The burden, trends, and risk factors for osteoarthritis in China from 1990 to 2019 and its predictions until 2035

doi:10.21203/rs.3.rs-4846817/v1

Objective: This research estimated osteoarthritis (OA) disease burden, trends, and risk variables in China from 1990 to 2019 and projected OA burden till 2035.

Methods: From the 2019 Global Burden of Disease (GBD) Study, data on incident cases, prevalence, disability-adjusted life years (DALY), and age-standardized rates (ASR) for OA were retrieved. We used estimated annual percentage change (EAPC) to do trend analysis, while a Bayesian age-period-cohort model was used to forecast OA tends to 2035. Besides, the burden attributed to OA risk factors was also assessed.

Results: The incidence, prevalence and DALY of OA increased 36%, 35%, and 40%, respectively, in China from 1990 to 2019. In China, knee OA was the most widespread, followed by hand OA, and hip OA is the least prevalent. The disease burden of OA was more severe among females than males, with differences in distribution between age groups. The ASIR of OA patients had risen steadily over the last 30 years and was expected to decrease somewhat over the next 15 years. In 2019, all risk variables contributed 12% of OA DALYs. The primary risk variables for OA DALYs were a high body mass index and advancing age.

Conclusion: The disease burden of OA in China has increased steadily over the last 30 years, but this trend will become less prominent or perhaps reverse during the next 15 years. High body mass index (BMI) has gradually emerged as an important factor in the disease burden of OA.

Osteoarthritis

Global disease burden

Trend

Prediction

Risk factor

China

Osteoarthritis (OA), the most common degenerative joint disease, has devastated elderly health and imposed a substantial economic burden.¹ In 2019 the OA prevalence and incidence were 6330.1 (per 100,000) and 509.8 (per 100,000) compared to 5880.6 (per 100,000) and 472.5 (per 100,000) in 1990.^2–4 As the global population ages, this growth may continue. As a prevalent and disabling disease, OA carried a tremendous burden and had a significant impact on the individuals affected, healthcare system, and society. In recent years, the burden of OA has been described in a few review papers from a few national studies, but detailed information is not available for all countries or regions.^5–7 However, no research has yet revealed annual national trends in the disability-adjusted life years (DALY), prevalence, and incidence of OA in China. Therefore, we reported the prevalence, incidence, and DALY of OA in the Chinese region and focused on the current status and long-term trends (by gender and age) of high BMI-associated OA over the past 30 years to assist China establish effective public health initiatives.

Data source

Based on the Global Burden of Disease study (GBD), this study examined the incidence, prevalence, disability-adjusted life years (DALY), and age-standardized rates (ASR) of OA in China. GBD 2019 is the Institute of Health Metrics and Evaluation's (IHME) most extensive epidemiological study to date. The full data sources and methods could be accessed through the Institute of Health's Global Health Data Exchange query tool (http://ghdx.healthdata.org/ghd-results-tool). Briefly, most of the data for China were obtained from the Chinese Center for Disease Control and Prevention, the National Central Cancer Registry, and the Cancer Incidence in Five Continents (CI5).¹⁰

Since this research employed de-identified GBD 2019 data, the institutional review board waived ethical approval and informed consent.¹¹

Disease burden

This study evaluated the disease burden of OA based on prevalence, incidence, and DALY, by DisMod-MR 2.1, a Bayesian meta-regression tool that evaluates all available incidence, prevalence, and remission data to assess the consistency of epidemiological parameters. DALY represented years lost to premature death (YLL) and disability (YLD).¹² YLL is calculated by multiplying the total number of fatalities by the average life expectancy at death. YLD is computed by multiplying the number of event incidences by the disability weight and the average number of years of disability. DALY represents a negative measure of life quality, which is a reasonable indicator and is best suited for assessing the burden of disease in a population.^13–14

Prediction model

In this study, a Bayesian APC model was used to predict the incidence of OA in China. The APC model can be stated as log(λ(A˙P)) = f(A) + g(P) + h(C), where A, P, and C represent the mean age, period, and birth cohort units of observation, and λ(A˙P) is the morbidity or mortality rate for women in birth cohort C at age A in period P.¹⁵ The study population was divided into six 5-year age groups and nine 5-year calendar groups. Because OA prevalence is 0% under 30, we consolidated the two age groups 20–24 years and 25–29 years into one age group of 30 years.

