An Outcome-Focused Measure of Mental Health Care Quality based on Standardized Patient-Reported Symptoms

PROJECT SUMMARY / ABSTRACT There are various psychological, cognitive, behavioral, medication and neurostimulation treatments that can improve the outcomes of people with common depressive and anxiety disorders. However, in usual practice, there is large variability in provider characteristics and delivery of treatments. Routine treatments are

often poorly characterized and structured clinical data on patients are scarce. The effectiveness and quality of routine mental health services in the community are not accurately monitored and are poorly understood. It will be necessary to implement monitoring of treatment quality so that treatment and outcomes can be

improved. At present, healthcare organizations, payers, and policy makers usually know little about the quality of care they support. Similarly, patients and their families have very limited information on quality to guide their choice of provider or treatment organization. This study develops, tests and validates a new, transdiagnostic outcome-focused mental health quality

measure. This measure is based on routine, regular patient reports of their symptoms. The measure can be aggregated at the provider, clinic, organization or plan level; inform choice of provider; and be used to improve routine delivery of services and health equity and reduce disparities among patients with common psychiatric

disorders. The quality measure is broadly relevant across community settings and populations, and suitable for endorsement by regulatory and governing bodies. The study is guided by partnership with stakeholders and end-users of quality measurement. The project aims to: 1) analyze existing data with responses to a wide

variety of items that are known to assess depression or anxiety, and empirically select symptom items for a transdiagnostic outcome-focused quality measure; 2) inform risk adjustment and benchmarking of the quality measure by studying the effects on outcomes of patient, provider, and practice factors, including social

determinants of health, baseline symptom severity, and diagnoses; and, 3) fully specify an outcome-focused quality measure that includes risk adjustment and benchmarks for improvement; and study, at practices nationally, its feasibility and psychometric properties, the effect of treatment characteristics on the quality of

care, and the effect of quality on health-related quality of life. The study leverages a unique existing database that contains more than 5 million symptom assessments from 500,000 patients collected during treatment episodes with more than 5,000 providers and 200 real- world practices. These patient-reported outcomes are supplemented with data on the characteristics of

patients, providers and treatment organizations. Analyses use psychometric methods, item response theory, and hierarchical and longitudinal modeling to study symptom data from patients over time during episodes of care. Results support development of an outcomes-focused quality measure than can be used to routinely

assess and improve the quality of mental health services.