Long term outcomes of alcohol use and use disorders: a prospective twin study

The prevalence of alcohol use and alcohol use disorders (AUD) peak in young adulthood, and decline with higher age. The long-term effect of this heavy alcohol use is not sufficiently understood. Previous research has shown that alcohol use is associated with anxiety, depression, low educational attainment and other adverse social outcomes. These associations may be causal in one or both directions, or they could be due to underlying risk factors shared by alcohol use and the outcomes. Previous research in this area is limited by difficulties in distinguishing between effects of alcohol use and effects of factors leading to alcohol use. In this project, we were to investigate long-term stability and change in alcohol use, and long-term psychological and social consequences of alcohol use and use disorders. The project followed 11,727 twins born between 1967 and 1991 in governmental registries with information on health service utilization, welfare benefits, employment and demographics. The participants have responded to one or more questionnaire studies, and 2 772 of the participants have been interviewed with structured diagnostic psychiatric interviews on two occasions ten years apart. The longitudinal nature of the data made it possible to follow the participants from young adulthood until 47 years of age. Associations and variations in outcome were investigated with epidemiological methods. We further used twin methods to control for genetic and shared environmental effects of unobserved confounders, and construct directional models, bringing us a step closer to understanding causality. The first paper in the projects indicated that alcohol use disorder in young adulthood is associated with higher levels of sick leave, but that this association is probably due to a higher prevalence of anxiety and depression among individuals with alcohol use disorder, rather than due to the alcohol use in itself. We investigated how AUD is associated with depression, and how the two disorders co-develop over time. The results show that the genetic risk factors for depression are stable, whereas different genetic risk factors influence AUD in young and middle adulthood. This implies that one cannot identify one set of risk genes for AUD. We further found that the environmental effects on AUD were largely stable, but no stability in environmental effects on depression. This implies that interventions aimed at prevention or treatment of alcohol misuse among young adults can have lasting effects, whereas efforts to alleviate depression need to be more continuous. The environmental correlation between the disorders increased over time, indicating possible causal effects between the disorders. We did not find support for the hypothesis that depression in young adulthood explained the increased risk for AUD later in life. We further investigate the associations between personality disorders and AUD. The results indicated that antisocial and borderline personality disorders were most strongly associated with AUD, whereas other personality disorders did not explain AUD once these two were taken into account. The symptoms conduct disorder and self-harm had the strongest association with AUD, and the results of biometric modelling indicates that conduct disorder before age 18 is causally associated with AUD, not merely an indicator of early risk. We combined the diagnostic interviews with data from primary and specialist care recoded in Norwegian health registries in order to investigate the health service utilization among individuals with alcohol use disorder or other mental disorders. The results indicated that only 7% of individuals qualifying for a diagnosis of alcohol use disorder were recorded with that diagnosis by the health services. The corresponding numbers for depression and anxiety were 36% and 21%. We found a strong correlation between the genetic risk factors for each of the disorders, whether they were measured at interview or in health registries. This indicates that one most likely can generalized genetic findings between health registry studies and interview studies. We have investigated how anxiety disorders are related to alcohol use disorder. By using so-called direction of causation models, we have substantiated the claim that social anxiety is a cause of later alcohol use disorders. Alternatively, one could imagine that alcohol use leads to anxiety, that all kinds of anxiety disorders lead to alcohol use disorders, or that certain personality characteristics lead to all kinds of psychopathology, in the absence of direct effects between the disorders. We found no indications that other anxiety disorders or that anxiety in general had a causal influence on alcohol use disorder; the effect was specific to social anxiety disorder.

Virkninger: Økt forskningkompetanse og økt internasjonalt og nasjonalt samarbeid. Forberdring av metoder for å studere mediering og årsaksretninger. Økt forståelse av hvordan genetisk risiko for alkoholmisbruk virker gjennom livet og henger sammen med risiko for andre psykiske lidelser og med personlighetstrekk. Økt oppmerksomhet knyttet til manglende behandling av psykiske lidelser. Effekter: Selv om det kan ta lang tid fra kunnskap frembringes til den leder til endret praksis, kan vi se konturene av noen effekter: 1) Vi forstår nå i større grad forstår hva slags psykologisk problematikk som ligger til grunn for alkoholmisbruk. 2) Sosial angstlidelse er spesielt viktigere for alkoholmisbruk. Behandling av denne angstlidelsen vil være spesielt virksomt og bør prioriteres høyere. 3) Det er et betydelig avvik mellom behov for og bruk av psykiske helsetjenester. Dette betyr at man må få flere til å søke psykisk helsehjelp og at kapasiteten til å ta disse imot må utvides.

Alcohol use and alcohol use disorders peak in young adulthood, and later decline. Alcohol use is associated with anxiety and mood disorders, low educational attainment, unemployment and being unwed. These associations may be causal in one or both directio ns, or they could be due to underlying risk factors shared by alcohol use and the outcomes. Previous research in this area is limited by difficulties in distinguishing between effects of alcohol use and effects of factors leading to alcohol use. Our aims are thus to investigate long-term stability and change in alcohol use, and long-term psychological and social consequences of alcohol use and use disorders. We will establish associations, investigate how variation in outcomes is affected by personality, mental disorders, life-events, and demography, and approach possible causal mechanisms. The project uses longitudinal data from 2,284 twins born between 1967 and 1979 who responded to two questionnaires and two psychiatric interviews between 1998 and 20 11 (age 19-44). These data will be linked to registry data on treatment, social benefits, employment, education, and demography. Associations and variations in outcome will be investigated with epidemiological methods. We will further use twin methods to control for genetic and shared environmental effects of unobserved confounders, and construct directional models, bringing us a step closer to understanding causality. Knowledge about development and consequences of alcohol use is needed to design ration al interventions. This project will provide information on which social policy planning can be based. Even a small reduction in alcohol related harms can lead to improved quality of life and save substantial amounts of money. The results will also contrib ute to a broader understanding of alcohol use disorders and drinking behaviour in general. Results will be published in top ranking scientific journals, and communicated to health personnel and authorities.