Age, period and cohort effects on adult physical activity levels from 1991 to 2011 in China
© Zang and Ng. 2016
Received: 24 December 2015
Accepted: 16 March 2016
Published: 20 April 2016
To date no work has differentiated the effects of age, period, and cohort on physical activity (PA) among Chinese adults, while also considering biological, behavioral, economic, and environmental factors over time.
We used data from the China Health and Nutrition Survey (CHNS) between 1991 and 2011 (20 years). The outcomes of interest are metabolic equivalent of task (MET) hours per week from work and domestic activities. Age, individual characteristics, household size, asset ownership, urbanization were included as covariates. Analyses for adult (≥20y) males (n = 29,343) and females (n = 31,094) was conducted to explicitly assess differences in PA due to age vs period effects, and implicitly assess differences by cohorts due to the period-specific experiences across individuals of varying ages.
The mean age of the sample rose from 41.31 to 50.8 years and PA decreased from 427.75 ± 264.35 MET hours per week (MET-hr/wk) in 1991 to 245.99 ± 206.65 MET-hr/wk in 2011, with much steeper declines for women compared to men. For both genders, we found non-linear decreases in PA with age over time. Controlling for age effects, negative period effects on PA were observed in each survey year, and were substantial from 1993 to 2000 for males and from 1993 to 2011 for females. The interaction between survey year and age (P < 0.05) were observed from 2004 to 2011. Higher community urbanicity, vehicle ownership, TV and computer ownership, overweight and obese, higher education served as negative predictors. Bicycle ownership, bigger household size, non-professional jobs, being married and having more children (for women) were positive predictors of PA (P < 0.05). Furthermore, at any given age, individuals who were younger at baseline had higher mean PA compared with individuals older at baseline.
This study followed a large cohort of adults over a significant portion of their lives. Strong age and secular trends were observed, resulting in an increasing number of participants who have or are likely to lower their PA levels. These trends suggest that tackling the rapid PA decline among its population is of high priority for China’s public health outlook as its population ages and continues to experience significant economic and environmental changes.
Over the last few decades, China has experienced reductions in mortality and morbidity attributable to infectious diseases, but concurrently has had rapid increases in the prevalence of chronic non-communicable diseases (NCDs) . Alongside this epidemiological transition is China’s rapidly aging population. Consequently, the burden of NCDs in China has dramatically increased and contributes to a significant and growing proportion of deaths as well as rising health care costs in recent years [2, 3].
There is now clear consensus that physical activity (PA) helps reduce the risk of NCDs such as cardiovascular and cerebrovascular diseases, diabetes, obesity, hypertension, cognitive function and cancers [4, 5], and also has positive effects on many aspects of mental health . However, while the benefits of PA are well accepted, unprecedented social and economic changes in China have led to dramatically lower levels of PA at work, as jobs have become less physically demanding over time. Moreover, economic and technological improvements have facilitated house work and transportation, increasing sedentary time across the country in recent years . These changes have been found to be strongly associated with large increases in the prevalence of overweight and obesity across all ages, regions, and levels of socioeconomic status [8, 9], as well as in the rise of NCDs  in China.
Indeed, engagement in various types or forms of PA is likely related to age, life-course, socioeconomic factors, assets ownership, the built environment and urbanization. Therefore, measurements of these factors are important for understanding how changes in PA may alter risk for disease or mortality. For a country like China that has an aging population, has and continues to experience rapid economic and environmental changes, understanding how these factors relate to PA is particularly important. However, most PA data available in China are cross-sectional [6, 10–12], so past studies using such data to study the associations between exposures and PA between age groups may represent differences between birth cohorts rather than effects of exposures [6, 10–12]. In contrast, studies using longitudinal designs may reflect temporal (period) factors. Thus, it is difficult to distinguish among effects of aging, cohorts, and period when either a cross-sectional or a longitudinal design is used.
To address these gaps, this study describes changes in PA among Chinese adults using 20-years of longitudinal data. We explicitly assess differences in PA within individuals over time (age effect) and population-wide differences in PA overtime (period effect), and implicitly assess differences in the experienced period effect across individuals of varying ages (cohort effect). We do so while also adjusting for social, economic, and environmental factors over time to determine how these factors contribute to changes in PA levels.
