Self-reported eating rate and metabolic syndrome in Japanese people: cross-sectional study

Objectives To examine the association between self-reported eating rate and metabolic syndrome. Design Cross-sectional study. Setting Annual health checkup at a health check service centre in Japan. Participants A total of 56 865 participants (41 820 male and 15 045 female) who attended a health checkup in 2011 and reported no history of coronary heart disease or stroke. Main outcome measure Metabolic syndrome was defined by the joint of interim statement of the International Diabetes Federation and the American Heart Association/National Heart, Lung, and Blood Institute. Results In multiple logistic regression models, eating rate was significantly and positively associated with metabolic syndrome. The multivariable-adjusted ORs (95% CI) for slow, normal and fast were 0.70 (0.62 to 0.79), 1.00 (reference) and 1.61 (1.53 to 1.70), respectively, in men (p for trend <0.001), and 0.74 (0.60 to 0.91), 1.00 (reference) and 1.27 (1.13 to 1.43), respectively, in women (p for trend <0.001). Of metabolic syndrome components, abdominal obesity showed the strongest association with eating rate. The associations of eating rate and metabolic syndrome and its components were largely attenuated after further adjustment for body mass index; however, the association of slow eating with lower odds of high blood pressure (men and women) and hyperglycaemia (men) and that of fast eating with higher odds of lipid abnormality (men) remained statistically significant. Conclusions Results suggest that eating rate is associated with the presence of metabolic syndrome and that this association is largely accounted for by the difference in body mass according to eating rate.

This large scale study was the first study to elucidate the association between eating rate 44 and metabolic syndrome in men and women, separately. 45 This is the only study to report the association between eating rate and metabolic 46 syndrome defined by using waist circumference. 47 Eating rate was assessed by self-report questionnaire. behaviors including eating rate. 5 Observational studies showed that obese people ate at a 58 faster rate than non-obese people, 6 and reducing eating rate may be a simple and 59 effective therapy for obesity. 7

153
The prevalence of MetS was 18.5% in men and 12.9% in women. systolic and diastolic blood pressures and lower HDL-C level in both men and women.

158
The ORs of the presence of MetS across eating rate are shown in Table 2.

159
Faster eating was associated with higher presence of MetS in age-and 160 multivariable-adjusted models. The trend was more apparent in men than in women.

161
The  with slow eating in both sexes, but they were associated with fast eating in men only.

174
Additional adjustment for BMI largely attenuated these associations, but the 175  In this large population of Japanese men and women, we found that eating rate MetS defined by using waist circumference.

187
The present finding for MetS is consistent with that of a study among Korean 188 men reporting that eating rate was positively associated with MetS, which was defined 189 by using BMI instead of waist circumference. 20  elucidated that women took more bites, smaller bite size and slower bites than men in 213 eating the same amount of doughnut, irrespectively of body size. 26 Another study 214 showed that objectively measured eating speed in men with self-reported slow eating 215 was faster than that in women with self-reported fast eating. 27 Taken together, eating 216 rate may have a greater impact on metabolism in men than that in women.

217
Although mechanisms whereby eating rate influences metabolism is not fully 218 elucidated, overeating may link fast eating to MetS. Fast eating gives few satiety signal 219 from oral cavity to the brain, 28, 29 induces less satiation and satiety due to a lack of 220 stomach expansion 30 and alters the circulating levels of certain gut hormones. 31, 32 In 221 these pathways, fast eating leads to excess energy intake, 33, 34 resulting in overweight 222 and MetS. Because fast eating has been associated with obesity even after adjusting for 223 total energy intake, [8][9][10][11]14 there may be other pathways. One study showed that 224 interleukin-1β and interleukin-6 were higher among those who ate fast than among 225 those who ate slowly, even after accounting for energy intake and BMI. 35   behaviors including eating rate. 5 Observational studies showed that obese people ate at a 59 faster rate than non-obese people, 6 and reducing eating rate may be a simple and 60 effective therapy for obesity. 7

61
During the past decade, several cross-sectional studies have found a positive 62 association between eating rate and overweight [8][9][10][11] or insulin resistance. [11][12][13][14][15] Similarly, a 63 few longitudinal studies showed that eating fast was associated with an increased risk of 64 weight gain 16,17 and type 2 diabetes. 18 In addition, some cross-sectional studies have 65 reported that fast eating was positively associated with hypertriglyceridemia and low 66 high-density lipoprotein cholesterol (HDL-C). 11,14,19 Therefore it is conceivable that 67

153
The prevalence of MetS was 18.5% in men and 12.9% in women. systolic and diastolic blood pressures and lower HDL-C level in both men and women.

