Increasing Utilization of Standardized Tools for Measurement-Based Care in the Management of Generalized Anxiety Disorder

Generalized anxiety disorder (GAD) is a common mental health condition encountered in primary care settings. GAD screening, diagnosis, and management are challenging, among other issues that capture the attention of primary care physicians (PCPs). Measurement-based care (MBC) involves the systematic assessment of patients’ symptoms and treatment progress using standardized tools. Generalized Anxiety Disorder 7-Item (GAD-7) is a well-known screening and symptom-monitoring tool for GAD. It quantifies subjective symptoms objectively by measuring the patient’s anxiety level. We aimed to increase utilization of GAD-7 in outpatient clinics to improve diagnosis and management of GAD through educational interventions and by educating PCPs to access the GAD-7 tool in the EPIC electronic medical record (EMR) with ease.

This study employed a quasi-experimental interrupted time series design over 12 months. The intervention involved displaying posters educating family physicians on accessing GAD-7 screening tools in EMR and using smart phrases to document GAD-7 results in two outpatient family medicine clinics. SlicerDicer measured total anxiety encounters and GAD-7 utilization 3 months before and 9 months after intervention. Statistical process control was used, and control charts were created using the statistical software JMP Pro-16. A Poisson regression model was used to detect statistically significant differences using statistical software SAS 9.4.

GAD-7 utilization increased from 4.5 in the preintervention period to 20.5 in the postintervention period. There was a significant increase in GAD-7 utilization over time after implementing our QI intervention. The control chart phase analysis showed a shift in the process with an increase in the average rate of GAD-7 utilization from 11.5 per 100 encounters per week in the preintervention phase to an average of 35.8 per 100 encounters per week postintervention phase. The effect was sustained over a year postintervention. The Poisson regression model also showed a 21% increase (p < 0.0001) in the incidence rate ratio in the post-intervention period as compared with the preintervention.

Utilization of GAD-7 as an application of MBC improved with the implementation of a bundled intervention using quality improvement tools. Other programs may replicate this in their similar quality improvement endeavors.