diabetes research and clinical practice 107 (2015) 424432

Contents available at ScienceDirect

Diabetes Research
and Clinical Practice
jou rnal hom ep ag e: w ww.e l s e v i er . c om/ loca te / d i ab r es

Type 2 diabetes mellitus incidence in Chinese:

Contributions of overweight and obesity
Chao Wang a, Jianxin Li a, Haifeng Xue a,b, Ying Li a,
Jianfeng Huang a, Jingzhuang Mai c, Jichun Chen a, Jie Cao a,
Xianping Wu d, Dongshuang Guo e, Ling Yu f, Dongfeng Gu a,*
a
Department of Epidemiology, Fuwai Hospital, National Center for Cardiovascular Diseases, China;
Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
b
Department of Food and Environment, School of Public Health, Qiqihar Medical University, Qiqihar, China
c
Guangdong Provincial Peoples Hospital and Cardiovascular Institute, Guangzhou, China
d
Sichuan Centre for Disease Control and Prevention, Chengdu, China
e
Yuxian Peoples Hospital, Yuxian, China
f
Fujian Provincial Peoples Hospital, Fuzhou, China

article info

abstract

Article history:

Aims: To estimate the incidence of Type 2 diabetes mellitus (T2DM) and the number of those

Received 10 March 2014

with T2DM attributable to overweight and obesity in China.

Received in revised form

Methods: We conducted a prospective cohort study among 15 680 participants (46.4%, men)

20 June 2014

aged 3574 years. The mean duration of follow-up was 8.0 years. We examined the

Accepted 14 September 2014

relationship between overweight, obesity and risk of T2DM by Cox proportional hazards

Available online 8 October 2014

models. Population attributable risk (PAR) of overweight and obesity was also calculated.

Keywords:

using PAR, incidence of T2DM and the population size of China in 2010.

Overweight

Results: During a mean follow-up of 8.0 years, the age-standardized incidence of T2DM was

Moreover, we estimated the number of T2DM events attributed to overweight and obesity

Obesity body mass index

9.5 per 1000 person-years in men and 9.2 in women. Overweight accounted for 28.3% (95%

Type 2 diabetes mellitus

confidence interval [CI]: 20.1, 36.2) of incident T2DM among men and 31.3% (95% CI: 25.5,

Population attributable risk

36.9) among women. The corresponding PAR of obesity was 10.1% (95% CI: 6.0, 14.2) among
men and 16.8% (95% CI: 12.0, 21.6) among women. Approximately 3.32 million (95% CI: 2.47,
4.24) incident T2DM were attributable to overweight and obesity in Chinese adults who were
35 to 74 years in 2010.
Conclusion: Our results indicate that incident T2DM is mainly attributable to overweight and
obesity in China. It is extremely important to advocate healthy lifestyle and prevent
excessive weight gain for reducing T2DM burden in China.
# 2015 Published by Elsevier Ireland Ltd.

* Corresponding author at: Tel.: +86 10 68331752; fax: +86 10 88363812.

E-mail address: gudongfeng@vip.sina.com (D. Gu).
http://dx.doi.org/10.1016/j.diabres.2014.09.059
0168-8227/# 2015 Published by Elsevier Ireland Ltd.

diabetes research and clinical practice 107 (2015) 424432

1.

Introduction

Type 2 diabetes mellitus (T2DM) is a global health problem that

threatens almost all nations in the world. The International
Diabetes Federation estimated the total number of people with
diabetes was 381.8 million in 2013 and is expected to rise to
591.9 million by 2035 [1]. China is the worlds most populous
country and the largest developing country, which has
enjoyed impressive economic developments over the past
two decades. However, the prevalence of diabetes has been
increasing sharply with rapid economic development, nutrition transition, changing lifestyle and so on [25]. In 1994, the
prevalence of diabetes was 2.5%, almost threefold increase
compared with 1980 (0.9%) [6]. Yang and colleagues reported
that the prevalence of total diabetes was 9.7% in the Chinese
population in 2007 [7]. The most recent national survey in 2010
reported that the rate of diabetes was elevated to 11.6%,
representing an estimated 113.9 million adults with diabetes
in China [8]. These numbers indicate that China has overtaken
India as the global epicenter of the diabetes epidemic. T2DM is
a huge challenge to our nations health and economy [2].
Notably, one important factor that contributes to the rapid
growth of diabetes in China and other developing countries is
the increasing prevalence of overweight and obesity [3,9]. The
prevalence of overweight and obesity increased significantly
among Chinese nearly over the last two decades [1014]. In
2010, the prevalence of overweight and obesity (body mass
index, BMI cut off points: 25.029.9 kg/m2 for overweight,
30 kg/m2 for obesity) in Chinese adults aged over 18 years old
was 27.9% and 5.1%, respectively [14].
Nowadays, several prospective studies have examined the
relationship between overweight/obesity and incidence of
T2DM [15,16]. However, few cohort studies have been
conducted, and the incidence of overweight and obesityattributable T2DM has not been rigorously estimated in the
general population in China or other developing countries. In
the present report, we aim to investigate the association
between overweight, obesity and incidence of T2DM among
Chinese adults, and estimate the population attributable risks
(PARs) and incidence of T2DM attributable to overweight and
obesity in China in 2010, using data from a prospective cohort
study of Chinese adults. These estimates could aid the
prevention and control of T2DM in China.

