This longitudinal population study used administrative data to explore uptake of psychiatric medication and hospitalizations for mental or behavioural disorders in several Swedish samples. The main sample of over 20,000 PhD students consisted of those starting a PhD program in Sweden between 2006 and 2017 and who had administrative data available the year before starting their PhD studies. Two matched control groups were also constructed: one including over 7 million individuals from the general population who were alive in 2006 and never enrolled in PhD studies in Sweden; and one including over 305,000 individuals from the highly educated population who additionally had a Swedish master's degree. These were matched to the PhD population by gender, birth year, and—for the highly educated control group—field and year of the master’s degree. All participants were identified using administrative data from Statistics Sweden and the National Board of Health and Welfare.