Abstract
Background/Objectives
A potential risk factor for cardio-metabolic diseases is irregular or inconsistent eating, however, research on this topic is scarce. We aimed to study associations between irregular consumption of energy intake in meals and cardio-metabolic risk factors.
Subject and Methods
Dietary intake data were derived from 5-day estimated diet diaries of 1768 participants of the National Survey of Health and Development (NSHD). Energy intakes during predefined meals (breakfast, lunch, dinner, between meals) and daily totals were analysed using a score for irregularity based on the deviation from the 5-day mean energy intake. Logistic regression models were used to calculate odds ratios (ORs) and 95% confidence intervals (CI) for having the metabolic syndrome or one of its components. Models were adjusted for sex, physical activity, socio-economic status, marital status, and smoking.
Results
Irregularity scores of energy intake ranged from 0-160 and were highest for between meals. An increased risk of the metabolic syndrome was associated with more irregular energy intake during breakfast (OR 1.34 (0.99-1.81); p trend 0.04) and between meals (OR 1.36 (1.01-1.85); p trend 0.04). Moreover, increased waist circumference was associated with irregular energy intake during breakfast (OR 1.90 (1.47; 2.45); p trend <0.01), evening meal (OR 1.36 (1.06; 1.75); p trend 0.02), and daily total (OR 1.34 (1.04; 1.72); p trend 0.01). No significant associations were found for the other components of the metabolic syndrome.
Conclusions
Individuals with a more irregular intake of energy, especially during breakfast and between meals, appeared to have an increased cardio-metabolic risk.
Keywords: meal patterns, irregularity, energy intake, cardio-metabolic risk, obesity, NSHD
Introduction
An eating behaviour that has emerged recently as a potential risk factor for chronic disease risk is irregular or inconsistent eating 1. The ready availability of food, particularly prepared food, along with the move from family meals at set times, has led to more irregular eating than in the past 2, 3. Food intake is determined and affected by many factors, one of which is time of day or chronobiology. Chronobiology refers to time-dependent variations in biological functions 4 and includes three components of time: (i) clock time (time of day); (ii) frequency (e.g. events per time span); and (iii) regularity, events at specific times. This research focusses on the third aspect of time: on regularity, or more specifically the disruption of regularity: irregularity. Surprisingly, only a few observational studies have investigated the association between meal irregularity and BMI, the metabolic syndrome or cardiovascular risk and those that do were focussed on regular meal frequency showing conflicting results; Sierra-Johnson et al found in a group of 60 year old men and women in Sweden that eating meals regularly is inversely associated with the metabolic syndrome, insulin resistance, and serum concentrations of γ- transferase 5. Shin et al reported that consuming a variable number of meals per day was not associated with the metabolic syndrome compared to regularly eating three times a day6. These observational studies assessed irregularity of meal frequency by questionnaire rather than using detailed data from dietary assessment to investigate the impact of intake that differs from day to day.
Evidence from intervention trials is scarcer but that available showed beneficial effects of eating regular meals on cardio-metabolic risk factors. Farshchi et al investigated the effect of a regular vs. irregular meal pattern on dietary thermogenesis, insulin sensitivity and fasting lipid profiles in lean 7 and obese women 8 in randomized controlled intervention studies. The irregular meal pattern had deleterious effects on insulin sensitivity and plasma cholesterol, which are known risk factors for CVD 8.
More research has been conducted on other aspects of chronobiological eating behaviour, such as the timing of eating occasions, which refers to the actual clock time. A recent study by Almoosawi et al has shown that the timing of energy and nutrient intake has shifted slightly over 17 years in the Medical Research Council (MRC) National Survey of Health and Development (NSHD), also known as the 1946 British Birth Cohort, with a greater proportion of intake later in the day in more recent years 9. Almoosawi et al also showed that the time of day when specific nutrients were consumed was associated with risk for the metabolic syndrome 10 as well as diabetes in this cohort 11. We aimed to study the irregular consumption of energy intake in and between meals in relation to cardio-metabolic disease risk using detailed dietary data from NSHD.
Subjects and Methods
Study members
The NSHD is a longitudinal study based on a social class-stratified sample of 5632 singleton births occurring within marriage in England, Scotland, and Wales during one week in March 1946 12, 13. Data have been collected in this cohort on more than 20 occasions across the life course; adult dietary data were collected when participants were aged 36 (1982), 43 (1989), 53 (1999) and 60-64 years (2006-11). For these analyses dietary data when cohort members were 53 years were included as at this time point data was available for both dietary intake and metabolic syndrome characteristics in a relatively large sample (n=1768).
This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the North Thames Multicentre Research Ethics Committee at age 53 years. Written informed consent was obtained from all subjects.
Dietary assessment and non-dietary variables
Dietary data were collected by research nurses providing food diaries during home visits to be completed over five consecutive days14. All food and drinks consumed both at home and away were recorded using household measures, and portion sizes were estimated using detailed guidance notes and photographs provided at the beginning of the diary14. Cohort members were asked to complete the diet diary in the days following the home visit and encouraged to complete three weekdays and two weekend days. Diet diaries were pre-structured and contained the following predefined meal slots: first thing, breakfast, mid-morning, lunch, tea, evening meal, late evening, and extras. The meal slots for first thing, mid-morning, tea, late evening and extras were collapsed into one meal slot for ‘between meals’. Diaries were coded at the Dunn Nutrition Unit in Cambridge by a team with extensive experience in coding and analysing dietary data using the program Diet In Data Out (DIDO)15, a dietary assessment system developed in house and incorporating McCance and Widdowson’s ‘The composition of foods’ 4th edition 16, and its supplements 17, 18. Only those diet diaries with 4 or more recording days were included in these analyses.
Information on demographic and socio-economic variables at age 53 years, including sex, socio-economic status (SES: no job, professional, intermediate, skilled non-manual, skilled manual, partly skilled, unskilled, armed forces), marital status (married, single, divorced/widowed/separated), physical activity (categories: none, 1-4 times per month, 5 or more times per month of sports or recreational activity), and smoking (current, ex-smoker, never smoker) were obtained through structured questionnaires 12.
Calculating scores for irregularity of meals based on variability in energy intake
To calculate a score for meal irregularity, energy intakes variance per meal were used as a proxy. To calculate the irregularity scores per predefined meal, the absolute difference of the individual energy intake from the 5-day mean energy intake of that predefined meal was divided by the 5-day mean energy intake of that predefined meal slot, multiplied by 100 and then averaged over the 5-days; this served as a measure of irregularity of energy intake for that predefined meal, with a low score indicating a more regular meal pattern of energy intake and a higher score indicating more irregularity of energy intake. This was done for each predefined meal (breakfast, lunch, evening meal and between meals) as well as for the daily total. The daily total meal irregularity score of energy intake was based on the deviation from total daily individual energy intake compared with the 5-day mean total daily energy intake calculated the same way as described above.
The analyses were performed for intake of energy in kcal. The daily total energy intake was also expressed as the percentage of estimated energy requirements (EAR) for which a population approach was used: for men, the current EAR is 2605 kcal/d and for women 2079 kcal/d for an average BMI of 22.5 kg/m2 and an average physical activity level (PAL) of 1.63 19 as the measure of physical activity levels collected at age 53 years was not precise enough to use individual levels.
