Abstract
[Purpose]
Depression is a growing mental health issue among older adults in South Korea, where rapid population aging has intensified the need for effective preventive strategies. Although physical activity is recognized as a protective factor against depression, limited research has examined how specific physical activity domains relate to depression among older Korean adults. This study investigated the associations between occupational, recreational, and commuting-related physical activity and depression severity and evaluated the effects of activity intensity and sociodemographic factors.
[Methods]
Data were drawn from the 2022 Korea National Health and Nutrition Examination Survey and included 1,331 adults aged 65 or older. Depression severity was assessed using Patient Health Questionnaire-9 (PHQ-9). Physical activity levels were quantified using metabolic equivalent (MET-minute) thresholds. Data were statistically analyzed using descriptive statistics, chi-square tests, and ordinal logistic regression.
[Results]
Of the participants, 85.27% reported no depression, while 14.73% experienced mild to severe symptoms. Chi-square analysis showed that vigorous-intensity occupational activity and moderate-intensity recreational activity were significantly associated with depression severity (p < 0.05). However, in the ordinal logistic regression model adjusting for covariates, only vigorous occupational activity remained statistically significant (β = 1.651, OR = 5.21, p = 0.013). Gender, household income, and education were significant predictors, whereas self-rated health and commuting activity were not. The positive regression coefficient indicated that the participants engaging in vigorous-intensity occupational activity had higher odds of belonging to more severe depression categories.
[Conclusion]
These findings highlight the importance of designing tailored interventions for vulnerable subgroups, including older women individuals with a low socioeconomic status.
Keywords: late-life mental health, MET, PHQ-9 depressive symptoms, sociodemographic disparities
INTRODUCTION
With rapid society development, population aging has become an increasingly prominent issue. The World Health Organization [1] has projected that the global population aged 65 years and older will reach 1.6 billion by 2050, accounting for approximately 16% of the total population. However, this demographic shift occurs more rapidly in South Korea; as of 2022, individuals aged 65 years and older already constituted 17.5% of the population, surpassing the global average [2]. With this accelerated aging trend, studies should focus on South Korea’s older population.
Population aging has profoundly altered demographic structures, leading to a substantial rise in the prevalence and severity of health challenges among older adults, especially in the mental health domain [3]. Among these challenges, depression stands out as one of the most common mental health issues [4]. It is a multifaceted disorder affecting mental and physical health; it is characterized by persistent low mood, diminished interest, and reduced pleasure lasting for at least 2 weeks [5]. Pathophysiologically, it is associated with neurotransmitter imbalances (e.g., serotonin, norepinephrine, and dopamine), decreased brain-derived neurofactor (BDNF) levels, and neuro-inflammatory responses [6-8]. It considerably diminishes the quality of life of older adults and increases the risk of chronic diseases such as cardiovascular disease, diabetes, and Alzheimer’s disease [9,10].
Physical activity is widely recognized as a pivotal intervention that can improve mental health [11]. Certain physical activities have been effective in alleviating depressive symptoms in older adults. For example, aerobic exercises such as walking and swimming help mitigate mild depression in this demographic [12]. In addition, moderate-intensity activities, including gardening and walking, not only reduce mild to moderate depressive symptoms but also offer high sustainability and accessibility because they require no specialized equipment or substantial physical exertion [13,14]. Even minor increases in light physical activity, such as 30 additional minutes per day, can considerably decrease the risk of depression among older adults [15]. Furthermore, the scores of older individuals who engage in regular exercise on the PHQ-9 depression scale are lower than those of their sedentary counterparts [16].
Despite the growing body of evidence demonstrating that physical activity exhibits protective effects against depression, important gaps remain in the existing literature, particularly in the context of rapidly aging societies such as South Korea. Although previous research consistently reported that the intensity and type of physical activity are associated with depressive symptoms among older adults, most studies focused on overall activity volume or general exercise participation rather than examining domain-specific patterns of physical activity in detail. Moreover, findings derived from Western populations may not be directly generalizable to Korean older adults, whose cultural context, daily activity patterns, and social environments differ substantially.
Given that different physical activity domains (e.g., recreational, commuting, household, and occupational activities) and varying intensity levels may distinctly influence depressive symptoms, a more nuanced understanding of these relationships is needed. However, few empirical studies have comprehensively examined the associations between physical activity domains, intensity levels, and depression among Korean older adults. Therefore, this study aimed to investigate the relationship between domain-specific physical activity (including type and intensity) and depressive symptoms among older adults in South Korea. Specifically, this study aimed to examine how different physical activity domains were associated with depression and identify whether varying intensity and volume levels of physical activity differentially influenced depressive symptoms in this population.
METHODS
Selection of Research Materials and Subjects
Original data from the 2022 Korea National Health and Nutrition Examination Survey (KNHANES), a comprehensive nationwide assessment of health and nutrition status conducted annually by the Korea Centers for Disease Control and Prevention (KCDC) since 1998, were used in this study. The 2022 dataset, which included Patient Health Questionnaire-9 (PHQ-9) results, represented the most current and authoritative resource for evaluating the health and mental well-being of South Korea’s population, thereby reflecting recent trends in depressive symptoms. The research subjects were selected as follows: the original 2022 survey dataset with 6,265 participants from 17 regions nationwide. After 906 individuals under the age of 18 years and 3,976 individuals under the age of 65 years were excluded, 1,669 older adults were retained. In addition, 52 individuals with incomplete or missing responses to exercise-related or PHQ-9 questions were excluded. Thus, 1,331 older adults were included in the final analysis.
