Search

In the causal mediation framework, a number of parametric regression-based approaches have been introduced in recent years for estimating natural direct and indirect effects for a binary outcome in an exact manner, without invoking simplifying assumptions based on the rareness or commonness of the outcome. However, most of these works have focused on a binary mediator. In this article, we aim at a continuous mediator and introduce an exact approach for the estimation of natural effects on the odds ratio, risk ratio, and risk difference scales. Our approach relies on logistic and linear models for the outcome and mediator, respectively, and uses numerical integration to calculate the nested counterfactual probabilities underlying the definition of natural effects. Formulas for the delta method standard errors for all effects estimators are provided. The performance of our proposed exact estimators was evaluated in simulation studies that featured scenarios with different levels of outcome rareness/commonness, including a marginally but not conditionally rare outcome scenario. Furthermore, we evaluated the merit of Firth's penalization to mitigate the bias in the logistic regression coefficients estimators for the smallest outcome prevalences and sample sizes investigated. Using a SAS macro provided, we implemented our approach to assess the effect of placental abruption on low birth weight mediated by gestational age. We found that our exact natural effects estimators worked properly in both simulated and real data applications.

In the causal mediation framework, several parametric-regression–based approaches have been introduced in the last decade for estimating natural direct and indirect effects. For a binary outcome, a number of proposed estimators use a logistic model and rely on specific assumptions or approximations that may be delicate or not easy to verify in practice. To circumvent the challenges prompted by the rare outcome assumption in this context, an exact closed-form natural-effects estimator on the odds ratio scale was recently introduced for a binary mediator. In this work, we further push this exact approach and extend it for the estimation of natural effects on the risk ratio and risk difference scales. Explicit formulas for the delta method standard errors are provided. The performance of our proposed exact estimators is demonstrated in simulation scenarios featuring various levels of outcome rareness/commonness. The total effect decomposition property on the multiplicative scales is also examined. Using a SAS macro (SAS Institute, Inc., Cary, North Carolina) we developed, our approach is illustrated to assess the separate effects of exposure to inhaled corticosteroids and placental abruption on low birth weight mediated by prematurity. Our exact natural-effects estimators are found to work properly in both simulations and the real data example.

Background The prevalence of vertebral deformities in long-term survivors of childhood acute lymphoblastic leukemia (ALL) is unknown. Our objectives were to identify the prevalence of vertebral deformities and their risk factors among long-term childhood ALL survivors. Methods/Results We recruited 245 (49% male) long-term childhood ALL survivors from the Preventing Late Adverse Effects of Leukemia Cohort (French-Canadian ALL survivors treated between the years 1987 and 2010 with the Dana Farber Cancer Institute clinical trials protocols, who did not experience disease relapse and/or receive hematopoietic stem cell transplant). Median age at recruitment was 21.7 years (range, 8.5-41) and median time since diagnosis was 15.1 years (range, 5.4-28.2). All participants underwent spine radiograph and dual-energy X-ray absorptiometry scans. The prevalence of vertebral deformity was 23% with 88% classified as grade 1 according to the Genant method. The majority of vertebral deformities were clinically silent. Regression analysis confirmed male sex (risk ratio [RR] = 1.94; 95% confidence interval [CI], 1.16-3.24; P = 0.011), higher glucocorticoid cumulative dose (RR = 1.05; 95% CI, 1.00-1.10; P = 0.032), and back pain (RR = 2.44; 95% CI, 1.56-3.84; P < 0.001) as predictors of prevalent vertebral deformity. Sex differences in vertebral deformity predictors emerged. Conclusions We report a significant prevalence of vertebral deformities in this young cohort. Male sex, cumulative glucocorticoid dose, and back pain were identified as predictors of prevalent vertebral deformity. Back pain emerging as a strong predictor of vertebral deformity underscores the importance of ongoing bone health surveillance in survivors with persistent vertebral deformities treated with these earlier protocols.

