Alcohol, tobacco, and illegal drug use contribute significantly to global mortality and morbidity. However, comprehensive cross-national epidemiological data have been lacking, particularly in less-developed economies. The World Health Organization (WHO) has consistently highlighted this data gap as a critical barrier to effective global health policy. This research paper addresses this gap by presenting data on lifetime alcohol, tobacco, cannabis, and cocaine use from the first 17 countries participating in the WHO's World Mental Health (WMH) Survey Initiative. This initiative is particularly important because it includes a number of less-developed economies, offering a more global perspective than previous studies. Cross-national research in this area is inherently challenging due to variations in healthcare systems, cultural norms, and data collection methods. Previous studies, often limited to specific regions or relying on unreliable official statistics (especially for illegal drugs), have provided fragmented insights. While regional comparisons have been conducted within Europe and certain areas of the Americas, a truly global, age-inclusive, and comprehensive study of concurrent alcohol, tobacco, and illegal drug use was absent before this initiative. The WMH Surveys present a unique opportunity to compare countries with varying drug policies, including those with strict prohibition (e.g., the US) and those with harm reduction approaches (e.g., the Netherlands). The inclusion of both cocaine source and consumer countries further enhances the study's value. The paper's objectives are to document lifetime drug use, focusing on young adults; examine sociodemographic correlates; and investigate age of onset (AOO) distributions of alcohol, tobacco, cannabis, and cocaine use.

The research paper acknowledges the historical challenges of cross-national epidemiological studies, particularly concerning substance use. Early comparisons relied on nation-level consumption data (e.g., taxation records) correlated with mortality rates from related causes, but this method suffers from limitations in the comparability of death certification practices across countries and the inherent unreliability of official statistics for illegal drugs. The development of standardized survey research methods for alcohol, tobacco, and illegal drug use offered improvements, but cross-national harmonization of these protocols remained a major obstacle. Qualitative rapid assessment methods, valuable for within-country planning, lack the standardization needed for international comparisons. The paper cites several previous efforts, including regional studies in Europe and the Americas, and reports from the WHO on alcohol and tobacco use. These previous initiatives, however, either focused on specific regions, single substances, or specific age groups, lacking the comprehensive, global scope of the current study.

The study employed a large-scale, multi-national approach using data from 18 surveys conducted across 17 countries. These countries were selected based on the availability of collaborators and funding. The surveys were carried out in the Americas (Colombia, Mexico, USA), Europe (Belgium, France, Germany, Italy, Netherlands, Spain, Ukraine), the Middle East and Africa (Israel, Lebanon, Nigeria, South Africa), Asia (Japan, China), and Oceania (New Zealand). Multistage probability samples were utilized in all countries (except Japan, which used an unclustered two-stage probability sample). Face-to-face interviews (except in France, which used telephone interviews) were conducted by trained lay interviewers using the WHO Composite International Diagnostic Interview (CIDI) Version 3.0. This standardized instrument assessed lifetime use of alcohol, tobacco, cannabis, and cocaine. Participants who reported use were also asked about their age of onset for each substance. Age of first tobacco use was not assessed in several European countries and Japan. Sociodemographic data (age, sex, education, marital status, employment, and household income) were also collected. Sample sizes varied across countries, ranging from 2,372 to 12,992 participants, with a total sample size of 85,052. Response rates ranged from 46% to 88% across countries. Internal subsampling was used to reduce respondent burden and a complex weighting scheme was applied to account for differential sampling and to adjust the sample distributions to match population sociodemographic and geographic distributions. Extensive interviewer training and quality control protocols were implemented to maintain data quality and consistency across countries. Data analysis included estimation of cumulative lifetime incidence, AOO distributions using the two-part actuarial method, and examination of sociodemographic correlates using discrete-time survival analysis and logistic regression. All analyses accounted for the complex survey design.

The study revealed substantial cross-national variation in lifetime drug use prevalence. Lifetime alcohol use was highest in the Americas, Europe, Japan, and New Zealand, while considerably lower rates were observed in the Middle East, Africa, and China. Lifetime tobacco use was most common in the US, Lebanon, Mexico, and several European countries, with significantly lower rates in African nations. Cannabis use was exceptionally high in the US and New Zealand (both 42%), compared to other countries. The US was an outlier in cocaine use (16%), with significantly lower rates in other regions. Males consistently exhibited higher lifetime use rates than females across all substances. A notable sex-cohort interaction was detected: younger cohorts showed higher use rates across all substances, with a decreasing male-female gap in recent cohorts. Age of onset distributions demonstrated remarkable similarity across countries for each substance type, with medians ranging from 16-19 years for alcohol and 18-19 years for cannabis. Cocaine onset was typically later. The period of risk for initiation, however, appeared to be extending into adulthood in recent cohorts, particularly for illegal drugs. Analyses of cohorts showed higher lifetime drug use among younger cohorts. The relative-odds of first use in any given year of life was consistently higher in more recent cohorts. This pattern held across all drug types and countries, suggesting a temporal trend of increasing risk over time. Analysis of age-specific cumulative incidence of drug use among the youngest cohort (22–29 years old) revealed considerable cross-national differences, particularly for illegal drugs, with the highest rates observed in the US and New Zealand for cannabis. Sociodemographic analysis revealed consistent associations across countries and drug types. Males were more likely to use all drugs than females. Younger adults had higher rates of use compared to older adults. Higher income was positively associated with lifetime use of both legal and illegal drugs. Marital status showed a significant relationship with illegal drug use, with never-married and previously married individuals having higher odds of lifetime use than currently married individuals. The association between education and drug use showed inconsistent patterns across different drug types.

The findings demonstrate significant global variation in drug use patterns, with the US showing strikingly high rates of both legal and illegal drug use. This variation is not simply explained by drug policies, as countries with strict policies did not consistently have lower use rates than countries with more liberal policies. The observed increase in drug use among younger cohorts and the diminishing male-female gap suggest a potential societal shift in substance use trends. The extension of the period of risk for initiation into adulthood, particularly for illegal drugs, has critical implications for prevention strategies that typically focus on adolescents. The study highlights the importance of considering factors beyond drug policies when addressing substance use, such as sociodemographic factors. The robust associations between drug use and income, age, sex, and marital status underscore the complexity of the problem and the need for multi-faceted approaches to prevention and intervention.

This study provides crucial, large-scale data on global drug use patterns. The findings reveal significant regional differences, with the US showing exceptionally high use rates. Younger cohorts display higher levels of use, and the traditional sex differences are diminishing. The extension of the risk period into young adulthood necessitates a reevaluation of prevention strategies. The lack of simple correlation between drug policies and use rates underscores the complexity of the issue, demanding a multifaceted approach to combatting substance use. Future research should explore the temporal dynamics of these trends and investigate the efficacy of various drug policies in diverse contexts.

The study is limited by its focus on a specific set of 17 countries, potentially impacting the generalizability of findings. Variability in response rates across countries could introduce bias. While efforts were made to ensure honest reporting, social stigma and legal practices might still influence self-reported drug use. The retrospective nature of AOO data introduces potential recall bias. The lack of a standardized measure of urbanicity across all countries limits the analysis of the relationship between drug use and urbanization. Despite these limitations, the study’s large, internationally representative sample and standardized methods offer a valuable contribution to understanding global drug use patterns.