# Initiation of antipsychotic medication among refugees, non-refugee migrants, second-generation migrants, and Swedish-born adults with incident non-affective psychotic disorders

## TL;DR

The study found that migrants in Sweden with non-affective psychosis were more likely to start with long-acting antipsychotics, possibly due to concerns about medication adherence.

## Contribution

This is the first study to compare antipsychotic initiation patterns across different migrant groups and Swedish-born individuals with non-affective psychosis.

## Key findings

- Refugees had lower initial antipsychotic use compared to other groups, but this difference disappeared after adjusting for sociodemographic and clinical factors.
- Migrants were more likely to start with long-acting injectable and first-generation antipsychotics compared to Swedish-born individuals.
- Younger age, higher education, and inpatient care were associated with antipsychotic initiation in both refugees and Swedish-born individuals.

## Abstract

It is not known if there are differences in antipsychotic initiation between migrants and native-born individuals diagnosed with non-affective psychotic disorder. This study aimed to determine (1) potential differences in initiation rate and type of first antipsychotic medication between refugees, non-refugee migrants, second-generation migrants, and Swedish-born young adults with incident non-affective psychosis and (2) which sociodemographic and clinical factors are associated with initiation.

This register-based cohort included 12,960 adults aged 18–35 years, residing in Sweden during 2007–2018, with an incident diagnosis of a non-affective psychotic disorder in inpatient or specialised outpatient care. Sociodemographic and clinical factors associated with antipsychotic initiation were determined using modified Poisson regression models yielding Relative Risk, RRs, and 95% Confidence Intervals, CI.

Initiation of antipsychotic use was slightly less common among refugees (65.6%) compared to non-refugee migrants (70.2%), second-generation migrants (71.0%), and Swedish-born individuals (71.1%). However, after adjustment for sociodemographic and clinical factors, there was no difference in initiation rates between refugees and Swedish-born individuals (adjusted RR 0.96; 95% CI 0.92-1.00). Olanzapine was most common initial antipsychotic in all groups. However, compared to the Swedish-born, refugees (1.47; 1.10–1.97), non-refugee migrants (1.70; 1.26–2.27) and second-generation migrants (1.43; 1.05–1.97) were more likely to initiate the use with long-acting injectable antipsychotics, and also with first-generation antipsychotics, particularly haloperidol. Sociodemographic factors associated with initiation were similar among refugees and Swedish-born individuals, including younger age, higher education and inpatient care.

Our finding that migrants were more likely to initiate long-acting antipsychotics suggests that clinical teams anticipate medication non-adherence among migrants.

The online version contains supplementary material available at 10.1007/s00127-025-02887-3.