Although medical research frequently involves an exposure variable with three or more discrete levels, detailed presentations of mediation techniques for dealing with multicategorical (multilevel) exposures are sparse. In this paper, we study two causal mediation approaches applicable to such a type of exposure for continuous mediator and outcome: the closed-form regression-based approach of Valeri and VanderWeele, and the marginal structural model-based approach of Lange, Vansteelandt, and Bekaert. While the consideration of multicategorical exposures is found explicitly addressed in the literature for the latter approach, this is, to our knowledge, not yet the case for the former. We first illustrate the application of the two aforementioned approaches to assess the dose–response relationship between maternal intake of inhaled corticosteroids and birthweight, where this relationship is potentially mediated by gestational age. More specifically, we provide a precise roadmap for the application of the regression-based approach and of the marginal structural model-based approach on our cohort of pregnancies. Expressions for the natural direct and indirect effects associated with our categorical exposure are provided and, for the regression-based approach, analytic formulas for standard error calculation using the delta method are presented for these effects. Second, a simulation study which mimics our data is presented to add to current knowledge on these causal mediation techniques. Results from this study highlight the relevance to assess robustness of mediation results obtained from multicategorical exposures, most notably for the least prevalent of exposure categories.

Decades of research have shown a strong association between wellbeing, health, and social resources. LGBTQ2+ communities are among those who historically have been excluded from accessing quality social resources. However, little is known about how access to different types of resources influences mental health and wellbeing.Data were drawn from an online sample of 3890 LGBTQ2+ people aged 18 years and older in Quebec, Canada. We identified key social resource patterns (from family of origin, friends, partner, neighbourhood, and LGBTQ2+ community) and investigated differences in socio-demographic and health outcomes across classes.A five-class solution best fitted the data, highlighting distinctive patterns in access to five key social resources: moderate friend support access (42.14%), overall high support access (23.51%), high friend support access (18.06%), only close ties support access (10.90%) and overall low support access (5.39%). Marginalized groups (trans and non-binary people, racialized or disabled people, immigrants) were less likely to access diverse, high-quality social resources. Accessing diverse social resources, particularly close ties (e.g., family of origin), was associated with better health outcomes. In the absence of close ties, having at least one other social resource was associated with better health outcomes compared to having limited access to all resources.We found a major imbalance in social resource access among LGBTQ2+ people. Creating safe spaces for LGBTQ2+ people and ensuring access to high-quality social resources is important in sustaining their health and wellbeing.

Background Despite greater acceptance of sexual and gender diversity and the scientific consensus that same-gender attraction, creative gender expression, and transness are not mental illnesses, LGBTQI2+ persons are still commonly told that they can or should change their sexual orientation, gender identity, or gender expression (SOGIE). The aim of this study was to describe the prevalence of SOGIE conversion efforts, including their sociodemographic correlates, among LGBTQI2+ persons. Methods Using community-based sampling, we assessed SOGIE conversion attempts and involvement in conversion services of 3,261 LGBTQI2+ persons aged 18 years and older in Quebec, Canada. Results A quarter of respondents experienced SOGIE conversion attempts, and fewer than 5% were involved in conversion services. Over half of those who were involved in SOGIE conversion services consented to them, but the services’ goals were made clear and explicit to only 55% and 30% of those who engaged in SO and GIE conversion, respectively. The results also suggest that family plays a key role in SOGIE conversion attempts and services utilization, and that indigenous, intersex, transgender, non-binary, and asexual persons, people of colour, as well as individuals whose sexual orientation is not monosexual (i.e., bisexual, pansexual) were more likely to have been exposed to conversion attempts and involved in conversion services. Conclusions This study found that the prevalence of conversion efforts is substantial. Interventions to protect LGBTQI2+ people from such attempts should focus not only on legal bans, but also on supporting families who need to be counseled in accepting sexual and gender diversity. Health professionals need to be adequately trained in LGBTQI2+ affirmative approaches. Religious therapists should consult with colleagues and undergo supervision to ensure that their religious beliefs do not interfere with their practice.

Survivors of childhood acute lymphoblastic leukemia (cALL) are at higher risk of developing cardiometabolic complications. We aimed at exploring the associations between biomarkers of inflammation, oxidative stress, endothelial function, endotoxemia and cardiometabolic risk factors. We conducted a cross-sectional analysis in 246 cALL survivors (mean age, 22.1 ± 6.3 years; mean time since diagnosis, 15.5 ± 5.2 years) and evaluated the associations using a series of logistic regressions. Using structural equation models, we also tested if the relationship between endotoxemia and cardiometabolic complications was mediated by the latent (unobserved) variable inflammation inferred from the observed biomarkers CRP, TNF-α and IL-6. High leptin-adiponectin ratio was associated with obesity [adjusted OR = 15.7; 95% CI (6.2–39.7)], insulin resistance [20.6 (5.2–82.1)] and the metabolic syndrome [11.2 (2.6–48.7)]. Higher levels of plasminogen activator inhibitor-1 and tumor necrosis factor-α were associated with obesity [3.37 (1.6–7.1) and 2.34 (1.3–4.2), respectively] whereas high C-reactive protein levels were associated with insulin resistance [3.3 (1.6–6.8)], dyslipidemia [2.6 (1.4–4.9)] and MetS [6.5 (2.4–17.9)]. Our analyses provided evidence for a directional relationship between lipopolysaccharide binding protein, related to metabolic endotoxemia, inflammation and cardiometabolic outcomes. Identification of biomarkers and biological mechanisms could open new avenues for prevention strategies to minimize the long-term sequelae, improve follow-up and optimize the quality of life of this high-risk population.

