This is a real head-scratcher for me, just based on the abstract. I can’t explain or write-off the researchers findings at this point. I hope they administered the food frequency questionnaire more than once. If not, I can’t take this seriously.
•Of 9689 middle-aged Australian women, 10% developed type 2 diabetes over 15 years.
•Carbohydrate restriction was associated with a 27% higher risk of type 2 diabetes.
•This association was attenuated after adjustment for BMI.
•The association was comparable for women with and without prior gestational diabetes.
•Women should be advised to avoid carbohydrate restricted diets low in fruit and grains.
Background and aims
Low-carbohydrate diets (LCDs) are increasingly popular but may be nutritionally inadequate. We aimed to examine if carbohydrate restriction in midlife is associated with risk of developing type 2 diabetes (T2DM), and if this association differs by previous gestational diabetes (GDM) diagnosis.
Methods and results
Dietary intake was assessed for 9689 women from the Australian Longitudinal Study on Women’s Health in 2001 (aged 50–55) and 2013 (aged 62–67) via validated food frequency questionnaires. Average long-term carbohydrate restriction was assessed using a low-carbohydrate diet score (highest quartile (Q4) indicating lowest proportion of energy from carbohydrates). Incidence of T2DM between 2001 and 2016 was self-reported at 3-yearly surveys. Log-binomial regression was used to estimate relative risks (RR) and 95% CIs. During 15 years of follow-up, 959 women (9.9%) developed T2DM. Carbohydrate restriction was associated with T2DM after adjustment for sociodemographic factors, history of GDM diagnosis and physical activity (Q4 vs Q1: RR 1.27 [95% CI 1.10, 1.48]), and this was attenuated when additionally adjusted for BMI (1.10 [0.95, 1.27]). Carbohydrate restriction was associated with lower consumption of fruit, cereals and high-fibre bread, and lower intakes of these food groups were associated with higher T2DM risk. Associations did not differ by history of GDM (P for interaction >0.15).
Carbohydrate restriction was associated with higher T2DM incidence in middle-aged women, regardless of GDM history. Health professionals should advise women to avoid LCDs that are low in fruit and grains, and to consume a diet in line with current dietary recommendations.
Steve Parker, M.D.