BACKGROUND: Psychotic disorders are known to exhibit heterogeneity with regards to illness course and prognosis, yet few studies have examined long-term healthcare use.

OBJECTIVE: To determine the prevalence and predictors of healthcare use for psychiatric disorders at 9 years after the first episode of psychosis (FEP).

METHODS: National registers were used to identify all Swedish residents aged 18-35 years with FEP between 2006 and 2013. The 12-month period-prevalence of secondary healthcare use was determined at each year of the 9-year follow-up, categorised according to main diagnosis (psychotic disorder vs other psychiatric disorder vs none vs censored). Multinomial logistic regression models were used to examine associations between baseline characteristics and healthcare use at 9 years and derive predicted probabilities and 95% CIs for the four outcome groups, for each predictor variable.

FINDINGS: Among 7733 individuals with FEP, 31.7% were treated in secondary healthcare for psychotic disorders at the 9-year follow-up, 24.1% were treated for other psychiatric disorders, 35.7% did not use healthcare services for psychiatric disorders and 8.5% were censored due to death/emigration. Having an initial diagnosis of schizophrenia was associated with the highest probability of secondary healthcare use for psychotic disorder at 9 years (0.50, 95% CI (0.46 to 0.54)] followed by inpatient treatment at first diagnosis (0.37, 95% CI (0.35 to 0.38)).

CONCLUSION: Although 56% of individuals with FEP were treated for psychiatric disorders in secondary healthcare 9 years later, a substantial proportion were treated for non-psychotic disorders.

CLINICAL IMPLICATIONS: Individuals with an initial diagnosis of schizophrenia, who received their first diagnosis in inpatient settings, may need more intensive treatment to facilitate remission and recovery.