Мета: Метою даної статті є розробка комплексу функціональних моделей контролю і діагностики технічного стану авіаційного двигуна ТВ3-117 в польотних режимах на основі нотації IDEF0. Методи: У статті застосовано метод системного моделювання на етапі проектування інформаційної системи контролю та діагностики технічного стану авіаційного двигуна ТВ3-117 в польотних режимах. Результати: Запропоновано підхід щодо формалізації процесу контролю і діагностики технічного стану авіаційного двигуна ТВ3-117 на основі методології системного аналізу і IDEF-технологій. Розроблено комплекс функціональних моделей процесу контролю і діагностики авіаційного двигуна на основі технології IDEF0, що дозволило виділити основний спектр завдань і сформулювати вимоги щодо їх реалізації в складі експертної системи контролю і діагностики технічного стану авіаційного двигуна ТВ3-117. Обговорення: Отримані результати застосовуються в рамках концепції інтелектуалізації процесу контролю і діагностики технічного стану авіаційного двигуна ТВ3-117 в польотних режимах, одним з пунктів якої є інтелектуальна обробка і зберігання інформації результатів льотних випробувань та експлуатації авіаційного двигуна ТВ3-117 на основі вимог сучасних баз даних і баз знань, з можливістю їх інтеграції в сучасні CASE-технології.

Summary Background The remarkable progress in the treatment of childhood acute lymphoblastic leukemia (cALL) has led to a survival rate reaching 90%. This success story is unfortunately linked to increased risk of impaired skeletal mass accumulation during childhood and adolescence, predisposing the patients to osteoporosis and pathological fractures at adulthood. Objective This study aims at characterizing the vitamin D status and bone health biomarkers in a well-characterized cohort of cALL survivors. Results Food frequency questionnaires reveal that (i) the total vitamin D intake varies greatly (44–2132 IU/d), (ii) only 16.8% of the participants consume vitamin D supplements, and (iii) 74% of survivors' intakes are below the Recommended Daily Intakes (400 IU/d). For the 42 participants taking vitamin D supplements, the median (2.5–97.5%iles) intake is 600 IU/d (21.2–1972 IU/d). Sixteen participants are vitamin D deficient ( Conclusion The present study shows that the prevalence of vitamin D insufficiency or deficiency is not greater in cALL survivors compared to the general Canadian population despite low vitamin D food and supplement intakes. Furthermore, there seem to be no overt imbalance in the gender- and age-adjusted serum bone turnover marker concentrations.

Building on the work of Schwartz et al, Joint Bayesian analysis of birthweight and censored gestational age using finite mixture models in Statistics in Medicine, we propose a Bayesian finite mixture of bivariate regression model for causal mediation analyses. Using an identifiability condition within each component of the mixture, we express the natural direct and indirect effects of the exposure on the outcome as functions of the component-specific regression coefficients. On the basis of simulated data, we examine the behavior of the model for estimating these effects in situations where the associations between exposure, mediator, and outcome are confounded or not. Additionally, we demonstrate that this mixture model can be used to account for heterogeneity arising through unmeasured binary or categorical mediator-outcome confounders. Considering gestational age as a potential mediator, we then illustrate our mediation mixture model to estimate the natural direct and indirect effects of exposure to inhaled corticosteroids during pregnancy on birthweight using a cohort of asthmatic women from the province of Quebec (Canada).

