quantitative assessment of the association of seafood consumption with CHD mortality and nonfatal MI in people without a history of CHD, but concluded that insufficient evidence supported a quantitative assessment of seafood consumption for secondary prevention. From this study, it was estimated that, compared to not eating seafood, eating a small amount of seafood—as little as half a serving per week—was associated with a reduction in risk of cardiovascular death of 17 percent (95% CI 8.8-25.0) and a reduction in risk of nonfatal MI of 27 percent (95% CI 21-34). Each additional weekly serving of seafood was associated with a further decrease in the risk of cardiovascular death of 3.9 percent (95% CI 1.1-6.6), but no additional benefit was statistically significant for the risk of nonfatal MI. The Agency for Healthcare Research and Quality (AHRQ) reviews are systematic reviews that synthesize observational and experimental studies in a qualitative way (see Appendix B). The conclusions of the AHRQ reviews are also based on intervention studies in groups at risk. In contrast, the studies by Whelton et al., He et al., and Konig et al. are meta-analyses that quantitatively combined observational studies. Meta-analyses are usually considered stronger evidence than systematic reviews.
Taken together, these meta-analyses of observational studies suggest a negative association between seafood consumption and CHD or death, particularly in individuals without a prior history of CHD. Recent data suggest that even small amounts of seafood consumption may be associated with a decreased risk for CHD or death (Schmidt et al., 2005a,b). These results should, however, be interpreted with caution, as they are based on observational studies and are thus subject to residual confounding. In other words, based on observational studies only, it is difficult to exclude the possibility that seafood intake may just be a marker for healthier lifestyle, and that no causal association exists between seafood consumption and cardiovascular protection (see Appendix Tables B-2a and B-2b).
The only reported studies of the association between seafood consumption and stroke have been observational (see Appendix Table B-2b). He et al. (2004a) quantitatively assessed the relationship between seafood consumption and risk of stroke using a meta-analysis of nine cohorts from eight studies. Pooled RR and 95 percent CI of risk for stroke were estimated by variance-based meta-analysis. These results demonstrated that consumption of seafood was inversely related to stroke risk, particularly ischemic stroke. Even infrequent seafood consumption (as seldom as 1 to 3 times per month) may be protective against the incidence of ischemic stroke compared to seafood consumption less than once per month. The pooled RRs for all stroke, compared to individuals who consumed seafood less than once a month,