Obesity mediates the opposite association of education and diabetes in Chinese men and women: Results from the REACTION study

Yuanyue Zhu; Chunyan Hu; Lin Lin; Shuangyuan Wang; Hong Lin; Yanan Huo; Qin Wan; Yingfen Qin; Ruying Hu; Lixin Shi; Qing Su; Xuefeng Yu; Li Yan; Guijun Qin; Xulei Tang; Gang Chen; Min Xu; Yu Xu; Tiange Wang; Zhiyun Zhao; Zhengnan Gao; Guixia Wang; Feixia Shen; Zuojie Luo; Li Chen; Qiang Li; Zhen Ye; Yinfei Zhang; Chao Liu; Youmin Wang; Shengli Wu; Tao Yang; Huacong Deng; Lulu Chen; Tianshu Zeng; Jiajun Zhao; Yiming Mu; Weiqing Wang; Guang Ning; Yufang Bi; Yuhong Chen; Jieli Lu

doi:10.1111/1753-0407.13325

. 2022 Oct 11;14(11):739–748. doi: 10.1111/1753-0407.13325

Show available content in

Obesity mediates the opposite association of education and diabetes in Chinese men and women: Results from the REACTION study

Yuanyue Zhu ^1,^2,^†, Chunyan Hu ^1,^2,^†, Lin Lin ^1,^2,^†, Shuangyuan Wang ^1,², Hong Lin ^1,², Yanan Huo ³, Qin Wan ⁴, Yingfen Qin ⁵, Ruying Hu ⁶, Lixin Shi ⁷, Qing Su ⁸, Xuefeng Yu ⁹, Li Yan ¹⁰, Guijun Qin ¹¹, Xulei Tang ¹², Gang Chen ¹³, Min Xu ^1,², Yu Xu ^1,², Tiange Wang ^1,², Zhiyun Zhao ^1,², Zhengnan Gao ¹⁴, Guixia Wang ¹⁵, Feixia Shen ¹⁶, Zuojie Luo ⁵, Li Chen ¹⁷, Qiang Li ¹⁸, Zhen Ye ⁶, Yinfei Zhang ¹⁹, Chao Liu ²⁰, Youmin Wang ²¹, Shengli Wu ²², Tao Yang ²³, Huacong Deng ²⁴, Lulu Chen ²⁵, Tianshu Zeng ²⁵, Jiajun Zhao ²⁶, Yiming Mu ²⁷, Weiqing Wang ^1,², Guang Ning ^1,², Yufang Bi ^1,², Yuhong Chen ^1,^2,^✉, Jieli Lu ^1,^2,^✉

^¹Department of Endocrine and Metabolic Diseases, Shanghai Institute of Endocrine and Metabolic Diseases, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

^²Shanghai National Clinical Research Center for Metabolic Diseases, Key Laboratory for Endocrine and Metabolic Diseases of the National Health Commission of the PR China, Shanghai Key Laboratory for Endocrine Tumors, State Key Laboratory of Medical Genomics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

^³Jiangxi Provincial People's Hospital Affiliated to Nanchang University, Nanchang, China

^⁴The Affiliated Hospital of Luzhou Medical College, Luzhou, China

^⁵The First Affiliated Hospital of Guangxi Medical University, Nanning, China

^⁶Zhejiang Provincial Center for Disease Control and Prevention, Hangzhou, China

^⁷Affiliated Hospital of Guiyang Medical University, Guiyang, China

^⁸Xinhua Hospital Affiliated to Shanghai Jiaotong University School of Medicine, Shanghai, China

^⁹Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

^¹⁰Sun Yat‐sen Memorial Hospital, Sun Yat‐sen University, Guangzhou, China

^¹¹The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China

^¹²The First Hospital of Lanzhou University, Lanzhou, China

^¹³Fujian Provincial Hospital, Fujian Medical University, Fuzhou, China

^¹⁴Dalian Municipal Central Hospital Affiliated of Dalian Medical University, Dalian, China

^¹⁵The First Hospital of Jilin University, Changchun, China

^¹⁶The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China

^¹⁷Qilu Hospital of Shandong University, Jinan, China

^¹⁸The Second Affiliated Hospital of Harbin Medical University, Harbin, China

^¹⁹Central Hospital of Shanghai Jiading District, Shanghai, China

^²⁰Jiangsu Province Hospital on Integration of Chinese and Western Medicine, Nanjing, China

^²¹The First Affiliated Hospital of Anhui Medical University, Hefei, China

^²²Karamay Municipal People's Hospital, Xinjiang, China

^²³The First Affiliated Hospital of Nanjing Medical University, Nanjing, China

^²⁴The First Affiliated Hospital of Chongqing Medical University, Chongqing, China

^²⁵Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

^²⁶Shandong Provincial Hospital affiliated to Shandong University, Jinan, China

^²⁷Chinese People's Liberation Army General Hospital, Beijing, China

Correspondence , Jieli Lu and Yuhong Chen, Department of Endocrine and Metabolic Diseases, Shanghai Institute of Endocrine and Metabolic Diseases, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 200 025, Shanghai, China. Shanghai National Clinical Research Center for Metabolic Diseases, Key Laboratory for Endocrine and Metabolic Diseases of the National Health Commission of the PR China, Shanghai Key Laboratory for Endocrine Tumors, State Key Laboratory of Medical Genomics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 200 025, Shanghai, China. Email: jielilu@hotmail.com and chenyh70@126.com

^†

Yuanyue Zhu, Chunyan Hu, and Lin Lin contributed equally to this work.

^✉

Corresponding author.

PMCID: PMC9705800 PMID: 36217863

Abstract

Background

Evidence regarding the impact of education on diabetes risk is scarce in developing countries. We aimed to explore the association between education and diabetes within a large population in China and to identify the possible mediators between them.

Methods

Information on educational level and lifestyle factors was collected through questionnaires. Diabetes was diagnosed from self‐report and biochemical measurements. A structural equation model was constructed to quantify the mediation effect of each mediator.

Results

Compared with their least educated counterparts, men with college education had a higher risk of diabetes (odds ratio [OR] 1.19; 95% confidence interval [CI], 1.12–1.27), while college‐educated women were less likely to have diabetes (OR 0.77; 95% CI, 0.73–0.82). Obesity was the strongest mediator in both genders (proportion of mediation: 11.6% in men and 23.9% in women), and its association with education was positive in men (β[SE] 0.0387 [0.0037]) and negative in women (β[SE] −0.0824 [0.0030]). Taken together, all behavioral factors explained 12.4% of the excess risk of diabetes in men and 33.3% in women.

Conclusions

In a general Chinese population, the association between education level and diabetes was positive in men but negative in women. Obesity was the major mediator underlying the education disparities of diabetes risk, with a stronger mediation effect among women.