425

cross-sectional study of cardiovascular disease (CVD) risk

factors during 20002001. The study used a 4-stage stratified
sampling method to select a nationally representative sample
of the general population in China. Details of the study design
and methods of ChinaMUCA [17,18] and China Cardiovascular
Health Study [19] have been described elsewhere. At baseline,
a total of 27 020 participants were included in the two studies.

2.2.

Baseline examinations

2.

Methods

During baseline examinations, a standard questionnaire

assessing demographic characteristics, education level, lifestyle such as smoking, drinking and work-related physical
activity, medical history and other risk factors was administered by trained research staff. High-school graduation was
defined as having had 12 or more years of schooling. Cigarette
smoking was defined as having smoked at least 400 cigarettes
or 500 g tobacco leaves in a lifetime or 1 cigarette per day for 1
year or more [20]. Alcohol consumption was defined as
drinking alcohol at least once a week during the last year.
Work-related physical activity was assessed on the basis of
subjects occupation [21]. Participants completed a comprehensive health examination which included evaluation of
anthropometric indexes, blood pressure and collection of
biological specimens for assessment after the interview. All
study investigators and staff members were trained and
certified. Body weight and height were measured to the
nearest 0.5 kg and 0.5 cm, respectively, with the participant
wearing lightweight clothing and no shoes. Blood pressure
was measured three times for all participants in sitting
position after resting for 5 min according to a standard
protocol. BMI was calculated as weight (kg)/height squared
(m2). Hypertension was defined as an average systolic blood
pressure (SBP) 140 mmHg and/or an average diastolic blood
pressure (DBP) 90 mmHg and/or use of antihypertensive
medication within the past 2 weeks.
Overnight fasting blood samples were collected for blood
glucose, total cholesterol (TC), triglyceride (TG) and high
density lipoprotein-cholesterol (HDL-C). Diabetes was defined
as a fasting glucose concentration 7.0 mmol/L or the use of
insulin or oral hypoglycemic agents, and/or a self-reported
history of T2DM. Impaired fasting glucose (IFG) was defined as
a fasting glucose concentration 6.1 mmol/L and <7.0 mmol/L
without diabetes [22]. Dyslipidemia was defined as TC
6.22 mmol/L, TG  2.26 mmol/L or HDL-C < l.04 mmol/L in
terms of criteria recommended by U.S. Adult Treatment Panel
III [23,24].

2.1.

Study population

2.3.

All study participants came from two prospective cohort

studies: China Multicenter Collaborative Study of Cardiovascular Epidemiology (ChinaMUCA) and China Cardiovascular
Health Study. ChinaMUCA started in 1998, included 15 clusters
which were selected on the basis of the main characteristics of
the population in terms of geographical locations, socioeconomic status and dietary patterns. Approximately 1000
subjects with 50% women were sampled in each cluster. Of
the 15 clusters, 11 accepted the invitation to participate in the
follow-up study. China Cardiovascular Health Study was a

Follow-up data collection

The follow-up examination was conducted in 2007 and 2008.

Lifestyles and other risk factors were assessed with a standard
questionnaire. Fasting glucose and lipids levels and presence
of T2DM and IFG were determined as per baseline. A total of
21 556 individuals were followed up, with an overall response
rate of 79.8%. Of these, participants who had a history of
myocardial infarction/stroke (n = 325) or who had diabetes
(n = 1180) at baseline were excluded. Participants with incomplete or invalid data on baseline diabetes status (n = 512) or BMI
(n = 41) were also excluded.

426

diabetes research and clinical practice 107 (2015) 424432

After excluding those (n = 3818) without complete followup data for diabetes ascertainment, 15 680 participants (7276
men and 8404 women) were retained for this analysis with
follow-up data of 125 267 person-years. The mean duration of
follow-up was 8.0 years. The baseline characteristics of
participants included in this analysis were similar to those
who were lost to follow-up. Among participants who underwent follow-up examination and those who lost to follow-up,
the percentages of female were 51.9% and 50.5%, respectively;
the mean baseline ages were 48.5 and 50.0 years; the mean BMI
were 23.5 and 23.7 kg/m2; the percentages of participants who
smoked, 37.2% and 37.9%; who consumed alcohol, 25.2% and
24.6%. The study was approved by Fuwai Hospital Ethics
Committee. All study participants gave written informed
consent.

2.4.

Furthermore, the PARs and 95% CIs for overweight or

obesity (http://www.hsph.harvard.edu/faculty/spiegelman/
par.html) were estimated through the method described by
Spiegelman and colleagues [26,29,30]. The numbers of incident
T2DM attributable to overweight or obesity in each subgroup
were then calculated by multiplying the PARs according to the
subgroup-specific incidence rates of T2DM from the study and
population size in China in 2010. The total numbers of
overweight or obesity-attributable T2DM in men and women
were calculated by summing these estimates across appropriate subgroups. Statistical analyses were conducted with the
use of SAS 9.1.3 (SAS Institute, Cary, NC). A P value <0.05 was
considered statistically significant.

3.