Cardio-metabolic risk factors
Measures at age 53 years were taken during a home visit by a team of trained research nurses and included the following: height, weight, waist circumference, and blood pressure (BP) measured according to standard protocols. A non-fasting venous blood sample was taken from which HDL cholesterol, and triglycerides and HbA1c were measured12.
The definition of the metabolic syndrome and its components were based on a modified version of the Adult Treatment Panel III (ATP-III) definition of the metabolic syndrome 20. This was met when ≥3 of the following criteria were present: waist circumference ≥102cm in men and ≥88cm in women; HDL cholesterol <1.036mmol/l (40mg/dl) in men and <1.295mmol/l (50mg/dl) in women or specific treatment for this lipid abnormality; triglycerides ≥1.7mmol/l (150mg/dl) in men and women or specific treatment for this lipid abnormality; and systolic BP ≥130mm Hg or diastolic BP ≥85mm Hg in men and women or use of anti-hypertensive medication; and Hb1Ac level in the top gender-specific quarter of the distribution (>5.8% among men and women) or use of medication for diabetes. The latter was a modification of the ATPIII criteria since fasting blood glucose was not available for this population at that time. Data for the metabolic syndrome was missing for 366 individuals and this was mainly due to missing data for HDL cholesterol (n=361). BMI is strictly speaking not a component of the metabolic syndrome under this definition, but was also included as outcome in this study.
Statistical analyses
Characteristics of those with a more irregular daily energy intake were compared to those with a more regular daily energy intake which was based on tertiles of the total daily regularity score of energy intake; to compare continuous characteristics across tertiles ANOVA was used and a chi-square test was used for categorical data.
The main outcome of this study was the risk of having the metabolic syndrome. Therefore, tertiles of irregularity scores of energy intake were entered into binary logistic regression models that calculated odds ratios (ORs) and 95% confidence intervals (CI) for having the metabolic syndrome or one of the ATP-III metabolic syndrome components or BMI; this was done for each meal as well as for the daily total irregularity score of energy intake. All models were adjusted for sex, socio-economic status (SES), marital status, physical activity, and smoking as confounding variables (model 1). A second model was run further adjusting for energy intakes during the other meals and daily total irregularity scores of energy intake were adjusted for EAR to overcome the possible effects of regular overconsumption of energy; for this, energy intake from all other meals were used but the one in the model to avoid possible over-adjustment in the model (i.e. the model for irregularity score for breakfast was adjusted for energy intake from lunch, evening meal and between meals). Effect modification by sex was evaluated by testing the coefficient of the interaction term. To test for linear trends across tertiles, each individual was assigned the median intake scores of their respective tertile and the resulting variable was fitted as a continuous variable in the model. The continuous components of the metabolic syndrome, waist circumference, BP (systolic and diastolic), HDL cholesterol, triglycerides, and BMI were further explored using multiple linear regression models adjusting for the same covariates as above.
To assess the possible impact of under- or over-reporting of energy intake sensitivity analyses were included categorizing cohort members as under-reporters or over-reporters based on the methods described by McCrory and using 2SD as cut-off points (thus as lower limit 59% and upper limit 141%) 21 and repeating the logistic regression models for plausible reporters only.
Data analysis was carried out using SPSS for MS Windows 20.0 (SPSS Inc, Chicago, IL) and a p value of <0.05 was considered statistically significant.
Results
Describing the sample
Cohort members with a more irregular total daily pattern of energy intake had a higher waist circumference and BMI (p for trend <0.05) and reported lower intakes of energy, carbohydrate, protein (only for g/d), and higher intakes of fat (only for g/d) and alcohol (Table 1). Irregularity score of energy intake ranged from 0-160 for breakfast, lunch, evening meal and between meals and 0-79 for the daily total irregularity score of energy intake. The percentage of plausible reporters was 83% of all cohort members and was lowest for those with a more irregular intake of energy (T3:74%) compared to those with a more regular intake of energy (T1:87%).
Table 1.
Descriptive characteristics of 1768 NSHD cohort members at age 53 years across tertile of daily irregularity score of energy intake
| Daily irregularity score: | T1 (<10.2) n=589 |
T2 (10.2-19.4) n=590 |
T3 (>19.4) n=589 |
||||
|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | p for trend | |
| Waist circumference (cm) | 90.0 | 12.9 | 89.0 | 13.2 | 91.3 | 13.3 | 0.009 |
| Hip circumference (cm) | 107.5 | 52.2 | 114.5 | 96.5 | 108.2 | 52.3 | 0.17 |
| Systolic BP (mm Hg) | 134.5 | 18.8 | 135.4 | 20.2 | 135.9 | 20.0 | 0.46 |
| Diastolic BP (mm Hg) | 83.6 | 11.9 | 84.1 | 12.6 | 84.5 | 12.4 | 0.51 |
| HDL cholesterol (mmol/l)* | 1.66 | 0.50 | 1.7 | 0.5 | 1.7 | 0.5 | 0.09 |
| Triglycerides (mmol/l)** | 2.02 | 1.26 | 2.1 | 1.5 | 2.2 | 1.6 | 0.38 |
| BMI (kg/m2) | 26.8 | 4.8 | 26.6 | 4.5 | 27.5 | 4.7 | 0.01 |
| Sex (% female) | 52.8% | 54.7% | 52.8% | 0.74 | |||
| Marital status (% married) | 78.4% | 80.3% | 77.8% | 0.03 | |||
| SES (% no job) | 6.3% | 7.6% | 6.6% | 0.39 | |||
| Smoking (% current) | 16.6% | 18.0% | 19.9% | 0.06 | |||
| Physical activity (% >=5x per month) | 31.4% | 36.9% | 35.5% | 0.05 | |||
| Total energy intake (kcal/d) | 2075 | 512 | 1983 | 484 | 1887 | 515 | <0.001 |
| Fat intake (g/d) | 78.8 | 26.6 | 76.7 | 24.4 | 72.8 | 24.7 | <0.001 |
| Fat intake (%en/d) | 33.7 | 6.0 | 34.5 | 5.7 | 34.5 | 6.4 | 0.03 |
| Carbohydrate intake (g/d) | 252.0 | 65.2 | 232.8 | 59.8 | 214.8 | 61.0 | <0.001 |
| Carbohydrate intake (%en/d) | 46.0 | 7.0 | 44.5 | 6.9 | 43.1 | 7.4 | <0.001 |
| Protein intake (g/d) | 80.4 | 17.0 | 78.4 | 19.0 | 74.1 | 18.1 | <0.001 |
| Protein intake (%en/d) | 15.9 | 2.8 | 16.0 | 2.7 | 16.0 | 2.8 | 0.45 |
| Alcohol intake (g/d) | 14.5 | 20.8 | 15.5 | 21.0 | 18.8 | 24.8 | 0.003 |
| Plausible reporters (%) | 86.9% | 86.7% | 73.9% | <0.001 | |||
| Energy intake during mealtime slots | |||||||
| Breakfast (kcal/d) | 310 | 150 | 278 | 143 | 244 | 142 | <0.001 |
| Lunch (kcal/d) | 562 | 209 | 542 | 206 | 487 | 203 | <0.001 |
| Evening meal (kcal/d) | 736 | 239 | 719 | 229 | 695 | 269 | 0.02 |
| In-between meals (kcal/d) | 468 | 292 | 443 | 271 | 456 | 334 | 0.38 |
Abbreviations: kilogram (kg), centimetre (cm), millimetre (mm), kilocalorie (kcal), per cent energy (% en), gram per day (g/d), blood pressure (BP), socio-economic status (SES)
HDL cholesterol n=1408 (T1 n=485; T2 n=461; T3 n=462)
Triglycerides n=1514 (T1 n=509; T2 n=503; T3 n=502)
Cohort members with a more irregular daily pattern of energy intake consumed less energy during breakfast, lunch, and during the evening meal compared to those with a more regular daily pattern (p trend <0.05). Other characteristics, such as the ratio men to women, SES, and smoking were similar across tertiles of daily total irregularity scores of energy intake except for marital status and physical activity: the highest number of married cohort members were in the second tertile and cohort members tended to be more physically active with higher irregular scores of energy intake (p for trend <0.05). Cohort members with a higher daily total irregularity score of energy intake also tended to have higher irregularity scores of energy intake for individual meals and the highest irregularity scores of energy intake were found for between meals (Table 2).
