Rehabilitation outcomes of patients with posterior circulation stroke after acute inpatient rehabilitation.

The study looked back at existing records rather than collecting data according to a plan set in advance, so what was recorded depended on what staff happened to write down, and other differences between patients may explain the results. Only patients well enough to manage at least an hour a day of rehabilitation were admitted, and more severely affected patients are the ones most likely to be referred to inpatient rehabilitation, so these figures do not describe patients who cannot tolerate that intensity or those managed outside a rehabilitation ward. There was no comparison group, so the improvements seen over the rehabilitation stay cannot be attributed to the rehabilitation programme. Only eight patients returned to work, which the authors say makes the return-to-work analysis unstable and the confidence intervals wide; it also means the predictors identified may not hold in another sample. This was one regional hospital without in-house endovascular services, over a fixed 20-month period, so the results may not transfer to other settings. Three-month follow-up relied partly on telephone calls, and the rehabilitation stay varied widely, from 2 to 135 days.

The authors declared that they received no financial support for this work or its publication. No other competing interests are stated in the paper. The paper does not say who designed the study or who wrote it up.

Do not read the functional and walking gains as proof that the rehabilitation programme caused them. There was no comparison group, the data were collected by looking back at records, and only patients well enough for an hour of therapy a day were included, so some of this recovery would have happened anyway. The 11.1% return-to-work figure is also not evidence that rehabilitation fails at vocational recovery; it is what was recorded in one small cohort at a single time point.