Intended for healthcare professionals

CCBYNC Open access
Research

Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)

BMJ 2016; 353 doi: https://doi.org/10.1136/bmj.i1246 (Published 12 April 2016) Cite this as: BMJ 2016;353:i1246

Re: Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)

Avery interesting historical dataset. Clearly the diet had no effect on deaths overall with table 6 showing about 105 in the intervention and 95 in the control group with the bulk of the deaths occurring in the over 65 age group (probably in nursing homes rather than psychiatric institutions) . What is most striking is the large fall in cholesterol in the control group ( in whom there is definitely no increased linoleic acid but saturated fat intake may be lower than their free living diet) and the strong relationship between the fall in cholesterol and the death rate with 60% of the deaths occurring in the group with a 80-100mg/dl fall in cholesterol. This suggests the people who died had an illness that caused their cholesterol to fall dramatically prior to death and the same thing is likely to have occurred in the intervention group overwhelming any effect of diet (positive or negative). I think 10 deaths difference between groups is just random noise. If we examine a statin meta analysis in the over 65s (Teng et al 2015) statins reduced nonfatal and total myocardial infarctions and a composite endpoint but total mortality was not different (0.96, 0.88-1.04) despite nearly 26,000 participants. So a study of 2200 people treated for a year or more (we don't know the mean duration of treatment of this group) with a dietary intervention is so underpowered no conclusions are possible.

Competing interests: No competing interests

13 April 2016
Peter M Clifton
Physician and researcher
University of South Australia
PO Box 2471 Adelaide SA 5001 Australia