Results

Statistical analysis

Overweight and obesity is defined as a BMI of 25.029.9 kg/m2

and as a BMI of 30.0 kg/m2 [25,26], respectively. Person-years of
follow-up were calculated from the date of baseline questionnaire interview to the date of outcome events or follow-up
interview for each participant, whichever came first. Agestandardized incidence was calculated with the use of the 10year age-specific incidence of T2DM and age distribution of the
Chinese population from year 2010 census data.
Baseline characteristics for different BMI categories were
compared using variance analysis or KruskalWallis H test for
continuous variables and chi-square test for categorical
variables. To examine the effects of BMI on T2DM among
men and women, hazard ratios (HRs) with 95% confidence
intervals (CIs) were calculated by Cox regression models. Age
was used as the primary time variable, with entry time defined
as the subjects age at the baseline and exit time as age at
diagnosis of T2DM, censoring, or death (whichever came first).
Two different regression models were applied: Model 1:
adjusted for cigarette smoking, alcohol drinking, geographic
region (north vs. south), urbanization (urban vs. rural), workrelated physical activity, education level and family history of
T2DM; Model 2: added the presence or absence of IFG,
hypertension and dyslipidemia to Model 1. The presence of
a linear trend was tested using the medians of BMI for each
group treated as a continuous variable. Moreover, we carried
out a series of sensitivity analyses for primary outcome. We
analyzed the association between BMI and incidence risk of
T2DM among participants without IFG and dyslipidemia at
baseline, respectively. Additionally, during the first three
years follow-up, 78 participants (38 men, 40 women) developed T2DM. After excluding the participants with T2DM at the
first three years of follow up, we investigated the association
between BMI and incidence risk of T2DM in the 4th to 8th years
of follow-up among participants. We also performed an
analysis when additional adjustment for abdominal obesity
(using the WC criterion, a WC  90 cm in men or WC  80 cm
in women was classified as abdominal obesity) was included
in the multivariable model [25]. The multivariable-adjusted
HRs of incident T2DM associated with overweight and obesity
compared with reference group were calculated for men and
women, and age subgroups with Cox proportional hazards
models [27,28].

Baseline characteristics for the different BMI categories were

displayed in Table 1. The mean age of the 15 680 participants
was 47.9 years, and 46.4% of participants were men. Prevalence of overweight and obesity were 26.6% and 3.2% for men
and 26.7% and 4.6% for women, respectively. In general, both
overweight or obesity in men and women was more prevalent
in northern China, as was a positive family history of diabetes,
higher SBP, DBP, fasting glucose, TC, TG, but lower HDL-C. This
group was also less likely to be physically active. Women with
overweight or obesity were older and more likely to have rural
residence. Furthermore, men with an elevated BMI were more
likely to have a higher education level but less likely to be
smokers than men with a BMI <25.0 kg/m2. Nonetheless,
women with an elevated BMI were less likely to have a higher
education level but more likely to be smokers than their
counterparts with a BMI < 25.0 kg/m2.
During a mean follow-up of 8.0 years, 1099 participants (536
men, 563 women) developed T2DM. Age-standardized incidence of T2DM was 9.5 and 9.2 per 1000 person-years in men
and women, respectively. Table 2 showed the incidence (per
1000 person-years) of T2DM increased with BMI category (6.0,
16.4, 33.2 among men, 4.9, 16.5, 31.4 among women with a BMI
of <25.0, 25.0 to 29.9, and 30.0 kg/m2, respectively). Compared with the men with BMI < 25.0 kg/m2, the multivariateadjusted HRs and 95% CIs (Model 2) of incident T2DM was 2.50
(2.04, 3.07) for overweight men and 4.59 (3.28, 6.44) for obesity
men. The corresponding HRs (95% CIs) were 2.54 (2.10, 3.09),
and 4.51 (3.44, 5.91) for incident T2DM among women. There
was a significant, doseresponse association between BMI and
incidence of T2DM in both men and women after adjustment
for multiple risk factors (P < 0.001 for trend) (Table 2). The
findings from each of our sensitivity analyses are presented in
Fig. 1. The results of the primary study were robust to the
sensitivity analysis.
In addition, the multivariable-adjusted HRs for incident
T2DM were significantly higher in overweight and obesity
than in BMI < 25.0 kg/m2 for men and women in all three age
groups (35 to 49, 50 to 59, and 60 years) (Table 3). We
calculated the PARs and absolute number for subjects in all
age groups. We found the prevalence of overweight and
obesity, and the PAR were higher for women than for men.
Overweight accounted for 28.3% (95% CI: 20.1, 36.2) of
incident T2DM among men and 31.3% (95% CI: 25.5, 36.9)

Characteristics

Number of participants
Age (years)
BMI (kg/m2)
SBP (mm Hg)
DBP (mm Hg)
Fasting glucose (mmol/L)
TC (mmol/L)
TG (mmol/L)
HDL-C (mmol/L)
Moderate/heavy PA
High school education
North China
Urban residents
Alcohol consumption
Cigarette smoking
Family history of T2DM
Impairing fasting glucose
Hypertension
Dyslipidemia

P*

Men
<25.0 kg/m 2

25.029.9 kg/m 2

30.0 kg/m 2

5109 (70.2)
48.4 (8.9)
21.6 (2.0)
120.7 (17.3)
77.4 (10.6)
5.01 (0.70)
4.76 (0.93)
1.12 (0.821.55)
1.38 (0.38)
2665 (52.3)
1608 (31.7)
2094 (41.0)
1924 (37.7)
2372 (46.5)
3735 (73.2)
230 (4.5)
307 (6.0)
938 (18.4)
1325 (26.0)

1938 (26.6)
48.3 (8.8)
26.9 (1.3)
129.5 (18.0)
84.8 (10.9)
5.20 (0.70)
4.93 (0.93)
1.58 (1.152.31)
1.16 (0.30)
599 (31.1)
830 (43.8)
1253 (64.7)
1053 (54.3)
925 (47.9)
1311 (67.8)
153 (7.9)
195 (10.1)
794 (41.0)
1001 (51.9)

