Risk of readmission among individuals with cannabis use disorder during a 15-year cohort study: the impact of socio-economic factors and psychiatric comorbidity.

BACKGROUND AND AIM: Cannabis use disorder (CUD) is one of the main reasons for seeking substance treatment in the Nordic countries, but there are few studies on readmission to care. We aimed to characterize CUD readmission and estimate the magnitude of how socio-economic factors and psychiatric comorbidity influence the risk of CUD readmission. DESIGN, SETTING AND PARTICIPANTS: This was a nation-wide cohort study carried out between 2001 and 2016 in Sweden. The participants were individuals with CUD, aged 17 years and above (n = 12 143). MEASUREMENTS: Information on predictors was obtained from registers and included education, income and psychiatric comorbidity assessed by six disease groups. The outcome measure was readmission, defined as a CUD visit to health-care at least 6 months after initial CUD diagnosis. Hazard ratios (HR) were estimated using Cox survival analyses and flexible parametric survival analyses to assess risk of readmission and how the risk varied with age. FINDINGS: The vast majority of CUD visits took place in outpatient care (~80%). Approximately 23% of the included individuals were readmitted to care during follow-up. The fully adjusted model showed an increased risk of readmission among those with schizophrenia and other psychotic disorders [HR = 1.54, 95% confidence interval (CI) = 1.29-1.84], low education (HR = 1.40, 95% CI = 1.24-1.57), personality disorders (HR = 1.27, 95% CI = 1.05-1.54) or mood disorders (HR = 1.27, 95% CI = 1.12-1.45). Flexible parametric modeling revealed increased risk of readmission mainly in individuals aged 18-35 years. CONCLUSIONS: The risk of readmission was highest among those with low education, schizophrenia and other psychotic disorders, mood-related disorders or personality disorders. Individuals aged 18-35 years showed the highest risk of readmission. Our findings highlight individuals with complex health-care needs.

Cannabis use disorder in relation to socioeconomic factors and psychiatric comorbidity: A cluster analysis of three million individuals born in 1970-2000.

BACKGROUND: Cannabis use disorder (CUD) is one of the main reasons for seeking substance use treatment. It is thus important to monitor and increase knowledge of individuals with CUD utilizing healthcare. We aimed to examine the number of CUD diagnoses over time, compare individuals with CUD with those without and identify subgroups based on CUD diagnosis, sex, birth year, socioeconomic factors and psychiatric comorbidity. METHODS: A Swedish, population-based study with 3,307,759 individuals, born in 1970-2000, with register data extending to 2016. K-mode cluster analysis was used to identify potential subgroups. RESULTS: The number of individuals with a CUD diagnosis was 14,046 (0.42%). CUD diagnoses increased over time (born 1990-1994: 61 per 100,000, born 1995-2000: 107 per 100,000, by 2016). A majority of those with a CUD had another psychiatric diagnosis (80%, compared with 19% for those without CUD). Four clusters were identified. Cluster 1 comprised mainly men with low income and substance use disorders, clusters 2, 3 and 4 comprised mainly women with higher proportions of mood-related, neurotic and stress-related and behavioural disorders. CONCLUSIONS: There was an increase in CUD diagnoses in Sweden over time, especially among younger birth cohorts. Individuals with CUD were more often male, from younger birth cohorts, with lower education and income than those without CUD. Men and women with CUD exhibited differences in education, income and psychiatric comorbidity. Our results demonstrate the importance of monitoring the impact of socioeconomic factors and psychiatric comorbidity in relation to CUD.

Cannabis use and the risk of anxiety and depression in women: A comparison of three Swedish cohorts.

BACKGROUND: The associations between cannabis use and anxiety or depression remain unclear. If cannabis affects these conditions, it is of interest to examine possible changes in cannabis use over time, in relation to anxiety and depression, as cannabis potency has increased in recent decades. METHODS: Cohorts from the Women and Alcohol in Gothenburg study (n = 1 100), from three time periods were used to examine associations over time between cannabis use and anxiety and depression. Logistic regression analyses were used and relative excess risk due to interaction (RERI) was calculated to examine potential additive interactions between period of cannabis use, cannabis use, and anxiety or depression. FINDINGS: Cannabis use was associated with anxiety in the oldest cohort (examined 1986-1992, born 1955/65), OR = 5.14 (1.67-15.80, 95 % CI), and with both anxiety and depression in the youngest cohort (examined 2000-2015, born 1980/93), OR = 1.66 (1.00-2.74, 95 % CI) and 2.37 (1.45-3.86, 95 % CI), respectively. RERI was significant between cannabis use and depression in the youngest cohort when compared with older cohorts (1.68 (0.45-2.92, 95 % CI)). LIMITATIONS: Cross-sectional data prevent statements on causality, though between-cohort comparisons are possible. CONCLUSION: The association between cannabis use and depression becomes more pronounced when adding the effect of period of use rather than looking at the effect of cannabis use itself. This study provides clinicians and public health workers with scientifically underpinned knowledge regarding the link between cannabis use and depression, of particular importance given increasing cannabis potency.

Cannabis use, subsequent other illicit drug use and drug use disorders: A 16-year follow-up study among Swedish adults.