Keywords: diabetes, education, gender‐dependent, mediation, obesity

Highlights

The association between education level and diabetes in Chinese adults is positive in men but negative in women.
Obesity is the major mediator underlying the gender disparity in the association between education and diabetes.
Combating obesity may help address the burden of diabetes across education levels in both genders.

graphic file with name JDB-14-739-g001.jpg

1. INTRODUCTION

China is the biggest developing country in the world and is currently undergoing a drastic nutrition transtion. The past decades have witnessed an alarming increase in diabetes prevalence, from 5.5% in 2001 ¹ to 12.8% in 2017, ² which is much faster than that in developed countries. To halt the striking rising trend of diabetes, it is essential to identify those at high risk and implement target population‐level strategies.

Low education level is a well‐established risk factor of diabetes in developed countries. ³ , ⁴ , ⁵ However, in developing countries such as China, the association varied with time and region. ⁶ , ⁷ , ⁸ Therefore, it is crucial to renew the related knowledge in a nationally representative cohort of the Chinese population.

Previous studies investigating the education–diabetes association have already identified several possible mediators. ⁹ , ¹⁰ , ¹¹ Unhealthy behaviors like obesity, smoking, and depressive symptoms were shown to play a role in this association. ⁹ , ¹² , ¹³ However, few studies assessed the specific mediating effect of each mediator separately. The knowledge gap added the difficulties in understanding the educational disparities of diabetes risk. Indeed, in most previous studies, the mediation effect of a certain variable was discovered by the difference in the estimates between adjusted and unadjusted models. ¹⁴ , ¹⁵ , ¹⁶ Admittedly, such a design has methodological issues as the inclusion of a single variable might alter the convergence of the whole statistical model, and subsequently affect the estimates of each mediation effect in an uncontrollable way. Therefore, an advanced statistical method tailored for mediation analysis is warranted for evaluating the path‐specific effects of interest. ¹⁷

Furthermore, evidence for sex differences in the education–diabetes association has been indicated in several studies, where the link between education and diabetes seemed stronger among women than men, ¹⁸ , ¹⁹ or even in the opposite direction. ³ , ²⁰ Therefore, it might be necessary to separate the discussion by gender. Hence, in the current study, we aimed to examine the relationship between education level and diabetes in both genders. We also tried to search for possible mediators, and individually assess the specific mediation effect of each mediator. Obesity, smoking history, unhealthy diet and activity status, non‐ideal sleep time, and depression are all well‐recognized behavioral risk factors of diabetes in previous literature; more importantly, they are all modifiable factors, which have the potential to be corrected. Given this, these variables were chosen to be the potential mediators in the current mediation study.

2. METHODS

2.1. Study population

The REACTION study is a multicenter, nationwide, population‐based study conducted in Chinese individuals aged 40 years or older. ²¹ Briefly, 259 657 participants were recruited from 25 communities across mainland China, with 253 490 individuals providing information on education level. Among them, 8609 participants were further excluded because of missing information on both diabetes diagnosis and biomedical measurement. Finally, a total of 244 881 individuals were included in the analysis. The study was approved by the Committee on Human Research at Ruijin Hospital, and written informed consent was obtained from each participant.

2.2. Data collection and clinical evaluation

Information on medical history, education level, residence, occupation, lifestyle factors (including smoking behavior, physical activity, and dietary patterns), and family history of diabetes was collected via in‐person interviews with a standard questionnaire. In the current study, participants were interviewed about their highest attained education. In response of the question, the participants were categorized into four groups (i.e., primary school or below, middle school, high school, and college or above). A validated questionnaire (Patient Health Questionnaire 9, PHQ‐9) was also administered to assess mental health. ²² Participants underwent measurements of height and weight using a standard protocol with light clothes and no shoes. Body mass index (BMI) was calculated as body weight in kilograms divided by body height in meters squared (kg/m²).

After an overnight fast of at least 10 h, blood samples were drawn for biochemical tests. Plasma was obtained at 0 and 2 h for the measurement of fasting glucose and 2 h postload glucose. Plasma glucose concentrations were evaluated at local hospitals using the glucose oxidase or hexokinase method within 2 h after blood sample collection. The Hemoglobin Capillary Collection System (Bio‐Rad Laboratories, Hercules, California) was used to collect finger capillary whole blood and shipped at 2–6°C to the central laboratory of the study to measure the level of glycosylated hemoglobin (HbA1c).

2.3. Definition of exposure, outcome, and covariates

2.3.1. Education levels

Education level was self‐reported and categorized into four groups, including primary education or below, middle school, high school, and college or above. In the current mediation study, education level was further classified into two categories based on receiving high school education or not. In China, primary and middle school education is compulsory; therefore, participants who completed high school were considered relatively well educated. ²³ This classification was also in accordance with that in our previous study. ²⁴

2.3.2. Diabetes

According to the American Diabetes Association (ADA) 2010 criteria, ²⁵ diabetes was defined as (1) a self‐reported previous diagnosis by health‐care professionals, (2) fasting plasma glucose level of 126 mg/dl (7.0 mmol/L) or higher, (3) 2‐h plasma glucose level of 200 mg/dl (11.1 mmol/L) or higher, or (4) HbA1c concentration of 6.5% or higher.

2.3.3. Covariates

Behavioral factors included obesity, smoking, physical activity, dietary patterns, and depression. For obesity, the cutoff point for Asian populations was used (BMI ≥ 25 kg/m²). ²⁶ For sleep duration, a sleep time of 6–8 h was considered ideal, as is verified by our previous study. ²⁷ Smoking status was binarized as ever/never smoker because diabetes risk changes remarkably once smoked. ²⁸ Healthy physical activity was defined according to the 2008 Physical Activity Guidelines for Americans. Participants who engaged in moderate‐intensity exercise for ≥150 min/week or vigorous‐intensity exercise for ≥75 min/week were considered physically active. ²⁹ Healthy dietary patterns were defined as a dietary score of 4. ³⁰ For depressive symptoms, each of the nine items of the PHQ‐9 is scored as 0, 1, 2, or 3, and the total score was summed; a PHQ‐9 score of 5–27 represented overt depression. ²²

Other covariates included in the analysis were age, economic‐geographic residence and family history of diabetes. Economic‐geographic residences of China were categorized into four groups according to the per capita disposable income of households from National Bureau of Statistics of China as Eastern, Northeastern, Central, and Western, which reflected different levels of economic development in China. ³¹ Age was defined as the age of the participants at recruitment. Family history of diabetes was defined as diabetes diagnosis among first‐degree relatives (i.e., direct blood relatives).