229 (3.2)
48.3 (9.2)
31.9 (1.8)
137.7 (20.1)
90.7 (12.0)
5.25 (0.70)
5.06 (0.94)
1.65 (1.202.42)
1.13 (0.26)
63 (27.6)
87 (38.7)
188 (82.1)
122 (53.3)
120 (52.9)
162 (70.7)
23 (10.0)
25 (10.9)
150 (65.5)
119 (52.4)

0.892
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.119
<0.001
<0.001
<0.001
<0.001
<0.001

P*

Women
<25.0 kg/m 2

25.029.9 kg/m 2

30.0 kg/m 2

5774 (68.7)
46.8 (8.6)
21.7 (2.0)
117.9 (18.6)
74.4 (10.1)
5.01 (0.65)
4.74 (0.97)
1.08 (0.821.50)
1.43 (0.34)
2249 (39.1)
1578 (27.7)
2309 (40.0)
2329 (40.3)
315 (5.5)
237 (4.1)
307 (5.3)
296 (5.1)
854 (14.8)
1211 (21.0)

2246 (26.7)
48.6 (8.5)
26.9 (1.3)
127.4 (20.5)
80.6 (10.6)
5.16 (0.69)
4.94 (0.95)
1.39 (1.011.93)
1.29 (0.30)
572 (25.5)
586 (26.4)
1360 (60.6)
1039 (46.3)
122 (5.4)
97 (4.3)
191 (8.5)
184 (8.2)
742 (33.0)
788 (35.1)

384 (4.6)
49.8 (8.4)
32.0 (1.9)
134.4 (20.1)
84.6 (10.7)
5.27 (0.69)
5.00 (1.01)
1.55 (1.152.07)
1.23 (0.28)
89 (23.3)
72 (19.1)
277 (72.1)
159 (41.4)
16 (4.2)
29 (7.6)
33(8.6)
44 (11.5)
191 (49.7)
162 (42.3)

<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.552
0.006
<0.001
<0.001
<0.001
<0.001

Continuous and categorical variables express as means (standard deviations) and number (percent), respectively. Triglyceride was shown as median value and interquartile range.
Reflecting overall group analysis of variance, chi square or KruskalWallis H test. BMI, body mass index; PA, physical activity;T2DM, type 2 diabetes mellitus; SBP, systolic blood pressure; DBP,
diastolic blood pressure; HDL-C, high density lipoprotein cholesterol; TG, triglycerides; TC, total cholesterol.
*

diabetes research and clinical practice 107 (2015) 424432

Table 1 Baseline characteristics according to different categories of BMI among Chinese adults aged 3574 years.

427

428

diabetes research and clinical practice 107 (2015) 424432

Table 2 Hazard ratio and age-standardized incidence for T2DM in different BMI groups of participants.
Variable

Total

Body

Mass

<25.0 kg/m 2
Men
No. of T2DM
Person-years of follow-up
Age-standardized rate per 1000 person-years
Model 1y
Model 2z
Women
No. of T2DM
Person-years of follow-up
Age-standardized rate per 1000 person-years
Model 1y
Model 2z

536
58059
9.5

563
67208
9.2

Index

P value for
linear trend

25.029.9 kg/m 2

30.0 kg/m 2

235
41213
6.0
1.00
1.00

246
15135
16.4
2.94 (2.423.56)
2.50 (2.043.07)

55
1711
33.2
5.62 (4.097.73)
4.59 (3.286.44)

<0.001
<0.001

207
46633
4.9
1.00
1.00

267
17675
16.5
2.93(2.423.54)
2.54 (2.103.09)

89
2900
31.4
5.76 (4.437.48)
4.51 (3.445.91)

<0.001
<0.001

y
Model 1 was adjusted cigarette smoking, alcohol drinking, geographic region (north vs. south), urbanization (urban vs. rural), work-related
physical activity, educational level and family history of T2DM.
z
Based on Model 1, Model 2 was adjusted for impaired fasting glucose, hypertension and dyslipidemia. BMI, body mass index; T2DM, type 2
diabetes mellitus.

among women, separately. The corresponding PAR of obesity

was 10.1% (95% CI: 6.0, 14.2) among men and 16.8% (95% CI:
12.0, 21.6) among women. It was estimated that a total of
3321 000 (95% CI: 2469 400, 4237 900) incident T2DM were
attributable to overweight and obesity in Chinese adults who
were 35 to 74 years in 2010: 1126 600 (95% CI: 800 100, 1441 000)
in men and 1193 900 (95% CI: 972 700, 1407 600) in women were
attributable to overweight, 452 800 (95% CI: 238 800, 565 300) in
men and 547 700 (95% CI: 457 800, 824 000) in women were
attributable to obesity.

4.

Discussion

In this study, we observe that overweight and obesity are

significantly associated with increased risk of incident T2DM
in Chinese men and women. The percentage of T2DM events
attributed to overweight is 28.3% in men and 31.3% in women,

and attributed to obesity is 10.1% among men and 16.8%

among women. We estimate that nearly 3.32 million incident
T2DM cases were attributable to overweight and obesity in
Chinese adults aged 3574 years in 2010. To our knowledge,
this is the first study that estimates the proportion and
number of overweight and obesity-attributable T2DM events
in Chinese population by a prospective cohort study. Our
findings clearly illustrate how important prevention of excess
weight is for reducing overweight and obesity-related incident
T2DM in China.
A number of other studies from developed countries have
provided strong evidence of an association between overweight/obesity and risk of diabetes [15,31,32]. The Nurses
Health Study had published several analyses on the association between BMI and T2DM, and have shown a major role of
excess weight in the development of T2DM [31,32]. Of note,
Stevens et al. indicated a 1.7% point increase in incidence of
diabetes is associated with a one-unit increase in BMI among

Fig. 1 Multivariate-adjusted hazard ratio of incident T2DM according to the BMI categories among participants by sex:
results of sensitivity analyses. (A) participants without impaired fasting glucose at baseline (B) participants without
dyslipidemia at baseline (C) participants without T2DM at the first three years of follow up (D) additional adjustment for
abdominal obesity in the multivariable model.

