Abstract
Abstract
Objective
This study investigates the prevalence of multi-morbidity, defined as the coexistence of two or more of three common chronic non-communicable diseases (NCDs)—hypertension, diabetes and dyslipidaemia—and assesses the effect of obesity on multi-morbidity among multi-ethnic populations in Xinjiang, northwest China, aiming to develop tailored prevention strategies for NCDs.
Methods
A retrospective cohort study was conducted, including 62 172 adults from 2018 to 2023. Restricted cubic spline curves were used to explore the dose-response effect between body mass index (BMI)/waist circumference (WC) and the incidence of multi-morbidity.
Results
Over 6 years, the incidence rates of multi-morbidity in our study increased significantly from 24.48% to 29.24%. The multi-morbidity rates in Hui and Uyghur ethnic populations were lower than those in Han Chinese. The HRs of central obesity on the incidence of multi-morbidity were 1.34 (1.29–1.39), whereas obesity was 1.50 (1.43–1.57). Both BMI and WC were non-linearly associated with the incidence of multi-morbidity. BMI showed a stronger association among Han Chinese, while WC had a greater effect on ethnic minorities.
Conclusions
Obesity significantly contributed to the development of multi-morbidity, with varied impacts across ethnic groups. Prevention and control strategies should prioritise BMI for Han Chinese and WC for ethnic minorities. This study highlighted the need for ethnicity-specific approaches to reduce the burden of multi-morbidity.
Keywords: Obesity, Adolescent, Chronic Disease, Multimorbidity, China
STRENGTHS AND LIMITATIONS OF THIS STUDY.
All anthropometric and biochemical indicators were measured by trained researchers to avoid self-report bias.
Multi-morbidity outcomes were identified via both physician clinical diagnosis and participant self-reported disease records.
A large multi-ethnic cohort from Yili, Xinjiang was enrolled for long-term retrospective follow-up analysis.
Restricted cubic spline curves and stratified Cox models were applied to assess nonlinear and ethnic heterogeneity associations.
Multiple population-level geographic subgroup analyses were conducted using ArcGIS visualisation tools.
Introduction
Multi-morbidity is generally defined as the coexistence of two or more chronic non-communicable diseases (NCDs) in a given individual.1 The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021 estimated that NCDs had already accounted for more than half of the global burden of disease.2 The prevalence of multi-morbidities had been rising globally since the early this century. A systematic review of 126 studies across 54 countries reported that 37.2% of adults experienced multi-morbidity, with the prevalence exceeding 50% among individuals aged 60 years and older. Women were disproportionately affected, with rates of 39.4% compared with 32.4% in men.3 4
NCDs were also the leading causes of disease burden in China, and this burden was increasing due to the ageing population and changes in lifestyles.5 6 Meta-analysis studies of cross-sectional and cohort studies showed that nearly one in four Chinese adults had multiple chronic conditions, with multi-morbidity prevalence increasing rapidly with ageing among individuals and wider society.7 8
In the context of global socioeconomic development, multi-morbidities will be an important public health problem that affects the health of the population in the long term.9 Individuals with multi-morbidities experienced reduced physical functioning and lower quality of life. They also faced mental and psychosocial disorders, increased comorbidities, disability, premature death and higher mortality.10 11 Older adults with multiple diseases were more likely to experience reduced quality of life, depressive disorders, frequent hospitalisations and could lead to premature death,12 which significantly reduced an individual’s quality of life and livelihood. Therefore, the multi-morbidity not only raised long-term public health concerns, but also represented an urgent pragmatic need to improve the capacity of the existing healthcare system.
The western China region, represented by the Xinjiang Uygur Autonomous Region (XUAR), was relatively less developed in terms of health conditions and had an outstanding burden of chronic diseases, due to its unique geographical, socioeconomic, ethnic and cultural characteristics. Understanding the prevalence and determinants of multi-morbidity in this region was critical for informing tailored prevention and control strategies. Thus, this study focused on Yili prefecture in Xinjiang, analysing data from 2018 to 2023 to explore the prevalence, trends and risk factors of multi-morbidity among its multi-ethnic population. In this study, multi-morbidity specifically refers to the coexistence of two or more of three common cardiometabolic conditions: hypertension, diabetes and dyslipidaemia.
Research design and methods
Study design and population inclusion criteria
This retrospective cohort study analysed data from adult residents of Yili, Xinjiang, who participated in health check-ups between 2018 and 2023. Participants were selected based on the availability of complete and reliable health records. Individuals were excluded if they met any of the following criteria: (1) participants with incorrect, duplicate or incomplete information; (2) those with extreme values in body mass index (BMI) and waist circumference (WC); (3) people with hypertension, diabetes and dyslipidaemia at baseline in 2018; (4) participants who did not participate in three or more surveys. The final cohort included 62 172 individuals for the study (figure 1).
Figure 1. Flowchart of the study participants.
