Introduction
Body Weight and Body Composition
Observational Studies
Study (year) | Population | Measurements | Findings |
---|---|---|---|
Fabry et al.[13] (1964) | 379 older males (60-64 yrs) | Frequency of food intake survey, calculation to determine overweight classification, triceps and subscapular skinfolds, and blood variables | Ingesting > 5meals/d, as compared to < 3 meals/d, significantly improves overweight classification and subcutaneous fat. |
Hedja & Fabry [14] (1964) | 89 males (30-50 yrs) | 2 week diet records along with height, body weight, and 12 site skinfold thickness | The group that ate less than 4 meals/day had a significantly greater body mass and skinfold averages than those that ate > 5 meals/day. |
Metzner et al. [15] (1977) | 948 males and 1,080 females (35-69 yrs) | 24 hour diet record interview, calculated adiposity index (i.e., calculated using triceps and subscapular skinfold measurements, height, and weight) | Adiposity index was inversely related (significantly) to meal frequency in both men and women after adjusting for caloric intake. In summary, as meal frequency increased, overweight classification decreased. |
Drummond et al. [16] (1998) | 42 males and 37 females (20-55 yrs) with a BMI from 18-30. (Suspected under-reporters were excluded from final analysis) | 7 day food diary; 7 day activity diary, 48 hour HR monitoring, 4 site skinfold thickness, height, and body weight. | Significant negative correlation between eating frequency and body weight was observed in males, but not females. Eating frequency was significantly correlated with total energy intake in females, but not in males. In both men and women no significant correlations between eating frequency and total energy expenditure were observed. |
Ruidavets et al. [17] (2002) | 330 males (45-64 yrs) | 3 day diet record, estimated physical activity (i.e., leisure, work related, and walking/cycling to work), body mass index, and waist-to-hip ratio | After eliminating under reporters (new sample size = 297) and restrained eaters (new sample size = 243), a significant negative correlation between eating frequency and BMI as well as waist-to-hip ratio was observed. |
Ma et al. [18] (2003) | 251 males and 248 females (20-70 yrs) | 24 hour dietary recalls, physical activity recalls, body weight, BMI, and physical activity recalls were collected every 3 months for 1 year | After adjusting for age, sex, physical activity, education, and total energy intake, participants reporting 4 or more eating episodes per day had a significantly lower risk of developing obesity than those eating 3 or fewer times per day. |
Franko et al. [19] (2008) | 1,209 black and 1,166 white female school children (9-19 yrs) | Multiple 3-day food diaries taken over several years, height, weight, and self reported physical activity | Girls between 9-19 years old, that ate 3 or more meals per day had significantly lower BMI-for-age Z scores. |
Study (year) | Population | Measurements | Findings |
---|---|---|---|
Dreon et al. [20] (1988) | 155 sedentary, overweight males (i.e., 120-140% of ideal weight) (30-59 yrs) | 7 day diet records, physical activity questionnaires, VO2 max treadmill test, resting metabolic rate via indirect calorimetry, hydrostatic weighing, and body mass. | Meal frequency did not have a significant effect on percent body fat, total weight, fat-free mass, or resting metabolic rate. |
Kant et al. [21] (1995) | 2,580 males and 4,567 females (25-74 yrs) | Baseline 24-hour dietary recall that assessed meal frequency and compared to follow-up interview several years later. Body weight, BMI, and physical activity were also assessed. | When regression analysis accounted for various covariates (i.e., age, energy intake, level of physical activity, smoking status, race, education, baseline BMI, alcohol intake, and level of morbidity), no significant differences between weight change and meal frequency were reported either at baseline or the follow-up. |
Summerbell et al. [22] (1996) | 187 males and females (divided into 4 different age groups (adolescent, working age, middle aged, and elderly). Suspected under-reporters were excluded from final analysis | 7 day dietary records and BMI | After removing suspected under-reporters from the analysis, only the adolescent group demonstrated a significant inverse relationship between meal frequency and BMI. |
Anderson & Rossner [23] 1996) | 86 obese and 61 normal weight males (20-60 yrs) | Multiple 24 hour dietary recalls (12 total) and BMI | No significant differences in food intake patterns were observed after suspected under-reporters were excluded from final analysis (obese: n = 23; normal weight: n = 44). |
Crawley & Summerbell [24] (1997) | 298 males and 433 females (16-17 yrs) | 4 day dietary record and BMI | Initial analysis in both males and females revealed that there was a significant inverse relationship between feeding frequency and BMI. Removing suspected under-reporters still yielded a significant inverse relationship. However, after removing overweight male dieters and under-weight/normal weight females who believed they were overweight, no significant relationship between meal frequency and BMI was observed. |
Titan et al. [25] (2001) | 6,890 males and 7,776 females (45-75 yrs) | Food frequency questionnaire, BMI, waist-hip ratio (WHR), and self-reported occupational physical activity | After adjusting for confounding variables (i.e., smoking status, age, occupational activity, etc), no consistent significant association in males and females was observed when comparing individuals who ate 1-2 as compared to greater than 6 times per day to BMI or WHR. |
Bertéus Forslund et al. [26] (2002) | 83 obese and 94 normal weight reference women (37-60 yrs) | Meal pattern questionnaire and BMI | The obese women consumed a significantly greater 6.1 meals/day as opposed to the reference group (non-overweight women) which consumed 5.2 meals/day. |
Pearcey and de Castro [27] (2002) | 7 male and 12 female "weight gaining" college students and 7 males and 12 female "weight stable" matched controls (no age range reported) | 7 day food intake diary, 7 day physical activity diary, and BMI | The observed weight gain in the "weight gaining" adults was attributed to the significantly greater intake of fat, carbohydrate, and overall food per meal, but not meal frequency. |
Yannakoulia et al.[28] (2007) | 64 pre and 50 post-menopausal women (including normal weight, overweight, and obese) (24-74 yrs) (Suspected under-reporters were excluded from analysis) | 3 day food records, activity records, self-reported physical activity assessment, BMI, WHR, and body composition (dual x-ray absorptiometery) | There was no association between adiposity indices and eating frequency in pre-menopausal women, but there was a significant positive correlation between body fat percentage and meal frequency in post-menopausal women. Eating frequency was positively correlated with energy intake in both groups of women. |
Howarth et al. [2] (2007) | 1,792 younger (20-59 yrs) and 893 older (60-69 yrs) males and females (Suspected under-reporters were excluded from analysis) | Two 24 hour diet records and BMI | After adjusting for sex, age, smoking status, ethnicity, income, etc in both age groups, eating frequency was positively associated with energy intake. Older and younger individuals who ate more than three and six times a day, respectively, had a significantly higher BMI (i.e., in the overweight category) than those who ate less than three and six, respectively. |
Duval et al. [29] (2008) | 69 non-obese (BMI b/w 20-29 kg/m2), premenopausal women (48-55 yrs) (Suspected under-reporters were excluded from analysis) | 7 day food diaries, body composition (dual x-ray absorptiometry), peak VO2, resting energy expenditure (REE) via indirect calorimetry, and physical activity energy expenditure (PAEE) using an accelerometer | A significant positive correlation was observed between eating frequency and total energy intake. There was an initial significant negative correlation between eating frequency and each of the following: BMI, body fat percentage and fat mass. However, after adjusting for PAEE and peak oxygen consumption, the associations were no longer significant. |
Experimental Studies
Blood Markers of Health
-
One large meal per day
-
10 meals per day given every two hours
-
Three meals per day
Metabolism
-
Diet induced thermogenesis (i.e., DIT or also known as the thermic effect of food)
-
Resting metabolic rate/total energy expenditure
-
Protein Metabolism