Background: Cognitive assessment is a required component of the Medicare Annual Wellness Visit (AWV). In this prospective study, we evaluated acceptability and usefulness of a patient-reported outcome measure (the PROMIS® Cognitive Function Screener, or PRO-CS) to screen for cognitive impairment during the AWV. We compared two versions of the PRO-CS: Abilities and Concerns.
Methods: We developed PRO-CS Abilities and PRO-CS Concerns using items from the PROMIS Cognitive Function item banks. We partnered with a large health system in Pennsylvania to implement an electronic health record (EHR)-integrated version of the 4-item PRO-CS into their AWV workflow. PRO-CS Abilities was implemented in June 2022 and then replaced with PRO-CS Concerns in October 2022. We used EHR data to evaluate scores on Abilities versus Concerns and their association with patient characteristics. We gathered feedback from providers on experiences with the PRO-CS and conducted cognitive interviews with patients to evaluate their preferences for Abilities versus Concerns.
Results: Between June 2022 and January 2023, 3,088 patients completed PRO-CS Abilities and 2,614 patients completed PRO-CS Concerns. Mean T-scores for Abilities (54.8) were slightly higher (indicating better cognition) than for Concerns (52.6). 10% of scores on Abilities and 13% of scores on Concerns indicated concern for cognitive impairment (T-score < 45). Both Abilities and Concerns were associated with clinical characteristics as hypothesized, with lower scores for patients with cognitive impairment diagnoses and those requiring assistance with instrumental activities of daily living. Abilities and Concerns had similar negative correlations with depression (r= -0.31 versus r= -0.33) and anxiety (r= -0.28 for both), while Abilities had a slightly stronger positive correlation with self-rated health (r = 0.34 versus r = 0.28). In interviews, providers commented that the PRO-CS could be useful to facilitate conversations about cognition, though several providers noted potential limitations of patient self-report. Feedback from patients indicated a preference for PRO-CS Concerns.
Conclusions: Our findings suggest potential utility of the PRO-CS for cognitive screening in the Medicare AWV. PRO-CS Abilities and Concerns had similar associations with patient clinical characteristics, but the Concerns version was more acceptable to patients.
One of the barriers to effective cognitive screening during the Medicare Annual Wellness Visit (AWV) is the lack of a brief, reliable, and validated screening tool for cognitive impairment that can be easily integrated into the electronic health record. Use of a patient-reported outcome measure for this purpose could alert clinicians to changes in cognition that require further assessment. Previous research has demonstrated that patient-reported cognitive abilities are potentially a separate construct from patient-reported cognitive concerns, and it is unclear which construct is optimal for use in a cognitive screener. In this prospective study, we evaluated acceptability and usefulness of a patient-reported outcome measure (the PROMIS® Cognitive Function Screener, or PRO-CS) to screen for cognitive impairment during the AWV. We compared two versions of the PRO-CS: Abilities and Concerns. Our findings suggest the PRO-CS was acceptable to providers and patients, though some providers had reservations about the use of patient self-report for cognitive screening. PRO-CS Abilities and PRO-CS Concerns had similar associations with patient clinical characteristics, but feedback from patients and experts indicated a preference for PRO-CS Concerns. Given the time and resource constraints in the Medicare AWV, our findings suggest the PRO-CS may be useful as the first step to engage providers, patients, and families in conversations about cognition. Further research is needed to understand the impact of the PRO-CS on patient-provider communication, as well as subsequent diagnostic testing and referrals.
Keywords: Alzheimer’s disease; Cognition; Cognitive function; Cognitive impairment; Dementia; Medicare; Primary care; Screening.
© 2024. RAND Corporation.