BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.
When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.
The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.
BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).
If you have any questions on BMJ Open’s open peer review process please email [email protected] BMJ Open
Fat energy intake as a potential mediator of spicy food intake frequency increased the risk of abdominal obesity in Chinese adult population: the Henan Rural Cohort Study ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2018-028736
Article Type: Original research
Date Submitted by the 21-Dec-2018 Author:
Complete List of Authors: Yang, Kaili; Henan Provincial People’s Hospital, Henan Eye Hospital, Henan Eye Institute, People's Hospital of Zhengzhou University, Zhengzhou, 450003, China; Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, Zhengzhou, Henan, PR. China. Li, Yuqian; Department of Clinical Pharmacology, School of Pharmaceutical Science, Zhengzhou University, Zhengzhou, Henan, PR. China, Xue, Yuan; Department of Nutrition and Food Hygiene, College of Public Health, Zhengzhou University, Zhengzhou, Henan, PR. China. Wang, Ling; Department of Nutrition and Food Hygiene, College of Public Health, Zhengzhou University, Zhengzhou, Henan, PR. China. Liu, Xiaotian; Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, Zhengzhou, Henan, PR. China. Tu, Runqi ; Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University Dong, Xiaokang; College of Public Health, Zhengzhou University, Department of Epidemiology and Biostatistics Mao, Zhenxing Li, Wenjie; Zhengzhou University, College of Public Health Wang, Chongjian; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health
Spicy food intake frequency, abdominal obesity, fat energy intake, Keywords: mediator analyses, rural population
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 32 BMJ Open
1 2 3 Title: 4 5 Fat energy intake as a potential mediator of spicy food intake frequency increased the 6 7 8 risk of abdominal obesity in Chinese adult population: the Henan Rural Cohort Study 9 10 Running title: 11 12 Spicy food intake and abdominal obesity 13 14 Authors: 15 16 Kaili Yang1, 2†, Yuqian Li3†, Yuan Xue4, Ling Wang4, Xiaotian Liu2, Runqi Tu2, 17 18 For peer review only 2 2 4 2* 19 Xiaokang Dong , Zhenxing Mao , Wenjie Li , Chongjian Wang 20 21 Affiliation: 22 23 1 Henan Provincial People’s Hospital, Henan Eye Hospital, Henan Eye Institute, 24 25 People's Hospital of Zhengzhou University, Zhengzhou, 450003, China. 26 27 2 Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou 28 29 30 University, Zhengzhou, Henan, PR. China. 31 3 32 Department of Clinical Pharmacology, School of Pharmaceutical Science, 33 34 Zhengzhou University, Zhengzhou, Henan, PR. China. 35 36 4 Department of Nutrition and Food Hygiene, College of Public Health, Zhengzhou 37 38 University, Zhengzhou, Henan, PR. China. 39 40 † Contributed equally to this work. 41 42 * 43 Correspondence author 44 45 Dr. Chongjian Wang 46 47 Department of Epidemiology and Biostatistics 48 49 College of Public Health, Zhengzhou University 50 51 100 Kexue Avenue, Zhengzhou, 450001, Henan, PR China 52 53 54 Phone: +86 371 67781452 55 56 Fax: +86 371 67781868 57 58 E-mail: [email protected] 59 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 32
1 2 3 Highlights 4 5 Spicy food intake frequency increased the risk of abdominal obesity in low 6 7 8 income region. 9 10 An inversed U-shaped dose-response relationship between spicy food intake 11 12 frequency and abdominal obesity was found in Chinese rural populations. 13 14 15 This is the first study that found fat energy intake may be a potenial biomarker, 16 17 which help to clarify the mechanisms and facilitate developing novel preventive 18 For peer review only 19 and therapeutic approaches for obesity in low income district. 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 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 32 BMJ Open
1 2 3 Abstract 4 5 6 Objectives Recent data that related to the relationship between spicy food intake 7 8 frequency and abdominal obesity was limited, especially in in low income areas. So 9 10 the study was to explore the relationship between spicy food intake frequency and 11 12 abdominal obesity, and then assess whether the associations were mediated by 13 14 15 different types energy intake in Chinese rural adults. 16 17 Design Cross-sectional study. 18 For peer review only 19 Setting Chinese rural population. 20 21 22 Participants subjects from the Henan Rural Cohort Study (n=28 773). 23 24 Primary outcome measures The effects of spicy food intake frequency on 25 26 abdominal obesity through restricted cubic spline and Logistic regression, and the 27 28 29 mediation effect was analyzed by bootstrap method. 30 31 Results: The adjusted percentages of abdominal obesity were 47.32%, 51.93%, 32 33 50.66%, and 50.29% in sub-groups of spicy food intake at never, 1-2 d/wk, 3-5 d/wk, 34 35 and 6-7 d/wk, respectively. The dose-response association between spicy food intake 36 37 38 frequency and abdominal obesity demonstrated an inverse U-shaped (P < 0.01). 39 40 Compared to subjects with never having spicy food, the adjusted ORs (95% CIs) of 41 42 1-2 d/wk, 3-5 d/wk and 6-7 d/wk were 1.186(1.093, 1.287), 1.127(1.038, 1.224) and 43 44 45 1.104(1.044, 1.169), respectively. Furthermore, the fat energy intake partially 46 47 mediated the relationship between spicy food and abdominal obesity, and the direct 48 49 and indirect effect were 1.107(1.053, 1.164) and 1.007(1.003, 1.012), separately. 50 51 52 Conclusions The data indicated that spicy food intake increased the risk of abdominal 53 54 obesity, and fat energy intake may be a potenial biomarker mediating this association 55 56 in Chinese rural populations. Clarifying the mechanisms will facilitate developing 57 58 novel preventive and therapeutic approaches for abdominal obesity. 59 60 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 32
1 2 3 Keywords Spicy food intake frequency, abdominal obesity, fat energy intake, 4 5 6 mediator analyses, rural population 7 8 Clinical Trial Registration 9 10 The Henan Rural Cohort Study has been registered at Chinese Clinical Trial 11 12 Register (Registration number: ChiCTR-OOC-15006699). 13 14 15 http://www.chictr.org.cn/showproj.aspx?proj=11375 16 17 18 For peer review only 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 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 32 BMJ Open
1 2 3 Strengths and limitations of this study 4 5 6 First study estimated the relationship between spicy food intake frequency and 7 8 abdominal obesity mediated by energy intake in low income dict. 9 10 Subgroup analyses according to basic characteristics were analyzed. 11 12 The causal association and it could not distinguish the effect of the spicy food by 13 14 15 itself from other accompanying factors. 16 17 18 For peer review only 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 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 32
1 2 3 Introduction 4 5 6 As a disorder of energy metabolism, obesity is defined as abnormal or excessive fat 7 8 accumulation that may impair health, of which the excessive accumulation of visceral 9 10 fat is an independent risk factor of cardiovascular diseases (1). Although great efforts 11 12 have been made to control obesity in China, the prevalence of obesity still continue 13 14 15 increasing, especially in rural areas with limited resource (2). According to the data of 16 17 China Health and Nutrition Survey (CHNS) in 2011, the prevalence of abdominal 18 For peer review only 19 obesity was 44.0% in rural China (3), and the recent study found age-standardized 20 21 (4) 22 prevalence of abdominal obesity was 43.71% in the general Chinese rural adults . 23 24 Ongoing reliable estimates are needed to establish effective programs of obesity and 25 26 related disease prevention and control programs for obesity management and reducing 27 28 29 the disease complications. It is commonly acknowledged that, besides genetic factors, 30 31 there are many modifiable risk factors contributing to obesity. Physical activity and 32 33 lifestyle factors particularly dietary behavior are considered to dramatically being 34 35 closely related with obesity, especially in developing countries (1). 36 37 38 Spicy food, as important food seasonings could improve the food’s flavor and taste, 39 40 has long been widely consumed in China (5). The consumption of spicy food affects 41 42 human health outcomes related to obesity, including several vascular and nonvascular 43 44 45 complications, such as hypertension, irritable bowel syndrome, lipid disorders, some 46 47 cancers and even mortality (6-10). Waist circumference is the most practical and 48 49 simplest index to measure the magnitude of the accumulation of fat in the abdomen. 50 51 52 An observational study provided evidence that the frequency of spicy food 53 54 consumption had presented positive associated with waist circumference (11), which 55 56 indicated that spicy food intake might play a role in the regulation of abdominal 57 58 obesity. To date, the study on exploring the associations between spicy food intake 59 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 32 BMJ Open
1 2 3 frequency and abdominal obesity in low income district is still limited. Furthermore, 4 5 6 spicy food intake could affect the energy intake, while the extra energy intake is a risk 7 8 factor for obesity. Population studies have shown that spicy food consumption inhibits 9 10 obesity by regulating energy metabolism (12), and Choi et al (13) found that the energy 11 12 intake in chili pepper nonusers was significantly lower than that in chili pepper users. 13 14 15 However, the associations among spicy food intake frequency, energy intake and 16 17 abdominal obesity have not been reported. 18 For peer review only 19 Therefore, the study aimed to illustrate the effects of spicy food intake frequency 20 21 22 on abdominal obesity in Chinese rural population, and then explore whether energy 23 24 intake mediated the relationship. 25 26 Methods 27 28 29 Study subjects 30 31 The current study subjects were from the Henan Rural Cohort Study, which was 32 33 registered on the Chinese Clinical Trial Register (Registration number: 34 35 ChiCTR-OOC-15006699). Briefly, the study was carried out in Yuzhou County in 36 37 38 Xuchang city, Suiping County in Zhumadian city, Tongxu County in Kaifeng city, 39 40 Xinxiang County in Xinxiang city and Yima County in Sanmenxia city of Henan 41 42 province in China between July 2015 and September 2017. A total of 29 867 people 43 44 45 aged 18-79 years old and with complete information of spicy food intake and waist 46 47 circumference measurements were recruited into the current study. To better estimate 48 49 the relationship between the frequency of spicy food intake and abdominal obesity, 50 51 52 365 subjects with serious diseases (heart failure, kidney failure and malignant tumor), 53 54 650 participants with hepatitis or tuberculosis or other infectious diseases, 51 women 55 56 who were pregnant or lactating, and 28 subjects under weight control management 57 58 during the last 6 months were excluded. Finally, 28 773 participants were included in 59 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 32
1 2 3 the current analysis. The study was conducted according to the Declaration of 4 5 6 Helsinki guidelines, and all procedures were approved by the “Zhengzhou University 7 8 Life Science Ethics Committee” (Code: [2015] MEC (S128)), and the informed 9 10 consent was obtained from all participants. 11 12 Assessment of spicy food intake frequency 13 14 15 A standardized questionnaire regarding dietary intake was administrated by 16 17 well-trained staffs through face to face interviews. Participants were further asked 18 For peer review only 19 “During the past month, how often did you have hot spicy foods in one week?” the 20 21 22 frequency among 0 to 7 days a week was selected by the participants. Based on the 23 24 observation of the existing epidemiological studies (14), the participants were divided 25 26 into four groups according to spicy food intake frequency: Never (reference), 1-2 27 28 29 d/wk, 3-5 d/wk and 6-7 d/wk. The test-retest reliability and the internal consistency of 30 31 questionnaire was accredited through a pilot study, and the internal consistency 32 33 Cronbach’s alpha coefficients (ICC) was 0.978 for spicy food intake frequency, which 34 35 indicated that the current variable could provide a reasonable estimation of spicy food 36 37 38 intake. 39 40 Assessment of covariates 41 42 Detailed information on socio-demographic and lifestyle factors was collected 43 44 45 through questionnaire. The following variables were included in the current study: age, 46 47 gender, education, tobacco use (Current smoker was defined as smoking at least 1 48 49 cigarette per day for 6 consecutive months), alcohol use (Current drinking was 50 51 52 defined as alcohol drinking of at least 12 times per year), and marital status. 53 54 Individual dietary intake data were collected by asking each household member to 55 56 report all food consumed category, quantity, meal type, and dining place) in a dietary 57 58 recall method and 3 day 24-h dietary record basis. For each individual, the mean total 59 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 32 BMJ Open
1 2 3 daily energy intake and the proportions from (protein, fat and carbohydrate) were 4 5 6 derived from dietary data according to the Chinese Food Composition Table. Physical 7 8 activity was divided into low, middle and high according to the international physical 9 10 activity questionnaire. Body weight with light clothing was measured to the nearest 11 12 0.1 kg using a weight measurement device (VBODY HBF-371, OMRON, Japan). 13 14 15 Height was measured to the nearest 0.1 cm without shoes using a standard right-angle 16 17 device and a fixed measurement tape. The value of body mass index (BMI) was 18 For peer review only 19 estimated as body weight (kg) divided by height square (m2). 20 21 22 Assessment of outcomes 23 24 Waist circumference was measured at a point midway between the lowest rib and 25 26 the iliac crest in a horizontal plane to the nearest 0.1 cm using non-elastic tape. All 27 28 29 measurements were conducted twice by trained research staff according to a standard 30 31 protocol from the Working Group on Obesity in China (WGOC) and the mean values 32 33 were used for the statistical analyses. Abdominal obesity was defined according to 34 35 guidelines of the International Diabetes Federation for Chinese populations as a waist 36 37 38 circumference ≥90 cm for men and ≥80 cm for women (15). 39 40 Statistical Analysis 41 42 Continuous variables were described as means ± standard deviation (SD) and 43 44 45 categorical variables were presented as proportion. Analysis of variance and 46 47 chi-square test were used to evaluate the differences of general characteristics across 48 49 four groups of spicy food intake frequency. The crude, age- and sex- adjusted 50 51 52 percentage of abdominal obesity in different groups were estimated. Restricted cubic 53 54 spline in logistic regression was used to explore the dose-response relationship 55 56 between continuous spicy food intake frequency and abdominal obesity. Logistic 57 58 regression was used to estimate the association of categorical spicy food intake 59 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 32
1 2 3 frequency for abdominal obesity based on the odds ratios (ORs) and 95% confidence 4 5 6 intervals (CIs). 7 8 The current study estimated the magnitude of change in the regression coefficient 9 10 with and without adjustment for different types of energy intake to examine the extent 11 12 of the energy intake explained by the association between spicy food intake frequency 13 14 15 and abdominal obesity. Six models were performed in the analysis. Model 1 was 16 17 crude model. Model 2 was adjusted for age, gender, education, marital status, tobacco 18 For peer review only 19 use, alcohol use and physical activity. Model 3 was adjusted for model 2 plus total 20 21 22 energy intake. Model 4 was adjusted for model 2 plus protein energy intake. Model 5 23 24 was adjusted for model 2 plus fat energy intake. Model 6 was adjusted for model 2 25 26 plus carbohydrate energy intake. The mediation analyses were conducted using the 27 28 29 bootstrap method. A conceptual model to illustrate the proposed association between 30 31 spicy food intake frequency (predictor variable), energy intake (mediator) and 32 33 abdominal obesity (dependent variable) was conducted. Stratified subgroup analyses 34 35 were performed according to the demographic characteristics and the spicy food types 36 37 38 to investigate whether the effects of spicy food intake frequency on the risk of 39 40 abdominal obesity were changed. The statistical analyses of the current data were 41 42 performed using SPSS 23.0 software package and P < 0.05 (two-tailed) was 43 44 45 considered as statistically significant. 46 47 Results 48 49 Demographic characteristics 50 51 52 Table 1 summarized the general characteristics of the participants. The mean age of 53 54 28 773 subjects was 55.39 ± 12.36 years, and the mean value of waist circumference 55 56 was 83.72±10.45cm. The characteristic of age, gender, education level, tobacco and 57 58 alcohol use, marital status, BMI, waist circumference, systemic blood pressure (SBP), 59 60 10
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 32 BMJ Open
1 2 3 diastolic blood pressure (DBP), protein energy, fat energy and carbohydrate energy 4 5 6 according to the spicy food intake frequency were different (P < 0.05). 7 8 Distribution of abdominal obesity in subgroups 9 10 The percentages of abdominal obesity by the categories of spicy flavor and spicy 11 12 food intake were depicted in Figure 1. Among the participants, there were 49.15% 13 14 15 subjects with abdominal obesity, and the crude percentages of abdominal obesity were 16 17 47.38%, 51.70%, 50.41% and 50.10% in sub-groups of spicy food intake at never, 1-2 18 For peer review only 19 d/wk, 3-5 d/wk, and 6-7 d/wk, respectively. The corresponding age- and sex- adjusted 20 21 22 percentage of abdominal obesity were 47.32%, 51.93%, 50.66%, and 50.29%, 23 24 respectively. 25 26 Association between spicy food intake frequency and abdominal obesity 27 28 29 The risk of abdominal obesity increased gradually as more frequency intake of 30 31 spicy food. The non-linear association test in six adjusted models all demonstrated 32 33 that this dose-response relationship was inversed U-shaped in all participants (P < 34 35 0.01, Figure 2). 36 37 38 Table 2 summarized the ORs of spicy food intake frequency for abdominal obesity. 39 40 Compared with those subjects with never having spicy food, the crude OR (95% CI) 41 42 of 1-2 d/wk, 3-5 d/wk and 6-7 d/wk were 1.113(1.031, 1.203), 0.992(0.918, 1.072) 43 44 45 and 1.038(0.985, 1.095), respectively. The OR (95% CI) was 1.009(0.992, 1.027) for 46 47 a one-day increment in spicy food intake frequency (Ptrend=0.308). After adjustment 48 49 for potential cofounders, the ORs (95% CIs) of 1-2 d/wk, 3-5 d/wk and 6-7 d/wk were 50 51 52 1.183(1.090, 1.283), 1.129(1.040, 1.226) and 1.118(1.057, 1.183), respectively. The 53 54 adjusted OR (95% CI) was 1.036(1.017, 1.055) for a one-day increment in spicy food 55 56 intake frequency. Further analyses were performed to explore whether the ORs (95% 57 58 CIs) were changed with and without adjustment for energy intake, and the results 59 60 11
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 32