Table 2.
Irregularity scores of energy intake per meal per tertile of daily irregularity score of energy intake presented as mean and standard deviations
| Tertile of daily total irregularity score | ||||||
|---|---|---|---|---|---|---|
|
| ||||||
| T1 (<10.2) | n=589 | T2 (10.2-19.4) n=590 | T3 (>19.4) | n=589 | ||
| Mean | SD | Mean | SD | Mean | SD | |
| Breakfast | 23.7 | 27.8 | 30.8 | 33.2 | 42.3 | 40.0 |
| Lunch | 27.9 | 19.6 | 33.8 | 23.5 | 43.9 | 27.7 |
| Evening meal | 26.4 | 17.6 | 31.9 | 19.3 | 41.6 | 24.3 |
| Between meals | 36.1 | 21.4 | 43.9 | 22.9 | 52.2 | 25.4 |
Cardio-metabolic risk factors
No effect modification by sex was observed and therefore results were presented for men and women combined (Table 3).
Table 3.
Logistic regression models for cardio-metabolic risk outcomes per meal in NSHD cohort members at age 53 years per tertile of irregularity score of energy intake presented per meal
| Outcome | Model 0 | Model 1 | Model 2 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Irregularity score of | T1 ** | T2 ** | T3 ** | p trend* | T1 ** | T2 ** | T3 ** | p trend* | T1 ** | T2 ** | T3 ** | p trend* | |
| Metabolic Syndrome (>3 components) # | |||||||||||||
| Breakfast | 1.00 (ref) | 0.96 (0.71;1.29) | 1.42 (1.05;1.91) | 0.01 | 1.00 (ref) | 0.98 (0.72;1.33) | 1.34 (0.99;1.81) | 0.04 | 1.00 (ref) | 0.99 (0.73;1.34) | 1.38 (1.01;1.87) | 0.03 | |
| Lunch | 1.00 (ref) | 0.88 (0.65;1.18) | 0.94 (0.70;1.27) | 0.78 | 1.00 (ref) | 0.92 (0.68;1.24) | 0.92 (0.68;1.25) | 0.64 | 1.00 (ref) | 0.92 (0.68;1.24) | 0.92 (0.68;1.25) | 0.65 | |
| Evening meal | 1.00 (ref) | 1.09 (0.80;1.47) | 1.30 (0.97;1.75) | 0.08 | 1.00 (ref) | 1.07 (0.79;1.46) | 1.26 (0.93;1.70) | 0.13 | 1.00 (ref) | 1.07 (0.78;1.46) | 1.27 (0.94;1.72) | 0.11 | |
| Between meals | 1.00 (ref) | 1.09 (0.80;1.47) | 1.33 (0.98;1.78) | 0.06 | 1.00 (ref) | 1.11 (0.82;1.51) | 1.36 (1.01;1.85) | 0.04 | 1.00 (ref) | 1.11 (0.82;1.51) | 1.38 (1.01;1.86) | 0.04 | |
| Daily total | 1.00 (ref) | 0.83 (0.61;1.12) | 1.07 (0.80;1.43) | 0.53 | 1.00 (ref) | 0.86 (0.63;1.16) | 1.09 (0.81;1.47) | 0.47 | 1.00 (ref) | 0.86 (0.63;1.16) | 1.09 (0.81;1.47) | 0.38 | |
| Increased Waist circumference # | |||||||||||||
| Breakfast | 1.00 (ref) | 1.17 (0.91;1.52) | 1.81 (1.41;2.33) | <0.001 | 1.00 (ref) | 1.16 (0.90;1.51) | 1.90 (1.47;2.45) | <0.001 | 1.00 (ref) | 1.16 (0.90;1.51) | 1.86 (1.43;2.40) | <0.001 | |
| Lunch | 1.00 (ref) | 1.23 (0.96;1.58) | 1.24 (0.96;1.59) | 0.14 | 1.00 (ref) | 1.24 (0.96;1.59) | 1.25 (0.97;1.61) | 0.12 | 1.00 (ref) | 1.23 (0.95;1.58) | 1.21 (0.93;1.56) | 0.21 | |
| Evening meal | 1.00 (ref) | 1.12 (0.87;1.45) | 1.34 (1.04;1.72) | 0.02 | 1.00 (ref) | 1.17 (0.91;1.51) | 1.36 (1.06;1.75) | 0.02 | 1.00 (ref) | 1.17 (0.91;1.52) | 1.31 (1.02;1.69) | 0.04 | |
| Between meals | 1.00 (ref) | 1.07 (0.83;1.38) | 1.17 (0.91;1.50) | 0.23 | 1.00 (ref) | 1.07 (0.83;1.38) | 1.20 (0.93; 1.54) | 0.15 | 1.00 (ref) | 1.08 (0.84;1.39) | 1.20 (0.93;1.55) | 0.15 | |
| Daily total | 1.00 (ref) | 0.97 (0.75;1.25) | 1.30 (1.02;1.66) | 0.02 | 1.00 (ref) | 0.99 (0.76;1.28) | 1.34 (1.04;1.72) | 0.01 | 1.00 (ref) | 0.99 (0.76; 1.28) | 1.34 (1.04;1.72) | 0.03 | |
| Increased BMI (≥25 kg/m2) | |||||||||||||
| Breakfast | 1.00 (ref) | 0.96(0.76;1.21) | 1.49 (1.17;1.89) | <0.001 | 1.00 (ref) | 0.97 (0.76;1.23) | 1.46 (1.14;1.87) | <0.01 | 1.00 (ref) | 0.97 (0.76;1.23) | 1.45 (1.13;1.86) | <0.01 | |
| Lunch | 1.00 (ref) | 1.12 (0.89;1.43) | 1.11 (0.87;1.40) | 0.46 | 1.00 (ref) | 1.18 (0.93;1.51) | 1.11 (0.87;1.42) | 0.48 | 1.00 (ref) | 1.17 (0.92;1.50) | 1.08 (0.85;1.38) | 0.65 | |
| Evening meal | 1.00 (ref) | 1.36 (1.08;1.72) | 1.65 (1.29;2.09) | <0.001 | 1.00 (ref) | 1.34 (1.05;1.70) | 1.62 (1.27;2.06) | <0.001 | 1.00 (ref) | 1.33 (1.05;1.69) | 1.58 (1.24;2.02) | <0.001 | |
| Between meals | 1.00 (ref) | 1.03 (0.81;1.30) | 1.36 (1.07;1.72) | 0.01 | 1.00 (ref) | 1.03 (0.81;1.30) | 1.35 (1.06;1.72) | 0.01 | 1.00 (ref) | 1.04 (0.82;1.32) | 1.35 (1.06;1.73) | 0.01 | |
| Daily total | 1.00 (ref) | 0.96 (0.76;1.21) | 1.41 (1.11;1.79) | <0.01 | 1.00 (ref) | 0.99 (0.78;1.26) | 1.45 (1.14;1.85) | <0.01 | 1.00 (ref) | 0.99 (0.78;1.26) | 1.45 (1.14;1.85) | 0.01 | |
| Increased Systolic BP # | |||||||||||||
| Breakfast | 1.00 (ref) | 0.84 (0.66;1.05) | 1.18 (0.93;1.49) | 0.07 | 1.00 (ref) | 0.86 (0.68;1.09) | 1.12 (0.88;1.42) | 0.22 | 1.00 (ref) | 0.85 (0.67;1.08) | 1.08 (0.85;1.38) | 0.36 | |
| Lunch | 1.00 (ref) | 0.85 (0.67;1.08) | 0.86 (0.68;1.09) | 0.26 | 1.00 (ref) | 0.88 (0.69;1.12) | 0.84 (0.66;1.07) | 0.17 | 1.00 (ref) | 0.88 (0.69; 1.12) | 0.82 (0.64;1.04) | 0.12 | |
| Evening meal | 1.00 (ref) | 0.93 (0.74;1.18) | 0.91 (0.72;1.16) | 0.48 | 1.00 (ref) | 0.88 (0.69;1.12) | 0.88 (0.69;1.12) | 0.33 | 1.00 (ref) | 0.88 (0.70;1.12) | 0.85 (0.67;1.08) | 0.20 | |
| Between meals | 1.00 (ref) | 0.92 (0.73;1.16) | 0.95 (0.76;1.21) | 0.75 | 1.00 (ref) | 0.92 (0.73;1.17) | 0.93 (0.74;1.18) | 0.60 | 1.00 (ref) | 0.92 (0.73;1.17) | 0.93 (0.74;1.19) | 0.61 | |