Risk factor

A comparative risk assessment approach is utilized to evaluate exposure and attributable DALY for behavioral, environmental, occupational, and metabolic risks in GBD 2019. Unfortunately, only high BMI was explicitly directly linked to OA in GBD 2019.¹⁶

Statistical analysis

To estimate trends in OA incidence, we utilized age-standardized incidence rates (ASR) and estimated annual percentage changes (EAPC). The incidence rate that remains after age is taken out of the equation and is identified as the age-standardized incidence rate (ASR).¹⁷ The OA ASR only compares the incidence of osteoarthritis in various nations, regions, or historical eras within the same region. If the age structure of the population differs substantially between two regions, comparing incidence rates does not demonstrate if the high incidence in one region is due to age structure or other variables. Hence, age-standardized incidence rates are necessary.

The EAPC is used to summarize trends in ASR over a specific period. The regression line is adjusted to the natural logarithm of the ASR value: y = α + βx + ε, where y = ln (ASR) and x = calendar year. With a 95% confidence interval (CI), EAPC is estimated as 100 (exp () − 1) using a linear regression model. When both the top and lower bounds of the EAPC and its 95% CI are 0, the ASR falls, rises, or remains unchanged.¹⁸

Analyses were performed using the statistical software R (Version 4.2.1); R packages included ggplot2, Colorbreed, BAMP, BAPC, and INLA. A two-sided P value less than 0.05 was considered statistically significant.

The Burden of OA in China, 1990-2019

Table 1 illustrates the incidence, prevalence, and DALY of OA, knee OA, hip OA, hand OA, and other OA by gender between 1990 and 2019. As shown in Table 1, from 1990 to 2019, the respective numbers of incidence and prevalence of OA increased exponentially from 4.6 million to 10.7 million, 51.8 million to 132.8 million, while the number of DALY increased from 1.8 million to 4.7 million.

In 2019, the age-standardized incidence (ASIR), age-standardized rates of prevalence (ASPR), and DALY for OA were 509.8, 6330.1, and 224.8, respectively. From 1990 to 2019, the ASR of OA have increased significantly. The EAPCs for ASIR, ASPR, and DALY were 0.36, 0.35, and 0.4, respectively. EAPC is a concise and commonly utilized measure of the trend of ASR over a particular period, as we discussed in the previous paragraphs. We can assume that the ASR trending downward throughout the EAPC is less than 0, and upward throughout the EAPC is more than 0. Except for hand OA, all OA categories had an EAPC larger than 0 in both genders, suggesting an upward trend in ASR for most OA from 1990 to 2019. In contrast, in hand OA, only men had EAPCs below 0 for DALY, ASPR, and ASIR, at -0.93, -0.94, and -0.96.

Female patients’ numbers have climbed by 159% over the past 30 years, while the rate of new cases has increased at a rate of 136.2%, as opposed to 153.1% and 127.7% for males. In comparison to males, women had EAPCs for ASIR, ASPR, and age-standardized DALY of 0.46 (0.38-0.54), 0.46 (0.37-0.54), and 0.56 (0.34-0.78), which were all higher than the corresponding values for males of 0.2 (0.1-0.29), 0.18 (0.08-0.27), and 0.19 (0.08-0.3).

During the past 30 years, female incidence have grown by 136.2% and prevalence by 159%, compared to 127.7% and 153.1% for men. Women had EAPCs for ASIR, ASPR, and DALY of 0.46, 0.46, and 0.56, compared to 0.2 , 0.18, and 0.19.

Proportion of disease burden by OA category in China, 1990-2019

In Fig 1, we described the percentage of the incidence, prevalence, and DALY for each type of OA in 1990 and 2019, the largest proportion of people have knee OA. As indicated in Fig 1(a), knee OA accounted for 80% of incidence in 1990, followed by hand OA at 10%, OA other at 8%, and OA hip at 2%. Fig 1(d) demonstrated that the makeup of OA incidence in 2019 is almost identical from 1990, with knee OA (80%) and hip OA (2%). As illustrated in Fig 1(d), in 1990, knee accounted for 75% of all OA prevalence, followed by hand OA at 12% and other OA at 10%. As shown in Fig 1(e), knee OA accounted for 73% of the prevalence in 2019, down 2% from 1990. OA hand and other grew by 1%, while OA hip remained stable. In contrast, as shown in Fig 1(c), the distribution of DALY in 1990 was 75% for knee OA, 12% for OA hand, 10% for OA other, and at least 3% for OA hip. As shown in Fig 1(f), the distribution of DALY in 2019 was not significantly different from that in 1990.

Trends in OA burdens, 1990 to 2019

The burden of total OA disease distribution by age in 1990 and 2019 is shown in Fig 2(a). OA prevalence is 0 before 30 years and rises with age, peaking at 50–54 years. Except for the 75-79 age range, OA incidence declines with age. Nonetheless, prevalence rises progressively with age, peaking at 90-94 years (95+ for women). The age-standardized DALY also increase with age, reaching a maximum in the 80-84 age group (85-89 for women). The graphs of prevalence, incidence, and DALY are similar in 1990 and 2019, but the graph on the left is slightly larger than the one on the right, suggesting that the number of women with prevalence, new cases, and new DALY is slightly larger than the one on the right. Women have more incidence, prevalence and DALY than men.