Supplemental material, SMM902794 Supplemental Material for Comparing two counterfactual-outcome approaches in causal mediation analysis of a multicategorical exposure: An application for the estimation of the effect of maternal intake of inhaled corticosteroids doses on birthweight by Mariia Samoilenko, Nadia Arrouf, Lucie Blais and Geneviève Lefebvre in Statistical Methods in Medical Research

More than two thirds of survivors have long-term adverse effects, and no study proposes a portrait of physical activity level in childhood acute lymphoblastic leukemia survivors. The aims of this study were to present the cardiorespiratory fitness (CRF) levels of survivors detailed overview sedentary activities portrait.A total of 247 childhood acute lymphoblastic leukemia survivors were included in our study. Survivors underwent a cardiopulmonary exercise test on ergocycle and completed physical activity and sedentary questionnaires to assess their leisure physical and sedentary activities and total daily energy expenditure.Up to 67% of survivors (84% below 18 y and 60% 18 y of age or above) did not fulfill the physical activity guidelines. Their CRF was reduced by almost 16% in regard to their predicted maximum oxygen consumption (VO2peak). Almost three quarters of the survivors (70% below 18 y and 76% 18 y of age or above) spent2 hours/day in leisure sedentary activities. Adult survivors who received high doses of anthracyclines and those who received radiation therapy had decreased odds to spend ≥2 hours/day in sedentary activities.Our results showed that survivors, especially children, were not active enough and had a reduced CRF. This study highlights the importance of promoting physical activity in survivors, especially because they are exposed to an increased risk of chronic health problems, which could be mitigated by physical activity.

Introduction: As the survival rate of childhood acute lymphoblastic leukemia (ALL) continues to improve, the physical deconditioning is becoming an increasingly common problem in survivors. The aim...

Background Childhood acute lymphoblastic leukemia (cALL) is the most frequent pediatric cancer. Over the past decades, treatment of cALL has significantly improved, with cure rates close to 90%. However intensive chemotherapy and cranial radiotherapy (CRT) during a critical period of a child’s development have been shown to lead to significant long-term side effects including cardiometabolic complications. Using the PETALE (Prévenir les effets tardifs des traitements de la leucémie aiguë lymphoblastique chez l’enfant) cALL survivor cohort, we investigated the association between combined cumulative corticosteroids (CS) doses and CRT exposures and obesity, insulin resistance, (pre-)hypertension, and dyslipidemia jointly. Methods A Bayesian multivariate latent-t model which accounted for our correlated binary outcomes was used for the analyses (n = 241 survivors). CS doses were categorized as low (LD) or high (HD). Combined exposure levels investigated were: 1) LD/no CRT; 2) LD/CRT, and; 3) HD/CRT. We also performed complementary sensitivity analyses for covariate adjustment. Results Prevalence of cardiometabolic complications ranged from 12.0% for (pre-)hypertension to 40.2% for dyslipidemia. The fully adjusted odds ratio (OR) for dyslipidemia associated with LD/CRT (vs. LD/No CRT) was OR = 1.98 (95% credible interval (CrI): 1.02 to 3.88). LD/CRT level also led to a 0.15 (95% CrI: 0.00 to 0.29) excess risk to develop at least one cardiometabolic complication. Except for obesity, adjusted results for the highest exposure category HD/CRT were generally similar to those for LD/CRT albeit not statistically significant. White blood cell count at diagnosis, a proxy for cALL burden at diagnosis, was found associated with insulin resistance (OR = 1.08 for a 10-unit increase (× 109/L), 95% CrI: 1.02 to 1.14). Conclusions Our results indicated that combined LD/CRT exposure is a likely determinant of dyslipidemia among cALL survivors. No evidence was found to suggest that high doses of CS lead to additional risk for obesity, insulin resistance, (pre-)hypertension, and dyslipidemia beyond that induced by CRT. The multivariate model selected for analyses was judged globally useful to assess potential exposure-related concomitance of binary outcomes. Electronic supplementary material The online version of this article (10.1186/s12874-019-0725-9) contains supplementary material, which is available to authorized users.

