Predicting stroke recovery: Three- and six-month rates of patient-centeredfunctional outcomes based on the Orpington Prognostic Scale

OBJECTIVE: To provide recovery rates after stroke for specific functions us ing the Orpington Prognostic Scale (OPS). DESIGN: Prospective cohort. SETTING: Hospital and community. PARTICIPANTS: 413 stroke survivors entered the study 3 to 14 days after suf fering a stroke. MEASUREMENTS: A cohort of hospitalized stroke survivors were recruited 3 to 14 days after stroke and assessed at 1, 3, and 6 months poststroke for neu rological, functional, and health status. Baseline OPS score was used to pr edict five functional outcomes at 3 and 6 months using development and vali dation datasets and receiver operating characteristic (ROC) curves. RESULTS: In 413 stroke survivors, functional recovery rates at 3 and 6 mont hs were similar. Baseline OPS predicted significant differences in recovery rates for all five outcomes (P < .0001 for all five outcomes at 3 and 6 mo nths). Personal care dependence was present at 3 months in only 3% of perso ns with baseline OPS scores of 3.2 or less compared with over 50% with OPS of 4.8 or higher. Independent personal care, meal preparation, and self-adm inistration of medication were achieved by 80% who had baseline OPS scores of 2.4 or lower compared with less than 20% when OPS scores were 4.4 or hig her. Independent community mobility was achieved in 50% of those who had OP S scores of 2.4 or lower but only 3% of those with OPS scores of 4.4 or hig her. The area under ROC curves assessing OPS scores against each of the fiv e outcomes ranged from 0.805 to 0.863 at 3 months and 0.74 to 0.806 at 6 mo nths. CONCLUSION: OPS scores can predict widely differing rates of functional rec overy in five important functional abilities. These estimates can be useful to survivors, families, providers, and healthcare systems who need to plan for the future.