2.4. Statistical analysis

The total effect of education on diabetes was estimated with multivariable logistic regression; odds ratios (ORs) and 95% confidence intervals (95% CIs) of education level (four categories) for diabetes are presented for men and women separately in Table 2. Model 1 was the crude model, model 2 was further adjusted for age, economic‐geographic residence, and family history of diabetes. In model 3, obesity (yes/no), ever smoker (yes/no), healthy diet (yes/no), ideal sleep time (yes/no), healthy physical activity (yes/no), and depression (yes/no) were also adjusted.

TABLE 2.

Association between education level and diabetes in men and women

Men	Primary school or below	Middle school	High school	College or above
Case (%)	15 033 (27.4)	11 880 (23.0)	8424 (20.4)	2458 (19.6)
Model 1	Ref	1.12 (1.07–1.16)	1.26 (1.21–1.32)	1.39 (1.32–1.46)
Model 2	Ref	1.26 (1.20–1.31)	1.39 (1.33–1.46)	1.40 (1.33–1.47)
Model 3	Ref	1.14 (1.09–1.21)	1.24 (1.17–1.31)	1.19 (1.12–1.27)
Women
Case (%)	4782 (25.2)	8208 (27.3)	6625 (29.8)	4167 (31.8)
Model 1	Ref	0.79 (0.77–0.81)	0.68 (0.66–0.70)	0.65 (0.62–0.68)
Model 2	Ref	1.02 (0.99–1.05)	0.87 (0.85–0.90)	0.77 (0.73–0.81)
Model 3	Ref	0.98 (0.94–1.01)	0.86 (0.83–0.90)	0.77 (0.73–0.82)

Open in a new tab

Note: Model 1: crude model. Model 2: adjusted for age, sex, and economic‐geographic residence. Model 3: further adjusted for obesity (yes/no), ever smoker (yes/no), healthy diet (yes/no), ideal sleep time (yes/no), healthy physical activity (yes/no), and depression (yes/no).

For the behavioral factors of interest, multivariable logistic regression was used to estimate their independent effect on diabetes one by one, separately in men and women. Age, education level, economic‐geographic residence, and family history of diabetes were adjusted for the analysis of each behavioral factor. The variables signifiantly associated with diabetes were included in a mediation model, and the mediation effect was further quantified. In this study, mediation analysis was conducted with a structural equation model (SEM), which is a recommended approach for mediation analysis with multiple mediators. ³² , ³³ SEMs are usually elaborated by path diagrams, with nodes representing the variables and arrows representing the relationship between them. ³³ A schematic diagram is shown in Supplement Figure S2. In the current model, the pathway between education and mediators was denoted as path α, the one between mediators and diabetes as path β, and the overall association of education with diabetes was marked as path c. Age, economic‐geographic residence, and family history of diabetes were used as controlling variables. The indirect effect was obtained by multiplying the coefficients of path α and β. The proportion of mediation for each mediator was calculated by dividing the total effect into indirect effect. SAS version 9.4 (SAS Institute) was used to conduct multivariate logistic regression analyses, and R version 4.0.3 (R Foundation for Statistical Computing) was used to construct an SEM for analyzing the complexity of associations between mediators and outcome using the “Lavvan” package. ³⁴

3. RESULTS

The distribution of behavioral factors by education level is shown in Table 1 (education level of four categories) and Supplement Table S1 (education level of two categories: high school education or not). Generally, participants with a higher education level tended to adopt an more ideal lifestyle compared with their disadvantaged counterparts: They were more likely to have a healthier diet, sleep time, and physical activity and less likely to smoke. However, the prevalence of both obesity and diabetes changed with education level in the opposite direction by gender. Compared with those with the lowest education level, the prevalence of obesity was 49.6% versus 39.0% in men, and 32.1% versus 45.9% in women with the highest education level, respectively (Figure 1A). Similarly, Figure 1B shows that the prevalence of diabetes rises with increasing education level in a graded manner among men (25.2% for primary school or below and 31.8% for college or above), while in women it decreases as education level rises (27.4% for primary school or below and 19.6% for college or above).

TABLE 1.

Baseline characteristics according to education level in men and women

	Men				Women
	Primary school or below	Middle school	High school	College or above	Primary school or below	Middle school	High school	College or above
Number, n (%)	19 009	30 117	22 257	13 106	54 860	51 702	41 301	12 529
Age (years)	62.24 ± 9.75	57.31 ± 9.29	56.85 ± 9.49	58.55 ± 10.84	60.79 (9.64)	55.33 (8.60)	54.36 (8.00)	54.61 (9.77)
Body mass index (kg/m²)	24.21 ± 3.62	24.86 ± 3.48	24.94 ± 3.51	25.09 ± 3.37	24.86 (3.74)	24.71 (3.64)	24.21 (3.51)	23.84 (3.36)
Obesity, n (%)	7291 (39.0)	13 919 (47.0)	10 500 (48.0)	6370 (49.6)	24 850 (45.9)	21 855 (43.0)	15 222 (37.5)	3934 (32.1)
Sleep time (h)	2.59 ± 0.54	2.49 ± 0.55	2.43 ± 0.56	2.40 ± 0.55	2.53 (0.56)	2.41 (0.56)	2.36 (0.55)	2.34 (0.55)
Ideal sleep time (%)	5920 (35.8)	11 973 (45.3)	10 161 (50.3)	6590 (54.2)	19 360 (40.2)	24 166 (52.1)	21 648 (56.6)	6828 (58.6)
Ever smoker, n (%)	7627 (40.1)	12 798 (42.5)	8366 (37.6)	3577 (27.3)	840 (1.5)	752 (1.5)	516 (1.2)	94 (0.8)
Healthy diet, n (%)	7882 (54.3)	14 368 (57.9)	11 945 (62.8)	7486 (65.5)	22 257 (52.6)	25 828 (59.2)	23 716 (65.6)	7319 (66.5)
Healthy physical activity, n (%)	1542 (8.5)	3589 (12.3)	3468 (16.0)	2775 (21.7)	4354 (8.3)	6695 (13.3)	6325 (15.7)	2192 (18.0)
Depression, n (%)	498 (2.9)	837 (3.1)	736 (3.6)	523 (4.2)	2360 (4.7)	2282 (4.8)	2200 (5.7)	832 (7.0)
Diabetes, n (%)	4782 (25.2)	8208 (27.3)	6625 (29.8)	4167 (31.8)	15 033 (27.4)	11 880 (23.0)	8424 (20.4)	2458 (19.6)
Family history of diabetes, n (%)	965 (5.3)	3202 (11.0)	3191 (14.8)	2107 (16.5)	3364 (6.4)	7022 (14.0)	8091 (20.1)	2807 (23.0)
Economic‐geographic residence (%)
Eastern	10 766 (56.6)	17 492 (58.1)	12 884 (57.9)	7141 (54.5)	29 173 (53.2)	27 382 (53.0)	22 961 (55.6)	6529 (52.1)
Northeastern	4715 (24.8)	4759 (15.8)	2904 (13.0)	1348 (10.3)	13 463 (24.5)	6404 (12.4)	3856 (9.3)	1089 (8.7)
Central	3093 (16.3)	6069 (20.2)	4888 (22.0)	3537 (27.0)	10 619 (19.4)	13 207 (25.5)	10 583 (25.6)	3497 (27.9)
Western	435 (2.3)	1797 (6.0)	1581 (7.1)	1080 (8.2)	1605 (2.9)	4709 (9.1)	3901 (9.4)	1414 (11.3)

Open in a new tab

Note: Continuous variables are represented as mean ± SD and categorical variables are shown as case (%).