As shown in Fig 2(b), hand OA incidence, prevalence, and DALY changes between 1990 and 2019 are similar. Like a symmetrical triangle, DALY and prevalence grow with age for both men and women. Nonetheless, incidence trends differently. In 1990, female incidence initially rises with age and peaks at 50-54, then steadily drops until it reaches its lowest value in the 70-74 age group, then rises again. In 1990, males had a similar incidence tendency to women, however the first peak value was in the 55-59 age group and the second low value was in the 80-84 age group. In 2019, incidence is similar to 1990. The graph shows that men and females had similar prevalence, DALY, and incidence of hand OA.

As depicted by Fig 2(c), in 1990, female hip OA DALYs steadily grew from the 30-34 age group to a maximum at 95+ years of age, while men followed a similar pattern. By 2019, DALY for women peak in the 85-89 age range and then stay practically flat, whereas DALY for males peak in the 80-84 age group and remain almost constant. Hip and knee OA incidence rise and fall. Between 1990 and 2019, hip OA incidence for men and women grew from the 30-34 age group to a high in the 60-64 age group, then steadily reduced. Hip OA rates have steadily climbed from 30 to 95+. Unlike knee OA, the yellow graph is much bigger than the blue graph in hip OA, demonstrating that males had a higher incidence, new cases, and DALY than women.

Fig 2(d) showed the trends in knee OA and total OA are generally consistent. Age gradually increases and decreases prevalence, incidence, and DALY. The peak age of knee OA is inconsistent, unlike total OA. For women, the greatest DALY was in the 80-84 age range in 1990 and in the 75-79 age range in 2019. For males, it was continuously in the 65-69 age range. Knee OA was more common in men and women aged 50–54 from 1990 to 2019. Between 1990 and 2019, the greatest age range for men and women was 80-84. Fig 2(a) showed that the graph on the left is significantly larger than the graph on the right, which indicates that women had greater incidence, prevalence, and DALY for knee OA than males.

Other OA prevalence and DALY increase with age and show a left-right symmetrical pattern (Fig 2(e)), similar to hand OA. Incidence varies. In 1990, incidence rose from 30-34 years for both men and women, peaking at 50-59 years, then falling and stabilizing after 65-69 years. In the other OA, the incidence patterns in 2019 and 1990 are almost same and the graphs on both sides are essentially symmetrical, suggesting that men and females had similar prevalence, DALY, and incidence.

Predictions of the OA disease burden

Bayesian APC modeling using bamp package predicts China's OA prevalence from 2020 to 2035. Incidence rates were computed using a 5-year diagnostic period and 5-year age groups for all ages. World standard population ASIR per 100,000 person-years was determined. Log-linear age-smoothed exponential growth and limited linear trend prediction to a cyclical cohort model match existing patterns in predicting new cases and incidence rates by sex and age from 1990 to 2035.

Fig 3(a-c) revealed male and female 5-year ASIRs. The total OA 5-year total ASIR rose yearly from 1995-1999 to 612.8 in 2015-2019, with the greatest growth rate in 2000-2004. We expect the total OA 5-year total ASIR to steadily diminish after 2020-2024, reaching 609 in 2030-2035. The overall trend matches total OA since knee OA accounts for about 70% of the prevalence. It is noteworthy to observe that the 5-year total ASIR for hand OA peaked at 56.7 in 2005-2009 and has slowly declined since then, reaching 48.6 in 2030-2035 according to our forecasts. Hip OA and other OA are closely connected, and the model predicts that both will continue to rise, reaching 11.8 and 42.4 in 2030-2035.

In the 2015-2024 era, men had a significantly higher 5-year total ASIR drop than females for OA. Males changed 4.6% and females 4.3%. Females will have a larger 5-year total ASIR reduction in knee OA than men. According to the projected graph, male 5-year total ASIR declines by 0.73% from 517.8 in 2020-2024 to 514 in 2030-2035, while female ASIR decreases by 1.1% from 508 in 2015-2019 to 502.6 in 2030-2035.

Risk factor

Table 2 provided the 2019 DALY contribution of three primary OA risk variables by year, 5-year age interval, and gender. OA was most associated with high BMI. High BMI accounted with 11.58% of 2019 OA DALY, up from 5.3% in 1990. The high body mass index (BMI) proportion of the population climbs progressively after 30 years, peaking at 13.97 percent at 50-54 years, and declining year afterwards. By gender, the proportion of men with a high BMI in OA is 9.86%, while for women it is as high as 12.76%.

