JMIR Publications

ABSTRACT

Background:

Clinical guideline recommends screening for depression in the general adult population but recognizes that optimum interval for screening is unknown. Ideal screening intervals should match with patient risk profiles.

Objective:

This study describes an approach that mined patient-reported data from a large clinical trial with predictive modeling to improve case identification for primary care patients at high risk for depression to match depression screening interval with patient risk profile.

Methods:

This paper analyzed data from a large safety-net primary care study for diabetes and depression. A regression-based data mining technique was used to examine 53 demographics, clinical, and patient-reported variables to develop three prediction models for major depression at 6, 12, and 18 months from baseline. Predictors among the strongest predictive power that also require low information collection efforts were selected to develop the prediction models. Predictive accuracy was measured by the area under the receiver operating curve (AUROC) and was evaluated by 10-fold cross-validation. The effectiveness of the prediction algorithms in supporting clinical decision making for five “typical” types of patients was demonstrated.

Results:

The analysis included 923 patients. Five patient-reported variables were selected in the prediction models: 1) Patient Health Questionnaire 2-item score; 2) the Sheehan Disability Scale; 3) previous problems with depression; 4) the diabetes symptoms scale; and 5) diabetes emotional burden. All three depression prediction models have AUROC>0.80, comparable with published depression prediction studies. Among the 5 “typical” types of patients, the algorithms suggest patients who reported impaired daily functioning by health status are at an elevated risk for depression in all three periods.

Conclusions:

This study demonstrated leveraging patient-reported data and prediction models can help improve case identification and clinical decision about depression screening interval for diabetes patients. Implementation of this approach can be coupled with application of modern technologies such as tele- and mobile-health assessment for collecting patient-reported data to improve privacy, reduce stigma and costs, and promote a personalized depression screening that matches screening interval with patient risk profiles.