1. Background

With the growth of the older population, the average life expectancy of Koreans has increased significantly to 82.4 years, but the healthy life expectancy is estimated to be 64.9 years, 17.5 years lower than the life expectancy [1]. This means that older adults live in an unhealthy state due to diseases, frailty, and lying sick in bed during the period. Since most health problems in older adults are chronic diseases, active improvement of lifestyle habits and self-management skills for chronic disease management are required [2]. However, individual efforts alone are insufficient, and the role of the community is important in improving health-related lifestyle habits and enhancing the competency in chronic disease self-management in older adults.

Most older adults aged 65 and older have one or more chronic diseases [2], so the management of geriatric diseases and health-promoting behaviors that enable independent daily life are required. Among health-promoting behaviors, physical activity is a useful method for a healthy old age and independent life by prolonging healthy life expectancies. In other words, physical activity can reduce changes in physical functions due to the aging process, prevent cognitive decline, increase independence, and alleviate depression, thereby improving the quality of life in older adults [3]. Despite the positive effects of such physical activity, it has been reported that 41.9% of older adults in Korea do not participate in physical activity and participation in physical activity decreases gradually with increasing age [2]. According to the World Health Organization (WHO)[4], the lack of physical activity is the fourth leading cause of death worldwide. Since the lack of physical activity is such a serious level, active measures are required to maintain and promote physical activity to lead a healthy life, and a prerequisite for such measures is to identify factors related to physical activity in older adults.

The physical activity of older adults is reported to be affected by sociodemographic factors such as gender, age, educational level, and economic status [3, 5]. Although an individual's biological, genetic and socioeconomic characteristics determine health status levels, health may also be improved by the physical environment or institutional support which provides motivation for health promotion. On the other hand, prolonged exposure to a harmful environment or more limited social support compared to others may result in relatively poorer health conditions or more serious health problems [6]. Individual factors that influence physical activity in older adults are variables that increase or decrease the likelihood that individuals would participate in physical activity, and they include individuals' psychological characteristics, such as self-efficacy, attitude, belief, and motivation and sociodemographic characteristics, such as gender, age, and education level [7]. Social factors are various social factors that influence individual behavior and they include social support, social network, social capital, and neighborhood factors [8]. Since physical activities are performed within a physical environment, physical environmental factors are very important for continuous or regular physical activities, and they include ease of spatial access to exercise facilities, convenience of exercise facilities, and the stability of the surrounding environment [9]. Therefore, it is necessary to gain an understanding of the complex effects of individual, social and environmental factors before developing a plan for improving physical activity among older adults.

In recent years, the composition ratio of the older population has been becoming different according to the residential areas (urban and non-urban areas) across the world. As a result, the exercise participation rate and factors affecting physical activity among older adults may vary according to the residential areas [10]. Therefore, in order to increase physical activity among older adults, it is necessary to develop strategies for promoting physical activity in consideration of the individual, social, and environmental characteristics of older adults in each residential area. Previous studies on the health of older adults according to the place of residence or residential area were mostly focused on individual factors such as demographic characteristics and health-related characteristics [11, 12] and a study considered the impact of social support on health status together with other factors [13]. Most previous studies on physical activity in older adults limited research to one type of area, such as urban or rural areas, or investigated related factors without the division of residential areas [3, 5, 8, 14]. Although a small number of studies compared predictive factors for physical activity among urban and rural older adults [15], few studies have sufficiently considered the social and environmental characteristics reflecting the physical and human health resources of residential areas.

In particular, in Korea, since economic development has been concentrated in the Metropolitan Area around Seoul, distinct disparities in health between the metropolitan and the non-metropolitan areas have been observed in recent years, and the health gap between regions has been widening especially in non-urban areas [16]. With respect to these aspects, it is considered a meaningful research attempt to examine the problem of regional health disparities among older adults since this investigation allows us to identify differences in the social structural conditions. The physical and social environment of a residential area affects the resources and choices available to residents. In particular, for older adults with characteristics such as physical frailty, a lack of resources, and limited accessibility, environmental factors such as medical services, various social facilities, and community capacity [17] affect their physical activities, which may lead to regional disparities in health status.

