Walnuts and diabetes management

Walnuts

Nikolas Bakalar opens his mass media article entitled "Walnuts for Diabetes" by stating, "eating walnuts may reduce the risk for Type 2 diabetes in women, a large new study concludes." Bakalar bases this statement on a peer-reviewed journal article published in the Journal of Nutrition entitled, "Walnut Consumption Is Associated with Lower Risk of Type 2 Diabetes in Women." In the journal article, researchers an Pan, Qi Sun, JoAnn E. Manson, Walter C. Willett, and Frank B. Hu conducted a study on a large sample derived from the Nurses' Health Study, surveying 58,063 women aged 52 -- 77 and 79,893 women aged 35 -- 52. The large sample size ensures the generalizability and external validity of the Pan, et al. (2012) research. Bakalar's "Walnuts for Diabetes" summarizes the main points of Pan et al.'s (2012) research to make it digestible for a mass market audience unfamiliar with scientific jargon. In this sense, Bakalar's article has a larger audience, but it is still necessary to know whether Bakalar captured the essence of the original scientific research.

The original paper published in the Journal of Nutrition was sponsored in part by a grant by the California Walnut Commission, which Bakalar points out in the body of his media article. In the peer-reviewed journal article by Pan, et al. (2012), readers need to use a magnifying glass to find out this information. The sponsorship data is not mentioned as a potential limitation to the research, when it clearly illustrates a potential bias in the research hypothesis.

Bakalar briefly summarizes the findings of the Journal of Nutrition article without any critique or analysis. The summary mentions part of the reason why walnuts have been singled out for research. That is, walnuts contain high quantities of omega-3 and omega-6 oils.

Causation between walnuts and diabetes risk has not been established in the original article, and it would be helpful if Bakalar would tell his readers that eating walnuts will not necessarily cause a person to not have diabetes, if nothing else should change in that person's lifestyle. It is also clear that the original study fails to offer any causation between walnut consumption and a lowered diabetes risk. The authors do tersely point out the limitations of the study, but like Bakalar, refrain from acknowledging the validity issues with the research.

Pan et al. (2012) do mention, though, that there are several other variables at stake that could cause the walnut-eating female population to exhibit a lower risk for diabetes vs. women who consumed no walnuts. For example, the authors note that habitual nut consumption in general (not just walnuts) was "associated with several healthy lifestyle practices," (p. 516-517). Bakalar does mention that the researchers controlled for some diet and lifestyle factors such as eating fish and found that even when other factors were controlled for, walnut consumption was still linked with diabetes risk reduction. Neither Pan et al. (2012) nor Bakalar mention the fact that walnuts are relatively expensive and that income is also directly related to the incidence rates of diabetes. Such information would make the original research, and its summary article, more credible. Bakalar also fails to mention the fact that the Pan et al. (2012) study was conducted on a large but restricted sample of only white female nurses.