Age group
(years)

Prevalence of
overweight (%)

Hazard risk
(95% CI)y

Population
attributable
risk (%)z

Absolute number of
incident T2DM (103)*

Prevalence of
obesity (%)

Hazard risk
(95% CI)y

Population
attributable
risk (%)z

Absolute number
of incident
T2DM (103)*

Men
3549
5059
60

27.0
26.3
25.7

2.17 (1.642.89)
3.12(2.214.41)
2.61 (1.464.70)

24.0 (13.134.4)
34.6 (19.148.4)
29.5 (3.152.1)

441.2 (240.8632.4)
333.8 (184.2466.9)
351.6 (36.9620.9)

3.1
3.0
3.9

5.25(3.308.35)
5.88(3.1311.04)
4.25 (1.7310.45)

10.7 (4.117.2)
7.4 (2.512.2)
15.5 (3.027.6)

196.7 (75.4316.2)
71.4 (24.1117.7)
184.7 (35.8328.9)

Total

26.6

2.50 (2.043.07)

28.3 (20.136.2)

1126.6 (800.11441.0)

3.2

4.59 (3.286.44)

10.1 (6.014.2)

452.8 (238.8565.3)

Women
3549
5059
60

24.6
30.7
30.4

2.49 (1.883.29)
2.21 (1.623.01)
2.81 (1.465.42)

31.0 (24.936.8)
26.5 (15.736.7)
47.0 (29.361.6)

440.2(353.6522.6)
316.3 (187.4438.1)
437.4(272.7573.3)

3.7
6.0
6.5

4.72 (3.206.95)
4.45(2.946.73)
3.20 (1.158.91)

17.8 (9.625.7)
18.7 (11.725.6)
7.7 (0.016.9)

252.8(136.3365.0)
223.2 (139.7305.6)
71.7 (0.0157.3)

Total

26.7

2.54 (2.103.09)

31.3 (25.536.9)

1193.9 (972.71407.6)

4.6

4.51 (3.445.91)

16.8 (12.021.6)

547.7 (457.8824.0)

CI, confidence interval.

y
Adjusted for cigarette smoking, alcohol drinking, geographic region (north vs. south), urbanization (urban vs. rural), work-related physical activity, educational status, family history of T2DM,
impaired fasting glucose, hypertension and dyslipidemia.
z
Adjusted for age, calendar year of follow-up, cigarette smoking, alcohol drinking, geographic region (north vs. south), urbanization (urban vs. rural), work-related physical activity, educational level,
family history of T2DM, impaired fasting glucose, hypertension and dyslipidemia.
*
Population numbers for men aged 3549, 5059, and 60 years were 177 776 200, 81 446 172, and 66 986 350, respectively. Population numbers for women aged 3549, 5059, and 60 years were
170 598 276, 78 619 473, and 65 766 611, respectively.

diabetes research and clinical practice 107 (2015) 424432

Table 3 Prevalence of overweight and obesity, hazard ratio, population attributable risk, and absolute number of T2DM events attributable to overweight or obesity in
China, 2010.

429

430

diabetes research and clinical practice 107 (2015) 424432

Chinese in a prospective study between 1983 and 1994 [33]. In

the present study, both overweight and obese men and
women were more likely to have T2DM, data consistent with
the results of the previous studies. It is known that adipocytes
(fat cells) secrete a lot of adipocyte hormones and adipokines,
which might increase the risk of diabetes via several pathways
such as increasing insulin resistance [34].
To date, only a few studies have reported estimates of the
risk of T2DM attributable to overweight and obesity [3538].
Laaksonen et al. reported that overweight/obesity was an
important predictor of T2DM (PAR = 77%) in Finnish cohorts.
They estimated that 77% of all new cases could have been
avoided, if everyones BMI had been less than 25.0 kg/m2 in
Finland [35]. In a longitudinal study of a Spanish population,
the population risk of diabetes attributable to excess weight
(BMI  25.0 kg/m2) was estimated as 46% for men and 61% for
women. If only obesity was targeted, 17% and31% of T2DM in
men and women could have been avoided [36]. Notably, the
PARs estimates (28.3% among men and 31.3% among women
attributable to overweight, and 10.1% among men and 16.8%
among women attributable to obesity) were lower in our
study. This might be explained by the higher prevalence of
overweight/obesity or relative risk of diabetes associated with
excess weight in these previous studies. The difference
between the studies might also be due to the length of
follow-up time, racial difference and the variables selected for
adjustment. Additionally, both the current and Spanish study
shown the PARs was higher in women than men. The similar
gender difference of PARs in different populations could be
explained by the more prevalent for overweight/obesity in
women. Weight gains in pregnancy and not returning to the
optimal weight might be causes of the higher prevalence of
overweight/obesity in women. Furthermore, gender differences in the level of physical activity and food intake may be
taken consideration [21,3842]. Several studies have shown
that Chinese men were more likely to be physically active than
women [21,39]. Chinas obesity epidemic might recently have
been driven by decreasing levels of physical activity [39]. Du
et al. also reported that lack of physical activity was associated
with larger risk of obesity in Chinese adults [40]. They found a
1-SD (14 metabolic equivalent task hours per day, MET-h/d)
greater physical activity was associated with a 0.15-unit (95%
CI: 0.14, 0.16) lower BMI (in kg/m2) and 0.48 (95% CI: 0.45, 0.50)
percentage points less body fat. Moreover, excessive energy
intake, particularly from energy-dense foods is often ascribed
as one of main causes that have led to obesity epidemic [42,43].
Healthy lifestyle habits, including physical activity and
healthy eating, can lower the risk of obese and its related
diseases. Besides difference in prevalence of overweight/
obesity, gender differences in HRs might also play a role in the
gender difference of PARs. In the Spanish study, compared
with the men with BMI <25.0 kg/m2, the multivariate-adjusted
HR of incident T2DM was 1.94 for overweight/obesity men,
lower than 2.83 among women [36]. The higher HR of T2DM
associated with overweight/obesity in women might partly
explain higher PAR of T2DM in women than men (46% and 61%
in men and women, respectively).
Moreover, Zhao et al. have shown that the PAR of diabetes
associated with overweight and obesity were 20.3% and 13.9%,
respectively, in China by a cross-sectional study [44]. Several