| Daily total | 1.00 (ref) | 1.02 (0.81;1.28) | 1.11 (0.88;1.40) | 0.36 | 1.00 (ref) | 1.03 (0.82;1.31) | 1.10 (0.87;1.40) | 0.42 | 1.00 (ref) | 1.03 (0.82;1.31) | 1.10 (0.87;1.40) | 0.39 | |
| Increased Diastolic BP # | |||||||||||||
| Breakfast | 1.00 (ref) | 0.94 (0.75;1.19) | 1.28 (1.02;1.62) | 0.02 | 1.00 (ref) | 0.98 (0.77;1.24) | 1.19 (0.94;1.51) | 0.12 | 1.00 (ref) | 0.97 (0.77;1.24) | 1.14 (0.89;1.45) | 0.24 | |
| Lunch | 1.00 (ref) | 0.99 (0.78;1.25) | 0.98 (0.78;1.24) | 0.86 | 1.00 (ref) | 1.04 (0.82; 1.32) | 0.95 (0.75;1.21) | 0.62 | 1.00 (ref) | 1.03 (0.81;1.31) | 0.92 (0.72;1.17) | 0.45 | |
| Evening meal | 1.00 (ref) | 0.96 (0.76;1.21) | 1.12 (0.89;1.42) | 0.28 | 1.00 (ref) | 0.89 (0.70;1.13) | 1.07 (0.84;1.35) | 0.49 | 1.00 (ref) | 0.90 (0.71;1.14) | 1.05 (0.82;1.33) | 0.62 | |
| Between meals | 1.00 (ref) | 0.98 (0.77;1.23) | 1.05 (0.83; 1.32) | 0.66 | 1.00 (ref) | 0.98 (0.77;1.24) | 1.02 (0.81;1.30) | 0.83 | 1.00 (ref) | 0.99 (0.78;1.26) | 1.02 (0.80;1.30) | 0.84 | |
| Daily total | 1.00 (ref) | 0.94 (0.75;1.19) | 1.05 (0.83;1.320 | 0.65 | 1.00 (ref) | 0.95 (0.75;1.21) | 1.03 (0.81;1.31) | 0.75 | 1.00 (ref) | 0.95 (0.75;1.21) | 1.03 (0.81;1.31) | 0.84 | |
| Decreased HDL cholesterol # | |||||||||||||
| Breakfast | 1.00 (ref) | 0.88 (0.57;1.36) | 1.10 (0.72;1.68) | 0.57 | 1.00 (ref) | 0.88 (0.57; 1.36) | 1.07 (0.70;1.65) | 0.65 | 1.00 (ref) | 0.89 (0.57;1.37) | 1.10 (0.71;1.70) | 0.56 | |
| Lunch | 1.00 (ref) | 1.14 (0.75;1.73) | 0.77 (0.49;1.21) | 0.20 | 1.00 (ref) | 1.18 (0.78;1.79) | 0.74 (0.47;1.17) | 0.14 | 1.00 (ref) | 1.19 (0.78;1.80) | 0.76 (0.48;1.20) | 0.18 | |
| Evening meal | 1.00 (ref) | 1.18 (0.77;1.82) | 1.06 (0.69;1.65) | 0.85 | 1.00 (ref) | 1.22 (0.79;1.88) | 1.06 (0.68;1.65) | 0.88 | 1.00 (ref) | 1.23 (0.79;1.89) | 1.11 (0.71;1.73) | 0.71 | |
| Between meals | 1.00 (ref) | 1.25 (0.81;1.94) | 1.27 (0.82;1.97) | 0.32 | 1.00 (ref) | 1.26 (0.81; 1.96) | 1.32 (0.85;2.05) | 0.25 | 1.00 (ref) | 1.25 (0.81;1.95) | 1.33 (0.86;2.07) | 0.23 | |
| Daily total | 1.00 (ref) | 0.61 (0.39;0.95) | 0.89 (0.59;1.34) | 0.69 | 1.00 (ref) | 0.62 (0.40;0.98) | 0.91 (0.60;1.37) | 0.77 | 1.00 (ref) | 0.62 (0.40;0.98) | 0.91 (0.60;1.37) | 0.99 | |
| Increased Triglycerides # | |||||||||||||
| Breakfast | 1.00 (ref) | 0.77 (0.60;0.98) | 1.20 (0.94; 1.54) | 0.05 | 1.00 (ref) | 0.78 (0.60;1.01) | 1.07 (0.82;1.39) | 0.37 | 1.00 (ref) | 0.79 (0.61;1.02) | 1.11 (0.85;1.45) | 0.25 | |
| Lunch | 1.00 (ref) | 0.85 (0.67;1.09) | 0.87 (0.68;1.12) | 0.35 | 1.00 (ref) | 0.92 (0.71;1.20) | 0.81 (0.62;1.05) | 0.11 | 1.00 (ref) | 0.92 (0.71;1.19) | 0.80 (0.61;1.04) | 0.09 | |
| Evening meal | 1.00 (ref) | 0.95 (0.74; 1.22) | 1.24 (0.97;1.59) | 0.06 | 1.00 (ref) | 0.88 (0.68; 1.14) | 1.16 (0.89;1.50) | 0.20 | 1.00 (ref) | 0.87 (0.67;1.13) | 1.15 (0.89;1.50) | 0.21 | |
| Between meals | 1.00 (ref) | 1.14 (0.89;1.46) | 1.18 (0.92; 1.51) | 0.21 | 1.00 (ref) | 1.17 (0.90;1.51) | 1.20 (0.92;1.55) | 0.20 | 1.00 (ref) | 1.17 (0.91;1.52) | 1.24 (0.95;1.60) | 0.13 | |
| Daily total | 1.00 (ref) | 0.91 (0.71;1.16) | 0.96 (0.75;1.23) | 0.79 | 1.00 (ref) | 0.94 (0.73;1.22) | 0.95 (0.73;1.23) | 0.73 | 1.00 (ref) | 0.94 (0.73;1.22) | 0.95 (0.73;1.23) | 0.97 | |
Model 0 unadjusted
Model 1 adjusted for sex, physical activity, SES, marital status, and smoking
Model 2 adjusted for sex, physical activity, SES, marital status, smoking, energy intake during other meals or EAR for daily total
P for trend based on median intake scores per tertile
The metabolic syndrome was defined as ≥3 of the following criteria present: waist circumference ≥102cm in men and ≥88cm in women; HDL cholesterol <1.036mmol/l (40mg/dl) in men and <1.295mmol/l (50mg/dl) in women or specific treatment for this lipid abnormality; triglycerides ≥1.7mmol/l (150mg/dl) in men and women or specific treatment for this lipid abnormality; and systolic blood pressure ≥130mm Hg or diastolic blood pressure ≥85mm Hg in men and women or use of anti-hypertensive medication; and Hb1Ac level in the top gender-specific quarter of the distribution (>5.8% among men and women) or use of medication for diabetes
Abbreviations: NSHD (National Health and Development Survey); BMI (body mass index); BP (blood pressure); HDL (high density lipoprotein), SES (socio-economic status)
Cut-off values for tertiles of irregularity scores were:
Breakfast: <4.1 (n=589), 4.1-42.4 (n=590), >42.4 (n=589)
Lunch: <18.9 (n=589), 18.9-43.8 (n=590), >43.8 (n=589)
Evening meal: <19.6 (n=589), 19.6-40.9 (n=590), >40.9 (n=589)
Between meals: <26.7 (n=589), 26.7-56.9 (n=590), >56.9 (n=589)
Daily total: <10.2 (n=589), 10.2-19.4 (n=590), >19.4 (n=589)
More irregular intake of energy at breakfast was associated with a higher risk of the metabolic syndrome (model 1: OR for top versus bottom tertile 1.34 (0.99; 1.81); p value for trend across 3 groups=0.04), as was an irregular intake of energy between meals (model 1: OR 1.36 (1.01; 1.85); p for trend 0.04). These associations were somewhat stronger after further adjustment for energy intake at other meals; for breakfast the OR was 1.38 (95% CI 1.01; 1.87; p for trend 0.03) and for between meals 1.38 (95% CI 1.01; 1.86; p for trend 0.04).