Data source and study procedures
The data were derived from electronic health check-up records of adult residents in Yili, Xinjiang, collected between 2018 and 2023 as part of a government-funded free health examination programme. The programme was implemented at the township level and covered demographic characteristics, lifestyle behaviours, anthropometric measurements, blood pressure (BP) and biochemical indicators.
For individuals with mobility difficulties, home-visit examinations were conducted to ensure coverage. All examinations were performed by trained medical staff following standardised operating procedures. Abnormal findings were further evaluated by physicians, and confirmed chronic disease cases were registered in the chronic disease management system. The present study is a secondary analysis of anonymised data extracted from hospital health examination records. All participants had provided written informed consent at the time of the health examination for their data to be used for medical research and health management purposes. The analysis used de-identified data and did not involve any identifiable personal information.
Measurements and definitions
Anthropometric measurements
Height was measured to the nearest 0.1 cm using a calibrated stadiometer with participants standing barefoot in an upright position. Weight was measured to the nearest 0.1 kg using a calibrated digital scale with participants wearing light clothing and no shoes. All instruments were calibrated daily. Measurements were conducted by trained nurses and medical staff who received standardised training before data collection, and inter-observer reliability was periodically assessed to ensure consistency.
BMI was calculated as weight (kg) divided by height squared (m²). According to the criteria of the Working Group on Obesity in China (WGOC), BMI was categorised as follows: underweight (<18.5 kg/m²), normal weight (18.5–23.9 kg/m²), overweight (24.0–27.9 kg/m²) and obesity (≥28.0 kg/m²). These criteria are widely used in studies of Chinese populations and reflect differences in body composition between Asian and Western populations, as Asians tend to have higher body fat percentages at lower BMI levels. Therefore, lower BMI thresholds are recommended for identifying obesity-related health risks in Asian populations.
WC was measured to the nearest 0.1 cm at the midpoint between the lowest rib and the iliac crest at the end of normal expiration using a non-elastic measuring tape. Central obesity was defined according to the WHO Asian-specific criteria as WC≥90 cm for men and ≥80 cm for women.
Blood pressure measurement and definition
BP was measured using a standardised electronic sphygmomanometer after the participant had rested for at least 5 min in a seated position. Three consecutive measurements of systolic blood pressure (SBP) and diastolic blood pressure (DBP) were recorded, and the average value was used for analysis.
People with hypertension was defined as SBP≥140 mm Hg and/or DBP≥90 mm Hg, or current use of antihypertensive medication, according to established national guidelines. This threshold is consistent with commonly used diagnostic criteria in epidemiological studies and national hypertension guidelines in China during the study period.
Biochemical measurements and definitions
Fasting venous blood samples were collected after at least 8 hours of fasting. Fasting blood glucose (FBG) and lipid profiles were analysed using standardised automated biochemical analysers in certified laboratories. People with diabetes was defined as FBG≥7.0 mmol/L, self-reported physician diagnosis or current use of antidiabetic medication. In the health examination programme, FBG was measured once during each visit as part of routine clinical practice. Dyslipidaemia was defined as the presence of one or more of the following lipid abnormalities: Total cholesterol≥5.2 mmol/L, triglycerides (TG)≥1.7 mmol/L, low-density lipoprotein cholesterol (LDL-c)≥3.4 mmol/L, high-density lipoprotein cholesterol<1.0 mmol/L.
Lifestyle behaviours assessment and definitions
Lifestyle behaviours were assessed using a structured questionnaire administered during the health examination. Smoking status was classified according to WHO STEPwise approach to surveillance criteria as people who never smoker, people who are former smoker or people who are current smoker. People who are current smoker were defined as individuals who had smoked at least 100 cigarettes in their lifetime and were currently smoking. Alcohol consumption was categorised into four groups: never, rarely (<12 times per year), often (≥1 time per week but not daily) and daily drinking. This classification was adapted from commonly used categories in population-based health surveys and lifestyle assessments in China.
Physical activity was assessed based on self-reported frequency of moderate-intensity activities (eg, walking, running, cycling or jogging) according to the WHO 2020 physical activity guidelines.13 Physical inactivity was defined as engaging in moderate-intensity physical activity fewer than 5 days per week and for less than 30 min per day.
Definition of multi-morbidity
Multi-morbidity was defined as the coexistence of two or more of the following chronic non-communicable diseases: hypertension, diabetes and dyslipidaemia.
Statistical analysis
First, we categorised participants into three age groups: the young group (18–34 years old), the middle-aged group (35–64 years old) and the older group (65 years old and above) and employed the trend χ2 test to assess the trend of multi-morbidity rates with age using R software (V.4.4.1). Similarly, we conducted a time trend analysis for the multi-morbidity rates of sex and ethnicity. Subsequently, the geographical distribution of average multi-morbidity rates across 18 townships in Yili, Xinjiang, from 2018 to 2023 was visualised using ArcGIS. These maps also clearly displayed the distribution differences of multi-morbidity rates by sex and age groups among the townships. In addition, we explored the relationship between BMI or WC and multi-morbidity using a restricted cubic spline (RCS) approach, including an analysis of the potential effects of sex and ethnicity on these relationships. Stratified analyses by sex and ethnicity were conducted to explore potential heterogeneity in the association between obesity indicators and multi-morbidity. Ultimately, we used the Cox proportional hazards regression model to identify the associated risk factors for the multi-morbidity, with results expressed as HRs and their corresponding 95% CIs.
