Exploring patient medication adherence and data mining methods in clinical big data: A contemporary review

Area of Science:

• Health Informatics
• Data Science
• Machine Learning

Background:

• Patient medication adherence data are increasingly consolidated from claims databases and electronic health records (EHRs).
• Data mining, particularly machine learning (ML), is crucial for extracting actionable insights from this data.
• Nonadherence leads to increased health risks and medical costs.

Purpose of the Study:

• To review the application of ML techniques in analyzing EHR data for medication adherence.
• To explore the structure of medical databases relevant to medication adherence.
• To utilize supervised and unsupervised ML paradigms to study adherence and its consequences.

Main Methods:

• Comprehensive literature review of EHR applications in medication adherence using ML.
• Analysis of medical database structures for adherence data.
• Application of supervised and unsupervised ML techniques to adherence data.

Main Results:

• Medical databases (e.g., SEER, NHANES) and ML are effective for medication adherence analysis in clinical big data.
• ML facilitates the excavation of patient medication logs for adherence insights.
• Findings support clinical decision-making, risk stratification, and research.

Conclusions:

• Advanced data mining and ML have transformed medication adherence research, improving patient care.
• Personalized interventions and research are key to advancing therapeutic approaches.