INTRODUCTION
In 2025, South Korea is expected to become a superaged society, which is defined as having ≥20.0% of the population aged 65 years or older, and the older population is expected to show a dramatic increase in the forth-coming decades [1]. Since aging is related to physical frailty and functional limitations, the economic and health care burden related to the older population will noticeably increase in Korean society [2-5]. Based on this consideration, maintaining good physical functioning in the older population and thereby extending independent community-based life is necessary. It is broadly accepted that engaging in physical activity (PA) induces various health benefits in the older population [3,6,7]. Despite the well-known positive influence, globally, 60-70% of the older population does not meet PA guidelines, which emphasizes the need to promote PA in the older population [8]. As a countermeasure to compensate for the lack of PA, the promotion of recreational PA is recommended due to the low likelihood of the older population being active at home and work [9,10]. Recreational PA is characterized by intentional engagement in exercise and is regarded as the most important physical activity domain in public health [11]. Considering all of these factors, promoting recreational PA may contribute to extending independent community-based life by leading to the fulfillment of PA guidelines.
Recently, many researchers have employed social-ecological models to explain PA engagement, indicating the importance of the living environment in which the physical form of communities, large- and small-scale built and natural features, and the transportation system affect PA engagement [12-15]. In particular, Kahn et al. [6] highlighted that the availability and accessibility of places to engage in PA may be associated with higher levels of recreational PA. However, such studies, especially those investigating the relationship between living environment and recreational PA, are limited in South Korea. Most of the studies focused only on the total quantity of moderate-to-vigorous PA and did not focus on a specific PA domain, such as recreational PA. Thus, the present study explored the relationship between living environment and recreational-PA engagement in the older Korean population.
METHODS
Study design and subjects
For this cross-sectional study, a database composed of general health, nutritional status, and lifestyle data of the Korean population gathered during the Korea National Health and Nutritional Examination Survey (KNHANES) of 2014-2019 was utilized. The analysis of this study involved 7,594 subjects (2,413 men and 5,181 women) from the KNHANES that were aged ≥60 years, which have been used to represent an older population in many previous studies [16-19]. Each subject provided written informed consent. This study was performed under the principles of the Declaration of Helsinki and approved by the Institutional Review Board of Silla University (IRB No.: 1041449-202203-HR-001).
Living environments and RMVPA
In this study, to evaluate subjects’ living environments, their regions, towns and dwelling types were assessed. Regions were sorted into two conditions: (1) special regions (one special region and six metropolitan regions) and (2) other regions. Special regions comprised Seoul, South Korea's capital city, and Busan, Incheon, Daegu, Daejeon and Gwangju, the largest cities in each province. Towns were sorted into two conditions: (1) dongs and (2) eups and myeons. A dong is an urbanized area within a city, and eups and myeons are rural areas within a city. Dwelling types were divided into two conditions: (1) apartments and (2) houses.
To evaluate recreational PA, recreational moderate-to-vigorous physical activity (RMVPA) was adopted because that PA category primarily contributes to health promotion [20]. RMVPA was assessed by using the Global Physical Activity Questionnaire (GPAQ), which has been employed and recommended by the World Health Organization [21]. The reliability and validity of the questionnaire have been verified in several nations [22-25]. In this study, RMVPA was sorted into tertiles to evaluate the quantity of PA among the subjects.
Other parameters and covariates
Height was assessed to the nearest 0.1 cm in a standing position without shoes, and body mass was assessed to the nearest 0.1 kg using a digital electronic scale while wearing light clothing. Body mass index (BMI) was calculated as weight divided by height squared (kg/m2).
Parameters that have been considered or reported to affect the association between living conditions and RMVPA were selected as covariates: age, marital status, household income, educational level, medication use, alcohol consumption, smoking status, muscle strength, and nutrition. Marriage status was self-reported. Household income was assorted by using tertiles. Self-reported alcohol consumption was categorized as never, ≤1 time/week, 2-3 times/week, or ≥4 times/week; educational level was categorized as primary, middle and high school, or college or beyond; and smoking status was categorized as never, former, or current smoker [26,27]. Medication use was recorded based on a self-reported prescription from a physician. Muscle strength was evaluated by utilizing handgrip strength, which was assessed three times each for the right and left hands, and then the average value was adopted [28]. Nutritional data, including total energy, carbohydrate, protein and fat intake, were gathered by using a food frequency questionnaire composed of 63 food items that are considered critical energy sources and nutrients. The questionnaire was designed as an open-ended survey for reporting a variety of dishes and foods using the 24-h recall method with various measuring aids [29].