factors that might play a potential role in the discrepancy

between these studies are different age, various definitions of
overweight and obesity, and study design. For instance, Zhao
et al. used data from adults aged over 18 years old and
collected in 2002. Overweight and obesity were defined by a
BMI range of 24.027.9 kg/m2 and 28.0 kg/m2, respectively
[44]. Nevertheless, the present study was conducted among
participants aged 3574 years since 1998 and 20002001. We
defined overweight and obesity by a BMI range of 25.029.9 kg/
m2 and 30.0 kg/m2, respectively. Notably, in order to
calculate PAR, they estimated the association between BMI
and the risk for T2DM based on a cross-sectional survey. It is
not possible to conclude that overweight and obesity causes,
or even precedes, T2DM. We excluded participants with a
history of CVD or T2DM at baseline to avoid possible reverse
causation effects. Additionally, the prognostic factors are
likely to bias the effect of overweight and obesity. For example,
patients with diabetes might undergo treatment. A fundamental principle in the treatment of T2DM patients has been
the recommendation to lose weight by diet and lifestyle
intervention [45]. Furthermore, several oral agents might
cause weight loss in T2DM patients. Metformin is associated
with a lack of weight gain and even weight loss in some
overweight patients [46,47]. Thus patients who received
treatment and intervention might lose weight. Zhaos study
was thus likely to underestimate the relative risk of T2DM
associated with overweight and obesity compared with the
current study. By contrast, the effects from overweight and
obesity were not influenced by prognostic factors in our
prospective cohort study. Overall, the current study has shown
that the proportion of avoidable new cases would reach 10.1% in
men and 16.8% in women, if only obesity was targeted. An
estimated 38.4% of incident T2DM cases would be avoided in
men and 48.1% in women, if the BMI was below 25.0 kg/m2.
These results also indicated that an effective and preventive
measure should be aimed at decreasing the prevalence of both
overweight and obese individuals, and not to focus exclusively
on already obese individuals. Our findings have important
public health implications, since overweight/obesity can be
prevented and controlled by lifestyle.
The major strengths of the present study included enrolling
a population-based sample including both male and female
participants. Given the population-based prospective design,
our study enabled the impact of overweight and obesity to be
estimated in terms of their PAR, which can help policy makers
determine strategies for diabetes control in China. In addition,
the participants with a history of CVD or T2DM at baseline
were excluded to avoid possible reverse causation effects.
Furthermore, to ensure data validity and reliability, we
implemented strict quality assurance and quality control
program at every phase of the study.
There are also several study limitations. First, there was a
lack of collection of other potential risk factors such as the
dietary habits, leisure time physical activity and fitness level.
Second, many subjects were not included in our analysis due
to loss to follow-up (approximately 20%), which might
influence our estimates.
Nonetheless, this impact was limited because the baseline
characteristics of subjects who participated in follow-up were
similar to those who were lost to follow-up. Finally, we

diabetes research and clinical practice 107 (2015) 424432

ascertained the incidence of T2DM by fasting glucose

concentration or the use of insulin or oral hypoglycemic
agents and not by oral glucose tolerance test (OGTT), which
might underestimate the incidence of T2DM. However, fasting
glucose is recommended over the OGTT in epidemiological
studies for its convenience and acceptability among participants and also for its cost-effectiveness [48,49].
In conclusion, the study documented that overweight and
obesity are strong risk factors for T2DM. We estimated that in
2010, approximately 3.32 million incident T2DM cases could
have been prevented by eliminating overweight and obesity in
adults aged 3574 years in China. Preventing excessive weight
gain is crucial for reducing the T2DM burden in China.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Grant support
This study was supported by grants from the Ministry of
Science and Technology of the Peoples Republic of China
(2006BAI01A01, 2011BAI11B03, and 2011BAI09B03).

