Also, they noted, the analysis was limited to survivors – the researchers had identified 467 subjects initially, of whom 120 died, another 51 were too disabled afterward to participate, and others dropped out or were excluded for other reasons – allowing for the possibility of survival bias "if certain triggers affect prognosis" after subarachnoid hemorrhage.
A third limitation, Dr. Vlak and colleagues wrote, was the assumption of a 24-hour hazard period for alcohol: "Many patients who reported exposure in the hazard period had a daily intake of at least [one glass] per day and therefore did not contribute to the [relative risk] in the main analysis."
The study was funded by the University Medical Center Utrecht. The authors reported no relevant disclosures.