Optimal Doses of Specific Antipsychotics for Relapse Prevention in a Nationwide Cohort of Patients With Schizophrenia

Heidi Taipale; Antti Tanskanen; Jurjen J. Luykx; Marco Solmi; Stefan Leucht; Christoph U. Correll; Jari Tiihonen


Schizophr Bull. 2022;48(4):774-784. 

In This Article


The cohort was identified from nationwide Hospital Discharge register and included all persons hospitalized due to schizophrenia during 1972–2014 in Finland and who were alive at January 1, 1996 (N = 61 889). Schizophrenia was defined as International Classification of Diseases (ICD) version 10 codes F20 and F25 (corresponding ICD-9 and ICD-8 code 295*). Data were linked via personal identification numbers on all inpatient stays (dates and diagnoses 1972, from Hospital Discharge register), dispensed medications (from Prescription register 1995–2017), and deaths (causes of death register 1972–2017). Observation time for medication use started on January 1, 1995 when the Prescription Register was established. We identified the first-episode schizophrenia cohort as persons who had their first diagnosis of schizophrenia during 1996–2014, without a previous diagnosis and who had not used antipsychotics in the one year before their first diagnosis (to ensure they are incident cases). Concerning patients with first-episode schizophrenia, the follow-up started on January 1, 1996 for those diagnosed before that date, and on the date of diagnosis for those diagnosed later, and ended at death or December 31, 2017. Patients who died between 1972 and 1996 were not included, as we could not follow their medication use. Relapse was defined as rehospitalization based on the main diagnosis of F20–F29. Relapse was treated as a recurrent event, ie possible to happen multiple times for the same person.

Antipsychotic use was defined as Anatomical Therapeutic Chemical (ATC) classification code N05A excluding lithium. Antipsychotic dispensings were derived from the Prescription register, and included ATC code, date of dispensing, product information (name of drug product, strength, package size), the number of packages dispensed, and dispensed amount in defined daily doses (DDDs) as defined by the WHO (WHO DDD).[9] Prescribed doses are not available in the Prescription register. Episodes when antipsychotic use started and ended, were modeled with the PRE2DUP method.[10] The method is based on the mathematical modeling of personal drug purchasing behavior from the dispensing data. By utilizing information on dispensing dates and amounts in DDDs, together with drug package level information and expert-opinion derived control parameters, the method constructs antipsychotic use episodes with start and end dates. Irregularities in dispensing events, caused by eg stockpiling and time periods spent in hospital care when drugs are provided by the unit treating and not recorded in the Prescription register, are taken into account.[10,11] Each ATC code and by separating oral and long-acting injectable antipsychotic (LAI) formulations were modeled separately.

After formation of antipsychotic use episodes, we calculated temporal dose estimates at each dispensing date as the sum of the dispensed DDD amount of the two previous dispensings divided by the outpatient time of these dispensings. This temporal dose estimate was expressed as DDDs/day dose. The dose estimate was valid until the next dispensing when the dose was reevaluated (ie dose changes were possible only at dispensing dates). Antipsychotic episodes including three or more dispensings were processed by considering dispensings in a stepwise manner, starting from the second dispensing and proceeding till the last dispensing. For antipsychotic episodes including only one or two dispensings, dose estimates were calculated as dispensed DDDs divided by outpatient time (ie, no dose changes were possible within these short episodes).

Antipsychotic episodes were subdivided into time periods when a certain dose (as DDDs/day) was consistently used in the following categories: <0.6, 0.6–<0.9, 0.9–<1.1, 1.1–<1.4, 1.4–<1.6, ≥1.6. Categories were formed around most frequent clinically relevant doses (0.5, 1.0, and 1.5 DDDs/day) with small variation and inaccuracy allowed (ie 0.9–<1.1 describes 1.0 DDDs/day use). We focused on antipsychotic monotherapy only and excluded time periods when more than one antipsychotic was used at the same time (ie antipsychotic polytherapy). A person may contribute person-time to multiple or even all dose categories of a drug or of multiple antipsychotic drugs used in monotherapy during the follow-up (Supplementary Figure 1).