1 2 3 showed that spicy food intake frequency were still positively associated with 4 5 6 abdominal obesity in model 3-6. With adjustment for fat energy intake, the ORs (95% 7 8 CIs) were 1.186(1.093, 1.287), 1.127(1.038, 1.224) and 1.104(1.044, 1.169), 9 10 separately. The adjusted OR (95% CI) was 1.051(1.032, 1.070) of one-day increment 11 12 in spicy food intake frequency. 13 14 15 Mediating role of energy intake 16 17 The mediate analyses were performed to explore the association between spicy 18 For peer review only 19 food intake frequency, energy intake and abdominal obesity, and the results were 20 21 22 presented in Supplementary Table 1. The direct effect and indirect effect mediated by 23 24 fat energy intake on the associations between spicy food intake frequency and 25 26 abdominal obesity were 1.107(1.053, 1.164) and1.007(1.003, 1.012), separately. The 27 28 29 adjusted OR (95% CI) of spicy food intake frequency for fat energy intake was 30 31 26.536(22.714, 30.358), and the adjusted OR (95% CI) of fat energy intake for 32 33 abdominal obesity was 1.007(1.003, 1.012). The mediator analyses indicated fat 34 35 energy intake were partial mediated the relationship between spicy food and 36 37 38 abdominal obesity (Figure 3). There were not found significant effects mediated by 39 40 protein energy intake, carbohydrate energy intake and total energy intake. 41 42 Subgroup analyses between spicy food intake frequency and abdominal obesity 43 44 45 Table 3 presents that the subgroup ORs (95% CIs) of spicy food intake frequency 46 47 for abdominal obesity. The trend associations were changed in subgroup subjects with 48 49 age below 30 years, above 61 years, male, Illiterate and Elementary, 50 51 52 Widowed/single/divorced/separation, Ex-Smoker, Smoker, Ex-drinker, Drinker, and 53 54 Low physical activity (P trend > 0.05). 55 56 Discussion 57 58 The present epidemiology survey focused on a rural population provided new 59 60 12
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 32 BMJ Open
1 2 3 evidence on the current burden of abdominal obesity in China. Overall, the prevalence 4 5 6 of abdominal obesity was much higher in Chinese rural adults than previous national 7 8 studies in China (3). There were 57.28% participants who have had spicy food, which 9 10 is consistent with previous studies demonstrating that consumption of spicy food is 11 12 popular in Chinese diet (16).Compared with those subjects with never having spicy 13 14 15 food, the participants with frequency of spicy food at 1-2 d/wk, 3-5 d/wk and 6-7 16 17 d/wk intake were positively associated with increased risk of abdominal obesity and 18 For peer review only 19 an inverse U-shaped relationship between spicy food intake frequency and abdominal 20 21 22 obesity was observed in the current study. Furthermore, the study found that fat 23 24 energy intake partially mediated the association between spicy food intake and 25 26 abdominal obesity. 27 28 29 Previous intervention studies have shown a beneficial effect of spicy food 30 31 consumption on weight management in intervention studies with a small sample size 32 33 from Western countries (17), but data is scant regarding the impact of spicy food on 34 35 obesity in Asian populations. A prospective study in CHNS data found that the 36 37 38 cumulative average chili intake was inversely associated with the risk of 39 40 overweight/obesity, which was independent of overall dietary pattern, energy intake 41 42 and lifestyle factors (18), while results of clinical trials showed that the indicators of 43 44 45 obesity were not significantly different in the study groups over baseline as well as 46 47 between the placebo and capsaicin groups (19). Additionally, our former study, which 48 49 combined cross-sectional and meta-analysis study found that spicy flavor and spicy 50 51 52 food frequency were positively associated with the general obesity in Chinese rural 53 54 populations (14). BMI and waist circumference are considered practical and effective 55 56 indexes in evaluating obesity (20). BMI is closely related to body fat (21), and could 57 58 reflect the degree of obesity deducting the influence of different heights. However, 59 60 13
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 32
1 2 3 BMI cannot accurately reflect the distribution of fat mass in the body while waist 4 5 6 circumference is the most practical and simplest indicator for evaluating abdominal 7 8 fat accumulation (22). Studes exploring the associations between spicy food intake 9 10 frequency and abdominal obesity are still limited. Soren et al (23) reported that the 11 12 abdominal adiposity decreased to a greater extent in the capsinoid group than in the 13 14 15 placebo group, although the mean change in waist girth was not significant, using 16 17 double-blind, randomized trial. The fat oxidation increased in the capsinoid group, 18 For peer review only 19 which was analogous to studies of capsinoids in animals (24).The China Kadoorie 20 21 22 Biobank (CKB) study found that the values of waist circumference increased with 23 24 spicy food intake frequency (14), which were consistent with the current results. The 25 26 potential explanation is that excessive fat intake with spicy foods may increase the 27 28 29 risk of obesity in consideration of spicy food is more meat-based in Chinese cuisines. 30 31 The non-linear association test demonstrated that the dose-response relationship 32 33 between spicy food intake frequency and abdominal obesity was inversed U-shaped in 34 35 all participants. The possible reason was that people daily having spicy food intake 36 37 38 are used to being tolerate the effects of spicy food are less likely to have more food. 39 40 However, these preliminary findings need to be further verified in large population 41 42 and multicentre study. 43 44 45 Over the past decades, a substantial amount of evidence from animal studies and 46 47 clinical trials suggest that spicy food consumption may reduce energy intake and 48 49 enhances fat oxidation (25, 26). However, the prospective study using CHNS data found 50 51 52 that high spicy food intake was positively associated with energy intake. Compared 53 54 with non-spicy food intake, the mean energy intake was greater by more than 55 56 200kcal/day in subjects consumed more than 50 gram/day of chili (18). In addition, 57 58 capsaicin is a basic component of and responsible for approximately 70% of the burn 59 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 32 BMJ Open
1 2 3 in spicy food. Janssens et al (27) found that the satiety and fullness would increase 4 5 6 when capsaicin was added into the human diet. Moreover, several human studies have 7 8 indicated that capsaicin play an important role in energy balance through changing the 9 10 intake of fat, carbohydrate, and protein (28, 29). In subgroup analyses, there were not 11 12 found trend association between spicy food intake and abdominal obesity subjects 13 14 15 above 61years, which might be considered that Henan were severely affected by 16 17 1959-1961 Chinese famine, and the potential famine effect would have effect on 18 For peer review only 19 obesity. Previous study of our team found that marital status, tobacco use, alcohol use, 20 21 22 and physical activity were associated with abdominal obesity, and female were more 23 24 probability of being abdominal obesity, which could explain that changes of the 25 26 current association in some extent (4). The variety of chili types and the lifestyles in 27 28 29 both genders partly account for the relationships between spicy food intake each day 30 31 increment and abdominal obesity in subgroup analyses. To some extent, the current 32 33 epidemiological study indicated that the spicy food might affect abdominal obesity 34 35 through increasing energy intake. Further studies should be conducted to examine the 36 37 38 detail contribution of protein, fat, and carbohydrate energy intake in the effect of 39 40 spicy food intake on abdominal obesity. 41 42 This is the first study to estimate the relationship between spicy food intake 43 44 45 frequency and abdominal obesity mediated by energy intake. However, there are 46 47 several limitations need to be noticed. Firstly, as a cross-sectional study, it was unable 48 49 to reveal a causal association and it could not distinguish the effect of the spicy food 50 51 52 by itself from other accompanying factors. Studies with long term follow-up are 53 54 needed to validate the associations. Secondly, assessing dietary behaviors in 55 56 population based studies through a food frequency questionnaire might have reporting 57 58 and recall bias, and the variety of chili types might affected the associations. Thirdly, 59 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 32
1 2 3 some residents such as college students and migrant workers were not included in the 4 5 6 current study for studying or working outside. And these groups were more likely to 7 8 be young and healthy, which might lead to the study overestimate the ratio of 9 10 abdominal obesity in the rural population. Finally, the participants of the current study 11 12 were from one province in the middle area of China, which may not be a 13 14 15 representative sample of the total Chinese rural population. However, the rural 16 17 population of Henan province accounts for 9% of total rural Chinese population. 18 For peer review only 19 Therefore, the results of this study could give some insights on further exploring the 20 21 22 association between spicy food consumption and abdominal obesity. 23 24 In conclusion, the present study indicated that the frequency of spicy food intake 25 26 was positively associated with abdominal obesity in Chinese rural residents, and fat 27 28 29 energy intake may be a potential mediator linking the increased incidence of 30 31 abdominal obesity in rural Chinese population. Therefore, multicenter, prospective 32 33 and intervention studies are needed to establish likely causality and elucidate potential 34 35 mechanisms. 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 32 BMJ Open
1 2 3 Acknowledgements 4 5 6 The authors would like to thank the participants, the coordinators, and administrators 7 8 for their supports during the study. To Tanko Abdulai, the authors would like to 9 10 express their gratitude for his critical reading of the manuscript. 11 12 13 Author contributions 14 15 C.J.W. conceived and designed the study. K.L.Y., Y.Q.L., Y.X., L.W., K.X.D., 16 17 18 Z.X.M. and W.J.L.For coordinated peer data collection.review K.L.Y., only Y.Q.L., X.T.L. and R.Q.T. 19 20 conducted the analyses. Y.Q.L. and K.L.Y. wrote the manuscript. All authors have 21 22 approved the final manuscript. 23 24 25 Funding 26 27 This research was supported by the National Key Research and Development 28 29 30 Program Precision Medicine Initiative of China (Grant NO: 2016YFC0900803), 31 32 National Natural Science Foundation of China (Grant NO: 81573243, 81602925, 33 34 U1304821), Science and Technology Foundation for Innovation Talent of Henan 35 36 Province (Grant NO: 164100510021, 154200510010), Henan Natural Science 37 38 39 Foundation (Grant NO: 182300410293), High-level Personnel Special Support 40 41 Project of Zhengzhou University (Grant NO: ZDGD13001). The funders had no role 42 43 in the study design, data collection and analysis, decision to publish, or preparation of 44 45 46 the manuscript. 47 48 Ethics approval 49 50 Ethics approval was obtained from the “Zhengzhou University Life Science Ethics 51 52 53 Committee”, and written informed consent was obtained for all participants. Ethic 54 55 approval code: [2015] MEC (S128). 56 57 58 59 60 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 32
1 2 3 Disclosure 4 5 6 All authors have completed the ICMJE uniform disclosure form at 7 8 www.icmje.org/coi_disclosure.pdf and declare: no support from any organization for 9 10 the submitted work; no financial relationships with any organizations that might have 11 12 an interest in the submitted work in the previous three years; no other relationships or 13 14 15 activities that could appear to have influenced the submitted work. 16 17 Data sharing statement 18 For peer review only 19 The datasets used and analyzed during the current study are available from the 20 21 22 corresponding author on reasonable request. 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 60 18
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 32 BMJ Open
1 2 3 References 4 5 6 1. World Health Organization . Obesity and overweight. Available at: 7 8 http://www.who.int/mediacentre/factsheets/fs311/en/. (Accessed December 2018) 9 10 2. Reynolds K, Gu D, Whelton PK, et al. InterASIA Collaborative Group. Prevalence 11 12 and risk factors of overweight and obesity in China. Obesity (Silver Spring). 2007; 13 14 15 15: 10-18. 16 17 3. Du P, Wang HJ, Zhang B, et al. Prevalence of abdominal obesity among Chinese 18 For peer review only 19 adults in 2011. J Epidemiol. 2017; 27:282-286. 20 21 22 4. Liu X, Wu W, Mao Z, et al. Prevalence and influencing factors of overweight and 23 24 obesity in a Chinese rural population: the Henan Rural Cohort Study. Sci Rep. 25 26 2018; 8(1):13101. 27 28 29 5. Li J, Wang R, Xiao C. Association between chilli food habits with iron status and 30 31 insulin resistance in a Chinese population. J Med Food. 2014; 17: 472-478. 32 33 6. Li Q, Cui YT, Jin RB, et al. Enjoyment of Spicy Flavor Enhances Central 34 35 Salty-Taste Perception and Reduces Salt Intake and Blood Pressure. Hypertension. 36 37 38 2017; 70: 1291-1299. 39 40 7. Khayyatzadeh SS, Kazemi-Bajestani SMR, Mirmousavi SJ, et al. Dietary behaviors 41 42 in relation to prevalence of irritable bowel syndrome in adolescent girls. J 43 44 45 Gastroenterol Hepatol. 2018; 33:404-410. 46 47 8. Xue Y, He T, Yu K, et al. Association between spicy food consumption and lipid 48 49 profiles in adults: a nationwide population-based study. Br J Nutr. 2017; 118: 50 51 52 144-153. 53 54 9. Aggarwal BB, Van Kuiken ME, Iyer LH, et al. Molecular targets of nutraceuticals 55 56 derived from dietary spices: potential role in suppression of inflammation and 57 58 tumorigenesis. Exp Biol Med. 2009; 234: 825-849. 59 60 19
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 32
1 2 3 10. Lv J, Qi L, Yu CQ, et al. Consumption of spicy foods and total and cause specific 4 5 6 mortality: population based cohort study. BMJ. 2015; 351:h3942. 7 8 11. Sun D, Lv J, Chen W, et al. Spicy food consumption is associated with adiposity 9 10 measures among half a million Chinese people: the China Kadoorie Biobank 11 12 study. BMC Public Health. 2014; 14: 1293. 13 14 15 12. Ludy MJ, Moore GE, Mattes RD. The effects of capsaicin and capsiate on energy 16 17 balance: critical review and meta-analyses of studies in humans. Chem Senses. 18 For peer review only 19 2012; 37: 103-121. 20 21 22 13. Choi SE, Chan J. Relationship of 6-n-propylthiouracil taste intensity and chili 23 24 pepper use with body mass index, energy intake, and fat intake within an 25 26 ethnically diverse population. J Acad Nutr Diet. 2015; 115: 389-396. 27 28 29 14. Yang K, Li Y, Mao Z, et al. Relationship between spicy flavor, spicy food intake 30 31 frequency and general obesity in a rural adult Chinese population: the RuralDiab 32 33 Study. Nutr Metab Cardiovasc Dis. 2018; 28:252-261. 34 35 15. International Diabetes Federation. The IDF consensus worldwide definition of the 36 37 38 metabolic syndrome. Available at: 39 40 https://www.idf.org/our-activities/advocacy-awareness/resources-and-tools/60:id 41 42 fconsensus-worldwide-definitionof-the-metabolic-syndrome.html. (Accessed 43 44 45 December 2018) 46 47 16. Gonlachanvit S. Are rice and spicy diet good for functional gastrointestinal 48 49 disorders? J Neurogastroenterol Motil. 2010; 16:131-138. 50 51 52 17. Whiting S, Derbyshire E, Tiwari BK. Capsaicinoids and capsinoids. A potential 53 54 role for weight management? A systematic review of the evidence. Appetite. 55 56 2012; 59:341-348.. 57 58 59 60 20
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 32 BMJ Open
1 2 3 18. Shi Z, Riley M, Taylor AW, et al. Chilli consumption and the incidence of 4 5 6 overweight and obesity in a Chinese adult population. Int J Obes (Lond). 2017; 7 8 41:1074-1079. 9 10 19. Deshpande J, Jeyakodi S, Juturu V. Tolerability of Capsaicinoids from Capsicum 11 12 Extract in a Beadlet Form: A Pilot Study. J Toxicol. 2016; 2016: 6584649. 13 14 15 20. Lean MEJ, Vlachou P, Govan L, et al. Different associations between body 16 17 composition and alcohol when assessed by exposure frequency or by quantitative 18 For peer review only 19 estimates of consumption. J Hum Nutr Diet. 2018; 31(6):747-757. 20 21 22 21. Gallagher D, VisserM, Sepulveda D, et al. How useful is body mass index for 23 24 comparison of body fatness across age, sex, and ethnic groups? Am J Epidemiol. 25 26 1996; 143: 228–239. 27 28 29 22. Misra A, Wasir JS, Vikram NK. Waist circumference criteria for the diagnosis of 30 31 abdominal obesity are not applicable uniformly to all populations and ethnic 32 33 groups. Nutrition. 2005; 21:969‐976. 34 35 23. Snitker S, Fujishima Y, Shen H, et al. Effects of novel capsinoid treatment on 36 37 38 fatness and energy metabolism in humans: possible pharmacogenetic 39 40 implications. Am J Clin Nutr. 2009; 89: 45-50. 41 42 24. Haramizu S, Mizunoya W, Masuda Y, et al. Capsiate, a Nonpungent Capsaicin 43 44 45 Analog, Increases EnduranceSwimming Capacity of Mice by Stimulation of 46 47 Vanilloid Receptors. Biosci Biotechnol Biochem. 2006; 70(4):774-781. 48 49 25. Tremblay A, Arguin H, Panahi S. Capsaicinoids: a spicy solution to the 50 51 52 management of obesity? Int. J. Obes. 2016; 40(8): 1198-204. Int J Obes (Lond). 53 54 2016; 40(8):1198-1204. 55 56 26. Varghese S, Kubatka P, Rodrigo L, et al. Chili pepper as a body weight-loss food. 57 58 Int J Food Sci Nutr. 2017; 68(4):392-401. 59 60 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 32
1 2 3 27. Janssens PL, Hursel R, Westerterp-Plantenga MS. Capsaicin increases sensation 4 5 6 of fullness in energy balance, and decreases desire to eat after dinner in negative 7 8 energy balance. Appetite. 2014; 77:44-49. 9 10 28. Yoshioka M, St-Pierre S, Drapeau V, et al. Effects of red pepper on appetite and 11 12 energy intake. Br J Nutr. 1999; 82(2): 115-123. 13 14 15 29. Westerterp-Plantenga MS, Smeets A, Lejeune MP. Sensory and gastrointestinal 16 17 satiety effect of capsaicin on food intake. Int J Obes (Lond). 2005; 29(6):682-688. 18 For peer review only 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 60 22
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 32 BMJ Open
1 2 3 Figure Legends 4 5 6 Figure 1 The percentage of abdominal obesity by different categories. 7 8 Figure 2 ORs (solid lines) and 95% CIs (dashed lines) of spicy food for abdominal 9 10 obesity from restricted cubic splines. (A: crude model, B: adjusted for age, gender, 11 12 education, marital status, tobacco use, alcohol use and physical activity, C: adjusted 13 14 15 for B plus total energy intake, D: adjusted for B plus protein energy intake, E: 16 17 adjusted for B plus fat energy intake, F: adjusted for B plus carbohydrate energy 18 For peer review only 19 intake.) 20 21 22 Figure 3 The mediation analyses between spicy food intake frequency and abdominal 23 24 obesity by fat energy intake. 25 26 Supplement Table 1 The mediation analyses between spicy food intake frequency 27 28 29 and abdominal obesity by energy intake. 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 60 23
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 32
1 2 3 Table 1 Characteristics of the participants according to the spicy food intake frequency 4 5 Variable Never 1-2 d/wk 3-5 d/wk 6-7 d/wk 6 P 7 (N=12291) (N=3252) (N=3241) (N=9989) 8 Age (years, mean ± SD) 58.19±11.68 51.58±13.21 51.46±13.56 54.46±11.64 <0.001 9 Gender, n (%) <0.001 10 11 Male 4813(39.16) 1245(38.28) 1449(44.71) 4214(42.19) 12 Female 7478(60.84) 2007(61.72) 1792(55.29) 5775(57.81) 13 14 Education, n (%) <0.001 15 Illiterate and Elementary 5948(48.39) 1093(33.61) 1126(34.74) 4550(45.55) 16 17 Middle school 4435(36.08) 1345(41.36) 1345(41.5) 4078(40.82) 18 High school and aboveFor peer1908(15.52) review814(25.03) only770(23.76) 1361(13.62) 19 20 Tobacco use, n (%) <0.001 21 Nonsmoker 9293(75.61) 2425(74.57) 2169(66.92) 6819(68.27) 22 Ex- Smoker 1033(8.4) 213(6.55) 256(7.9) 739(7.4) 23 24 Smoker 1965(15.99) 614(18.88) 816(25.18) 2431(24.34)
25 Alcohol use, n (%) <0.001 26 27 Nondrinker 10216(83.12) 2510(77.18) 2335(72.05) 7251(72.59) 28 Ex-drinker 657(5.35) 122(3.75) 143(4.41) 479(4.80) 29 30 Drinker 1418(11.54) 620(19.07) 763(23.54) 2259(22.61) 31 Physical activity, n (%) 0.091 32 Low 4376(35.60) 1117(34.35) 1028(31.72) 2601(26.04) 33 34 Middle 4365(35.51) 1048(32.23) 1215(37.49) 3959(39.63) 35 High 3550(28.88) 1087(33.43) 998(30.79) 3429(34.33) 36 37 Marital status, n (%) <0.001 38 Married/cohabiting 10844(88.23) 2972(91.39) 2943(90.81) 9161(91.71) 39 40 Widowed/single/divorced/separation 1447(11.77) 280(8.61) 298(9.19) 828(8.29) 41 BMI(kg/m2, mean ± SD) 24.57±3.51 25.00±3.62 24.84±3.66 24.86±3.56 <0.001 42 43 Waist circumference(cm, mean ± SD) 83.48±10.40 84.05±10.71 83.84±10.83 83.86±10.31 0.009 44 SBP(mmHg, mean ± SD) 127.86±20.53 125.58±19.97 124.18±19.54 123.50±19.22 <0.001 45 DBP(mmHg, mean ± SD) 77.76±11.59 78.46±11.92 77.40±11.85 76.71±11.50 <0.001 46 47 Total energy (kcal, mean ± SD) 2371.08±660.96 2371.07±649.17 2606.19±691.99 2606.19±691.99 <0.001
