of having any mental (OR = 1.75 [1.08, 2.85]), mood (OR = 2.50 [1.39, 4.48]) and anxiety disorders (OR = 2.40 [1.33, 4.31]). Mean functional health scores demonstrated a dose–response association across health status categories, with the highest scores among healthy youth and the lowest scores among multimorbid youth (all p < .05). Conclusion Chronic physical illness and mental disorders are prevalent in youth. Youths with 2 physical illnesses have a higher likelihood of mental disorders. Higher functional health scores protected against all mental disorders. Mental health interventions for youth should promote strong overall functional health.

In causal inference, interference takes place when the intervention on one unit affects the outcome of other units. Most of the previous methods for estimating network causal effects assume that the covariate information is complete, which may lead to biased estimates when missingness exists. In this study, we consider the partial and direct interference setting. Specifically, the whole population can be divided into different clusters. Within each cluster, the outcome of each unit is dependent on the intervention received by other units, but not dependent on the confounders or outcomes of other units within the same cluster or of those in different clusters. We also assume that the confounders are subject to non-ignorable missingness, and a confounder is considered as missing if any component of it is missing. We propose three consistent estimators for the direct, indirect, total, and overall effect of the intervention on the outcome, and derive the asymptotic results accordingly. A comprehensive study is carried out as well to investigate the finite sample properties of the proposed estimators. We illustrate the proposed methods by analyzing the dataset collected from an acid rain program, which was launched to reduce air pollution in the United States by encouraging the scrubber’s installation on power plants, where the records of some operating characteristics of the power generating facilities are subject to missingness.

Background Evidence suggests that having a chronic physical illness (CPI; eg, asthma, diabetes, and epilepsy) is an independent risk factor for suicidality (ie, suicidal ideation or attempts) among youth. Less is known about the mechanisms linking CPI and suicidality. Some evidence suggests that mental illness (eg, depression and anxiety) or neurodevelopmental disorder (eg, attention-deficit/hyperactivity disorder) mediates or moderates the CPI-suicidality association. Missing from the knowledge base is information on the association between having co-occurring CPI and mental illness or neurodevelopmental disorder (MIND) on youth suicidality. Objective This study uses epidemiological data from the 2019 Canadian Health Survey of Children and Youth (CHSCY) to study the intersection of CPI, MIND, and suicidality in youth. We will estimate prevalence, identify predictors, and investigate psychosocial and service use outcomes for youth with CPI-MIND comorbidity versus other morbidity groups (ie, healthy, CPI only, and MIND only). Methods Conducted by Statistics Canada, the CHSCY collected data from 47,850 children (aged 1-17 years) and their primary caregiving parent. Measures of youth CPI, MIND, family environment, and sociodemographics are available using youth and parent informants. Information on psychiatric services use is available via parent report and linkage to national administrative health data found in the National Ambulatory Care Reporting System and the Discharge Abstract Database, which allow the investigation of hospital-based mental health services (eg, emergency department visits, hospitalizations, and length of stay in hospital). Questions about suicidality were restricted to youths aged 15-17 years (n=6950), which form our analytic sample. Weighted regression-based analyses will account for the complex survey design. Results Our study began in November 2023, funded by the American Foundation for Suicide Prevention (SRG-0-008-22). Access to the linked CHSCY microdata file was granted in May 2024. Initial examination of CHSCY data shows that approximately 20% (1390/6950) of youth have CPI, 7% (490/6950) have MIND, 7% (490/6950) seriously considered suicide in the past year, and 3% (210/6950) had attempted suicide anytime during their life. Conclusions Findings will provide estimates of suicidality among youth with CPI-MIND comorbidity, which will inform intervention planning to prevent loss of life in this vulnerable population. Modeling correlates of suicidality will advance understanding of the relative and joint effects of factors at multiple levels—information needed to target prevention efforts and services. Understanding patterns of psychiatric service use is vital to understanding access and barriers to services. This will inform whether use matches need, identifying opportunities to advise policy makers about upstream resources to prevent suicidality. Importantly, findings will provide robust baseline of information on the link between CPI-MIND comorbidity and suicidality in youth, which can be used by future studies to address questions related to the impact of the COVID-19 pandemic and associated countermeasures in this vulnerable population of youth. International Registered Report Identifier (IRRID) DERR1-10.2196/57103

In causal inference, properly selecting the propensity score (PS) model is a popular topic and has been widely investigated in observational studies. In addition, there is a large literature concerning the missing data problem. However, there are very few studies investigating the model selection issue for causal inference when the exposure is missing at random (MAR). In this paper, we discuss how to select both imputation and PS models, which can result in the smallest RMSE of the estimated causal effect. Then, we provide a new criterion, called the ``rank score" for evaluating the overall performance of both models. The simulation studies show that the full imputation plus the outcome-related PS models lead to the smallest RMSE and the rank score can also pick the best models. An application study is conducted to study the causal effect of CVD on the mortality of COVID-19 patients.

How to deal with missing data in observational studies is a common concern for causal inference. When the covariates are missing at random (MAR), multiple approaches have been provided to help solve the issue. However, if the exposure is MAR, few approaches are available and careful adjustments on both missingness and confounding issues are required to ensure a consistent estimate of the true causal effect on the response. In this article, a new inverse probability weighting (IPW) estimator based on weighted estimating equations (WEE) is proposed to incorporate weights from both the missingness and propensity score (PS) models, which can reduce the joint effect of extreme weights in finite samples. Additionally, we develop a triple robust (TR) estimator via WEE to further protect against the misspecification of the missingness model. The asymptotic properties of WEE estimators are proved using properties of estimating equations. Based on the simulation studies, WEE methods outperform others including imputation-based approaches in terms of bias and variability. Finally, an application study is conducted to identify the causal effect of the presence of cardiovascular disease on mortality for COVID-19 patients.

Purpose Adolescent depression is a significant public health concern, and studying its multifaceted factors using traditional methods possess challenges. This study employs random forest (RF) algorithms to determine factors predicting adolescent depression scores. Methods This study utilized self-reported survey data from 56,008 Canadian students (grades 7–12) attending 182 schools during the 2021/22 academic year. RF algorithms were applied to identify the correlates of (i) depression scores (CESD-R-10) and (ii) presence of clinically relevant depression (CESD-R-10 ≥ 10). Results RF achieved a 71% explained variance, accurately predicting depression scores within a 3.40 unit margin. The top 10 correlates identified by RF included other measures of mental health (anxiety symptoms, flourishing, emotional dysregulation), home life (excessive parental expectations, happy home life, ability to talk to family), school connectedness, sleep duration, and gender. In predicting clinically relevant depression, the algorithm showed 84% accuracy, 0.89 sensitivity, and 0.79 AUROC, aligning closely with the correlates identified for depression score. Conclusion This study highlights RF’s utility in identifying important correlates of adolescent depressive symptoms. RF’s natural hierarchy offers an advantage over traditional methods. The findings underscore the importance and additional potential of sleep health promotion and school belonging initiatives in preventing adolescent depression.