Predicting changes in hypertension control using electronic health records from a chronic disease management program

Abstract

Objective: Common chronic diseases such as hypertension are costly and difficult to manage. Our ultimate goal is to use data from electronic health records to predict the risk and timing of deterioration in hypertension control. Towards this goal, this work predicts the transition points at which hypertension is brought into, as well as pushed out of, control.

Method: In a cohort of 1294 patients with hypertension enrolled in a chronic disease management program at the Vanderbilt University Medical Center, patients are modeled as an array of features derived from the clinical domain over time, which are distilled into a core set using an information gain criteria regarding their predictive performance. A model for transition point prediction was then computed using a random forest classifier.

Results: The most predictive features for transitions in hypertension control status included hypertension assessment patterns, comorbid diagnoses, procedures and medication history. The final random forest model achieved a c-statistic of 0.836 (95% CI 0.830 to 0.842) and an accuracy of 0.773 (95% CI 0.766 to 0.780).

Conclusions: This study achieved accurate prediction of transition points of hypertension control status, an important first step in the long-term goal of developing personalized hypertension management plans.

Keywords: hypertension control; predictive modeling; visualization.

Figure 1

An illustration of the predictive modeling pipeline.

Figure 2

Physician assessments of hypertension control status for a single patient over time. Blue circles are in-control assessments, red circles are out-of-control assessments. The background color bands indicate the type of episodes, where blue means positive episode and red negative episode. The transition points indicate different adjacent assessments. In this example, there are three transitions.

Figure 3

An illustration of the visualization tool developed to model temporal assessment patterns with the MyHealthTeam patient cohort.

Figure 4

Visualization of blood pressure values and clinician determination of blood pressure control status. The rectangles highlight the potential discrepancies between blood pressure values and clinician determination.

Figure 5

The c-statistic for individual feature concepts when varying the observation window. The larger observation windows tend to have better c-statistics. The performance trends of most feature concepts plateau after approximately 1 year.