Abstract 10: The PROMIS Physical Function Scale- a Promising Scale for use in Stroke

Abstract

Background: Functional status is a critically important outcome measure after stroke. The ability to electronically capture this through self-report is increasing. The Patient Reported Outcomes Measurement Information System (PROMIS) uses computer adaptive testing (CAT), which can increase score precision and reduce patient burden. The number of questions depends on the patient’s responses. The objective of this study was to evaluate the performance of the PROMIS physical function scale (PROMIS PF) compared to the validated Stroke Impact Scale 16 (SIS16) in an ambulatory stroke clinic. Methods: Patient-reported PROMIS PF (CAT version) and SIS16 were electronically collected on patients seen in the stroke clinic using the Knowledge Program platform. Patient completion rates were > 70%. Distribution of scores was compared to assess ceiling/floor effects. Internal consistency was assessed by calculating the SEM for PROMIS PF and Cronbach’s alpha for SIS16. Correlations with NIHSS were performed to assess convergent validity. Results: The PROMIS PF and SIS16 were collected on 5,691 patients from Jan 2010 - May 2014. Mean SIS16 score = 81.4 (SD 21.5), mean PROMIS PF score = 42.8 (SD 11.2). The SIS16 had a22% ceiling effect, <1 % ceiling effect was seen with PROMIS PF (Figure). Patients completed 16 SIS16 items and a median of 4 [IQR 4, 5] PROMIS PF items. Cronbach’s alpha for SIS16 = 0.959, SEM for PROMIS PF = 2.3; both values suggest excellent internal consistency. Test-retest correlation was 0.867 for SIS16 and 0.87 for PROMIS PF. Correlations between NIHS and the 2 scales were -0.461 (p<0.001) for PROMIS PF and -0.559 (p<0.0001) for SIS16. Conclusion: The use of the PROMIS system to obtain electronic patient-reported functional status in ambulatory stroke clinic is feasible. PROMIS PF is an option for measurement of physical function in patients with stroke. It had similar test characteristics as the SIS16 but with lower patient burden and no ceiling effect.