Clarissa Janousch

The co-occurrence of substance use disorders (SUDs) and other psychiatric disorders (oPDs) is a major global public health and policy concern in adolescents and young adults. Past research has commonly examined single disorder pairs, with comorbidities considered discrete events. Yet, new models of psychopathology stress the need for dimensional and longitudinal perspectives that account for complex, changing patterns of mental health issues. Despite these developments, several key research gaps remain. First, few studies have investigated intricate multi-disorder comorbidity patterns (i.e., multimorbidity) beyond simple disorder pairs. Second, longitudinal studies with closely spaced measures spanning key developmental periods, such as adolescence to adulthood, are scarce, limiting insights into onset, remission, and relapse. Third, most research relies on self-reported data, which can introduce bias, and few studies have utilized large-scale, registry-based data to track comorbidity at the population level. Finally, there is a lack of research on how sociodemographic and contextual factors (e.g., sex, age, healthcare utilization) influence these patterns and their transitions over time.This project will address these gaps by examining the heterogeneity, development, and transition of co-occurring SUDs and oPDs over time. Using a large register-based cohort in Sweden (~3 million individuals born 1970-2000), the project will pursue three overall aims: (1) Identify and characterize distinct patterns of multimorbidity, (2) investigate developmental trajectories including onset, remission, and relapse across the lifespan, and (3) examine intra-individual transitions between different multimorbidity profiles. For all investigations, all mentioned factors related to these patterns and transitions will be examined. This project will substantially enhance our knowledge of multimorbidity in SUDs and oPDs and the underlying mechanisms within the general population. It will contribute to identifying population subgroups at increased risk and clarifying how multimorbidity develops across different phases of the lifespan. Ultimately, the results will help guide targeted prevention efforts and support the development of evidence-based public health policies.

Substance use is highly prevalent among Swiss adolescents and young adults, with potential long-term consequences for cognitive well-being. Substances such as alcohol, tobacco, and cannabis have been proposed to be particularly neurotoxic during the highly neuroplastic and sensitive period of adolescent brain maturation. Indeed, several cross-sectional and a few longitudinal studies suggest that an early age of onset and frequent use of these substances during adolescence is associated with impaired cognitive performance and less educational success in later life. These previous studies came with notable limitations, however. First, cross-sectional studies typically used small and highly selective samples, assessed substance use by self-report only, and could, by design, not address the directionality of effects. Second, the few longitudinal studies tended to be underpowered, relied on self-reported substance use only, started their substance use assessments past the age of typical initiation, or were limited to one or few adolescent substance use assessments. Most of these studies also had a low prevalence of adolescent substance use, limiting statistical power to detect associations with later-life cognition. Third, sex-differences were neglected in nearly all studies, although animal studies suggest differences in developmental vulnerability between sexes. Finally, previous studies did not adequately address polysubstance use and concurrent use or chronic consumption of specific substance combinations, which are, however, normative patterns of use in adolescents today.We propose to address these limitations by leveraging a unique, large-scale cohort study with 9 main assessments from ages 7 to 24 to investigate the developmental neurotoxicity of early substance use on young adult cognitive functions. The Zurich Project on the Social Development from Childhood to Adulthood (z-proso, N=1,675, 52% males), funded by the Swiss National Science Foundation (SNSF, No. 198052), has investigated the social, educational, and psychological development of participants at ages 7, 8, 9, 11, 13, 15, 17, 20, and 24. Importantly, up to 6 self-reports of substance use (prevalence and frequency) per person are available between ages 11 and 24 in z-proso. At age 20 (in 2018) and 24 (in 2022), the study also collected hair samples. Funded by the SNSF (No. 189008), the age 20 hair samples were already assayed for ~100 illegal substances and medications and their main metabolites by state-of-the-art toxicological analysis. Part of the here proposed work is to additionally assay the already collected age 24 hair samples.The fine-grained z-proso substance use questionnaire revealed a high burden from substance use. For example, self-reported 12-month prevalence at age 20 was: cannabis use=56%, cocaine=11%, Ecstasy=12%, codeine=13%, and use of at least 2 illegal substances=25%. The age 20 hair tests suggested that 3-month self-reports underestimated the prevalence of most substances by 30–60%, demonstrating the urgent need of objective measurement of substance use. Importantly, at age 24 (data collection was finished in August 2022), three well-established tests of the Core Battery of the Cambridge Neuropsychological Test Automated Battery (CANTAB) were also administered, assessing four core domains of young adults’ cognition: attention, working memory, visuo-spatial declarative memory, and executive functions.Thus, the richly characterized unique z-proso data set will allow us to predict young adults’ cognitive performance across core domains, using both the self-reported and objectively quantified measures of adolescent and young adult substance use. It will also allow us to investigate potential sex difference in the developmental neurotoxicity of the substances. Importantly, at all of its 9 main assessments, z-proso extensively characterized sociodemographics, child characteristics (e.g., symptoms of anxiety, depression, pro- and antisocial behaviors, ADHD, and later also psychosis), earlier indicators of cognitive functions (e.g., school success), and other potential confounders/covariates. The longitudinal data also allow for an examination of the directionality of effects and to apply propensity score modeling techniques that are suitable for causal analysis of observational studies. Our research aims include to test:1) cross-sectional associations between substance use and cognition at age 24,2) longitudinal associations between objectively quantified substance use in hair (age 20 and 24) and cognition at age 24,3) associations of different substance use trajectories between ages 11 and 20 and cognition at age 24, 4) cross-sectional and longitudinal associations of cognition and real-life functioning at age 24, and5) the prediction of cognition from adolescent and young adult substance use with the use of artificial intelligence.New findings revealed that legalization of cannabis in the US has driven young adults’ rates of cannabis use to an all-time high (1). In this era, it is crucial to understand the potential cognitive costs of early substance use. The present study has outstanding potential to overcome limitations of previous work, and to generate important new knowledge on the impact of early life substance use on cognition. Its insights can inform ongoing debates of legalizing cannabis and other substances, including in Switzerland, which has high rates of substance use among young people, and where a legalized cannabis market is expected within the next few years. Of note, we aim not only to investigate the effects of alcohol, tobacco, and cannabis use but also of substances such as Ecstasy, cocaine, prescription opioids, and polysubstance use, which have been scarcely investigated in longitudinal study designs and which were used at high prevalence rates at age 20 in the z-proso cohort.The proposed project leverages a rich, long-term-longitudinal existing data collection in a cost-effective way to examine important questions about the role of substance use in young adult cognition at a unique scale. Results of the proposed project have the potential to provide substantial and significant new insights into the study of the developmental neurotoxicity of adolescent substance use on cognition. Thus, we expect that our findings will inform addiction research, prevention, and policymaking in important new ways. Our proposal requests funding for one PhD student, one postdoc, consultant costs for an international expert on causal modeling, and assaying the age 24 hair data to analyze the above research questions with the exceptional z-proso data set.