Measurement of Variables
Depression scoring criteria (PHQ-9 score)
The association between physical activity patterns and depressive symptoms in older adults was investigated using PHQ-9, which includes nine items evaluating symptom-related aspects and one item assessing overall functional impairment, including symptoms such as low mood, loss of interest, sleep disturbances, fatigue, changes in appetite, feeling of worthlessness, concentration difficulties, psychomotor retardation, and suicidal ideation. The participants were instructed to rate the frequency of depressive symptoms over the previous 2 weeks by using a four-point scale: not at all (0 points), have a few days (1 point), more than half of the days (2 points), and nearly every day (3 points). Total scores ranging from 0 to 27 were classified as follows: 0–4, no depression; 5–9, mild depression; 10–14, moderate depression; 15–19, moderately severe depression; and 20–27, severe depression. PHQ-9 is widely recognized for its reliability and validity, particularly in older adults. In the Korean context, it has demonstrated high internal consistency (Cronbach’s α = 0.91) and validity for assessing depressive symptoms in this demographic [17].
Physical activity types, levels, and intensity
Physical activity behavior was assessed on the basis of the participants’ self-reported engagement in physical activities over the past 7 days. Data on the type of physical activity (recreation, commuting-related, and work-related) and its intensity (moderate or vigorous intensity) were collected. The intensity of physical activity was quantified using the metabolic equivalents of task (MET) system. MET-minutes per week was calculated using the following formula: Physical activity volume (MET-min/week) = MET value × activity time (min) × activity frequency (days/week).
The MET value represented the ratio of energy expenditure during an activity to the energy expenditure at rest. Moderate-intensity activities typically have MET values between 3.0 and 6.0, while vigorous-intensity activities have MET values greater than 6.0 [18]. According to a previous study [19], MET values for moderate- and vigorous-intensity activities were standardized at 4 and 8, respectively. MET-minutes per week was calculated as follows: moderate-intensity MET = 4 × activity time (min) × activity frequency (days/week) and vigorous-intensity MET = 8 × activity time (min) × activity frequency (days/week).
Statistical Analysis
Data were statistically analyzed using SPSS 26.0 (IBM Corp., Armonk, NY, USA), and statistical significance was set at p < 0.05. The participants’ characteristics and the distribution of key study variables were summarized using descriptive statistics. Descriptive statistics were primarily reported as frequencies and percentages because several variables were operationalized using established categorical classifications (e.g., PHQ-9 depression severity groups and BMI categories) and primary analyses were based on categorical comparisons. Two complementary analytical approaches were utilized to examine the relationship between physical activity and depression. First, physical activity was analyzed according to type and intensity (e.g., occupational vs. recreational activity and moderate vs. vigorous intensity), which reflects the context and qualitative characteristics of the activity. Second, physical activity was examined on the basis of the total activity volume by using MET-minutes per week, which represents the overall quantity of energy expenditure accumulated through physical activity regardless of its context. Through these approaches, different dimensions of physical activity behavior can be captured; therefore, different statistical patterns may be produced. The associations between depression levels, various physical activity types (recreational, commuting, and occupational), and physical activity intensities based on MET classifications were examined using chi-square tests. Group differences in the distribution of depression severity were assessed. Given that PHQ-9 depression severity is an ordinal categorical variable (none, mild, moderate, moderately severe, and severe), ordinal logistic regression models were used to further analyze the independent effects of physical activity type and intensity on depression severity. In the proportional odds model, an odds ratio (OR) of > 1 indicated the increased odds of being in a higher (more severe) depression category, whereas an OR of < 1 indicated lower odds of more severe depression. The proportional odds assumption underlying the ordinal logistic regression model was assessed using the test of parallel lines. A non-significant result (p > 0.05) denoted that the proportional odds assumption was satisfied, and the ordinal logistic regression model was appropriate for the data. Sociodemographic variables, including gender, household income, education level, and self-rated health, were included as covariates. Regression outcomes were reported using regression coefficients (β), Wald statistics, p-values, and 95% confidence intervals (CI). The overall model fit was assessed using likelihood ratio tests. Additional continuous descriptive statistics for key variables (i.e., age, BMI, PHQ-9 scores, and MET-min/week) were calculated (Supplementary Table S1) to enhance the characterization of the study sample.
RESULTS
Demographic Characteristics of the Participants
Table 1 includes 1,331 older Korean adults, of whom 45.08% were men and 54.92% were women. Regarding the body mass index (BMI) distribution, 36.21% of the participants had a BMI of ≥ 25. A family history of chronic diseases was reported by 30.28% of the participants, and the health of 75.73% of the participants was rated as good. Income distribution was relatively balanced between low- and high-income groups. In terms of educational attainment, 47.63% completed only primary school. Supplementary Table S1 provides additional continuous descriptive statistics, including mean and standard deviation values for age, BMI, and PHQ-9 scores stratified by sex and supplementary MET-min/week summaries for physical activity variables.