48 Fat energy (kcal, mean ± SD) 649.83±166.88 664.74±167.84 689.14±172.15 704.09±182.27 <0.001 49 50 Carbohydrate energy (kcal, mean ± SD) 1423.31±477.60 1407.78±472.68 1445.63±479.46 1570.50±500.73 <0.001 51 Protein energy (kcal, mean ± SD) 297.93±97.99 298.55±95.55 311.61±97.83 331.60±103.27 <0.001 52 53 Abbreviations: BMI, body mass index; SBP, systemic blood pressure; DBP, diastolic blood pressure. 54 55 56 57 58 59 60 24
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 32 BMJ Open
1 2 3 4 5 Table 2 The ORs (95% CIs) of spicy food intake frequency for abdominal obesity 6 7 Never (N=12291) 1-2 d/wk (N=3252) 3-5 d/wk (N=3241) 6-7 d/wk (N=9989) Each day increment Ptrend 8 9 Model 1 1.000 1.113(1.031, 1.203) 0.992(0.918, 1.072) 1.038(0.985, 1.095) 1.009(0.992, 1.027) 0.308 10 11 Model 2 1.000 1.183(1.090, 1.283) 1.129(1.040, 1.226) 1.118(1.057, 1.183) 1.036(1.017, 1.055) <0.001 12 For peer review only 13 Model 3 1.000 1.185(1.092, 1.286) 1.132(1.043, 1.229) 1.122(1.059, 1.187) 1.059(1.039, 1.078) <0.001 14 15 Model 4 1.000 1.185(1.092, 1.286) 1.131(1.042, 1.228) 1.117(1.055, 1.182) 1.055(1.036, 1.074) <0.001 16 17 Model 5 1.000 1.186(1.093, 1.287) 1.127(1.038, 1.224) 1.104(1.044, 1.169) 1.051(1.032, 1.070) <0.001 18 19 Model 6 1.000 1.185(1.091, 1.286) 1.133(1.043, 1.23) 1.129(1.066, 1.195) 1.059(1.040, 1.079) <0.001 20 21 Model 1: crude model. 22 23 Model 2: adjusted for age, gender, education, marital status, tobacco use, alcohol use and physical activity. 24 25 Model 3: adjusted for model 2 plus total energy intake. 26 27 Model 4: adjusted for model 2 plus protein energy intake. 28 29 Model 5: adjusted for model 2 plus fat energy intake. 30 31 Model 6: adjusted for model 2 plus carbohydrate energy intake. 32 33 34 35 36 37 38 39 40 41 25 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 26 of 32
1 2 3 4 5 Table 3 The subgroup ORs (95% CIs) of spicy food intake frequency for abdominal obesity 6 Variable Never 1-2 d/wk 3-5 d/wk 6-7 d/wk Each day increment P trend 7 Age(year) 8 <30 1.000 0.717(0.470-1.095) 0.721(0.478-1.087) 0.981(0.671-1.433) 1.003(0.886-1.136) 0.959 9 31~40 1.000 1.547(1.205-1.986) 1.331(1.032-1.716) 1.394(1.112-1.747) 1.096(1.020-1.178) 0.013 10 41~50 1.000 1.477(1.235-1.765) 1.336(1.113-1.603) 1.221(1.070-1.394) 1.057(1.013-1.104) 0.011 11 51~60 1.000 1.042(0.888-1.223) 1.140(0.970-1.340) 1.141(1.025-1.269) 1.046(1.010-1.084) 0.012 12 61~70 For1.000 peer1.406(1.188-1.665) review1.179(0.998-1.394) only1.011(0.910-1.122) 1.006(0.972-1.041) 0.738 13 >70 1.000 0.945(0.718-1.243) 1.030(0.782-1.357) 0.947(0.798-1.124) 0.985(0.931-1.041) 0.590 14 Gender 15 Male 1.000 1.334(1.167-1.524) 1.192(1.049-1.355) 0.990(0.902-1.088) 0.994(0.964-1.025) 0.697 16 Female 1.000 1.109(1.000-1.230) 1.055(0.948-1.176) 1.196(1.112-1.287) 1.058(1.033-1.083) <0.001 17 Education 18 Illiterate and Elementary 1.000 1.181(1.028-1.356) 1.028(0.896-1.178) 1.042(0.957-1.135) 1.011(0.983-1.04) 0.442 19 Middle school 1.000 1.251(1.100-1.422) 1.146(1.008-1.303) 1.113(1.017-1.219) 1.032(1.002-1.063) 0.037 20 High school and above 1.000 1.173(0.986-1.396) 1.34(1.120-1.604) 1.168(1.008-1.353) 1.061(1.012-1.113) 0.015 21 Marital status 22 Married/cohabiting 1.000 1.188(1.090-1.294) 1.144(1.05-1.247) 1.108(1.044-1.176) 1.033(1.013-1.053) 0.001 23 Widowed/single / divorced/separation 1.000 1.122(0.842-1.495) 0.944(0.705-1.263) 1.093(0.900-1.327) 1.024(0.961-1.091) 0.467 24 Tobacco use 25 Nonsmoker 1.000 1.154(1.050-1.269) 1.110(1.006-1.225) 1.158(1.083-1.238) 1.048(1.025-1.071) <0.001 26 Ex- Smoker 1.000 1.074(0.787-1.465) 1.235(0.928-1.644) 1.097(0.895-1.345) 1.037(0.971-1.109) 0.281 27 Smoker 1.000 1.386(1.138-1.687) 1.094(0.912-1.312) 0.941(0.821-1.079) 0.968(0.926-1.012) 0.150 28 Alcohol use 29 Nondrinker 1.000 1.155(1.053-1.268) 1.045(0.950-1.150) 1.120(1.050-1.194) 1.035(1.013-1.057) 0.002 30 Ex-drinker 1.000 1.077(0.710-1.634) 1.211(0.824-1.780) 1.098(0.847-1.424) 1.036(0.952-1.128) 0.413 31 Drinker 1.000 1.363(1.118-1.662) 1.347(1.117-1.623) 1.069(0.926-1.233) 1.010(0.965-1.058) 0.668 32 Physical activity 33 Low 1.000 1.176(1.023-1.353) 1.222(1.056-1.413) 1.070(0.964-1.188) 1.028(0.993-1.063) 0.116 34 Middle 1.000 1.270(1.103-1.464) 1.098(0.961-1.255) 1.127(1.029-1.234) 1.035(1.005-1.067) 0.023 35 High 1.000 1.115(0.963-1.29) 1.059(0.909-1.234) 1.105(0.998-1.224) 1.031(0.997-1.066) 0.076 36 Adjusted model (except where it is the variable of interest): adjusted for age, gender, education, marital status, tobacco use, alcohol use, physical activity and fat energy intake. 37 38 39 40 41 26 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 27 of 32 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 1 The percentage of abdominal obesity by different categories. 33 34 99x78mm (300 x 300 DPI) 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 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 32
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 Figure 2 ORs (solid lines) and 95% CIs (dashed lines) of spicy food for abdominal obesity from restricted 22 cubic splines. (A: crude model, B: adjusted for age, gender, education, marital status, tobacco use, alcohol 23 use and physical activity, C: adjusted for B plus total energy intake, D: adjusted for B plus protein energy 24 intake, E: adjusted for B plus fat energy intake, F: adjusted for B plus carbohydrate energy intake.) 25 26 631x283mm (300 x 300 DPI) 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 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 32 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 Figure 3 The mediation analyses between spicy food intake frequency and abdominal obesity by fat energy 22 intake 23 24 184x83mm (300 x 300 DPI) 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 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 32
1 2 3 Supplement Table 1 The mediation analyses between spicy food intake frequency and abdominal obesity by energy intake. 4 5 Energy intake Direct effect, OR(95%CI) Path a, β(95%CI) Path b, OR(95%CI) Indirect effect, OR(95%CI) 6 1.047(1.028, 1.067) 7 Total 64.316(58.799, 69.832) 1.001(0.999, 1.002) 0.999(0.996, 1.001) 8 Protein 1.117(1.063, 1.175) 18.501(16.190, 20.812) 0.999(0.996, 1.001) 0.997(0.993,1.002) 9 1.107(1.053, 1.164) 10 Fat 26.536(22.714, 30.358) 1.003(1.001, 1.004) 1.007(1.003, 1.012) 11 Carbohydrate 1.181(1.130, 1.234) 8.955(7.826, 10.834) 1.001(0.999, 1.003) 1.001(0.998, 1.004) 12 Adjusted model: adjusted for age, gender, education, marital status, tobacco use, alcohol use, and physical activity. 13 14 15 16 17 18 For peer review only 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 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 32 BMJ Open
1 2 STROBE 2007 (v4) Statement—Checklist of items that should be included in reports of cross-sectional studies 3
4 5 Item Section/Topic Recommendation Reported on page # 6 # 7 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 3 8 9 (b) Provide in the abstract an informative and balanced summary of what was done and what was found 2 10 11 Introduction 12 Background/rationale 2 Explain the scientificFor background peer and rationale forreview the investigation being reported only 6 13 Objectives 3 State specific objectives, including any prespecified hypotheses 7 14 15 Methods 16 Study design 4 Present key elements of study design early in the paper 7 17 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data 7 19 collection 20 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of participants 7 21
22 23 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if 8, 9 24 25 applicable 26 Data sources/ 8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe 8, 9 27 measurement comparability of assessment methods if there is more than one group 28 Bias 9 Describe any efforts to address potential sources of bias 9 29 30 Study size 10 Explain how the study size was arrived at 7 31 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and 8, 9 32 why 33 34 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 9, 10 35 (b) Describe any methods used to examine subgroups and interactions 10 36
37 (c) Explain how missing data were addressed 38 (d) If applicable, describe analytical methods taking account of sampling strategy 10 39 (e) Describe any sensitivity analyses 10 40 41 Results 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 32 of 32
1 2 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, 10 3 confirmed eligible, included in the study, completing follow-up, and analysed 4 5 (b) Give reasons for non-participation at each stage 6 (c) Consider use of a flow diagram 7 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential 10, 11 8 9 confounders 10 (b) Indicate number of participants with missing data for each variable of interest 24 11 Outcome data 15* Report numbers of outcome events or summary measures 11 12 Main results 16 (a) Give unadjustedFor estimates peer and, if applicable, reviewconfounder-adjusted estimates only and their precision (eg, 95% confidence 11, 12 13 14 interval). Make clear which confounders were adjusted for and why they were included 15 (b) Report category boundaries when continuous variables were categorized 9 16 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period 17 18 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses 11, 12 19 Discussion 20 Key results 18 Summarise key results with reference to study objectives 13 21 22 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and 15, 16 23 magnitude of any potential bias 24 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from 13, 14, 15 25 similar studies, and other relevant evidence 26 27 Generalisability 21 Discuss the generalisability (external validity) of the study results 14, 15 28 Other information 29 30 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on 17 31 which the present article is based 32 33 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies. 34 35 36 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE 37 checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 38 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org. 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open
Fat energy intake as a potential mediator of spicy food intake frequency increased the risk of abdominal obesity based on a cross-sectional study in Chinese adult population: the Henan Rural Cohort Study
ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2018-028736.R1
Article Type: Original research
Date Submitted by the 11-Jul-2019 Author:
Complete List of Authors: Yang, Kaili; Henan Provincial People’s Hospital, Henan Eye Hospital, Henan Eye Institute, People's Hospital of Zhengzhou University, Zhengzhou, 450003, China; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Li, Yuqian; Zhengzhou University, Department of Clinical Pharmacology, School of Pharmaceutical Science Xue, Yuan; Zhengzhou University, Department of Nutrition and Food Hygiene, College of Public Health Wang, Ling; Zhengzhou University, Department of Nutrition and Food Hygiene, College of Public Health Liu, Xiaotian; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Tu, Runqi ; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Dong, Xiaokang; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Mao, Zhenxing ; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Li, Wenjie; Zhengzhou University, Department of Nutrition and Food Hygiene, College of Public Health Wang, Chongjian; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health
Primary Subject Nutrition and metabolism Heading:
Secondary Subject Heading: Epidemiology, Nutrition and metabolism, Public health
Spicy food intake frequency, abdominal obesity, fat energy intake, Keywords: mediator analyses, rural population
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 33 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 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 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 33
1 2 3 Title: 4 5 Fat energy intake as a potential mediator of spicy food intake frequency increased the 6 7 8 risk of abdominal obesity based on a cross-sectional study in Chinese adult 9 10 population: the Henan Rural Cohort Study 11 12 Running title: 13 14 Spicy food intake and abdominal obesity 15 16 Authors: 17 18 Kaili Yang1, 2†, ForYuqian Lipeer3†, Yuan Xue review4, Ling Wang4, Xiaotianonly Liu2, Runqi Tu2, 19 20 Xiaokang Dong2, Zhenxing Mao2, Wenjie Li4, Chongjian Wang2* 21 22 Affiliation: 23 24 1 25 Henan Provincial People’s Hospital, Henan Eye Hospital, Henan Eye Institute, 26 27 People's Hospital of Zhengzhou University, Zhengzhou, 450003, China. 28 29 2 Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou 30 31 University, Zhengzhou, Henan, PR. China. 32 33 3 Department of Clinical Pharmacology, School of Pharmaceutical Science, 34 35 Zhengzhou University, Zhengzhou, Henan, PR. China. 36 37 4 Department of Nutrition and Food Hygiene, College of Public Health, Zhengzhou 38 39 University, Zhengzhou, Henan, PR. China. 40 41 † 42 Contributed equally to this work. 43 * 44 Correspondence author 45 46 Dr. Chongjian Wang 47 48 Department of Epidemiology and Biostatistics 49 50 College of Public Health, Zhengzhou University 51 52 100 Kexue Avenue, Zhengzhou, 450001, Henan, PR China 53 54 Phone: +86 371 67781452 55 56 Fax: +86 371 67781868 57 58 59 E-mail: [email protected] 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 33 BMJ Open
1 2 3 Abstract 4 5 Objectives Recent data that related to the association between spicy food intake 6 7 8 frequency and abdominal obesity was limited, especially in in low income areas. So 9 10 the study aimed to explore the relationship between spicy food intake frequency and 11 12 abdominal obesity, and then assess whether the associations were mediated by energy 13 14 15 intake in Chinese rural adults. 16 17 Design Cross-sectional study. 18 For peer review only 19 Setting Chinese rural population. 20 21 22 Participants Subjects from Henan Rural Cohort Study (n=28 773). 23 24 Primary outcome measures The effects of spicy food intake frequency on 25 26 abdominal obesity were presented by restricted cubic spline and Logistic regression, 27 28 and the mediation effect was analyzed through bootstrap method. 29 30 31 Results: The numbers of abdominal obesity were 5997(48.79%), 1674(51.48%), 32 33 1575(48.60%), and 4968(49.73%) in sub-groups of spicy food intake at never, 1-2 34 35 d/wk, 3-5 d/wk, and 6-7 d/wk, respectively. An inverse U-shaped association was 36 37 38 found between spicy food intake frequency and abdominal obesity (P < 0.01). 39 40 Compared to subjects who never had spicy food, the adjusted ORs (95% CIs) of 1-2 41 42 d/wk, 3-5 d/wk and 6-7 d/wk were 1.186(1.093, 1.287), 1.127(1.038, 1.224) and 43 44 45 1.104(1.044, 1.169), respectively. Furthermore, the fat energy intake partially 46 47 mediated the relationship between spicy food intake frequency and abdominal 48 49 obesity, and the direct and indirect effect were 1.107(1.053, 1.164) and 1.007(1.003, 50 51 1.012), separately. 52 53 54 Conclusions The data indicated that spicy food intake increased the risk of abdominal 55 56 obesity, and fat energy intake may be a mediator of this association in Chinese rural 57 58 59 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 33
1 2 3 populations. Clarifying the mechanisms will facilitate developing novel preventive 4 5 6 and therapeutic approaches for abdominal obesity. 7 8 Keywords Spicy food intake frequency, abdominal obesity, fat energy intake, 9 10 mediator analyses, rural population 11 12 Clinical Trial Registration 13 14 15 The Henan Rural Cohort Study has been registered at Chinese Clinical Trial 16 17 Register (Registration number: ChiCTR-OOC-15006699). 18 For peer review only 19 http://www.chictr.org.cn/showproj.aspx?proj=11375 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 60 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 33 BMJ Open
1 2 3 Strengths and limitations of this study 4 5 6 This is the first study to estimate the relationship between spicy food intake 7 8 frequency and abdominal obesity mediated by energy intake. 9 10 A large, population-based sample. 11 12 The study included available information on a broad range of covariates. 13 14 15 An inversed U-shaped relationship between spicy food intake frequency and 16 17 abdominal obesity was found in Chinese rural populations. 18 For peer review only 19 The survey did not provide relevant causal association between spicy food intake 20 21 22 frequency and abdominal obesity. 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 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 33
1 2 3 Introduction 4 5 6 As a disorder of energy metabolism, obesity is defined as abnormal or excessive fat 7 8 accumulation that may impair health, and the excessive accumulation of visceral fat is 9 10 an independent risk factor of cardiovascular diseases (1). Although great efforts have 11 12 been made to control weight in China, the prevalence of obesity still continue 13 14 15 increased, especially in rural areas (2). According to the data of China Health and 16 17 Nutrition Survey (CHNS) in 2011, the prevalence of abdominal obesity was 44.0% in 18 For peer review only 19 rural China (3), and recent study also found age-standardized prevalence of abdominal 20 21 (4) 22 obesity was 43.71% in Chinese rural adults . Ongoing reliable programs are needed 23 24 to manage obesity and reduce the disease complications. It is commonly 25 26 acknowledged that, there are many modifiable risk factors contributing to obesity 27 28 29 besides genetic factors. Physical activity and lifestyle factors particularly dietary 30 31 behaviors are considered as being closely related with obesity, especially in 32 33 developing countries (1). 34 35 Spicy food, which could improve the food’s flavor and taste, has long been widely 36 37 38 consumed in China (5). Several studies have reported that the consumption of spicy 39 40 food could affect human health outcomes related to obesity, such as hypertension, 41 42 irritable bowel syndrome, lipid disorders, cancers and even mortality (6-10). As the 43 44 45 most practical and simplest index, waist circumference is used to measure the 46 47 magnitude of fat accumulation in the abdomen. An observational study conducted in 48 49 five urban and five rural areas of China provided evidence that the frequency of spicy 50 51 (11) 52 food consumption was positive associated with waist circumference , which 53 54 indicated that spicy food intake might play a role in the regulation of abdominal 55 56 obesity. According to the sixth national population census, the rural districts account 57 58 for 63.90% of China population, and the education level, economic development, 59 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 33 BMJ Open
1 2 3 behavior factors, food choices and dietary pattern in rural areas were different from 4 5 (12) 6 urban areas of China . Thus, this is meaningful to explore the associations between 7 8 spicy food intake frequency and abdominal obesity in rural populations. Furthermore, 9 10 population studies have shown that spicy food consumption inhibits obesity by 11 12 regulating energy metabolism (13), and Choi et al (14) found that the energy intake in 13 14 15 chili pepper nonusers was significantly lower than that in chili pepper users in the 16 17 New York City area. But the associations among spicy food intake frequency, energy 18 For peer review only 19 intake and abdominal obesity have not been reported. 20 21 22 Therefore, the study aimed to illustrate the effects of spicy food intake frequency 23 24 on abdominal obesity in Chinese rural population, and then explore whether energy 25 26 intake could mediate the relationship. 27 28 29 Methods 30 31 Study subjects 32 33 The current study subjects were from the Henan Rural Cohort Study (15), which was 34 35 registered on the Chinese Clinical Trial Register (Registration number: 36 37 38 ChiCTR-OOC-15006699). Briefly, the study was carried out in Yuzhou County in 39 40 Xuchang city, Suiping County in Zhumadian city, Tongxu County in Kaifeng city, 41 42 Xinxiang County in Xinxiang city and Yima County in Sanmenxia city of Henan 43 44 45 province in China between July 2015 and September 2017. A total of 29 867 people 46 47 aged 18-79 years old and with complete information of spicy food intake and waist 48 49 circumference measurements were recruited into the current study. To better estimate 50 51 52 the relationship between the frequency of spicy food intake and abdominal obesity, 53 54 365 subjects with serious diseases (heart failure, kidney failure and malignant tumor), 55 56 650 participants with hepatitis or tuberculosis or other infectious diseases, 51 women 57 58 who were pregnant or lactating, and 28 subjects under weight control management 59 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 33
1 2 3 during the last 6 months were excluded. Finally, 28 773 participants were included in 4 5 6 the current analysis. The study was conducted according to the Declaration of 7 8 Helsinki guidelines, and all procedures were approved by the “Zhengzhou University 9 10 Life Science Ethics Committee” (Code: [2015] MEC (S128)), and the informed 11 12 consent was obtained from all participants. 13 14 15 Patient and public involvement 16 17 Neither patients nor the public were involved in developing this project. 18 For peer review only 19 Assessment of spicy food intake frequency 20 21 22 Food Frequency Questionnaire (FFQ) regarding dietary intake was administrated 23 24 by well-trained staffs through face to face interviews. Participants were further asked 25 26 “During the past month, how often did you have spicy foods in one week?” the 27 28 29 frequency among 0 to 7 days a week was selected by the participants. Based on the 30 31 observation of the existing epidemiological studies (16), the participants were divided 32 33 into four groups according to spicy food intake frequency: Never (reference), 1-2 34 35 d/wk, 3-5 d/wk and 6-7 d/wk. The types of foods were asked to confirm whether they 36 37 38 belong to spicy foods. The test-retest reliability and the internal consistency of 39 40 questionnaire was accredited through a pilot study with 76 subjects, and the internal 41 42 consistency Cronbach’s alpha coefficients (ICC) was 0.978, which indicated that the 43 44 45 current variable could provide a reasonable estimation of spicy food intake. 46 47 Assessment of covariates 48 49 Detailed information on socio-demographic and lifestyle factors was collected 50 51 52 through questionnaire. The following variables were included in the current study: 53 54 age, gender, education, tobacco use (Current smoker was defined as smoking at least 55 56 1 cigarette per day for 6 consecutive months), alcohol use (Current drinking was 57 58 defined as alcohol drinking of at least 12 times per year), and marital status. 59 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 33 BMJ Open
1 2 3 Individual dietary intake data was collected by asking each household member to 4 5 6 report all food consumed category, quantity, meal type, and dining place) in a dietary 7 8 recall method. The quantity of food consumed including staple food, livestock, 9 10 poultry, fish, eggs, dairy, fruits, vegetables, beans, nuts, pickles, cereal and animal oil 11 12 over the previous year was recorded. For each individual, the mean of total daily 13 14 15 energy intake and the proportions of protein, fat and carbohydrate were derived from 16 17 dietary data according to the Chinese Food Composition Table (2009). The 3 day 24-h 18 For peer review only 19 record was conducted in a small sample to valid the FFQ and the results demonstrated 20 21 22 that the questionnaire could be used as a representative tool to conduct a dietary 23 24 evaluation of a rural population (17). Physical activity was divided into low, middle 25 26 and high according to the international physical activity questionnaire (18). Blood 27 28 29 pressure on the right arm supported at heart-level with sitting position was measured 30 31 by electronic sphygmomanometers (Omron HEM-7071A, Japan) in triplicate. The 32 33 average of the three measures was used for analysis. Body weight with light clothing 34 35 was measured to the nearest 0.1 kg using a weight measurement device (VBODY 36 37 38 HBF-371, OMRON, Japan). Height was measured to the nearest 0.1 cm without shoes 39 40 using a standard right-angle device and a fixed measurement tape. The value of body 41 42 mass index (BMI) was estimated as body weight (kg) divided by height square (m2). 43 44 45 Assessment of outcomes 46 47 Waist circumference was measured at a point midway between the lowest rib and 48 49 the iliac crest in a horizontal plane to the nearest 0.1 cm using non-elastic tape. All 50 51 52 measurements were conducted twice by trained research staff according to a standard 53 54 protocol from the Working Group on Obesity in China (WGOC) and the mean values 55 56 were used for statistical analyses. Abdominal obesity was defined as a waist 57 58 59 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 33
1 2 3 circumference ≥90 cm for men and ≥80 cm for women according to the guidelines 4 5 6 of the International Diabetes Federation for Chinese populations (19). 7 8 Statistical Analysis 9 10 Continuous variables were described as means ± standard deviation (SD) and 11 12 13 categorical variables were presented as proportion. Analysis of variance and 14 15 chi-square test were used to evaluate the differences of general characteristics of four 16 17 groups of spicy food intake. The crude, age- and sex- adjusted percentage of 18 For peer review only 19 20 abdominal obesity in different groups were estimated. Restricted cubic spline was 21 22 used to explore the dose-response relationship between continuous spicy food intake 23 24 frequency and abdominal obesity. Logistic regression analysis was used to estimate 25 26 27 the association of categorical spicy food intake frequency and abdominal obesity 28 29 based on the odds ratios (ORs) and 95% confidence intervals (CIs). 30 31 To examine the extent of the association between spicy food intake frequency and 32 33 abdominal obesity mediated by energy intake, the current study estimated the 34 35 36 magnitude of change in the regression coefficient with and without adjustment for 37 38 different types of energy intake. Six models were performed in the analyses. Model 1 39 40 was the crude model. Model 2 was adjusted for age, gender, education, marital status, 41 42 43 tobacco use, alcohol use and physical activity. Model 3 was adjusted for model 2 plus 44 45 total energy intake. Model 4 was adjusted for model 2 plus protein energy intake. 46 47 Model 5 was adjusted for model 2 plus fat energy intake. Model 6 was adjusted for 48 49 50 model 2 plus carbohydrate energy intake. The mediation analyses were conducted 51 52 through bootstrap method. A conceptual model to illustrate the proposed association 53 54 between spicy food intake frequency (predictor variable), energy intake (mediator) 55 56 and abdominal obesity (dependent variable) was conducted. Stratified subgroup 57 58 59 analyses were performed, according to the demographic characteristics and the types 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 33 BMJ Open
1 2 3 of spicy food, to investigate whether the effects of spicy food intake frequency on the 4 5 6 risk of abdominal obesity were changed. The statistical analyses of the current data 7 8 were performed using SPSS 23.0 software package and P < 0.05 (two-tailed) was 9 10 considered as statistically significant. 11 12 Results 13 14 15 Demographic characteristics 16 17 Table 1 summarized the general characteristics of the participants. The mean age of 18 For peer review only 19 28 773 subjects was 55.39 ± 12.36 years, and the mean value of waist circumference 20 21 22 were 85.44±10.55 cm for male and 82.54±10.22 cm for female. The characteristic of 23 24 age, gender, education level, tobacco and alcohol use, marital status, BMI, waist 25 26 circumference, systemic blood pressure (SBP), diastolic blood pressure (DBP), 27 28 29 protein energy, fat energy and carbohydrate energy were different according to the 30 31 spicy food intake frequency groups (P < 0.05). 32 33 Distribution of abdominal obesity in subgroups 34 35 The percentages of abdominal obesity by the categories of spicy flavor and spicy 36 37 38 food intake frequency were depicted in Figure 1. Among the participants, there were 39 40 49.15% subjects with abdominal obesity, and the numbers of abdominal obesity were 41 42 5997(48.79%), 1674(51.48%), 1575(48.60%), and 4968(49.73%) in sub-groups of 43 44 45 spicy food intake at never, 1-2 d/wk, 3-5 d/wk, and 6-7 d/wk, respectively. The 46 47 corresponding age- and sex- adjusted percentage of abdominal obesity were 47.32%, 48 49 51.93%, 50.66%, and 50.29%, respectively. 50 51 52 Association between spicy food intake frequency and abdominal obesity 53 54 The ORs of spicy food intake frequency for abdominal obesity showed first went 55 56 up and then declined after reaching the peak. The inversed U-shaped dose-response 57 58 relationships were demonstrated in all non-linear association tests (P < 0.01, Figure 59 60 10
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 33
1 2 3 2). 4 5 6 Table 2 summarized the ORs of spicy food intake frequency for abdominal obesity. 7 8 Taking the subjects who never consumed spicy food as the reference group, the crude 9 10 ORs (95% CIs) for 1-2 d/wk, 3-5 d/wk and 6-7 d/wk were 1.113(1.031, 1.203), 11 12 0.992(0.918, 1.072) and 1.038(0.985, 1.095), respectively. The ORs (95% CIs) was 13 14 15 1.009(0.992, 1.027) for one-day increment in the frequency of spicy food intake 16 17 frequency (Ptrend=0.308). After adjustment for potential cofounders, the ORs (95% 18 For peer review only 19 CIs) of 1-2 d/wk, 3-5 d/wk and 6-7 d/wk were 1.183(1.090, 1.283), 1.129(1.040, 20 21 22 1.226) and 1.118(1.057, 1.183), respectively. The adjusted OR (95% CI) was 23 24 1.036(1.017, 1.055) for one-day increment in the frequency of spicy food intake. 25 26 Further analyses were performed to explore whether the ORs (95% CIs) were changed 27 28 29 with and without adjustment for energy intake, and the results showed that spicy food 30 31 intake frequency were still positively associated with abdominal obesity in model 3-6. 32 33 With adjusting for fat energy intake, the ORs (95% CIs) were 1.186(1.093, 1.287), 34 35 1.127(1.038, 1.224) and 1.104(1.044, 1.169), separately. The adjusted OR (95% CI) 36 37 38 was 1.051(1.032, 1.070) for one-day increment in the frequency of spicy food intake. 39 40 Mediating role of energy intake 41 42 The mediation analyses were performed to explore the association among the 43 44 45 frequency of spicy food intake, energy intake and abdominal obesity, and the related 46 47 results were presented in Supplementary Table 1. The direct effect and indirect effect 48 49 mediated by fat energy intake on the associations between spicy food intake 50 51 52 frequency and abdominal obesity were 1.107(1.053, 1.164) and1.007(1.003, 1.012), 53 54 separately. The adjusted OR (95% CI) of spicy food intake frequency for fat energy 55 56 intake was 26.536(22.714, 30.358), and the adjusted OR (95% CI) of fat energy intake 57 58 for abdominal obesity was 1.007(1.003, 1.012). The mediation analyses indicated fat 59 60 11
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 33 BMJ Open
1 2 3 energy intake partially mediated the relationship between spicy food intake frequency 4 5 6 and abdominal obesity (Figure 3). There were no significant effects mediated by 7 8 protein energy intake, carbohydrate energy intake and total energy intake. 9 10 Subgroup analyses between spicy food intake frequency and abdominal obesity 11 12 Table 3 presents that the subgroup ORs (95% CIs) of spicy food intake frequency 13 14 15 for abdominal obesity. There were no significantly statistical trend associations in 16 17 subgroup subjects with age below 30 years, above 61 years, male, illiterate and 18 For peer review only 19 Elementary, widowed/single/divorced/separation, ex-smoker, smoker, ex-drinker, 20 21 22 drinker, and low physical activity (P trend > 0.05). 23 24 Discussion 25 26 The present epidemiology survey provided the new evidences on the current burden 27 28 29 of abdominal obesity in Chinese rural population. Overall, the prevalence of 30 31 abdominal obesity was higher in Chinese rural adults than previous national studies in 32 33 China (3). There were 57.28% of participants who consumed spicy food, which 34 35 demonstrated that the consumption of spicy food is popular in Chinese diet 36 37 38 (20).Compared with subjects who never had spicy food, the participants with 39 40 consuming spicy food at 1-2 d/wk, 3-5 d/wk and 6-7 d/wk were positively associated 41 42 with the increased risk of abdominal obesity. Additionally an inverse U-shaped 43 44 45 relationship between spicy food intake frequency and abdominal obesity was 46 47 observed in the current study. Furthermore, the study also found fat energy intake 48 49 partially mediated the association between spicy food intake frequency and abdominal 50 51 52 obesity. 53 54 Previous intervention study has shown a beneficial effect of spicy food 55 56 consumption on weight management in a small sample size from Western countries 57 58 (21), but data is scant regarding the effect of spicy food on obesity in Asian 59 60 12
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 33
1 2 3 populations. A prospective study in CHNS data found that the cumulative average 4 5 6 chili intake was inversely associated with the risk of overweight/obesity, which was 7 8 independent of overall dietary pattern, energy intake and lifestyle factors (22). While 9 10 results of clinical trial showed that the indicators of obesity were not significantly 11 12 different between the placebo and capsaicin groups (23). Additionally, the former study 13 14 15 combining cross-sectional and meta-analysis study found that spicy food frequency 16 17 were positively associated with the general obesity in Chinese rural populations (16). 18 For peer review only 19 Both the BMI and waist circumference are considered as practical and effective 20 21 (24) (25) 22 indexes in evaluating obesity . BMI is closely related to body fat , and could 23 24 reflect the degree of obesity deducting the influence of different heights. However, 25 26 BMI cannot accurately reflect the distribution of fat mass in the body while waist 27 28 29 circumference is regarded as the most practical and simplest indicator for evaluating 30 31 abdominal fat accumulation (26). Studes exploring the associations between spicy food 32 33 intake frequency and abdominal obesity are still limited. Snitker et al (27) reported that 34 35 the abdominal adiposity decreased to a greater extent in capsinoid group than in 36 37 38 placebo group, although the mean change in waist girth was not significant. The fat 39 40 oxidation increased in the capsinoid group, which was analogous to the animal study 41 42 conducted by Haramizu et al (28). Although the Snitker et al and Haramizu et al 43 44 45 investigations were conducted using capsinoids, which are non-pungent 46 47 capsaicin-related substances, with the same structure might help to understand the 48 49 effects of pungent capsaicinoid analogs. The China Kadoorie Biobank (CKB) study 50 51 52 found that the values of waist circumference increased with spicy food intake 53 54 frequency increasing (11), which was consistent with the current results. The non-linear 55 56 association tests demonstrated that the dose-response relationship between spicy food 57 58 intake frequency and abdominal obesity was inversed U-shaped in all participants. 59 60 13
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 33 BMJ Open
1 2 3 The possible reason was that people consumed the daily spicy food tend to have a 4 5 6 better degree of tolerance of spicy food and were less likely to have more other foods 7 8 to relieve the hot. However, these preliminary findings need to be further verified in 9 10 other population and multicenter study. 11 12 Over the past decades, a substantial amount of evidence from animal studies and 13 14 15 clinical trials suggest that spicy food consumption may reduce energy intake and 16 17 enhance fat oxidation (29, 30). However, the prospective study using CHNS data found 18 For peer review only 19 that high spicy food intake was positively associated with energy intake. Compared 20 21 22 with non-spicy food intake, the mean energy intake was greater with more than 23 24 200kcal/day in subjects consumed more than 50 gram/day of chili (22). In addition, 25 26 capsaicin is a basic component and responsible for approximately 70% of the burn in 27 28 (31) 29 spicy food. Janssens et al found that the satiety and fullness would increase when 30 31 capsaicin was added into the human diet. Moreover, several human studies have 32 33 indicated that capsaicin play an important role in energy balance through changing the 34 35 intake of fat, carbohydrate, and protein (32, 33). Though metabolic studies indicated a 36 37 38 lipolytic effect of spices, the study showed that fat energy intake mediated the 39 40 association between spicy food intake frequency and abdominal obesity. The possible 41 42 mechanism underlying the cholesterol-lowering activity was that spicy food could 43 44 45 stimulate fecal excretion of bile acids and increases the resistance of serum 46 47 lipoproteins to oxidation (34). While the effect of spicy food on weight management 48 49 regarding the mechanisms underlying energy intake balance and appetite changes 50 51 (35) (36) 52 except of lipid metabolism . In addition, Kawada et al found the 53 54 supplementation of capsaicin lowered serum triacylglycerol concentration and 55 56 stimulated lipid mobilization from adipose tissue but not affected lipogenesis in obese 57 58 rats. As a less pungent structural analog of capsaicin, nonivamide could prevent 59 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 33
1 2 3 weight gain without affecting blood total cholesterol, low-density lipoprotein 4 5 (37) 6 cholesterol, or triglycerides in a 12-week intervention trial . Considering the 7 8 defects of the cross-sectional study, further investigations are warranted to explore the 9 10 pathophysiological mechanisms underlying the effect of spicy food consuming on 11 12 abdominal obesity and lipid metabolism. In subgroup analyses, there were no trends 13 14 15 toward the association between spicy food intake frequency and abdominal obesity 16 17 subjects above 61years. The possible reason might be considered that adults in Henan 18 For peer review only 19 were severely affected by Chinese famine occurred in 1959-1961, and the potential 20 21 22 famine effect would have effect on the prevalence of obesity. Previous study of our 23 24 team found that marital status, tobacco use, alcohol use, and physical activity were 25 26 associated with abdominal obesity, and female tended to have the abdominal obesity, 27 28 (4) 29 which could explain the associations of subgroup to some extent . The variety of 30 31 chili types and the lifestyles in both genders partly account for the relationships 32 33 between spicy food intake each day increment and abdominal obesity in subgroup 34 35 analyses. Additionally, the current study showed the participants with higher spicy 36 37 38 food intake tended to have lower values of SBP and DBP, which is consistent with the 39 40 findings of He et al (38). In terms of obesity and hypertension both related with chronic 41 42 disease, the detail association between spicy food intake and hypertension would be 43 44 45 explored in future. To some extent, the current epidemiological study indicated that 46 47 the spicy food might affect abdominal obesity through increasing energy intake. 48 49 Further studies should be conducted to examine the detailed contribution of protein, 50 51 52 fat, and carbohydrate energy intake in the effect of spicy food intake on abdominal 53 54 obesity. 55 56 This is the first study to estimate the relationship between spicy food intake 57 58 frequency and abdominal obesity mediated by energy intake. However, there are 59 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 33 BMJ Open
1 2 3 several limitations need to be noticed. Firstly, as a cross-sectional study, it was unable 4 5 6 to reveal a causal association between spicy food intake frequency and abdominal 7 8 obesity. Studies with long term follow-up are needed to validate the associations. 9 10 Secondly, assessing dietary behaviors in population based studies through FFQ might 11 12 have reporting and recall bias, and the variety of chili types might affected the 13 14 15 associations. But a validity study with small participants who completed the 3 day 16 17 24-hour recalls demonstrated the current FFQ could be used as a representative tool to 18 For peer review only 19 conduct a dietary evaluation of a rural population. Thirdly, some residents such as 20 21 22 college students and migrant workers were not included in the current study. And 23 24 these groups were more likely to be young and healthy, which might lead to the study 25 26 overestimate the ratio of abdominal obesity in the rural population. Finally, the 27 28 29 participants of the current study were from one province in the middle area of China, 30 31 which may not be a representative sample of the total Chinese rural population. 32 33 However, the rural population of Henan province accounts for 9% of total rural 34 35 Chinese population. Therefore, the results of this study could give some insights on 36 37 38 further exploring the association between spicy food consumption and abdominal 39 40 obesity. 41 42 In conclusion, the present study indicated that the frequency of spicy food intake 43 44 45 was positively associated with abdominal obesity in Chinese rural residents, and fat 46 47 energy intake may be a potential mediator linking the increased prevalence of 48 49 abdominal obesity in rural Chinese population. Therefore, multicenter, prospective 50 51 52 and intervention studies are needed to further explore the possible causality 53 54 associations and elucidate potential mechanisms. 55 56 57 58 59 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 33
1 2 3 Acknowledgements 4 5 6 The authors would like to thank the participants, the coordinators, and administrators 7 8 for their supports during the study. To Dr. Tanko Abdulai, the authors would like to 9 10 express their gratitude for his critical reading of the manuscript. 11 12 13 Author contributions 14 15 C.J.W. conceived and designed the study. K.L.Y., Y.Q.L., Y.X., L.W., K.X.D., 16 17 18 Z.X.M. and W.J.L.For coordinated peer data collection.review K.L.Y., only Y.Q.L., X.T.L. and R.Q.T. 19 20 conducted the analyses. Y.Q.L. and K.L.Y. wrote the manuscript. All authors have 21 22 approved the final manuscript. 23 24 25 Funding 26 27 This research was supported by the National Key Research and Development 28 29 30 Program Precision Medicine Initiative of China (Grant NO: 2016YFC0900803), 31 32 National Natural Science Foundation of China (Grant NO: 81573243, 81602925, 33 34 U1304821), Science and Technology Foundation for Innovation Talent of Henan 35 36 Province (Grant NO: 164100510021, 154200510010), Henan Natural Science 37 38 39 Foundation (Grant NO: 182300410293), High-level Personnel Special Support 40 41 Project of Zhengzhou University (Grant NO: ZDGD13001). The funders had no role 42 43 in the study design, data collection and analysis, decision to publish, or preparation of 44 45 46 the manuscript. 47 48 Ethics approval 49 50 Ethics approval was obtained from the “Zhengzhou University Life Science Ethics 51 52 53 Committee”, and written informed consent was obtained for all participants. Ethic 54 55 approval code: [2015] MEC (S128). 56 57 58 59 60 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 33 BMJ Open
1 2 3 Disclosure 4 5 6 All authors have completed the ICMJE uniform disclosure form at 7 8 www.icmje.org/coi_disclosure.pdf and declare: no support from any organization for 9 10 the submitted work; no financial relationships with any organizations that might have 11 12 an interest in the submitted work in the previous three years; no other relationships or 13 14 15 activities that could appear to have influenced the submitted work. 16 17 Data sharing statement 18 For peer review only 19 The datasets used and analyzed during the current study are available from the 20 21 22 corresponding author on reasonable request. 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 60 18
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 33
1 2 3 References 4 5 6 1. World Health Organization . Obesity and overweight. Available at: 7 8 http://www.who.int/mediacentre/factsheets/fs311/en/. (Accessed December 2018) 9 10 2. Reynolds K, Gu D, Whelton PK, et al. InterASIA Collaborative Group. Prevalence 11 12 and risk factors of overweight and obesity in China. Obesity (Silver Spring). 2007; 13 14 15 15: 10-18. 16 17 3. Du P, Wang HJ, Zhang B, et al. Prevalence of abdominal obesity among Chinese 18 For peer review only 19 adults in 2011. J Epidemiol. 2017; 27:282-286. 20 21 22 4. Liu X, Wu W, Mao Z, et al. Prevalence and influencing factors of overweight and 23 24 obesity in a Chinese rural population: the Henan Rural Cohort Study. Sci Rep. 25 26 2018; 8(1):13101. 27 28 29 5. Li J, Wang R, Xiao C. Association between chilli food habits with iron status and 30 31 insulin resistance in a Chinese population. J Med Food. 2014; 17: 472-478. 32 33 6. Li Q, Cui YT, Jin RB, et al. Enjoyment of Spicy Flavor Enhances Central 34 35 Salty-Taste Perception and Reduces Salt Intake and Blood Pressure. 36 37 38 Hypertension. 2017; 70: 1291-1299. 39 40 7. Khayyatzadeh SS, Kazemi-Bajestani SMR, Mirmousavi SJ, et al. Dietary behaviors 41 42 in relation to prevalence of irritable bowel syndrome in adolescent girls. J 43 44 45 Gastroenterol Hepatol. 2018; 33:404-410. 46 47 8. Xue Y, He T, Yu K, et al. Association between spicy food consumption and lipid 48 49 profiles in adults: a nationwide population-based study. Br J Nutr. 2017; 118: 50 51 52 144-153. 53 54 9. Aggarwal BB, Van Kuiken ME, Iyer LH, et al. Molecular targets of nutraceuticals 55 56 derived from dietary spices: potential role in suppression of inflammation and 57 58 tumorigenesis. Exp Biol Med. 2009; 234: 825-849. 59 60 19
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 33 BMJ Open
1 2 3 10. Lv J, Qi L, Yu CQ, et al. Consumption of spicy foods and total and cause specific 4 5 6 mortality: population based cohort study. BMJ. 2015; 351:h3942. 7 8 11. Sun D, Lv J, Chen W, et al. Spicy food consumption is associated with adiposity 9 10 measures among half a million Chinese people: the China Kadoorie Biobank 11 12 study. BMC Public Health. 2014; 14: 1293. 13 14 15 12. Zhai FY, Du SF, Wang ZH, et al. Dynamics of the Chinese diet and the role of 16 17 urbanization, 1991–2011. Obes Rev. 2014; 15 Suppl 1:16-26. 18 For peer review only 19 20 13. Ludy MJ, Moore GE, Mattes RD. The effects of capsaicin and capsiate on energy 21 22 balance: critical review and meta-analyses of studies in humans. Chem Senses. 23 24 2012; 37: 103-121. 25 26 14. Choi SE, Chan J. Relationship of 6-n-propylthiouracil taste intensity and chili 27 28 29 pepper use with body mass index, energy intake, and fat intake within an 30 31 ethnically diverse population. J Acad Nutr Diet. 2015; 115: 389-396. 32 33 15. Liu X, Mao Z, Li Y, et al. The Henan Rural Cohort: a prospective study of 34 35 36 chronic non-communicable diseases. Int J Epidemiol. 2019; 26. 37 38 16. Yang K, Li Y, Mao Z, et al. Relationship between spicy flavor, spicy food intake 39 40 frequency and general obesity in a rural adult Chinese population: the RuralDiab 41 42 43 Study. Nutr Metab Cardiovasc Dis. 2018; 28:252-261. 44 45 17. Xue Y, Yang KL, Wang BY, et al. Reproducibility and Validity of a Food 46 47 48 Frequency Questionnaire in the Henan Rural Cohort Study. Public Health Nutr. 49 50 2019;(accept) 51 52 18. Craig CL, Marshall AL, Sjöström M, et al. International physical activity 53 54 55 questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35: 56 57 1381e95. 58 59 60 20
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 33
1 2 3 19. International Diabetes Federation. The IDF consensus worldwide definition of the 4 5 6 metabolic syndrome. Available at: 7 8 https://www.idf.org/our-activities/advocacy-awareness/resources-and-tools/60:id 9 10 fconsensus-worldwide-definitionof-the-metabolic-syndrome.html. (Accessed 11 12 December 2018) 13 14 15 20. Gonlachanvit S. Are rice and spicy diet good for functional gastrointestinal 16 17 disorders? J Neurogastroenterol Motil. 2010; 16:131-138. 18 For peer review only 19 21. Whiting S, Derbyshire E, Tiwari BK. Capsaicinoids and capsinoids. A potential 20 21 22 role for weight management? A systematic review of the evidence. Appetite. 23 24 2012; 59:341-348.. 25 26 22. Shi Z, Riley M, Taylor AW, et al. Chilli consumption and the incidence of 27 28 29 overweight and obesity in a Chinese adult population. Int J Obes (Lond). 2017; 30 31 41:1074-1079. 32 33 23. Deshpande J, Jeyakodi S, Juturu V. Tolerability of Capsaicinoids from Capsicum 34 35 Extract in a Beadlet Form: A Pilot Study. J Toxicol. 2016; 2016: 6584649. 36 37 38 24. Lean MEJ, Vlachou P, Govan L, et al. Different associations between body 39 40 composition and alcohol when assessed by exposure frequency or by quantitative 41 42 estimates of consumption. J Hum Nutr Diet. 2018; 31(6):747-757. 43 44 45 25. Gallagher D, VisserM, Sepulveda D, et al. How useful is body mass index for 46 47 comparison of body fatness across age, sex, and ethnic groups? Am J Epidemiol. 48 49 1996; 143: 228–239. 50 51 52 26. Misra A, Wasir JS, Vikram NK. Waist circumference criteria for the diagnosis of 53 54 abdominal obesity are not applicable uniformly to all populations and ethnic 55 56 groups. Nutrition. 2005; 21:969‐976. 57 58 59 60 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 33 BMJ Open
1 2 3 27. Snitker S, Fujishima Y, Shen H, et al. Effects of novel capsinoid treatment on 4 5 6 fatness and energy metabolism in humans: possible pharmacogenetic 7 8 implications. Am J Clin Nutr. 2009; 89: 45-50. 9 10 28. Haramizu S, Mizunoya W, Masuda Y, et al. Capsiate, a Nonpungent Capsaicin 11 12 Analog, Increases EnduranceSwimming Capacity of Mice by Stimulation of 13 14 15 Vanilloid Receptors. Biosci Biotechnol Biochem. 2006; 70(4):774-781. 16 17 29. Tremblay A, Arguin H, Panahi S. Capsaicinoids: a spicy solution to the 18 For peer review only 19 management of obesity? Int. J. Obes. 2016; 40(8): 1198-204. Int J Obes (Lond). 20 21 22 2016; 40(8):1198-1204. 23 24 30. Varghese S, Kubatka P, Rodrigo L, et al. Chili pepper as a body weight-loss food. 25 26 Int J Food Sci Nutr. 2017; 68(4):392-401. 27 28 29 31. Janssens PL, Hursel R, Westerterp-Plantenga MS. Capsaicin increases sensation 30 31 of fullness in energy balance, and decreases desire to eat after dinner in negative 32 33 energy balance. Appetite. 2014; 77:44-49. 34 35 32. Yoshioka M, St-Pierre S, Drapeau V, et al. Effects of red pepper on appetite and 36 37 38 energy intake. Br J Nutr. 1999; 82(2): 115-123. 39 40 33. Westerterp-Plantenga MS, Smeets A, Lejeune MP. Sensory and gastrointestinal 41 42 satiety effect of capsaicin on food intake. Int J Obes (Lond). 2005; 43 44 45 29(6):682-688. 46 47 34. Ahuja KD1, Ball MJ. Effects of daily ingestion of chilli on serum lipoprotein 48 49 oxidation in adult men and women. Br J Nutr. 2006; 96(2): 239-242 50 51 52 35. Yoneshiro T, Aita S, Kawai Y, et al. Nonpungent capsaicin analogs (capsinoids) 53 54 increase energy expenditure through the activation of brown adipose tissue in 55 56 humans. Am J Clin Nutr 2012; 95(4): 845e50 57 58 59 60 22
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 33
1 2 3 36. Kawada T, Hagihara K, Iwai K. Effects of capsaicin on lipid metabolism in rats 4 5 6 fed a high fat diet. J Nutr 1986, 116, 1272-1278. 7 8 37. Hochkogler CM, Lieder B, Rust P, et al. A 12-week intervention with 9 10 nonivamide, a TRPV1 agonist, prevents a dietary-induced body fatgain and 11 12 increases peripheral serotonin in moderately overweight subjects. Mol Nutr Food 13 14 15 Res. 2017; 61 (5). 16 17 38. He T, Wang M, Tian Z, et al. Sex-dependent difference in the association between 18 For peer review only 19 frequency of spicy food consumption and risk of hypertension in Chinese adults. 20 21 22 Eur J Nutr. 2018; 4. 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 60 23
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 33 BMJ Open
1 2 3 Figure Legends 4 5 6 Figure 1 The percentage of abdominal obesity by different categories. 7 8 Figure 2 ORs (solid lines) and 95% CIs (dashed lines) of spicy food for abdominal 9 10 obesity from restricted cubic splines. (A: crude model, B: adjusted for age, gender, 11 12 education, marital status, tobacco use, alcohol use and physical activity, C: adjusted 13 14 15 for B plus total energy intake, D: adjusted for B plus protein energy intake, E: 16 17 adjusted for B plus fat energy intake, F: adjusted for B plus carbohydrate energy 18 For peer review only 19 intake.) 20 21 22 Figure 3 The mediation analyses between spicy food intake frequency and abdominal 23 24 obesity by fat energy intake. 25 26 Supplement Table 1 The mediation analyses between spicy food intake frequency 27 28 29 and abdominal obesity by energy intake. 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 60 24
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 33
1 2 3 Table 1 Characteristics of the participants according to the spicy food intake frequency 4 5 Variable Never 1-2 d/wk 3-5 d/wk 6-7 d/wk 6 P 7 (N=12291) (N=3252) (N=3241) (N=9989) 8 Age (years, mean ± SD) 58.19±11.68 51.58±13.21 51.46±13.56 54.46±11.64 <0.001 9 Gender, n (%) <0.001 10 11 Male 4813(39.16) 1245(38.28) 1449(44.71) 4214(42.19) 12 Female 7478(60.84) 2007(61.72) 1792(55.29) 5775(57.81) 13 14 Education, n (%) <0.001 15 Illiterate and Elementary 5948(48.39) 1093(33.61) 1126(34.74) 4550(45.55) 16 17 Middle school 4435(36.08) 1345(41.36) 1345(41.5) 4078(40.82) 18 High school and aboveFor peer1908(15.52) review814(25.03) only770(23.76) 1361(13.62) 19 20 Tobacco use, n (%) <0.001 21 Nonsmoker 9293(75.61) 2425(74.57) 2169(66.92) 6819(68.27) 22 Ex- Smoker 1033(8.4) 213(6.55) 256(7.9) 739(7.4) 23 24 Smoker 1965(15.99) 614(18.88) 816(25.18) 2431(24.34)
25 Alcohol use, n (%) <0.001 26 27 Nondrinker 10216(83.12) 2510(77.18) 2335(72.05) 7251(72.59) 28 Ex-drinker 657(5.35) 122(3.75) 143(4.41) 479(4.80) 29 30 Drinker 1418(11.54) 620(19.07) 763(23.54) 2259(22.61) 31 Physical activity, n (%) 0.091 32 Low 4376(35.60) 1117(34.35) 1028(31.72) 2601(26.04) 33 34 Middle 4365(35.51) 1048(32.23) 1215(37.49) 3959(39.63) 35 High 3550(28.88) 1087(33.43) 998(30.79) 3429(34.33) 36 37 Marital status, n (%) <0.001 38 Married/cohabiting 10844(88.23) 2972(91.39) 2943(90.81) 9161(91.71) 39 40 Widowed/single/divorced/separation 1447(11.77) 280(8.61) 298(9.19) 828(8.29) 41 BMI(kg/m2, mean ± SD) 24.57±3.51 25.00±3.62 24.84±3.66 24.86±3.56 <0.001 42 43 Waist circumference(cm, mean ± SD) 44 Male 84.79±10.28 87.23±10.57 86.51±11.08 85.28±10.58 <0.001 45 Female 82.64±10.39 82.07±10.31 81.68±10.13 82.83±9.99 <0.001 46 47 SBP(mmHg, mean ± SD) 127.86±20.53 125.58±19.97 124.18±19.54 123.50±19.22 <0.001
48 DBP(mmHg, mean ± SD) 77.76±11.59 78.46±11.92 77.40±11.85 76.71±11.50 <0.001 49 50 Total energy (kcal, mean ± SD) 2371.08±660.96 2371.07±649.17 2446.38±657.54 2606.19±691.99 <0.001 51 Fat energy (kcal, mean ± SD) 649.83±166.88 664.74±167.84 689.14±172.15 704.09±182.27 <0.001 52 53 Carbohydrate energy (kcal, mean ± SD) 1423.31±477.60 1407.78±472.68 1445.63±479.46 1570.50±500.73 <0.001 54 Protein energy (kcal, mean ± SD) 297.93±97.99 298.55±95.55 311.61±97.83 331.60±103.27 <0.001 55 56 Abbreviations: BMI, body mass index; SBP, systemic blood pressure; DBP, diastolic blood pressure. 57 58 59 60 25
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 33 BMJ Open
1 2 3 4 5 Table 2 The ORs (95% CIs) of spicy food intake frequency for abdominal obesity 6 7 Never (N=12291) 1-2 d/wk (N=3252) 3-5 d/wk (N=3241) 6-7 d/wk (N=9989) Each day increment Ptrend 8 9 Model 1 1.000 1.113(1.031, 1.203) 0.992(0.918, 1.072) 1.038(0.985, 1.095) 1.009(0.992, 1.027) 0.308 10 11 Model 2 1.000 1.183(1.090, 1.283) 1.129(1.040, 1.226) 1.118(1.057, 1.183) 1.036(1.017, 1.055) <0.001 12 For peer review only 13 Model 3 1.000 1.185(1.092, 1.286) 1.132(1.043, 1.229) 1.122(1.059, 1.187) 1.059(1.039, 1.078) <0.001 14 15 Model 4 1.000 1.185(1.092, 1.286) 1.131(1.042, 1.228) 1.117(1.055, 1.182) 1.055(1.036, 1.074) <0.001 16 17 Model 5 1.000 1.186(1.093, 1.287) 1.127(1.038, 1.224) 1.104(1.044, 1.169) 1.051(1.032, 1.070) <0.001 18 19 Model 6 1.000 1.185(1.091, 1.286) 1.133(1.043, 1.23) 1.129(1.066, 1.195) 1.059(1.040, 1.079) <0.001 20 21 Model 1: crude model. 22 23 Model 2: adjusted for age, gender, education, marital status, tobacco use, alcohol use and physical activity. 24 25 Model 3: adjusted for model 2 plus total energy intake. 26 27 Model 4: adjusted for model 2 plus protein energy intake. 28 29 Model 5: adjusted for model 2 plus fat energy intake. 30 31 Model 6: adjusted for model 2 plus carbohydrate energy intake. 32 33 34 35 36 37 38 39 40 41 26 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 28 of 33
1 2 3 4 5 Table 3 The subgroup ORs (95% CIs) of spicy food intake frequency for abdominal obesity 6 Variable Never 1-2 d/wk 3-5 d/wk 6-7 d/wk Each day increment P trend 7 Age(year) 8 <30 1.000 0.717(0.470-1.095) 0.721(0.478-1.087) 0.981(0.671-1.433) 1.003(0.886-1.136) 0.959 9 31~40 1.000 1.547(1.205-1.986) 1.331(1.032-1.716) 1.394(1.112-1.747) 1.096(1.020-1.178) 0.013 10 41~50 1.000 1.477(1.235-1.765) 1.336(1.113-1.603) 1.221(1.070-1.394) 1.057(1.013-1.104) 0.011 11 51~60 1.000 1.042(0.888-1.223) 1.140(0.970-1.340) 1.141(1.025-1.269) 1.046(1.010-1.084) 0.012 12 61~70 For1.000 peer1.406(1.188-1.665) review1.179(0.998-1.394) only1.011(0.910-1.122) 1.006(0.972-1.041) 0.738 13 >70 1.000 0.945(0.718-1.243) 1.030(0.782-1.357) 0.947(0.798-1.124) 0.985(0.931-1.041) 0.590 14 Gender 15 Male 1.000 1.334(1.167-1.524) 1.192(1.049-1.355) 0.990(0.902-1.088) 0.994(0.964-1.025) 0.697 16 Female 1.000 1.109(1.000-1.230) 1.055(0.948-1.176) 1.196(1.112-1.287) 1.058(1.033-1.083) <0.001 17 Education 18 Illiterate and Elementary 1.000 1.181(1.028-1.356) 1.028(0.896-1.178) 1.042(0.957-1.135) 1.011(0.983-1.04) 0.442 19 Middle school 1.000 1.251(1.100-1.422) 1.146(1.008-1.303) 1.113(1.017-1.219) 1.032(1.002-1.063) 0.037 20 High school and above 1.000 1.173(0.986-1.396) 1.34(1.120-1.604) 1.168(1.008-1.353) 1.061(1.012-1.113) 0.015 21 Marital status 22 Married/cohabiting 1.000 1.188(1.090-1.294) 1.144(1.05-1.247) 1.108(1.044-1.176) 1.033(1.013-1.053) 0.001 23 Widowed/single / divorced/separation 1.000 1.122(0.842-1.495) 0.944(0.705-1.263) 1.093(0.900-1.327) 1.024(0.961-1.091) 0.467 24 Tobacco use 25 Nonsmoker 1.000 1.154(1.050-1.269) 1.110(1.006-1.225) 1.158(1.083-1.238) 1.048(1.025-1.071) <0.001 26 Ex- Smoker 1.000 1.074(0.787-1.465) 1.235(0.928-1.644) 1.097(0.895-1.345) 1.037(0.971-1.109) 0.281 27 Smoker 1.000 1.386(1.138-1.687) 1.094(0.912-1.312) 0.941(0.821-1.079) 0.968(0.926-1.012) 0.150 28 Alcohol use 29 Nondrinker 1.000 1.155(1.053-1.268) 1.045(0.950-1.150) 1.120(1.050-1.194) 1.035(1.013-1.057) 0.002 30 Ex-drinker 1.000 1.077(0.710-1.634) 1.211(0.824-1.780) 1.098(0.847-1.424) 1.036(0.952-1.128) 0.413 31 Drinker 1.000 1.363(1.118-1.662) 1.347(1.117-1.623) 1.069(0.926-1.233) 1.010(0.965-1.058) 0.668 32 Physical activity 33 Low 1.000 1.176(1.023-1.353) 1.222(1.056-1.413) 1.070(0.964-1.188) 1.028(0.993-1.063) 0.116 34 Middle 1.000 1.270(1.103-1.464) 1.098(0.961-1.255) 1.127(1.029-1.234) 1.035(1.005-1.067) 0.023 35 High 1.000 1.115(0.963-1.29) 1.059(0.909-1.234) 1.105(0.998-1.224) 1.031(0.997-1.066) 0.076 36 Adjusted model (except where it is the variable of interest): adjusted for age, gender, education, marital status, tobacco use, alcohol use, physical activity and fat energy intake. 37 38 39 40 41 27 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 29 of 33 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 1 The percentage of abdominal obesity by different categories. 40 90x90mm (300 x 300 DPI) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 2 ORs (solid lines) and 95% CIs (dashed lines) of spicy food for abdominal obesity from restricted cubic splines. (A: crude model, B: adjusted for age, gender, education, marital status, tobacco use, alcohol 40 use and physical activity, C: adjusted for B plus total energy intake, D: adjusted for B plus protein energy 41 intake, E: adjusted for B plus fat energy intake, F: adjusted for B plus carbohydrate energy intake.) 42 43 90x90mm (300 x 300 DPI) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 33 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 3 The mediation analyses between spicy food intake frequency and abdominal obesity by fat energy intake. 40 41 90x90mm (300 x 300 DPI) 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 33
1 2 3 Supplement Table 1 The mediation analyses between spicy food intake frequency and abdominal obesity by energy intake. 4 5 Energy intake Direct effect, OR(95%CI) Path a, β(95%CI) Path b, OR(95%CI) Indirect effect, OR(95%CI) 6 7 Total 1.047(1.028, 1.067) 64.316(58.799, 69.832) 1.001(0.999, 1.002) 0.999(0.996, 1.001) 8 Protein 1.117(1.063, 1.175) 18.501(16.190, 20.812) 0.999(0.996, 1.001) 0.997(0.993,1.002) 9 10 Fat 1.107(1.053, 1.164) 26.536(22.714, 30.358) 1.003(1.001, 1.004) 1.007(1.003, 1.012) 11 Carbohydrate 1.181(1.130, 1.234) 8.955(7.826, 10.834) 1.001(0.999, 1.003) 1.001(0.998, 1.004) 12 Adjusted model: adjusted for age, gender, education, marital status, tobacco use, alcohol use, and physical activity. 13 Path a stands for the effect of spicy food intake frequency on energy intake. 14 Path b stands for the effect of energy intake on abdominal obesity adjusted for spicy food intake frequency. 15 16 17 18 For peer review only 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 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 33 BMJ Open
1 2 STROBE 2007 (v4) Statement—Checklist of items that should be included in reports of cross-sectional studies 3
4 5 Item Section/Topic Recommendation Reported on page # 6 # 7 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 1, 2,3 8 9 (b) Provide in the abstract an informative and balanced summary of what was done and what was found 2,3 10 11 Introduction 12 Background/rationale 2 Explain the scientificFor background peer and rationale forreview the investigation being reported only 5, 6 13 Objectives 3 State specific objectives, including any prespecified hypotheses 6 14 15 Methods 16 Study design 4 Present key elements of study design early in the paper 6 17 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data 6, 7 19 collection 20 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of participants 6, 7 21
22 23 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if 7, 8, 9 24 25 applicable 26 Data sources/ 8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe 7, 8, 9 27 measurement comparability of assessment methods if there is more than one group 28 Bias 9 Describe any efforts to address potential sources of bias 7, 8, 9 29 30 Study size 10 Explain how the study size was arrived at 6, 7 31 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and 7, 8, 9 32 why 33 34 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 9, 10 35 (b) Describe any methods used to examine subgroups and interactions 9, 10 36
37 (c) Explain how missing data were addressed 38 (d) If applicable, describe analytical methods taking account of sampling strategy 39 (e) Describe any sensitivity analyses 40 41 Results 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 34 of 33
1 2 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, 10 3 confirmed eligible, included in the study, completing follow-up, and analysed 4 5 (b) Give reasons for non-participation at each stage 6 (c) Consider use of a flow diagram 7 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential 10 8 9 confounders 10 (b) Indicate number of participants with missing data for each variable of interest 11 Outcome data 15* Report numbers of outcome events or summary measures 10 12 Main results 16 (a) Give unadjustedFor estimates peer and, if applicable, reviewconfounder-adjusted estimates only and their precision (eg, 95% confidence 11, 12 13 14 interval). Make clear which confounders were adjusted for and why they were included 15 (b) Report category boundaries when continuous variables were categorized 16 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period 17 18 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses 12 19 Discussion 20 Key results 18 Summarise key results with reference to study objectives 12 21 22 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and 15, 16 23 magnitude of any potential bias 24 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from 13, 14, 15, 16 25 similar studies, and other relevant evidence 26 27 Generalisability 21 Discuss the generalisability (external validity) of the study results 13, 14, 15 28 Other information 29 30 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on 17 31 which the present article is based 32 33 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies. 34 35 36 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE 37 checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 38 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org. 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open
The association of the frequency of spicy food intake and the risk of abdominal obesity in rural Chinese adults: a cross-sectional study ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2018-028736.R2
Article Type: Original research
Date Submitted by the 12-Sep-2019 Author:
Complete List of Authors: Yang, Kaili; Henan Provincial People’s Hospital, Henan Eye Hospital, Henan Eye Institute, People's Hospital of Zhengzhou University, Zhengzhou, 450003, China; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Li, Yuqian; Zhengzhou University, Department of Clinical Pharmacology, School of Pharmaceutical Science Xue, Yuan; Zhengzhou University, Department of Nutrition and Food Hygiene, College of Public Health Wang, Ling; Zhengzhou University, Department of Nutrition and Food Hygiene, College of Public Health Liu, Xiaotian; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Tu, Runqi ; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Dong, Xiaokang; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Mao, Zhenxing ; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health Li, Wenjie; Zhengzhou University, Department of Nutrition and Food Hygiene, College of Public Health Wang, Chongjian; Zhengzhou University, Department of Epidemiology and Biostatistics, College of Public Health
Primary Subject Nutrition and metabolism Heading:
Secondary Subject Heading: Epidemiology, Nutrition and metabolism, Public health
Spicy food intake frequency, abdominal obesity, fat energy intake, Keywords: mediator analyses, rural population
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 34 BMJ Open
1 2 3 Title: 4 5 The association of the frequency of spicy food intake and the risk of abdominal 6 7 8 obesity in rural Chinese adults: a cross-sectional study 9 10 Running title: 11 12 Spicy food intake and abdominal obesity 13 14 Authors: 15 16 Kaili Yang1, 2†, Yuqian Li3†, Yuan Xue4, Ling Wang4, Xiaotian Liu2, Runqi Tu2, 17 18 Xiaokang DongFor2, Zhenxing peer Mao2, Wenjie review Li4, Chongjian only Wang2* 19 20 Affiliation: 21 22 1 Henan Provincial People’s Hospital, Henan Eye Hospital, Henan Eye Institute, 23 24 25 People's Hospital of Zhengzhou University, Zhengzhou, 450003, China. 26 2 27 Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou 28 29 University, Zhengzhou, Henan, PR. China. 30 31 3 Department of Clinical Pharmacology, School of Pharmaceutical Science, 32 33 Zhengzhou University, Zhengzhou, Henan, PR. China. 34 35 4 Department of Nutrition and Food Hygiene, College of Public Health, Zhengzhou 36 37 University, Zhengzhou, Henan, PR. China. 38 39 † Contributed equally to this work. 40 41 * 42 Correspondence author 43 44 Dr. Chongjian Wang 45 46 Department of Epidemiology and Biostatistics 47 48 College of Public Health, Zhengzhou University 49 50 100 Kexue Avenue, Zhengzhou, 450001, Henan, PR China 51 52 Phone: +86 371 67781452 53 54 Fax: +86 371 67781868 55 56 E-mail: [email protected] 57 58 59 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 34
1 2 3 Abstract 4 5 Objectives Recent data relating to the association between spicy food intake 6 7 8 frequency and abdominal obesity are limited, especially in low income areas. 9 10 Therefore, the study explored the relationship between spicy food intake frequency 11 12 and abdominal obesity, and assessed the role of energy intake as a mediator of these 13 14 15 associations in a rural Chinese adult population. 16 17 Design Cross-sectional study. 18 For peer review only 19 Setting Rural Chinese adult population. 20 21 22 Participants Subjects from Henan Rural Cohort Study (n=28 773). 23 24 Primary outcome measures The effects of spicy food intake frequency on 25 26 abdominal obesity were analyzed by restricted cubic spline and logistic regression, 27 28 and the mediation effect was analyzed using the bootstrap method. 29 30 31 Results: The adjusted percentages of abdominal obesity were 47.32%, 51.93%, 32 33 50.66%, and 50.29% in the spicy food intake sub-groups of never, 1–2 d/wk, 3–5 34 35 d/wk, and 6–7 d/wk, respectively. An inverse U-shaped association was found 36 37 38 between spicy food intake frequency and abdominal obesity (P < 0.01). Compared to 39 40 subjects who never consumed spicy food, the adjusted ORs (95% CIs) in the 1–2 41 42 d/wk, 3–5 d/wk and 6–7 d/wk subgroups were 1.186(1.093, 1.287), 1.127(1.038, 43 44 45 1.224) and 1.104(1.044, 1.169), respectively. Furthermore, the increased odds of 46 47 abdominal obesity related to more frequent spicy food intake was mediated by higher 48 49 fat energy intake; the direct and indirect effects were 1.107(1.053, 1.164) and 50 51 1.007(1.003, 1.012), respectively. 52 53 54 Conclusions The data indicated that spicy food intake increased the risk of abdominal 55 56 obesity, and fat energy intake may be a mediator of this association in rural Chinese 57 58 59 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 34 BMJ Open
1 2 3 populations. Clarifying the mechanisms will facilitate the development of novel 4 5 6 preventive and therapeutic approaches for abdominal obesity. 7 8 Keywords Spicy food intake frequency, abdominal obesity, fat energy intake, 9 10 mediator analyses, rural population 11 12 Clinical trial registration 13 14 15 The Henan Rural Cohort Study has been registered at Chinese Clinical Trial 16 17 Register (Registration number: ChiCTR-OOC-15006699). 18 For peer review only 19 http://www.chictr.org.cn/showproj.aspx?proj=11375 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 60 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 34
1 2 3 Strengths and limitations of this study 4 5 6 This is the first study to estimate the relationship between spicy food intake 7 8 frequency and abdominal obesity mediated by energy intake. 9 10 A large, population-based sample was investigated. 11 12 The study included available information on a broad range of covariates. 13 14 15 However, the survey did not clarify a relevant causal association between spicy 16 17 food intake frequency and abdominal obesity. 18 For peer review only 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 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 34 BMJ Open
1 2 3 Introduction 4 5 6 As a disorder of energy metabolism, obesity is defined as abnormal or excessive 7 8 fat accumulation that may impair health, and the excessive accumulation of visceral 9 10 fat is an independent risk factor of cardiovascular diseases (1). Although great efforts 11 12 have been made to control the weight of the population in China, the prevalence of 13 14 15 obesity continues to increase, especially in rural areas (2). According to the data 16 17 obtained in the China Health and Nutrition Survey (CHNS) in 2011, the prevalence of 18 For peer review only 19 abdominal obesity was 44.0% in rural China (3). A recent study also showed that the 20 21
22 age-standardized prevalence of abdominal obesity was 43.71% in rural Chinese adults 23 24 (4). Ongoing and reliable programs are needed to manage obesity and reduce the 25 26 associated complications. It is commonly acknowledged that many modifiable risk 27 28 29 factors contribute to obesity besides genetic factors. Physical activity and lifestyle 30 31 factors, particularly dietary behavior, are considered to be closely related to obesity, 32 33 especially in developing countries (1). 34 35 Spices, which improve the flavor and taste of food, have long been widely 36 37 38 consumed in Chinese cuisine (5). Several studies have indicated that the consumption 39 40 of spicy food affects obesity-related human health outcomes, such as hypertension, 41 42 irritable bowel syndrome, lipid disorders, cancers and even mortality (6-10). As the 43 44 45 most practical and simplest index, waist circumference is used to measure the 46 47 magnitude of fat accumulation in the abdomen. An observational study conducted in 48 49 five urban and five rural areas of China provided evidence that the frequency of spicy 50 51 (11) 52 food consumption was positively associated with waist circumference , which 53 54 indicated that spicy food intake plays a role in the regulation of abdominal obesity. 55 56 According to the sixth national population census, the rural districts account for 57 58 63.90% of the population of China, and the education level, economic development, 59 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 34
1 2 3 behavior factors, food choices and dietary patterns in rural areas differed from those 4 5 (12) 6 found in urban areas of China . However, the associations between spicy food 7 8 intake frequency and abdominal obesity in rural populations remain to be fully 9 10 elucidated. Furthermore, population studies have shown that spicy food consumption 11 12 inhibits obesity by regulating energy metabolism (13), and Choi et al. (14) found that the 13 14 15 energy intake in chili pepper non-users was significantly lower than that in chili 16 17 pepper users in the New York City area. However, the associations among spicy food 18 For peer review only 19 intake frequency, energy intake and abdominal obesity have not been reported. 20 21 22 Therefore, the study investigated the effects of spicy food intake frequency on 23 24 abdominal obesity in a rural Chinese population, and assessed the role of energy 25 26 intake as a mediator of these associations. 27 28 29 Methods 30 31 Study participants 32 33 The participants in the current study were from the Henan Rural Cohort Study 34 35 (15), which was registered on the Chinese Clinical Trial Register (Registration number: 36 37 38 ChiCTR-OOC-15006699). Briefly, the study was carried out between July 2015 and 39 40 September 2017 in the following locations in Henan Province in China: Xuchang 41 42 City, Yuzhou County; Zhumadian City, Suiping County; Kaifeng City, Tongxu 43 44 45 County; Xinxiang City, Xinxiang County; and Sanmenxia City, Yima County. A total 46 47 of 29 867 people aged between 18 and 79 years and with complete information for 48 49 spicy food intake and waist circumference measurements were recruited into the 50 51 52 current study. To better estimate the relationship between the frequency of spicy food 53 54 intake and abdominal obesity, 365 subjects with serious diseases (heart failure, kidney 55 56 failure and malignant tumor), 650 participants with hepatitis or tuberculosis or other 57 58 infectious diseases, 51 women who were pregnant or lactating, and 28 subjects under 59 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 34 BMJ Open
1 2 3 weight control management during the previous 6 months were excluded. Finally, 28 4 5 6 773 participants were included in the current analysis. This study was conducted 7 8 according to the Declaration of Helsinki guidelines, and all procedures were approved 9 10 by the Zhengzhou University Life Science Ethics Committee [Code: [2015] MEC 11 12 (S128)]. Informed consent was obtained from all participants. 13 14 15 Patient and public involvement 16 17 Neither patients nor the public were involved in developing this project. 18 For peer review only 19 Assessment of spicy food intake frequency 20 21 22 The Food Frequency Questionnaire (FFQ) regarding dietary intake was delivered 23 24 by well-trained staff through face-to-face interviews. Participants were also asked 25 26 “During the past month, how often did you have spicy foods in one week?” and a 27 28 29 frequency between 0 to 7 days per week was selected by the participants. Based on 30 31 the observation of the existing epidemiological studies (16), the participants were 32 33 divided into four groups according to spicy food intake frequency: Never (reference), 34 35 1–2 d/wk, 3–5 d/wk and 6–7 d/wk. The participants were asked to confirm that the 36 37 38 food types contained spices. The test-retest reliability and the internal consistency of 39 40 the questionnaire was accredited through a pilot study with 76 subjects. The internal 41 42 consistency Cronbach’s alpha coefficient (ICC) was 0.978, which indicated that the 43 44 45 current variable provides a reasonable estimation of spicy food intake frequency. 46 47 Assessment of covariates 48 49 Detailed information on socio-demographic and lifestyle factors was collected 50 51 52 using a questionnaire. The following variables were included in the current study: age, 53 54 gender, education, tobacco use (current smoker was defined as smoking at least 1 55 56 cigarette per day for 6 consecutive months), alcohol use (current drinker was defined 57 58 as alcohol drinking of at least 12 times per year), and marital status. Individual dietary 59 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 34
1 2 3 intake data were collected by asking each household member to report the category of 4 5 6 all food consumed, quantity, meal type, and dining place) using a dietary recall 7 8 method. The quantity of food consumed, including staple food, livestock, poultry, 9 10 fish, eggs, dairy, fruits, vegetables, beans, nuts, pickles, cereal and animal oil over the 11 12 previous year was recorded. For each individual, the mean total daily energy intake 13 14 15 and the proportions of protein, fat and carbohydrate were derived from dietary data 16 17 according to the Chinese Food Composition Table (2009). The 3 day 24-h record was 18 For peer review only 19 conducted in a small sample to validate the FFQ and the results demonstrated that the 20 21 22 questionnaire provided a representative tool to conduct a dietary evaluation of a rural 23 24 population (17). Physical activity was divided into low, middle and high according to 25 26 the international physical activity questionnaire (18). Blood pressure recorded on the 27 28 29 right arm supported at heart-level in the sitting position was measured using electronic 30 31 sphygmomanometers (Omron HEM-7071A, Japan). All recordings were obtained in 32 33 triplicate and the mean value was used for analysis. Body weight with light clothing 34 35 was measured to the nearest 0.1 kg using a weight measurement device (VBODY 36 37 38 HBF-371, OMRON, Japan). Height was measured to the nearest 0.1 cm without shoes 39 40 using a standard right-angle device and a fixed measurement tape. The body mass 41 42 index (BMI) was estimated as body weight (kg) divided by the height squared (m2). 43 44 45 Assessment of outcomes 46 47 Waist circumference was measured to the nearest 0.1 cm at a point midway 48 49 between the lowest rib and the iliac crest in a horizontal plane using a non-elastic 50 51 52 tape. All measurements were obtained in duplicate by trained research staff according 53 54 to a standard protocol from the Working Group on Obesity in China (WGOC) and the 55 56 mean values were used for statistical analyses. Abdominal obesity was defined as a 57 58 59 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 34 BMJ Open
1 2 3 waist circumference ≥90 cm for men and ≥80 cm for women according to the 4 5 6 guidelines of the International Diabetes Federation for Chinese populations (19). 7 8 Statistical analysis 9 10 Continuous variables were described as means ± standard deviation (SD) and 11 12 13 categorical variables were presented as proportions. Analysis of variance and 14 15 chi-squared tests were used to evaluate the differences in the general characteristics of 16 17 the four spicy food intake frequency subgroups. The crude, age- and sex- adjusted 18 For peer review only 19 20 percentages of abdominal obesity in the different groups were estimated. Restricted 21 22 cubic spline analysis was used to explore the dose-response relationship between 23 24 continuous spicy food intake frequency and abdominal obesity. Logistic regression 25 26 27 analysis was used to estimate the association of categorical spicy food intake 28 29 frequency and abdominal obesity based on the odds ratios (ORs) and 95% confidence 30 31 intervals (CIs). 32 33 To examine the extent of the association between spicy food intake frequency 34 35 36 and abdominal obesity mediated by energy intake, we estimated the magnitude of 37 38 change in the regression coefficient with and without adjustment for different types of 39 40 energy intake. The following models were evaluated in the analyses: Model 1, the 41 42 43 crude model; Model 2, adjusted for age, gender, education, marital status, tobacco 44 45 use, alcohol use and physical activity; Model 3, adjusted for model 2 plus total energy 46 47 intake; Model 4, adjusted for model 2 plus protein energy intake; Model 5, adjusted 48 49 50 for model 2 plus fat energy intake; and Model 6, adjusted for model 2 plus 51 52 carbohydrate energy intake. The mediation analyses were conducted using the 53 54 bootstrap method. A conceptual model to illustrate the proposed association between 55 56 spicy food intake frequency (predictor variable), energy intake (mediator) and 57 58 59 abdominal obesity (dependent variable) was conducted. Stratified subgroup analyses 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 34
1 2 3 were performed, according to the demographic characteristics and the types of spicy 4 5 6 food, to investigate changes in the effects of spicy food intake frequency on the risk of 7 8 abdominal obesity. The statistical analyses of the current data were performed using 9 10 SPSS 23.0 software package and P < 0.05 (two-tailed) was considered to indicate 11 12 statistical significance. 13 14 15 Results 16 17 Demographic characteristics 18 For peer review only 19 Table 1 summarizes the general characteristics of the participants. The mean age 20 21 22 of 28 773 subjects was 55.39 ± 12.36 years, and the mean values of waist 23 24 circumference were 85.44 ± 10.55 cm for males and 82.54 ± 10.22 cm for females. 25 26 There were significant differences in the characteristics of age, gender, education 27 28 29 level, tobacco and alcohol use, marital status, BMI, waist circumference, systemic 30 31 blood pressure (SBP), diastolic blood pressure (DBP), protein energy, fat energy and 32 33 carbohydrate energy between the spicy food intake frequency subgroups (P < 0.05). 34 35 Distribution of abdominal obesity in subgroups 36 37 38 The percentages of abdominal obesity by the categories of spicy food intake 39 40 frequency are shown in Figure 1. Among the participants, 49.15% had abdominal 41 42 obesity, of which 5997(48.79%), 1674(51.48%), 1575(48.60%), and 4968(49.73%) 43 44 45 were categorized in the never, 1–2 d/wk, 3–5 d/wk, and 6–7 d/wk, sub-groups of spicy 46 47 food intake frequency, respectively. The corresponding age- and sex- adjusted 48 49 percentages of abdominal obesity were 47.32%, 51.93%, 50.66%, and 50.29%, 50 51 52 respectively. 53 54 Association between spicy food intake frequency and abdominal obesity 55 56 The ORs of the spicy food intake frequency subgroups for abdominal obesity 57 58 first increased and then declined after reaching the peak. All non-linear association 59 60 10
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 34 BMJ Open
1 2 3 tests revealed this inverted U-shaped dose-response relationship (P < 0.01, Figure 2). 4 5 6 Table 2 summarizes the ORs of spicy food intake frequency for abdominal 7 8 obesity. Taking the participants who never consumed spicy food as the reference 9 10 group, the crude ORs (95% CIs) for the 1–2 d/wk, 3–5 d/wk and 6–7 d/wk subgroups 11 12 were 1.113(1.031, 1.203), 0.992(0.918, 1.072) and 1.038(0.985, 1.095), respectively. 13 14 15 For each one-day increment in the frequency of spicy food intake, the OR (95% CI) 16 17 was 1.009(0.992, 1.027) (Ptrend=0.308). After adjustment for potential confounders, 18 For peer review only 19 the ORs (95% CIs) of 1–2 d/wk, 3–5 d/wk and 6–7 d/wk subgroups were 1.183(1.090, 20 21 22 1.283), 1.129(1.040, 1.226) and 1.118(1.057, 1.183), respectively. The adjusted OR 23 24 (95% CI) was 1.036(1.017, 1.055) for each one-day increment in the frequency of 25 26 spicy food intake. In further analyses of changes in the ORs (95% CIs) with and 27 28 29 without adjustment for energy intake, spicy food intake frequency remained positively 30 31 associated with abdominal obesity in Models 3–6. After adjusting for fat energy 32 33 intake, the ORs (95% CIs) were 1.186(1.093, 1.287), 1.127(1.038, 1.224) and 34 35 1.104(1.044, 1.169), respectively. The adjusted OR (95% CI) was 1.051(1.032, 1.070) 36 37 38 for each one-day increment in the frequency of spicy food intake. 39 40 Mediating role of energy intake 41 42 The role of energy intake in mediating the association between spicy food intake 43 44 45 frequency and the risk of abdominal obesity was analyzed, and the related results are 46 47 presented in Supplementary Table 1. The ORs (95% CIs) for direct and indirect 48 49 effects mediated by fat energy intake on the associations between spicy food intake 50 51 52 frequency and abdominal obesity were 1.107(1.053, 1.164) and1.007(1.003, 1.012), 53 54 respectively. The adjusted OR (95% CI) of spicy food intake frequency for fat energy 55 56 intake was 26.536(22.714, 30.358), and the adjusted OR (95% CI) of fat energy intake 57 58 for abdominal obesity was 1.007(1.003, 1.012). Our analyses indicated that fat energy 59 60 11
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 34
1 2 3 intake partially mediated the relationship between spicy food intake frequency and 4 5 6 abdominal obesity (Figure 3). There were no significant effects mediated by protein 7 8 energy intake, carbohydrate energy intake and total energy intake. 9 10 Subgroup analyses between spicy food intake frequency and abdominal obesity 11 12 The ORs (95% CIs) of spicy food intake frequency subgroups for abdominal 13 14 15 obesity are presented in Table 3. There were no statistically significant trend 16 17 associations in subgroup participants aged ≤ 30 years, > 61 years, male, illiterate 18 For peer review only 19 20 and Elementary level educated, widowed/single/divorced/separated, ex-smoker, 21 22 smoker, ex-drinker, drinker, and low physical activity (P trend > 0.05). 23 24 Discussion 25 26 27 The present epidemiology survey provides new evidence for the current burden 28 29 of abdominal obesity in rural Chinese populations. Overall, the prevalence of 30 31 abdominal obesity was higher in a rural population of Chinese adults than that in 32 33 previous national studies in China (3). In total, 57.28% of participants consumed spicy 34 35 (20) 36 food, demonstrating the popularity of spicy food in the Chinese diet .Compared 37 38 with participants who never had spicy food, the participants in the spicy food intake 39 40 frequency subgroups (1–2 d/wk, 3–5 d/wk and 6–7 d/wk) were positively associated 41 42 43 with an increased risk of abdominal obesity. Moreover, an inverted U-shaped 44 45 relationship between spicy food intake frequency and abdominal obesity was 46 47 observed in the current study. The study also showed that fat energy intake partially 48 49 50 mediated the association between spicy food intake frequency and abdominal obesity. 51 52 A previous intervention study showed a beneficial effect of spicy food 53 54 consumption on weight management in a small sample size from Western countries 55 56 (21), although data regarding the effect of spicy food on obesity in Asian populations 57 58 59 are scarce. A prospective study of CHNS data showed that the cumulative average 60 12
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 34 BMJ Open
1 2 3 chili intake was inversely associated with the risk of overweight/obesity, which was 4 5 (22) 6 independent of overall dietary pattern, energy intake and lifestyle factors . In 7 8 contrast, clinical trials have also shown no significant differences in the indicators of 9 10 obesity between the placebo and capsaicin groups (23). Additionally, a previous study 11 12 combining cross-sectional and meta-analysis showed that spicy food intake frequency 13 14 15 was positively associated with general obesity in rural Chinese populations (16). Both 16 17 the BMI and waist circumference are considered as practical and effective indexes in 18 For peer review only 19 evaluating obesity (24). BMI is closely related to body fat (25), and reflects the degree of 20 21 22 obesity without the influence of differences in height. However, BMI does not 23 24 accurately reflect the distribution of fat mass in the body, whereas waist 25 26 circumference is regarded as the most practical and simplest indicator for evaluating 27 28 (26) 29 abdominal fat accumulation . In this study, we found that the mean waist 30 31 circumference was in the desirable range for males, but elevated for females. This 32 33 phenomenon might be explained by asymmetrical age structure and hormone 34 35 differences (15, 27, 28). Peri- and post-menopausal women are known to experience 36 37 38 relatively unfavorable metabolic changes compared with those in men of similar age, 39 40 making women more likely to show a tendency for abdominal weight gain during the 41 42 menopausal transition (27, 28). Studies exploring the associations between spicy food 43 44 (29) 45 intake frequency and abdominal obesity are still limited. Snitker et al. and 46 47 Haramizu et al. (30) conducted investigations using capsinoids, which are non-pungent 48 49 capsaicin analogs (13). Snitker et al. (29) reported that abdominal adiposity decreased to 50 51 52 a greater extent in the capsinoid group than in the placebo group, although the mean 53 54 change in waist girth was not significant. Fat oxidation increased in the capsinoid 55 56 group, a pattern analogous to that observed in the animal study conducted by 57 58 Haramizu et al. (30). Furthermore, it should be noted that the effects of capsinoids 59 60 13
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 34
1 2 3 delivered to humans in capsules and mice being fed capsiate solution prior to an 4 5 6 endurance test cannot necessarily be extrapolated to those in free-living humans 7 8 consuming capsaicin-containing spicy food as part of their regular diet. Thus, further 9 10 studies are required to fully elucidate the effects of spicy food components on 11 12 abdominal adiposity. The China Kadoorie Biobank (CKB) study showed that waist 13 14 15 circumference increased with the frequency of spicy food intake (11), which is 16 17 consistent with the current results. In addition, the non-linear association tests 18 For peer review only 19 demonstrated an inverted U-shaped dose-response relationship between spicy food 20 21 22 intake frequency and abdominal obesity in all participants. It can be speculated that 23 24 this effect is related to the observation that participants who consume spicy foods 25 26 more regularly tend to become less sensitive to the oral tactile effects and rate spicy 27 28 (31, 32) 29 stimuli as having lower burn intensity . However, these preliminary findings 30 31 require further verification in other populations and multicenter study. 32 33 Accumulating evidence from animal studies and clinical trials suggested that 34 35 spicy food consumption may reduce energy intake and enhanced fat oxidation (33, 34). 36 37 38 However, the prospective study using CHNS data showed that high amounts spicy 39 40 food intake was positively associated with energy intake. The mean energy intake of 41 42 participants who consumed more than 50 gram/day of chili was greater (> 43 44 (22) 45 200kcal/day) compared with that of the non-spicy food intake group . In addition, 46 47 capsaicin is a basic component of spicy food and responsible for approximately 70% 48 49 of the burn. Janssens et al (35) found that satiety and fullness increased when capsaicin 50 51 52 was added into the human diet. Moreover, several human studies have indicated that 53 54 capsaicin plays an important role in energy balance by changing the intake of fat, 55 56 carbohydrate, and protein (36, 37). Although metabolic studies have indicated a lipolytic 57 58 effect of spices, these studies have shown that fat energy intake mediates the 59 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 34 BMJ Open
1 2 3 association between spicy food intake frequency and abdominal obesity. It can be 4 5 6 speculated that the mechanism underlying the cholesterol-lowering activity involves 7 8 the capacity of spicy food to stimulate fecal excretion of bile acids and increased 9 10 resistance of serum lipoproteins to oxidation (38). However, the effect of spicy food on 11 12 weight management in terms of the mechanisms underlying energy intake balance and 13 14 15 changes in appetite and lipid metabolism remain to be clarified (39). In addition, 16 17 Kawada et al (40) found that capsaicin supplementation lowered serum triacylglycerol 18 For peer review only 19 concentration and stimulated lipid mobilization from adipose tissue but did not affect 20 21 22 lipogenesis in obese rats. Nonivamide, which is a less pungent structural analog of 23 24 capsaicin, was found to prevent weight gain without affecting blood total cholesterol, 25 26 low-density lipoprotein cholesterol, or triglycerides in a 12-week intervention trial (41). 27 28 29 Considering the defects of this cross-sectional study, further investigations are 30 31 warranted to explore the pathophysiological mechanisms underlying the effect of 32 33 spicy food consumption on abdominal obesity and lipid metabolism. In our subgroup 34 35 analyses, there was no association between spicy food intake frequency and 36 37 38 abdominal obesity in participants aged over 61years, possibly as a result of the severe 39 40 effects of the Chinese famine occurred in 1959–1961 on the prevalence of obesity in 41 42 adults in Henan Province. We previously reported that marital status, tobacco use, 43 44 45 alcohol use, and physical activity were associated with abdominal obesity, particularly 46 47 in women, which could explain the associations of the subgroups to some extent (4). 48 49 The variety of chili types and the lifestyles of both sexes partly account for the 50 51 52 relationships between each additional day of spicy food intake and abdominal obesity 53 54 in the subgroup analyses. Additionally, the current study showed that the participants 55 56 with higher spicy food intake tended to have lower SBP and DBP, which is consistent 57 58 with the findings of He et al (42). In terms of the associations of both obesity and 59 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 34
1 2 3 hypertension with chronic disease, a possible association between spicy food intake 4 5 6 and hypertension warrant further investigations. To some extent, the current 7 8 epidemiological study indicated that the spicy food might affect abdominal obesity by 9 10 increasing energy intake. These effects might vary depending on the type of fats 11 12 consumed (e.g., saturated, trans, monounsaturated, polyunsaturated); thus, further 13 14 15 studies focusing on the different types of fats consumed should be conducted to 16 17 examine the contribution of fat energy intake to the effects of spicy food intake on 18 For peer review only 19 abdominal obesity. 20 21 22 This is the first study to estimate the relationship between spicy food intake 23 24 frequency and abdominal obesity mediated by energy intake. However, several 25 26 limitations should be noticed. First, as a cross-sectional study, we were unable to 27 28 29 reveal a causal association between spicy food intake frequency and abdominal 30 31 obesity and studies with long term follow-up are needed to validate the associations. 32 33 Second, assessing dietary behaviors in population-based studies through FFQ might 34 35 have reporting and recall bias, and the variety of chili types might affect the 36 37 38 associations. However, a validation study with a small number of participants who 39 40 completed the 3 day 24-h recall demonstrated that the current FFQ is a representative 41 42 tool to conduct a dietary evaluation of a rural population. Third, some residents, such 43 44 45 as college students and migrant workers, were not included in the current study. These 46 47 groups are more likely to be young and healthy, which might lead to an overestimate 48 49 of the proportion of abdominal obesity in the rural population. Finally, the participants 50 51 52 in the current study were from one province in the central region of China, which may 53 54 not be representative of the total Chinese rural population. However, the rural 55 56 population of Henan Province accounts for 9% of total rural Chinese population. 57 58 Therefore, the results of this study provide some insights into the association between 59 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 34 BMJ Open
1 2 3 spicy food consumption and abdominal obesity. 4 5 6 In conclusion, the present study indicates that the frequency of spicy food intake 7 8 is positively associated with abdominal obesity in the rural Chinese population, and 9 10 implicate fat energy intake as a potential mediator linking the increased prevalence of 11 12 abdominal obesity in this population. Therefore, multicenter, prospective and 13 14 15 intervention studies are needed to further explore the possible causal associations and 16 17 elucidate potential mechanisms. 18 For peer review only 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 60 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 34
1 2 3 Acknowledgements 4 5 6 The authors would like to thank the participants, coordinators, and administrators for 7 8 their support during this study. The authors would like to express their gratitude to Dr. 9 10 Tanko Abdulai for his critical reading of the manuscript. 11 12 13 Author contributions 14 15 C.J.W. conceived and designed the study. K.L.Y., Y.Q.L., Y.X., L.W., K.X.D., 16 17 18 Z.X.M. and W.J.L.For coordinated peer data collection.review K.L.Y., only Y.Q.L., X.T.L. and R.Q.T. 19 20 conducted the analyses. Y.Q.L. and K.L.Y. wrote the manuscript. All authors have 21 22 approved the final manuscript. 23 24 25 Funding 26 27 This research was supported by the National Key Research and Development 28 29 30 Program Precision Medicine Initiative of China (Grant NO: 2016YFC0900803), 31 32 National Natural Science Foundation of China (Grant NO: 81573243, 81602925, 33 34 U1304821), Science and Technology Foundation for Innovation Talent of Henan 35 36 Province (Grant NO: 164100510021, 154200510010), Henan Natural Science 37 38 39 Foundation (Grant NO: 182300410293), High-level Personnel Special Support 40 41 Project of Zhengzhou University (Grant NO: ZDGD13001). The funders had no role 42 43 in the study design, data collection and analysis, decision to publish, or preparation of 44 45 46 the manuscript. 47 48 Ethics approval 49 50 Ethics approval was obtained from the “Zhengzhou University Life Science Ethics 51 52 53 Committee”, and written informed consent was obtained from all participants. Ethics 54 55 approval code: [2015] MEC (S128). 56 57 58 59 60 18
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 34 BMJ Open
1 2 3 Disclosure 4 5 6 All authors have completed the ICMJE uniform disclosure form at 7 8 www.icmje.org/coi_disclosure.pdf and declare: no support from any organization for 9 10 the submitted work; no financial relationships with any organizations that might have 11 12 an interest in the submitted work in the previous three years; no other relationships or 13 14 15 activities that could appear to have influenced the submitted work. 16 17 Data sharing statement 18 For peer review only 19 The datasets used and analyzed during the current study are available from the 20 21 22 corresponding author on reasonable request. 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 60 19
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 34
1 2 3 References 4 5 6 1. World Health Organization . Obesity and overweight. Available at: 7 8 http://www.who.int/mediacentre/factsheets/fs311/en/. (Accessed December 2018) 9 10 2. Reynolds K, Gu D, Whelton PK, et al. InterASIA Collaborative Group. Prevalence 11 12 and risk factors of overweight and obesity in China. Obesity (Silver Spring). 2007; 13 14 15 15: 10-18. 16 17 3. Du P, Wang HJ, Zhang B, et al. Prevalence of abdominal obesity among Chinese 18 For peer review only 19 adults in 2011. J Epidemiol. 2017; 27:282-286. 20 21 22 4. Liu X, Wu W, Mao Z, et al. Prevalence and influencing factors of overweight and 23 24 obesity in a Chinese rural population: the Henan Rural Cohort Study. Sci Rep. 25 26 2018; 8(1):13101. 27 28 29 5. Li J, Wang R, Xiao C. Association between chilli food habits with iron status and 30 31 insulin resistance in a Chinese population. J Med Food. 2014; 17: 472-478. 32 33 6. Li Q, Cui YT, Jin RB, et al. Enjoyment of Spicy Flavor Enhances Central 34 35 Salty-Taste Perception and Reduces Salt Intake and Blood Pressure. 36 37 38 Hypertension. 2017; 70: 1291-1299. 39 40 7. Khayyatzadeh SS, Kazemi-Bajestani SMR, Mirmousavi SJ, et al. Dietary behaviors 41 42 in relation to prevalence of irritable bowel syndrome in adolescent girls. J 43 44 45 Gastroenterol Hepatol. 2018; 33:404-410. 46 47 8. Xue Y, He T, Yu K, et al. Association between spicy food consumption and lipid 48 49 profiles in adults: a nationwide population-based study. Br J Nutr. 2017; 118: 50 51 52 144-153. 53 54 9. Aggarwal BB, Van Kuiken ME, Iyer LH, et al. Molecular targets of nutraceuticals 55 56 derived from dietary spices: potential role in suppression of inflammation and 57 58 tumorigenesis. Exp Biol Med. 2009; 234: 825-849. 59 60 20
