[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.197.171.35. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Citations 0
Letters
April 15, 1998

Dietary Fat and Ischemic Stroke

Author Affiliations
 

Margaret A.WinkerMD, Senior EditorIndividualAuthorPhil B.FontanarosaMD, Senior EditorIndividualAuthor

JAMA. 1998;279(15):1171-1173. doi:10-1001/pubs.JAMA-ISSN-0098-7484-279-15-jbk0415

To the Editor.—In their analysis of the Framingham Heart Study, Dr Gillman and colleagues1 suggest that "[q]uestions of potential bias and confounding naturally accompany findings that are surprising or counterintuitive." The finding that they refer to is "an inverse association between fat intake and the development of ischemic stroke among men." Indeed, the data in Table 2 of the article suggest that a more complete examination of the results is needed before their conclusion can be accepted. The group made up of subjects with the lowest percentage intake of energy from fat differs markedly from the group with the highest percentage of energy from fat in the following ways. The energy intake of the lowest-fat group is about 54.8% of that in the highest-fat group (7879 kJ/d compared with 14389 kJ/d). The lowest-fat group's alcohol intake is about 190.4 g/wk compared with 160.0 g/wk for the highest-fat group. The fruit and vegetable intake is 4.6 servings per day for the lowest-fat group and 5.7 per day for highest-fat group. Had ranges and SDs been presented for these variables, it would have been clearer as to whether some subjects in the lowest-fat group had particularly high-risk nutritional patterns. It would not be surprising or counterintuitive to find that poorer nutrition quality, ie, subjects who consume less energy, with large alcohol intake and fewer fruits and vegetables as parts of total energy consumption, would be at greater risk for stroke. A higher rate of stroke in the subjects with the highest alcohol intake would help to decide whether Framingham is similar in this way to the Physicians' Health Study,2 the Honolulu Heart Study,3 the Scottish Heart Health Study,4 and the Busselton study.5

First Page Preview View Large
First page PDF preview
First page PDF preview
×