Abstract
OBJECTIVE: This investigation estimated the incremental cost-effectiveness of high-intensity training (HIT) compared with conventional physical therapy in individuals with subacute stroke, based on the additional personnel required to deliver the therapy. DESIGN: Secondary analysis from a pilot study and subsequent randomized controlled trial. SETTING: Outpatient laboratory setting. PARTICIPANTS: Data were collected from individuals with locomotor impairments 1-6 months poststroke (N=44) who participated in HIT (n=27) or conventional physical therapy (n=17). INTERVENTIONS: Individuals performing HIT practiced walking tasks in variable contexts (stairs, overground, treadmill) while targeting up to 80% maximum heart rate reserve. Individuals performing conventional therapy practiced impairment-based and functional tasks at lower intensities (<40% heart rate reserve). MAIN OUTCOME MEASURES: Costs were assessed based on personnel use with availability of similar equipment. Incremental cost-effectiveness ratios (ICERs) and cost-effectiveness acceptability curves were calculated for quality-adjusted life years (QALYs) derived from the Medical Outcomes Short Form-36 questionnaire and gains in self-selected speeds (SSSs). RESULTS: Personnel costs were higher after HIT (mean, $1420±234) vs conventional therapy (mean, $1111±219), although between-group differences in QALYs (0.05 QALYs; 95% confidence interval [CI], 0.0-0.10 QALYs) and SSS (0.20 m/s; 95% CI, 0.05-0.35 m/s) favored HIT. ICERs were $6180 (95% CI, -$96,364 to $123,211) per QALY and $155 (95% CI, 38-242) for a 0.1 m/s gain in SSS. CONCLUSIONS: Additional personnel to support HIT are relatively inexpensive but can add substantial effectiveness to subacute rehabilitation. Future research should evaluate patient factors that increase the likelihood of improvement to maximize the cost-effectiveness of treatment post stroke.