Background Most childhood acute lymphoblastic leukemia (ALL) survivors develop chronic treatment-related adverse effects several years after the end of therapy. A regular practice of physical activity and a good cardiorespiratory fitness have the potential to reduce the risk of chronic disease and improve quality of life. The aim of this study was to evaluate in a cohort of ALL survivors, the association between a good cardiorespiratory fitness or the respect of physical activity guidelines and major long-term health outcomes. Methods In total, 247 ALL survivors underwent a cardiopulmonary exercise test, completed a physical activity questionnaire and a battery of clinical examinations. We calculated the odds ratio to obtain the preventive fraction (PF) to evaluate the effects of the cardiorespiratory fitness and physical activity levels on health outcomes (ie, obesity, metabolic health, cardiac health, cognitive health and mood, bone health). Results Despite their young age, 88% of the participants presented at least one adverse health outcome, and 46% presented ≥3. Their cardiorespiratory fitness was also lower than expected with a median VO2 peak reaching 84% of the predicted value. In the analyses using cardiorespiratory fitness, statistically significant PFs were observed for obesity (0.30), low-high-density lipoprotein-cholesterol (0.21) and depression (0.26). In the physical activity level analyses, statistically significant PFs were observed for obesity, depression, and low bone mineral density, with a PF of 0.55, 0.81, and 0.60, respectively. Conclusions Our results indicate that a good cardiorespiratory fitness and physical activity level induced a preventive action for most health outcomes studied and was associated with a lower late adverse effects prevalence in ALL survivors.

In their accompanying article, Samoilenko and Lefebvre (Am J Epidemiol. 2019;188(7):1203–1205) correctly note 2 typographical errors in the formulas presented in a 2011 paper on placental abruption by Ananth and VanderWeele (Am J Epidemiol. 2011;174(1):99–108). Fortunately, to the best of our knowledge, researchers are using our methods papers and Dr. VanderWeele’s 2015 book on mediation analysis (Explanation in Causal Inference; Oxford University Press, New York, New York), rather than the paper on placental abruption, to carry out their direct and indirect effect analyses; and in our methods papers and the book, the formulas are correct. The formulas discussed by Samoilenko and Lefebvre and in our work make reference to a “rare outcome assumption.” In evaluating this assumption, it is important to note that the outcome is to be relatively rare across all strata defined by the exposure and the mediator—a point that is often neglected.

Our objectives were to assess the prevalence of cardiometabolic complications in children, adolescents, and young adult survivors of childhood acute lymphoblastic leukemia (cALL), to identify their predictors and the risk compared to the Canadian population. We performed a cardiometabolic assessment of cALL survivors from the PETALE cohort (n = 247, median age at visit of 21.7 years). In our group, overweight and obesity affected over 70% of women. Pre-hypertension and hypertension were mostly common in men, both adults (20%) and children (19%). Prediabetes was mainly present in women (6.1% of female adult survivors) and 41.3% had dyslipidemia. Cranial radiation therapy was a predictor of dyslipidemia (RR: 1.60, 95% CI: 1.07–2.41) and high LDL-cholesterol (RR: 4.78, 95% CI: 1.72–13.28). Male gender was a predictor for pre-hypertension and hypertension (RR: 5.12, 95% CI: 1.81–14.46). Obesity at the end of treatment was a predictor of obesity at interview (RR: 2.07, 95% CI: 1.37–3.14) and of metabolic syndrome (RR: 3.04, 95% CI: 1.14–8.09). Compared to the general population, cALL survivors were at higher risk of having the metabolic syndrome, dyslipidemia, pre-hypertension/hypertension and high LDL-cholesterol, while the risk for obesity was not different. Our results support the need for early screening and lifestyle intervention in this population.

Building on the work of Schwartz et al, Joint Bayesian analysis of birthweight and censored gestational age using finite mixture models in Statistics in Medicine, we propose a Bayesian finite mixture of bivariate regression model for causal mediation analyses. Using an identifiability condition within each component of the mixture, we express the natural direct and indirect effects of the exposure on the outcome as functions of the component-specific regression coefficients. On the basis of simulated data, we examine the behavior of the model for estimating these effects in situations where the associations between exposure, mediator, and outcome are confounded or not. Additionally, we demonstrate that this mixture model can be used to account for heterogeneity arising through unmeasured binary or categorical mediator-outcome confounders. Considering gestational age as a potential mediator, we then illustrate our mediation mixture model to estimate the natural direct and indirect effects of exposure to inhaled corticosteroids during pregnancy on birthweight using a cohort of asthmatic women from the province of Quebec (Canada).

