Using machine learning to develop a five-item short form of the Children’s Depression Inventory

Background There are many adolescents experience depression that is not detected and treated in time. The Children’s Depression Inventory (CDI) is widely used in China, but no localized revision or simplified form has been formulated. Consequently, the present study aimed to formulate an abbreviated version of the CDI with only five items, using a large sample of 20,675 Chinese children aged 7 to 15 years in Sichuan Province. Methods First, different versions of the short-form scales were identified by backward elimination. Then, the area under the ROC curve (AUC) of five machine learning (ML) algorithms on the short-form scales were compared. Finally, the prediction performance of each short-form scale was evaluated with the metric of the naïve Bayes (NB). Results The study identified a five-item short-form CDI with a judgment threshold of 4 as the most appropriate scale considering all assessment indicators. The scale had 81.48% fewer items than the original version, indicating good predictive performance (AUC = 0.81, Accuracy = 0.83, Recall = 0.76, Precision = 0.71). Based on the test of 315 middle school students, the results showed that the five-item CDI had good measurement indexes (Cronbach’s alpha = 0.72, criterion-related validity = 0.77). Conclusions This five-item short-form CDI is the first shortened and revised version of the CDI in China based on large local data samples.