BackgroundIn studies investigating the relapse rate of psychotic symptoms in patients diagnosed with schizophrenia there is a discrepancy between discontinuation studies finding a relapse rate up to 90% after discontinuation of antipsychotic medication and long-term follow-up studies finding approx. 30% of patients living without antipsychotic medication and psychotic symptoms. Long-term follow-up studies often have multiple follow-up assessments, but little is known about the use of medication in the intervals between the follow-up points. While register studies can follow large cohorts of patients, they are unable to investigate psychopathology and level of functioning in patients who discontinue their medication. In this study we use data from a clinical cohort with information on participants symptoms and functioning and combine them with register data on the individual participants prescriptions and hospitalizations.MethodsThe present study represents a combination of a clinical study from early intervention settings and register-based information on antipsychotic drug use and hospital contacts. For the present study, patients were included 18 months into their 24 months early intervention treatment and followed up 3 ½ year later. At baseline and follow-up we performed clinical assessments with all patients and via the Danish National Hospital Register and the Danish National Prescription Register, we had complete nationwide information for all patients identifying all redeemed prescriptions for antipsychotic drugs from 6 to 42 months after inclusion into the study. Based on medication information from the Danish National Prescription Register, we divided participants in the following four groups: 1) Non-users, 2) compliant on medication, 3) stopped but resumed later with medication, and 4) stopped with medication.ResultsOf the 316 participants included in this study 94.3% had I diagnosis of schizophrenia. In the 3 years preceding the 5 years follow-up 28.2% did not redeem any prescriptions for antipsychotics drugs while 21.2% discontinued their treatment during the follow-up, 20.9% discontinued their treatment but resumed later and 29.7% remained in stable treatment. At the 5 years follow-up the 30.3% of the Never-users where in competitive employment, the mean psychotic symptom score were 1.4 SD (1.4) and negative symptoms 1.1 SD (0.9). Whiles these results were worse for patients Compliant on medication (17%, 1.9 SD (1.3), 1.8 SD (1.0)), Stopped but resumed medication (10.6%, 22.4 SD (1.4), 1.5 SD (1.0)) and Stopped medication (17%, 1.6 SD (1.3), 1.3 SD (1.0)), respectively. Of the Never-user 23.6% were in remission of both positive and negative symptoms, while this was only the case for 12.8% of those compliant on medication.DiscussionThis study is a naturalistic cohort study and we are unable to draw any conclusion regarding the causality between symptoms remission and use of antipsychotic medication. The study shows that a substantial proportion of patients, for several years, can discontinue their medical treatment without being re-hospitalized and with lower symptoms burden then patients who continue their medical treatment. Some patients discontinue their treatment but resume it later. These patients have approximately the same functional level and psychotypological scores as those who are compliant with their medical treatment and are treated with equivalent doses of antipsychotic at the time of the follow-up.