Prevalence and risk factors of diabetes among adults aged 45 years or older in China: A national cross‐sectional study

Abstract Background Although there is preponderance of literature on disease burden of diabetes in developed countries, limited investigations have been conducted in less developed regions including China. This study aimed to explore the current prevalence and risk factors for diabetes, pre‐diabetes, awareness, treatment and control of diabetes in China. Methods We included 12,458 adults from the China Health and Retirement Longitudinal Study. We estimated prevalence of diabetes and pre‐diabetes in the overall sample and by socio‐demographics. Bivariate associations of diabetes, pre‐diabetes, awareness, control and treatment of diabetes with health and function measures were evaluated by chi‐squared test and multivariate logistic regression analysis. Results We found that the prevalence of diabetes and pre‐diabetes was 13.21% and 25.16%. The prevalence of diabetes increased with advanced age (12.37%, 15.98% and 16.52% among persons who were 45 to 55, 55 to 65 and ≥65 years old, respectively), educational background (14.52%, 15.52% and 15.58% among persons who were illiterate, had primary education and had secondary or above education) and weight (8.18%, 17.05% and 22.54% among persons with a body mass index of 18.5 to 24.9, 25.0 to 29.9 and ≥30.0, respectively). The prevalence of diabetes was higher among urban residents than among rural residents (19.04% vs. 12.85%). We also observed that aged between 55 and 65 years, obesity, history of hypertension and coronary heart disease, and inactivity were significant risk factors of awareness of diabetes. Conclusion Our results indicated that diabetes is high prevalent in adults aged 45 years or above in China. The potentially modifiable risk factors should be further studied to develop interventions and strategies aimed at prevention and treatment of diabetes among middle‐aged and older Chinese adults.


| INTRODUC TI ON
Diabetes is one of the fastest-growing chronic diseases in the world. In 2017, approximately 425 million adults worldwide had the condition, and about half of these cases had not been diagnosed. In China, the number of diabetic patients is estimated to increase from 20.8 million in 2000 to 42.3 million in 2030. 1 The increase is thought to be the most rapid worldwide, and it might be related to China's recent rapid economic development and urbanization, which are contributing to socioeconomic and epidemiological transition. Previous research has shown that non-communicable diseases are initially more common in population subgroups of high socioeconomic status (SES) and then, with increasing development, become more common in lower SES groups. 2 However, these evidences are less shown in developing countries.
Studies have indicated that the grim situation regarding diabetes could be mitigated through appropriate management and education.
However, diabetes awareness is low, especially in developing countries, and blood glucose control in particular remains a challenge for healthcare providers. In China, less than one-third (30.1%) of diabetic patients were aware of their disease condition, 3 and this may affect the estimate of diabetes prevalence. Research has found that adequate glycemic control can significantly decrease the risk of diabetes-related complications, causing a delay in disease progression. Basic information about the prevalence, treatment and diabetic educational condition should be analysed to provide evidence for further prevention.
Although previous studies have examined the prevalence of awareness, treatment and control of diabetes in different populations, the results were inconsistent due to the discrepancy in geographic regions and diagnosis criteria. Moreover, many researches were reported based on self-reports. In addition, diabetes is a major risk factor for cardiovascular disease, which has become the leading cause of death in China. IDF reported that over 10% of health costs in China could be related to diabetes. 4 It is known that social and economic factors such as income, educational level, occupation, levels of physical activity, being overweight or obese, health behaviours, living conditions and other demographic factors are strong influence factors of pre-diabetes and diabetes, 5,6 while the physiological or genetic factors linking diabetes and cardiovascular disease are understudied. Therefore, we used data from a nationally representative cohort of middle-aged and older adults in China to explore the prevalence and demographic as well as physiological risk factors of pre-diabetes, diabetes, awareness, control and treatment of diabetes. More attention should be given to implement preventive strategy and interventions to improve the overall management of diabetes among community-dwelling elders in China.