AIMS: To examine the association between cannabis use and subsequent other illicit drug use and drug use disorders (harmful use and dependence). DESIGN, SETTING, PARTICIPANTS: We used survey data from a population-based cohort in Stockholm County (collected 1998-2000), with linkage to the National Patient Register. The study base comprised participants aged 20-64 years (N = 10 345), followed-up until 2014. Cox and logistic regression analyses were conducted to test associations between self-reported cannabis use and risk of subsequent other illicit drug use (three-year follow-up) and drug use disorders (16-year follow-up). FINDINGS: The odds ratio (OR) for other illicit drug use onset at three-year follow-up for lifetime cannabis users was 7.00 (4.47-10.35, 95% CI) and for recent cannabis users 34.41 (19.14-61.88, 95% CI). Adjusting for age and AUDIT score attenuated the association, for lifetime users: OR = 5.48 (3.69-8.13, 95% CI) and OR = 5.65 (3.80-8.41, 95% CI), and for recent users: OR = 18.32 (9.88-33.99, 95% CI) and OR = 20.88 (11.19-38.95, 95% CI). For cannabis users only, the hazard ratio (HR) for drug use disorders at 16-year follow-up was 0.89 (0.31-2.61, 95%CI). For cannabis and other illicit drug users, the corresponding HR was 7.27 (3.85-13.75, 95% CI). CONCLUSIONS: There was no independent association between cannabis use and subsequent drug use disorders. The association with subsequent drug use disorders was rather explained by other illicit drug use, which cannabis users were at higher risk of at the three-year follow-up.

Hälsoutvecklingen i Stockholm bättre än i övriga landet - Jämförelse av hälsoläget utifrån globala sjukdomsbördeprojektet.

Previous studies in Sweden have focused on a number of indicators to assess and compare health conditions at regional levels over time. In this study we aimed to give a more complete picture of the health situation in Stockholm County compared to the rest of Sweden, by using the DALY measure (disability-adjusted life years). DALY combines life lost to premature death (YLL) and years lived with disability (YLD) in one measure, and also allow comparisons of fatal and non-fatal conditions. This approach reveals that low back and neck pain and ischemic heart disease dominated the disease burden in 2015. Moreover, the health progress in Stockholm County has been better than the rest of Sweden since 1990, and the main reason is the decrease in premature death (YLL). This can partly be explained by a decrease in risk factors such as unhealthy diets, high blood pressure, tobacco smoking, high BMI and physical inactivity. The development of YLD has been relatively constant since 1990 in both Stockholm County and the rest of Sweden, implying that Sweden has been more successful in preventing death than reducing disability.

Alcohol-attributed disease burden and alcohol policies in the BRICS-countries during the years 1990-2013.

BACKGROUND: We aimed to assess alcohol consumption and alcohol-attributed disease burden by DALYs (disability adjusted life years) in the BRICS countries (Brazil, Russia, India, China and South Africa) between 1990 and 2013, and explore to what extent these countries have implemented evidence-based alcohol policies during the same time period. METHODS: A comparative risk assessment approach and literature review, within a setting of the BRICS countries. Participants were the total populations (males and females combined) of each country. Levels of alcohol consumption, age-standardized alcohol-attributable DALYs per 100 000 and alcohol policy documents were measured. RESULTS: The alcohol-attributed disease burden mirrors level of consumption in Brazil, Russia and India, to some extent in China, but not in South Africa. Between the years 1990-2013 DALYs per 100 000 decreased in Brazil (from 2124 to 1902), China (from 1719 to 1250) and South Africa (from 2926 to 2662). An increase was observed in Russia (from 4015 to 4719) and India (from 1574 to 1722). Policies were implemented in all of the BRICS countries and the most common were tax increases, drink-driving measures and restrictions on advertisement. CONCLUSIONS: There was an overall decrease in alcohol-related DALYs in Brazil, China and South Africa, while an overall increase was observed in Russia and India. Most notably is the change in DALYs in Russia, where a distinct increase from 1990-2005 was followed by a steady decrease from 2005-2013. Even if assessment of causality cannot be done, policy changes were generally followed by changes in alcohol-attributed disease burden. This highlights the importance of more detailed research on this topic.

Low cardiorespiratory fitness in young adulthood and future risk of disability pension: a follow-up study until 59†years of age in Swedish men.

BACKGROUND: There is a strong belief in the potential of increased physical activity to improve the health of populations. The objective of the present study was to estimate the association between low cardiorespiratory fitness in young adulthood and subsequent health impairment until middle age, measured by disability pension. METHODS: The study utilised data on cardiorespiratory fitness and a number of covariates (social background, health behaviours and psychological characteristics) from the Swedish Conscription Cohort of 1969/1970 including 49,321 men born in 1949-1951, and data on disability pension from 1971 to 2008 (20-59 years of age) through record linkage with two national insurance databases. Cox proportional-hazards regressions yielded hazard ratios (HR) with 95% confidence intervals (CI). RESULTS: Having low-level or mid-level cardiorespiratory fitness in late adolescence was associated with an increased HR of disability pension across the follow-up (HR for low fitness: 1.85, CI 1.71 to 2.00; HR for mid-level fitness: 1.40, CI 1.31 to 1.50). The association was stronger with earlier disability pensions than with later disability pensions, which was also seen after multiple confounding adjustments. At the same time, these adjustments revealed considerable confounding of the association by individual differences in psychological characteristics in particular. CONCLUSIONS: Lower levels of cardiorespiratory fitness in young adulthood were found to be associated with an increased HR of disability pension throughout the follow-up until 59 years of age, even after adjustment for important confounding factors measured in late adolescence. Increased physical fitness may thus have a lowering effect on the risk of disability pension.