Statistical analysis
Data are presented as the mean±standard deviation (SD), percentage and number, or odds ratio and 95% confidence interval (CI). The Kol-mogorov‒ Smirnov test was employed to identify the normality of continuous variables (Table 1). The independent t test or the Mann‒ Whitney U test was adopted to compare variables between males and females (Table 2). The overall and sex-specific distributions of living conditions by RMVPA tertiles were explored using chi-square (Table 3). Logistic regression was adopted to identify the sex-specific odds ratio for the associations of the regions, towns and dwelling types with the middle and highest tertile of RMVPA (Tables 4 and 5). Model 1 was unadjusted; Model 2 was adjusted for Model 1 plus age; Model 3 was adjusted for Model 2 plus marriage status, household income and educational level; Model 4 was adjusted for Model 3 plus medication use; Model 5 was adjusted for Model 4 plus drinking and smoking status; and Model 6 was adjusted for Model 5 plus muscle strength and nutrition. Statistical analyses were conducted with SPSS software, version 26.0 (IBM Corp., Armonk, NY, USA). p <0.05 was considered to indicate statistical significance.
Table 1
The normality tests of continuous variables
Table 2
Characteristics of study subjects
| Variables | Total (n=7,594) | Male (n=2,413) | Female (n=5,181) | p |
|---|---|---|---|---|
| RMVPA (min/wk)1 | 73.1±195.0 | 104.4±257.6 | 58.6±155.4 | <0.001 |
| Dwelling region | 0.472 | |||
| One special and six metropolitan regions | 3,325 (43.8) | 1,071 (44.4) | 2,254 (43.5) | |
| Other regions | 4,269 (56.2) | 1,342 (55.6) | 2,927 (56.5) | |
| Dwelling town | 0.395 | |||
| Dongs | 5,652 (74.4) | 1,811 (75.1) | 3,841 (74.1) | |
| Eups and myeons | 1,942 (25.6) | 602 (24.9) | 1,340 (25.9) | |
| Dwelling type | 0.535 | |||
| Apartment | 4,624 (60.9) | 1,457 (60.4) | 3,167 (61.1) | |
| House | 2,970 (39.1) | 9,560 (39.6) | 2,014 (38.9) | |
| Age (y)1 | 70.0±6.5 | 69.9±6.4 | 70.0±6.5 | 0.447 |
| Height (cm)1 | 161.4±8.3 | 160.3±8.1 | 161.9±8.4 | <0.001 |
| Body weight (kg)1 | 62.8±10.9 | 61.8±10.6 | 63.3±11.0 | <0.001 |
| Body mass index (kg/m2) | 24.1±3.3 | 24.0±3.3 | 24.1±3.3 | 0.201 |
| Handgrip strength (kg)1 | 27.4±8.9 | 26.2±8.4 | 28.0±9.0 | <0.001 |
| Total energy intake (kcal/d)1 | 1,807.6±751.6 | 1,499.1±399.8 | 1,951.2±829.9 | <0.001 |
| Carbohydrate intake (g/d)1 | 285.8±114.2 | 247.9±75.7 | 303.4±124.4 | <0.001 |
| Protein intake (g/d)1 | 63.4±32.1 | 52.8±20.3 | 68.3±35.3 | <0.001 |
| Fat intake (g/d)1 | 36.8±27.1 | 29.0±16.5 | 40.5±30.2 | <0.001 |
| Marriage status | <0.05 | |||
| Married | 7,519 (99.0) | 2,381 (98.7) | 5,138 (99.2) | |
| Unmarried | 75 (1.0) | 32 (1.8) | 43 (0.8) | |
| Medication | ||||
| Hypertension | 3,720 (49.0) | 1,124 (46.6) | 2,596 (50.1) | <0.01 |
| Diabetes | 1,427 (18.8) | 474 (19.6) | 953 (18.4) | 0.194 |
| Hyperlipidemia | 1,994 (26.3) | 433 (17.9) | 1,561 (30.1) | <0.001 |
| Household income | <0.001 | |||
| Low | 3,092 (40.9) | 817 (34.0) | 2,275 (44.1) | |
| Lower-middle | 2,091 (27.7) | 698 (29.0) | 1,393 (27.0) | |
| Upper middle | 1,365 (17.9) | 487 (20.2) | 869 (16.9) | |
| High | 1,019 (13.5) | 403 (16.8) | 616 (12.0) | |
| Education level | <0.001 | |||
| Primary school | 4,194 (55.5) | 851 (35.5) | 3,343 (64.9) | |
| Middle school | 1,209 (16.0) | 443 (18.5) | 766 (14.9) | |
| High school | 1,364 (18.1) | 642 (26.8) | 722 (14.0) | |
| College | 783 (10.4) | 463 (19.3) | 320 (6.2) | |
| Alcohol consumption | <0.001 | |||
| Never | 3,732 (49.1) | 698 (28.9) | 3,034 (58.6) | |
| ≤Once a week | 2,750 (36.2) | 875 (36.3) | 1,875 (36.2) | |
| 2-3 Times/week | 637 (8.4) | 447 (18.5) | 190 (3.7) | |
| ≥4 Times/week | 475 (6.3) | 393 (16.3) | 82 (1.6) | |
| Smoking | <0.001 | |||
| Never | 5,365 (70.6) | 496 (20.6) | 4,869 (94.0) | |
| Former smoking | 1,574 (20.7) | 1,400 (58.0) | 174 (3.4) | |
| Current smoking | 655 (8.6) | 517 (21.4) | 138 (2.7) |
Table 3
Sex-specific differences of dwelling conditions from the lowest to highest recreational moderate-to-vigorous recreational physical activity tertiles
Table 4
Sex specific odds ratio for the association between the dwelling regions, towns or types and the middle tertile of recreational moderate-to-vigor ous physical activity
Table 5
Sex specific odds ratio for the association between the dwelling regions, towns or types and the highest tertile of recreational moderate-to-vigorous physical activity
RESULTS
Table 1 displays the normality test results of continuous variables. All the variables, including RMVPA, age, height, body weight, body mass index, handgrip strength, and total energy, carbohydrate, protein and fat intake, had nonnormal distributions.
Table 2 exhibits the characteristics of the study subjects. The average RMVPA was 73.1 min/week, and the quantity of RMVPA in male subjects was greater than that in females (p <0.001). There were no significant differences in living conditions between males and females, including for region, town and dwelling type, age and BMI. Height and body weight in female subjects were greater than those in male subjects (p < 0.001 for both).
Table 3 shows the sex-specific differences of living conditions from the lowest to highest RMVPA tertiles. For the differences among RMVPA categories, in both male and female subjects, all living conditions, such as regions, towns and dwelling types, indicated significant differences (p <0.001 for all). The percentages of lowest tertiles of RMVPA in other regions, eups and myeons, and houses were higher than those in one special and six metropolitan regions, dongs and apartments. However, the percentages of middle and highest tertiles in one special and six metropolitan regions, dongs and apartments are higher than those in other regions, eups and myeons, and houses.