Given the large sample size, statistical significance was interpreted alongside effect sizes and 95% CIs to avoid overemphasis on clinically trivial differences. We also evaluated the magnitude and consistency of associations rather than relying solely on P values.
Results
Baseline characteristics
Table 1 shows the baseline characteristics of 62 172 participants in this study. In general, 17 412 individuals (28.01%) were identified as having multi-morbidity, defined as two or more chronic NCDs. The mean age of the study participants was 54 (48,66) years, with participants in the multi-morbidity group being significantly older than those without multi-morbidity (p<0.001). Of the total participants, 28 303 (45.50%) were male, with a higher proportion of males in the non-multi-morbidity group compared with the multi-morbidity group (p<0.001). Participants in the multi-morbidity group had higher levels of BMI, WC, SBP, DBP, TG and LDL-c. The study population encompassed multiple ethnic groups, including Han, Hui, Mongolian, Uyghur and Kazakh ethnic groups, although the distribution of ethnicities between the multi-morbidity and non-multi-morbidity groups was not significantly different. We also examined the ethnicity distribution of participants excluded at baseline due to pre-existing chronic diseases and did not observe substantial differences compared with the included cohort. People who smoke, drink alcohol daily and rarely exercise constituted a similar proportion in both groups. The multi-morbidity group had a significantly higher proportion of people with diabetes, hypertension and dyslipidaemia than the other group (p<0.001).
Table 1. Demographic and clinical characteristics of the study population (Median (P25, P75)/n (%)).
| Characteristics | Total (n=62 172) | Multi-morbidity (n=17 412) | Non-multi-morbidity (n=44 760) | H/χ² | P† value |
|---|---|---|---|---|---|
| Age, years | 54 (48,66) | 62 (52, 74) | 52 (46, 61) | 5353.8 | <0.001 |
| Male | 28 303 (45.52) | 7591 (43.60) | 20 712 (46.27) | 36.1 | <0.001 |
| Ethnicities | 479.7 | <0.001 | |||
| Han | 52 934 (85.14) | 15 621 (29.51) | 37 313 (70.49) | ||
| Hui | 1713 (2.76) | 435 (25.39) | 1278 (74.61) | ||
| Mongolian | 745 (1.20) | 107 (14.36) | 638 (85.64) | ||
| Uyghur | 2103 (3.38) | 484 (23.02) | 1619 (76.98) | ||
| Kazakh | 4428 (7.12) | 709 (16.01) | 3719 (83.99) | ||
| Others | 249 (0.40) | 56 (24.00) | 193 (76.00) | ||
| Smoking status | 208.5 | <0.001 | |||
| Never | 48 196 (77.52) | 13 744 (28.52) | 34 452 (71.48) | ||
| Quit | 2270 (3.65) | 859 (37.84) | 1411 (62.16) | ||
| Smoking | 11 706 (18.83) | 2809 (16.13) | 8897 (19.88) | ||
| Drinking status | 79.4 | <0.001 | |||
| Never | 49 070 (78.93) | 14 019 (28.57) | 35 051 (71.43) | ||
| Rarely | 11 075 (17.81) | 2754 (24.87) | 8321 (75.13) | ||
| Often | 1537 (2.47) | 465 (30.25) | 1072 (69.75) | ||
| Daily | 490 (0.79) | 174 (35.51) | 316 (64.49) | 2211.7 | <0.001 |
| Exercise status | |||||
| Never | 34 166 (54.95) | 7044 (20.62) | 27 122 (79.38) | ||
| Rarely | 4020 (6.47) | 1188 (29.55) | 2832 (70.45) | ||
| Often | 6078 (9.78) | 2486 (40.90) | 3592 (59.10) | ||
| Daily | 17 908 (28.80) | 6694 (37.38) | 11 214 (62.62) | ||
| Diabetes | 6349 (10.21) | 5716 (90.03) | 633 (9.97) | 13 488.0 | <0.001 |
| Hypertension | 20 340 (32.72) | 15 536 (76.39) | 4804 (23.61) | 35 083.0 | <0.001 |
| Dyslipidaemia | 42 783 (68.81) | 16 449 (38.45) | 26 334 (61.55) | 7416.1 | <0.001 |
| BMI, kg/m2 | 25.00 (23.10, 27.30) | 25.90 (23.80, 28.10) | 24.60 (22.80, 26.90) | 1763.1 | <0.001 |
| BMI categories | 1546.5 | <0.001 | |||
| Underweight/normal | 23 224 (37.35) | 4634 (19.95) | 18 590 (80.05) | ||
| Overweight | 27 257 (43.84) | 8178 (30.00) | 19 079 (70.00) | ||
| Obesity | 11 691 (18.80) | 4600 (39.34) | 7091 (60.66) | ||
| WC, cm | 86.00 (80, 92) | 88.00 (83, 95) | 85.00 (80, 90) | 2162.7 | <0.001 |
| WC categories | 2033.2 | <0.001 | |||
| Normal | 34 058 (54.78) | 7189 (21.11) | 26 869 (78.89) | ||
| Central obesity | 28 114 (45.22) | 10 223 (36.36) | 17 891 (63.64) | ||
| SBP, mm Hg | 125 (117, 134) | 136 (127, 144) | 120 (114, 130) | 10 524.0 | <0.001 |