Regardless of the type of OA, the trend is consistent with total OA. Knee OA, which accounts for the highest proportion of arthritis, had a high BMI contribution to total DALY from 7.1% in 1990 to 15.7% in 2019. Looking at the age subgroups, knee OA likewise starts to rise gradually from the age of 30 years upwards in terms of high BMI, reaching a peak of 17.4% at the age of 50-54 years, and then declines gradually. A gender analysis shows it is higher for women at 16.27% than for men at 14.77%.

Hip OA provided a substantially lesser proportion of total DALY than knee OA, with its high BMI percentage rising from 2.26% in 1990 to 5.1% in 2019 and again reaching 5.9% at 50-54. Nevertheless, BMI to total DALY for hip OA was similar from 35-69 years old, ranging from 5-6%.

This thorough study covers the epidemiology and development of OA in China from 1990 to 2019, with forecasts of ASIR values 15 years into the future. Trend analysis and risk factor assessments over 29 years shed light on OA etiology and prevention.

This research describes China's long-term trends in ASIR, ASPR, and OA DALY: The 2019 ASR for OA incidence, prevalence, and DALY was 472.5, 5880.6, and 206.4, up 37.3, 449.5, and 18.4 from 1990. Additional study shows that males over 29 had ASIR, ASPR, and DALY increases of 31.9, 400.5, and 15.7, with growth rates of 8.3%, 8.3%, and 9.3%. Female growth rates were 6.8%, 6.9%, and 8.35%. From 1990 to 2019, women carried more OA illness than men. Nevertheless, males had higher growth rates of ASIR, ASPR, and DALY than women, indicating that men with OA will have a higher disease burden in the future.

OA view. In 1990, knee OA accounted for 75% of the DALY for OA in China, and despite a 1% drop in 2019, it remains the largest contributor to the disease burden. Obesity and population aging may be responsible. China has the most overweight and obese population in the world, with over 50% of adults and 20% of schoolchildren. Studies also reveal that symptomatic OA risk rises by 9–13% each kilogram added.^19,20 And as the knee joint is the major weight-bearing joint in the body, it is not surprising that high BMI is becoming an important cause of DALY due to OA.

Only hand OA had an EAPC below 0 for its age-standardized DALY, ASPR, and ASIR from 1990 to 2019. Hand OA's ASIR is dropping during this period. Additional analysis shows that only men have hand OA EAPC levels below 0, whereas females have values above 0. This shows that ASIR for hand OA is growing in women and decreasing in men, like other forms of OA. The fast economic growth and technological improvement of the last 30 years has replaced many easy and repetitive labor occupations with machines and electronic devices, freeing men's hands and lowering hand OA ASIR. This may explain why hand OA in women has increased ASIR.

Our data demonstrates that women had a larger burden of OA illness linked with high BMI than males, based on prevalence, incidence, and standardized DALY. The literature suggests numerous causes. (1) Post-menopausal estrogen insufficiency increases subcutaneous adipose tissue free fatty acid accumulation, which may enhance OA-induced discomfort. Abdominal subcutaneous fat impacts knee biomechanics and inflammatory response. (2) Biomechanically, estrogen modulates fat distribution, with more estrogen receptors in subcutaneous fat than visceral tissue and more androgen receptors in visceral tissue than subcutaneous tissue. Female subcutaneous fat increases knee joint stress and impact impulse. (3) Women may be more worried about their health than males, so they go to the hospital more often and are more sensitive to joint abnormalities.^21–25

The burden of OA illness is now higher in Chinese women than in men, but the growth rates of ASIR, ASPR, and age-standardized DALY are higher in men. With such growth rates, men will eventually outburden women. Our forecasts show that the 5-year total ASR for men is approximately the same from 2005 to 2009, then begins to surpass that of girls, which is consistent with our earlier findings. After this spike, both sexes show a steady fall in ASR, which we assume is due to the aging population decreasing and the fat population peaking and declining.^26,27

This detailed update to the 2019 GBD report identifies critical risk factors for policymaking. The incidence, prevalence, and DALY of OA in China have generally trended upward since 1990. During the next 15 years, the ASIR of OA in China will progressively slow down and level off due to high BMI and an aging population, which will force the incidence of OA upward. Obesity prevention is crucial. Lifestyle adjustments, notably weight management, may minimize the burden of OA and avoid flare-ups.

XFZ contributed to the conception and design of the study. SZ and JC contributed to acquisition of data, analysis and interpretation of data. HJW and JMZ make contribution to drafting the article. GLJ and ZGZ contributed to revising it critically for important intellectual content. XFZ finally approve the version to be submitted.

Acknowledgement of other contributors

The authors wish to thank all individuals who have contributed to the Global Burden of Disease (GBD) Study 2019.

