In this example, the conclusion of the paper, that higher intake of “processed meat” is associated with increased risk of dementia is not supported by the methodology and is not relevant in establishing causation, especially since unprocessed red meat consumption is shown to have a negative association, and total meat, no association.
The food frequency questionnaires and 24 hour dietary recall were not precise enough to establish confounding with related food items, such as breading of meat, and related carbohydrate/sugar intake was not controlled for, nor was diabetes or insulin resistance. The research question isn’t relevant considering a much more abundant body of research linking insulin resistance and dementia exists and was not mentioned or considered in the study plan.
Adapted from: Zhang, Huifeng, Darren C Greenwood, Harvey A Risch, David Bunce, Laura J Hardie, and Janet E Cade. “Meat Consumption and Risk of Incident Dementia: Cohort Study of 493,888 UK Biobank Participants.” The American Journal of Clinical Nutrition, March 22, 2021, nqab028. doi.org/10.1093/ajcn/nqab028.