| DBP, mm Hg | 78 (70, 81) | 80 (74, 88) | 76 (70, 80) | 4027.1 | <0.001 |
| TG, mmol/L | 1.57 (1.23, 2.16) | 1.88 (1.41, 2.56) | 1.47 (1.2, 1.98) | 2964.2 | <0.001 |
| TC, mmol/L | 1.62 (1.32, 2.06) | 1.62 (1.31, 2.07) | 1.62 (1.32, 2.06) | 4.1 | <0.05 |
| HDL-c, mmol/L | 1.62 (1.32, 2.06) | 1.62 (1.31, 2.07) | 1.62 (1.32, 2.06) | 4.1 | <0.05 |
| LDL-c, mmol/L | 2.54 (1.86, 3.3) | 2.76 (2.06, 3.58) | 2.46 (1.79, 3.18) | 885.1 | <0.001 |
| Follow-up time, years | 3.34 (2.39, 3.92) | 2.74 (1.88, 3.54) | 3.36 (2.41, 3.92) | 154.5 | <0.001 |
Multi-morbidity was defined as the co-occurrence of ≥2 of three cardiometabolic diseases (hypertension, diabetes, and dyslipidaemia).
Kruskal–Wallis H test and χ² test were used for between-group comparisons.
P<0.05.
P<0.001.
Prebronchodilator.
BMI, body mass index; BP, blood pressure; DBP, diastolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WC, waist circumference.
Trends in multi-morbidity rates across different groups from 2018 to 2023
Figure 2A indicates that the multi-morbidity increased progressively over time across all age groups, with higher multi-morbidity rates of the older age group than that of the other two groups. Figure 2B shows the multi-morbidity rates by sex from 2018 to 2023, indicating that the rates in both males and females generally increased over time, with females consistently exhibiting higher multi-morbidity rates than males throughout the study period. Figure 2C illustrates the multi-morbidity rates of different ethnic groups over time from 2018 to 2023. The multi-morbidity rates of different ethnic groups fluctuated during the study period, but generally maintained an upward trend, with significant differences in the multi-morbidity trends of the Han, Kazakh, Mongolian and Uyghur ethnic groups. Detailed data are provided in online supplemental table S1.
Figure 2. Trend of multi-morbidity rates in different groups from 2018 to 2023.
Analysis of the geographical distribution of multi-morbidity rates
Average multi-morbidity rates during the 6 years for each township
In figure 3A, multi-morbidity rates varied significantly between different townships. Twelve townships had average rates between 20.00% and 29.99%, five townships had rates between 30.00% and 39.99% and one township had a rate of 48.42%.
Figure 3. Average multi-morbidity rates of different groups during the 6 years for each township.
Average multi-morbidity rates by sex during the six years for each township
Figure 3B reveals that in the majority of townships, the average multi-morbidity rates for males ranged from 20.00% to 29.99%. However, three townships reported elevated rates between 30.00% and 39.99% range, and two townships had particularly high male multi-morbidity rates exceeding 40.00%, reaching the 40.00%–49.99% range.
In figure 3C, the multi-morbidity rates among women in most townships also clustered between 20.00% and 29.99%. Five townships, however, exhibited a heightened rate of multi-morbidity in women, which was situated between 30.00% and 39.99%. Moreover, two other townships stood out with a female multi-morbidity rate surpassing 40.00%, spanning the 40.00%–49.99% range.
Average multi-morbidity rates by age group during the 6 years for each township
According to the data in figure 3D, there was no township where the average multi-morbidity rate of the older was in the range of 10.00%–19.99%, indicating that the average multi-morbidity rate of the older was generally high. Two townships had an average multi-morbidity rate for older adults in the 30.00%–39.99% range, while all other townships had an average multi-morbidity rate for older adults above 40.00%.
As shown in figure 3E, the average multi-morbidity rate of the middle-aged population in nine townships ranged between 20.00% and 29.99%, and four townships had a lower multi-morbidity rate of 10.00%–19.99%, whereas the other four townships had higher average multi-morbidity rates for the middle-aged population, in the 30.00%–39.99% range, and only one township had average multi-morbidity rates for the middle-aged population that was higher than 40.00%.