Conflict of interest

No conflict of interests to declare.

Declaration of Funding

All sources of funding should be declared as an acknowledgement at the end of the text.

Role of the funding source

XFZ was supported by a series of initiatives, including the Clinical Frontier Technology Program of the First Affiliated Hospital of Jinan University, China (No. JNU1AF-CFTP-2022-a01204), the National Key Research and Development Program of China (2022YFE0206200), the Funding by Science and Technology Projects in Guangzhou (202102010079, 202206010158, 202201020087), the Fundamental Research Funds for the Central Universities (21621103), the National Natural Science Foundation of China (82172440), and the Natural Science Foundation of Guangdong Province (2023A1515011553). It should be noted that the opinions expressed in this publication belong solely to the authors and are not necessarily representative of the aforementioned organizations.

Ethics

all cases of this manuscript were collected from the GBD 2019 database, a publicly available and anonymous database

Studies involving humans or animals

Data availability statement: The datasets generated during and/or analysed during the current study are available from the online GBD 2019 database

Garstang, S.V. and T.P. Stitik, Osteoarthritis - Epidemiology, risk factors, and pathophysiology. American Journal of Physical Medicine & Rehabilitation, 2006. 85(11): p. S2-S11.
Singh, A. and B. Antony, Response to the letter to editor by Chen et al. regarding: Burden of osteoarthritis in India and its states, 1990-2019: findings from the Global Burden of Disease Study 2019. Osteoarthritis and Cartilage, 2022. 30(10): p. 1413-1414.
Forouzanfar, M.H., et al., Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet, 2016. 388(10053): p. 1659-1724.
Vos, T., et al., Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet, 2015. 386(9995): p. 743-800.
Blueher, M., Obesity: global epidemiology and pathogenesis. Nature Reviews Endocrinology, 2019. 15(5): p. 288-298.
Usenbo, A., et al., Prevalence of Arthritis in Africa: A Systematic Review and Meta-Analysis. PLoS One, 2015. 10(8): p. e0133858.
Salih, S. and P. Sutton, Obesity, knee osteoarthritis and knee arthroplasty: a review. BMC sports science, medicine & rehabilitation, 2013. 5(1): p. 25-25.
Berenbaum, F., F. Eymard, and X. Houard, Osteoarthritis, inflammation and obesity. Current Opinion in Rheumatology, 2013. 25(1): p. 114-118.
Safiri, S., et al., Global, regional and national burden of osteoarthritis 1990-2017: a systematic analysis of the Global Burden of Disease Study 2017. Annals of the Rheumatic Diseases, 2020. 79(6): p. 819-828.
James, S.L.G., et al., Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet, 2018. 392(10159): p. 1789-1858.
Chugh, S.S., et al., Worldwide Epidemiology of Atrial Fibrillation A Global Burden of Disease 2010 Study. Circulation, 2014. 129(8): p. 837-847.
Gao, T., et al., Disability adjusted life year (DALY): A useful tool for quantitative assessment of environmental pollution. Science of the Total Environment, 2015. 511: p. 268-287.
Murray, C.J. and A.D. Lopez, Global mortality, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet (London, England), 1997. 349(9063): p. 1436-42.
Murray, C.J.L., et al., Disability-adjusted life years (DALY) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet, 2012. 380(9859): p. 2197-2223.
Fosse, E. and C. Winship, Analyzing Age-Period-Cohort Data: A Review and Critique, in Annual Review of Sociology, Vol 45, K.S. Cook and D.S. Massey, Editors. 2019. p. 467-492.
Naghavi, M., et al., Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet, 2015. 385(9963): p. 117-171.
Liu, Q., et al., Changes in the global burden of depression from 1990 to 2017: Findings from the Global Burden of Disease study. J Psychiatr Res, 2020. 126: p. 134-140.
Liu, Z., et al., The trends in incidence of primary liver cancer caused by specific etiologies: Results from the Global Burden of Disease Study 2016 and implications for liver cancer prevention. J Hepatol, 2019. 70(4): p. 674-683.
Barba, C., et al., Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet, 2004. 363(9403): p. 157-163.
Hunter, D.J. and S. Bierma-Zeinstra, Osteoarthritis. Lancet, 2019. 393(10182): p. 1745-1759.
Costa-Font, J. and M. Gyori, The weight of patriarchy? Gender obesity gaps in the Middle East and North Africa (MENA). Social Science & Medicine, 2020. 266.
Huh, D., et al., Female Overweight and Obesity in Adolescence: Developmental Trends and Ethnic Differences in Prevalence, Incidence, and Remission. Journal of Youth and Adolescence, 2012. 41(1): p. 76-85.
Wang, L., et al., Regional specific differences in prevalence of overweight/obesity in China: findings from China Kadoorie Biobank study in 10 areas in China. Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi, 2015. 36(11): p. 1190-4.
Afshin, A., et al., Health Effects of Overweight and Obesity in 195 Countries over 25 Years. New England Journal of Medicine, 2017. 377(1): p. 13-27.
You, W. and M. Henneberg, Relaxed natural selection contributes to global obesity increase more in males than in females due to more environmental modifications in female body mass. Plos One, 2018. 13(7).
Li, H., T. Zhou, and C. Jia, The influence of the universal two-child policy on China's future population and ageing. Journal of Population Research, 2019. 36(3): p. 183-203.
Chooi, Y.C., C. Ding, and F. Magkos, The epidemiology of obesity. Metabolism-Clinical and Experimental, 2019. 92: p. 6-10.

