A new study in the Lancet tells me that my low carb diet is going to make me die sooner (Sara B Seidelmann, et.al. Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis The Lancet, August 16, 2018). Turned out that people who were eating relatively lower carbs back 25 years ago were more likely to have died than those who ate a “moderate” amount of carbohydrates.
The study is one of the weakest forms of study which used one of the weakest forms of evidence and reaches marginal conclusions.
- The definition of low carbs is less than 40%. That’s not ketogenic levels at all.
- Association studies do not show causation (Studying Research – Part 5 – Types of Studies). The rooster crowing does not cause the sun to rise.
- People who did Low Carb diets back in the Atkins diet days were more likely to have done so for health reasons at the time (Wikipedia – Atkins Diet).
- The study showed that these people were also more likely to be obese, diabetic and smokers. Teasing out these differences can be tough.
- What someone ate as their regular diet 25 years ago is not a very good indication of what they ate in the following 25 years (Studying Research – Part 4 – Methodological Problems).
- A Food Frequency Questionaire used to determine diet. There are many issues with this methodology (Food Frequency Questionnaire at a Glance).
- All-cause mortality rates don’t filter out irrelevant causes (What does all-cause mortality mean?).
- The hazard ratio for low carb eaters was 1.20 which means that the low carb eaters were 20% more likely to die (Wikipedia – Hazard ratio).
- The hazard ratio for high carb eaters was even higher. High carb is defined as 70% or higher.
- The study was slanted towards plant based sources rather than animal based sources.
- The study mashed together studies which controlled for different elements.
- The society as a whole has shifted in what we have eaten over this 25 year period. How is that factored into this?
- The study lead, Sara B Seidelmann, is a well published cardiologist but this seems to be her only paper on this particular subject (list of her papers). The same with most of the rest of the authors.
Let’s compare the Hazard Ratio in the paper (HR=1.2) with Type 2 Diabetes (Tancredi M, et.al. Excess Mortality among Persons with Type 2 Diabetes.
N Engl J Med. 2015 Oct 29;373(18):1720-32.):
As compared with controls, the hazard ratio for death from any cause among patients younger than 55 years of age who had a glycated hemoglobin level of 6.9% or less (≤52 mmol per mole of nonglycated hemoglobin) was 1.92 (95% CI, 1.75 to 2.11)
No, low carb is not killing me. Diabetes was killing me.