TITLE:
Comparative Analysis of Monotherapy and Bi-Therapy in Antipsychotic Treatment
AUTHORS:
Rocco de Filippis, Abdullah Al Foysal
KEYWORDS:
Antipsychotics, Monotherapy, Bi-Therapy, Schizophrenia Treatment, Side Effects, Machine Learning in Psychiatry
JOURNAL NAME:
Open Access Library Journal,
Vol.11 No.11,
November
7,
2024
ABSTRACT: Background: Antipsychotic medications are crucial for managing psychiatric disorders such as schizophrenia, but their use can lead to side effects. This study compares the efficacy and side effects of monotherapy versus bi-therapy in the treatment of schizophrenia. Bi-therapy, also known as dual therapy or combination therapy, refers to the use of two medications simultaneously to treat a medical condition. Objective: This paper aims to evaluate the comparative efficacy and side effects of monotherapy (atypical antipsychotics) and bi-therapy (typical and atypical antipsychotics) over 12 months in schizophrenia patients. The objective is to compare monotherapy and bi-therapy in terms of symptom control (measured by PANSS), functional outcomes (measured by GAF), and side effects to determine which approach provides better overall treatment success in schizophrenia. Methods: A total of 100 schizophrenia patients were randomly assigned to two groups: Group A (monotherapy) and Group B (bi-therapy). The Positive and Negative Syndrome Scale (PANSS) and Global Assessment of Functioning (GAF) were used to assess symptom severity and functional outcomes at baseline and after 12 months. Side effects were also tracked. A machine learning model (Random Forest) was applied to identify key predictors of treatment success. Results: Group A (monotherapy) showed significant improvements in PANSS scores with fewer side effects. Group B (bi-therapy) showed greater symptom reduction but more pronounced side effects. Machine learning analysis identified PANSS scores at 12 months and side effects as the most important predictors of treatment success. Conclusion: Monotherapy with atypical antipsychotics offers a favorable balance of efficacy and side effects, making it a suitable option for many patients. Bi-therapy, while offering better symptom control, leads to more side effects and should be considered for treatment-resistant cases. Further studies are needed to optimize personalized treatment strategies using machine learning techniques.