A new Scottish study reveals that teenagers at risk of depression are significantly more likely to experiment with cigarettes, vapes, alcohol, cannabis and other drugs by age 14 or 15, with maternal attachments playing a crucial role in mitigation.
Scottish teenagers who show signs of depression are far more likely to have experimented with cigarettes, vapes, alcohol, cannabis and other drugs by the age of 14 or 15, according to a newly published analysis of the Growing Up in Scotland cohort, which also found that relationships with mothers and fathers do not appear to shape that risk in the same way. The paper, published on 7 September in the International Journal of Mental Health and Addiction, examined six measures of lifetime use, meaning whether a young person had ever tried each substance rather than whether they used it regularly. More than half of the adolescents in the sample had consumed alcohol at least once, while use of the other substances was much less common.
That matters beyond the academic debate because Scottish policymakers have long treated family life and mental health as central to prevention. A Scottish Government evidence review has said strong bonds with parents are linked to lower rates of problematic substance use, while weak monitoring and high parent-child conflict are associated with greater alcohol and drug use. The same review notes that mental health problems are correlated with psychoactive substance use in adolescents and young adults. What the new study adds is a national, cohort-based test of those ideas in mid-adolescence, using the same family context to look across smoking, vaping, drinking, drunkenness, cannabis and other drugs.
The clearest pattern was in the maternal models. After adjusting for a wide range of social and demographic factors, the researchers found that teenagers classed as being at elevated risk of depression had odds of substance use that were between 1.86 and 4.39 times higher than those of peers not deemed at risk. Stronger attachment to mothers was associated with lower odds across every substance category, with odds ratios between 0.73 and 0.81. Higher levels of mother-youth conflict, by contrast, were linked to higher odds across the board, with odds ratios ranging from 1.19 to 1.81. Depression was screened using three items from the short-form Composite International Diagnostic Interview, while attachment and conflict were built from established survey scales and showed good internal consistency.
The paternal picture was narrower, but not negligible. In the father-based models, depression still predicted every substance outcome, with odds ratios between 1.93 and 3.96. Better father-youth attachment was associated with reduced odds for smoking, vaping, drinking, drunkenness and cannabis use, with odds ratios between 0.73 and 0.85, but it was not statistically significant for other drugs. Conflict with fathers only showed up for three outcomes: smoking, drinking and other drug use, where the odds ratio reached 1.76. In the discussion, the authors argue that fathers may now exert more day-to-day influence than older literature assumed, even if their effects are less uniform than mothers’. They point to research suggesting fathers still spend about seven hours less per week than mothers on household and parenting activity, which may help explain why the pattern differs.
The findings come from Sweep 10 of the Growing Up in Scotland birth cohort, an official longitudinal study that originally followed a nationally representative sample of 5,217 children born between June 2004 and May 2005. The sweep used here drew on data from 2,943 families collected in two phases, first between January and July 2019 and then, after a pause in March 2020, again from August 2020 as the pandemic disrupted face-to-face interviewing. The research had been circulating in academic settings well before journal publication. University of Edinburgh records show it was presented at the Edinburgh Mental Health Conference on 22 April 2024, under the university’s School of Health in Social Science, Edinburgh Neuroscience and the Child and Adolescent Mental Health Research Centre. A poster record was then logged for the APS Global Psychological Science Summit in October 2024. Programme materials for that meeting listed lead author Shuya Xie with affiliations to both the University of Edinburgh and the Centre for Student Development at Guangzhou Information Technology Vocational School.
The paper also picked up several familiar inequalities, though less consistently than the family relationship findings. Boys were more likely to report vaping and cannabis use, while lower parental education was tied to smoking and vaping. Being the only child in a household was associated with greater odds of smoking and vaping in both model sets, and with drunkenness in the maternal models. Rural residence was linked to smoking in both models and, in the maternal analysis, also to vaping and drinking. There is one notable wrinkle: the paper appears internally inconsistent on ethnicity. Its results summary says maternal “other” ethnicity predicted a greater likelihood of drinking, but the discussion section describes the same factor as protective and interprets it through the so-called migrant paradox. That ambiguity sits awkwardly alongside earlier Scottish evidence reviews, which have instead stressed family structure, conflict and supervision as the more established social risk markers.
The study does not settle why depression and substance use travel together. The authors discuss the possibility that some teenagers use nicotine, alcohol or drugs to blunt low mood, but they also note that a shared vulnerability could be at work, with common underlying risks driving both. What the data do show is that depression remained a strong marker even after extensive controls, and that the association looked especially pronounced for “other drugs”, defined in the study as drugs other than cannabis. The authors are careful about the weaknesses of the evidence. The work relied on self-report questionnaires; conflict was reported by parents rather than adolescents; the attachment measure captured security on a single scale rather than distinct styles; and the dataset did not include current or lifetime parental substance use, a potentially important missing influence. Because part of the survey moved from in-home interviews to telephone and online collection during Covid-19, the researchers also say the mode change could have affected how candidly teenagers answered sensitive questions.
Even so, the practical message is plain enough. The authors say the results strengthen the case for early identification that does not stop at the teenager alone, but also looks at the emotional climate at home and at both parents rather than defaulting to mothers. They point to the Strengthening Families Program as one example of how that could translate into practice, citing 12 randomised controlled trials that found a roughly 50 per cent reduction in substance misuse among high-risk children through lower conflict and better family relationships. For Scottish services, where official evidence has already linked supervision, disclosure and parent-child conflict to later harms, the study suggests that depression screening and family-based support may be most useful when treated as part of the same prevention problem.
Disclaimer: This content is for informational purposes only and is not intended to be a substitute for professional medical judgment, advice, diagnosis, or treatment.
