Timing of Eating and Its Role in Managing Diabetes

From a chronobiological point, glucose metabolism in humans follow a circadian rhythm through diurnal variation of glucose tolerance that typically peaks during day-light hours, when food consumption usually happens and reduces when it comes to night-dark hours when fasting usually occurs¹³. Several hormones involved in glucose metabolism, such as insulin and cortisol, exhibit circadian oscillation^14,15. For example, experiments in rodents have shown the importance of the circadian system in glucose metabolism with changes in insulin sensitivity and insulin secretion patterns inducing highly rhythmic changes, thus affecting blood glucose levels^16–18. Therefore, insulin secretion and sensitivity are closely regulated by circadian control and have strong effects on glucose metabolism. Unusual meal timings can cause glucose intolerance by affecting the phase relationship between the central circadian pacemaker and peripheral oscillators in cells of the liver and pancreas in rodents¹⁹. Similarly in humans, timed meal intake is also driven by the SCN, play a role in synchronization of circadian rhythms in peripheral tissues, thereby affecting glucose metabolism^7,8,20. The effects of diet on circadian rhythmicity clearly involves a relationship between factors such as meal timings and nutrients (chrononutrition), that can contribute to circadian perturbance and influence the manifestation of metabolic disorders such as type 2 diabetes (Fig. 1).

Since the middle of the 20th century, eating patterns have shifted towards later eating times with over one-third of the caloric intake consumed after 6 pm²¹. This late-eating pattern, common in the modern lifestyle today, may lead to circadian misalignment and therefore exert a negative impact on glucose control. Circadian misalignment is also increasingly found in night shift workers. As human beings are diurnal species and generally sleep at night, shift workers are prone to developing sleep disturbances when the relationship between the light-dark phase and food intake is desynchronized. Considerable epidemiological evidences have shown that the disruption of the biological circadian clock is negatively associated with various metabolic diseases such as obesity, cardiovascular disease, gastrointestinal problems and diabetes^4,22. Shift workers tend to have a higher basal metabolic index (BMI) compared with day workers^23,24. Several studies displayed a negative relationship between shift work and metabolism. Shift workers exhibited a lowered glucose and lipid tolerance following a change from day to night shift work^25–27. Insulin resistance was also demonstrated to be more recurrent in shift workers who were 50 years or younger²⁸. Another study equated the risk of shift work to the risk of smoking one pack of cigarettes a day in spite of controlling for other confounders and risk factors²⁹. The complexity of the phenotypic expression of obesity is influenced by numerous factors such as stress, social rhythm, altered patterns of transcriptional genes, altered glucose and lipid homoeostasis, disruption of the central and peripheral oscillators and a decreased thermogenic response during night eating³⁰.

Emerging evidence also suggests that chrononutrition is influenced by an individual’s “chronotype”. Chronotype is a behaviour manifestation of an individual’s internal circadian clock system, whereby they can be classified to have a preference for the morning or evening⁹. Individuals with an “evening chronotype” or also termed “later chronotype” are biologically driven to consume foods later in the day³¹.

There has been a relatively strong association between breakfast skipping and insulin resistance or type 2 diabetes shown in some studies. In a 16-year follow-up cohort study, using the Cox proportional hazards analysis, it was shown that after adjusting for known risk factors of T2D including BMI, US men who skipped breakfast had a 21% higher risk of developing T2D compared with men who consumed breakfast³². In another randomised crossover trial, 10 women underwent a 2-week intervention with a 2-week washout period of either skipping or consuming breakfast. The study demonstrated that omitting breakfast led to a significantly higher energy intake, fasting total and LDL cholesterol, and a significantly lower postprandial insulin sensitivity³³. Individuals who have a preference for late-night dinner consumption (later chronotypes) may have a tendency to skip breakfast the following morning³⁴. This may be due to insufficient time to eat during the day or due to reduced hunger in the morning. A large cross-sectional study in healthy Japanese individuals, after adjusting for body mass index (BMI), showed that breakfast skipping was not the sole cause of hyperglycaemia but also a result of late-night dinner consumption³⁵. Kobayashi et al. reported that healthy males who skipped breakfast and then consumed large meals at lunch and dinner, had greater postprandial glucose, especially after dinner³⁶. An acute study in type 2 diabetics, who were classified as later chronotypes and skipped breakfast, had poorer glycaemic control as indicated by their significantly higher HbA1c values³¹. A cross-sectional study by Sakai et al. reported that there was an independent association of having a late-night dinner and skipping breakfast with poor glycaemic control in individuals with type 2 diabetes³⁷.

The omission of breakfast further compounded by late-night meal consumption, may delay circadian rhythms. These individuals do not consume breakfast due to lack of appetite signalling brought about by the disruption of biological clocks³⁸. Due to inappropriate time of feeding, the misalignment of the circadian clock leads to worsening of glycaemic control and an increased risk of developing type 2 diabetes.

However, there is another body of evidence suggesting that breakfast is not the most important meal of the day, rather, it is ‘just another meal’ as there has yet to be an established causal relationship between skipping breakfast and its negative metabolic implications. The authors highlighted that in most breakfast studies comparing the effects of consuming or skipping breakfast, the duration of overnight fast was not accounted for³⁹. This was an important confounder that should be considered. For example, two breakfast consumers both had their last meals at 2200 h the night before, one ate breakfast the next morning at 0600 h compared with one that consumed breakfast only at 1000 h, the difference in duration of overnight fast could potentially be the cause for a prominent difference metabolically. Conversely, regardless of time of meal consumed, similar durations of overnight fast may have resulted in similar metabolic profiles.

Likewise, despite routine evidences indicating that skipping breakfast resulted in an increased BMI and an overcompensation of energy consumed later during the day, no causality has been established^40,41. Zilberter and Zilberter also discussed indirect potential benefits of omitting breakfast in relation to intermittent fasting where the suppression of appetite could lead to voluntary caloric reduction and caloric restriction may have profound metabolic effects on human health^39,42,43.

Evidence suggests that meal ingestion in the morning and late evening (time of day) influence glucose metabolism in humans. Earlier studies, using mixed meals or glucose infusion, have reported circadian responses of reduced glucose tolerance and insulin sensitivity in healthy participants for the evening rather than in the morning^44–47. An acute study by Sato et al. examined the effects of a late evening meal on diurnal variation of blood glucose in healthy individuals, assessed by continuous glucose monitoring (CGMS™)⁴⁸. There was an increase in blood glucose after the late evening meal which shifted towards later at night, with peaking of blood glucose observed during sleep⁴⁸. Late evening meals may cause postprandial hyperglycaemia with this decrease in glucose tolerance from morning towards the night.

A cohort study was the first to show the relationship between late-night dinner consumption and glycaemic control in type 2 diabetics, whereby having a late dinner meal after 8 pm was independently associated with an increase in HbA1c³⁷. In some acute trials, both healthy individuals^49,50 and type 2 diabetics⁵¹ showed significantly higher blood glucose and insulin values after night-time meals. These studies deduce that the disruption of the circadian rhythm led to the exacerbation of the physiological nocturnal decrease of glucose tolerance. Peter et al. showed that type 2 diabetic subjects who ate three identical meals had glucose excursions that were higher in the morning than in the evening⁵². There was increased glucose tolerance in response to the first and third meals of the day, irrespective of glycaemic control. There was also a change in circadian variation in insulin sensitivity in type 2 diabetics⁵². Type 2 diabetics exhibit a different daily circadian pattern from healthy individuals, with increased insulin sensitivity towards the night, and higher glucose excursions in the morning than in the evening.

Methods to improve glycaemic control at dinner has been reported by some authors. Dinner that was divided into two smaller meals reduced post-meal glycaemic excursions, due to the “second-meal effect” phenomenon, by enhancing β-cell responsiveness⁵³ at the second dinner meal induced by the first meal^49,51. It was also suggested that if meal size and carbohydrate quantities were smaller, postprandial glucose can be ameliorated in both healthy and type 2 diabetics.

In summary, time of day is indicative of having an influence on the postprandial glucose response to a meal. There is a defined circadian pattern for postprandial glycaemia for similar meals consumed either in the evening or morning. These studies illustrate that glucose metabolism is not only affected by what and how much you eat alone, but also when the meal is consumed. However, more data from well-designed epidemiological studies is necessary to prove causality.

Current evidence suggests that the time of day in which the amount of calories is consumed, can affect glycaemic control. Some animal studies have shown that there is an impairment of peripheral clock gene expressions due to skipping breakfast or reduced food intake in the first meal of the day, along with high-caloric dinners (despite no differences in daily total caloric intake), resulting in higher daily glucose excursions^54,55.

