JMIR Publications

ABSTRACT

Background:

Medication non-adherence can compound into severe medical problems for patients. Identifying patients that are likely to become non-adherent may help reduce these problems. Data-driven machine learning models can predict medication adherence by using selected indicators from patients’ past health records. Sources of data for these models traditionally fall under two main categories: a) proprietary data from insurance claims, pharmacy prescriptions, or electronic medical records and b) survey data collected from representative groups of patients. Models developed using these data sources often are limited because they are: proprietary, subject to high cost, have limited scalability, or lack timely accessibility. These limitations suggest health social forums might be an alternate source of data for adherence prediction. Indeed, this data is accessible, affordable, timely and available at scale. However, it can also be inaccurate.

Objective:

This paper proposes a medication adherence machine learning model for fibromyalgia therapies that can mitigate the inaccuracy of health social forum data.

Methods:

Transfer learning is a machine learning technique that allows knowledge acquired from one dataset to be transferred to another dataset. In this study, predictive adherence models for the target disease were first developed by using accurate but limited survey data. These models were then used to predict medication adherence from health social forum data. Random forest (RF), an ensemble machine learning technique, was used to develop the predictive models. This transfer learning methodology is demonstrated here by examining data from the Medical Expenditure Panel Survey (MEPS) and the PatientsLikeMe health social forum.

Results:

When the models are carefully designed, less than a 5% difference in accuracy is observed between the MEPS and the PatientsLikeMe medication adherence predictions for fibromyalgia treatments. This design must take into consideration the mapping between the predictors and the outcomes in the two datasets.

Conclusions:

This study exemplifies the potential and the limitations of transfer learning in medication adherence predictive models based on survey data and health social forum data. The proposed approach can make timely medication adherence monitoring cost-effective and widely accessible. Additional investigation is needed to improve the robustness of the approach and to extend its applicability to other therapies and other sources of data.