Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: Chronobiol Int. 2022 Apr 5;39(7):976–983. doi: 10.1080/07420528.2022.2053702

Meal Timing, Distribution of Macronutrients, and Inflammation among African-American Women: A Cross-Sectional Study

Samantha C Truman 1, Michael D Wirth 2, Swann Arp Adams 3, Gabrielle M Turner-McGrievy 4, Kelly E Reiss 5, James R Hébert 6
PMCID: PMC9177623  NIHMSID: NIHMS1789247  PMID: 35379042

Abstract

Chronic low-grade inflammation is an underlying risk factor for numerous chronic diseases, including cancer. Eating earlier in the day has been associated with a reduction in levels of inflammatory markers and inflammation-related health outcomes (e.g., obesity, metabolic disorders). This cross-sectional study of 249 obese African-American women examined the effect of various mealtime-related factors associated with macronutrient consumption in relation to chronic inflammation and Breast Imaging Reporting and Data System (BI-RAD) readings. During 2011 and 2013, a single 24-hour dietary recall was administered, blood samples were assayed for c-reactive protein (CRP) and interleukin-6 (IL-6), and BI-RAD ratings were assessed to determine the influence of mealtime on chronic inflammation and breast cancer risk score. Multiple linear and logistic regression models were used to assess these relationships. Higher carbohydrate consumption at breakfast was associated with a significantly lower CRP vs. higher carbohydrate consumption at dinner (6.99, vs. 9.56 mg/L, respectively, p = 0.03). Additionally, every 1-unit increase in percent energy consumed after 5PM resulted in a BI-RAD reading indicating a possibly suspicious abnormality (OR: 1.053, 95% CI: 1.003–1.105), suggesting an increase in breast cancer risk. Timing of energy and macronutrient intake may have important implications for reducing the risk of diseases associated with chronic inflammation.

Keywords: timing, food intake, inflammation, breast cancer, obesity

Introduction

Circadian rhythms are known to influence many physiological mechanisms and behavioral processes. The suprachiasmatic nucleus (SCN), located in the anterior hypothalamus, houses the central circadian clock (Scheer, Hilton, Mantzoros, & Shea, 2009). The “internal desynchronization” of the circadian clock has been associated with numerous chronic diseases including cancer (Bahijri et al., 2013; Froy & Miskin, 2010; Scheer et al., 2009). While light-dark cycling is considered the dominant external cue for synchronizing the circadian clock, meal timing also may exert indirect control on the SCN, thus influencing peripheral clocks including those in the gut (Asher & Sassone-Corsi, 2015; Ribas-Latre & Eckel-Mahan, 2016). The importance of understanding how to properly control circadian rhythms is underlined by the World Health Organization’s classification of circadian disruption as a probable carcinogen (Adams, 2013).

“Eat like a king in the morning, a prince at noon, and a peasant at dinner” is a quote by medieval Jewish philosopher, Maimonides (Asher & Sassone-Corsi, 2015). This proverb and similar other ones can be traced back as far as to the Ancient Greeks and Romans (Almoosawi, Vingeliene, Karagounis, & Pot, 2016). Differences in the time-of-day that macronutrients are consumed, as well as late-night eating, have been associated with obesity (Ruddick‐Collins, Morgan, & Johnstone, 2020). For example, in a weight-loss study among overweight or obese individuals, participants were randomized to a diet arm with 700 kcal at breakfast, 500 kcal at lunch, and 200 kcal at dinner (labeled the breakfast group). The dinner group consumed the inverse pattern. Compared to the dinner group, the breakfast group had statistically significantly greater reductions in body weight (11% vs. 4%). This study also noted greater reductions for metabolic markers including glucose and insulin in the breakfast group compared to the dinner group (Keim, Van Loan, Horn, Barbieri, & Mayclin, 1997). There is some evidence indicating that greater caloric consumption in the morning compared to at night or adherence to an early time-restricted feeding protocol (eTRF) are associated with lower levels of inflammatory markers (Marinac et al., 2016, 2015). For example, one study found that each 10% increase in calories consumed during the evening (i.e., after 5PM) resulted in a 3% increase in c-reactive protein (CRP) (Marinac et al., 2015). However, not all studies involving shifting of calories to earlier in the day or some type of fasting approach have found associations with inflammatory biomarkers (Bhutani, Klempel, Kroeger, Trepanowski, & Varady, 2013; Sutton et al., 2018; Trepanowski et al., 2018).

In addition to obesity and inflammation, people who consume their last meal before 9PM have been shown to have a 15% lower odds of developing breast cancer compared to those who consumed a meal after 10PM (Kogevinas et al., 2018). This same study showed a reduction of nearly 25% for prostate cancer using the same comparisons (Marinac et al., 2015). In a study by Marinac and colleagues, an overnight fasting period of >13 hours (which tends to indicate greater energy consumption earlier in the day) was associated with an increased risk of breast cancer recurrence (hazard ratio = 1.36, 95% confidence interval [95%CI] = 1.05–1.76) (Marinac et al., 2015). It is possible that increased inflammation due to consumption of food later at night may be one of the mechanisms through which cancer develops. Studies have shown that elevated CRP is correlated with an increased risk of breast cancer (Guo et al., 2015). It also has been shown that breast cancer grows at different times of day (Oh et al., n.d.; Stevens, Brainard, Blask, Lockley, & Motta, 2014; Yang et al., 2009; You et al., 2005).