| Data source and study population
We use data collected from the survey of the China Health and Retirement Longitudinal Study (CHARLS), a national survey representative of the middle-aged and elderly population (45 years old and above) in China. CHARLS adopted a four-stage stratified cluster sampling to recruit participants. In the first stage, 150 county-level units from 28 provinces were selected to provide a mix of urban and rural settings, with a wide variation in the level of economic development. The second stage randomly chose 939 primary sampling units (PSUs) (470 villages and 469 communities) from the above countylevel units. All the dwellings in each selected PSU were outlined on Google Earth maps using the CHARLS-GIS software, which was specifically designed for the study. The third stage randomly selected 24 mapped households from each PSU. The last stage randomly selected 1 adult aged 45 years and older from each household. 7 Finally, a total of 21,097 respondents were successfully enrolled in 2015.
Of these respondents, cross-section blood samples of 13

| Definitions
In the present study, diabetes was defined as (1) a self-reported previous diagnosis by healthcare professionals (2) fasting plasma glucose (FPG) >126 mg/dl and/or HbA1c >6.5%. The cut-off points for diagnosis of diabetes were based on current recommendations from the American Diabetes Association. 8 Pre-diabetes was defined as 6.1 mmol/L<FPG <7.0 mmol/L. 9 Awareness was defined as the proportion of individuals with self-reported physician-diagnosed diabetes among all participants with diabetes. Treatment was defined as the percentage of diabetic patients who had taken diabetic medication. Control was characterized as the rate of participants with an HbA1c level under 7.0% among diabetic patients who were treated with diabetic medications.

| Blood sample collection and quality control
The key advantage in using data collected from CHARLS is that blood samples were measured in the survey. Nearly two-thirds blood samples of individuals were collected by medically trained staff from the China Center for Disease Prevention and Control (CDC). Participants were asked to fast overnight. After collection, plasma for glucose assay was separated from blood samples and stored at −20°C, and whole blood for HbA1C assay was stored immediately and during shipment at 4°C. All the blood samples were transported within 2 weeks to the CDC, where samples were placed at −80°C in a deep freezer before assay. Blood assays were performed at the Youanmen Center for Clinical Laboratory of Capital Medical University during February 2013 and June 2013. The laboratory used quality control samples daily during the testing of the CHARLS study samples, and all test results were within the target range (within two SDs of mean quality control concentrations). Glucose was measured using an enzymatic colorimetric test, and HbA1c was analysed using boronate affinity chromatography.

| Other data collection
We sent two interviewers to each county-level unit to interview about 72 households located in three communities. The interviewers were trained at Peking University by CHARLS staff members, and the interviews took place in respondents' homes with the use of CAPI technology. The interviewers who conducted the countylevel interviews described above also carried equipment for and conducted measurements of health functioning and performance in respondents' households. These included the anthropometric measurements of height, weight, waist circumference, lower right leg length and arm length, lung capacity, grip strength, speed of repeated chair stand, blood pressure, walking speed and balance tests. 7 Our study included age and gender as individual attributes. We used two variables of socioeconomic status: level of education and residence area. We classified level of education as illiterate, primary education or above, and secondary education or above. All medical conditions including hypertension, diabetes and cardiac disease were diagnosed or assessed by a physician or using a self-reported history of diagnosed condition. Disability was assessed by five activities of daily living (ADL) tasks (dressing, bathing, eating, getting out of bed and toileting). For each ADL task, participants were asked, 'Do you have difficulty in' performing the task? Those participants who responded, 'I have difficulty but can still do it', 'Yes, I have difficulty and need help' or 'I cannot do it' to one or more of the ADL tasks were considered having ADL disability.
Height and weight were measured and used to calculate body mass index (BMI: kg/m 2 ) using standard formula. The survey also collected information on body mass index (BMI), which is the ratio of weight in kilograms to height in metres squared. We use BMI to identify whether respondents are underweight (BMI < 18.5), normal (18.5 < BMI < 24), overweight (24 < BMI < 28) and obese (BMI ≥ 28)

