The National Quality Forum (NQF) commissioned two background papers to inform its work on patient-reported outcome performance measures (PRO-PMs). This first paper is on methodological issues in the selection, use, and administration of patient-reported outcomes for performance measurement in health care settings. The authors are from the Feinberg School of Medicine, Northwestern University.
According to the paper, two major challenges to using PROs for purposes of accountability and performance improvement must be addressed. First, PROs have not been widely adopted in clinical use; thus, they are unfamiliar to many health care professionals, payers, and others in health care systems. Second, little is known about the best set of responsive questions to aggregate for the purpose of measuring performance of the health care entity. Some PROs can be used across patient populations and chronic conditions but others have been developed for very specific diseases and conditions. Well-known surveys such the Behavioral Risk Factor Surveillance System (BRFSS) and National Health and Nutrition Examination Survey (NHANES) are health risk assessments that measure health behaviors (now with large longitudinal national and state data sets), but component items have not been frequently used to measure healthcare provider performance.
Measurement of patient ratings is a complex concept that is related to perceived needs, expectations of care, and experience of care. Patient ratings can cover the spectrum of patient engagement, from experience to shared decision-making to self-management to full activation. Patient satisfaction is a multidimensional construct that includes patient concerns about the disease and its treatment, issues of treatment affordability and financial burden for the patient, communication with health care providers, access to services, satisfaction with treatment explanations, and confidence in the physician. Another potentially important predictor of health outcomes is patient activation, or the degree to which patients are motivated and have the relevant knowledge, skills, and confidence to make optimal health care decisions. The Consumer Assessment of Healthcare Providers and Systems (CAHPS) is the best known and most frequently used measure of patient experience.
Methodological issues that must be considered include the source of PRO data (self or proxy), mode of administration (self or interviewer), method of administration (written “paper and pencil”, telephonic, interactive voice response, web-based, touchscreen, or other electronic; and measurement equivalence if multiple methods are used), setting of administration (clinic, home, or other), and scoring (using raw scores through classical test theory or applying probabilistic analysis using modern test and item response theory).
Additional barriers that must be addressed include administering PROs in vulnerable populations; literacy, health literacy and numeracy; language and cultural differences; differences in functional abilities; response shift; and the impact of non-responders to items and questionnaires.
Finally, the authors of the paper list characteristics of PROs that should be considered when selecting them for performance measurement: conceptual model, reliability, validity, interpretability of scores, burden (on both respondent and administrator), availability of alternate modes and methods of administration, availability of language and cultural adaptations, and integration within electronic health records.
Ignatius Bau 