The purpose of this study was to assess the association of meal-timing behaviors on both the biological markers of chronic inflammation, CRP and interleukin-6 (IL-6), and breast cancer assessment score (Breast Imaging Reporting and Data System [BI-RADS]) among obese African-American women. Specifically, it was hypothesized that those women consuming a higher percentage of calories after 5PM would have higher levels of inflammation and a higher BI-RAD reading (higher probability of a malignant breast lesion) than those consuming a smaller percentage of calories after 5PM. In addition, as an exploratory analysis, macronutrient intake among meal-types (i.e., breakfast, lunch, and dinner) were assessed for their association with the study outcomes.

Materials and Methods

Study Sample

This cross-sectional study was derived from baseline data collected from 2011 to 2015 through the Sistas Inspiring Sistas Through Activity and Support (SISTAS) study. SISTAS was a randomized diet and exercise-based intervention study that aimed to reduce breast cancer risk among obese African-American women (Bevel et al., 2018). For initial inclusion in the SISTAS trial, participants had to self-identify as African American; be 30 years of age or older; have obesity (body mass index ≥ 30 kg/m2); be willing to be randomized; have not had a previous cancer diagnosis, except non-melanoma skin cancer; have no serious, unstable co-morbidities that would make participation in a diet and physical activity intervention difficult or risky; have no inflammation-related conditions such as rheumatoid arthritis or Crohn’s disease; and have no stable hormone replacement therapy usage. However, it should be stressed that this is a baseline-only analysis that did not utilize the intervention design. The sample size for this analysis was based on individuals with available data from baseline measurements (Bevel et al., 2018). The study protocol was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the University of South Carolina Institutional Review Board (Pro00008713). All individuals gave their informed consents for inclusion before they participated in the study.

Measures

Participants (n = 249) completed a detailed questionnaire that included information on demographics, medical history, self-efficacy for diet, and perceived stress. Anthropometric measures such as height (stadiometer), weight and body fat percentage (Tanita® TBF-300WA Body Composition Analyzer) were measured. Blood pressure was determined using a sphygmomanometer.

A single 24-hour dietary recall (24HR) interview was telephone-administered during a weekday and was administered using the Nutrition Data System for Research software (NDSR, 2015), licensed from the Nutrition Coordinating Center (NCC) at the University of Minnesota. The 24HR occurred around the time of the clinic visit when blood samples were drawn for characterization of inflammatory markers. Meal-timing variables included: (1) the percentage of calories consumed in the evening (between 5:00PM and 3:59AM), (2) the number of kilocalories and percentage of macronutrients consumed at breakfast, lunch, and dinner, and (3) the number of kilocalories and percentage of macronutrients consumed during meal-time bins 4:00AM–9:59AM, 10:00AM–2:59PM, 3:00PM–7:59PM, and 8:00PM–3:59AM. Meal-time bins were determined based on a previous study conducted by St-Onge and colleagues (St-Onge, Pizinger, Kovtun, & Roychoudhury, 2018). Fasting times were not examined in this research. Preferably, consecutive days of dietary reporting are ideal for estimating average nighttime fasting. Given that such a data collection protocol was not utilized and that the definitions used for characterizing meal-timing have been used previously, as indicated above, the research team felt usage of the meal-timing definitions employed in this study were appropriate.

The Dietary Inflammatory Index (DII®) was calculated to quantify one’s dietary inflammatory potential. More negative scores indicate a more anti-inflammatory diet, while positive scores indicate a more pro-inflammatory diet. The DII was not a major focus of this work, so details on its calculation can be found elsewhere (Shivappa, Steck, Hurley, Hussey, & Hébert, 2014). Briefly, DII food parameters (micro and micronutrients and a few whole foods) were estimated from the 24HR. These were assigned article-effect scores for their association with inflammation. A global mean and standard deviation for these food parameters were derived for dietary data from 11 countries. The global mean was subtracted from the 24HR-derived values and divided by the global standard deviation. These resulting z-scores were converted to a percentile and centered by doubling the value and subtracting 1. These were then multiplied by the article-effect score and then summed across all food parameters to create the overall DII score. The global standard values and 24HR reported values were converted to a density format per 1,000 calories consumed. The greater the DII score, the more pro-inflammatory the diet; more negative values represent more anti-inflammatory diets. BI-RADS were collected through medical chart reviews and confirmed by a radiologist blinded to intervention assignment.

Non-fasting blood samples were collected at the same time as the anthropometric measures. CRP and high-sensitivity IL-6 were assayed using ELISA kits (R&D systems). Peripheral whole blood was collected on EDTA, centrifuged immediately (1,000g for 15 min), and stored on ice for transport back to the processing lab. Upon receipt in the lab, the samples were aliquoted, put on liquid nitrogen, and stored in a −80°C freezer for batching of samples until assay. Additional details can be found elsewhere (St-Onge et al., 2018). Although other markers of inflammation have been associated with chronic diseases, CRP and IL-6 are among the more commonly studied inflammatory markers and are relatively easily quantified (Ansar & Ghosh, 2013; Prasad, Sung, & Aggarwal, 2012). Generally, in epidemiological research, those with a CRP value > 10 mg/L are excluded from analyses of chronic outcomes, as values >10 mg/L are indicative of an acute infection (Ansar & Ghosh, 2013). Given the generally high values of CRP in this population of obese women, we did not consider it appropriate to employ 10 mg/L as a cut-point for exclusion as such an approach would have excluded a large percentage of the participants. Therefore, the decision was made to remove those with a CRP value greater than 30 mg/L (n=9) from the CRP analyses, as 30 mg/L still falls into the range considered for severe acute infection.

