Lay Summary

We compared the inpatient and outpatient registers in identifying persons with psychotic disorders as well as characteristics and outcome between the individuals treated in inpatient and outpatient care. Compared to the outpatients, the inpatients had more often broad schizophrenia and less often affective psychosis, had younger onset age and worked less in the end of the follow-up. Due to increasing proportion of patients treated in outpatient care inclusion of both inpatients and outpatients in research is important.

Background

Representativeness of research may vary depending on sources of data. Inpatients with psychoses likely have different severity of illness compared to outpatients without hospitalization. However, nowadays increasing proportion of patients are treated in outpatient care, which emphasizes the importance of including outpatients in research data.

Methods

This study is based on the Northern Finland Birth Cohort 1966. Information on the diagnoses of psychotic disorders were collected from the inpatient and outpatient treatments of the Care Register for Health Care and various outpatient registers until year 2013. Outcome was based on register information on education, employment, pensions, and mortality until year 2018. The most severe register diagnosis was used as a primary diagnosis of the individual. We evaluated how well the diagnoses were detected in inpatient and outpatient registers and compared onset age and outcome between persons treated in inpatient care and persons treated solely in outpatient care.

Results

We identified 266 inpatients and 94 outpatients. Of all persons diagnosed with broad schizophrenia, 74% could be identified using the inpatient register and 88% using the outpatient registers (p=0.003). The corresponding figures were 59% and 74% for affective psychoses (p=0.048) and 73% and 70% for other psychoses (p=0.603). Broad schizophrenia was more often detected in both inpatient and outpatient registers than affective psychoses or other psychoses (62% vs. 33% and 42%). Compared to the outpatients, the inpatients had more often broad schizophrenia and less often affective psychosis or other psychosis (corresponding proportions 59%, 20%, and 43% for the inpatients and 30%, 43% and 28% for the outpatients; p<0.001), and younger mean age of illness onset (31.7 vs. 39.6 years; p<0.001). During 2018, 17% of the inpatients and 29% of the outpatients worked at least 70% of time (p=0.028). Education or socioeconomic status did not differ. Six inpatients and three outpatients had died during the 5-year follow-up.

Conclusion

Inclusion of both inpatients and outpatients enables representative samples in research on psychosis. Inclusion of outpatients has become more important since psychiatric care has shifted towards outpatient care.