Acknowledgements
This work was supported by grants of the National Science &
Technology Pillar Program (2006BAI01A01, 2011BAI09B03 and
2011BAI11B03) from the Ministry of Science and Technology of
the Peoples Republic of China.

references

[1] Guariguata L, Whiting DR, Hambleton I, Beagley J,

Linnenkamp U, Shaw JE. Global estimates of diabetes
prevalence in adults for 2013 and projections for 2035 for
the IDF Diabetes Atlas. Diabetes Res Clin Pract
2014;103:13749.
[2] Hu FB. Globalization of diabetes: the role of diet, lifestyle,
and genes. Diabetes Care 2011;34:124957.
[3] Hossain P, Kawar B, El Nahas M. Obesity and diabetes in the
developing worlda growing challenge. N Engl J Med
2007;356:2135.
[4] Hu D, Sun L, Fu P, Xie J, Lu J, Zhou J, et al. Prevalence and
risk factors for type 2 diabetes mellitus in the Chinese adult
population: the InterASIA Study. Diabetes Res Clin Pract
2009;84:28895.
[5] Hu D, Fu P, Xie J, Chen CS, Yu D, Whelton PK, et al.
Increasing prevalence and low awareness, treatment and
control of diabetes mellitus among Chinese adults: the
InterASIA study. Diabetes Res Clin Pract 2008;81:2507.
[6] Pan XR, Yang WY, Li GW, Liu J. Prevalence of diabetes and
its risk factors in China, 1994. National Diabetes Prevention
and Control Cooperative Group. Diabetes Care
1997;20:16649.
[7] Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J, et al. Prevalence of
diabetes among men and women in China. N Engl J Med
2010;362:1090101.

431

[8] Xu Y, Wang L, He J, Bi Y, Li M, Wang T, et al. Prevalence

and control of diabetes in Chinese adults. JAMA
2013;310:94859.
[9] Finucane MM, Stevens GA, Cowan MJ, Danaei G, Lin JK,
Paciorek CJ, et al. National, regional, and global trends in
body-mass index since 1980: systematic analysis of health
examination surveys and epidemiological studies with 960
country-years and 9.1 million participants. Lancet
2011;377:55767.
[10] Wildman RP, Gu D, Muntner P, Wu X, Reynolds K, Duan X,
et al. Trends in overweight and obesity in Chinese adults:
between 1991 and 19992000. Obesity (Silver Spring)
2008;16:144853.
[11] Wang W, Wu ZS, Zhao D, Wu GX, Wang WH, Liu J, et al. The
trends of body mass index and overweight in population
aged 2564 in Beijing during 19841999. Zhonghua Liu Xing
Bing Xue Za Zhi 2003;24:2725.
[12] Li F, Fan JG, Cai XB. Study on the prevailing trend of
overweight and obesity in Shanghai Bao-Steel Company
based population from 1995 to 2002. Zhonghua Yu Fang Yi
Xue Za Zhi 2007;41:3841.
[13] Xi B, Liang Y, He T, Reilly KH, Hu Y, Wang Q, et al. Secular
trends in the prevalence of general and abdominal obesity
among Chinese adults, 19932009. Obes Rev 2012;13:28796.
[14] National Center for Chronic and Non-communicable
Disease Control and Prevention. Chinese Center for Disease
Control and Prevention. Report on chronic disease risk
factor surveillance in China 2010. Beijing: Military Medical
Science Press; 2012.
[15] Abdullah A, Peeters A, de Courten M, Stoelwinder J. The
magnitude of association between overweight and
obesity and the risk of diabetes: a meta-analysis of
prospective cohort studies. Diabetes Res Clin Pract
2010;89:30919.
[16] Wannamethee SG, Shaper AG, Walker M. Overweight and
obesity and weight change in middle aged men: impact on
cardiovascular disease and diabetes. J Epidemiol
Community Health 2005;59:1349.
[17] The Collaborative Study Group on Trends of Cardiovascular
Diseases in China and Preventive Strategy. Current status
of major cardiovascular risk factors in Chinese populations
and their trends in the past two decades. Chin J Cardiol
2001;29:749.
[18] Li JX, Fan S, Li Y, Chen J, Cao J, Huang J, et al. Incidence of
obesity and its modifiable risk factors in Chinese adults
aged 3574 years: a prospective cohort study. Zhonghua Liu
Xing Bing Xue Za Zhi 2014;35:34953.
[19] Gu D, Reynolds K, Wu X, Chen J, Duan X, Muntner P, et al.
Prevalence, awareness, treatment, and control of
hypertension in china. Hypertension 2002;40:9207.
[20] Chen Z, Li Y, Zhao LC, Zhou BF, Yang J, Wang ZW, et al. A
study on the association between tea consumption and
stroke. Zhonghua Liu Xing Bing Xue Za Zhi 2004;25:66670.
[21] Muntner P, Gu D, Wildman RP, Chen J, Qan W, Whelton PK,
et al. Prevalence of physical activity among Chinese adults:
results from the International Collaborative Study of
Cardiovascular Disease in Asia. Am J Public Health
2005;95:16316.
[22] Alberti KG, Zimmet PZ. Definition, diagnosis and
classification of diabetes mellitus and its complications.
Part 1: Diagnosis and classification of diabetes mellitus
provisional report of a WHO consultation. Diabet Med
1998;15:53953.
[23] Expert Panel on Detection, Evaluation, and Treatment of
High Blood Cholesterol in Adults. Executive summary of
the third report of The National Cholesterol Education
Program (NCEP) expert panel on detection, evaluation, and
treatment of high blood cholesterol in adults (adult
treatment panel III). JAMA 2001;285:248697.