Table 1 The incidence, prevalence and age-standardized DALY of OA in China by gender, between 1990 and 2019

OA Types	Sex	Incidence (95% UI)					Prevalence (95% UI)					DALY (95% UI)
OA Types	Sex	Count (1990)	ASR (1990) per 100,000	Count (2019)	ASR (2019) per 100,000	EAPC	Count (1990)	ASR (1990) per 100,000	Count (2019)	ASR (2019) per 100,000	EAPC	Count (1990)	ASR (1990) per 100,000	Count (2019)	ASR (2019) per 100,000	EAPC
OA	Both	4590967.2 (4044932-5178959.7)	472.5 (417.5-531.4)	10681310.8 (9375594.2-12079546)	509.8 (450.7-573.7)	0.36 (0.29-0.44)	51760492 (45946299.7-58263041.5)	5880.6 (5247.8-6570.4)	132807005.7 (117994900-149081986.9)	6330.1 (5657.3-7085.4)	0.35 (0.28-0.42)	1819337.7 (903102.1-3648573.9)	206.4 (103.1-414.4)	4724884.5 (2347242.7-9536082.4)	224.8 (112.4-452.3)	0.4 (0.32-0.48)
	Male	1917756.6 (1680159.6-2182078.7)	384.9 (339.8-436.4)	4367685.2 (3826418-4963492.4)	416.8 (367.1-471.1)	0.2 (0.1-0.29)	21277932.3 (18850028.9-23951455.3)	4844.7 (4325.1-5418.7)	53860985.1 (47758494-60622830.8)	5245.2 (4680.4-5900.4)	0.18 (0.08-0.27)	742590.9 (370776.8-1475681.8)	168.6 (84-336.3)	4724884.5 (2347242.7-9536082.4)	184.3 (91.8-370.7)	0.19 (0.08-0.3)
	Female	2673210.6 (2359382.9-2993781.4)	564.6 (498.2-633.4)	6313625.6 (5538885.4-7118134.4)	603.2 (530.5-675.7)	0.46 (0.38-0.54)	30482559.8 (27125647.1-34200961.7)	6888.8 (6150.7-7714.2)	78946020.6 (70244562.7-88677086.1)	7366.4 (6571.1-8254)	0.46 (0.37-0.54)	742590.9 (370776.8-1475681.8)	243.1 (121.4-490)	2827915.1 (1409988.8-5745787.5)	263.4 (132-535.2)	0.56 (0.34-0.78)
Hand	Both	438451.1 (327147.5-592527.2)	45.7 (34.5-61.4)	1097964.8 (815105.3-1482495.8)	52.7 (39.7-70.5)	-0.27 (-0.65-0.11)	6989896.9 (5054130.5-9635407.6)	828 (604-1130.2)	19637407.6 (14317681.6-26649670.3)	953.8 (701.2-1284.6)	-0.15 (-0.49-0.2)	222254.4 (108156.5-469745.9)	26.1 (12.9-54.4)	623067.1 (304032.9-1298390.3)	30.1 (14.7-62.8)	-0.14 (-0.48-0.2)
	Male	221563.7 (163956-304993.4)	45.1 (33.7-61.1)	558828.9 (412325.9-774392.2)	53.2 (39.9-72.4)	-0.96 (-1.61--0.32)	3309589.4 (2371071.1-4589826)	805.5 (591.1-1102.7)	9365722.4 (6861093.8-12639309.6)	942.6 (698.3-1283.1)	-0.94 (-1.55--0.32)	106199.8 (51710.3-225850.8)	25.6 (12.6-53.7)	299558.4 (144668.8-630898.5)	30 (14.6-63.7)	-0.93 (-1.54--0.31)
	Female	216887.4 (162084.4-290327.8)	46.1 (34.9-61.5)	539135.9 (401510.2-730203.8)	52.1 (39.3-69.3)	0.49 (0.29-0.68)	3680307.4 (2667092.7-5078701.7)	855.2 (622.8-1180.4)	10271685.2 (7461795.2-13831518.8)	968.1 (709.8-1300.4)	0.55 (0.33-0.76)	116054.6 (56701.1-239861.4)	26.8 (13.2-55.2)	323508.7 (158255.2-684800.1)	30.4 (15-64.4)	0.56 (0.34-0.78)
Hip	Both	81539.8 (59898.5-106021.1)	8.3 (6.1-10.8)	243109.5 (177943.7-319067.9)	11.4 (8.4-14.9)	1.25 (1.14-1.35)	1503678.4 (1148134.4-1911933.6)	172.5 (132.2-218.9)	4787069.3 (3667374.2-6064140.1)	234.1 (179.8-297)	1.16 (1.06-1.26)	48358.1 (22107-98907.3)	5.5 (2.5-11.3)	153024.7 (70238.4-319576)	7.5 (3.4-15.5)	1.16 (1.06-1.26)
	Male	81539.8 (59898.5-106021.1)	8.9 (6.5-11.6)	131749.1 (96482.9-173132.2)	12.4 (9.1-16.1)	1.23 (1.15-1.31)	794577 (604207.3-1003589.5)	185 (141.9-234)	2517333.9 (1927047.1-3208627.4)	253.4 (195.2-321.2)	1.14 (1.06-1.22)	25794.9 (11836.7-52681)	5.9 (2.7-12.2)	81129.2 (36960.3-168843.2)	8.1 (3.7-16.9)	1.14 (1.06-1.22)
	Female	36796.7 (26896.1-47796.7)	7.7 (5.6-10)	111360.4 (81141.4-146693.3)	10.5 (7.7-13.7)	1.29 (1.15-1.43)	709101.5 (538823.4-899786.5)	161.2 (123.3-205.1)	2269735.4 (1734404.9-2870992.5)	216.4 (165.8-274.2)	1.18 (1.05-1.31)	22563.2 (10314.5-46242.6)	5.1 (2.3-10.4)	71895.5 (33315.8-148629.3)	6.8 (3.2-14.1)	1.18 (1.06-1.31)
Knee	Both	3683886.1 (3175286.6-4208424)	377.2 (326.8-429.8)	8425777.1 (7233585.4-9675256.2)	402.3 (347.2-459.9)	0.45 (0.35-0.55)	42570836.3 (36337788.6-49302893.1)	4797 (4111.6-5535.6)	108120120.3 (91899240.7-125849501.1)	5125.5 (4385.1-5925.2)	0.48 (0.37-0.6)	1365167.2 (671564.5-2805799.8)	152.7 (75.3-312.9)	3464753.2 (1688168.2-7064613.5)	163.7 (80.1-333.9)	0.5 (0.38-0.61)
	Male	1442949.3 (1234966.9-1663580.3)	287.4 (246.6-329.3)	3193112.7 (2720383.8-3690619.2)	305 (262-350.2)	0.37 (0.25-0.49)	15825667.2 (13418655.2-18339941.7)	3521.9 (3008.8-4059.5)	38953941.9 (32878748.7-45456914.9)	3743.8 (3182.6-4347.7)	0.4 (0.27-0.54)	513592.7 (250457.6-1056793.5)	113.4 (55.5-232.4)	1260941.9 (610715.2-2586024.5)	120.7 (58.9-247.7)	0.41 (0.27-0.55)
	Female	2240936.8 (1949238.9-2551157)	471.8 (410.4-537.2)	5232664.4 (4514246.4-5989180.3)	500 (432.7-569.9)	0.46 (0.37-0.55)	26745169.2 (22904337.3-30791888.6)	6017.2 (5163.4-6954.8)	69166178.4 (59077673.6-80018657.7)	6430.8 (5513.8-7423.6)	0.52 (0.41-0.62)	851574.5 (421598.2-1743510.8)	190.7 (94.5-389.7)	2203811.3 (1079034.2-4486984.6)	204.6 (100.4-419)	0.53 (0.43-0.64)
Other	Both	387090.2 (301842-480936.4)	41.4 (32.7-51.1)	484039.5 (234460.1-1006719.9)	43.4 (34.2-53.7)	0.18 (0.17-0.2)	5784078 (4284793-7448527.8)	702.4 (531.1-890.4)	15280407.1 (11352173.7-19624729.2)	745.8 (563.1-950.3)	0.23 (0.22-0.25)	183558 (88122.4-378717.6)	22.1 (10.8-46)	914459.4 (710382.8-1145920.1)	23.5 (11.5-49.1)	0.24 (0.22-0.25)
	Male	208500.5 (160828.6-261781.1)	43.5 (34.4-53.8)	483994.5 (372941-610941.2)	46.2 (36.4-57.6)	0.22 (0.21-0.24)	3025309.1 (2227270.2-3934933.8)	749.9 (568.9-954.9)	7984316.1 (5891064.5-10367338.8)	805.7 (605.8-1033.9)	0.27 (0.25-0.29)	97003.5 (46516.2-201254.5)	23.8 (11.5-50)	255339.9 (123192.5-529035.9)	25.6 (12.4-53.5)	0.27 (0.25-0.29)
	Female	178589.7 (140946.7-220316.8)	39 (30.9-48.2)	430464.9 (338864.2-535100.5)	40.6 (32-50)	0.16 (0.14-0.17)	2758768.9 (2069884.3-3513752.5)	657.6 (499.6-829.3)	7296091 (5474002.3-9268653.7)	689.8 (520.9-873.4)	0.19 (0.17-0.21)	86554.5 (41892.9-178889.1)	20.5 (10-42.6)	228699.6 (111335.3-474821.2)	21.6 (10.6-44.8)	0.2 (0.19-0.22)

