Investigating the Effect of Adherence to Antipsychotic Therapy on the Length of Stay and Number of Hospitalizations in Patients with Schizophrenia – A Descriptive Analysis

BACKGROUND: Optimizing long-term outcomes in schizophrenia treatment requires effective pharmacological interventions. Medication adherence is known to influence clinical outcomes, yet there is a scarcity of studies examining its correlation with factors like Length of Stay (LOS) and re-hospitalization frequency. These outcomes are crucial indicators of how medication adherence affects overall patient well-being. PURPOSE: This study aims to describe the effect of medication adherence on the length of stay (LOS) and number of hospitalizations in patients with schizophrenia. PATIENTS AND METHODS: A total of 157 subjects from the West Java Psychiatric Hospital were included in this cross-sectional retrospective study. Data, including demographics, comorbidities, duration of illness, antipsychotic adherence, LOS, and the number of hospitalizations, were collected from the patients’ medical records. All the data were analyzed using the Chi-Square (χ²) test with a significance level set at p < 0.05. RESULTS: Our findings showed that 88% of all schizophrenia inpatients were in the nonadherence group. The highest (40.7%) LOS (>30 days) was found in the non-adherence group with discontinued therapy/stopped therapy group, while the highest percentage of patients with less than five hospitalizations was identified in the obedient and regular therapy group (94.4%). In the statistical results, we observed a significant association between therapy adherence (p = 0.043) and therapy regimen (p = 0.014) with gender. Additionally, the distinction between male and female schizophrenia patients demonstrated statistical significance (p = 0.000). CONCLUSION: In this study, therapy adherence and therapy regimen were found to have a significant association with gender, as well as differences between the number of male and female schizophrenia patients that were statistically significant. While other variables may exhibit clinical associations, their statistical significance has not been fully depicted. The results of this study could be preliminary study for subsequent observational studies.