Increased waist circumference was associated with a more irregular intake of energy at breakfast (model 1: OR 1.90 (1.47; 2.45); p for trend <0.01), at evening meals (model 1: OR 1.36 (1.06; 1.75); p for trend <0.01) as well as for the daily total (model 1: OR 1.34 (1.04; 1.72); p for trend 0.01). These associations were similar or stronger after adjustment for energy intake at other meals or total energy intake (Table 3, model 2). For BMI, similar observations were made: increased BMI was associated with a more irregular intake of energy at breakfast (model 1: OR 1.46 (1.14; 1.87); p for trend <0.01), evening meal (model 1: OR 1.62 (1.27; 2.06); p for trend <0.001), between meals (model 1: 1.35 (1.06;1.72); p for trend=0.01) and daily total (model 1: OR 1.45 (1.14; 1.85); p for trend <0.01).
For systolic and diastolic BP, HDL cholesterol and plasma triglycerides, no statistical significant associations were observed.
Sensitivity analyses including only plausible reporters of energy intake (n=1441, 83%) showed similar results though results were somewhat attenuated, for example cohort members with irregular intake of energy during breakfast appeared to have a higher risk of the metabolic syndrome (model 1: OR 1.45 (1.02;2.06)).
Secondary outcomes
The continuous components of the metabolic syndrome were further explored using linear regression models. As no effect modification by sex was observed, results are presented for men and women combined (Table 4).
Table 4.
Linear regression outcomes of irregularity scores of energy intake (continuously) with continuous outcome variables per meal in NSHD cohort members at age 53 years
| Outcome | Model 0 | Model 1 | Model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Irregularity score of | b | 95% CI | P value | b | 95% CI | P value | b | 95% CI | P value | |
| Waist circumference (cm) | ||||||||||
| Breakfast | 0.048 | (0.030;0.065) | <0.001 | 0.028 | (0.013;0.044) | <0.001 | 0.027 | (0.011;0.43) | <0.001 | |
| Lunch | 0.033 | (0.009;0.058) | 0.01 | 0.014 | (−0.008;0.036) | 0.21 | 0.010 | (−0.012;.032) | 0.38 | |
| Evening meal | 0.054 | (0.026;0.083) | <0.001 | 0.037 | (0.012;0.062) | 0.004 | 0.033 | (0.007;0.059) | 0.01 | |
| Between meals | 0.037 | (0.011;0.063) | 0.01 | 0.030 | (0.008;0.053) | 0.008 | 0.032 | (0.009;0.054) | 0.01 | |
| Daily total | 0.143 | (0.063;0.223) | <0.001 | 0.119 | (0.049;.0189) | 0.001 | 0.119 | (0.049;0.189) | <0.001 | |
| BMI (kg/m2) | ||||||||||
| Breakfast | 0.008 | (0.002;0.014) | 0.01 | 0.010 | (0.004;0.016) | 0.002 | 0.009 | (0.003;0.016) | <0.001 | |
| Lunch | 0.009 | (0.000;0.018) | 0.05 | 0.009 | (0.001;0.018) | 0.04 | 0.008 | (−0.001;0.017) | 0.09 | |
| Evening meal | 0.020 | (0.010;0.030) | <0.001 | 0.021 | (0.011;0.031) | <0.001 | 0.019 | (0.009;0.030) | <0.001 | |
| Between meals | 0.015 | (0.006;0.024) | 0.001 | 0.015 | (0.006;0.024) | 0.001 | 0.015 | (0.006;0.024) | <0.001 | |
| Daily total | 0.051 | (0.023;0.080) | <0.001 | 0.054 | (0.026;0.082) | <0.001 | 0.054 | (0.026;0.082) | <0.001 | |
| Systolic BP (mm Hg) | ||||||||||
| Breakfast | 0.037 | (0.010;0.063) | 0.01 | 0.024 | (−0.003;0.050) | 0.08 | 0.018 | (−0.009;0.045) | 0.19 | |
| Lunch | 0.017 | (−0.020;0.054) | 0.38 | 0.007 | (−0.030;0.044) | 0.72 | −0.003 | (−0.041;0.035) | 0.89 | |
| Evening meal | 0.029 | (−0.014;0.072) | 0.19 | 0.020 | (−0.023;0.063) | 0.36 | 0.006 | (−0.038;0.050) | 0.79 | |
| Between meals | −0.003 | (−0.042;0.036) | 0.87 | −0.006 | (−0.044;0.033) | 0.78 | −0.004 | (−0.043;0.034) | 0.82 | |
| Daily total | 0.010 | (−.0111;0.131) | 0.87 | −0.006 | (−0.125;0.114) | 0.93 | −0.006 | (−0.125;0.114) | 0.93 | |
| Diastolic BP (mm Hg) | ||||||||||
| Breakfast | 0.024 | (0.008;0.041) | <0.001 | 0.015 | (−0.001;0.032) | 0.07 | 0.012 | (−0.004;0.029) | 0.15 | |
| Lunch | 0.014 | (−0.009;0.038) | 0.23 | 0.007 | (−0.016;0.030) | 0.54 | 0.002 | (−0.022;0.025) | 0.88 | |
| Evening meal | 0.035 | (0.009;0.062) | 0.01 | 0.029 | (0.002;0.055) | 0.03 | 0.023 | (−0.004;0.050) | 0.10 | |
| Between meals | 0.003 | (−0.021;0.027) | 0.80 | 0.000 | (−0.023;0.024) | 0.97 | 0.001 | (−0.023;0.025) | 0.93 | |
| Daily total | 0.017 | (−0.058;0.093) | 0.66 | 0.005 | (−0.069;0.079) | 0.89 | 0.005 | (−0.069;0.079) | 0.89 | |
| HDL cholesterol (mmol/l) | ||||||||||
| Breakfast | 0.000 | (−0.001;0.000) | 0.20 | 0.000 | (0.000;0.001) | 0.40 | 0.000 | (0.000;0.001) | 0.40 | |
| Lunch | 0.000 | (−0.001;0.001) | 0.54 | 0.000 | (−0.001;0.001) | 0.34 | 0.001 | (−0.001;0.002) | 0.34 | |
| Evening meal | −0.001 | (−0.002;0.001) | 0.26 | 0.000 | (−0.001;0.001) | 0.84 | 0.000 | (−0.001;0.001) | 0.95 | |
| Between meals | 0.000 | (−0.001;0.001) | 0.92 | 0.000 | (−0.001;0.001) | 0.73 | 0.000 | (−0.001;0.001) | 0.76 | |
| Daily total | 0.000 | (−0.003;0.004) | 0.94 | 0.001 | (−0.002;0.004) | 0.63 | 0.001 | (−0.002;0.004) | 0.63 | |
| Triglycerides (mmol/l) | ||||||||||
| Breakfast | 0.004 | (0.002;0.006) | <0.001 | 0.003 | (0.000;0.005) | 0.02 | 0.003 | (0.000;0.005) | 0.02 | |
