1. Machine learning analysis of patients' perceptions towards generic medication in Greece: a survey-based study.
- Author
-
Kassandros, Konstantinos, Saranti, Evridiki, Misailidou, Evropi, Tsiggou, Theodora-Aiketerini, Sissiou, Eleftheria, Kolios, George, Constantinides, Theodoros, and Kontogiorgis, Christos
- Subjects
PATIENTS' attitudes ,MACHINE learning ,CONVENIENCE sampling (Statistics) ,GENERIC drugs ,MEDICAL personnel - Abstract
Introduction: This survey-based study investigates Greek patients' perceptions and attitudes towards generic drugs, aiming to identify factors influencing the acceptance and market penetration of generics in Greece. Despite the acknowledged cost-saving potential of generic medication, skepticism among patients remains a barrier to their widespread adoption. Methods: Between February 2017 and June 2021, a mixed-methods approach was employed, combining descriptive statistics with advanced machine learning models (Logistic Regression, Support Vector Machine, Random Forest, Gradient Boosting, and XGBoost) to analyze responses from 2,617 adult participants. The study focused on optimizing these models through extensive hyperparameter tuning to predict patient willingness to switch to a generic medication. Results: The analysis revealed healthcare providers as the primary information source about generics for patients. Significant differences in perceptions were observed across demographic groups, with machine learning models successfully identifying key predictors for the acceptance of generic drugs, including patient knowledge and healthcare professional influence. The Random Forest model demonstrated the highest accuracy and was selected as the most suitable for this dataset. Discussion: The findings underscore the critical role of informed healthcare providers in influencing patient attitudes towards generics. Despite the study's focus on Greece, the insights have broader implications for enhancing generic drug acceptance globally. Limitations include reliance on convenience sampling and self-reported data, suggesting caution in generalizing results. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF