Psychotic disorders are frequently preceded by depressive disorders, and it has been hypothesized that treatment of depression in youth may reduce risk for later psychosis. Using quasi-experimental methods, we estimated the causal relationship between the treatment of adolescent depression with selective serotonin reuptake inhibitors (SSRIs) and the risk of later psychosis.

We used data linkage from multiple national Finnish registries for all individuals (n = 697,289) born between 1987 and 1997 to identify depression diagnosed before age 18, cumulative SSRI treatment within three years of diagnosis, and diagnoses of non-affective psychotic disorders by end of follow-up (age 20–29). We used instrumental variable analyses, exploiting variability in prescribing across hospital districts to estimate causal effects. Analyses were conducted using two-stage least squares modelling. Sensitivity analyses examined effects stratified by confounders and effects of specific SSRIs.

Our final sample included 22,666 individuals diagnosed with depression in adolescence, of whom 60.2% (n = 13,650) had used SSRIs. 10.7% of adolescents with depression went on to be diagnosed with a non-affective psychotic disorder. SSRI treatment for adolescent depression was not associated with a reduced risk of developing a psychotic disorder (one-year β = 0.04,CI:−0.01 to 0.09; two-years β = 0.02,CI:−0.06 to 0.09; three-years β = −0.02,CI:−0.08 to 0.05).

Our quasi-experimental investigation does not support the hypothesis that treatment of adolescent depression reduces the subsequent risk of psychosis. Our findings question the assumption that treatment of common mental health disorders in youth may impact the risk of developing severe mental illnesses in adulthood.

People with severe mental illness (SMI) have an elevated risk of obesity but the causes and mechanisms are unclear. We explored the familial association between parental SMI and body mass index (BMI) in middle-aged offspring. Our objective was to determine if the offspring of either parent with SMI have an increased risk for obesity.

The Northern Finland Birth Cohort 1966 is a cohort study of offspring with expected date of birth in 1966. The data include originally 12 068 mothers and 12 231 children from the provinces of Lapland and Oulu in Finland. The final study sample included 5050 middle-aged offspring. Parental SMI was used as exposure in the study. BMI measured at the age of 46 years was used as a primary outcome.

Risk for obesity was elevated in the offspring of mothers with SMI [overweight: adjusted odds ratio (OR) 1.93 (1.29–2.90), obese class I: 1.97 (1.20–3.25), obese classes II–III: 2.98 (1.67–5.33)]. For the offspring of either parent with SMI, statistically significant results were found in obese class I and obese classes II–III [overweight: adjusted OR 1.21 (0.94–1.54), obese class I: 1.52 (1.03–1.08), obese classes II–III: 1.53 (1.01–2.32)].

We found an elevated risk of obesity in the middle-aged offspring of either parent with SMI, especially in the offspring of mothers with SMI. Thus, there might be a common familial pathway leading to the co-occurrence of obesity and SMI.