The data from figure 3F reveals that the average multi-morbidity rate of the youth population was at a low level in all townships, with 15 townships having an average multi-morbidity rate of youth in the range of 0.00%–9.99%, and three townships had average youth multi-morbidity rates in the 10.00%–19.99% range.
Analysis of multi-morbidity risk factors
In figure 4A, people with central obesity showed a higher rate of multi-morbidity compared with people being absent of central obesity (HRs 1.34, 95% CIs 1.29 to 1.39). In figure 4B, individuals having obesity had significantly increased multi-morbidity compared with the individuals without obesity (HRs 1.50, 95% CIs 1.43 to 1.57). Data from both figures highlighted a significant correlation between age and daily alcohol consumption with the likelihood of developing multi-morbidities (p<0.05), with the highest risk being observed among older individuals and those who drank alcohol daily. In contrast, the risk of multi-morbidity was significantly lower among the Hui and Uyghur ethnic minorities compared with the Han majority.
Figure 4. Forest plot for cox proportional‑hazards regression for multi-morbidity of three chronic non-communicable diseases (NCDs). (A) Stratified by central obesity; (B) Stratified by obesity.
Assessment of the impact of different body weight status on multi-morbidity rates
In figure 5A, the survival graph clearly demonstrated the effect of obesity on the probability of morbidity. The survival probability of individuals without obesity was consistently higher than individuals having obesity throughout the study period. Over time, the survival probability decreased in both groups, but the rate of decline was faster in the group having obesity, indicating a higher risk of multi-morbidity in individuals having obesity. The HRs were higher in the group having obesity than in the group without obesity and this difference was statistically significant. We also observed similar trend of the effect of central obesity on the probability of survival in figure 5B.
Figure 5. Kaplan-Meier plot of event-free survival. (A) Stratified by obesity; (B) Stratified by central obesity.
The relationship between BMI and WC with multi-morbidity rates
RCS stratified by ethnicity
As shown in figure 6, the RCS results showed a significant nonlinear dynamic association between BMI and risk of multi-morbidity (P for non-linear<0.001), as well as a nonlinear trend between WC and risk of multi-morbidity in the overall population (p<0.05). In the RCS plot of the relationship between BMI and multi-morbidity, multi-morbidity increased with increasing BMI. However, the RCS plot depicting the relationship between WC and multi-morbidity revealed an increase in multi-morbidity with rising WC within the range of HRs>1, while a decrease was observed in the range of HRs<1 as WC continued to increase. Ethnicity was used as a stratification variable in the analyses to explore the heterogeneity of the association of BMI and WC with multi-morbidity in different ethnic groups. In the ethnically stratified analysis of BMI, the risk of multi-morbidity was significantly higher in the Han Chinese population than in other ethnic groups at HRs>1, whereas the risk of multi-morbidity was significantly lower in the Han Chinese population than in other ethnic groups at HRs<1. In the ethnically stratified analyses of WC, the risk of multi-morbidity was almost always significantly higher in the other ethnic groups than in the Han Chinese.
Figure 6. Restricted cubic spline (RCS) of body mass index (BMI) and waist circumference (WC) on multi-morbidity by ethnicity.
RCS stratified by sex
The results depicted in figure 7A indicate that for women, the risk of multi-morbidity exhibited a clear upward trend with increasing BMI, with HRs values showing a significant increase, particularly after a BMI of more than 25. Men exhibited a similar trend, although at some BMI values, the HRs increased slightly less rapidly in men compared with women. Figure 7B demonstrates that the association between WC and the risk of multi-morbidities was significant for both males and females in sex-stratified analyses; however, the HRs differed between the sexes. Specifically, women had higher HRs values than men, suggesting that women might be at a higher risk of multi-morbidities than men at the same level of WC. In the WC range of 70–90 cm, the risk ratio was approximately 1.5 for women, while it was lower for men. This sex difference in the risk of multi-morbidity became more pronounced as WC increased.
Figure 7. Restricted cubic spline (RCS) of body mass index (BMI) and waist circumference (WC) on multi-morbidity by. (A) Stratified by BMI; (B) Stratified by WC.
Strengths and limitations
This study possessed several prominent methodological strengths. To our knowledge, it represents the first investigation to explore NCD multi-morbidity based on a large multi-ethnic sample across Xinjiang, China. All anthropometric indicators including weight, height and WC, alongside biochemical laboratory markers, were measured on-site by professionally trained researchers rather than relying on participant self-report, which effectively minimised misclassification bias stemming from self-reported data. Furthermore, the identification of multi-morbidity outcomes achieved relatively high accuracy and comprehensiveness by integrating two sources of diagnostic information: participants’ self-reported chronic disease histories and formal physician diagnoses obtained through standardised clinical examinations.