| Lunch | 0.003 | (0.000;0.006) | 0.09 | 0.001 | (−0.002;0.004) | 0.68 | 0.000 | (−0.003;0.003) | 0.80 | |
| Evening meal | 0.003 | (−0.001;0.006) | 0.15 | 0.001 | (−0.003;.004) | 0.65 | 0.001 | (−0.003;0.004) | 0.77 | |
| Between meals | 0.003 | (0.000;0.006) | 0.03 | 0.004 | (0.001;0.006) | 0.02 | 0.004 | (0.001;0.007) | 0.02 | |
| Daily total | 0.007 | (−0.003;.0.016) | 0.19 | 0.005 | (−0.005;0.014) | 0.35 | 0.005 | (−0.005;0.014) | 0.35 | |
Model 0 unadjusted
Model 1 adjusted for sex, physical activity, SES, marital status, and smoking
Model 2 adjusted for sex, physical activity, SES, marital status, smoking, energy intake during other meals or EAR for daily total
Abbreviations: NSHD (National Health and Development Survey); BMI (body mass index); BP (blood pressure); HDL (high density lipoprotein), SES (socio-economic status)
Cohort members with a more irregular energy intake during all mealtimes had higher waist circumferences (Table 4; model 0). After adjusting for sex, physical activity, SES, marital status and smoking, this finding remained statistically significant for all meals except for lunch (Table 4; model 1); similar results were seen after further adjustment for energy intakes during the other meals or EAR (Table 4; model 2).
Cohort members with a more irregular energy intake during breakfast, evening meal, between meals and daily total, had higher BMIs (Table 4; model 0). This remained after adjusting for sex, physical activity, SES, marital status and smoking (Table 4; model 1) as well as after further adjustment for energy intakes during the other meals (Table 4; model 2).
No statistical significant associations were observed for meal regularity of energy intake at the different meals and systolic BP, diastolic BP, and HDL cholesterol in adjusted models, although the unadjusted model for regularity of energy intake during breakfast was statistically significantly associated with systolic and diastolic BP as well as HDL cholesterol (Table 4, model 0).
Cohort members with a more irregular energy intake at breakfast had higher plasma triglycerides in adjusted models.
Discussion
This innovative study investigating meal irregularity of energy intake derived from detailed dietary information showed that having a more irregular energy intake at breakfast as well as between meals was associated with a higher risk of the metabolic syndrome. Moreover, a more irregular energy intake at breakfast, as well as between meals, was associated with a higher waist circumference and BMI.
To the best of our knowledge, this is the first study investigating meal irregularity using variability in energy intake during meals as a proxy for irregularity. Previous studies investigated another aspect of regularity, namely regular meal frequency and showed beneficial effects of regular meals on cardio-metabolic risk factors and BMI in both observational studies 5, 22; and intervention trials 7, 8. The studies of Farshshi et al showed that irregular meal consumption resulted in greater insulin resistance and higher fasting lipid profiles of total and LDL cholesterol in lean women7 and these effects appeared to be more pronounced in obese women8. The suggested underlying biological mechanism for these beneficial effects of meal regularity on cardio-metabolic risk factors could be related to de-synchronisation of circadian rhythms that are found in cardiovascular function, glucose metabolism and the gastro-intestinal tract 23, 24. Regulating and keeping the number of meals per day constant could minimise fluctuations in insulin concentration and plasma glucose. Since a considerable drop in glucose can induce hunger, a more regular eating regime could potentially reduce this effect. It may also be that more regular eating can induce more stable and constant plasma levels of intestinal satiety hormones, such as glucagon-like-peptide-1, cholecystokinin and peptide YY 24. In this study, meal irregularity of energy intake was associated with BMI and waist circumference, and these could potentially be mediating factors between meal regularity and cardio-metabolic disease risk. However, more research is needed to elucidate the exact underlying mechanisms of the results seen in this study.
Despite the possible beneficial effects of regular meal consumption, regular over-consumption of energy could be a potential risk factor for overweight, obesity and other cardio-metabolic risk factors. Therefore, we also adjusted our analyses for energy requirements in an additional model (model 2). It should be noted that we could only use this approach for the daily total energy regularity score since currently there are no specific energy requirements for individuals’ mealtimes. Analyses on the associations between the irregularity scores of energy intake per meal were further adjusted for energy intake during the other meals in an additional model (model 2), which showed similar or stronger associations compared to model 1.