432

diabetes research and clinical practice 107 (2015) 424432

[24] Vijayaraghavan K. Treatment of dyslipidemia in patients

with type 2 diabetes. Lipids Health Dis 2010;9:144.
[25] Gu D, Reynolds K, Wu X, Chen J, Duan X, Reynolds RF, et al.
Prevalence of the metabolic syndrome and overweight
among adults in China. Lancet 2005;365:1398405.
[26] Black MH, Sacks DA, Xiang AH, Lawrence JM. The relative
contribution of prepregnancy overweight and obesity,
gestational weight gain, and IADPSG-defined gestational
diabetes mellitus to fetal overgrowth. Diabetes Care
2013;36:5662.
[27] Gu D, Kelly TN, Wu X, Chen J, Samet JM, Huang JF, et al.
Mortality attributable to smoking in China. N Engl J Med
2009;360:1509.
[28] He J, Gu D, Chen J, Wu X, Kelly TN, Huang JF, et al.
Premature deaths attributable to blood pressure in China: a
prospective cohort study. Lancet 2009;374:176572.
[29] Hedderson M, Ehrlich S, Sridhar S, Darbinian J, Moore S,
Ferrara A. Racial/ethnic disparities in the prevalence of
gestational diabetes mellitus by BMI. Diabetes Care
2012;35:14928.
[30] Spiegelman D, Hertzmark E, Wand HC. Point and interval
estimates of partial population attributable risks in cohort
studies: examples and software. Cancer Causes Control
2007;18:5719.
[31] Carey VJ, Walters EE, Colditz GA, Solomon CG, Willett WC,
Rosner BA, et al. Body fat distribution and risk of noninsulin-dependent diabetes mellitus in women. The
Nurses Health Study. Am J Epidemiol 1997;145:6149.
[32] Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon
CG, et al. Diet, lifestyle, and the risk of type 2 diabetes
mellitus in women. N Engl J Med 2001;345:7907.
[33] Stevens J, Truesdale KP, Katz EG, Cai J. Impact of body mass
index on incident hypertension and diabetes in Chinese
Asians, American Whites, and American Blacks: the
Peoples Republic of China Study and the Atherosclerosis
Risk in Communities Study. Am J Epidemiol
2008;167:136574.
[34] Bray GA. Medical consequences of obesity. J Clin Endocrinol
Metab 2004;89:25839.
[35] Laaksonen MA, Knekt P, Rissanen H, Harkanen T, Virtala E,
Marniemi J, et al. The relative importance of modifiable
potential risk factors of type 2 diabetes: a meta-analysis of
two cohorts. Eur J Epidemiol 2010;25:11524.
[36] Huerta JM, Tormo MJ, Chirlaque MD, Gavrila D, Amiano P,
Arriola L, et al. Risk of type 2 diabetes according to
traditional and emerging anthropometric indices in Spain,
a Mediterranean country with high prevalence of obesity:

[37]

[38]

[39]

[40]

[41]

[42]
[43]

[44]

[45]
[46]

[47]

[48]

[49]

results from a large-scale prospective cohort study. BMC

Endocr Disord 2013;13:7.
Davin C, Vollenweider P, Waeber G, Paccaud F, MarquesVidal P, et al. Cardiovascular risk factors attributable to
obesity and overweight in Switzerland. Nutr Metab
Cardiovasc Dis 2012;22:9528.
Hosseinpanah F, Rambod M, Azizi F. Population
attributable risk for diabetes associated with excess weight
in Tehranian adults: a population-based cohort study. BMC
Public Health 2007;7:328.
Bell AC, Ge K, Popkin BM. Weight gain and its predictors in
Chinese adults. Int J Obes Relat Metab Disord
2001;25:107986.
Du H, Bennett D, Li L, Whitlock G, Guo Y, Collins R, et al.
Physical activity and sedentary leisure time and their
associations with BMI, waist circumference, and
percentage body fat in 0.5 million adults: the China
Kadoorie Biobank study. Am J Clin. Nutr 2013;97:48796.
Azizi F, Azadbakht L, Mirmiran P. Trends in overweight,
obesity and central fat accumulation among Tehranian
adults between 19981999 and 20012002: Tehran lipid and
glucose study. Ann Nutr Metab 2005;49:38.
Kanter R, Caballero B. Global gender disparities in obesity: a
review. Adv Nutr 2012;3:4918.
Wansink B, Cheney MM, Chan N. Exploring comfort food
preferences across age and gender. Physiol Behav
2003;79:73947.
Zhao W, Zhai Y, Hu J, Wang J, Yang Z, Kong L, et al.
Economic burden of obesity-related chronic diseases in
Mainland China. Obes Rev 2008;9(Suppl 1):627.
Pi-Sunyer FX. Weight loss in type 2 diabetic patients.
Diabetes Care 2005;28:15267.
Lee A, Morley JE. Metformin decreases food consumption
and induces weight loss in subjects with obesity with type
II non-insulin-dependent diabetes. Obes Res 1998;6:4753.
Stumvoll M, Nurjhan N, Perriello G, Dailey G, Gerich JE.
Metabolic effects of metformin in non-insulin-dependent
diabetes mellitus. N Engl J Med 1995;333:5504.
Gu D, Reynolds K, Duan X, Xin X, Chen J, Wu X, et al.
Prevalence of diabetes and impaired fasting glucose in the
Chinese adult population: International Collaborative
Study of Cardiovascular Disease in Asia (InterASIA).
Diabetologia 2003;46:11908.
Expert Committee on the Diagnosis and Classification of
Diabetes Mellitus. Report of the expert committee on the
diagnosis and classification of diabetes mellitus. Diabetes
Care 2002;26:S520.