Despite these advantages, the research also carries several noteworthy limitations, with special consideration required for statistical interpretation driven by the study’s substantial sample size of 62 172 participants. The large cohort scale substantially increased statistical power, creating a risk of overpowering whereby minor, clinically trivial differences could still yield statistically significant associations. In response to this potential bias, we did not implement additional formal statistical adjustment procedures specifically designed to mitigate excessive statistical power; instead, we adopted a conservative interpretative strategy during all analytical stages, consistently evaluating statistical significance alongside corresponding HRs, 95% CIs and effect magnitudes rather than drawing conclusions solely based on P values. This comprehensive approach was applied throughout trend analyses, restricted cubic spline modelling and Cox proportional hazards regression to avoid overinterpreting weak statistical signals lacking practical clinical meaning. Beyond statistical considerations, several diagnostic and data collection limitations remain. Diabetes diagnoses were confirmed only through single-measurement FBG results combined with participant self-reporting, which may underestimate true diabetes prevalence within the cohort. All lifestyle exposure data, including smoking, alcohol intake and physical activity frequency, were collected via questionnaire self-report, introducing inherent recall bias. Additionally, uneven access to routine healthcare services across townships may create disparities in chronic disease detection during annual physical examinations, meaning certain chronic conditions identified at participants’ initial health check-ups may have remained undiagnosed for years prior to enrolment.
Discussion
This study highlighted significant disparities in the incidence and multi-morbidity of three chronic NCDs across ethnic groups in Yili, Xinjiang, with a strong association between multi-morbidity and obesity.
Our findings showed that multi-morbidity rates escalated with age in Yili, Xinjiang, over a 6-year period. This trend was largely attributed to the ageing population, as the number of older Chinese was on the rise and was expected to double in both number and as a percentage of the total population within the next two decades.7 14 Although the prevalence estimated fluctuates across and within different age groups, our research indicated that a significant number of individuals aged 65 years and above exhibited the highest levels of multi-morbidity. It was a well-documented fact that multi-morbidity prevalence rose among the older,15 16 stemming from the gradual decline in metabolic functions that occurred throughout the lifespan of older adults.17 In our study, females exhibited a higher prevalence of multi-morbidity compared with males, aligning with numerous studies conducted in both developed and developing nations.3 18 19 This disparity might stem from women’s longer life expectancy, which exposed them to risk factors for a more extended period. Additionally, the high rates of widowhood and solitary living among women might contribute to the development of certain psychiatric and psychological conditions.20
Geographical disparities in multi-morbidity rates might be influenced by variations in environmental factors, lifestyle behaviours, economic status and access to healthcare services across different regions in Yili, Xinjiang. In addressing the prevention and control of chronic disease multi-morbidities, it was crucial to concentrate on chronic diseases with high incidence rates, such as obesity, hypertension and hyperlipidaemia, and to take regional factors into full account, particularly in areas of Xinjiang with high prevalence and concentration.21 22
Furthermore, older adults having obesity or central obesity, as well as those who consume alcohol daily, were at a greater risk of chronic NCDs multi-morbidity. Obesity of central obesity and alcohol drinking were established major risk factors for chronic NCDs multi-morbidity. These findings emphasised the necessity of comprehensive lifestyle interventions for the prevention and management of chronic NCDs multi-morbidity.23–25
Ethnic differences in multi-morbidity prevalence were also observed, with Han majority adults experiencing higher rates than Hui and Uyghur ethnic minority groups. These ethnic disparities may also be influenced by differences in socioeconomic status, cultural practices and traditional dietary patterns among ethnic groups in Xinjiang, which may affect lifestyle behaviours and cardiometabolic risk profiles. Similar ethnic disparities in chronic NCDs had been documented in Western countries.26 27 Further research is warranted to explore the precise nature of the relationship between ethnicity and multi-morbidities. The differing impacts of BMI and WC on the Han Chinese population and other ethnic groups may be related to variations in body fat distribution, genetic susceptibility and lifestyle behaviours among different ethnic populations.28 29 In addition, differences in dietary patterns, physical activity and socioeconomic conditions among ethnic groups may influence the relative importance of overall adiposity and central obesity in the development of multi-morbidity. Previous studies have suggested that central obesity, measured by WC, may be a stronger predictor of cardiometabolic risk, whereas BMI reflects overall adiposity.29 In multi-ethnic populations, differences in body composition and metabolic characteristics may lead to heterogeneous associations between obesity indicators and multi-morbidity. Alternatively, it could be because the obesity threshold for minority populations was higher than the current standard in China (BMI≥28 kg/m²), which warrants further investigation in future studies on obesity and its metabolic diseases in ethnic minorities.30
Conclusions
This study highlights the substantial public health challenge posed by chronic NCDs and their associated multi-morbidities in Xinjiang, China, particularly in ethnically diverse border regions. Both general obesity and central obesity were significantly associated with an increased risk of developing multi-morbidity. Therefore, obesity control is important and necessary for reducing the burden of NCDs in this region.