Statistical Analysis

Participants were dichotomized (yes/no) into groups based on whether they consumed 30% or more of their total calories after 5PM. A 30% cut-off was chosen based on the findings from Marinac and colleagues (Marinac et al., 2015). Chi-square tests and t-tests were used to test for significant differences of the covariates between these two groups and to assess the association between meal type (breakfast, lunch, and dinner), energy intake before and after 5PM, and various meal-time bins in relation to total kilocalories (kcal), macronutrients (fat, protein, and carbohydrates), and the DII. It should be noted that sample sizes within the aforementioned categorizations may not be the same as some of the participants did not eat during certain time-bins throughout the day. See Table 1 for more details on the sample sizes per time bin or meal type.

Table 1.

Percent calories from macronutrients, energy intake, and the DII according to meal timing categories, SISTAS Study, South Carolina, 2011–2015.

Meal and Timing Categories Sample Size Energy Intake (kcal) % kcal fat % kcal protein % kcal carb DII
Mean ± SD Mean % ± SD Mean % ± SD Mean % ± SD Mean ± SD
Meal Type
Breakfast 212 363.9 ± 240.4 27.9 ± 18.0 12.8 ± 7.6 59.4 ± 22.3 0.54 ± 2.30
Lunch 211 558.7 ± 362.6e 35.2 ± 15.4e 19.5 ± 12.4e 45.2 ± 19.5e 0.76 ± 2.29
Dinner 240 601.1 ± 344.4e 34.8 ± 14.5e 21.1 ± 12.0e 42.9 ± 17.5e 0.75 ± 2.33
Before or After 5PM
Before 248 927.3 ± 499.6 33.8 ± 12.4 15.9 ± 7.1 50.2 ± 15.0 0.62 ± 2.49
After 242 701.9 ± 433.3f 35.1 ± 15.3 18.2 ± 10.6f 45.3 ± 18.2f 0.77 ± 2.37
Mealtime Bins
4AM – 9:59AM 166 346.4 ± 242.8 26.7 ± 18.3 12.6 ± 7.4 60.7 ± 22.3 0.38 ± 2.16
10AM – 2:59PM 223 572.9 ± 364.2a 33.3 ± 15.7a 16.7 ± 10.9a 49.9 ± 19.5a 0.71 ± 2.47
3PM – 7:59PM 211 673.8 ± 443.2a,b 33.6 ± 15.1a 18.8 ± 11.5a,b 46.6 ± 18.8a 0.70 ± 2.31
8PM – 3:59AM 159 455.2 ± 362.9a,b,c 33.2 ± 19.4a 14.5 ± 12.3c 51.0 ± 24.9a 0.91 ± 2.25a
Open in a new tab

Abbreviations used: carb, carbohydrate; DII, Dietary Inflammatory Index; kcal, kilocalories; SD, standard deviation.

Mean values were significantly different (p<0.05) as indicated by the following mealtime bins:

a

Significant difference from 4AM – 9:59AM,

b

Significant difference from 10AM – 2:59PM,

c

Significant difference from 3PM – 7:59PM,

d

Significant different from 8PM – 3:59AM

Mean values were significantly different from breakfast:

e

p<0.05

Sample sizes are variable as not all participants consumed during eat of the defined periods. The “Before” row values represent the respective column values only for foods consumed before 5PM. Conversely, the “After” row values represent the respective column values only for foods consumed after 5PM.

Mean values were significantly different compared to before 5PM: p<0.05

Least square (i.e., multivariable-adjusted) means for CRP and IL-6 by the exposure measures were obtained through linear regression. Exposures included the definition of calories consumed after 5PM noted in the previous paragraph. Additionally, individuals were assigned to the meal-type (breakfast, lunch, dinner) that contributed the most to each of the following factors: kilocalories, protein, carbohydrates, fat, and DII score. For example, if a participant consumed 50% of her kcals from carbohydrates at breakfast, 30% at lunch, and 20% at dinner, she was classified into the “breakfast” group for the meal that contributed the most to their daily carbohydrate intake. Additionally, for the meal-type classification, we restricted analyses to just those reporting consumption of all three major meals (i.e., breakfast, lunch, and dinner). Given only a single 24HR was used, it was difficult to determine if those only eating one or two of the major meals typically consume food in such a pattern or if these were outlier days in which meals were missed or skipped. Therefore, the decision was made to include only those reporting all three of the major meals. Confounder selection was based on a 10% change in the beta coefficient of the primary exposure for each model (full vs. parsimonious model). All models were checked for violations of linear regression, and none were found.