A cohort study reported that when most of the day’s calorie requirements were consumed at dinner, there was a 2-fold greater incidence of diabetes in older men and women⁵⁶. In a randomized, parallel-arm designed study, the effect of consuming a high calorie meal in the morning (i.e. breakfast) versus a high calorie meal in the evening, was assessed in overweight and obese women (with metabolic syndrome)⁵⁷. The group who consumed more calories at breakfast saw a greater reduction in fasting blood glucose and insulin when compared with consuming more calories at dinner⁵⁷. This was taking into account the fact that there was a similar daily calorie intake for both arms. Furthermore, a reduction in glycaemic and insulinemic responses for oral glucose tolerance test (OGTT) after a high calorie breakfast compared with a high calorie dinner was also observed⁵⁷. In a crossover study, type 2 diabetic participants who were given a high-caloric breakfast/low caloric dinner (contrasting arm was a low caloric breakfast/high-caloric dinner), was associated with reduction in postprandial hyperglycaemia, increase in insulin and glucagon-like peptide-1 (GLP-1) for the whole day⁵⁸. The results of this study by Jakubowicz et al. demonstrated the diurnal variation of glycaemic control in type 2 diabetics⁵⁸. The amount of calories consumed at breakfast or dinner seemed to have an influence on the daily rhythm of postprandial glycaemic excursions and insulin levels.

For most people, it would be unrealistic to avoid eating at night. With night-time eating being associated with poor glucose control and the increased risk of type 2 diabetes, manipulating meal composition of late meals or reducing the portion size may be crucial strategies to positively impact on postprandial glucose.

The quantity, quality type, and rate of digestion of dietary carbohydrates are the primary determinants of postprandial glucose levels and insulin response^59,60. Therefore, they all play an important role in the diet we consume, when it comes to circadian rhythmicity and glycaemic control. Epidemiological studies have shown that the time of consumption of carbohydrate-rich meals, if at the start of the day, has protective benefits against the development of diabetes^61,62. Acute trials have reported that, specifically nocturnal consumption of carbohydrates, an increased absorption of dietary cabohydrates resulted in a higher postprandial glucose profile the following morning^63,64.

Epidemiological studies has reported that consumption of more carbohydrates than fats in the morning prevents the development of diabetes and metabolic syndrome^61,62. The effect of manipulating fats, as well as carbohydrates, in day and night meals on postprandial glycaemic response has been undertaken in a few experimental studies. A randomized crossover trial in healthy men compared whether consuming a high carbohydrate diet or a high fat diet during different timings in a 24 h period, would produce different plasma glucose responses⁷⁴. A more rapid rise in plasma glucose was observed with the high carbohydrate diet compared with the high fat diet. There was also a circadian pattern in plasma glucose concentration, with the circadian effect coming from the high fat diet consumption⁷⁴. A recent randomised crossover trial comparing two isocaloric meals, differing in total sugar and saturated fat, was undertaken during a simulated night shift work in overweight males with high fasting lipids⁷⁵. Although this study resulted in no significant changes in circadian gene expressions, modifying a meal by reducing saturated fat and sugar for a dinner meal was associated with improved glucose response⁷⁵. Whilst the quality of fat ingested is known to influence metabolism, there is a lack of consistent information on the degree of saturation and chain length of fatty acids influencing postprandial glycaemia and lipidemia. This further highlights the need to investigate the chronobiology of dietary fat intake on glucose homoeostasis.

A recent crossover study on healthy participants examined if a high protein meal could attenuate postprandial glucose in the morning and at night⁷⁶. The effect of a high protein meal showed a significant modulation of the glucose response at night, with a significantly lower incremental area under the curve (iAUC) compared with a standard meal⁷⁶. There were no differences in iAUC glucose for morning between the high protein test meal and standard meals. In addition, there were no differences noted for insulin responses between meal type in the morning or night⁷⁶. These findings suggest that increasing the amount of proteins in a meal can reduce postprandial glucose at night. This can be beneficial for people who are late chronotypes or late-night eaters, who are more predisposed to glycaemic excursions and therefore reduce the risk of hyperglycaemia. The glucose-attenuation property of dietary protein seems to be also influenced by the timing of consumption. However, there are limited studies to support this and more needs to be done on the glycaemic and insulinemic impact of protein in meals, in accordance to timing of day.

An overall summary of key studies reporting the effects of meal timings and dietary factors on glycaemic control is shown in Table 1.