| Statistical analysis
Age-and gender-standardized prevalence of diabetes was calculated by the direct method using population census data of China in 2010. Characteristics of the study participants were depicted as the mean (95% CIs) for continuous variables and percentages (95% CIs) for categorical variables among all participants and in subgroups of sex. The χ 2 test (for categorical variables) was used to analyse the differences in the socioeconomic characteristics, physical function and biomarkers between diabetic and pre-diabetic individuals. The associations between awareness, control of glucose, treatment of diabetes (dependent variable) and independent variables were analysed using multivariate logistic regression analysis and reported as adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs). Statistical significance was defined as a p-value <.05. All analyses were performed using STATA software (version 14.0; Stata Corp LP. TX). Table 1 summarizes and compares the general characteristics between male and female. The mean age of the study subjects was 60.66 years (SD 9.74 years). There were 5889 men (47.27%) and 6569 women (52.73%). The diastolic blood pressures and fasting glucose were not significantly different between different sex. All other selected characteristics showed significant differences between male and female (p < .05).

| Prevalence of pre-diabetes and diabetes
Prevalence rates of pre-diabetes and diabetes in different subgroups of participants are shown in Table 2 alongside 95% confidence limits.
The overall unadjusted prevalence of diabetes was 13.21% (95% CI: 12.62%-13.81%), and the overall prevalence of pre-diabetes was 25.16% (95% CI: 24.39%-25.92%) in this cohort. Prevalence of diabetes is 12.37% in the 45 ~ 55 year group and increases with age, and more than one quarter of respondents (26.45%) have pre-diabetes among those aged between 55 and 65 years. There were no differences in the prevalence of diabetes and pre-diabetes among different sex and participants with different educational background. The overall standardized prevalence of diabetes adjusted for age and sex was 14.76%, and 12.59%, 15.96% and 16.69% among participants aged between 45 ~ 55 year, 55 ~ 65 year and over 65 years, respectively. Higher prevalence of diabetes and pre-diabetes is found among respondents who live in urban areas than those who live in rural area. Prevalence of diabetes and pre-diabetes increases with BMI, from 11.11% (95% CI: 10.30%-11.91%) in BMI < 18.5 group to 22.54% (95% CI: 20.59%-24.49%) in BMI > = 28 group. There is a higher pre-diabetes prevalence than diabetes prevalence for those who had a history of hypertension, stroke, coronary heart disease, ADL disability and lower/upper extremity functional limitation (p < .001). Figure 1 summarizes diabetes prevalence by sex categories. Moreover, we observed differences in systolic BP, diastolic BP, platelets and fasting glucose between normal, pre-diabetes and diabetes participants (Table 3).     16 Similarly, in this study, we found the prevalence of diabetes among Chinese adults aged 45 years or older in 2015 was 13.21% (13.6% in male and 11.54% in female), and the overall prevalence of pre-diabetes was 25.16%. The prevalence of diabetes in our present study was also relatively higher than in some regional surveys in China reported in recent years, such as in Kazakh adults in Xin Jiang (7.3%, 11.6% and 10.5% in the age groups of 45-54, 55-64 and 65 years or older, respectively). 15 However, the prevalence of diabetes among Chinese rural population aged 40 years and older was higher, climbing to 24.7%. 9 The difference might be due to the discrepancy in the socioeconomic status of different region, as well as age, lifestyles and dietary habits among different population. 9 In our study, we also found that age, residence, BMI, history of hypertension, stroke and coronary heart disease, having ADL disability and lower/upper extremity functional limitation were significantly associated with higher risk of diabetes and pre-diabetes.

| Factors related with awareness, treatment and control of diabetes
The prevalence of diabetes was higher in male population aged 45 ~ 55 years than in female. In other age groups, the prevalence of female was greater. It is said that more women are overweight or obese after the age of 45 years, whereas more males are overweight at younger age. 17 Because obesity is the major risk factor of diabetes in both sexes, the prevalence of diabetes among male in all age groups might be higher than female, and the prevalence of diabetes among female at middle-aged and older might be greater. It is not surprising that the prevalence of patterns of diabetes across regions