Logistic regression was used to assess the effects of percent total calorie consumption after 5PM (as a continuous metric) in relation to BI-RAD readings. BI-RAD readings were obtained from 42 of the participants and dichotomized (negative/benign and probably benign/suspicious abnormality). These analyses were conducted utilizing SAS (version 9.4, Cary, NC). Given only baseline data were used, missing data was minimal. No additional procedures were undertaken to account for baseline missing data which was assumed to be missing at random. Lastly, given the exploratory nature of the study, adjustments for multiple comparisons were not performed.

Results

Baseline characteristics have been described for this population and can be found elsewhere.[19] Briefly, average age, body fat percentage, and body mass index (BMI) of the participants was 50.2±11.4 years, 46.0±6.0%, and 39.3±7.9 kg/m2 respectively. About 73% of the 249 women consumed more than 30% of their calories after 5:00PM. There were no meaningful differences between those consuming more or those consuming less than 30% of calories after 5PM for anthropometric or demographic measures (data not tabulated).

Participants had a higher energy intake at both lunch and dinner compared to breakfast (p≤0.05, Table 1). The highest average energy, fat, and protein intake was during 3:00PM–7:59PM time bin. However, carbohydrate intake was highest in the morning (4:00AM–9:59AM). The most pro-inflammatory DII score was observed during 8:00PM–3:59AM.

The adjusted CRP and IL-6 values were not statistically different when comparing the meal types (i.e., breakfast, lunch, and dinner) with the highest energy consumption or the meal-types with the most pro-inflammatory DII value (Table 2, below this paragraph). As for fat intake, those whose breakfast contributed the most to their fat intake compared to dinner, appeared to have higher CRP values (8.23 vs. 6.30 mg/L, p=0.09); i.e., this was not statistically significant. A similar relationship was observed when comparing fat intake at lunch compared to dinner (8.07 vs. 6.30, mg/L, p=0.09). The inverse occurred for carbohydrate intake; participants who consumed the most carbohydrates at breakfast had a significantly lower CRP compared to those with their highest consumption at dinner (6.99 vs. 9.56 mg/L, p=0.02, respectively). A similar relationship was observed when comparing lunch to dinner (6.79 vs. 9.56 mg/L, p=0.05, respectively). If considering adjustment for multiple comparisons (i.e., 3 possible comparisons between breakfast, lunch, and dinner), neither of these two statistically significant results would retain statistical significance.

Table 2.

Adjusted mean CRP and IL-6 by meal types with the highest intake from each category, SISTAS Study, South Carolina, 2011–2015.

Inflammatory Markers Breakfast Lunch Dinner P-values
Mean (95% CI) Mean (95% CI) Mean (95% CI) B vs. L B vs. D L vs. D
Meal type with the highest calories (kcal)
n 18 72 83
CRP (mg/L)a 6.27 (3.41–9.14) 7.17 (5.82–8.50) 8.04 (6.77–9.32) .58 .27 .35
IL-6 (pg/mL)b 2.55 (1.75–3.36) 2.73 (2.33–3.13) 2.80 (2.42–3.18) .70 .58 .79
Meal type with the highest Dietary Inflammatory Index
n 54 60 59
CRP (mg/L)a 8.15 (6.54–9.75) 7.41 (5.93–8.89) 7.03 (5.53–8.53) .51 .32 .72
IL-6 (pg/mL)b 2.53 (2.06–3.00) 2.88 (2.45–3.32) 2.80 (2.36–3.32) .25 .39 .79
Meal type with the highest fat intake
n 49 60 64
CRP (mg/L)a 8.23 (6.59–9.89) 8.07 (6.63–9.50) 6.30 (4.82–7.78) .88 .09 .09
IL-6 (pg/mL)b 2.62 (2.14–3.10) 2.85 (2.42–3.28) 2.75 (2.31–3.20) .47 .67 .75
Meal type with highest protein intake
n 23 64 89
CRP (mg/L)a 7.38 (4.92–9.82) 7.49 (6.24–8.74) 7.56 (6.11–9.01) .90 .93 .95
IL-6 (pg/mL)b 2.24 (1.54–2.93) 3.04 (2.63–3.45) 2.66 (2.29–3.03) .05 .28 .16
Meal type with highest carbohydrate intake
n 103 33 37
CRP (mg/L)a 6.99 (5.87–8.10) 6.79 (4.76–8.82) 9.56 (7.69–11.43) .87 .02 .05
IL-6 (pg/mL)b 2.77 (2.43–3.11) 2.50 (1.92–3.08) 2.91 (2.36–3.46) .42 .66 .30
Open in a new tab

Abbreviations used: B, breakfast; CI, confidence interval; CRP, C-reactive protein; DII, Dietary Inflammatory Index; D, dinner; IL-6, interleukin-6; kcal, kilocalories; L, lunch; mg/L, milligrams per liter; pg/mL, picograms per milliliter

a

Adjusted for body mass index and age.

b

Adjusted for aspiring use, age, body mass index, and diastolic blood pressure.

Note: Individuals were assigned to the meal (i.e., breakfast, lunch, or dinner) that contributed the highest number of kcal, DII, fat intake, protein intake, or carbohydrate intake. For example, if participate X consumed 500, 1000, and 700 kcals at breakfast, lunch, and dinner respectively, then this individual was assigned to the lunch group. A similar process was repeated for each category (i.e., kcal, DII, etc.).

P-values were estimated using multivariable linear regression with calculation of least squares means.

CRP and IL-6 concentrations were not statistically significant when comparing the mean inflammatory values between participants who did or did not consume 30% or more of their total energy after 5PM (data not tabulated). Among the 42 women with BI-RAD readings, 30 (71%) had a negative/benign reading and 12 (29%) had, at least, a suspicious reading. Each one-percentage increase in calories consumed after 5PM was significantly associated with a suspicious BI-RAD reading (versus benign) (OR=1.053, 95% CI=1.003–1.105) after adjustment for age, BMI, self-efficacy for diet, and family history of breast cancer (data not tabulated).

Discussion

This study suggests that consumption of energy, specifically carbohydrates, later in the day (e.g., at dinner) compared to earlier in the day (e.g., at breakfast) may be associated with increased inflammatory cytokines. Also, consuming more energy after 5PM was associated with increased odds of potential breast abnormalities. As noted in the Introduction, evidence for the association between chrononutrition-related constructs and inflammation is inconsistent across studies (Bhutani et al., 2013; Marinac et al., 2016, 2015; Sutton et al., 2018; Trepanowski et al., 2018). In addition to that, representation of African Americans in such research is limited. Hence, novelty of this study includes analysis of meal-timing, distribution of macronutrients, and inflammation among African-American women. The findings from this study are consistent with previous literature, indicating that a greater proportion of energy and macronutrient intake is consumed at later times in the day in a US population (Almoosawi et al., 2016). While not indicated in this study, higher evening energy intake has been associated with higher concentrations of CRP, which contributes to an elevated risk of breast cancer and other chronic diseases (Marinac et al., 2015). Specifically, our findings tend to be in line with other research indicating that consuming a majority of calories earlier in the day or eTRF protocols are not associated with lower levels of pro-inflammatory cytokines (Almoosawi et al., 2016; Kessler et al., 2020; McAllister, Pigg, Renteria, & Waldman, 2020; Sutton et al., 2018; Trepanowski et al., 2018; Wilkinson et al., 2020). It must be acknowledged that many of these were clinical trials of shorter duration; still, it is conceivable that they were not of sufficient duration to produce changes in systemic inflammation. Our study design and definitions align with work from Marinac and colleagues whose study included adult women from the National Health and Nutrition Examination Study (NHANES). In the study conducted by Marinac and colleagues, every 10% increase in the proportion of calories consumed after 5PM was associated with a 3% higher CRP value (p=0.02) (Marinac et al., 2015). The average CRP values of women in the current study was much higher than the study by Marinac and colleagues. It is possible that other processes related to the obesity (which was a requirement of study entry into SISTAS) was driving the CRP values more so than meal-timing would.

Although evening energy consumption was not found to be associated with CRP and IL-6, an increase in energy consumption after 5PM was associated with a BI-RAD reading of probably benign or suspicious abnormalities. This fits somewhat with previous research. For example, previous work has shown that higher energy consumption in the evening was associated with an increased risk in breast cancer (Marinac et al., 2015). Using data from the Multicase-Control Study in Spain (MCC-Spain), the adjusted OR for combined prostate and breast cancer for consuming supper before 9PM compared to later than 10PM was 0.82 (95%CI=0.67–1.00) (Kogevinas et al., 2018). Previous findings, plus the ones contributed by the current paper, provide evidence for the potential impact of meal-timing on breast cancer. However, due to the nature of the current study design we cannot assess the temporality of this association.

Exploratory analyses were undertaken to examine the impact of macronutrient intake by meal type (i.e., breakfast, lunch, and dinner). Our results indicate that a greater consumption of carbohydrates consumed during breakfast or lunch rather than dinner may play an important role in inflammatory control. There are possible biological rationales for this. Greater carbohydrate intake early in the day, has been associated with both an overall lower energy intake throughout the day and lower rates of obesity (Almoosawi et al., 2016; de Castro, 2004). Carbohydrate consumption results in an increase in blood glucose levels, causing insulin to be secreted to essentially lower glucose levels (Sutton et al., 2018). In congruence with circadian rhythms found in metabolism, insulin sensitivity tends to be highest in the morning; thus, the body will more readily produce insulin when consuming carbohydrates in the morning compared to the evening (Sutton et al., 2018). With our population comprising of women with obesity, there may already be a low prevalence of insulin sensitivity. However, this was not testable in the current study. Therefore, consuming carbohydrates outside the peak insulin sensitivity hours (in the evening) coupled with an already decreased insulin sensitivity, it is plausible that a higher inflammatory response would occur in response to chronic high levels of glucose in the blood (Stafeev, Vorotnikov, Ratner, Menshikov, & Parfyonova, 2017). It should be noted that age, BMI and total energy intake did not differ across meal types that contributed the most to carbohydrate intake. Hence, these important biological factors do not explain the observed associations. Interestingly, the inverse was true of fat intake, with a greater consumption of fat in the morning resulting in a higher CRP concentration. It is unclear how increased fat intake in the morning compared to at night would be associated with higher levels of inflammation. However, a previous study found an association between high fat consumption at breakfast and an increase in daily fat consumption,(Holt, Delargy, Lawton, & Blundell, 1999) providing a plausible explanation for this finding. As with what was observed for carbohydrates, age, BMI, and total energy did not differ by the meal type that contributed the most to fat intake.