The sex-specific odds ratios for the associations of regions, towns and dwelling types with the middle tertile of RMVPA are shown in Table 4. Regarding male subjects’ dwelling regions, in Models 1, 2, 3, 4, 5, and 6, relative to other regions, special regions had odds ratios of 1.59 (1.30-2.93), 1.59 (1.30-1.93), 1.57 (1.29-1.92), 1.58 (1.29-1.93), 1.57 (1.29-1.92), and 1.56 (1.27-1.91) for the highest tertile of RMVPA, respectively (p <0.001 for all). Regarding male subjects’ dwelling towns, in Models 1, 2, 3, 4, 5, and 6, relative to eups and myeons, dongs had odds ratios of 1.94 (1.54-2.43), 1.94 (1.55-2.44), 1.89 (1.50-2.39), 1.90 (1.51-2.40), 1.93 (1.52-2.44), and 1.94 (1.53-2.46) for the highest tertile of RMVPA, respectively (p <0.001 for all). Regarding male subjects’ dwelling types, in Models 1, 2, 3, 4, 5, and 6, relative to houses, apartments had odds ratios of 1.45 (1.19-1.78), 1.46 (1.19-1.78), 1.40 (1.14-1.73), 1.40 (1.14-1.73), 1.41 (1.14-1.74), and 1.42 (1.15-1.75) for the highest tertile of RMVPA, respectively (p <0.001 for all). Concerning female subjects’ dwelling regions, in Models 1, 2, 3, 4, 5, and 6, relative to other regions, special regions had odds ratios of 1.77 (1.56-2.01), 1.70 (1.49-1.93), 1.64 (1.44-2.87), 1.63 (1.43-1.86), 1.64 (1.44-1.87), and 1.66 (1.46-1.90) for the highest tertile of RMVPA, respectively (p <0.001 for all). Concerning female subjects’ dwelling towns, in Models 1, 2, 3, 4, 5, and 6, relative to eups and myeons, dongs had odds ratios of 2.04 (1.76-2.36), 1.90 (1.64-2.20), 1.80 (1.55-2.10), 1.80 (1.55-2.10), 1.80 (1.55-2.10), and 1.78 (1.53-2.08) for the highest tertile of RMVPA, respectively (p <0.001 for all). Concerning female subjects’ dwelling types, in Models 1, 2, 3, 4, 5, and 6, relative to houses, apartments had odds ratios of 1.39 (1.22-1.58), 1.32 (1.16-1.50), 1.21 (1.06-1.39), 1.21 (1.06-1.39), 1.22 (1.07-1.40), and 1.24 (1.08-1.43) for the highest tertile of RMVPA, respectively (p <0.01 for all).
The sex-specific odds ratios for the associations of regions, towns and dwelling types with the highest tertile of RMVPA are shown in Table 5. Regarding male subjects’ dwelling regions, in Models 1, 2, 3, 4, 5, and 6, relative to other regions, special regions had odds ratios of 1.89 (1.55-2.30), 1.87 (1.54-2.27), 1.81 (1.48-2.21), 1.83 (1.50-2.23), 1.83 (1.49-2.23), and 1.80 (1.47-2.20) for the highest tertile of RMVPA, respectively (p <0.001 for all). Regarding male subjects’ dwelling towns, in Models 1, 2, 3, 4, 5, and 6, relative to eups and myeons, dongs had odds ratios of 2.67 (2.09-3.40), 2.63 (2.06-3.36), 2.48 (1.94-3.18), 2.51 (1.96-3.22), 2.57 (2.00-3.30), and 2.57 (1.99-3.31) for the highest tertile of RMVPA, respectively (p <0.001 for all). Regarding male subjects’ dwelling types, in Models 1, 2, 3, 4, 5, and 6, relative to houses, apartments had odds ratios of 1.65 (1.36-2.01), 1.62 (1.33-1.97), 1.50 (1.22-1.84), 1.51 (1.22-1.85), 1.51 (1.22-1.86), and 1.52 (1.23-1.87) for the highest tertile of RMVPA, respectively (p <0.001 for all). Concerning female subjects’ dwelling regions, in Models 1, 2, 3, 4, 5, and 6, relative to other regions, special regions had odds ratios of 1.99 (1.73-2.30), 1.92 (1.66-2.21), 1.86 (1.60-2.15), 1.83 (1.58-2.12), 1.84 (1.58-2.13), and 1.85 (1.60-2.14) for the highest tertile of RMVPA, respectively (p <0.001 for all). Concerning female subjects’ dwelling towns, in Models 1, 2, 3, 4, 5, and 6, relative to eups and myeons, dongs had odds ratios of 2.54 (2.13-3.03), 2.37 (1.99-2.84), 2.27 (1.89-2.72), 2.24 (1.86-2.86), 2.24 (1.87-2.69), and 2.23 (1.85-2.67) for the highest tertile of RMVPA, respectively (p <0.001 for all). Concerning female subjects’ dwelling types, in Models 1, 2, 3, 4, 5, and 6, relative to houses, apartments had odds ratios of 1.40 (1.20-1.60), 1.32 (1.14-1.53), 1.21 (1.04-1.41), 1.19 (1.02-1.39), 1.20 (1.03-1.40), and 1.21 (1.03-1.41) for the highest tertile of RMVPA, respectively (p <0.05 for all).
DISCUSSION
The purpose of the present study was to explore the relationship between living environments, such as regions, towns and dwelling types, and RMVPA engagement in a sample taken from an older Korean population. The percentages of lowest tertiles of RMVPA in other regions, eups and myeons, and houses were higher than those in one special and six metropolitan regions, dongs and apartments. However, the percentages of middle and highest tertiles showed the opposite results (p <0.001 for all). Subjects living in special regions, dongs and apartments were 1.56, 1.94, 1.42, 1.66, 1.78, and 1.24 times more likely to be in the middle RM-VPA tertile in male and female subjects, respectively, relative to those in other regions, eups and myeons, and houses (p <0.01 for all). Those living in special regions, dongs and apartments were 1.80, 2.57, 1.52, 1.85, 2.23, and 1.21 times more likely to be in the highest RMVPA tertile in male and female subjects, respectively, relative to those in other regions, eups and myeons, and houses (p <0.05 for all). These findings suggest that living in an apartment in an urbanized large city may positively contribute to RMVPA engagement in the older Korean population, regardless of sex.
The present study showed that older population dwelling in an apartment in a large, urbanized city may be more likely to participate in RM-VPA than those dwelling in a house in a small rural area. This phenomenon can be explained by two previous studies that demonstrated why and how an exercise facility-rich environment may encourage recreational PA as follows. First, the inconvenience of accessing exercise facilities is mentioned as the reason for not participating in regular recreational PA. Nearby facilities decrease some of the barriers to engagement in recreational PA in that travel time and traffic-associated barriers are decreased [13,30,31]. Second, exercise facilities work as a psychological stimulus to change exercise habits. People dwelling near exercise facilities may frequently pay attention to exercise [31]. In the present study, even though the authors could not directly investigate whether accessing exercise facilities is convenient for older population in an apartment in an urbanized large city, it can be presumed that recently, in South Korea's urbanized large cities, there were many public or private exercise facilities, and it was easy to access these facilities via a well-organized transportation system, even for older population. In addition, because most recently built apartment buildings have resident-only exercise facilities, dwelling in apartments also positively affects participation in RMVPA. However, in the case of small rural areas, there is still a lack of exercise facilities, and it would be difficult for older population to access those facilities without private cars. Considering all of these factors, it is suggested that frequent exposure to exercise facilities in an apartment building in a large, urbanized city may increase motivation for recreational PA, and easy availability and accessibility to exercise facilities for older population may allow them to carry out habitual engagement in RMVPA.
On the one hand, Mohammad and his colleagues reported that better availability of physical activity destinations is important, but many things, including building design, transport infrastructure, residential density and street connectivity, affect people's behavior [30,32,33]. Given the reports, to clearly confirm the relationship between the living environment and RMVPA, all potential factors mentioned above should be considered. However, in the present study, the relationship between the living environment and RMVPA was investigated based on information regarding where the subjects lived. Thus, further investigation, which considers all of the potential factors, would be helpful to confirm the conclusion yielded by the present study. Additionally, since RMVPA data were collected by utilizing a questionnaire, there is a possibility of self-selection bias. Objective assessment of RMVPA would provide stronger evidence. Although those limitations exist in the present study, a strong strength is also recognized. The authors adjusted for several potential covariates, including age, marriage status, household income, educational level, medication use, alcohol consumption, smoking status, muscle strength, and nutrition, that may affect the RMVPA engagement of older population.
In conclusion, in these aged participants, living in an apartment in an urbanized large city was positively associated with the quantity of RM-VPA, regardless of sex. This finding suggests that enough nearby exercise facilities and accessibility improvement may contribute to an increase in recreational PA engagement in the older Korean population.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
