Zelano J, Redfors P, Åsberg S, Kumlien E. Association between poststroke epilepsy and death: A nationwide cohort study. European Stroke Journal. 2016;1(4):272–278. DOI:10.1177/2396987316669000
When re-examining the data set of our previous study (1) for the purpose of a meta-analysis, the authors discovered errors in some reported hazard ratios (HR). They have found that this was caused by a coding error in early analyses, which were then by mistake delivered as final versions from our external statistical consultant. The error did not affect the direction of any reported associations, or any conclusions in the paper, but the risk of death after PSE was somewhat underestimated in the affected analyses. We sincerely apologize for this error.
The HR for death after development of PSE in all patients was correctly reported at 1.69 (95%CI: 1.64–1.74), but the HR for subgroups were incorrect. Table 3 should read:
| Age | Stroke type | n | HR (95% CI) |
|---|---|---|---|
| All | All | 106455 | 1.69 (1.64–1.74) |
| <45 | ICH | 400 | 2.01 (0.77–5.23) |
| <45 | Infarction | 1631 | 1.90 (0.90–4.00) |
| >=45 | ICH | 9795 | 1.28 (1.16–1.40) |
| >=45 | Infarction | 94629 | 1.84 (1.78–1.90) |
In the results, the sentence “When stratified by stroke type and age, the risk of death was significantly increased (HR: 1.54, 95% CI: 1.38–1.72) in patients aged 45 or above with PSE after ischemic stroke.” should read “When stratified by stroke type and age, the risk of death was significantly increased in patients aged 45 or above.” In the results and discussion, the correct HR of the adjusted analysis is not 1.36 (1.20–1.55), but 1.71 (95%CI: 1.65–1.78).