Reply to I Shiue

Dear Sir:

We appreciate Shiue's insightful comments and question about the possibility of an association between consumption of a Mediterranean-style diet (MEDI) and hemorrhagic stroke in our multiethnic population-based cohort, the Northern Manhattan Study (NOMAS) (1). We agree that more studies are needed to compare the relative associations of a MEDI across stroke subtypes to inform public health messages. Currently in NOMAS, we do not have sufficient statistical power to examine the association between a MEDI and hemorrhagic stroke after other vascular risk factors are controlled for, because we have only 26 confirmed hemorrhagic strokes in the available sample. However, to address Shiue's question, we conducted an exploratory analysis and observed an inverse trend between adherence to a MEDI and the risk of hemorrhagic stroke, but the association was not significant. As previously described, consumption of a MEDI was evaluated on a 10-point scale (score of 0–9), with a greater score indicating consumption of a diet more reflective of a MEDI. We observed that for every 1-point increase in the MEDI score, the HR of hemorrhagic stroke over a mean of 10 y of follow-up was 0.85 (95% CI: 0.67, 1.08; P = 0.19). Therefore, the effect of a MEDI on hemorrhagic stroke risk was more protective than that for ischemic stroke, although neither was statistically significant. We will revisit this examination in the future when there is enough power for a rigorous multivariate-adjusted analysis, and we agree with Shiue that this question should be addressed in other large population-based cohort studies.