Covariates in studies examining longitudinal relationships between substance use and mental health problems among youth: A meta-epidemiologic review
Related Experiment Video
Updated: May 13, 2025

Author Spotlight: Understanding Adolescent Social Adversity Effects on Neurodevelopment in Mice
Published on: March 15, 2024
Jillian Halladay1, Steph Kershaw2, Emma K Devine2
1McMaster University School of Nursing, 1280 Main St W, Hamilton, ON L8S 4L8, USA; The Peter Boris Centre for Addictions Research, McMaster University/St. Joseph's Healthcare Hamilton, 100 West 5Th St, Hamilton, ON L8N 3K7, Canada; The Matilda Centre for Research in Mental Health and Substance Use, The University of Sydney, Australia.
Introduction:
This meta-epidemiological review examines covariate selection and reporting practices in observational studies analyzing longitudinal relationships between youth substance use and mental health problems (internalizing and externalizing).
Methods:
Sixty-nine studies published in high-impact journals from 2018 to 2023 were included. Studies were included if they explored prospective relationships between substance use and mental health among youth (12-25 years) and used repeated measures designs. Data extraction focused on study characteristics, covariates and their selection methods, and reporting practices.
Results:
There were 574 covariates included across studies; 33 were included as moderators and 18 were included as mediators. At the study level, the most common covariate domains included demographics (90 % of included studies had at least one demographic, mostly sex), substance-related variables (67 %; mostly alcohol or smoking), internalizing symptoms (39 %; mostly depression), family-related variables (29 %; mostly parental substance use or mental illness), and externalizing symptoms (19 %; mostly conduct). 93 % of studies had unique sets of lower-order covariates. Across all studies (n = 69), only 35 % provided details for how, and why, all covariates were selected with only 12 % reporting selecting covariates a priori, and none being pre-registered. Only 60 % mentioned confounding and only 13 % mentioned risk of confounding in their conclusions.
Conclusions:
The findings highlight the need for improved covariate selection and reporting practices. Establishing a core set of covariates and adhering to standardized reporting guidelines would enhance the comparability and reliability of research findings in this field. Researchers can use this review to identify and justify the inclusion and exclusion of commonly reported covariates when analyzing relationships between youth substance use and mental health problems.
Related Concept Videos
Stress and Mental Health
Individuals with depression often experience challenges in both their personal and professional...
Longitudinal Studies
Longitudinal Research
Substance Use Disorders Affecting Sleep
Understanding the concepts of physical dependence,...
Observational Studies
There are three types of observational studies – Prospective, retrospective, and cross-sectional.
Prospective Study
Prospective studies, also known as longitudinal or cohort studies, are carried out by collecting future data from groups sharing similar characteristics. One...
Drug Abuse and Addiction: Pharmacological Phenomena

