Preliminary validation of a single self-report question as a screening tool for depression in older adult populations: Analyses using the Minimum Data Set Depression Rating Scale

Abstract

Objectives: The primary objective of this research was to inform the criterion validity of the single self-report depression screening question in the interRAI-Contact Assessment (CA) against the validated Depression Rating Scale (DRS) in the RAI-Home Care (HC) instrument. To achieve this objective, two overarching research questions were addressed: 1) What is the DRS cut-point best suited for the detection of a possible depressed mood state?; and 2) What are the consistencies in the prevalence rates of the CA’s single self-report depression question and the HC’s DRS detection measure? A secondary objective of this research was to explore the relationships of multiple depression-related outcomes with selected possible predictor variables to indirectly aid in the identification of a possible depressed mood state. Methods: Four datasets were obtained from the interRAI organization to inform this research. Three datasets, which house and concurrently administer both the self-report item and the DRS, were used to determine the best-suited DRS detection threshold through both univariate and bivariate analyses. The validity of the CA’s self-report item was informed through bivariate analyses with the HC’s DRS measure using the phi-correlation and the c statistic from the bivariate logistic regression model. Spearman and point biserial correlations and bivariate logistic regression modeling informed the relationships of the possible predictor variables with the depression-related outcomes. Results: The DRS one-plus threshold was determined to be the cut-point best-suited for the detection of a possible depressed mood state. Several predictor variables proved statistically significant but were not consistent across the three datasets. The CA’s single question did not evidence a strong association with the HC’s DRS measure. Conclusions: According to the resulting significance of the predictor variables, the experience and expression of depression across the three explored samples differ; the results from one sample cannot be generalized to explain the experience of the other samples. The DRS one-plus threshold is supported for the detection of a possible depressed mood state. However, the validity of the single self-report question in comparison to the DRS measure cannot be supported with the study’s results. However, due to recognized study limitations, the overall validity of the measure is not conclusive. Future research directions are recommended.