Therefore, based on the social ecological model (SEM) [18], which considers multidimensional effects including social and environmental factors as well as individual factors in relation to the determinants of health behaviors of individuals, this study aimed to make an integrative investigation of overall factors influencing physical activity in older adults by region. Through this investigation, this study attempted to provide basic data for the development of community programs to promote physical activity among older adults and strategies to create a social environment to reduce regional health inequalities.

2. Purpose

The main purpose of this study was to identify individual, social and environmental factors related to physical activity in older adults in urban and non-urban areas by using the data from the Community Health Survey since the data allows a comparison of health status levels between regions.

3. Conceptual Framework

The conceptual framework of this study was based on the social ecological model. This model has five hierarchical levels of individual, interpersonal, community, organizational, and policy/enabling environments, and it also has the concept of expansion from individuals to society [18]. Since this model presents a comprehensive system for understanding the effects of various levels on physical activity, it has been used as an effective theoretical basis in several previous studies on physical activity [19]. Since the use of the data from the Community Health Survey involves limitations in linking variables in relation to organizational and policy areas with data obtained from a questionnaire survey on health behaviors related to chronic diseases, a modified conceptual model reflecting individual, social, and environmental factors was constructed (Figure 1).

Figure 1 Conceptual framework of this study. Click for larger image

1. Study Design

This study is a descriptive research to investigate individual, social, and environmental factors related to physical activities among older adults in urban and non-urban areas and to identify influencing factors for physical activity.

2. Participants

This study used data from the 2017 Community Health Survey among 228,381 adults aged 19 and older. The samples of the Community Health Survey were extracted by the complex sampling design method, and the information on the elements of complex sampling design, such as sampling weight, stratified variables, and clustered variables, is included in the raw data. If analysis is conducted by selecting or deleting some part of the data instead of using the total data, the standard error bias of the estimate may occur due to the omission of information on complex sampling design included in the deleted data. Therefore, among the total respondents, older adults aged 65 and older without missing data were first selected, and 23,043 elderly people in urban areas and 34,063 elderly people in non-urban areas were designated as subpopulations. Then, data of the subpopulations was used without further deleting other data.

3. Research variables

The research variables used in this study were individual, social, and environmental factors including sociodemographic and health-related characteristics which were selected among variables available in the Community Health Survey data by referring to the items reported as factors related to physical activity in previous studies [3, 5, 6, 7, 8, 9, 14, 15].

4. Data Collection and Analysis

The Community Health Survey is a cross-sectional study using multi-stage area probability sampling by the two-stage stratified two-stage cluster systematic sampling method. After stratification of dividing the population into regions, administrative units (dong/eup · myeon), and residential types (apartment/single house) to represent all the regions and the entire population in the country, sample households are extracted proportionally to the number of households in the population of each strata by the systematic sampling method. The data is collected on the selected households and all members of those households through interviews and the Community Health Survey (CHS) data is publicly available data. Therefore, this study was conducted using the 2017 CHS data after the investigators signed the agreement for access to personal information for research or statistical purposes and submitted a plan on the use of CHS data to request the data.

5. Statistical Analysis

Data analysis was carried out using IBM SPSS Statistics 23.0 by applying an analysis method for complex sample survey data considering the strata, cluster and weights. A χ² test was conducted to examine differences in individual, social and environmental factors and physical activity in older adults between urban and non-urban areas. Participants were divided into the inactive, minimally active, and health-enhancing physical activity groups according to the amount of physical activity, and the level of physical activity was set as the dependent variable. Based on the results of previous studies [3, 5, 6, 7, 8, 9, 14, 15], individual, social, and environmental factors which were expected to affect physical activity were entered into the logistic regression model to calculate the odds ratios and determine statistical significance with a 95% confidence interval. A value of p<.05 was considered statistically significant.