Importantly, our findings provide new evidence that the impact of obesity indicators on multi-morbidity differs across ethnic groups. BMI showed a stronger association with multi-morbidity among Han Chinese, whereas WC demonstrated a stronger relationship among ethnic minority populations. These results suggest potential ethnic heterogeneity in adiposity-related metabolic risk and highlight the importance of considering ethnic differences when developing chronic disease prevention strategies.
In addition, the geographical analysis revealed clustering of multi-morbidity in certain townships, suggesting that local environmental conditions, lifestyle patterns and access to healthcare resources may contribute to regional disparities in disease burden. These findings provide valuable information for policymakers and public health practitioners to identify high-risk populations and regions.
Given that general obesity and central obesity share common behavioural and metabolic determinants, prevention strategies should prioritise comprehensive weight management rather than targeting BMI or WC separately. Evidence-based interventions that have been shown to improve obesity-related outcomes include promotion of healthy dietary patterns, increased physical activity and reduction of harmful alcohol consumption.
Importantly, public health programmes should also consider how to effectively reach ethnic minority populations. Community-based health promotion, culturally appropriate dietary guidance and locally adapted physical activity programmes may help improve the accessibility and effectiveness of obesity prevention strategies in minority communities. These approaches may contribute to reducing ethnic disparities in the burden of multi-morbidity.
Overall, our findings underscore the need for region-specific and ethnicity-sensitive public health strategies to reduce obesity and prevent the development of multi-morbidity in multi-ethnic populations in northwest China.
Supplementary material
Footnotes
Funding: This work was supported by National Science and Technology Innovation 2030 Major Program for Four Major Chronic Diseases: 2023ZD0508500, 2023ZD0508504.
Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-102146).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by The 4th Division Hospital of Xinjiang Production and Construction Corps (No. BTSSHEC-C-2024-001-016). Participants gave informed consent to participate in the study before taking part. This study used anonymizedanonymised secondary data extracted from hospital health examination records. All participants had provided written informed consent at the time of the health examination for their health information to be used for medical research and health management purposes. The present study involved analysis of de-identified data and did not include any identifiable personal information. All procedures were conducted in accordance with the principles of the Declaration of Helsinki.
Map disclaimer: The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Patient and public involvement: Patients and/or the public were involved in the design, or conduct, or reporting or dissemination plans of this research. Refer to the Research design and methods section for further details.
Data availability statement
Data are available upon reasonable request.
References
- 1.Geneva: World Health Organization; 2016. Multimorbidity: technical series on safer primary care.https://www.who.int/publications/i/item/9789241511650 Available. [Google Scholar]
- 2.Collaborators GCoD Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2024;403:2100–32. doi: 10.1016/S0140-6736(24)00367-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chowdhury SR, Chandra Das D, Sunna TC, et al. Global and regional prevalence of multimorbidity in the adult population in community settings: a systematic review and meta-analysis. eClinicalMedicine. 2023;57:101860. doi: 10.1016/j.eclinm.2023.101860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chen Z, Chen J, Collins R, et al. China Kadoorie Biobank of 0.5 million people: survey methods, baseline characteristics and long-term follow-up. Int J Epidemiol. 2011;40:1652–66. doi: 10.1093/ije/dyr120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhou M, Wang H, Zeng X, et al. Mortality, morbidity, and risk factors in China and its provinces, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. The Lancet. 2019;394:1145–58. doi: 10.1016/S0140-6736(19)30427-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang Y, Zhao L, Gao L, et al. Health policy and public health implications of obesity in China. Lancet Diabetes Endocrinol. 2021;9:446–61. doi: 10.1016/S2213-8587(21)00118-2. [DOI] [PubMed] [Google Scholar]
- 7.Hu Y, Wang Z, He H, et al. Prevalence and patterns of multimorbidity in China during 2002-2022: A systematic review and meta-analysis. Ageing Res Rev. 2024;93:102165. doi: 10.1016/j.arr.2023.102165. [DOI] [PubMed] [Google Scholar]
- 8.Reynolds K, Gu D, Whelton PK, et al. Prevalence and risk factors of overweight and obesity in China. Obesity (Silver Spring) 2007;15:10–8. doi: 10.1038/oby.2007.527. [DOI] [PubMed] [Google Scholar]
- 9.The Lancet Making more of multimorbidity: an emerging priority. The Lancet. 2018;391:1637. doi: 10.1016/S0140-6736(18)30941-3. [DOI] [PubMed] [Google Scholar]
