Screening for Depression in Daily Life: Development and External Validation of a Prediction Model Based on Actigraphy and Experience Sampling Method (Preprint)

Abstract

BACKGROUND In many countries, depressed individuals often first visit primary care settings for consultation, but a considerable number of clinically depressed patients remains unidentified. Introducing additional screening tools may facilitate the diagnostic process. We aimed to examine whether Experience Sampling Method (ESM)-based measures of depressive affect and behaviors can discriminate depressed from non-depressed individuals. In addition, the added value of actigraphy-based measures was examined. OBJECTIVE The aim of our study was to examine whether ESM-assessed depression-related affect and behavior can discriminate between depressed and non-depressed individuals, whether actigraphy data can discriminate between depressed and non-depressed individuals on its own, and whether it has added value with respect to the use of ESM. METHODS We used data from two samples to develop and validate prediction models. The development dataset included 14 days of ESM and continuous actigraphy of currently depressed (n=43) and non-depressed individuals (n=82). The validation dataset included 30 days of ESM and continuous actigraphy of currently depressed (n=27) and non-depressed individuals (n=27). Backward stepwise logistic regression analyses were applied to build the prediction models. Performance of the models was assessed with goodness of fit indices, calibration curves, and discriminative ability (AUC - the area under the receiver operating characteristic curve). RESULTS In the development dataset, the discriminative ability was good for the actigraphy model (AUC=.790) and excellent for the ESM (AUC=.991) and combined-domains model (AUC=.993). In the validation dataset, the discriminative ability was reasonable for the actigraphy model (AUC=.648) and excellent for the ESM (AUC=.891) and combined-domains model (AUC=.892). CONCLUSIONS ESM was a good diagnostic predictor and is easy to calculate, and it therefore holds promise for implementation in clinical practice. Actigraphy showed no added value to ESM as a diagnostic predictor, but might still be useful when ESM use is restricted.