We used data from the China Health and Nutrition Survey (CHNS), a prospective household-based study that includes multiple ages and cohorts across nine rounds of surveys between 1989 and 2011 in nine diverse provinces and three megacities (Beijing, Shanghai and Chongqing were added in 2011) . A multistage, stratified sampling design was used to ensure that the CHNS provided representation of rural, urban and suburban areas varying substantially in geography, economic development, public resources and health indicators . The CHNS is the only large scale, longitudinal study of its kind in China. Further information on survey procedures and rationale of the CHNS is in the cohort profile . The CHNS was approved by the institutional review committees of the University of North Carolina at Chapel Hill, the National Institute of Nutrition and Food Safety and China Center for Disease Control and Prevention.
Our analysis included the eight rounds of survey data collected in 1991, 1993, 1997, 2000, 2004, 2006, 2009 and 2011. We did not use data from 1989 (n = 4753) because the sample at that wave only included adults 20–45 years old. We excluded participants who had fewer than two waves of data collection (including participants from the three mega-cities), as we would be unable to assess temporal trends for these individuals (n = 10,527). Additionally, observations were excluded from the analysis if a participant was pregnant, disabled or retired at the time of data collection. We focused our analyses on adults who were 20–59y in 1991. We omitted adults >60y in 1991 from our analysis due to the small and shrinking sample size over time as this cohort aged and some respondents passed away. Consequently, the inclusion of the >60y birth cohort (that would have been >80y by 2011) would have introduced potential bias.
Our final analysis sample included 13,245 participants, with an average of 4.5 observations collected on each subject (60,437 observations). The data are an unbalanced panel, with some individuals being absent in certain waves and returning in future waves as the sample was replenished over time [13, 14]. If a community had <20 households recruited, the CHNS team randomly selected additional households to obtain at least 20 households in each community. Individuals who were members of ‘old’ (previously sampled) and ‘new’ households helped replenish the sample. Relative to previous rounds of data collection, retention was between 80 and 88 % across all surveys after 1991 .
The outcomes of interest are the metabolic equivalent of task (MET) hours per week (MET-hr/wk). MET is defined as the ratio of a person’s working metabolic rate relative to his or her resting (basal) metabolic rate . We derived PA measures by multiplying (sub)activity-specific MET values obtained from the literature [13, 16] by the self-reported time reported spent in these (sub)activities among individuals in the CHNS to obtain the MET-hrs/week measurement. The MET-hr/wk measurement accounts for both the average intensity of each (sub)activity, and the time spent in each activity. In the CHNS, questions about occupational and domestic activities started in 1991, but questions about active leisure (from various forms of sports and exercise such as walking/strolling, jogging/running, swimming, badminton, table tennis, basketball, dance, wushu), and travel (from walking, bicycling, taking public transport, and driving) were only included starting in 1997. Measurements of sedentary leisure time among adults were only available starting in 2004. Due to our interest in measuring the changes in PA across periods, the primary outcome used is limited to work MET-hrs/wk from both market-based and home-based work, plus domestic MET-hrs/wk from preparing food, buying food, laundry and childcare (which may omit other important domestic activities such as cleaning). We term this “work and domestic PA”. Details on how these values were calculated are described elsewhere [17, 18]. To provide context, we also include descriptive findings on the amount of active leisure PA and travel PA in MET-hrs/wk and sedentary leisure time(hrs/wk) derived from self-reported frequency and time spent on weekends vs weekdays in: various forms of sports and exercise (such as walking/strolling, jogging/running, swimming, badminton, table tennis, basketball, dance, wushu); on walking, bicycling, taking public transport, and driving, and; on watching television/movies/videos, playing video/computer games and reading/writing, respectively.
Age at the baseline survey was used to define age groups for analysis (20–29y, 30–39y, 40–49y, 50–59y). Age and age-squared were included in longitudinal models to account for the nonlinear association of PA with age.
We included a number of individual level measures that may be associated with PA levels, including: weight status (Normal weight/Overweight/Obese, defined using BMI of 25–30 as overweight and BMI >30 as obese), occupation, marriage status, education, and number of children.
Household socio-economic factors
We also accounted for a number of household factors such as household size (number of persons reported to be living in the household at the time of the survey), and household income.
We were interested if ownership of certain technologies relate to PA and so included measurements of bicycle ownership, vehicle ownership, and TV and computer ownership.
A multi-component urbanicity index created from community survey data [19, 20] reflected population size and density, community infrastructure, and economic and environment characteristics. An increase in the value of the index over time represents greater urbanization.
Cross-sectional univariate descriptive statistics of the primary outcomes and explanatory variables were calculated and reported as mean ± SD or %, stratified by gender (due to potential differences related to gender roles) for each survey year. For context, we also show the estimated MET-hrs/wk from active leisure and travel activities, as well as sedentary leisure time (in hrs/wk).
As the CHNS is a longitudinal study, attrition, migration, modifications to the sampling methods (including replenishment) over time, and other factors may result in cohort membership varying over time. In order to limit potential bias as a result of possible sample selection issues, we used a Heckman-two-step approach  to calculate an “inverse Mill’s ratio” for inclusion as a time-varying variable in all the models described below. This “inverse Mill’s ratio” is the inverse of the predicted probability that an individual was included in each survey given his/her province, community urbanicity and their interaction.
A series of longitudinal mixed effects models with fixed and random individual-level effects and random slopes were used to explicitly assess differences in PA within individuals over time (age effect), population-wide differences in PA overtime (period effect), and implicitly assess differences in the experienced period effect across individuals of varying ages (cohort effect). Mixed effects models were chosen because they can accommodate unbalanced data and continuous covariates, and because the hierarchical nature of the models overcome the identifiability problem by not assuming that age-period-cohort (APC) effects are linear and additive at the same level of analysis [22, 23].
We conducted four models. Model 1 included age and age-squared, indicator variables representing survey year, and age and age-squared interacted with survey year. The year coefficients describe the period effect, and interacted terms describe how period effects may differ by age (cohort effect). Model 2 layers on individual level covariates (occupation, weight status, marital status, education and, including number of kids for female to model 1. In addition to providing information about how these variables relate to PA, comparing model 2 to model 1 tells us whether these individual factors explain some of the period effects. Model 3 layers on household socio-economic measures such as household income, household size, household ownership and urbanization index to model 1. Similarly, comparing model 3 to model 1 tells us whether these household factors explain some of the period effects. Model 4 includes the full set of variables.
We also estimated a set of models stratified by baseline age group (which represent cohorts), and the comparison of coefficients across baseline age-stratified models provide insights into period-specific effects of covariates. Age and age-squared were omitted from these stratified models. All models were conducted using Stata’s XTMIXED program . Results were considered significant at p < 0.05.
Cross-sectional univariate descriptives of the China Health and Nutrition Survey across survey years, mean ± SD
All Participants (N)
41.31 ± 14.29
42.54 ± 14.65
42.11 ± 13.78
43.34 ± 13.36
46.37 ± 13.23
47.62 ± 13.03
48.89 ± 13.06
50.80 ± 12.92
Work & domestic MET-hours/week
427.75 ± 264.35
354.43 ± 222.28
356.38 ± 231.15
304.74 ± 210.46
263.08 ± 227.32
262.23 ± 232.78
249.46 ± 215.88
245.99 ± 206.65
Active leisure and travel MET-hours/week
11.00 ± 15.35
12.51 ± 16.30
9.69 ± 18.43
9.28 ± 16.97
9.88 ± 17.70
11.25 ± 21.88
Sedentary leisure time (hours/week)
18.47 ± 15.87
18.13 ± 15.06
19.57 ± 15.77
18.64 ± 15.98
40.68 ± 13.94
41.77 ± 14.26
40.90 ± 13.06
42.27 ± 12.74
45.69 ± 12.75
46.98 ± 12.43
48.30 ± 12.73
50.29 ± 12.70
Work & domestic MET-hours/week
396.42 ± 244.45
333.59 ± 208.01
353.81 ± 221.81
304.07 ± 200.34
284.95 ± 231.70
281.37 ± 236.20
257.41 ± 210.57
261.34 ± 206.88
Active leisure and travel MET-hours/week
12.49 ± 17.15
13.89 ± 18.45
10.52 ± 21.43
10.26 ± 19.32
10.23 ± 19.30
11.94 ± 23.25
Sedentary leisure time (hours/week)
20.38 ± 16.43
20.25 ± 16.75
21.17 ± 16.84
19.21 ± 20.05
41.94 ± 14.60
43.31 ± 14.98
43.33 ± 14.36
44.40 ± 13.86
46.98 ± 13.61
48.18 ± 13.50
49.42 ± 13.33
51.24 ± 13.10
Work & domestic MET-hours/week
458.75 ± 279.26
375.42 ± 233.93
358.96 ± 240.17
305.40 ± 220.03
243.63 ± 221.58
245.80 ± 228.55
242.28 ± 220.35
232.43 ± 205.53
Active leisure and travel MET-hours/week
9.50 ± 13.14
11.24 ± 13.94
8.67 ± 13.82
8.12 ± 13.57
9.45 ± 15.50
10.43 ± 20.09
Sedentary leisure time (hours/week)
17.47 ± 15.44
17.15 ± 14.12
18.60 ± 14.96
17.56 ± 14.65
Clearly, work and domestic PA remained key contributors of PA among Chinese adults. Meanwhile, the prevalence of overweight and obese in Chinese adults nearly tripled between 1991 and 2001 from 12.2 to 32.9 %.
Mixed effects estimates on work and domestic PA levels (MET-hrs/wk) among Chinese adult men, coefficient (95 % CI)
Survey year (ref: 1991)
0.1 0 (−0.42,0.62)
Inverse Mill’s Ratio
TV and computer ownership
Occupation (ref: professional job)
Education (ref: < primary school)
Middle and high school
College and above
Number of observations
Number of individuals
In Model 2, we found that higher community urbanicity, vehicle ownership, TV and computer ownership were negative predictors of work and domestic PA. Meanwhile bicycle ownership and bigger household size were positive predictors of work and domestic PA (Table 2, Model 2). Given that these measurements of technologies that should not directly affect work or domestic PA are associated suggests that there is strong correlation across various domains of PA.
In Model 3, we found that being overweight or obese was negatively related to PA. Compared to professional jobs (e.g., doctor, professor, lawyer, architect, engineer, editor), other occupations (farmer, officer/soldier/athlete, unskilled worker, skilled worker, students) were positive predictors of work and domestic PA. Being married was positively associated, but higher education was negatively associated with work and domestic PA levels (Table 2, Model 3). In comparing the coefficients from Model 3 to Model 2, we see that household factors account for much of the period effects.
Indeed, after adjusting for potential confounding effects of household socio-economic factors and urbanicity, as well as individual measures (Table 2, Model 4), the direction of all of the period effects remained the same, but coefficients were slightly attenuated relative to Model 2. Moreover, the inclusion of both household and individual factors strengthen the cohort effects (how period effects vary by age) especially for 2004*age, 2006*age and 2011*age.
Mixed effects estimates on work and domestic PA levels (MET-hrs/wk) among Chinese adult women, coefficient (95 % CI)
Survey year (ref: 1991)
Inverse Mill’s Ratio
TV and computer ownership
Occupation (ref: professional job)
Number of children
Education (ref: < primary school)
middle and high school
college and above
Number of observations
Number of individuals
Likewise, after adjusting for potential confounding effects of household socio-economic factors and urbanicity as well as individual measures (Table 3, Model 4)., the direction of all of the period effects remained the same, but coefficients were slightly attenuated relative to Model 2. Moreover, the inclusion of both household and individual factors strengthen the negative cohort effects especially for 2004*age and 2006*age.
The present study described changes in the primary sources of PA (work and domestic activities) among Chinese adults between 1991 and 2011, based on a questionnaire on time use that has been used in earlier studies [9, 17, 18, 25–28]. We found that throughout this 20-year period, for both adult men and women in China, work and domestic PA levels fell by nearly half, and the decline was more pronounced for women. Period effects, along with higher community urbanicity, vehicle ownership, TV and computer ownership, higher education, being overweight or obese were negatively associated with work and domestic PA levels. Bicycle ownership, labor-intensive jobs, being married, and larger household size were found to be positively associated with work and domestic PA levels. Moreover, our implicit measurement of the cohort effects by including interactions of year with age showed that being older in 2004 and 2006 had stronger negative associations with lower work and domestic PA in particular.
The CHNS was designed to provide representation of rural, urban and suburban areas with a focus on examining household- and individual-level sociodemographic factors, diet, physical activity, health and behavior changes within the context of community-level urbanicity and social and economic change . While many of the measures related to health and behaviors are based on self-report, the CHNS is the only large-scale, longitudinal study of its kind in China . Other PA-related studies in China, such as the cross-sectional InterASIA study conducted in 2000–2001  as well as the multi-country World Health Survey studies in 2002–2003 that uses the International Physical Activity Questionnaire (IPAQ)  ask about participation and estimate prevalence of physical inactivity, so the measurements are not directly comparable. However, the overall findings are consistent with what we found here (occupational PA is the most prevalent).
China’s dramatic economic, demographic and social transformation continues to unfold with cohorts experiencing rapid environmental change at different ages. The changes include increased modernization of the Chinese food system; continued use and greater dissemination of modern technology in manufacturing, transportation and leisure; housing, health system and educational system changes; and new pressures on the labor, environmental and social systems [30, 31].
Since economic reform in China in the late 1980s to early 1990s, more and more workers have transitions from heavy industry to service industry jobs. Consequently, the portion of the population engaged in vigorous labor has fallen and replaced by work activities that are lighter in physical labor needs or sedentary in nature . Moreover, technological advances in the workplace have lessened the physical demands of many jobs, and studies in urban China have found that these technological improvements are associated with declines in PA .
Concurrently, the public transportation system was improved with more buses and subways, while increase wealth have resulted in greater vehicle ownership (rose by over 15-fold between 1991 and 2011) , and higher reports of people using private cars as their main mode of transportation [26, 33, 34]. The proportion of Chinese who own color televisions, have access to cable networks and computers have all also risen dramatically, as have screen time [18, 35, 36]. We found consistent results in this study that these transformations are associated with the decline in PA in these two decades.
Unsurprisingly, we found that higher educational attainment was negatively associated with work and domestic physical activity. Likewise, vehicle and computer ownership, having a professional job, and living in a more urbanized area were negatively associated. Conversely, bicycle ownership, non-professional jobs, being married and bigger household size were positively associated with work and domestic PA. Certainly these factors are correlated with each other as participants who had higher education are more likely to have less labor-intensive occupations, more likely to have private cars (over bicycles) and live in urban areas (authors own review of correlation coefficients across measures). Nonetheless, the results indicate that they are also independently associated with reduced work and domestic PA levels among Chinese adults.
In this study, we found that work and domestic PA among women declined more substantially than among men. This is consistent with earlier findings , and largely driven by that fact that Chinese women historically do the majority of domestic activities (e.g., cooking, laundry, childcare), and still do, but both time spent and energy expended in these activities have fallen dramatically for women, while not changing much for men. This is likely due to at least two reasons. First, Chinese women have been bearing fewer children as a result of both the one-child policy (which was only recently altered), as well as for economic reasons. As seen in the models, the positive relationship between the number of children and PA suggest that the shrinking childbirth rate lessens PA from tending to children and related housework. Secondly, modern home-technologies, such as washing machines, microwaves, electric cookers, vacuum cleaners help facilitate housework by both reducing time spent as well as energy needs.
A strength of this study is that this is the largest longitudinal analysis of physical activity to be conducted in a diverse sample of Chinese adults. To our knowledge, we are the first to describe 20-year changes in PA among Chinese while differentiating effects of age, period, and birth cohort and determine how physical activity is influenced by biological, behavioral, economic, and environmental factors over time.
Nonetheless, there are several limitations to this study. Our analysis only focuses on self-reported work and domestic activity, which forms the majority of time-use as well as energy expenditure in China as well as in other developing countries [17, 37]. These measures of work and domestic activity may be misreported or incomplete (e.g., missing cleaning as a domestic activity), and partly relies on MET-values derived from the US, which may not be appropriate.
Additionally, due to data limitations (missing 2 earlier waves) we were unable to include other forms of PA in our analyses. It would be particularly important to also consider the role of leisure-time activity. Past studies have shown that in China, participation in moderate or vigorous leisure-time physical activity was low at 7.9 and 28.9 %, among urban and rural Chinese adults, respectively . Other descriptive studies looking at trends over time using the CHNS also suggest that active leisure activity is increasing slightly, but insufficiently to make up the large reductions from the other domains of work, home and transportation . Much more needs to be done to move the needle of active leisure, which has been found to benefit not just physical health but also mental health . Investments into age and culturally appropriate efforts that use and encourage the provision of easily accessible communal spaces for tai-chi, wushu (martial arts), dance, badminton, jogging and other more popular regional or cultural sports should be made in both urban and rural settings. Greater commitments towards active transportation (e.g., bicycling via providing a bicycle-sharing system and improving road safety) should also be considered as it can concurrently address air pollution, congestion and climate change concerns. Many of these approaches are also mutually reinforcing, as improved air quality and improved traffic safety removes barriers to exercising outdoors. Indeed “whole of government” and “whole of society” approaches will be necessary .
The urgency of this issue is highlighted by estimates that in 2007, the direct and indirect costs of physical inactivity in China was 6.7 billion USD (in 2007 dollars) as the lack of PA contributes to between 12 and 19 % of the five major NCDs in China , and 8.3 % of all-cause mortality . A more recent study found that cardiometabolic risk prevalence was high in all age groups in China, particularly younger adults, and that low PA consistently predicted higher cardiometabolic risk across most outcomes and age-sex groups . Consequently, the economic cost of physical inactivity is expected to grow as PA levels continue to fall, given the strong age effect and secular trends found in this study, China’s rapidly aging population, as well as the much lower PA levels of the younger cohorts. This is a problem that has gained notice, and there are increasingly multi-sectoral, government policies and corporate social responsibility efforts underway to reduced inactivity, encourage PA across all domains of life, and thus to halt the growing prevalence of NCDs in China . Examples include the 2007 Ministry of Health campaign “ten thousand steps a day, the balance of eating and activity and a healthy life”, the 2003 Guidelines for Prevention and Control of Overweight and Obesity of Chinese Adults, and the 2007 School-age Children and Teenagers Overweight and Obesity Prevention and Control Guidelines . Efforts targeting children in particular are critical to encourage formation of healthier habits and lifestyles. Some intervention programs have also been shown to be effective and expanded across cities and provinces, such as the Take Ten school-based PA intervention. Originally a pilot study in Beijing in 2004, Take Ten was expanded and eventually prompted the Department of Education to institute a policy of one hour of PA in schools every day . It would be useful to examine what the long-term effects of such a policy would have on this younger cohort as they age.
In summary, this study followed a large cohort of adults over a significant portion of their lives. As China underwent dramatic economic transitions, we observed strong age and secular trends in work and domestic PA levels, resulting in an increasing number of men and women who have or are likely to have even lower PA levels both due to aging and the combination of period and cohort differences. These trends are highly relevant for health policy and preventive health measures in the China and other countries that are now facing similar challenges.
We would like to thank Dr. Shufa Du for helping arrange a visiting scholarship, Drs. Xiaodong Jia, and Shurong Zou for their assistance reviewing the manuscript, and funders: NIH (R01-HD30880, DK056350, R24 HD050924 and R01-HD38700, R01-HL108427), Fogarty International Center, Shanghai Municipal Bureau of Health (number: 20134053), Shanghai Rising-Star Program (number: 15YF1411100).
Sources of support
The work presented in this paper was supported by NIH (R01-HD30880, DK056350, R01-HD38700, and R01-HL108427) and the Fogarty International Center, NIH gave financial support for the CHNS data collection from 1989 to 2011. We are grateful to the Carolina Population Center and its NIH Center grant (P2C HD050924) for general support. The study also supported by the research grants of Shanghai Municipal Bureau of Health (number: 20134053), Shanghai Rising-Star Program (number: 15YF1411100). None of the funders, NIH, Fogarty International Center, Shanghai Municipal Bureau of Health, and Shanghai science and Technology Commission had any role in the design, analysis or writing of this paper.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Li Y, Wang L, Jiang Y, Zhang M, Wang L. Risk factors for noncommunicable chronic diseases in women in China: surveillance efforts. Bull World Health Organ. 2013;91(9):650–60.View ArticleGoogle Scholar
- Tang S, Ehiri J, Fau - Long Q, Long Q. China’s biggest, most neglected health challenge. Non-communicable diseases. 2013; 2:7. doi: 10.1186/2049-9957-2-7.
- Bloom DE, Cafiero ET, McGovern ME, Prettner K, Stanciole A, Weiss J, et al. The economic impact of non-communicable disease in China and India: estimates, projections, and comparisons, NBER Working Paper No. 19335, August 2013, JEL No. E13,I15,O40. 2013.Google Scholar
- World Health Organization. Global Recommendations on Physical Activity for Health. Geneva; 2010. Available at: http://www.who.int/dietphysicalactivity/publications/9789241599979/en/.
- Wu Z, Yao C, Zhao D, Wu G, Wang W, Liu J, et al. Sino-MONICA project: a collaborative study on trends and determinants in cardiovascular diseases in China, Part i: morbidity and mortality monitoring. Circulation. 2001;103(3):462–8.View ArticleGoogle Scholar
- Huang X, Yang H, Wang HH, Qiu Y, Lai X, Zhou Z, et al. The association between physical activity, mental status, and social and family support with five major non-communicable chronic diseases among elderly people: a cross-sectional study of a rural population in Southern China. Int J Environ Res Public Health. 2015;12(10):13209–23.View ArticleGoogle Scholar
- Katzmarzyk PT. Physical activity, sedentary behavior, and health: paradigm paralysis or paradigm shift? Diabetes. 2010;59(11):2717–25.View ArticleGoogle Scholar
- Jones-Smith JC, Gordon-Larsen P, Siddiqi A, Popkin BM. Cross-national comparisons of time trends in overweight inequality by socioeconomic status among women using repeated cross-sectional surveys from 37 developing countries, 1989–2007. Am J Epidemiol. 2011;173(6):667–75.View ArticleGoogle Scholar
- Monda KL, Adair LS, Zhai F, Popkin BM. Longitudinal relationships between occupational and domestic physical activity patterns and body weight in China. Eur J Clin Nutr. 2008;62(11):1318–25.View ArticleGoogle Scholar
- Muntner P, Gu D, Wildman RP, Chen J, Qan W, Whelton PK, et al. Prevalence of physical activity among Chinese adults: results from the International Collaborative Study of Cardiovascular Disease in Asia. Am J Public Health. 2005;95(9):1631–6.View ArticleGoogle Scholar
- Fan M, Su M, Tan Y, Liu Q, Ren Y, Li L, et al. Gender, age, and education level modify the association between body mass index and physical activity: a cross-sectional study in Hangzhou, China. PLoS One. 2015;10(5), e0125534.View ArticleGoogle Scholar
- Sun L, Jiang X, Zhao X, Zhang Y, Xu Y, Shang L. Physical activity level and associated factors among civil servants in Xi’an, China. J Sci Med Sport. 2015; S1440-2440. http://dx.doi.org/10.1016/j.jsams.2015.08.003.
- Popkin BM, Du S, Zhai F, Zhang B. Cohort profile: the China Health and Nutrition Survey--monitoring and understanding socio-economic and health change in China, 1989–2011. Int J Epidemiol. 2010;39(6):1435–40.View ArticleGoogle Scholar
- Zhang B, Zhai FY, Du SF, Popkin BM. The China Health and Nutrition Survey, 1989–2011. Obes Rev. 2014;15 Suppl 1:2–7.View ArticleGoogle Scholar
- Sallis JF, Haskell WL, Wood PD, Fortmann SP, Rogers T, Blair SN, et al. Physical activity assessment methodology in the Five-City Project. Am J Epidemiol. 1985;121(1):91–106.Google Scholar
- Ainsworth B, Haskell W, Herrmann S, Meckes N, Bassett Jr D, Tudor-Locke C, et al. The compendium of physical activities tracking guide. Healthy Lifestyles Research Center, College of Nursing & Health Innovation, Arizona State University. [cited; Available from: http://prevention.sph.sc.edu/tools/docs/documents_compendium.pdf.
- Ng SW, Popkin BM. Time use and physical activity: a shift away from movement across the globe. Obes Rev. 2012;13(8):659–80.View ArticleGoogle Scholar
- Ng SW, Norton EC, Popkin BM. Why have physical activity levels declined among Chinese adults? Findings from the 1991–2006 China Health and Nutrition Surveys. Soc Sci Med. 2009;68(7):1305–14.View ArticleGoogle Scholar
- Dahly DL, Adair LS. Quantifying the urban environment: a scale measure of urbanicity outperforms the urban–rural dichotomy. Soc Sci Med. 2007;64(7):1407–19.View ArticleGoogle Scholar
- Jones-Smith JC, Popkin BM. Understanding community context and adult health changes in China: development of an urbanicity scale. Soc Sci Med. 2010;71(8):1436–46.View ArticleGoogle Scholar
- Heckman JJ. Sample selection bias as a specification error. Econometrica. 1979;47:153–61.View ArticleGoogle Scholar
- Yang Y, Lee LC. Dynamics and heterogeneity in the process of human frailty and aging: evidence from the U.S. older adult population. J Gerontol B Psychol Sci Soc Sci. 2010;65B(2):246–55.View ArticleGoogle Scholar
- Zhang J, Chaaban J. The economic cost of physical inactivity in China. Prev Med. 2013;56(1):75–8.View ArticleGoogle Scholar
- StataCorp. Stata Statistical Software, release 11. College Station: StataCorp LP; 2009.Google Scholar
- Attard SM, Howard AG, Herring AH, Zhang B, Du S, Aiello AE, et al. Differential associations of urbanicity and income with physical activity in adults in urbanizing China: findings from the population-based China Health and Nutrition Survey 1991–2009. Int J Behav Nutr Phys Act. 2015;12:152.View ArticleGoogle Scholar
- Bell AC, Ge K, Popkin BM. The road to obesity or the path to prevention: motorized transportation and obesity in China. Obes Res. 2002;10(4):277–83.View ArticleGoogle Scholar
- Monda KL, Gordon-Larsen P, Stevens J, Popkin BM. China’s transition: the effect of rapid urbanization on adult occupational physical activity. Soc Sci Med. 2007;64(4):858–70.View ArticleGoogle Scholar
- Ng SW, Howard AG, Wang HJ, Su C, Zhang B. The physical activity transition among adults in China: 1991–2011. Obes Rev. 2014;15:27–36.View ArticleGoogle Scholar
- Guthold R, Ono T, Strong KL, Chatterji S, Morabia A. Worldwide variability in physical inactivity a 51-country survey. Am J Prev Med. 2008;34(6):486–94.View ArticleGoogle Scholar
- Hertel T, Zhai F. Impacts of the Doha Development Agenda on China: the role of labor markets and complementary education reforms. In: Bank TW, ed. Working Paper, 2005.Google Scholar
- Hertel TW, Winters LA. Poverty impacts of a WTO agreement: synthesis and overview. In: Bank W, ed. Washington, DC, 2005. [WWW document]. URL http://elibrary.worldbank.org/doi/abs/10.1596/1813-9450-3757. (accessed 11 January 2010).
- National Bureau of Statistics of the People’s Republic of China. Online database. http://www.stats.gov.cn/english/. (accessed on 17 March 2016)
- Wang H, Chen C, Huang C, Fu L. On-road vehicle emission inventory and its uncertainty analysis for Shanghai, China. Sci Total Environ. 2008;398(1–3):60–7.View ArticleGoogle Scholar
- Zhao H, Yang G, Zhu F, Jin X, Begeman P, Yin Z, et al. An investigation on the head injuries of adult pedestrians by passenger cars in China. Traffic Inj Prev. 2013;14(7):712–7.View ArticleGoogle Scholar
- Du SF, Wang HJ, Zhang B, Zhai FY, Popkin BM. China in the period of transition from scarcity and extensive undernutrition to emerging nutrition-related non-communicable diseases, 1949–1992. Obes Rev. 2014;15 Suppl 1:8–15.View ArticleGoogle Scholar
- Zhai FY, Du SF, Wang ZH, Zhang JG, Du WW, Popkin BM. Dynamics of the Chinese diet and the role of urbanicity, 1991–2011. Obes Rev. 2014;15 Suppl 1:16–26.View ArticleGoogle Scholar
- Forrest KY, Bunker CH, Kriska AM, Ukoli FA, Huston SL, Markovic N. Physical activity and cardiovascular risk factors in a developing population. Med Sci Sports Exerc. 2001;33(9):1598–604.View ArticleGoogle Scholar
- Wang H, Zhai F. Programme and policy options for preventing obesity in China. Obes Rev. 2013;14:134–40.View ArticleGoogle Scholar
- Lee IM, Shiroma EJ, Lobelo F, Puska P, Blair SN, Katzmarzyk PT, et al. Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy. Lancet. 2012;380(9838):219–29.View ArticleGoogle Scholar
- Adair LS, Gordon-Larsen P, Du SF, Zhang B, Popkin BM. The emergence of cardiometabolic disease risk in Chinese children and adults: consequences of changes in diet, physical activity and obesity. Obes Rev. 2014;15 Suppl 1:49–59.View ArticleGoogle Scholar
- Zhang Y-X, Zhou J-Y, Zhao J-S, Chu Z-H. The role of 1-h physical activity every day in preventing obesity in adolescents in Shandong, China. Eur J Pediatr. 2012;172(3):325–30.Google Scholar