- 10.Kanesarajah J, Waller M, Whitty JA, et al. Multimorbidity and quality of life at mid-life: A systematic review of general population studies. Maturitas. 2018;109:53–62. doi: 10.1016/j.maturitas.2017.12.004. [DOI] [PubMed] [Google Scholar]
- 11.Barnett K, Mercer SW, Norbury M, et al. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. 2012;380:37–43. doi: 10.1016/S0140-6736(12)60240-2. [DOI] [PubMed] [Google Scholar]
- 12.Salisbury C. Multimorbidity: time for action rather than words. Br J Gen Pract. 2013;63:64–5. doi: 10.3399/bjgp13X661020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bull FC, Al-Ansari SS, Biddle S, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54:1451–62. doi: 10.1136/bjsports-2020-102955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Luo Y, Su B, Zheng X. Trends and Challenges for Population and Health During Population Aging - China, 2015-2050. China CDC Wkly. 2021;3:593–8. doi: 10.46234/ccdcw2021.158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Marengoni A, Angleman S, Melis R, et al. Aging with multimorbidity: a systematic review of the literature. Ageing Res Rev. 2011;10:430–9. doi: 10.1016/j.arr.2011.03.003. [DOI] [PubMed] [Google Scholar]
- 16.Wang Z, Peng W, Li M, et al. Association between multimorbidity patterns and disability among older people covered by long-term care insurance in Shanghai, China. BMC Public Health. 2021;21:418. doi: 10.1186/s12889-021-10463-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ho I-S, Azcoaga-Lorenzo A, Akbari A, et al. Examining variation in the measurement of multimorbidity in research: a systematic review of 566 studies. Lancet Public Health. 2021;6:e587–97. doi: 10.1016/S2468-2667(21)00107-9. [DOI] [PubMed] [Google Scholar]
- 18.van den Bussche H, Koller D, Kolonko T, et al. Which chronic diseases and disease combinations are specific to multimorbidity in the elderly? Results of a claims data based cross-sectional study in Germany. BMC Public Health. 2011;11:101. doi: 10.1186/1471-2458-11-101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Alimohammadian M, Majidi A, Yaseri M, et al. Multimorbidity as an important issue among women: results of a gender difference investigation in a large population-based cross-sectional study in West Asia. BMJ Open. 2017;7:e013548. doi: 10.1136/bmjopen-2016-013548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Keteepe-Arachi T, Sharma S. Cardiovascular Disease in Women: Understanding Symptoms and Risk Factors. Eur Cardiol. 2017;12:10–3. doi: 10.15420/ecr.2016:32:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Peng W, Chen S, Chen X, et al. Trends in major non-communicable diseases and related risk factors in China 2002-2019: an analysis of nationally representative survey data. Lancet Reg Health West Pac . 2024;43:100809. doi: 10.1016/j.lanwpc.2023.100809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Tabano DC, Bol K, Newcomer SR, et al. The Spatial Distribution of Adult Obesity Prevalence in Denver County, Colorado: An Empirical Bayes Approach to Adjust EHR-Derived Small Area Estimates. EGEMS (Wash DC) 2017;5:24. doi: 10.5334/egems.245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hu F, Xu L, Zhou J, et al. Association between Overweight, Obesity and the Prevalence of Multimorbidity among the Elderly: Evidence from a Cross-Sectional Analysis in Shandong, China. IJERPH. 2020;17:8355. doi: 10.3390/ijerph17228355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Srivastava S, Joseph K J V, Dristhi D, et al. Interaction of physical activity on the association of obesity-related measures with multimorbidity among older adults: a population-based cross-sectional study in India. BMJ Open. 2021;11:e050245. doi: 10.1136/bmjopen-2021-050245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zhang C, Xiao S, Shi L, et al. Urban-Rural Differences in Patterns and Associated Factors of Multimorbidity Among Older Adults in China: A Cross-Sectional Study Based on Apriori Algorithm and Multinomial Logistic Regression. Front Public Health. 2021;9:707062. doi: 10.3389/fpubh.2021.707062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Quiñones AR, Newsom JT, Elman MR, et al. Racial and Ethnic Differences in Multimorbidity Changes Over Time. Med Care. 2021;59:402–9. doi: 10.1097/MLR.0000000000001527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Du YR, Liu L, Zhao Y, et al. Ethnic disparities in prevalence of chronic non-communicable diseases and its multimorbidity among older adults in rural southwest China. BMC Public Health. 2023;23:1217. doi: 10.1186/s12889-023-16161-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Pan XF, Wang L, Pan A. Epidemiology and determinants of obesity in China. Lancet Diabetes Endocrinol. 2021;9:373–92. doi: 10.1016/S2213-8587(21)00045-0. [DOI] [PubMed] [Google Scholar]
- 29.Wang Y, Xue H, Sun M, et al. Prevention and control of obesity in China. Lancet Glob Health. 2019;7:e1166–7. doi: 10.1016/S2214-109X(19)30276-1. [DOI] [PubMed] [Google Scholar]
- 30.Yang Y, He K, Zhang Y, et al. Ethnicity Disparities in the Prevalence, Awareness, Treatment, and Control Rates of Hypertension in China. Int J Hypertens. 2023;2023:1432727. doi: 10.1155/2023/1432727. [DOI] [PMC free article] [PubMed] [Google Scholar]