Another aspect to take into consideration is the fact that overweight or obese people tend to under-report their dietary intake more than people with a normal weight25. We did find that the percentage of plausible reporters of energy intake was somewhat lower for cohort members who reported a more irregular energy intake. However, our sensitivity analysis including only plausible reporters found that results of the logistic regression analyses were similar though somewhat attenuated suggesting under-reporting may have slightly biased our findings.
A challenge of studying meal patterns in general is that different definitions for a meal are used in different studies that make it less than straightforward to compare different studies on meal patterns. In the diet diaries used in NSHD, meal times were pre-structured in eight meal slots which were then consolidated into four (breakfast, lunch, evening meal, and between meals). Cohort members were asked to fill in their dietary intake accordingly. Thus for data analysis in this paper we were reliant on these specific pre-defined and subjective meals descriptors. The benefit of this is that these were the same for all cohort members. However, it could be speculated that having predefined meal slots could be difficult for some specific sub-groups of the population, like shift workers, who may follow a different pattern of meals over the day; this mode of working is associated with a greater risk for obesity and cardio-vascular disease 26. Unfortunately data on shift workers was not available.
Other studies on specific meal patterns have mostly been focussed on breakfast consumption showing that patterns of breakfast consumption are associated with health consequences27, 28. In this study, it was observed that higher irregularity of energy intake at one meal slot was correlated higher irregularity of energy intake at another time slot or daily total irregularity scores of energy intake suggesting that meal irregularity was not predominant for one meal specifically but that meal irregularity of energy intake at specific meals could have different metabolic consequences as most predominant associations were found for breakfast and between meals.
A limitation of this study was the missing data for the metabolic syndrome, mainly due to missing data of HDL cholesterol levels as well as non-fasting blood samples which were used to measure serum triglycerides; though some studies indicate that non-fasting blood samples may still be an indicator of cardiovascular disease risk 30, 31. A sensitivity analysis comparing those with missing data to the total population showed similar dietary intakes of energy, macronutrients and alcohol (data not shown), however, those with missing data for the metabolic syndrome seemed to have a higher BMI and waist circumference, smoked a little more and were somewhat less physically active. This may have led to some regression dilution bias 32.
On a general note, given the multiple tests there was the potential for false positive findings; however, if corrected for multiple testing and thus reducing the false positive error rate, the false negative error rate would increase which could lead to missing possible important findings 29. Additionally, correction for multiple testing is usually required when a large number of independent statistical tests is performed and it could be argued that in this study, meal irregularity of the different meal slots are correlated, as it was observed that cohort members who had an irregular daily total energy intake also seemed to have more irregular energy intake during breakfast, lunch, evening meal and between meals.
A major strength of this study is the use of detailed dietary data in a substantial number of subjects and to our knowledge this is the first investigation of meal irregularity using such data in relation to cardio-metabolic risk factors. Dietary intakes were assessed using 5-day estimated diet diaries, which have taken over from the now rarely performed weighed assessments as the gold standard for dietary assessment 33, since they provide similar energy intakes 34. Since the data were derived from a birth cohort, all cohort members were of the same age and age was therefore not a confounding variable. Nevertheless, the use of a cohort of one age does impact on the generalizability to other populations and therefore more research on regularity of meal patterns in other age groups and/or populations would be desirable. Notably, while this sample has its limitations in terms of generalizability, their dietary intakes were comparable to other national UK populations, like those of similar age in the National Diet and Nutrition Survey (NDNS) 35, 36. Providing unique insights into associations between meal irregularity of energy intake and cardio-metabolic risk factors, the results of this research could potentially be translated into practical public health messages, which are especially relevant as the obesity epidemic continues. Regularity of meals has been implied as a successful factor in weight loss strategies 37. However, dietary behaviour is complex and encompasses numerous characteristics 38. Thus more research, combining more elements of chronobiology and dietary behaviour is warranted.
Acknowledgements
We are indebted to all the members who took part in the NSHD.
This work was supported by the MRC grant numbers U1200632239 and U123092720. GKP and AMS designed the research; GP conducted the research and analysed data; GKP, RH, AMS wrote the paper.
Footnotes
Conflict of Interest The authors declare no conflict of interest.
References
- 1.Mesas AE, Munoz-Pareja M, Lopez-Garcia E, Rodriguez-Artalejo F. Selected eating behaviours and excess body weight: a systematic review. Obes Rev. 2012;13(2):106–35. doi: 10.1111/j.1467-789X.2011.00936.x. [DOI] [PubMed] [Google Scholar]
- 2.Samuelson G. Dietary habits and nutritional status in adolescents over Europe. An overview of current studies in the Nordic countries. European journal of clinical nutrition. 2000;54(Suppl 1):S21–8. doi: 10.1038/sj.ejcn.1600980. [DOI] [PubMed] [Google Scholar]
- 3.Berteus Forslund H, Torgerson JS, Sjostrom L, Lindroos AK. Snacking frequency in relation to energy intake and food choices in obese men and women compared to a reference population. International journal of obesity (2005) 2005;29(6):711–9. doi: 10.1038/sj.ijo.0802950. [DOI] [PubMed] [Google Scholar]
- 4.Ekmekcioglu C, Touitou Y. Chronobiological aspects of food intake and metabolism and their relevance on energy balance and weight regulation. Obes Rev. 2011;12(1):14–25. doi: 10.1111/j.1467-789X.2010.00716.x. [DOI] [PubMed] [Google Scholar]
- 5.Sierra-Johnson J, Unden AL, Linestrand M, Rosell M, Sjogren P, Kolak M, et al. Eating meals irregularly: a novel environmental risk factor for the metabolic syndrome. Obesity (Silver Spring) 2008;16(6):1302–7. doi: 10.1038/oby.2008.203. [DOI] [PubMed] [Google Scholar]
- 6.Shin A, Lim SY, Sung J, Shin HR, Kim J. Dietary intake, eating habits, and metabolic syndrome in Korean men. Journal of the American Dietetic Association. 2009;109(4):633–40. doi: 10.1016/j.jada.2008.12.015. [DOI] [PubMed] [Google Scholar]
- 7.Farshchi HR, Taylor MA, Macdonald IA. Regular meal frequency creates more appropriate insulin sensitivity and lipid profiles compared with irregular meal frequency in healthy lean women. European journal of clinical nutrition. 2004;58(7):1071–7. doi: 10.1038/sj.ejcn.1601935. [DOI] [PubMed] [Google Scholar]
- 8.Farshchi HR, Taylor MA, Macdonald IA. Beneficial metabolic effects of regular meal frequency on dietary thermogenesis, insulin sensitivity, and fasting lipid profiles in healthy obese women. The American journal of clinical nutrition. 2005;81(1):16–24. doi: 10.1093/ajcn/81.1.16. [DOI] [PubMed] [Google Scholar]
- 9.Almoosawi S, Winter J, Prynne CJ, Hardy R, Stephen AM. Daily profiles of energy and nutrient intakes: are eating profiles changing over time? European journal of clinical nutrition. 2012;66(6):678–86. doi: 10.1038/ejcn.2011.210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Almoosawi S, Prynne CJ, Hardy R, Stephen AM. Time-of-day and nutrient composition of eating occasions: prospective association with the metabolic syndrome in the 1946 British birth cohort. International journal of obesity (2005) 2012 doi: 10.1038/ijo.2012.103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Almoosawi S, Prynne CJ, Hardy R, Stephen AM. Diurnal eating rhythms: Association with long-term development of diabetes in the 1946 British birth cohort. Nutr Metab Cardiovasc Dis. 2013 doi: 10.1016/j.numecd.2013.01.003. [DOI] [PubMed] [Google Scholar]
- 12.Wadsworth M, Kuh D, Richards M, Hardy R. Cohort Profile: The 1946 National Birth Cohort (MRC National Survey of Health and Development) International journal of epidemiology. 2006;35(1):49–54. doi: 10.1093/ije/dyi201. [DOI] [PubMed] [Google Scholar]
- 13.Prynne C, Nip WF, Almoosawi S, Lennox A. MRC National Survey of Health and Development (MRC NSHD) Dietary data report 2006-11. 2012.