158
The ORs of the presence of MetS across eating rate are shown in Table 2.

159
Faster eating was associated with higher presence of MetS in age-and 160 multivariable-adjusted models. The trend was more apparent in men than in women.

161
The   MetS defined by using waist circumference.

187
The present finding for MetS is consistent with that of a study among Korean 188 men reporting that eating rate was positively associated with MetS, which was defined 189 by using BMI instead of waist circumference. 20      behaviors including eating rate. 5 Observational studies showed that obese people ate at a 59 faster rate than non-obese people, 6 and reducing eating rate may be a simple and 60 effective therapy for obesity. 7

61
During the past decade, several cross-sectional studies have found a positive 62 association between eating rate and overweight [8][9][10][11] or insulin resistance. [11][12][13][14][15] Similarly, a 63 few longitudinal studies showed that eating fast was associated with an increased risk of 64 weight gain 16,17 and type 2 diabetes. 18 In addition, some cross-sectional studies have 65 reported that fast eating was positively associated with hypertriglyceridemia and low 66 high-density lipoprotein cholesterol (HDL-C). 11,14,19 Therefore it is conceivable that 67 was defined by using body mass index (BMI), rather than waist circumference, and

162
The prevalence of MetS was 18.5% in men and 12.8 % in women. pressures and lower HDL-C level than those who ate slowly.

168
The ORs of the presence of MetS across eating rate are shown in Table 2.

169
Faster eating was associated with higher presence of MetS in age-and 170 multivariable-adjusted models. The trend was more apparent in men than in women.

171
The  In this large population of Japanese men and women, we found that eating rate   profiles in men. To our best knowledge, the present study is the first to report a positive 198 association between eating rate and MetS defined by using waist circumference.

199
The present finding for MetS is consistent with that of a study among Korean 200 men reporting that eating rate was positively associated with MetS, which was defined 201 by using BMI instead of waist circumference. 20 As regards MetS components, our study 202 is compatible with some cross-sectional studies showing that eating rate was associated 203 with higher BMI [8][9][10][11] and two longitudinal studies showing that eating rate was 204 associated with weight gain. 16,17 In a Korean study that elucidated the association 205 between eating rate and components of MetS for men and women separately, eating rate 206 was associated with obesity, high blood pressure, hyperglycemia and abnormal lipid 207 profile in men, whereas it was associated with only obesity in women. 11 Our results 208 were largely consistent with those in the Korean study (except for blood pressure in   1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  Notably, we found that the associations of MetS components with eating rate 211 were largely attenuated after adjustment for BMI, a finding compatible with those of a 212 cross-sectional study in Korea 11 and a prospective study in Japan. 18 This result indicates 213 that obesity is a mediator whereby fast eating deteriorates MetS components. We also 214 found, however, that some associations remained statistically significant even after 215 adjusting for BMI (dyslipidemia with fast eating and hyperglycemia with slow eating in 216 men, and high blood pressure with slow eating in both men and women). Similarly, the 217 above-mentioned Korean study 11 reported that high rate of eating remained an important 218 determinant for low HDL-C and high fasting plasma glucose after adjustment for BMI 219 in men. Therefore, there may be pathways other than weight gain that might underlie the 220 association between eating rate and MetS. 221 We found that the association between eating rate and MetS was stronger in 222 men than in women, consistent with a previous study in Korea. 11 Such sex difference 223 may reflect the difference in actual eating speed between men and women. One study 224 elucidated that women took more bites, smaller bite size and slower bites than men in 225 eating the same amount of doughnut, irrespectively of body size. 27 Another study 226 showed that objectively measured eating speed in men with self-reported slow eating  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59 16 was faster than that in women with self-reported fast eating. 28 Taken together, eating 228 rate may have a greater impact on metabolism in men than that in women.
BMI=body mass index; HDL-C=high to density lipoprotein cholesterol.
a Linear regression and logistic regression, assigning ordinal number (0 to 2) to eating rate, as appropriate.

162
The prevalence of MetS was 18.5% in men and 12.8 % in women. pressures and lower HDL-C level than those who ate slowly.

168
The ORs of the presence of MetS across eating rate are shown in Table 2.

169
Faster eating was associated with higher presence of MetS in age-and 170 multivariable-adjusted models. The trend was more apparent in men than in women.