To the Editor.
—There are several major problems with the article by Heinemann et al1 that severely reduce the validity of the findings. The fundamental issues of sample appropriateness, diagnosis, comorbidity, timing, and use of an activities of daily living measure of unknown reliability and validity are ignored.2 There are numerous inconsistencies in the sample size at different analysis stages, particularly relating to measures of improvement. Other problems include the nonstatement of the number deceased, and the lack of comment about the small number of patients admitted from home (15% were admitted from home, while 75% were discharged to home). The lack of sufficient information may be due to inadequate documentation and poor extraction of data, which, unfortunately, invalidates the results.The statistical assumptions of the ϰ2 test (independence of variables) seem to have been violated. Regression analysis, not including nonsignificant variables, and the reporting of raw
Vanclay F, Shah S. Validity of Outcome Following Stroke Rehabilitation. Arch Neurol. 1989;46(12):1270–1271. doi:10.1001/archneur.1989.00520480012007
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: