China has a high prevalence physical inactivity which leads to poor health outcomes. Among Chinese adults, the prevalence of leisure-time physical activity (LTPA) participation is low. Data collected in the China Health and Nutrition Survey (CHNS) from 2000 to 2015 were used to observe trends in LTPA participation among adults aged 18 years and above. The overall age-standardized LTPA prevalence increased from 7.13% in 2000 to 11.79% in 2011 before dropping to 7.33% in 2015. Over 80% of LTPA participants reported vigorous-intensity LTPA. Although LTPA prevalence and the proportion to reach the vigorous-intensity level had been increasing overall, they were still at a low level. Efforts to improve rates of LTPA among Chinese adults, especially for residents in rural areas and with low-income, through education or other interventions has the potential benefits for improving overall health.
Physical inactivity contributes 12%–19% to the risks associated with the 5 major noncommunicable diseases (NCDs) in China (1). Increasing physical activity (PA) is an effective way to improve both individual and population-level health outcomes. However, PA levels appear to be declining globally (2). Occupational, travel, domestic, and leisure-time PA are four dimensions of the overall PA and the first three showed a striking decrease. LTPA was the smallest contributor to overall PA, but it plays an important role to keep healthy (3). The proportion of LTPA in China was also much lower than that of Western countries (3-5) such as Brazil, the United States, Finland, etc. Increasing LTPA to compensate for declines in the other three domains of PA to maintain the current level of the overall PA and further promote its improvement is vitally important because only LTPA is self-directed and the other three types have passively declined due to high speed economic development, which include examples such as driving cars instead of walking to the office and using washing machines instead of washing by hand. To understand whether Chinese adults actively participate in LTPA to improve the decreasing total PA, CHNS provides the opportunity to investigate the trend of LTPA participation over time with the overall goal of informing health policies.
Details of the study design and methods of the CHNS have been previously reported (6). A multi-stage, stratified, random cluster sampling design was used across 10 rounds of surveys from 1989 to 2015. The samples in the present paper included 6 rounds of survey data collected in 2000, 2004, 2006, 2009, 2011, and 2015. Nine provincial-level administrative divisions (PLADs), including Shandong, Liaoning, Heilongjiang, Jiangsu, Henan, Guizhou, Hunan, Hubei, and Guangxi were chosen because they were present in all six surveys. Adults aged 18-100 years old with complete data on LTPA, socioeconomic status, and demographics in each survey year were considered eligible subjects. We excluded individuals who were disabled, pregnant, or lactating during each particular wave. Finally, our study consists of 59,504 observations in total.
Information on LTPA within the latest year included participation in sports such as martial arts, gymnastics, dancing, aerobics, jogging, swimming, soccer, basketball, tennis, badminton, volleyball, or other LTPA (table tennis, or doing Tai Chi, etc.). For each LTPA item, respondents reported the average hours per week spent in the last year. To measure energy expenditure for each LTPA item, metabolic equivalent of task hour per week (MET·h/w) were calculated. Therefore, the average MET-hours per week measurements comprise both the average intensity of each activity and the time spent in each one. Participants were classfied into four categories by LTPA (0 MET·h/w = inactive, 0.1–7.4 MET·h/w = low-intensity, 7.5–14.9 MET·h/w = moderate-intensity and ≥15.0 MET·h/w = vigorous-intensity) (7-8). LTPA prevalence was defined as the proportion of the number of LTPA participants among the total number of participants in each survey. Standard questionnaires were used by trained interviewers to collect sociodemographic characteristics and annual per household income, which was categorized into tertiles (high, middle, and low) according to each survey round. Descriptive analysis was stratified by gender, age group, and urban-rural residence in each round. The overall prevalence from 2000 to 2011 was standardized according to the age composition in 2015 of CHNS. Chi-square tests were used to test for differences among males and females, age groups, urban and rural residence, and LTPA levels. Trend Chi-square test was used for changes over time and income levels. p<0.05 were considered statistically significant. All tests were conducted in SAS software (version 9.4, SAS Institute, Inc., Cary, NC, USA).
Table 1 shows that the proportion of adults aged 18 to 44 years old declined in these 15 years and the proportion of 45–64 years old group and ≥65 years old group increased, consistent with observed aging of the population. Participants in urban area represented around 30% of the sample in each survey year. Figure 1 gives information about the LTPA prevalence of adults by demographic characterristics during the study period. The age-standardized LTPA prevalence increased from 7.13% in 2000 to 11.79% in 2011, then dropped to 7.33% in 2015. The proportion of participants that were male or living in urban areas were higher than that of females and rural residents, respectively. There was significantly statistical difference between males and females as well as urban and rural. Compared to groups of adults aged 45 to 64 years old and ≥65 years old, the group aged 18 to 44 years old had the highest LTPA prevalence. People at high income levels had approximately 7 to 15 percentage pointed higher LTPA prevalence than those at low income levels, and the income level trend tests were significant in each year. The increasing trend by years was statistically significant in the LTPA prevalence of total population, males and females groups, three age groups, rural group, high and low income groups, respectively. Figure 2 illustrates that from 2004 onward, over 80% of LTPA participants were able to reach the vigorous-intensity level and this proportion increased each year. Specifically, 60% of LTPA participants exercised at vigorous-intensity PA in 2000 and then rising to 90% in 2015.
Variable 2000 2004 2006 2009 2011 2015 Overall 11,308 9,618 9,662 9,892 9,427 9,597 Gender Male 5,781(51.12) 4,698(48.85) 4,625(47.87) 4,797(48.49) 4,512(47.86) 4543(47.34) Female 5,527(48.88) 4,920(51.15) 5,037(52.13) 5,095(51.51) 4,915(52.14) 5,054(52.66) Age group (yrs) 18–44 6,277(55.48) 4,178(43.44) 4,048(41.90) 3,701(37.41) 3,108(32.97) 2,945(30.69) 45–64 3,749(33.14) 3,942(40.99) 3,985(41.24) 4,393(44.41) 4,398(46.65) 4,452(46.39) ≥65 1,287(11.38) 1,498(15.57) 1,629(16.86) 1,798(18.18) 1,921(20.38) 2,200(22.92) Urban-rural Urban 3,388(29.96) 3,010(31.30) 3,035(31.41) 3,139(31.73) 3,059(32.45) 2,994(31.20) Rural 7,920(70.04) 6,608(68.70) 6,627(68.59) 6,753(68.27) 6,368(67.55) 6,603(68.80)
Table 1. Samples and proportion (%) of adults aged 18 years and above from 2000 to 2015, stratified by gender, age group and urban-rural.
Figure 1. LTPA prevalence (%) of adults aged 18 years and above in subgroups from 2000 to 2015. (A) gender for LTPA prevalence; (B) area for LTPA prevalence; (C) age for LTPA prevalence; (D) income level for LTPA prevalend.* indicates that the prevalence from 2000 to 2011 was standardized according to the age composition in 2015. † indicates a statistical significant difference by gender for LTPA prevalence in each survey round; § indicates a statistical significant difference between urban and rural for LTPA prevalence in each survey round; ¶ indicates a statistical significant difference among age groups for LTPA prevalence in each survey round; ** indicates a statistical significant income level trend for LTPA prevalence in each survey round; †† indicates statistical significant year trend for LTPA prevalence in each subtype group, such as male gourp, rural group, low income group.