Send to

Choose Destination
See comment in PubMed Commons below
Am J Epidemiol. 2004 Sep 1;160(5):436-44.

Problems with the assessment of dietary fat in prostate cancer studies.

Author information

Department of Epidemiology, College of Public Health, University of Iowa, 200 Hawkins Drive, Iowa City, IA 52242. USA.


The authors conducted a detailed review of studies on the association between prostate cancer and total dietary fat along with specific fatty acids. Overall, the 29 studies reporting actual dietary fat levels in grams of fat were heterogeneous, suggesting that pooling of the relative risks may be inappropriate. Heterogeneity was also seen by study design. More specifically, although the pooled estimate for prostate cancer and an increase of 45 g in total fat consumption per day was small (relative risk = 1.2), heterogeneity between studies was large, and the association was not supported by specific fatty acids. The strongest association was found among the five extremely inconsistent studies of alpha-linolenic fatty acid. The associations with advanced prostate cancer were more homogeneous and suggest a relation with total and saturated fat but none with specific fatty acids. This review highlights the inconsistent way in which total dietary fat and specific fatty acids have been measured and reported across epidemiologic studies of prostate cancer. The heterogeneity between studies was large, possibly because of the variation in the dietary instruments used and the corresponding databases (nondifferential misclassification), recall bias, differing case definitions, residual confounding, or potential selection bias in different studies.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Silverchair Information Systems
    Loading ...
    Support Center