6. Ethical Considerations

This study was conducted after receiving an exemption determination from the IRB of Catholic University of Pusan (IRB No. 2019-01-009) for the secondary analysis of the 2017 Community Health Survey data that was collected after obtaining informed consent from all participants prior to data collection.

1. Individual, Social and Environmental Factors and Physical Activity of Participants

The analysis results of differences in individual, social, and environmental factors and physical activities between older adults in urban and non-urban areas are shown in Table 1. In terms of demographic and sociological characteristics among individual factors, there were statistically significant differences between the two groups in age, household type, household income, educational level and economic activity, except for gender (p<.001). As for health-related characteristics, the percentages of people with multimorbidity of chronic diseases, current drinkers and obese people were 1.9%, 5.5%, and 3.6% higher, respectively, among older adults in urban areas than those in non-urban areas. The percentage of older adults with a daily sleep duration of less than 6 hours was 4.0 percent higher among older adults in urban areas than in non-urban areas.

Table 1 Individual, Social, Environmental Factors, and Physical Activity among the Older Adults by Regions (N=57,106) Click for larger image

With respect to social factors, the level of trust among neighbors and exchanges of help among neighbors were 16.6% and 38.0% higher, respectively, in older adults in non-urban areas than those in urban areas. The proportions of older adults who reported meeting with neighbors, relatives (family members) and friends two or more times a week were 33.8%, 8.0%, and 5.3% higher, respectively, in non-urban areas than in urban areas. On the other hand, the percentage of older adults regularly participating in social activities at least once a month for social activities, such as religious activities, socializing activities, leisure activities and charitable activities, was higher in urban areas.

As for environmental factors, the levels of satisfaction with living conditions, public transportation and medical services were 1.5%, 15.8%, and 14.7% higher, respectively, among older adults in urban areas than those in non-urban areas. On the other hand, the percentage of people satisfied with the natural environment was 7.4% higher among older adults in non-urban areas, and the utilization rate of public health institutions was also 31.5% higher among older adults in non-urban areas, compared to those in urban areas. In terms of physical activity, among older adults in urban areas, the minimally active group constituted the largest proportion (49.6%), but among older adults in urban areas, the inactive group accounted for the largest proportion (53.4%).

2. Influencing Factors for Physical Activity in Urban Older Adults

A multiple logistic regression analysis was performed to examine factors influencing physical activity in older adults in urban areas. As a result, among individual factors, age, gender, economic activity, current smoking status, current drinking status, obesity level, sleep duration and subjective health status were identified as significant influencing factors. Regarding social factors, exchanges of help among neighbors, the frequencies of meeting with neighbors and friends, and regular participation in religious, socializing, and leisure activities were found to be significant influencing factors. As for environment factors, the experience of a fall within the past year, satisfaction with the living conditions, satisfaction with public transportation, and the use of public health facilities were found to be significant influencing factors (Table 2).

Table 2 Multivariate Logistic Regression Analysis of Factors Influencing Physical Activity by Region Click for larger image

With respect to individual factors, smokers were 0.73 times less likely to be in the inactive group than non-smokers (95% CI=0.64~0.82), while drinkers were 1.17 times more likely to be in the inactive group than non-smokers (95% CI=1.10~1.28). Regarding the degree of obesity assessed by the body mass index, the normal weight group was 1.19 times more likely to be in the inactive group (95% CI=1.10~1.28) and in the health-enhancing physical activity group (95% CI=1.03~1.38), compared to the obese group. In terms of daily sleep duration, the normal sleeping group with a sleep duration of 7~8 hours per day was 1.56 times more likely to belong to the inactive group than the oversleeping group with a sleep duration of 9 hours or more per day (95% CI=1.33~1.83).

With respect to social factors, older adults meeting with neighbors less than once per month were 0.74 times less likely to belong to the minimally active group (95% CI=0.69~0.81) and 0.62 times less likely to belong to the health-enhancing physical activity group (95% CI=0.52~0.73), compared to those who met with neighbors two or more times a week. Older adults participating in religious activities were 1.16 times more likely to belong to the minimally active group (95% CI=1.08~1.25) than those who did not. Also, older adults participating in socializing activities were 1.18 times more likely to belong to the minimally active group (95% CI=1.10~1.27) and 1.44 times more likely to belong to the health-enhancing physical activity group (95% CI=1.23~1.69) than those who did not. In addition, older adults participating in leisure activities were 1.97 times more likely to belong to the minimally active group (95% CI=1.79~2.18) and 4.02 times more likely to belong to the health-enhancing physical activity group (95% CI=3.41~4.74) than those who did not.

In terms of environmental factors, those who had experienced a fall in the past year were 0.90 times less likely to belong to the inactive group than those who did not (95% CI=0.83~0.97). Older adults who were satisfied with their living conditions were 1.15 times more likely to belong to the minimally active group than those who were not (95% CI=1.02~1.29), and those satisfied with public transportation were 1.18 times more likely to belong to the minimally active group than those who were not (95% CI=1.07~1.31). In addition, older adults using public health facilities were 1.09 times more likely to be in the minimally active group than those who did not (95% CI=1.02~1.16).

3. Influencing Factors for Physical Activity in Non-urban Older Adults

The multiple logistic regression analysis was performed to analyze factors influencing physical activity among older adults in non-urban areas, and the results showed that among individual factors, age, gender, economic activity, the number of chronic diseases, current smoking status, current drinking status, obesity level, sleep duration, and subjective health status were identified as significant influencing factors. As for social factors, the frequencies of meeting with neighbors, family and friends and the regular participation in socializing and leisure activities were identified as influencing factors. Regarding environmental factors, the experience of a fall within the past year and satisfaction with medical services were found to affect physical activity in non-urban older adults (Table 2).

In terms of individual factors, those without chronic diseases were 0.74 times less likely to belong to the minimally active group (95% CI=0.65~0.95) than those with 4 or more chronic diseases. Among older adults, smokers were 0.82 times less likely to belong to the minimally active group than non-smokers (95% CI=0.73~0.93), whereas drinkers were 1.14 times more likely (95% CI=1.06~1.23) and 1.30 times more likely (95% CI=1.14~1.48) to belong to the minimally active group and the health-enhancing physical activity group than non-drinkers. In terms of the degree of obesity, the normal weight group was 1.22 times more likely to belong to the minimally active group (95% CI=1.12~1.34) and 1.19 times more likely to belong to the health-enhancing physical activity group (95% CI=1.03~1.38), respectively, compared to the obese group. Regarding the daily sleep duration, the normal sleeping group with an average sleep duration of 7~8 hours was 1.32 times more likely to belong to the minimally active group (95% CI=1.13~1.53) and 1.60 times more likely to belong to the health-enhancing physical activity group (95% CI=1.22~2.09), compared to the oversleeping group with a sleep duration of 9 hours or more.

In terms of social factors, when the frequency of meeting between neighbors was less than once a month, the likelihoods of belonging to the minimum activity group and the health-enhancing physical activity group were 0.74 times (95% CI=0.64~0.87) and 0.38 times lower (95% CI=0.28~0.52), respectively, compared to when the frequency of meeting between neighbors was two or more times a week. Older adults who participated in social activities were 1.20 times more likely to belong to the minimally active group than those who did not (95% CI=1.11 ~1.29). Also, older adults who participated in leisure activities were 1.81 times more likely to belong to the minimally active group (95% CI=1.61~2.03) and 2.09 times more likely to belong to the health-enhancing physical activity group (95% CI=1.68~2.60) than those who did not.

Regarding environmental factors, older adults who had experienced a fall in the past year were 0.91 times less likely to belong to the minimally active group (95% CI=0.83~0.99) than those without the experience of falls. Older adults who were satisfied with medical services were 0.84 times more likely to belong to the health-enhancing physical activity group than the inactive group (95% CI=0.71~0.99).