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 34 BMJ Open
1 2 3 10. Lv J, Qi L, Yu CQ, et al. Consumption of spicy foods and total and cause specific 4 5 6 mortality: population based cohort study. BMJ. 2015; 351:h3942. 7 8 11. Sun D, Lv J, Chen W, et al. Spicy food consumption is associated with adiposity 9 10 measures among half a million Chinese people: the China Kadoorie Biobank 11 12 study. BMC Public Health. 2014; 14: 1293. 13 14 15 12. Zhai FY, Du SF, Wang ZH, et al. Dynamics of the Chinese diet and the role of 16 17 urbanization, 1991–2011. Obes Rev. 2014; 15 Suppl 1:16-26. 18 For peer review only 19 20 13. Ludy MJ, Moore GE, Mattes RD. The effects of capsaicin and capsiate on energy 21 22 balance: critical review and meta-analyses of studies in humans. Chem Senses. 23 24 2012; 37: 103-121. 25 26 14. Choi SE, Chan J. Relationship of 6-n-propylthiouracil taste intensity and chili 27 28 29 pepper use with body mass index, energy intake, and fat intake within an 30 31 ethnically diverse population. J Acad Nutr Diet. 2015; 115: 389-396. 32 33 15. Liu X, Mao Z, Li Y, et al. The Henan Rural Cohort: a prospective study of 34 35 36 chronic non-communicable diseases. Int J Epidemiol. 2019; 26. 37 38 16. Yang K, Li Y, Mao Z, et al. Relationship between spicy flavor, spicy food intake 39 40 frequency and general obesity in a rural adult Chinese population: the RuralDiab 41 42 43 Study. Nutr Metab Cardiovasc Dis. 2018; 28:252-261. 44 45 17. Xue Y, Yang KL, Wang BY, et al. Reproducibility and Validity of a Food 46 47 48 Frequency Questionnaire in the Henan Rural Cohort Study. Public Health Nutr. 49 50 2019;(accept) 51 52 18. Craig CL, Marshall AL, Sjöström M, et al. International physical activity 53 54 55 questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35: 56 57 1381e95. 58 59 60 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 34
1 2 3 19. International Diabetes Federation. The IDF consensus worldwide definition of the 4 5 6 metabolic syndrome. Available at: 7 8 https://www.idf.org/our-activities/advocacy-awareness/resources-and-tools/60:id 9 10 fconsensus-worldwide-definitionof-the-metabolic-syndrome.html. (Accessed 11 12 December 2018) 13 14 15 20. Gonlachanvit S. Are rice and spicy diet good for functional gastrointestinal 16 17 disorders? J Neurogastroenterol Motil. 2010; 16:131-138. 18 For peer review only 19 21. Whiting S, Derbyshire E, Tiwari BK. Capsaicinoids and capsinoids. A potential 20 21 22 role for weight management? A systematic review of the evidence. Appetite. 23 24 2012; 59:341-348.. 25 26 22. Shi Z, Riley M, Taylor AW, et al. Chilli consumption and the incidence of 27 28 29 overweight and obesity in a Chinese adult population. Int J Obes (Lond). 2017; 30 31 41:1074-1079. 32 33 23. Deshpande J, Jeyakodi S, Juturu V. Tolerability of Capsaicinoids from Capsicum 34 35 Extract in a Beadlet Form: A Pilot Study. J Toxicol. 2016; 2016: 6584649. 36 37 38 24. Lean MEJ, Vlachou P, Govan L, et al. Different associations between body 39 40 composition and alcohol when assessed by exposure frequency or by quantitative 41 42 estimates of consumption. J Hum Nutr Diet. 2018; 31(6):747-757. 43 44 45 25. Gallagher D, VisserM, Sepulveda D, et al. How useful is body mass index for 46 47 comparison of body fatness across age, sex, and ethnic groups? Am J Epidemiol. 48 49 1996; 143: 228–239. 50 51 52 26. Misra A, Wasir JS, Vikram NK. Waist circumference criteria for the diagnosis of 53 54 abdominal obesity are not applicable uniformly to all populations and ethnic 55 56 groups. Nutrition. 2005; 21:969‐976. 57 58 27. Lovejoy JC. The menopause and obesity. Prim Care. 2003; 30(2):317-325. 59 60 22
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 34 BMJ Open
1 2 3 28. van der Leeuw J, Wassink AM, van der Graaf Y, et al. Age-related differences in 4 5 6 abdominal fat distribution in premenopausal and postmenopausal women with 7 8 cardiovascular disease. Menopause. 2013;20(4):409-417. 9 10 29. Snitker S, Fujishima Y, Shen H, et al. Effects of novel capsinoid treatment on 11 12 fatness and energy metabolism in humans: possible pharmacogenetic 13 14 15 implications. Am J Clin Nutr. 2009; 89: 45-50. 16 17 30. Haramizu S, Mizunoya W, Masuda Y, et al. Capsiate, a Nonpungent Capsaicin 18 For peer review only 19 Analog, Increases Endurance Swimming Capacity of Mice by Stimulation of 20 21 22 Vanilloid Receptors. Biosci Biotechnol Biochem. 2006; 70(4):774-781. 23 24 31. Byrnes NK, Hayes JE. Behavioral measures of risk tasking, sensation seeking and 25 26 sensitivity to reward may reflect different motivations for spicy food liking and 27 28 29 consumption. Appetite. 2016;103: 411-422. 30 31 32. Ludy MJ, Mattes RD. Comparison of sensory, physiological, personality, and 32 33 cultural attributes in regular spicy food users and non-users. Appetite. 2012; 34 35 58(1):19-27. 36 37 38 33. Tremblay A, Arguin H, Panahi S. Capsaicinoids: a spicy solution to the 39 40 management of obesity? Int. J. Obes. 2016; 40(8): 1198-204. Int J Obes (Lond). 41 42 2016; 40(8):1198-1204. 43 44 45 34. Varghese S, Kubatka P, Rodrigo L, et al. Chili pepper as a body weight-loss food. 46 47 Int J Food Sci Nutr. 2017; 68(4):392-401. 48 49 35. Janssens PL, Hursel R, Westerterp-Plantenga MS. Capsaicin increases sensation 50 51 52 of fullness in energy balance, and decreases desire to eat after dinner in negative 53 54 energy balance. Appetite. 2014; 77:44-49. 55 56 36. Yoshioka M, St-Pierre S, Drapeau V, et al. Effects of red pepper on appetite and 57 58 energy intake. Br J Nutr. 1999; 82(2): 115-123. 59 60 23
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 34
1 2 3 37. Westerterp-Plantenga MS, Smeets A, Lejeune MP. Sensory and gastrointestinal 4 5 6 satiety effect of capsaicin on food intake. Int J Obes (Lond). 2005; 7 8 29(6):682-688. 9 10 38. Ahuja KD, Ball MJ. Effects of daily ingestion of chilli on serum lipoprotein 11 12 oxidation in adult men and women. Br J Nutr. 2006; 96(2): 239-242. 13 14 15 39. Yoneshiro T, Aita S, Kawai Y, et al. Nonpungent capsaicin analogs (capsinoids) 16 17 increase energy expenditure through the activation of brown adipose tissue in 18 For peer review only 19 humans. Am J Clin Nutr 2012; 95(4): 845e50. 20 21 22 40. Kawada T, Hagihara K, Iwai K. Effects of capsaicin on lipid metabolism in rats 23 24 fed a high fat diet. J Nutr 1986, 116, 1272-1278. 25 26 41. Hochkogler CM, Lieder B, Rust P, et al. A 12-week intervention with 27 28 29 nonivamide, a TRPV1 agonist, prevents a dietary-induced body fatgain and 30 31 increases peripheral serotonin in moderately overweight subjects. Mol Nutr Food 32 33 Res. 2017; 61 (5). 34 35 42. He T, Wang M, Tian Z, et al. Sex-dependent di erence in the association between 36 ff 37 38 frequency of spicy food consumption and risk of hypertension in Chinese adults. 39 40 Eur J Nutr. 2018; 4. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 24
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 34 BMJ Open
1 2 3 Figure Legends 4 5 6 Figure 1 The percentage of abdominal obesity by different categories. 7 8 Figure 2 ORs (solid lines) and 95% CIs (dashed lines) of spicy food for abdominal 9 10 obesity from restricted cubic splines. (A: crude model, B: adjusted for age, gender, 11 12 education, marital status, tobacco use, alcohol use and physical activity, C: adjusted 13 14 15 for B plus total energy intake, D: adjusted for B plus protein energy intake, E: 16 17 adjusted for B plus fat energy intake, F: adjusted for B plus carbohydrate energy 18 For peer review only 19 intake.) 20 21 22 Figure 3 The mediation analyses between spicy food intake frequency and abdominal 23 24 obesity by fat energy intake. 25 26 Supplement Table 1 The mediation analyses between spicy food intake frequency 27 28 29 and abdominal obesity by energy intake. 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 60 25
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 34
1 2 3 Table 1 Characteristics of the participants according to the spicy food intake frequency 4 5 Variable Never 1-2 d/wk 3-5 d/wk 6-7 d/wk 6 P 7 (N=12291) (N=3252) (N=3241) (N=9989) 8 Age (years, mean ± SD) 58.19±11.68 51.58±13.21 51.46±13.56 54.46±11.64 <0.001 9 Gender, n (%) <0.001 10 11 Male 4813(39.16) 1245(38.28) 1449(44.71) 4214(42.19) 12 Female 7478(60.84) 2007(61.72) 1792(55.29) 5775(57.81) 13 14 Education, n (%) <0.001 15 Illiterate and Elementary 5948(48.39) 1093(33.61) 1126(34.74) 4550(45.55) 16 17 Middle school 4435(36.08) 1345(41.36) 1345(41.5) 4078(40.82) 18 High school and aboveFor peer1908(15.52) review814(25.03) only770(23.76) 1361(13.62) 19 20 Tobacco use, n (%) <0.001 21 Nonsmoker 9293(75.61) 2425(74.57) 2169(66.92) 6819(68.27) 22 Ex- Smoker 1033(8.4) 213(6.55) 256(7.9) 739(7.4) 23 24 Smoker 1965(15.99) 614(18.88) 816(25.18) 2431(24.34)
25 Alcohol use, n (%) <0.001 26 27 Nondrinker 10216(83.12) 2510(77.18) 2335(72.05) 7251(72.59) 28 Ex-drinker 657(5.35) 122(3.75) 143(4.41) 479(4.80) 29 30 Drinker 1418(11.54) 620(19.07) 763(23.54) 2259(22.61) 31 Physical activity, n (%) 0.091 32 Low 4376(35.60) 1117(34.35) 1028(31.72) 2601(26.04) 33 34 Middle 4365(35.51) 1048(32.23) 1215(37.49) 3959(39.63) 35 High 3550(28.88) 1087(33.43) 998(30.79) 3429(34.33) 36 37 Marital status, n (%) <0.001 38 Married/cohabiting 10844(88.23) 2972(91.39) 2943(90.81) 9161(91.71) 39 40 Widowed/single/divorced/separation 1447(11.77) 280(8.61) 298(9.19) 828(8.29) 41 BMI(kg/m2, mean ± SD) 24.57±3.51 25.00±3.62 24.84±3.66 24.86±3.56 <0.001 42 43 Waist circumference(cm, mean ± SD) 44 Male 84.79±10.28 87.23±10.57 86.51±11.08 85.28±10.58 <0.001 45 Female 82.64±10.39 82.07±10.31 81.68±10.13 82.83±9.99 <0.001 46 47 SBP(mmHg, mean ± SD) 127.86±20.53 125.58±19.97 124.18±19.54 123.50±19.22 <0.001
48 DBP(mmHg, mean ± SD) 77.76±11.59 78.46±11.92 77.40±11.85 76.71±11.50 <0.001 49 50 Total energy (kcal, mean ± SD) 2371.08±660.96 2371.07±649.17 2446.38±657.54 2606.19±691.99 <0.001 51 Fat energy (kcal, mean ± SD) 649.83±166.88 664.74±167.84 689.14±172.15 704.09±182.27 <0.001 52 53 Carbohydrate energy (kcal, mean ± SD) 1423.31±477.60 1407.78±472.68 1445.63±479.46 1570.50±500.73 <0.001 54 Protein energy (kcal, mean ± SD) 297.93±97.99 298.55±95.55 311.61±97.83 331.60±103.27 <0.001 55 56 Abbreviations: BMI, body mass index; SBP, systemic blood pressure; DBP, diastolic blood pressure. 57 58 59 60 26
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 34 BMJ Open
1 2 3 4 5 Table 2 The ORs (95% CIs) of spicy food intake frequency for abdominal obesity 6 7 Never (N=12291) 1-2 d/wk (N=3252) 3-5 d/wk (N=3241) 6-7 d/wk (N=9989) Each day increment Ptrend 8 9 Model 1 1.000 1.113(1.031, 1.203) 0.992(0.918, 1.072) 1.038(0.985, 1.095) 1.009(0.992, 1.027) 0.308 10 11 Model 2 1.000 1.183(1.090, 1.283) 1.129(1.040, 1.226) 1.118(1.057, 1.183) 1.036(1.017, 1.055) <0.001 12 For peer review only 13 Model 3 1.000 1.185(1.092, 1.286) 1.132(1.043, 1.229) 1.122(1.059, 1.187) 1.059(1.039, 1.078) <0.001 14 15 Model 4 1.000 1.185(1.092, 1.286) 1.131(1.042, 1.228) 1.117(1.055, 1.182) 1.055(1.036, 1.074) <0.001 16 17 Model 5 1.000 1.186(1.093, 1.287) 1.127(1.038, 1.224) 1.104(1.044, 1.169) 1.051(1.032, 1.070) <0.001 18 19 Model 6 1.000 1.185(1.091, 1.286) 1.133(1.043, 1.23) 1.129(1.066, 1.195) 1.059(1.040, 1.079) <0.001 20 21 Model 1: crude model. 22 23 Model 2: adjusted for age, gender, education, marital status, tobacco use, alcohol use and physical activity. 24 25 Model 3: adjusted for model 2 plus total energy intake. 26 27 Model 4: adjusted for model 2 plus protein energy intake. 28 29 Model 5: adjusted for model 2 plus fat energy intake. 30 31 Model 6: adjusted for model 2 plus carbohydrate energy intake. 32 33 34 35 36 37 38 39 40 41 27 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 28 of 34
1 2 3 4 5 Table 3 The subgroup ORs (95% CIs) of spicy food intake frequency for abdominal obesity 6 Variable Never 1-2 d/wk 3-5 d/wk 6-7 d/wk Each day increment P trend 7 Age(year) 8 ≤30 1.000 0.717(0.470-1.095) 0.721(0.478-1.087) 0.981(0.671-1.433) 1.003(0.886-1.136) 0.959 9 31~40 1.000 1.547(1.205-1.986) 1.331(1.032-1.716) 1.394(1.112-1.747) 1.096(1.020-1.178) 0.013 10 41~50 1.000 1.477(1.235-1.765) 1.336(1.113-1.603) 1.221(1.070-1.394) 1.057(1.013-1.104) 0.011 11 51~60 1.000 1.042(0.888-1.223) 1.140(0.970-1.340) 1.141(1.025-1.269) 1.046(1.010-1.084) 0.012 12 61~70 For1.000 peer1.406(1.188-1.665) review1.179(0.998-1.394) only1.011(0.910-1.122) 1.006(0.972-1.041) 0.738 13 >70 1.000 0.945(0.718-1.243) 1.030(0.782-1.357) 0.947(0.798-1.124) 0.985(0.931-1.041) 0.590 14 Gender 15 Male 1.000 1.334(1.167-1.524) 1.192(1.049-1.355) 0.990(0.902-1.088) 0.994(0.964-1.025) 0.697 16 Female 1.000 1.109(1.000-1.230) 1.055(0.948-1.176) 1.196(1.112-1.287) 1.058(1.033-1.083) <0.001 17 Education 18 Illiterate and Elementary 1.000 1.181(1.028-1.356) 1.028(0.896-1.178) 1.042(0.957-1.135) 1.011(0.983-1.04) 0.442 19 Middle school 1.000 1.251(1.100-1.422) 1.146(1.008-1.303) 1.113(1.017-1.219) 1.032(1.002-1.063) 0.037 20 High school and above 1.000 1.173(0.986-1.396) 1.34(1.120-1.604) 1.168(1.008-1.353) 1.061(1.012-1.113) 0.015 21 Marital status 22 Married/cohabiting 1.000 1.188(1.090-1.294) 1.144(1.05-1.247) 1.108(1.044-1.176) 1.033(1.013-1.053) 0.001 23 Widowed/single / divorced/separated 1.000 1.122(0.842-1.495) 0.944(0.705-1.263) 1.093(0.900-1.327) 1.024(0.961-1.091) 0.467 24 Tobacco use 25 Nonsmoker 1.000 1.154(1.050-1.269) 1.110(1.006-1.225) 1.158(1.083-1.238) 1.048(1.025-1.071) <0.001 26 Ex- Smoker 1.000 1.074(0.787-1.465) 1.235(0.928-1.644) 1.097(0.895-1.345) 1.037(0.971-1.109) 0.281 27 Smoker 1.000 1.386(1.138-1.687) 1.094(0.912-1.312) 0.941(0.821-1.079) 0.968(0.926-1.012) 0.150 28 Alcohol use 29 Nondrinker 1.000 1.155(1.053-1.268) 1.045(0.950-1.150) 1.120(1.050-1.194) 1.035(1.013-1.057) 0.002 30 Ex-drinker 1.000 1.077(0.710-1.634) 1.211(0.824-1.780) 1.098(0.847-1.424) 1.036(0.952-1.128) 0.413 31 Drinker 1.000 1.363(1.118-1.662) 1.347(1.117-1.623) 1.069(0.926-1.233) 1.010(0.965-1.058) 0.668 32 Physical activity 33 Low 1.000 1.176(1.023-1.353) 1.222(1.056-1.413) 1.070(0.964-1.188) 1.028(0.993-1.063) 0.116 34 Middle 1.000 1.270(1.103-1.464) 1.098(0.961-1.255) 1.127(1.029-1.234) 1.035(1.005-1.067) 0.023 35 High 1.000 1.115(0.963-1.29) 1.059(0.909-1.234) 1.105(0.998-1.224) 1.031(0.997-1.066) 0.076 36 Adjusted model (except where it is the variable of interest): adjusted for age, gender, education, marital status, tobacco use, alcohol use, physical activity and fat energy intake. 37 38 39 40 41 28 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 29 of 34 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 1 The percentage of abdominal obesity by different categories. 40 90x90mm (300 x 300 DPI) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 34
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 2 ORs (solid lines) and 95% CIs (dashed lines) of spicy food for abdominal obesity from restricted cubic splines. (A: crude model, B: adjusted for age, gender, education, marital status, tobacco use, alcohol 40 use and physical activity, C: adjusted for B plus total energy intake, D: adjusted for B plus protein energy 41 intake, E: adjusted for B plus fat energy intake, F: adjusted for B plus carbohydrate energy intake.) 42 43 90x90mm (300 x 300 DPI) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 34 BMJ Open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 3 The mediation analyses between spicy food intake frequency and abdominal obesity by fat energy intake. 40 41 90x90mm (300 x 300 DPI) 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 34
1 2 3 Supplement Table 1 The mediation analyses between spicy food intake frequency and abdominal obesity by energy intake. 4 5 Energy intake Direct effect, OR(95%CI) Path a, β(95%CI) Path b, OR(95%CI) Indirect effect, OR(95%CI) 6 7 Total 1.047(1.028, 1.067) 64.316(58.799, 69.832) 1.001(0.999, 1.002) 0.999(0.996, 1.001) 8 Protein 1.117(1.063, 1.175) 18.501(16.190, 20.812) 0.999(0.996, 1.001) 0.997(0.993,1.002) 9 10 Fat 1.107(1.053, 1.164) 26.536(22.714, 30.358) 1.003(1.001, 1.004) 1.007(1.003, 1.012) 11 Carbohydrate 1.181(1.130, 1.234) 8.955(7.826, 10.834) 1.001(0.999, 1.003) 1.001(0.998, 1.004) 12 Adjusted model: adjusted for age, gender, education, marital status, tobacco use, alcohol use and physical activity. 13 Path a stands for the effect of spicy food intake frequency on energy intake. 14 Path b stands for the effect of energy intake on abdominal obesity adjusted for spicy food intake frequency. 15 16 17 18 For peer review only 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 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 34 BMJ Open
1 2 STROBE 2007 (v4) Statement—Checklist of items that should be included in reports of cross-sectional studies 3
4 5 Item Section/Topic Recommendation Reported on page # 6 # 7 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 1, 2,3 8 9 (b) Provide in the abstract an informative and balanced summary of what was done and what was found 2,3 10 11 Introduction 12 Background/rationale 2 Explain the scientificFor background peer and rationale forreview the investigation being reported only 5, 6 13 Objectives 3 State specific objectives, including any prespecified hypotheses 6 14 15 Methods 16 Study design 4 Present key elements of study design early in the paper 6 17 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data 6, 7 19 collection 20 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of participants 6, 7 21
22 23 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if 7, 8, 9 24 25 applicable 26 Data sources/ 8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe 7, 8, 9 27 measurement comparability of assessment methods if there is more than one group 28 Bias 9 Describe any efforts to address potential sources of bias 7, 8, 9 29 30 Study size 10 Explain how the study size was arrived at 6, 7 31 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and 7, 8, 9 32 why 33 34 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 9, 10 35 (b) Describe any methods used to examine subgroups and interactions 9, 10 36
37 (c) Explain how missing data were addressed 38 (d) If applicable, describe analytical methods taking account of sampling strategy 39 (e) Describe any sensitivity analyses 40 41 Results 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 34 of 34
1 2 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, 10 3 confirmed eligible, included in the study, completing follow-up, and analysed 4 5 (b) Give reasons for non-participation at each stage 6 (c) Consider use of a flow diagram 7 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential 10 8 9 confounders 10 (b) Indicate number of participants with missing data for each variable of interest 11 Outcome data 15* Report numbers of outcome events or summary measures 10 12 Main results 16 (a) Give unadjustedFor estimates peer and, if applicable, reviewconfounder-adjusted estimates only and their precision (eg, 95% confidence 10, 11, 12 13 14 interval). Make clear which confounders were adjusted for and why they were included 15 (b) Report category boundaries when continuous variables were categorized 16 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period 17 18 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses 12 19 Discussion 20 Key results 18 Summarise key results with reference to study objectives 12 21 22 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and 15, 16 23 magnitude of any potential bias 24 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from 13, 14, 15, 16, 17 25 similar studies, and other relevant evidence 26 27 Generalisability 21 Discuss the generalisability (external validity) of the study results 13, 14, 15, 16, 17 28 Other information 29 30 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on 18 31 which the present article is based 32 33 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies. 34 35 36 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE 37 checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 38 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org. 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60