- 14.Prynne CJ, Paul AA, Mishra GD, Greenberg DC, Wadsworth ME. Changes in intake of key nutrients over 17 years during adult life of a British birth cohort. The British journal of nutrition. 2005;94(3):368–76. doi: 10.1079/bjn20041404. [DOI] [PubMed] [Google Scholar]
- 15.Price GM, Paul AA, Key FB, Harter AC, Cole D, Day KC, et al. Measurement of diet in a large national survey: comparison of computerised and manual coding in household measures. J Hum Nutr Diet. 1995;8:417–428. [Google Scholar]
- 16.Paul AA, Southgate DAT. McCance and Widdowson’s The Composition of Foods. 4th ed London: 1978. [Google Scholar]
- 17.Holland B, Unwin I, Buss DH. Cereals and Cereal Products: Third Supplement to McCance and Widdowson’s The Composition of Foods. 1988.
- 18.Holland B, Unwin I, Buss DH. Milk and Milk Products: Fourth Supplement to McCance and Widdowson’s The Composition of Foods. 1989.
- 19.Scientific Advisory Committee on Nutrition Dietary Recommendations for Energy. 2011.
- 20.Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112(17):2735–52. doi: 10.1161/CIRCULATIONAHA.105.169404. [DOI] [PubMed] [Google Scholar]
- 21.McCrory MA, McCrory MA, Hajduk CL, Roberts SB. Procedures for screening out inaccurate reports of dietary energy intake. Public health nutrition. 2002;5(6A):873–82. doi: 10.1079/PHN2002387. [DOI] [PubMed] [Google Scholar]
- 22.al-Isa AN. Obesity among Kuwait University students: an explorative study. J R Soc Promot Health. 1999;119(4):223–7. doi: 10.1177/146642409911900404. [DOI] [PubMed] [Google Scholar]
- 23.McFerran B, Mukhopadhyay A. Lay Theories of Obesity Predict Actual Body Mass. Psychological science. 2013 doi: 10.1177/0956797612473121. [DOI] [PubMed] [Google Scholar]
- 24.Garaulet M, Madrid JA. Chronobiological aspects of nutrition, metabolic syndrome and obesity. Adv Drug Deliv Rev. 2010;62(9-10):967–78. doi: 10.1016/j.addr.2010.05.005. [DOI] [PubMed] [Google Scholar]
- 25.Lissner L, Heitmann BL, Bengtsson C. Population studies of diet and obesity. The British journal of nutrition. 2000;83(Suppl 1):S21–4. doi: 10.1017/s000711450000091x. [DOI] [PubMed] [Google Scholar]
- 26.Erren TC, Pape HG, Reiter RJ, Piekarski C. Chronodisruption and cancer. Die Naturwissenschaften. 2008;95(5):367–82. doi: 10.1007/s00114-007-0335-y. [DOI] [PubMed] [Google Scholar]
- 27.Astbury NM, Taylor MA, Macdonald IA. Breakfast consumption affects appetite, energy intake, and the metabolic and endocrine responses to foods consumed later in the day in male habitual breakfast eaters. The Journal of nutrition. 2011;141(7):1381–9. doi: 10.3945/jn.110.128645. [DOI] [PubMed] [Google Scholar]
- 28.Timlin MT, Pereira MA. Breakfast frequency and quality in the etiology of adult obesity and chronic diseases. Nutrition reviews. 2007;65(6 Pt 1):268–81. doi: 10.1301/nr.2007.jun.268-281. [DOI] [PubMed] [Google Scholar]
- 29.Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology. 1990;1(1):43–6. [PubMed] [Google Scholar]
- 30.Stalenhoef AF, de Graaf J. Association of fasting and nonfasting serum triglycerides with cardiovascular disease and the role of remnant-like lipoproteins and small dense LDL. Current opinion in lipidology. 2008;19(4):355–61. doi: 10.1097/MOL.0b013e328304b63c. [DOI] [PubMed] [Google Scholar]
- 31.Nordestgaard BG, Langsted A, Freiberg JJ. Nonfasting hyperlipidemia and cardiovascular disease. Current drug targets. 2009;10(4):328–35. doi: 10.2174/138945009787846434. [DOI] [PubMed] [Google Scholar]
- 32.Fontana A, Copetti M, Mazzoccoli G, Kypraios T, Pellegrini F. A linear mixed model approach to compare the evolution of multiple biological rhythms. Statistics in medicine. 2013;32(7):1125–35. doi: 10.1002/sim.5712. [DOI] [PubMed] [Google Scholar]
- 33.Thompson FE, Subar AF. Dietary assessment methodology. In: Coulston A, Boushey C, editors. Nutrition in the Prevention and Treatment of Disease. 2nd edn Elsevier; Amsterdam: 2008. [Google Scholar]
- 34.Bingham SA, Gill C, Welch A, Cassidy A, Runswick SA, Oakes S, et al. Validation of dietary assessment methods in the UK arm of EPIC using weighed records, and 24-hour urinary nitrogen and potassium and serum vitamin C and carotenoids as biomarkers. International journal of epidemiology. 1997;26(Suppl 1):S137–51. doi: 10.1093/ije/26.suppl_1.s137. [DOI] [PubMed] [Google Scholar]
- 35.Henderson L, Gregory J, Irving K, Swan G. National Diet and Nutrition Survey: adults aged 19 to 64 years. Volume 2: Energy, protein, carbohydrate, fat and alcohol intake. Food Standards Agency, Department of Health London; 2002. [Google Scholar]
- 36.Bates B, Lennox AM, Prentice A, Bates C, Swan G. National Diet and Nutrition Survey: Headline results from Years 1, 2 and 3 (combined) of the Rolling Programme (2008/2009-2010/11) London, UK: 2012. [Google Scholar]
- 37.Erren TC, Reiter RJ. A generalized theory of carcinogenesis due to chronodisruption. Neuro endocrinology letters. 2008;29(6):815–21. [PubMed] [Google Scholar]
- 38.Steptoe A, Pollard TM, Wardle J. Development of a measure of the motives underlying the selection of food: the food choice questionnaire. Appetite. 1995;25(3):267–84. doi: 10.1006/appe.1995.0061. [DOI] [PubMed] [Google Scholar]