Asthma is a heterogeneous disease, and responses to asthma medications vary noticeably among patients. A substantively oriented objective of this study was to explore the potentially heterogeneous effects of exposure to maternal inhaled corticosteroids (ICS) on gestational age (GA) at delivery and birth weight (BW) using a cohort of 6,197 pregnancies among women with asthma (Quebec, Canada, 1998-2008). A methodologically oriented objective was to comprehensively describe the application of a Bayesian 2-component mixture-of-bivariate-regressions model to address this issue and estimate the effects of ICS on GA and BW jointly. Based on the proposed model, no association between ICS and GA/BW was found for a large proportion of asthmatic pregnancies. However, a positive association between ICS exposure and GA/BW was revealed in a small subset of pregnancies comprising mainly preterm and low-birth-weight infants. A novel application of this model was also subsequently performed using BW z score instead of BW as the outcome variable. In conclusion, the studied mixture-of-bivariate-regressions model was useful for detecting heterogeneity in the effect of ICS on GA and BW in our population of women with asthma. These analyses pave the way for analogous uses of this model for general assessment of exposure effect heterogeneity for these perinatal outcomes.

Childhood cancer survivorship issues represent an established public health challenge. Most late adverse effects (LAEs) have been demonstrated to be time and treatment dependent. The PETALE study is a multidisciplinary research project aiming to comprehensively characterize LAEs and identify associated predictive biomarkers in childhood acute lymphoblastic leukemia (cALL) survivors.cALL survivors treated at Sainte-Justine University Health Center with Dana-Farber Cancer Institution-ALL protocols 87-01 through 2005-01 were eligible. During Phase I of the study, the participants underwent comprehensive clinical, biologic, and psychosocial investigation targeting metabolic syndrome, cardiotoxicity, bone morbidity, neurocognitive problems, and quality of life issues. Whole-exome sequencing was performed for all participants. Subjects identified with an extreme phenotype during Phase I were recalled for additional testing (Phase II).Phase I included 246 survivors (recall rate 71.9%). Of those, 85 participants completed Phase II (recall rate 88.5%). Survivors agreeing to participate in Phase I (n = 251) were similar to those who refused (n = 31) in terms of relapse risk profile, radiotherapy exposure, and age at the time of study. Participants, however, tended to be slightly older at diagnosis (6.1 vs. 4.7 years old, P = 0.08), with a higher proportion of female agreeing to participate compared with males (93.2 vs. 86.5%, P = 0.07).The PETALE study will contribute to comprehensively characterize clinical, psychosocial, biologic, and genomic features of cALL survivors using an integrated approach. Expected outcomes include LAE early detection biomarkers, long-term follow-up guidelines, and recommendations for physicians and health professionals.

Background Recent research has suggested that long-term pediatric cancer survivors were at risk of important physical and psychological morbidities. To date, we do not know to what extent functional health status contributes to psychological risk and which domains are most important. The aim of this study was to systematically explore which functional domain could explain anxiety, depression, and distress symptoms. Procedure We used data available for 105 adolescents and 182 adults successfully treated for childhood acute lymphoblastic leukemia at two Canadian sites part of the PETALE cohort. Participants were ≥5 years postdiagnosis, aged 22 ± 6 years, 52% female, and 49% acute lymphoblastic leukemia high-risk status. The contribution of health functional status (15D/16D questionnaires) to self-reported anxiety, depression, and distress (Beck scales and distress thermometer) was evaluated using adjusted logistic regression models. Results Prevalence rates found for mild-severe anxiety, depression, and distress were 14%, 21%, and 30% among adolescents and 27%, 20%, and 19% among adults. Frequent health domains associated with psychological risk were sleeping and breathing in adolescents, and vitality/fatigue, discomfort/symptoms, mental function, and sleeping in adults. Mental function was systematically associated with psychological risk across age groups (median OR = 10.00, 95% CI 3.01-33.71). Exploratory mediation bootstrapping analyses suggested that the effect on psychological risk of overall health status and mental function problems was partly explained by social/work/school functioning. Conclusion The results identified important functional health domains that could be targeted for interventions preventing psychological risk: vitality/fatigue, discomfort/symptoms, sleeping, and mental function issues. Health domains probably affect mood partly by limiting social/work/school functioning.

Table S2. Adjusted odds ratios (ORs) for individual cardiometabolic outcomes. (DOCX 18 kb)

Table S2. Adjusted odds ratios (ORs) for individual cardiometabolic outcomes. (DOCX 18 kb)

Table S1. Cut-off values for cardiometabolic outcomes. (DOCX 25 kb)

Table S1. Cut-off values for cardiometabolic outcomes. (DOCX 25 kb)