Despite its strengths, this research has limitations. Greater energy intake, as well as late night eating, have both been associated with obesity (Almoosawi et al., 2016). Our study population was comprised of women with obesity who also had very high levels of inflammation. These factors may have contributed to the lack of statistically significant differences seen within this study population. Excess body weight is associated with chronic inflammation and is typically linked with unhealthy lifestyle behaviors, such as poor dietary habits (Pitsavos et al., 2007). However, age, BMI, and total energy intake were either adjusted for or were not statistical confounders. Regardless, the homogeneous population (i.e., obese African-American women) limit generalizability to other populations. In order to reduce participant burden, a major limitation in this study was the use of a single 24-hour dietary recall as this may be unable to capture the habitual intake of our participants (Hebert et al., 2002). A single 24HR is subject to daily fluctuation, thus dietary information on a single day may provide inaccurate estimates of usual dietary intake (Basiotis, Welsh, Cronin, Kelsay, & Mertz, 1987; Ma et al., 2009). Future work should consider using multiple 24HRs to account for inter-person variability and to reduce total error that is induced when using a single dietary recall. However, it should be noted that a single 24HR only has slightly lower variance than food frequency questionnaires (Hebert, Chen Backlund, Engle, Barone, & Biener, 1990; Hebert et al., 2002) which are more commonly used in epidemiological research. Additionally, the sample size for analyses involving BI-RADS was small and only 12 participants had the “outcome” of interest. It is worth noting, that despite limitations, the objective measures used to assess inflammation (CRP, IL-6, DII) and breast cancer risk (BI-RADS) increased the validity of this study. Additionally, this study included African Americans who are an underrepresented population in research and in chrononutrition research in particular. Previous work using NHANES data found that when comparing meal timing by race, breakfast was 28 minutes earlier among European Americans than African Americans, but lunch and dinner were more comparable.(Kant, 2018) Understanding how chrononutrition impacts health differently by race may inform future research in chrononutrition or clinical practice.

Conclusions

Findings from the present suggest the importance of restricting energy consumption, specifically carbohydrate intake, to certain time windows as a possible means to reduce chronic inflammation. Additionally, evidence presented in this report indicates a potential association between energy consumption later in the evening and breast abnormalities. However, due to the cross-sectional nature of this study a temporal sequence cannot established. The literature has shown that individuals who consume a higher carbohydrate breakfast tend to have a lower overall BMI (Ansar & Ghosh, 2013; de Castro, 2004). In combination with findings in the extant literature, this study suggests that altering timing of macronutrient intake is likely to provide benefits for reducing both obesity and chronic inflammation and, thus, reduce breast cancer risk. Further studies are needed to capture the habitual meal patterns to fully assess the relation of meal timing and composition with inflammation.

Table 3.

Adjusted mean CRP and IL-6 (95% confidence interval) by consumption of <30% vs. ≥30% of total energy (kcal) after 5PM, SISTAS Study, South Carolina, 2011–2015.

Inflammatory Markers Consumed <30% kcal after 5PM Consumed ≥30% kcal after 5PM P-value
<30% vs. ≥30%
n 68 181
CRP (mg/L)a 8.16 (6.76–9.57) 7.22 (6.33–8.10) 0.26
IL-6 (pg/mL)b 2.46 (2.05–2.86) 2.76 (2.51–3.01) 0.20
Open in a new tab

Abbreviations used: CRP, C-reactive protein; IL-6, interleukin-6; kcal, kilocalories; mg/L, milligrams per liter; pg/mL, picograms per milliliter

a

Adjusted for body mass index and age.

b

Adjusted for aspiring use, age, body mass index, and diastolic blood pressure.

P-values were estimated using multivariable linear regression with calculation of least squares means.

Funding details:

This research was funded by Susan G. Komen®, grant number GTDR17500160, and by The South Carolina Cancer Disparities Community Network Projected awarded by the National Institute of Health, grant number 1U54CA153461-01.

Footnotes

Disclosure statement:

Dr. Hébert owns controlling interest in Connecting Health Innovations LLC (CHI), a company that has licensed the right to his invention of the Dietary Inflammatory Index (DII®) from the University of South Carolina in order to develop computer and smart phone applications for patient counseling and dietary intervention in clinical settings. Dr. Michael Wirth is an employee of CHI.

References

  1. Adams P (2013). The breast cancer conundrum. Bulletin of the World Health Organization, 91(9), 626–627. 10.2471/BLT.13.020913 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Almoosawi S, Vingeliene S, Karagounis LG, & Pot GK (2016). Chrono-nutrition: A review of current evidence from observational studies on global trends in time-of-day of energy intake and its association with obesity. Proceedings of the Nutrition Society, 75(4), 487–500. 10.1017/S0029665116000306 [DOI] [PubMed] [Google Scholar]
  3. Ansar W, & Ghosh S (2013). C-reactive protein and the biology of disease. Immunol Res, 131–142. 10.1007/s12026-013-8384-0 [DOI] [PubMed] [Google Scholar]
  4. Asher G, & Sassone-Corsi P (2015). Time for food: The intimate interplay between nutrition, metabolism, and the circadian clock. Cell, 161(1), 84–92. 10.1016/j.cell.2015.03.015 [DOI] [PubMed] [Google Scholar]
  5. Bahijri S, Borai A, Ajabnoor G, Abdul Khaliq A, AlQassas I, Al-Shehri D, & Chrousos G (2013). Relative Metabolic Stability, but Disrupted Circadian Cortisol Secretion during the Fasting Month of Ramadan. PLOS ONE, 8(4), 1–6. 10.1371/journal.pone.0060917 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Basiotis PP, Welsh SO, Cronin FJ, Kelsay JL, & Mertz W (1987). Number of Days of Food Intake Records Required to Estimate Individual and Group Nutrient Intakes with Defined Confidence. The Journal of Nutrition, 117(9), 1638–1641. 10.1093/jn/117.9.1638 [DOI] [PubMed] [Google Scholar]
  7. Bevel M, Babatunde OA, Heiney SP, Brandt HM, Wirth MD, Hurley TG, … Adams SA (2018). Sistas Inspiring Sistas Through Activity and Support (SISTAS): Study Design and Demographics of Participants. Ethnicity & Disease, 28(2), 75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bhutani S, Klempel MC, Kroeger CM, Trepanowski JF, & Varady KA (2013). Alternate day fasting and endurance exercise combine to reduce body weight and favorably alter plasma lipids in obese humans. Obesity, 21(7), 1370–1379. 10.1002/oby.20353 [DOI] [PubMed] [Google Scholar]
  9. de Castro JM (2004). The time of day of food intake influences overall intake in humans. The Journal of Nutrition, 134(1), 104–111. 10.1093/jn/134.1.104 [DOI] [PubMed] [Google Scholar]
  10. Froy O, & Miskin R (2010). Effect of feeding regimens on circadian rhythms: implications for aging and longevity. Aging, 2(1), 7–27. 10.18632/aging.100116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Guo L, Liu S, Zhang S, Chen Q, Zhang M, Quan P, … Sun X (2015). C-reactive protein and risk of breast cancer: A systematic review and meta-analysis. Science Reports. 10.1038/srep10508 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hebert J, Chen Backlund JY, Engle A, Barone J, & Biener K (1990). Intra and inter-person sources of variability in fat intake in a feeding trial of 14 men. European Journal of Epidemiology, 6(1), 55–60. 10.1007/BF00155550 [DOI] [PubMed] [Google Scholar]
  13. Hebert J, Ebbeling C, Matthews C, Hurley T, Ma Y, Druker S, & Clemow L (2002). Systematic Errors in Middle-Aged Women’s Estimates of Energy Intake: Comparing Three Self-Report Measures to Total Energy Expenditure from Doubly Labeled Water. Annals of Epidemiology, 12(8), 577–586. [DOI] [PubMed] [Google Scholar]
  14. Holt SHA, Delargy HJ, Lawton CL, & Blundell JE (1999). The effects of high-carbohydrate vs high-fat breakfasts on feelings of fullness and alertness, and subsequent food intake. International Journal of Food Sciences and Nutrition, 50(1), 13–28. [DOI] [PubMed] [Google Scholar]
  15. Kant AK (2018). Eating patterns of US adults: Meals, snacks, and time of eating. Physiology & Behavior, 193, 270–278. 10.1016/j.physbeh.2018.03.022 [DOI] [PubMed] [Google Scholar]
  16. Keim NL, Van Loan MD, Horn WF, Barbieri TF, & Mayclin PL (1997). Weight loss is greater with consumption of large morning meals and fat- free mass is preserved with large evening meals in women on a controlled weight reduction regimen. Journal of Nutrition, 127(1), 75–82. 10.1093/jn/127.1.75 [DOI] [PubMed] [Google Scholar]
  17. Kessler K, Hornemann S, Rudovich N, Weber D, Grune T, Kramer A, … Pivovarova-Ramich O (2020). Saliva Samples as A Tool to Study the Effect of Meal Timing on Metabolic And Inflammatory Biomarkers. Nutrients, 12(2), 340. 10.3390/nu12020340 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kogevinas M, Espinosa A, Castelló A, Gómez-Acebo I, Guevara M, Martin V, … Romaguera D (2018). Effect of mistimed eating patterns on breast and prostate cancer risk (MCC-Spain Study). International Journal of Cancer, 143(10), 2380–2389. 10.1002/ijc.31649 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ma Y, Olendzki BC, Pagoto SL, Hurley TG, Magner RP, Ockene IS, … Hébert JR (2009). Number of 24-Hour Diet Recalls Needed to Estimate Energy Intake. Annals of Epidemiology, 19(8), 553–559. 10.1016/j.annepidem.2009.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Marinac CR, Nelson SH, Breen CI, Hartman SJ, Natarajan L, Pierce JP, … Patterson RE (2016). Prolonged Nightly Fasting and Breast Cancer Prognosis. JAMA Oncology, 2(8), 1049. 10.1001/jamaoncol.2016.0164 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Marinac CR, Sears DD, Natarajan L, Gallo LC, Breen CI, & Patterson RE (2015). Frequency and Circadian Timing of Eating May Influence Biomarkers of Inflammation and Insulin Resistance Associated with Breast Cancer Risk. PLOS ONE, 10(8), e0136240. 10.1371/journal.pone.0136240 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. McAllister MJ, Pigg BL, Renteria LI, & Waldman HS (2020). Time-restricted feeding improves markers of cardiometabolic health in physically active college-age men: a 4-week randomized pre-post pilot study. Nutrition Research, 75, 32–43. 10.1016/j.nutres.2019.12.001 [DOI] [PubMed] [Google Scholar]
  23. Oh E-Y, Yang X, Friedman A, Ansell CM, Du-Quiton J, Quiton DF, … Hrushesky WJM (n.d.). Circadian transcription profile of mouse breast cancer under light-dark and dark-dark conditions. Cancer Genomics & Proteomics, 7(6), 311–322. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21156964 [PubMed] [Google Scholar]
  24. Pitsavos C, Panagiotakos DB, Tzima N, Lentzas Y, Chrysohoou C, Das UN, & Stefanadis C (2007). Diet, exercise, and C-reactive protein levels in people with abdominal obesity: The ATTICA epidemiological study. Angiology, 58(2), 225–233. 10.1177/0003319707300014 [DOI] [PubMed] [Google Scholar]
  25. Prasad S, Sung B, & Aggarwal BB (2012). Age-associated chronic diseases require age-old medicine: Role of chronic inflammation. Preventive Medicine, 54, S29–S37. 10.1016/j.ypmed.2011.11.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ribas-Latre A, & Eckel-Mahan K (2016). Interdependence of nutrient metabolism and the circadian clock system: Importance for metabolic health. Molecular Metabolism, 5(3), 133–152. 10.1016/j.molmet.2015.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ruddick‐Collins LC, Morgan PJ, & Johnstone AM (2020). Mealtime: A circadian disruptor and determinant of energy balance? Journal of Neuroendocrinology, 32(7). 10.1111/jne.12886 [DOI] [PubMed] [Google Scholar]
  28. Scheer Fr. A. J., Hilton MF, Mantzoros Ch. S., & Shea SA (2009). Adverse metabolic and cardiovascular consequences of circadian misalignment. Proceedings of the National Academy of Sciences, 106(11), 4453–4458. 10.1073/pnas.0808180106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Shivappa N, Steck SE, Hurley TG, Hussey JR, & Hébert JR (2014). Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutrition, 17(8), 1689–1696. 10.1017/S1368980013002115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. St-Onge M-P, Pizinger T, Kovtun K, & Roychoudhury A (2018). Sleep and meal timing influence food intake and its hormonal regulation in healthy adults with overweight/obesity. European Journal of Clinical Nutrition. 10.1038/s41430-018-0312-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Stafeev IS, Vorotnikov AV, Ratner EI, Menshikov MY, & Parfyonova YV (2017). Latent Inflammation and Insulin Resistance in Adipose Tissue. International Journal of Endocrinology, 2017. 10.1155/2017/5076732 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Stevens RG, Brainard GC, Blask DE, Lockley SW, & Motta ME (2014). Breast cancer and circadian disruption from electric lighting in the modern world. CA: A Cancer Journal for Clinicians, 64(3), 207–218. 10.3322/caac.21218 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sutton EF, Beyl R, Early KS, Cefalu WT, Ravussin E, & Peterson CM (2018). Early Time-Restricted Feeding Improves Insulin Sensitivity, Blood Pressure, and Oxidative Stress Even without Weight Loss in Men with Prediabetes. Cell Metabolism, 27(6), 1212–1221.e3. 10.1016/j.cmet.2018.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Trepanowski JF, Kroeger CM, Barnosky A, Klempel M, Bhutani S, Hoddy KK, … Varady KA (2018). Effects of alternate-day fasting or daily calorie restriction on body composition, fat distribution, and circulating adipokines: Secondary analysis of a randomized controlled trial. Clinical Nutrition, 37(6), 1871–1878. 10.1016/j.clnu.2017.11.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Wilkinson MJ, Manoogian ENC, Zadourian A, Lo H, Fakhouri S, Shoghi A, … Taub PR (2020). Ten-Hour Time-Restricted Eating Reduces Weight, Blood Pressure, and Atherogenic Lipids in Patients with Metabolic Syndrome. Cell Metabolism, 31(1), 92–104.e5. 10.1016/j.cmet.2019.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Yang X, Wood PA, Oh E-Y, Du-Quiton J, Ansell CM, & Hrushesky WJM (2009). Down regulation of circadian clock gene Period 2 accelerates breast cancer growth by altering its daily growth rhythm. Breast Cancer Research and Treatment, 117(2), 423–431. 10.1007/s10549-008-0133-z [DOI] [PubMed] [Google Scholar]
  37. You S, Wood PA, Xiong Y, Kobayashi M, Du-Quiton J, & Hrushesky WJM (2005). Daily coordination of cancer growth and circadian clock gene expression. Breast Cancer Research and Treatment, 91(1), 47–60. 10.1007/s10549-004-6603-z [DOI] [PubMed] [Google Scholar]

RESOURCES