Daily eating occasions and risk of type 2 diabetes

From the Neuhouser Group, Public Health Sciences Division

In the US, the prevalence of type 2 diabetes mellitus (T2D) is projected to triple in the coming decades. Currently, the prevalence is astounding; over 23 million US citizens have T2D. Not only is it the seventh leading cause of death, the 2017 T2D expenditures and related health difficulties were approximately $327 billion dollars. Several studies have identified dietary behaviors, body weight, physical activity, and genetics as risk factors of T2D. Eating behaviors are unconfirmed risk factors.  Eating behaviors include daily breakfast consumption and time-restricted eating. It is also not clear whether eating frequency increases the risk of T2D.  If it does, one biological effect could include the chronic exposure to hyperglycemia, which would negatively affect pancreatic B-cell function, next glucose regulation could become impaired and increase the risk of T2D.

The scientific and clinical community would like to learn whether or not increased levels of glucose during the day due to frequent eating is hurtful in regard to T2D risk compared to elevated oscillations of glucose that occurs after eating full sized meals and less frequent eating. Therefore, it is very important to examine eating occasions as an exposure variable and modifiable risk factor for T2D.  The Neuhouser Group, from the Division of Public Health Sciences, evaluated the association between the number of daily eating occasions and the risk of T2D  and whether these associations varied by body mass index (BMI), age, or race/ethnicity in the Women’s Health Initiative Dietary Modification Clinical Trial (WHI-DM).  The study was published in a recent issue of Current Developments in Nutrition

The WHI-DM postmenopausal participants were recruited from 40 US clinical centers who were randomly assigned to a low-fat dietary pattern or a control group from 1993 to 1998. The DM intervention was delivered as a behavior modification program. Although T2D was not one of the primary outcomes of the trial, the intervention successfully increased the consumption of fruits and vegetables and lowered dietary fat intake as a percent of total energy. Demographics and covariate data (physical activity, smoking history, other lifestyle characteristics) were collected at baseline visits. Staff at the 40 WHI Clinical Centers measured waist circumference, height, and weight to obtain the BMI.  A food frequency questionnaire (FFQ) was used to assess baseline dietary intake in all participants. The participants completed the FFQ at baseline, 1 year after randomization, and thereafter annually to one-third to the participants on a rotating basis. A subsample of participants (4.6%) completed a 24 Hour (HR) recall at years 3 and 6.  The completed 24 HR recalls at year 3 were used to create a measure of eating frequency since the recall record also captures information on number of eating occasions per day in addition to meal contents. T2D outcomes were assessed by self-report. Cox proportional hazards were used to evaluate the association between tertiles of eating occasions and risk of T2D, and additional models.   evaluate the risk of T2D stratified by baseline BMI (<30/>30), race/ethnicity, and age. 

An estimated 15% of the subsample (15.4%=332) self-reported the incidence of diabetes after the start of the recall cohort. Women who reported 4 daily eating occasions were 38% more likely to develop diabetes than women who reported 1-3 daily eating occasions (HR: 1.38, 95% CI: 1.03-1.84).  Women, with a BMI less than 30 kg/m2, who reported 4 eating occasions per day had a 55% greater risk of developing diabetes than women who reported 1-3 eating occasions per day (HR:1.55, 95% CI: 1.00-2.39).  Women who were older than 60 years and reported 4 eating occasions were more likely to develop diabetes than those with less than 4 eating occasions (HR: 1.61; 95% CI: 1.11-2.33).

In this study, there were statically significant increases in T2D risk in women who reported 4 daily eating occasions compared to those who reported fewer daily eating occasions. Once stratified by BMI (less than 30 kg/m2) and age (older than 60 years), the associations were stronger.  There was no further increase in risk for women who reported 5 daily eating occasions, in fact the results were attenuated. However, included in the group who reported 5 or more daily eating occasions were those who ate up to 10 times daily so there is likely some unexplained variation in that group that was not possible to tease out further. This study could be considered novel in its approach and relates to a wider interest in the scientific community on both number of eating occasions and time of day of eating – "chrononutrition."  The Neuhouser group stated that two ways to further their research“include biomarkers of insulin resistance over the course of the day. This could provide more definitive evidence on whether there is an association between the total number of daily eating occasions and the risk of T2D. In addition, intervention studies testing chrononutriton could also add valuably to this line of research.” 

This research was supported by the National Heart, Lung, and Blood Institute, National Institutes of Health, and US Department of Health and Human Service.

Fred Hutch/UW Cancer Consortium member Marian Neuhouser contributed to this work.

Neuhouser ML, Wertheim BC, Perrigue MM, Hingle M, Tinker LF, Shikany JM, Johnson KC, Waring ME, Seguin-Fowler RA, Vitolins MZ, Schnall E. 2020. Associations of Number of Daily Eating Occasions with Type 2 Diabetes Risk in the Women's Health Initiative Dietary Modification Trial. Current Developments in Nutrition. 2020 Aug;4(8):nzaa126.