The prevalence of obesity and diabetes with education level in men and women. (A). The prevalence of obesity with education level in men and women. (B). The prevalence of diabetes with education level in men and women. The darkness of the bar color denotes the education level

The association between education level and diabetes is depicted by gender in Table 2. After adjusting for age, economic‐geographic residence, lifestyle factors, and obesity, the association of diabetes and education level remained positive in men and negative in women. Compared with the least‐educated individuals, the ORs (95% CI) of diabetes among men and women with college education were 1.19 (1.12–1.27) and 0.77 (0.73–0.82), respectively. When taking education level as a dichotomous variable (high school education or not), participants with high school education were at a 11% higher risk in men and 16% lower risk in women, which is shown in Supplement Table S2.

We then examined the effect of each behavioral factor on diabetes in both genders. Supplement Figure S1 summarizes the ORs of the potential factors on diabetes after controlling for age, economic‐geographic residence, family history of diabetes, and education. In men, obesity and being a smoker were significantly related to the risk of diabetes, while in women, significant variables were obesity, healthy diet, healthy physical activity, and ideal sleep time. Therefore, these factors were included in the mediation model as potential mediators. For men, the proportion of mediation was 11.6% for obesity and 0.8% for being an ever smoker (Table 3), whereas among women, the proportion for mediation was 23.9% for obesity, 3.1% for healthy diet, 2.1% for healthy physical activity, and 4.2% for ideal sleep time (Table 3). Overall, behavioral factors mediated 12.4% and 33.3% of the total association between education and diabetes in men and women.

TABLE 3.

Mediation analysis of the association between education level and diabetes in men and women

	X → M (α path)		M → Y (β path)		Indirect effect (α*β)		Total effect		Mediation (%)
Men	β	SE	β	SE	β	SE	β	SE
Obesity	0.0387	0.0037	0.1102	0.0030	0.0043	0.0004	0.0367	0.0034	11.6%
Ever smoker	−0.0806	0.0035	−0.0037	0.0031	0.0003	0.0003	0.0367	0.0034	0.8%
Women	β	SE	β	SE	β	SE
Obesity	−0.0824	0.0030	0.1121	0.0024	−0.0092	0.0004	−0.0387	0.0025	23.9%
Healthy diet	0.0669	0.0029	−0.0179	0.0027	−0.0012	0.0002			3.1%
Healthy physical activity	0.0500	0.0023	−0.0164	0.0035	−0.0008	0.0002			2.1%
Ideal sleep time	0.0783	0.0030	−0.0207	0.0024	−0.0016	0.0002			4.2%

Open in a new tab

Note: Age, economic‐geographic residence, and family history of diabetes were used as controlling variables.

4. DISCUSSION

In the present study, we found that the association of diabetes with education level was positive in men but negative in women. Moreover, obesity was the major mediator underlying the observed association. To our knowledge, this is the first mediation analysis in a Chinese population concerning the gender‐specific education–diabetes association. These results bring new perspectives to the prevention of diabetes at the population level.

Compared with other indicators, education is superior in representing socioeconomic levels. Education level can incorporate socioeconomic factors in both earlier and later life and remains relatively stabilized throughout the life course. Furthermore, information on education is less private and thus more easily to obtain and measure. ³⁵ As a result, education was employed as proxy to represent socioeconomic position in our study, as is the case in much of the proceeding literature. ³⁶ , ³⁷

To date, a number of studies have explored the association between educational level and diabetes, with some further discussing the potential mediators. However, the results were conflicting with diverse economic development and ethnicity. For example, in developed countries, the association between educational level and diabetes is unitedly negative5, 38 while in less developed countries, such as China and India, the results were far more discordant. ³⁹ , ⁴⁰ Even within the Chinese population, the relationship varied with time, region, and overall education level of the study population. In 2006, a study conducted in Nanjing reported that participants with higher education were at higher risk of diabetes after adjustment for gender,³⁹ while in a previous Qingdao study, low education was an established risk factor of diabetes. ⁴¹ Another study at approximately the same period using the China Kadoorie Biobank (CKB) population reported no liner association between education and diabetes. ²⁰ However, in contrast to all these studies, a gender‐specific gradient relationship between education and diabetes was observed in our study, with men with a high level of education and women with a low level of education having a high risk of diabetes. This finding is in line with what was found in Korean ⁴² and Swedish populations, ¹⁴ where the ORs of diabetes among men with higher education and women with lower education ranged from 1.64 to 2.3. It is indicated that gender differences in the education–diabetes association are most likely to occur in low‐ or middle‐income countries which are currently undergoing economic and nutritional transformation. ⁸ , ⁴³

The gendered relationship between education and diabetes could be a result of the variation of BMI with education, as Wu et al. put it. ²⁰ However, no further elucidation was given in that study. Similarly, we also found that obesity is the major mediator in both genders. Given its nonnegligible mediation effect size and the opposite connection with education level (positive in men and negative in women), it might be one possible explanation in the contrasting association between education and diabetes by gender.

Nevertheless, other mediating factors varied in type and quantity by gender. Apart from obesity, smoking for men, physical activity, diet, and sleep time for women also played important mediating roles in the education–diabetes relationship. For obesity alone, the mediation proportion of obesity was 23.9% in women, more than double that (11.6%) in men. One possible explanation is that the association between education and obesity is stronger in women (path α in Table 2). The association between education level and obesity was stronger among women, which was consistent with a study from Thailand. ⁴⁴ Taking all the results together, it could be speculated that behavioral factors have a stronger impact on diabetes among women, which is consistent with previous findings. ¹⁴ , ⁴⁵ In this case, maintaining a healthy lifestyle might be most useful for the prevention of diabetes among women with low education.

Furthermore, it is noteworthy that even when all the mediators are combined, they can only explain a modest percentage (33.3% for women and 12.4% for men) in the total effect of education on diabetes, which was reported before as well. ⁴⁶ , ⁴⁷ Thus, it is implied that other potential mediators such as marital status, household income, and insurance type still remained undiscovered, and the direct effect of education is remarkable. Nevertheless, efforts aiming at lifestyle modification should be encouraged in people with low education, particularly women.

The strengths of this study include the large nationally representative sample, the comprehensive inclusion of potential mediators, and the advanced statistical method for mediation analysis. However, there are some limitations to be addressed. Firstly, it was a cross‐sectional study, which limits our interpretation of causality underlying the association between education and diabetes. However, the highest educational level usually remained consistent after early adulthood ⁴⁸ and thus less likely led to a reversed causation. Secondly, information on education level, diet, and physical activity was self‐reported, which might result in a reported bias. Thirdly, people aged less than 40 years old were not included in the current study due to our study design. However, as shown in previous studies, the education–diabetes association is more evident among younger individuals, ⁴⁹ which means our results might underestimate, but not overestimate, the total effect of education on diabetes.

In conclusion, we found a gender‐specific association between education and diabetes in China, where women with low education and men with high education were at higher risk of diabetes. The mediation effect of obesity might be one possible explanation in this gender difference. Other behavioral factors such as physical activity and sleep time may also play a mediating role, but this is only evident in women. Therefore, mitigating obesity is of great potential to curb the educational disparities of diabetes risk in both genders, and promoting a healthy lifestyle is especially helpful for women.

DISCLOSURE

None declared.

Supporting information

Appendix S1 Supporting Information.

Click here for additional data file.^{(489.1KB, docx)}

ACKNOWLEDGEMENTS

This work was supported by the National Natural Science Foundation of China (Grant No. 81970728, 81930021, 81970691, 82170819, and 21904084), Shanghai Outstanding Academic Leaders Plan (Grant No. 20XD1422800), Shanghai Medical and Health Development Foundation (Grant No. DMRFP_I_01), Clinical Research Plan of SHDC (Grant No. SHDC2020CR3064B, SHDC2020CR1001A), Clinical Research Project of Shanghai Municipal Health Commission (20214Y0002) and Science and Technology Committee of Shanghai (Grant No. 19411964200 and 20Y11905100).

Zhu Y, Hu C, Lin L, et al. Obesity mediates the opposite association of education and diabetes in Chinese men and women: Results from the REACTION study. Journal of Diabetes. 2022;14(11):739‐748. doi: 10.1111/1753-0407.13325

Funding information National Natural Science Foundation of China, Grant/Award Numbers: 81970728, 81930021, 81970691, 82170819, 21904084; Science and Technology Committee of Shanghai, Grant/Award Numbers: 19411964200, 20Y11905100; Shanghai Medical and Health Development Foundation, Grant/Award Number: DMRFP_I_01; Clinical Research Plan of SHDC, Grant/Award Numbers: SHDC2020CR3064B, SHDC2020CR1001A; Shanghai Outstanding Academic Leaders Plan, Grant/Award Number: 20XD1422800; Clinical Research Project of Shanghai Municipal Health Commission, Grant/Award Number: 20214Y0002

Contributor Information

Yuhong Chen, Email: chenyh70@126.com.

Jieli Lu, Email: jielilu@hotmail.com.

REFERENCES

1. Chan JC, Malik V, Jia W, et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology. JAMA. 2009;301(20):2129‐2140. [DOI] [PubMed] [Google Scholar]
2. Li Y, Teng D, Shi X, et al. Prevalence of diabetes recorded in mainland China using 2018 diagnostic criteria from the American Diabetes Association: national cross sectional study. BMJ. 2020;369:m997. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Espelt A, Arriola L, Borrell C, Larrañaga I, Sandín M, Escolar‐Pujolar A. Socioeconomic position and type 2 diabetes mellitus in Europe 1999‐2009: a panorama of inequalities. Curr Diabetes Rev. 2011;7(3):148‐158. [DOI] [PubMed] [Google Scholar]
4. Williams ED, Magliano DJ, Zimmet PZ, et al. Area‐level socioeconomic status and incidence of abnormal glucose metabolism: the Australian diabetes, obesity and lifestyle (AusDiab) study. Diabetes Care. 2012;35(7):1455‐1461. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Sacerdote C, Ricceri F, Rolandsson O, et al. Lower educational level is a predictor of incident type 2 diabetes in European countries: the EPIC‐InterAct study. Int J Epidemiol. 2012;41(4):1162‐1173. [DOI] [PubMed] [Google Scholar]
6. Seiglie JA, Marcus ME, Ebert C, et al. Diabetes prevalence and its relationship with education, wealth, and BMI in 29 low‐ and middle‐income countries. Diabetes Care. 2020;43(4):767‐775. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Wang A, Stronks K, Arah OA. Global educational disparities in the associations between body mass index and diabetes mellitus in 49 low‐income and middle‐income countries. J Epidemiol Community Health. 2014;68(8):705‐711. [DOI] [PubMed] [Google Scholar]
8. Williams J, Allen L, Wickramasinghe K, Mikkelsen B, Roberts N, Townsend N. A systematic review of associations between non‐communicable diseases and socioeconomic status within low‐ and lower‐middle‐income countries. J Glob Health. 2018;8(2):020409. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Rodríguez López S, Tumas N. Educational disparities in diabetes: a mediation analysis through BMI among urban adults from Argentina. Nutr Clin Diet Hosp. 2020;40(2):128‐134. [Google Scholar]
10. Hauger H, Groth MV, Ritz C, et al. Socio‐economic differences in cardiometabolic risk markers are mediated by diet and body fatness in 8‐ to 11‐year‐old Danish children: a cross‐sectional study. Public Health Nutr. 2016;19(12):2229‐2239. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Steele CJ, Schottker B, Marshall AH, et al. Education achievement and type 2 diabetes‐what mediates the relationship in older adults? Data from the ESTHER study: a population‐based cohort study. BMJ Open. 2017;7(4):e013569. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Rahman M, Nakamura K, Hasan SMM, Seino K, Mostofa G. Mediators of the association between low socioeconomic status and poor glycemic control among type 2 diabetics in Bangladesh. Sci Rep. 2020;10(1):6690. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Smith PM, Smith BT, Mustard CA, Lu H, Glazier RH. Estimating the direct and indirect pathways between education and diabetes incidence among Canadian men and women: a mediation analysis. Ann Epidemiol. 2013;23(3):143‐149. [DOI] [PubMed] [Google Scholar]
14. Agardh EE, Ahlbom A, Andersson T, et al. Explanations of socioeconomic differences in excess risk of type 2 diabetes in Swedish men and women. Diabetes Care. 2004;27(3):716‐721. [DOI] [PubMed] [Google Scholar]
15. Maty SC, Everson‐Rose SA, Haan MN, Raghunathan TE, Kaplan GA. Education, income, occupation, and the 34‐year incidence (1965‐99) of type 2 diabetes in the Alameda County study. Int J Epidemiol. 2005;34(6):1274‐1281. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Loucks EB, Magnusson KT, Cook S, Rehkopf DH, Ford ES, Berkman LF. Socioeconomic position and the metabolic syndrome in early, middle, and late life: evidence from NHANES 1999‐2002. Ann Epidemiol. 2007;17(10):782‐790. [DOI] [PubMed] [Google Scholar]
17. Lange T, Rasmussen M, Thygesen LC. Assessing natural direct and indirect effects through multiple pathways. Am J Epidemiol. 2014;179(4):513‐518. [DOI] [PubMed] [Google Scholar]
18. Chung GK, Lai FTT, Yeoh EK, Chung RY. Gender‐specific trends of educational inequality in diagnosed diabetes from 1999 to 2014 in Hong Kong: a serial cross‐sectional study of 97,481 community‐dwelling Chinese adults. Popul Health Metr. 2021;19(1):37. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Iakunchykova O, Averina M, Wilsgaard T, et al. What factors explain the much higher diabetes prevalence in Russia compared with Norway? Major sex differences in the contribution of adiposity. BMJ Open Diabetes Res Care. 2021;9(1):e002021. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Wu H, Bragg F, Yang L, et al. Sex differences in the association between socioeconomic status and diabetes prevalence and incidence in China: cross‐sectional and prospective studies of 0.5 million adults. Diabetologia. 2019;62(8):1420‐1429. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Bi Y, Lu J, Wang W, et al. Cohort profile: risk evaluation of cancers in Chinese diabetic individuals: a longitudinal (REACTION) study. J Diabetes. 2014;6(2):147‐157. [DOI] [PubMed] [Google Scholar]
22. Kroenke K, Spitzer RL, Williams JB. The PHQ‐9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606‐613. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Meng D, Poon‐Mcbrayer KF, Farnsworth EB. The development of special education in China: A sociocultural review. Remedial Spec Educ. 2001;22(5):288‐298. [Google Scholar]
24. Hu C, Zhang Y, Lin L, et al. Gestational hyperglycemia and the risk of cardiovascular diseases among elderly Chinese women: findings from the REACTION study. J Diabetes. 2021;13(12):949‐959. [DOI] [PubMed] [Google Scholar]
25. American Diabetes Association . Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(Suppl 1):S62‐S69. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Pan WH, Yeh WT. How to define obesity? Evidence‐based multiple action points for public awareness, screening, and treatment: an extension of Asian‐Pacific recommendations. Asia Pac J Clin Nutr. 2008;17(3):370‐374. [PubMed] [Google Scholar]
27. Zheng R, Niu J, Wu S, et al. Gender and age differences in the association between sleep characteristics and fasting glucose levels in Chinese adults. Diabetes Metab. 2021;47(2):101174. [DOI] [PubMed] [Google Scholar]
28. Akter S, Goto A, Mizoue T. Smoking and the risk of type 2 diabetes in Japan: A systematic review and meta‐analysis. J Epidemiol. 2017;27(12):553‐561. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. U.S. Department of Health and Human Services . 2008. Physical Activity Guidelines for Americans.www.health.gov/paguidelines.
30. Lloyd‐Jones DM, Hong Y, Labarthe D, et al. Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association's strategic impact goal through 2020 and beyond. Circulation. 2010;121(4):586‐613. [DOI] [PubMed] [Google Scholar]
31. Jiang H, Sun Z, Guo H, et al. An assessment of urbanization sustainability in China between 1990 and 2015 using land use efficiency indicators. npj Urban Sustain. 2021;1(1):34. [Google Scholar]
32. Baron RM, Kenny DA. The moderator‐mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173‐1182. [DOI] [PubMed] [Google Scholar]
33. MacKinnon DP, Valente MJ. Mediation from multilevel to structural equation modeling. Ann Nutr Metab. 2014;65(2–3):198‐204. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Rosseel Y. The lavvan tutorial.
35. Darin‐Mattsson A, Fors S, Kåreholt I. Different indicators of socioeconomic status and their relative importance as determinants of health in old age. Int J Equity Health. 2017;16(1):173. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Mathisen J, Jensen AKG, Andersen I, Andersen GS, Hvidtfeldt UA, Rod NH. Education and incident type 2 diabetes: quantifying the impact of differential exposure and susceptibility to being overweight or obese. Diabetologia. 2020;63(9):1764‐1774. [DOI] [PubMed] [Google Scholar]
37. Oshio T, Kan M. Educational level as a predictor of the incidences of non‐communicable diseases among middle‐aged Japanese: a hazards‐model analysis. BMC Public Health. 2019;19(1):852. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Espelt A, Borrell C, Roskam AJ, et al. Socioeconomic inequalities in diabetes mellitus across Europe at the beginning of the 21st century. Diabetologia. 2008;51(11):1971‐1979. [DOI] [PubMed] [Google Scholar]
39. Xu F, Yin XM, Zhang M, Leslie E, Ware R, Owen N. Family average income and diagnosed type 2 diabetes in urban and rural residents in regional mainland China. Diabet Med. 2006;23(11):1239‐1246. [DOI] [PubMed] [Google Scholar]
40. Ramachandran A, Snehalatha C, Kapur A, et al. High prevalence of diabetes and impaired glucose tolerance in India: National Urban Diabetes Survey. Diabetologia. 2001;44(9):1094‐1101. [DOI] [PubMed] [Google Scholar]
41. Shang X, Li J, Tao Q, et al. Educational level, obesity and incidence of diabetes among Chinese adult men and women aged 18–59 years old: an 11‐year follow‐up study. PLoS One. 2013;8(6):e66479. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Park SJ, Kang HT, Nam CM, Park BJ, Linton JA, Lee YJ. Sex differences in the relationship between socioeconomic status and metabolic syndrome: the Korean National Health and nutrition examination survey. Diabetes Res Clin Pract. 2012;96(3):400‐406. [DOI] [PubMed] [Google Scholar]
43. Imkampe AK, Gulliford MC. Increasing socio‐economic inequality in type 2 diabetes prevalence‐‐repeated cross‐sectional surveys in England 1994–2006. Eur J Public Health. 2011;21(4):484‐490. [DOI] [PubMed] [Google Scholar]
44. Aekplakorn W, Hogan MC, Chongsuvivatwong V, et al. Trends in obesity and associations with education and urban or rural residence in Thailand. Obesity (Silver Spring). 2007;15(12):3113‐3121. [DOI] [PubMed] [Google Scholar]
45. Kumari M, Head J, Marmot M. Prospective study of social and other risk factors for incidence of type 2 diabetes in the Whitehall II study. Arch Intern Med. 2004;164(17):1873‐1880. [DOI] [PubMed] [Google Scholar]
46. Wikström K, Lindström J, Tuomilehto J, et al. Socio‐economic differences in dysglycemia and lifestyle‐related risk factors in the Finnish middle‐aged population. Eur J Public Health. 2010;21(6):768‐774. [DOI] [PubMed] [Google Scholar]
47. Demakakos P, Marmot M, Steptoe A. Socioeconomic position and the incidence of type 2 diabetes: the ELSA study. Eur J Epidemiol. 2012;27(5):367‐378. [DOI] [PubMed] [Google Scholar]
48. Härkönen J, Lindberg M, Karlsson L, Karlsson H, Scheinin NM. Education is the strongest socio‐economic predictor of smoking in pregnancy. Addiction. 2018;113(6):1117‐1126. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Kim SR, Han K, Choi JY, et al. Age‐ and sex‐specific relationships between household income, education, and diabetes mellitus in Korean adults: the Korea National Health and nutrition examination survey, 2008‐2010. PLoS One. 2015;10(1):e0117034. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix S1 Supporting Information.

Click here for additional data file.^{(489.1KB, docx)}

[jdb13325-bib-0001] 1. Chan JC, Malik V, Jia W, et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology. JAMA. 2009;301(20):2129‐2140. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0002] 2. Li Y, Teng D, Shi X, et al. Prevalence of diabetes recorded in mainland China using 2018 diagnostic criteria from the American Diabetes Association: national cross sectional study. BMJ. 2020;369:m997. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0003] 3. Espelt A, Arriola L, Borrell C, Larrañaga I, Sandín M, Escolar‐Pujolar A. Socioeconomic position and type 2 diabetes mellitus in Europe 1999‐2009: a panorama of inequalities. Curr Diabetes Rev. 2011;7(3):148‐158. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0004] 4. Williams ED, Magliano DJ, Zimmet PZ, et al. Area‐level socioeconomic status and incidence of abnormal glucose metabolism: the Australian diabetes, obesity and lifestyle (AusDiab) study. Diabetes Care. 2012;35(7):1455‐1461. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0005] 5. Sacerdote C, Ricceri F, Rolandsson O, et al. Lower educational level is a predictor of incident type 2 diabetes in European countries: the EPIC‐InterAct study. Int J Epidemiol. 2012;41(4):1162‐1173. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0006] 6. Seiglie JA, Marcus ME, Ebert C, et al. Diabetes prevalence and its relationship with education, wealth, and BMI in 29 low‐ and middle‐income countries. Diabetes Care. 2020;43(4):767‐775. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0007] 7. Wang A, Stronks K, Arah OA. Global educational disparities in the associations between body mass index and diabetes mellitus in 49 low‐income and middle‐income countries. J Epidemiol Community Health. 2014;68(8):705‐711. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0008] 8. Williams J, Allen L, Wickramasinghe K, Mikkelsen B, Roberts N, Townsend N. A systematic review of associations between non‐communicable diseases and socioeconomic status within low‐ and lower‐middle‐income countries. J Glob Health. 2018;8(2):020409. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0009] 9. Rodríguez López S, Tumas N. Educational disparities in diabetes: a mediation analysis through BMI among urban adults from Argentina. Nutr Clin Diet Hosp. 2020;40(2):128‐134. [Google Scholar]

[jdb13325-bib-0010] 10. Hauger H, Groth MV, Ritz C, et al. Socio‐economic differences in cardiometabolic risk markers are mediated by diet and body fatness in 8‐ to 11‐year‐old Danish children: a cross‐sectional study. Public Health Nutr. 2016;19(12):2229‐2239. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0011] 11. Steele CJ, Schottker B, Marshall AH, et al. Education achievement and type 2 diabetes‐what mediates the relationship in older adults? Data from the ESTHER study: a population‐based cohort study. BMJ Open. 2017;7(4):e013569. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0012] 12. Rahman M, Nakamura K, Hasan SMM, Seino K, Mostofa G. Mediators of the association between low socioeconomic status and poor glycemic control among type 2 diabetics in Bangladesh. Sci Rep. 2020;10(1):6690. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0013] 13. Smith PM, Smith BT, Mustard CA, Lu H, Glazier RH. Estimating the direct and indirect pathways between education and diabetes incidence among Canadian men and women: a mediation analysis. Ann Epidemiol. 2013;23(3):143‐149. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0014] 14. Agardh EE, Ahlbom A, Andersson T, et al. Explanations of socioeconomic differences in excess risk of type 2 diabetes in Swedish men and women. Diabetes Care. 2004;27(3):716‐721. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0015] 15. Maty SC, Everson‐Rose SA, Haan MN, Raghunathan TE, Kaplan GA. Education, income, occupation, and the 34‐year incidence (1965‐99) of type 2 diabetes in the Alameda County study. Int J Epidemiol. 2005;34(6):1274‐1281. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0016] 16. Loucks EB, Magnusson KT, Cook S, Rehkopf DH, Ford ES, Berkman LF. Socioeconomic position and the metabolic syndrome in early, middle, and late life: evidence from NHANES 1999‐2002. Ann Epidemiol. 2007;17(10):782‐790. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0017] 17. Lange T, Rasmussen M, Thygesen LC. Assessing natural direct and indirect effects through multiple pathways. Am J Epidemiol. 2014;179(4):513‐518. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0018] 18. Chung GK, Lai FTT, Yeoh EK, Chung RY. Gender‐specific trends of educational inequality in diagnosed diabetes from 1999 to 2014 in Hong Kong: a serial cross‐sectional study of 97,481 community‐dwelling Chinese adults. Popul Health Metr. 2021;19(1):37. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0019] 19. Iakunchykova O, Averina M, Wilsgaard T, et al. What factors explain the much higher diabetes prevalence in Russia compared with Norway? Major sex differences in the contribution of adiposity. BMJ Open Diabetes Res Care. 2021;9(1):e002021. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0020] 20. Wu H, Bragg F, Yang L, et al. Sex differences in the association between socioeconomic status and diabetes prevalence and incidence in China: cross‐sectional and prospective studies of 0.5 million adults. Diabetologia. 2019;62(8):1420‐1429. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0021] 21. Bi Y, Lu J, Wang W, et al. Cohort profile: risk evaluation of cancers in Chinese diabetic individuals: a longitudinal (REACTION) study. J Diabetes. 2014;6(2):147‐157. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0022] 22. Kroenke K, Spitzer RL, Williams JB. The PHQ‐9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606‐613. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0023] 23. Meng D, Poon‐Mcbrayer KF, Farnsworth EB. The development of special education in China: A sociocultural review. Remedial Spec Educ. 2001;22(5):288‐298. [Google Scholar]

[jdb13325-bib-0024] 24. Hu C, Zhang Y, Lin L, et al. Gestational hyperglycemia and the risk of cardiovascular diseases among elderly Chinese women: findings from the REACTION study. J Diabetes. 2021;13(12):949‐959. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0025] 25. American Diabetes Association . Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(Suppl 1):S62‐S69. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0026] 26. Pan WH, Yeh WT. How to define obesity? Evidence‐based multiple action points for public awareness, screening, and treatment: an extension of Asian‐Pacific recommendations. Asia Pac J Clin Nutr. 2008;17(3):370‐374. [PubMed] [Google Scholar]

[jdb13325-bib-0027] 27. Zheng R, Niu J, Wu S, et al. Gender and age differences in the association between sleep characteristics and fasting glucose levels in Chinese adults. Diabetes Metab. 2021;47(2):101174. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0028] 28. Akter S, Goto A, Mizoue T. Smoking and the risk of type 2 diabetes in Japan: A systematic review and meta‐analysis. J Epidemiol. 2017;27(12):553‐561. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0029] 29. U.S. Department of Health and Human Services . 2008. Physical Activity Guidelines for Americans.www.health.gov/paguidelines.

