Study on butter and mortality: A critical analysis

A prospective study of three American cohorts found that higher butter consumption was linked to an increased total and cancer mortality, while greater intake of plant-based oils was associated with lower mortality from all causes, cancer, and cardiovascular disease.¹ However, some key nuances of the study’s findings, which provide important context, may not have been adequately recognized.

Firstly, it is critical to consider the quantity of butter in question, as well as the way it was consumed: 

• Only total intakes of butter at the highest level (13-14g or ~1 tablespoon per day) demonstrated significant results with total mortality. 
• Butter intake, even at the highest level, was not linked to cardiovascular mortality risk. 
• When considering how butter was used, only butter used on food and bread, but not in cooking, was significant. Although, the authors controlled for the possible confounding effect of bread, it remains unclear what other foods butter was paired with. Thus, other dietary factors could have influenced the observed associations. 

Furthermore, understanding the broader dietary and lifestyle context is crucial when interpreting these results: 

• Overall diet quality: The association between butter consumption and increased total mortality was observed only in individuals with an unhealthy or poor-quality diet, not in those with a healthy diet. 
• Other key foods: The study did not assess whether individuals using plant oils also consumed more vegetables, such as using oil for salad dressing. While models adjusted for overall diet quality using the Alternative Healthy Eating Index, it did not isolate vegetable intake specifically. 
• Lifestyle habits: Participants with higher total butter intake had higher Body Mass Index (BMI) and energy intake and were more likely to currently smoke. They were also less likely to be physically active and to use multivitamins. In fact, the authors acknowledge that despite adjustment for certain confounding variables, residual confounding may still exist.
  • Smoking: Only cancer-related mortality showed a significant association with butter consumption, while no link was found with cardiovascular disease mortality. This may be influenced by smoking status, which was more common among those who consumed butter. Although the researchers adjusted for smoking in their analysis, this adjustment may not have fully accounted for its effect. 

Among other limitations mentioned by the authors, there is the possibility that participants may have mistakenly reported margarine intake as butter intake due to their similar appearance, as well as the limited generalizability of the study sample (healthcare professionals who were predominantly white).  

Importantly, previous robust studies have been conducted on this topic and found no adverse effects resulting from butter intake. In fact, stronger studies, such as a systematic review and meta-analysis and a large multi-country prospective cohort study (The PURE Study), support a moderate intake of butter as part of a healthy diet, with no adverse associations with health outcomes, including mortality.^2,3 

Similarly, according to the World Cancer Research Fund International, the leading health authority on diet and cancer, butter is not indicated with risk of cancer including colorectal, ovarian, breast, and prostate cancers.⁴ As well, butter is not mentioned as one of the Cancer Prevention Recommendations.⁵