Physical activity predicts fitness and walking capacity after stroke: a diagnostic accuracy study

Abstract Background and Purpose Clinicians need access to accurate self-reported tools that can assist with screening individuals who are at risk for fitness (V̇O 2 peak) and walking impairments (e.g., 6-minute walk test [6MWT] distance) post-stroke. The associations and diagnostic metrics between self-reported physical activity as measured by the Physical Activity Scale for People with Disabilities (PASIPD, MET-hours/week) and V̇O 2 peak (≥ 15 mL/kg/min) and 6MWT (≥ 350 m) among ≥ 6 months post-stroke were evaluated. Methods This is a secondary analysis from an aerobic exercise RCT. Participants’ age, sex, V̇O 2 peak, 6MWT distance and the PASIPD were collected at baseline. Multivariable logistic regression analyses evaluated the association between V̇O 2 peak (≥ 15 mL/kg/min), 6MWT (≥ 350 m) and the PASIPD (MET-hours/week). Predicted classifications and the Youden index identified cut points of the PASIPD. Results Eighty-five participants (n=53 males, aged 65.1 ± 9.5 years, 1.8 ± 1.2 years post-stroke) were included. Significant associations were found whereby one-unit increase in the PASIPD (MET-hours/week) was associated with a 21% increase in the odds of having a V̇O 2 peak ≥ 15 mL/kg/min [aOR = 1.21; 95% CI 1.07, 1.36, p=0.002] and 14% increase in the odds of having a 6MWT ≥ 350 m [aOR = 1.14; 95% CI 1.05, 1.23, p=0.001] with excellent area under the curve values (AUC: 0.82-0.93). Youden-derived PASIPD cut points of 8.9 and 10.6 MET-hours/week may identify individuals with fitness (≥ 15 mL/kg/min) and walking impairments (≥ 350 m)post-stroke (AUC: 0.69-0.71). Discussion and Conclusions Clinicians may use the self-reported PASIPD to identify individuals with fitness and walking impairments post-stroke.