[jdb13325-bib-0030] 30. Lloyd‐Jones DM, Hong Y, Labarthe D, et al. Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association's strategic impact goal through 2020 and beyond. Circulation. 2010;121(4):586‐613. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0031] 31. Jiang H, Sun Z, Guo H, et al. An assessment of urbanization sustainability in China between 1990 and 2015 using land use efficiency indicators. npj Urban Sustain. 2021;1(1):34. [Google Scholar]

[jdb13325-bib-0032] 32. Baron RM, Kenny DA. The moderator‐mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173‐1182. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0033] 33. MacKinnon DP, Valente MJ. Mediation from multilevel to structural equation modeling. Ann Nutr Metab. 2014;65(2–3):198‐204. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0034] 34. Rosseel Y. The lavvan tutorial.

[jdb13325-bib-0035] 35. Darin‐Mattsson A, Fors S, Kåreholt I. Different indicators of socioeconomic status and their relative importance as determinants of health in old age. Int J Equity Health. 2017;16(1):173. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0036] 36. Mathisen J, Jensen AKG, Andersen I, Andersen GS, Hvidtfeldt UA, Rod NH. Education and incident type 2 diabetes: quantifying the impact of differential exposure and susceptibility to being overweight or obese. Diabetologia. 2020;63(9):1764‐1774. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0037] 37. Oshio T, Kan M. Educational level as a predictor of the incidences of non‐communicable diseases among middle‐aged Japanese: a hazards‐model analysis. BMC Public Health. 2019;19(1):852. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0038] 38. Espelt A, Borrell C, Roskam AJ, et al. Socioeconomic inequalities in diabetes mellitus across Europe at the beginning of the 21st century. Diabetologia. 2008;51(11):1971‐1979. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0039] 39. Xu F, Yin XM, Zhang M, Leslie E, Ware R, Owen N. Family average income and diagnosed type 2 diabetes in urban and rural residents in regional mainland China. Diabet Med. 2006;23(11):1239‐1246. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0040] 40. Ramachandran A, Snehalatha C, Kapur A, et al. High prevalence of diabetes and impaired glucose tolerance in India: National Urban Diabetes Survey. Diabetologia. 2001;44(9):1094‐1101. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0041] 41. Shang X, Li J, Tao Q, et al. Educational level, obesity and incidence of diabetes among Chinese adult men and women aged 18–59 years old: an 11‐year follow‐up study. PLoS One. 2013;8(6):e66479. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0042] 42. Park SJ, Kang HT, Nam CM, Park BJ, Linton JA, Lee YJ. Sex differences in the relationship between socioeconomic status and metabolic syndrome: the Korean National Health and nutrition examination survey. Diabetes Res Clin Pract. 2012;96(3):400‐406. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0043] 43. Imkampe AK, Gulliford MC. Increasing socio‐economic inequality in type 2 diabetes prevalence‐‐repeated cross‐sectional surveys in England 1994–2006. Eur J Public Health. 2011;21(4):484‐490. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0044] 44. Aekplakorn W, Hogan MC, Chongsuvivatwong V, et al. Trends in obesity and associations with education and urban or rural residence in Thailand. Obesity (Silver Spring). 2007;15(12):3113‐3121. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0045] 45. Kumari M, Head J, Marmot M. Prospective study of social and other risk factors for incidence of type 2 diabetes in the Whitehall II study. Arch Intern Med. 2004;164(17):1873‐1880. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0046] 46. Wikström K, Lindström J, Tuomilehto J, et al. Socio‐economic differences in dysglycemia and lifestyle‐related risk factors in the Finnish middle‐aged population. Eur J Public Health. 2010;21(6):768‐774. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0047] 47. Demakakos P, Marmot M, Steptoe A. Socioeconomic position and the incidence of type 2 diabetes: the ELSA study. Eur J Epidemiol. 2012;27(5):367‐378. [DOI] [PubMed] [Google Scholar]

[jdb13325-bib-0048] 48. Härkönen J, Lindberg M, Karlsson L, Karlsson H, Scheinin NM. Education is the strongest socio‐economic predictor of smoking in pregnancy. Addiction. 2018;113(6):1117‐1126. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jdb13325-bib-0049] 49. Kim SR, Han K, Choi JY, et al. Age‐ and sex‐specific relationships between household income, education, and diabetes mellitus in Korean adults: the Korea National Health and nutrition examination survey, 2008‐2010. PLoS One. 2015;10(1):e0117034. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Obesity mediates the opposite association of education and diabetes in Chinese men and women: Results from the REACTION study

在中国男性和女性中肥胖介导了相反的教育对糖尿病关联:REACTION研究的结果

Yuanyue Zhu

Chunyan Hu

Lin Lin

Shuangyuan Wang

Hong Lin

Yanan Huo

Qin Wan

Yingfen Qin

Ruying Hu

Lixin Shi

Qing Su

Xuefeng Yu

Li Yan

Guijun Qin

Xulei Tang

Gang Chen

Min Xu

Yu Xu

Tiange Wang

Zhiyun Zhao

Zhengnan Gao

Guixia Wang

Feixia Shen

Zuojie Luo

Li Chen

Qiang Li

Zhen Ye

Yinfei Zhang

Chao Liu

Youmin Wang

Shengli Wu

Tao Yang

Huacong Deng

Lulu Chen

Tianshu Zeng

Jiajun Zhao

Yiming Mu

Weiqing Wang

Guang Ning

Yufang Bi

Yuhong Chen

Jieli Lu

Abstract

Background

Methods

Results

Conclusions

摘要

目的

材料和方法

结果

结论

1. INTRODUCTION

2. METHODS

2.1. Study population

2.2. Data collection and clinical evaluation

2.3. Definition of exposure, outcome, and covariates

2.3.1. Education levels

2.3.2. Diabetes

2.3.3. Covariates

2.4. Statistical analysis

TABLE 2.

3. RESULTS

TABLE 1.

FIGURE 1.

TABLE 3.

4. DISCUSSION

DISCLOSURE

Supporting information

ACKNOWLEDGEMENTS

Contributor Information

REFERENCES

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES