Bulletin of the World Health Organization

Twenty-year trends in the prevalence of disability in China

Xiaoying Zheng a, Gong Chen a, Xinming Song a, Jufen Liu b, Lijing Yan c, Wei Du a, Lihua Pang a, Lei Zhang a, Jilei Wu a, Bingzi Zhang a & Jun Zhang d

a. Institute of Population Research/WHO Collaborating Centre on Reproductive Health and Population Science, Peking University, Beijing 100871, China.
b. Institute of Reproductive and Child Health, Peking University, Beijing, China.
c. China International Centre for Chronic Disease Prevention, Beijing, China.
d. China Disabled Persons’ Federation, Beijing, China.

Correspondence to Xiaoying Zheng (e-mail: xzheng@pku.edu.cn).

(Submitted: 28 April 2011 – Revised version received: 03 July 2011 – Accepted: 03 July 2011 – Published online: 14 July 2011.)

Bulletin of the World Health Organization 2011;89:788-797. doi: 10.2471/BLT.11.089730


Disability is common in both developed and developing countries.1 An estimated 650 million people worldwide live with some form of disability.2 The way that disability is conceived has changed substantially in recent years, the current focus being on three areas of impairment: bodily functions and structures, activities and participation.2 In this study, disability is defined as one or more abnormalities in anatomical structure or the loss of a particular organ or function (either physical or psychological) affecting a person’s ability to carry out a normal activity and to participate fully in study, work, and community and social life. In the United States of America, about 50 million adults suffer from various forms of disability, including hearing loss, visual impairment, cognitive impairment and limited mobility.3 In China, about 85 million people have a disability or another condition that affects their daily lives and social activities.4 Furthermore, disability can lead to limited access to education, health care and rehabilitation services, to the detriment of the disabled people, their families and local communities, health-care systems and social security systems.1,5 Therefore, specific strategies, policy initiatives and sustainable programmes are needed to improve the health status of the disabled population.

Several studies have explored trends in disability among different populations. According to prevalence estimates, in the United States 51.2 million people aged 6 years or older (18.1% of the age group) reported having a disability in 2002, compared with 54.4 million (18.7%) in 2005.6 Freedman et al.7 systematically reviewed cross-sectional and cohort studies and reported a general decrease in the age-adjusted prevalence of disability in the population aged 65 years of age or older in the United States. A similar decrease in disability prevalence was reported for Spain’s population aged 65 years or older.8 In terms of life expectancy, the Netherlands reported an increase in the number of years lived with minor disabilities9 and Japan reported a decrease in the number of years lived with severe disabilities.10 These mixed findings may be attributable to different indicators, disability contexts and study populations.11

In China, two large-scale nationally representative household surveys were conducted from 1 April to 30 May in 1987 and 2006.12,13 The resulting figures have been used in epidemiological studies to conduct specific analyses of different types of disabilities,1421 rural/urban disparities in disability prevalence4 and disability-free life expectancy.22 Using different survey series, Phillips et al.23 described the prevalence of mental disorders in four provinces, and Chou and Leung24 evaluated self-reported disability among the elderly in Hong Kong Special Administrative Region. These studies have primarily provided a demographic profile of people with disability in China. However, there is no scientific evidence of the magnitude of the problem of disability across the entire nation that will allow for complex survey designs and age-adjusted analysis of trends in disability prevalence nationwide.

China has undergone rapid social, economic, political, institutional and demographic transitions. Thus, understanding the magnitude of the burden of disability and its trends over time is essential for improving the health of the population and enhancing the lives of people with disabilities.

In this study, our primary objective was to assess changes in the prevalence of disability in China during the 20-year period from 1987 through 2006. We also aimed to investigate any changes in the leading causes of disability that may have occurred over that time and that could potentially explain the changes in overall prevalence. Since 80% of all those who are disabled live in developing countries,25 most of which are going through an epidemiologic and sociodemographic transition, the findings from this study may be broadly reflective of disability trends in the developing world at large.


Data source

We obtained data from the 1987 and 2006 national surveys of the non-institutionalized population of China.12,13 Both surveys used multistage, stratified random cluster sampling, with probability proportional to size, to derive nationally representative samples. The surveys were approved by the State Council and conducted in all province-level administrative regions of mainland China by the Leading Group of China National Sample Survey on Disability and the National Bureau of Statistics. Within each region, sampling strata were defined based on subordinate administrative areas, local geographical characteristics or local gross domestic product, where appropriate, to allow for anticipated regional variability. Within each stratum, a four-stage sampling strategy was followed involving four natural administrative units (i.e. county, town, village and community), and sampling was conducted with probability proportional to cluster size. The sampling interval (i.e. the total population divided by the number of units at each stage) used the most up-to-date population and address information from the Ministry of Civil Affairs and Public Security in Beijing. Both surveys excluded the institutionalized population and together comprised a total of 424 counties (3169 communities) in 1987 and 734 (5964) in 2006.12,13 The final sample size was 1 579 316 in 1987 and 2 526 145 in 2006, figures representing 1.5 and 1.9 per 1000 non-institutionalized inhabitants of China, respectively. All survey respondents provided consent to participate to the Chinese government.

The survey protocol and survey questions were reviewed by leading national and international experts, and the sampling scheme was reviewed by experts from the Division of Statistics of the United Nations.12,13 A standardized protocol was adopted in both years to ensure the comparability of the survey results. Two pilot studies were conducted in different provinces before each survey. Strict quality control measures were implemented at every step during each survey, from the drafting of the sampling frame through field sampling, from the filling out of the questionnaires to the checking of the returned forms, and from data input to the checking of data quality.12,13


In both 1987 and 2006, trained field interviewers used a structured questionnaire to inquire about visual, hearing and speech disability, physical or intellectual disability and mental disability (Table 1). Those who responded “yes” to any of the corresponding questions and all children aged 6 years or less were referred to different designated physicians for further disability screening and confirmation. A designated physician performed medical examinations and followed diagnostic manuals to make the final diagnosis and assess the severity of the disability, if any, and to confirm its primary causes. Respondents with multiple positive answers were examined by multiple specialists (a separate doctor for each disability). The primary causes of each disability were recorded.

Statistical analyses

All data were entered into a custom-designed database and analysed using SAS Version 9.2 (SAS Institute, Cary, USA). Allowing for the complex sampling design, we constructed sample weights using standard weighting procedures.26 For each cause we calculated the weighted proportion of its contribution to a particular disability, and we ranked the proportions thus obtained to determine the five leading causes of each disability. We also calculated separately a given cause’s contribution to different disabilities. We used the SURVEYFREQ procedure27 to estimate the weighted prevalence of disability of various types, with 95% confidence intervals (CIs), for the overall population and for different population segments.

Allowing for changes in the age structure of the population over time, we calculated the age-adjusted prevalence of disability through direct standardization using the 1990 census-derived Chinese population as the standard.28 We estimated the annual change in age-standardized prevalence and its associated 95% CIs using the two-point percentage change annualized estimator (PCAE), which is based on the first and last rate only, not the intervening rates, but is robust whether the annual change in rate is linear or not.29 A PCAE-associated P-value < 0.05 was considered indicative of statistical significance.30


The 1987 and 2006 surveys were nationally representative. Compared with the 1987 study population, the 2006 study population included more adults aged 18–64 (65.7% versus 57.8%), more seniors aged 65 years or older (9.9% versus 5.7%), more males (50.7% versus 50.5%) and more urban residents (33.5% versus 15.6%).

In both surveys most disability occurred among rural residents (Table 2). Hearing and speech disabilities were predominant in both years and accounted for 44.5% of all disability in 1987 and for 38.0% in 2006 (Table 2).

Table 2 shows that in 1987 the crude number of disabled people was highest in the population aged 18–44 years. Almost 20 years later, the largest estimated number of disabled people had shifted from this group to that of people aged 65–74 years (12.9 million versus 18.8 million, respectively). Despite this, an estimated 17.0 million people aged 18–44 years had a disability in 2006 (Fig. 1). The weighted prevalence of disability increased with age in 1987, and this monotonic pattern persisted in 2006 among people 7 years of age or older (Fig. 1).

Fig. 1. Estimated number of disabled people and weighted prevalence of disability, by age group, China, 1987–2006
Fig. 1. Estimated number of disabled people and weighted prevalence of disability, by age group, China, 1987–2006

Between 1987 and 2006, the estimated number of disabled people rose from 52.7 million to 84.6 million. This corresponds to a weighted prevalence of 4.9% in 1987 and 6.5% in 2006 (Table 3). The overall age-adjusted prevalence of disability decreased significantly, by an annual average of 0.5% over the 20-year period (Table 3). It increased significantly for males and rural residents but decreased significantly for females and urban residents and in the areas of visual disability, hearing and speech disability, intellectual disability, physical disability and mental disability (Table 3).

The 20-year upward or downward trend in the prevalence of visual, hearing and speech, physical and intellectual disability did not vary by sex or place of residence (urban versus rural) (Table 4). The prevalence of mental disability increased significantly over time among males, females and rural residents but not among urban residents (Table 4).

Fig. 2 shows changes in the five leading causes of disability over time. For example, some medical conditions, such as pyogenic infection and meningitis, contributed less to their associated disabilities in 2006 than in 1987. Instead, new emerging medical causes of disability became more prevalent in 2006. Osteoarthritis and workplace injury, for example, became greater contributors to physical disability, and asphyxia to intellectual disability.

Fig. 2. Leading classified causes of each type of disability and corresponding weighted proportion, China, 1987 and 2006
Fig. 2. Leading classified causes of each type of disability and corresponding weighted proportion, China, 1987 and 2006
a New classified causes introduced into the 2006 survey.


This study demonstrates that many people suffer disabilities, and their numbers will increase because of the rapid changes occurring in everyday life in China. Although the prevalence of disability in China (6.5%) is much lower than in countries such as the United States (18.7%), more people with disabilities would probably be identified in China if national surveys measured disability in terms of the activities of daily living and instrumental activities of daily living, instead of the more narrowly defined impairments confirmed by a physician examination. The current findings could be only the tip of the iceberg; hidden beneath the surface could be millions of other lives affected by disability in China.

The weighted prevalence of disability increased over 20 years in China, as it did in the United States.6 This finding could be attributable primarily to the increase in the life expectancy of the disabled population over time.4,9,10,22 After controlling for such an ageing effect, we found that the adjusted prevalence of disability has decreased significantly over time in China. This finding is consistent with the decreasing trend observed in people aged 65 years and over in the United States7 and Spain8 and fills a knowledge gap in the existing literature on China.

Trends in disability in different population segments have been studied widely. The present study identified an increase in the age-adjusted prevalence of disability among males and an opposite trend among females in China. Such a disparity in disability by sex is not in line with previous findings.68 After systematically reviewing published studies that assessed disability trends among Americans aged either 65 years or older or 70 years or older, Freedman et al.7 indicated that the evidence for such disparity by gender and race was limited and mixed. Sagardui-Villamor et al.8 observed a greater reduction in disability prevalence among males aged 65 and over than among females in the same age group in Spain. Brault6 used United States census data to report that more females (prevalence: 20.1%) than males (prevalence: 17.3%) had some form of disability in 2005, whereas the 2006 survey in China showed a weighted prevalence of disability of 6.6% in males and 6.4% in females. Using life expectancy indicators for 2004, Hashimoto et al.10 reported more years without disability among females (73.0 years) than among males (69.7 years) in Japan. This similarity between Japan and China, both of which differ from the United States, suggests that the sex disparity in disability could be the result of cultural influences. Furthermore, Whitson et al.31 analysed cross-sectional data from 5888 community-dwelling people and suggested that sex-based causal differences (e.g. a higher prevalence of obesity in women) might explain 12.9% of the disparity in disability between the sexes. Similarly, it is possible that the dissimilar distribution of causal factors in males and females in the present study explains the observed disparity by sex. Nevertheless, such observed disparity highlights the need for further investigation among different population segments.

This study revealed that both physical and mental disabilities have increased significantly in China over the past two decades, whereas disabilities in vision, hearing and speech as well as intellectual disabilities have decreased significantly. This finding is consistent with those of previous descriptive Chinese studies.15,1821,23 In a study that examined the epidemiological distribution of physical disability, Luan and Liu21 reported that its prevalence varied among different regions in China. In describing the epidemiological profile of disability associated with mental disorders in four provinces of China, Phillips et al.23 indicated that this problem is being overlooked and suggested an urgent analysis of the situation. Like these studies, the current one enriches the literature by providing data on the magnitude of the change in the burden of different disabilities over time. Because mental disability is more prevalent among socioeconomically disadvantaged populations32 and physical disability is more prevalent among rural populations,21 the current findings alert experts to the importance of taking actions specifically designed for each target population to relieve the burden of disability in the community.

This study also demonstrates changes over time in the leading causes of disability in China. The fact that fewer infectious and genetic diseases were causes of disability in 2006 than in 1987 may point to improvements in public health in China during the 20-year period. For example, polio vaccination schemes33 may have contributed to a reduction in poliomyelitis, which causes physical disability. However, the finding of increasing trends in physical disability may further imply that the success of public health efforts such as polio vaccination could be outweighed by new challenges stemming from rapid modernization, industrialization and urbanization in transitional China, such as workplace injuries and other types of injuries.34 In 2004, about half a million industrial fatalities occurred in China, a figure representing one fourth of the fatalities worldwide.35 Therefore, it is not surprising that workplace injuries have become a leading cause of disability. Since the 1980s, large numbers of rural residents in China have migrated to cities to look for jobs.36 These migrant workers are at high risk of suffering workplace injuries that may result in disability or loss of life owing to a lack of labour force experience and to high financial stress.37 Once migrant workers are disabled due to workplace injury, they are more likely to move back to their rural homes because they become socioeconomically disadvantaged.36 Since increasing disability in the rural population goes hand in hand with a higher risk of physical disability,21 we speculate that injuries to migrant workers may be more or less attributable to the observed increase in disability in the rural population in China. However, we do not have sufficient data to confirm this speculation. Future research may help to determine whether workplace injuries have a greater impact on the rural than on the urban population. Nevertheless, the changes in causality found in this study underline the importance of continuing public-health efforts and developing specific programmes to eliminate various causes of disability.

The reported time trends in disability and the epidemiological transition to different new causal factors for various disabilities are expected to occur in settings in transition where health-care system reform, rapid economic growth and population ageing are taking place, as they are in China. The current findings may trigger actions to improve prevention and treatment efforts worldwide, especially in developing countries in transition. Furthermore, the findings point to the need to combat emerging challenges in addition to securing the success of population-wide measures such as polio vaccination. For example, occupational disability in China may be reduced through sustained commitment and initiatives to improve workplace safety within a framework of primary prevention (e.g. job training and safety education), secondary prevention (e.g. workplace hazard identification and prompt emergency response) and tertiary prevention (e.g. pharmacological treatment to delay complications).

The definition of disability and its operational measurement vary greatly among studies.710,23,24 In studies that rely on self-reported disability, uncontrolled factors such as different levels of participation in disability education and intervention programmes may influence reporting behaviour. Because the current study used medical confirmation instead of self-reporting, its findings are probably robust against this possibility. Moreover, this study was based on two representative nationwide population-based surveys with large sample sizes and standardized quality control measures, resulting in reliable prevalence estimates with small sampling errors.2,13 Although the 1987 survey used the International classification of impairments, disabilities, and handicaps38 and the 2006 survey used the International classification of functioning, disability and health39 to classify disability, both surveys employed the Chinese word canji,12,13 meaning both handicap38 and disability,39 which is consistent with the definition used in this study. A detailed comparison of the methods used in these two surveys is available elsewhere.22

Many factors could have contributed to the observed changes in the prevalence of disability, including a reduction in risk factors and changes in the availability of social security programmes among different populations, since such programmes provide people with the opportunity to get appropriate treatment and avoid disabling outcomes.40 For example, the fact that fewer rural than urban residents are covered by social security programmes because of the hukou (population registration) system4,36 may have resulted in the urban–rural disparity in the prevalence of disability observed in this study. However, the use of cross-sectional data limited our ability to investigate the possible underlying causes of the observed trends. Survey data included extensive demographic information but lacked details on exposure, such as access to medical care. Thus, to control for exposure instead of person–time we used population-level exposure, a somewhat more robust approach given that dramatic changes have taken place across the entire nation. The significant change in the overall prevalence of disability in China may therefore reflect the social, economic and environmental changes taking place in the country.


Although overall prevalence has dropped, the absolute number of disabled people continues to increase in China. There are disparities in rates of disability among certain population segments and new causes of disability are emerging. Thus, the health-care system, the community and individuals themselves are faced with an increased burden of disability and new challenges. Sustained efforts should be made to improve the prevention and treatment of disabilities in China and throughout the world.


We thank the provincial and municipal federations of disabled people for their support in data collection and management, and Iris Chi and Marilyn Flynn for their helpful comments and assistance.


Support for this study was obtained from: key National Project (973) of Study on the Mechanisms of Interaction between Environment and Genetics of Birth Defects in China (No. 2007CB5119001); key State funds for social science project (Research on Measuring Disability Prevention in China, No. 09&ZD072); National Baby Health Promotion Programme (No. FP2000NO13); Education Ministry Key Programme (No. 02185); National Yang Zi Scholar Programme, and projects 211 and 985 of Peking University (No. 20020903).

Competing interests:

None declared.