This study investigates the prognostic ability of the Orpington Prognostic Scale within 48 hours (OPS-1) after admission in predicting outcome at 6 months and 2 years in acute ischemic stroke and compares it with the 2 week OPS (OPS-2). All consecutive ischemic stroke patients (n = 117) were scored on the OPS, Barthel activities of daily living, Oxford handicap scale, European stroke scale, and Rivermead motor assessment at 48 hours, 2 weeks, 6 months, and 2 years post-stroke. Baseline OPS scores at 48 hours and 2 weeks were used to predict outcomes at 6 months and 2 years. The OPS-1 was an excellent predictor of length of hospital stay (P < .001), place of discharge (P < .01), and outcome at 6 months and 2 years (P < .0001, Fisher's exact). The OPS-2 was marginally better than the OPS-1 though this benefit was outweighed by the earlier stratification of the 48-hour measure. The sensitivity, specificity, and positive predictive values (PPV) of the "good" OPS-1 versus the OPS-2 at predicting independence at 6 months were 85% vs 92%, 85% vs 63% and 87% vs 92%, respectively. The sensitivity, specificity, and PPV of the "poor" OPS-1 versus OPS-2 were 48% v 35%, 97% v 100%, and 93% v 100% respectively. The OPS at 48 hours is a good predictor of outcome at 6 months and 2 years after ischemic stroke and allows early identification of 3 prognostic groups, which may help in identifying patients most likely to benefit from intensive rehabilitation.