ASR = age-standardized rates; EAPC = estimated annual percentage changes; OA = osteoarthritis; UI = uncertainty interval.

Table 2 The percentage of total DALYs from risk factors for OA, hip OA, and knee OA in China by year, by 5-year age group, and by gender in 2019.

Variables	High body mass index (95%UI)
Variables	OA	Hip OA	Knee OA
Year
1990	0.05 (0.13-0.01)	0.02 (0.06-0.00)	0.07 (0.17-0.02)
1995	0.06 (0.13-0.01)	0.03 (0.06-0.01)	0.08 (0.18-0.02)
2000	0.07 (0.14-0.02)	0.03 (0.07-0.01)	0.09 (0.2-0.03)
2005	0.08 (0.16-0.02)	0.03 (0.07-0.01)	0.11 (0.22-0.03)
2010	0.09 (0.18-0.03)	0.04 (0.08-0.01)	0.13 (0.24-0.04)
2015	0.11 (0.2-0.04)	0.05 (0.09-0.02)	0.14 (0.26-0.05)
2019	0.12 (0.21-0.05)	0.05 (0.1-0.02)	0.16 (0.28-0.06)
Age Groups
<30	0	0	0
30-34	0.09 (0.17-0.03)	0.05 (0.09-0.02)	0.13 (0.25-0.05)
35-39	0.12 (0.22-0.05)	0.05 (0.1-0.02)	0.15 (0.27-0.06)
40-44	0.13 (0.23-0.05)	0.05 (0.1-0.02)	0.16 (0.28-0.06)
45-49	0.13 (0.24-0.05)	0.06 (0.11-0.02)	0.17 (0.3-0.07)
50-54	0.14 (0.25-0.06)	0.06 (0.11-0.02)	0.17 (0.31-0.07)
55-59	0.13 (0.24-0.05)	0.06 (0.11-0.02)	0.17 (0.31-0.07)
60-64	0.12 (0.22-0.05)	0.06 (0.1-0.02)	0.17 (0.3-0.07)
65-69	0.11 (0.21-0.04)	0.05 (0.1-0.02)	0.16 (0.29-0.06)
70-74	0.1 (0.18-0.04)	0.05 (0.09-0.02)	0.15 (0.27-0.06)
75-79	0.09 (0.17-0.03)	0.04 (0.08-0.02)	0.13 (0.25-0.05)
80-84	0.07 (0.14-0.02)	0.03 (0.07-0.01)	0.11 (0.22-0.03)
85-89	0.07 (0.14-0.02)	0.04 (0.07-0.01)	0.11 (0.22-0.03)
90-94	0.07 (0.14-0.02)	0.04 (0.08-0.01)	0.11 (0.22-0.03)
95+	0.07 (0.14-0.02)	0.04 (0.08-0.01)	0.11 (0.23-0.03)
Gender
Both	0.12 (0.21-0.05)	0.05 (0.1-0.02)	0.16 (0.28-0.06)
Male	0.1 (0.19-0.03)	0.05 (0.1-0.02)	0.15 (0.28-0.05)
Female	0.13 (0.24-0.04)	0.05 (0.11-0.02)	0.16 (0.31-0.06)

UI = uncertainty interval.

No competing interests reported.

The burden, trends, and risk factors for osteoarthritis in China from 1990 to 2019 and its predictions until 2035

Status:

Version 1

Abstract

Figures

Introduction

Methods

Data source

Disease burden

Prediction model

Risk factor

Statistical analysis

Results

The Burden of OA in China, 1990-2019

Proportion of disease burden by OA category in China, 1990-2019

Trends in OA burdens, 1990 to 2019

Predictions of the OA disease burden

Risk factor

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Status:

Version 1