F I G U R E 1 Prevalence of diabetes among male and female population
resembles those of obesity. 18 Higher prevalence of diabetes and prediabetes among respondents who live in urban areas than those who live in rural area were observed in this study, which was consistent with previous findings. 5,19,20 Also, higher prevalence of diabetes and pre-diabetes was seen among participants with higher BMI, and both obesity and inactivity were strongly associated with decreased odds of awareness of diabetes among diabetic participants. Urbanization is associated with changes in lifestyle that lead to physical inactivity, an unhealthy diet, and obesity, all of which have been implicated as contributing factors in the development of diabetes. 15 Participants living in urban areas were also more likely to be well informed and report antidiabetic treatment and to control their diabetes better than rural residents. 21 These may reflect the problem of accessibility and the quality of health care among this disadvantaged group.
The government should put into place basic medical and health care covering more rural residents in the future health policy setting and ensure more rural resident have access to convenient and affordable basic medical and health services.
Additionally, we observed significant differences in systolic BP, diastolic BP, platelets and fasting glucose between normal among pre-diabetes and diabetes participants. Since the functional changes occurring in diabetes and hypertensive conditions significantly alter the haemodynamic stress on the heart and other organs, prior study found that people with diabetes and hypertension were associated with an increased risk of cardiovascular mortality compared with those with either condition alone. 22 Several pathogenic mechanisms have been proposed to explain the association between diabetes and hypertension, 23 including the incretin-mediated control of the renin-angiotensin-aldosterone system and alterations in calciumcalmodulin system. Therefore, the blood pressure and biomarkers of diabetic participants need to be closely monitored to reduce the mortality caused by cardiovascular disease among middle-aged and older Chinese adults.
Findings from our survey also showed that the overall proportion of patients who were aware of their diabetes condition was 39.29%; of those who were aware of their condition, 47.04% were receiving Additionally, we found that aged between 55 and 65 years, being obesity, having a history of hypertension and coronary heart disease, as well as being inactivity were both significantly associated with higher odds of awareness, which indicated that individual consciousness played an important role in the awareness of diabetes. Chinese adults aged 45 years and older. Second, a multi-stage stratified random cluster method was used in the sampling process, and complex weighted computation was used in data analysis. Third, venous blood samples were collected and transported to the same laboratory for testing using the same standardized protocol. All these features enhanced the accuracy of prevalence estimates.
Our study also had some limitations. First, we did not distinguish between type 1 and type 2 diabetes among these patients.
Second, our study findings are likely to have been influenced by additional factors that were not performed in participants of CHARLS, such as 2-h oral glucose-tolerance. Therefore, the prevalence of diabetes and pre-diabetics was probably underestimated.
Finally, because our study was a cross-sectional survey, we could not establish cause-and-effect relationships between the observed associations.
In summary, our study showed that diabetes and pre-diabetes were highly prevalent among adults aged 45 years or older in China.
More troublesome is the finding that awareness, treatment and control rates of diabetes remained relatively low compared with those of developed countries. Given the paucity of data on diabetes among Chinese middle-aged and older adults, which represent the world's largest ageing population, our study may serve as a basis for future research aimed at identifying physiological, behavioural, and psychosocial risk factors of diabetes, and finally improve the prevention, detection and treatment rates of diabetes in China.

CO N FLI C T O F I NTE R E S T
The authors declare no potential conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Anying Bai collected data, conceptualized the study, performed statistical analysis and wrote the manuscripts; Jing Tao conceptualized the study and reviewed/edited the manuscripts; Liyuan Tao contributed to the discussion and reviewed/edited the manuscripts; Jue Liu conceptualized the study, acquired the funding, administrated the project and reviewed/edited the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The CHARLS data are publicly available at http://charls.pku.edu.cn.