Table 1.
Demographic characteristics of the participants
| Variable | Category | Gender (%) |
Total (n = 1,331) | |
|---|---|---|---|---|
| Men (n = 600) | Women (n = 731) | |||
| Body mass index (BMI) | BMI < 23 | 230 (38.33) | 265 (36.25) | 495 (37.19) |
| 23 ≤ BMI < 25 | 169 (28.17) | 185 (25.31) | 354 (26.60) | |
| BMI ≥ 25 | 201 (33.50) | 281 (38.44) | 482 (36.21) | |
| Family history of chronic diseases | Yes | 229 (38.17) | 174 (23.80) | 403 (30.28) |
| No | 371 (61.83) | 557 (76.20) | 928 (69.72) | |
| Self-rated health | Yes | 475 (79.17) | 533 (72.91) | 1,008 (75.73) |
| No | 125 (20.83) | 198 (27.09) | 323 (24.27) | |
| Income | Low | 139 (23.17) | 170 (23.26) | 309 (23.22) |
| Mid-low | 142 (23.67) | 186 (25.44) | 328 (24.64) | |
| Mid-high | 149 (24.83) | 185 (25.31) | 334 (25.09) | |
| High | 170 (28.33) | 190 (25.99) | 360 (27.05) | |
| Education level | Primary school | 193 (32.17) | 441 (60.33) | 634 (47.63) |
| Middle school | 107 (17.83) | 108 (14.77) | 215 (16.15) | |
| High school | 174 (29.00) | 126 (17.24) | 300 (22.54) | |
| College school | 126 (21.00) | 56 (7.66) | 182 (13.67) | |
Depression and Physical Activity Types/Levels of the Participants
Table 2 shows the following results: 85.27% of the participants with depression, 10.67% with mild depression, 2.70% with moderate depression, 0.90% with moderately severe depression, and 0.45% with severe depression. Depression prevalence was higher among women than men (19.69% vs. 8.67%). Engagement in vigorous-intensity occupational or recreational activity was extremely low (0.75% and 2.70%, respectively), whereas moderate-intensity recreational activity was more common (18.41%). Such vigorous-intensity occupational activity was extremely rare in this sample (0.75%, n = 10). Given this very small subgroup size, the regression estimates for this variable should be cautiously interpreted because of the potential for sparse data bias and unstable effect estimates. Furthermore, 50.71% of the participants reported commuting-related physical activity, suggesting that older adults maintained a certain level of daily activity.
Table 2.
Depression and physical activity types/levels of the participants
| Variable | Category | Gender (%) |
Total (n = 1,331) | |
|---|---|---|---|---|
| Men (n = 600) | Women (n = 731) | |||
| Depression group | No depressive symptoms | 548 (91.33) | 587 (80.30) | 1,135 (85.27) |
| Mild depression | 40 (6.67) | 102 (13.95) | 142 (10.67) | |
| Moderate depression | 9 (1.50) | 27 (3.69) | 36 (2.70) | |
| Moderately severe depression | 2 (0.33) | 10 (1.37) | 12 (0.90) | |
| Severe depression | 1 (0.17) | 5 (0.68) | 6 (0.45) | |
| Vigorous-intensity physical activity (occupational) | Yes | 7 (1.17) | 3 (0.41) | 10 (0.75) |
| No | 593 (98.83) | 728 (99.59) | 1,321 (99.25) | |
| Vigorous-intensity physical activity (recreational) | Yes | 29 (4.83) | 7 (0.96) | 36 (2.70) |
| No | 571 (95.17) | 724 (99.04) | 1,295 (97.30) | |
| Moderate-intensity physical activity (occupational) | Yes | 36 (6.00) | 36 (4.92) | 72 (5.41) |
| No | 564 (94.00) | 695 (95.08) | 1,259 (94.59) | |
| Moderate-intensity physical activity (recreational) | Yes | 161 (26.83) | 84 (11.49) | 245 (18.41) |
| No | 439 (73.17) | 647 (88.51) | 1,086 (81.59) | |
| Physical activity (commuting) | Yes | 284 (47.33) | 391 (53.49) | 675 (50.71) |
| No | 316 (52.67) | 340 (46.51) | 656 (49.29) | |
Effect of Demographic Characteristics on Depression Levels
Gender, income, self-rated health, and education attainment were significantly associated with depression severity (p < 0.05; Table 3). Specifically, older women, those with lower income, and lower education levels demonstrated higher levels of depressive symptoms. Conversely, BMI and family history of chronic diseases were not significantly associated with depression severity.
Table 3.
Impact of demographic characteristics on depression levels
| Variable | Category | Depression group (%) |
Total | χ 2 | p | ||||
|---|---|---|---|---|---|---|---|---|---|
| No depressive symptoms | Mild depression | Moderate depression | Moderately severe depression | Severe depression | |||||
| Sex | Men | 548 (48.28) | 40 (28.17) | 9 (25.00) | 2 (16.67) | 1 (16.67) | 600 (45.08) | 32.835 | 0.000** |
| Women | 587 (51.72) | 102 (71.83) | 27 (75.00) | 10 (83.33) | 5 (83.33) | 731 (54.92) | |||
| Body mass index | BM < 23 | 418 (36.83) | 49 (34.51) | 19 (52.78) | 8 (66.67) | 1 (16.67) | 495 (37.19) | 11.925 | 0.155 |
| 23 ≤ BMI < 25 | 299 (26.34) | 41 (28.87) | 8 (22.22) | 3 (25.00) | 3 (50.00) | 354 (26.60) | |||
| BMI ≥ 25 | 418 (36.83) | 52 (36.62) | 9 (25.00) | 1 (8.33) | 2 (33.33) | 482 (36.21) | |||
| Family history of chronic diseases | Yes | 357 (31.45) | 36 (25.35) | 7 (19.44) | 3 (25.00) | 0 (0.00) | 403 (30.28) | 7.141 | 0.129 |
| No | 778 (68.55) | 106 (74.65) | 29 (80.56) | 9 (75.00) | 6 (100.00) | 928 (69.72) | |||
| Self-rated health | Yes | 869 (76.56) | 103 (72.54) | 28 (77.78) | 5 (41.67) | 3 (50.00) | 1,008 (75.73) | 11.038 | 0.026* |
| No | 266 (23.44) | 39 (27.46) | 8 (22.22) | 7 (58.33) | 3 (50.00) | 323 (24.27) | |||
| Income | Low | 244 (21.50) | 46 (32.39) | 10 (27.78) | 6 (50.00) | 3 (50.00) | 309 (23.22) | 29.118 | 0.004** |
| Mid-low | 278 (24.49) | 33 (23.24) | 15 (41.67) | 2 (16.67) | 0 (0.00) | 328 (24.64) | |||
| Mid-high | 289 (25.46) | 38 (26.76) | 4 (11.11) | 2 (16.67) | 1 (16.67) | 334 (25.09) | |||
| High | 324 (28.55) | 25 (17.61) | 7 (19.44) | 2 (16.67) | 2 (33.33) | 360 (27.05) | |||
| Education level | Primary school | 506 (44.58) | 86 (60.56) | 26 (72.22) | 11 (91.67) | 5 (83.33) | 634 (47.63) | 40.545 | 0.000** |
| Middle school | 187 (16.48) | 23 (16.20) | 3 (8.33) | 1 (8.33) | 1 (16.67) | 215 (16.15) | |||
| High school | 270 (23.79) | 24 (16.90) | 6 (16.67) | 0 (0.00) | 0 (0.00) | 300 (22.54) | |||
| College | 172 (15.15) | 9 (6.34) | 1 (2.78) | 0 (0.00) | 0 (0.00) | 182 (13.67) | |||
p < 0.05
p < 0.01
More specifically, the prevalence of depression in older women was markedly higher than that in men (71.83% vs. 28.17%). These gender differences were also pronounced in moderate depression (75.00% vs. 25.00%), moderately severe depression (83.33% vs. 16.67%), and severe (83.33% vs. 16.67%) depression categories; therefore, gender disparities persist across progressive levels of symptom severity rather than emerging only at the early stages. The prevalence of depression of low-income older adults was also higher (32.39%) than that of high-income individuals (17.61%). Furthermore, the proportions of moderate (27.78%), moderately severe (50.00%), and severe (50.00%) depression of those in the low-income group were considerably higher than those in their higher-income counterparts (moderate, 16.67%; moderately severe, 16.67%; and severe, 33.33%). These results suggested that socioeconomic vulnerability might contribute to the onset and exacerbation of depressive symptoms. Furthermore, the depression prevalence of older adults with only primary education was substantially higher (60.56%) than that of older adults with higher educational attainment (6.34%). The former also accounted for the largest shares of the moderate (72.22%), moderately severe (91.67%), and severe (83.33%) depression groups; thus, lower educational attainment was associated with greater symptom severity rather than a sole increase in prevalence. The prevalence of depression of the participants with self-rated poor health was higher (27.46%) than that of the participants with self-rated good health (72.54%). Those with self-rated poor health showed disproportionately higher proportions of moderate (22.22%), moderately severe (58.33%), and severe (50.00%) depression than the healthy participants did (moderate: 7.78%, moderately severe: 5.00%, severe: 5.00%). These findings suggested that subjective health perception might play a progressive role in predicting the occurrence and worsening of depression.
Relationship between the type/intensity of physical activity (occupational, recreational, and commuting) and depression in older Korean adults
Table 4 examines the relationship between various types/intensities of physical activity and depression levels among older adults in Korea. Vigorous-intensity occupational activity was significantly associated with depression levels (p < 0.01). Older adults who engaged in vigorous-intensity occupational activities exhibited a lower prevalence of depression (70% with no depression vs. 10% with severe depression). Moderate-intensity recreational activity was also significantly related to depression (p < 0.05). Older adults who participated in moderate-intensity recreational activities demonstrated a lower prevalence of depression (88.98% with no depression vs. 1.22% with severe depression). Conversely, vigorous-intensity recreational activity was not significantly linked with depression (p = 0.237) likely because of the low participation rate and limited sample size, which possibly reduced statistical power. Similarly, moderate-intensity occupational activity and commuting-related activity were not significantly associated with depression levels (p > 0.05).
Table 4.
Chi-square test for the relationship between type/intensity of physical activity and depression
| Variable | Category | Depression group (%) |
χ 2 | p | ||||
|---|---|---|---|---|---|---|---|---|
| No depressive symptoms | Mild depression | Moderate depression | Moderately severe depression | Severe depression | ||||
| Vigorous-intensity physical activity (occupational) | Yes | 7 (70.00) | 1 (10.00) | 0 (0.00) | 1 (10.00) | 1 (10.00) | 30.185 | 0.001** |
| No | 1,128 (85.39) | 141 (10.67) | 36 (2.73) | 11 (0.83) | 5 (0.38) | |||
| Vigorous-intensity physical activity (recreational) | Yes | 28 (77.78) | 7 (19.44) | 0 (0.00) | 1 (2.78) | 0 (0.00) | 5.527 | 0.237 |
| No | 1,107 (85.48) | 135 (10.42) | 36 (2.87) | 11 (0.85) | 6 (0.46) | |||
| Moderate-intensity physical activity (occupational) | Yes | 56 (77.78) | 12 (16.67) | 3 (4.17) | 0 (0.00) | 1 (1.39) | 5.842 | 0.211 |
| No | 1,079 (85.70) | 130 (10.33) | 33 (2.62) | 12 (0.95) | 5 (0.40) | |||
| Moderate-intensity physical activity (recreational) | Yes | 218 (88.98) | 19 (7.76) | 5 (2.04) | 0 (0.00) | 3 (1.22) | 10.057 | 0.039* |
| No | 917 (84.44) | 123 (11.33) | 31 (2.85) | 12 (1.10) | 3 (0.28) | |||
| Physical activity (commuting) | Yes | 581 (86.07) | 68 (10.07) | 18 (2.67) | 5 (0.74) | 3 (0.44) | 0.958 | 0.916 |
| No | 554 (84.45) | 74 (11.28) | 18 (2.74) | 7 (1.07) | 3 (0.46) | |||
p < 0.05
p < 0.01
Ordinal logistic regression analysis of physical activity type/intensity on depression in older Korean adults
In Table 5, after adjustments for gender, self-rated health, income, and education attainment, vigorous-intensity occupational activity remained significantly associated with depression risk. The regression coefficient among older adults engaging in high-intensity work-related physical activity was β = 1.651 (p = 0.013), indicating substantially greater log odds of belonging to a more severe depression category. The odds ratio (OR = 5.21) showed that individuals performing vigorous-intensity occupational activity were 5.21 times more likely to experience more severe depression than those who did not. Conversely, moderate-intensity recreational activity revealed only a minimal and statistically non-significant association with depression severity (β = 0.041, p = 0.859; OR = 1.04); therefore, it was not a meaningful predictor in this study. Male gender (β = -0.796, p < 0.001; OR = 0.45) emerged as a protective factor, indicating significantly reduced odds of increased depressive severity. By contrast, low income (β = 0.561, p = 0.019; OR = 1.75) and primary education only (β = 1.075, p = 0.003; OR = 2.93) were significant risk factors. Self-rated health was not significantly related to the regression model (p = 0.352).
Table 5.
Ordinal logistic regression analysis of physical activity type/intensity on depression
| Regression coefficient (β) | Standard error | Wald | DF | P | OR | 95% CI | |||
|---|---|---|---|---|---|---|---|---|---|
| Threshold for dependent variable | No depressive | 0.957 | 0.778 | 1.512 | 1 | 0.219 | - | -0.568 | 2.481 |
| Symptoms mild depression | 2.407 | 0.783 | 9.448 | 1 | 0.002 | - | 0.872 | 3.942 | |
| Moderate depression | 3.555 | 0.805 | 19.520 | 1 | 0.001 | - | 1.978 | 5.132 | |
| Independent variable | Moderately severe depression | 4.679 | 0.871 | 28.873 | 1 | 0.001 | - | 2.972 | 6.385 |
| Severe depression | Reference | ||||||||
| Sex | Men | -0.796 | 0.183 | 18.960 | 1 | 0.001 | 0.451 | -1.154 | -0.438 |
| Women | Reference | - | - | - | - | - | - | - | |
| Self-rated health | Yes | -0.164 | 0.176 | 0.865 | 1 | 0.352 | 0.849 | -0.509 | 0.181 |
| No | Reference | - | - | - | - | - | - | - | |
| Income | Low | 0.561 | 0.240 | 5.476 | 1 | 0.019 | 1.752 | 0.091 | 1.030 |
| Mid-low | 0.216 | 0.244 | 0.784 | 1 | 0.376 | 1.241 | -0.262 | 0.693 | |
| Mid-high | 0.127 | 0.245 | 0.268 | 1 | 0.605 | 1.135 | -0.354 | 0.608 | |
| High | Reference | - | - | - | - | - | - | - | |
| Education level | Primary school | 1.075 | 0.365 | 8.658 | 1 | 0.003 | 2.93 | 0.359 | 1.791 |
| Middle school | 0.700 | 0.401 | 3.047 | 1 | 0.081 | 2.014 | -0.086 | 1.486 | |
| High school | 0.536 | 0.387 | 1.921 | 1 | 0.166 | 1.709 | -0.222 | 1.294 | |
| College school | Reference | - | - | - | - | - | - | - | |
| Vigorous-intensity physical activity (occupational) | Yes | 1.651 | 0.663 | 6.199 | 1 | 0.013 | 5.211 | 0.351 | 2.951 |
| No | Reference | - | - | - | - | - | - | - | |
| Moderate-intensity physical activity (recreational) | Yes | 0.041 | 0.231 | 0.032 | 1 | 0.859 | 1.042 | -0.412 | 0.494 |
| No | Reference | - | - | - | - | - | - | - | |
Association between physical activity volume levels and depression in older Korean adults
Table 6 examines the relationship between physical activity intensity and depression among older Korean adults. In vigorous-intensity occupational activity, the proportion of individuals with no depression was higher in the MET ≥ 500 group (85.33%) than in the MET < 500 group (77.78%); conversely, the proportion of individuals with severe depression was higher in the MET < 500 group (11.11%). The chi-square test (χ2 = 23.231, p = 0.000) revealed that the distribution of depression severity significantly differed between the two groups. However, this bivariate result did not account for potential confounding factors and should therefore be cautiously interpreted when compared with the multivariate regression analysis. No significant difference was found between groups in vigorous-intensity recreational activity (χ2 = 4.509, p = 0.341). MET ≥ 500 and MET < 500 groups showed similarly high proportions of individuals without depression (80.00% vs 85.42%), indicating no meaningful correlation. Moderate-intensity recreational activity did not have a significant association with depression (χ2 = 5.842, p = 0.211). Although the MET ≥ 500 group showed a slightly higher proportion without depression (85.70% vs 77.78%), the difference was not statistically significant. Interestingly, moderate-intensity recreational activity exhibited a significant association (χ2 = 10.057, p = 0.039). However, the pattern was counterintuitive: the MET < 500 group exhibited a higher proportion without depression (88.98%) than the MET ≥ 500 group (84.44%). The MET ≥ 500 group also showed a higher proportion of mild depression (11.33% vs 7.76%). This pattern suggested the possibility of unmeasured confounding variables or reverse causality such as individuals with depressive symptoms engaging in more recreational activity because of necessity rather than choice.
Table 6.
Association between physical activity volume levels and depression
| Variable | Category | Depression group (%) |
χ 2 | p | ||||
|---|---|---|---|---|---|---|---|---|
| No depressive symptoms | Mild depression | Moderate depression | Moderately severe depression | Severe depression | ||||
| MET vigorous-intensity physical activity (occupational) | MET < 500 | 7 (77.78) | 1 (11.11) | 0 (0.00) | 0 (0.00) | 1 (11.11) | 23.231 | 0.001** |
| MET ≥ 500 | 1,128 (85.33) | 141 (10.67) | 36 (2.72) | 12 (0.91) | 5 (0.38) | |||
| MET vigorous-intensity physical activity (recreational) | MET < 500 | 28 (80.00) | 7 (20.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) | 4.509 | 0.341 |
| MET ≥ 500 | 1,107 (85.42) | 135 (10.42) | 36 (2.78) | 12 (0.93) | 6 (0.46) | |||
| MET moderate-intensity physical activity (occupational) | MET < 500 | 56 (77.78) | 12 (16.67) | 3 (4.17) | 0 (0.00) | 1 (1.39) | 5.842 | 0.211 |
| MET ≥ 500 | 1,079 (85.70) | 130 (10.33) | 33 (2.62) | 12 (0.95) | 5 (0.40) | |||
| MET moderate-intensity physical activity (recreational) | MET < 500 | 218 (88.98) | 19 (7.76) | 5 (2.04) | 0 (0.00) | 3 (1.22) | 10.057 | 0.039* |
| MET ≥ 500 | 917 (84.44) | 123 (11.33) | 31 (2.85) | 12 (1.10) | 3 (0.28) | |||
p < 0.05
p < 0.01
Ordinal logistic regression analysis of physical activity volume levels on depression in older Korean adults
For vigorous-intensity occupational activity, the participants below the MET ≥ 500 threshold demonstrated a regression coefficient of β = 1.09 (p = 0.167) relative to those who met the threshold (Table 7). The corresponding odds ratio (OR = 2.974) suggested higher odds of belonging to more severe depression categories among those who did not meet the recommended MET level; however, this association was not statistically significant. The 95% CI (−0.454, 2.635) included zero, and the OR interval included 1, indicating that the observed association likely occurred by chance. For moderate-intensity recreational activity, the MET < 500 group showed a regression coefficient of β = 0.045 (p = 0.846) compared with those meeting the threshold, with an OR of 1.046. This finding indicated minimal differences in depression severity based on recreational activity levels. The results from the MET-based activity volume analysis differed somewhat from those of the activity type/intensity analysis. This discrepancy likely reflected the fact that activity volume captures the total quantity of energy expenditure, whereas activity type/intensity reflects the contextual characteristics of physical activity behavior.
Table 7.
Ordinal logistic regression analysis of physical activity volume levels on depression
| Regression coefficient (β) | Standard Error | Wald | Df | OR | p | 95% CI |
|||
|---|---|---|---|---|---|---|---|---|---|
| LL | UL | ||||||||
| Threshold for dependent variable | No depressive symptoms | 0.959 | 0.778 | 1.522 | 1 | - | 0.217 | -0.565 | 2.484 |
| Mild depression | 2.408 | 0.783 | 9.462 | 1 | - | 0.002 | 0.874 | 3.943 | |
| Moderate depression | 3.552 | 0.804 | 19.498 | 1 | - | < 0.001 | 1.976 | 5.129 | |
| Moderately severe depression | 4.671 | 0.87 | 28.809 | 1 | - | < 0.001 | 2.965 | 6.377 | |
| Severe depression | Reference | - | - | - | - | - | - | - | |
| MET vigorous-intensity occupational activity | MET < 500 | 1.09 | 0.788 | 1.914 | 1 | 2.974 | 0.167 | -0.454 | 2.635 |
| MET ≥ 500 | Reference | - | - | - | - | - | - | - | |
| MET moderate-intensity physical activity (recreational) | MET < 500 | 0.045 | 0.231 | 0.038 | 1 | 1.046 | 0.846 | -0.408 | 0.498 |
| MET ≥ 500 | Reference | - | - | - | - | - | - | - | |
DISCUSSION
This study examined the associations between different types and intensities of physical activity and depressive symptoms in older Korean adults by using national representative data from the KNHANES. The findings indicated that vigorous-intensity occupational physical activity was significantly associated with greater odds of more severe depressive symptoms after adjustment for sociodemographic covariates. Conversely, moderate-intensity recreational activity was not a significant predictor of depression severity in the multivariate analysis. Although vigorous-intensity occupational physical activity showed a statistically significant association with depression severity in the regression analysis, this finding should be interpreted cautiously. Only a very small proportion of participants reported engaging in this type of activity (approximately 0.75%). Consequently, the estimated effect size might be sensitive to sparse data bias and statistical instability. Therefore, this result should be considered exploratory and hypothesis generating rather than definitive evidence of a causal relationship.
The demographic differences identified in this study further contextualize the findings. Unmarried older women showed a higher prevalence of depressive symptoms, consistent with prior epidemiological studies identifying gender-related vulnerability and the absence of marital support as important risk factors for late-life depression [20,21]. Although the present study did not examine the interaction between marital status and depression specifically among older women due to the small proportion of unmarried participants in the sample, existing Korea-based research suggests that older women may face a greater caregiving burden, increased social isolation, and lower participation in leisure-time physical activity, all of which are associated with elevated depression risk [22]. With the observed gender differences in depressive symptoms in this study, future research should further investigate how marital status and related social factors uniquely influence depression among older women. Similarly, the socioeconomic gradient—where higher income and higher education attainment predicted lower depression levels—aligns with long-standing evidence suggesting that socioeconomic resources enhance psychological resilience by improving access to health information, healthcare services, and opportunities for active lifestyles [23,24].
Previous studies suggested that certain forms of occupational or outdoor physical activity, such as farming or manual labor, may be associated with lower depressive symptom scores compared with sedentary occupations [25,26]. However, the present findings indicate a different pattern: older adults engaging in vigorous-intensity occupational activity showed higher odds of belonging to more severe depression categories after adjusting for socioeconomic factors. One possible explanation is that physically demanding occupational activity among older adults may reflect socioeconomic vulnerability or necessity-driven labor rather than voluntary health-promoting exercise.
A key issue in interpreting the present findings concerns the potential for reverse causality. Because the study employed a cross-sectional design, the temporal relationship between physical activity and depressive symptoms cannot be determined. It is possible that older adults experiencing depressive symptoms may engage in physically demanding occupational activities due to economic necessity rather than voluntary participation. In such cases, occupational physical activity may reflect underlying socioeconomic vulnerability rather than serving as a protective factor against depression. Therefore, the observed association should be interpreted cautiously, and future longitudinal studies are needed to clarify the directionality of this relationship.
An important point to consider is the distinction between activity type/intensity and total activity volume. In the present study, analyses based on activity type and intensity examine the contextual characteristics of physical activity (e.g., occupational versus recreational activity and the intensity level of engagement), whereas MET-based analyses reflect the overall amount of physical activity accumulated over time. These two analytical approaches capture different dimensions of physical activity behavior and may therefore yield different statistical results.
Additionally, the chi-square analyses represent unadjusted associations, whereas the ordinal logistic regression models account for potential confounding variables such as gender, income, education level, and self-rated health. Adjustment for these covariates may attenuate previously observed associations, which may explain why certain findings observed in the chi-square analyses were not retained in the multivariate models. These findings highlight that different operationalizations of physical activity (contextual characteristics versus accumulated volume) may capture distinct behavioral and physiological mechanisms related to depression.
The public health implications of these findings are substantial. As Korea is one of the fastest-aging societies globally, promoting physical activity represents an accessible, low-cost, and scalable strategy for improving mental health among older adults [27,28]. Potential community-level interventions include gardening programs, neighborhood walking groups, structured exercise prescriptions, and programs aimed at social connectedness for low-income older adults. Enhancing opportunities for social participation is particularly relevant, as social engagement is a robust predictor of emotional well-being and health aging [29].
This study has several limitations that provide meaningful directions for future research. First, the findings primarily reflected the characteristics of older adults in South Korea, limiting cross-cultural generalizability. Additionally, the sample included only individuals aged 65 years and older. Future research should incorporate broader age ranges, including middle-aged adults, and diverse cultural backgrounds to enhance external validity. Second, no neurobiological indicators were examined to clarify the mechanisms underlying the relationship between physical activity and depression. Given the increasing evidence that links physical activity with biomarkers such as BDNF and inflammatory factors, future studies should integrate biological measures to further elucidate causal pathways. Third, the cross-sectional design of the study limited the ability to infer causal relationships between physical activity and depressive symptoms. Because exposure and outcome were measured at the same time point, reverse causality could not be excluded. For example, older adults experiencing depressive symptoms may be more likely to engage in physically demanding occupational activities because of socioeconomic necessity rather than personal choice. Longitudinal or prospective cohort studies should clarify the temporal and causal relationships between occupational physical activity and depression. Fourth, although several key sociodemographic variables were controlled in the regression models, residual confounding may still occur. The KNHANES dataset includes additional health-related variables such as chronic pain status and medication use (including antidepressants or other psychotropic agents) that were not incorporated into the present analysis. These factors may influence physical activity engagement and depression severity. Therefore, the observed associations should be cautiously interpreted. Future studies should include more detailed clinical and treatment-related variables to clarify the independent relationship between domain-specific physical activity and depressive symptoms in older adults. Fifth, the prevalence of vigorous-intensity occupational physical activity in the sample was extremely low (approximately 0.75%). Such small cell counts may introduce sparse data bias and result in inflated odds ratios or unstable regression estimates. Thus, the observed association between vigorous occupational activity and depression severity should be interpreted cautiously and considered exploratory. Future studies with larger samples of individuals engaging in high-intensity occupational activity should confirm this relationship.
In conclusion, this study indicates that vigorous-intensity occupational physical activity was associated with higher odds of belonging to more severe depression categories after adjustments for sociodemographic factors. The findings of the present study suggest that the relationship between physical activity and depression in later life might differ depending on the context in which an activity occurred. Therefore, public health strategies should emphasize context-sensitive physical activity interventions and address socioeconomic inequalities to promote mental well-being in aging societies.
Footnotes
ACKNOWLEDGMENT
The authors declare no conflicts of interest. This manuscript was developed from and revised based on the master's thesis submitted by Zhao Fan.
REFERENCES
- 1.World Health Organization (WHO) Global Report on Aging and Health. Geneva: WHO; 2022. [Google Scholar]
- 2.Statistics Korea (KOSTAT) Annual Report on Population and Aging. Daejeon: Statistics Korea; 2022. [Google Scholar]
- 3.Santini ZI, Jose PE, Cornwell EY, Koyanagi A, Nielsen L, Hinrichsen C, Meilstrup C, Madsen KR, Koushede V. Social disconnectedness, perceived isolation, and symptoms of depression and anxiety among older Americans (NSHAP): a longitudinal mediation analysis. Lancet Public Health. 2020;5:e62–e70. doi: 10.1016/S2468-2667(19)30230-0. [DOI] [PubMed] [Google Scholar]
- 4.Hull L, Schneider C, Taylor M. Depression among aging populations: Challenges and solutions. JMIR Ment Health. 2022;45:215–30. [Google Scholar]
- 5.American Psychological Association (APA) Depression: Definitions and Symptoms. Washington, DC: APA; 2024. [Google Scholar]
- 6.Krishnan V, Nestler EJ. The molecular neurobiology of depression. Nature. 2008;455:894–902. doi: 10.1038/nature07455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sen S, Duman R, Sanacora G. Serum brain-derived neurotrophic factor, depression, and antidepressant medications: meta-analyses and implications. Biol Psychiatry. 2008;64:527–32. doi: 10.1016/j.biopsych.2008.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Miller AH, Maletic V, Charles L Raison CL. Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression. Biol Psychiatry. 2009;65:732–41. doi: 10.1016/j.biopsych.2008.11.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Diniz BS, Butters MA, Albert SM, Dew MA, Reynolds CF., 3rd Late-life depression and risk of vascular dementia and Alzheimer’s disease: systematic review and meta-analysis of community-based cohort studies. Br J Psychiatry. 2013;202:329–35. doi: 10.1192/bjp.bp.112.118307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Carney RM, Freedland KE. Depression and coronary heart disease. Nat Rev Cardiol. 2017;14:145–55. doi: 10.1038/nrcardio.2016.181. [DOI] [PubMed] [Google Scholar]
- 11.Pearce M, Garcia L, Abbas A, Strain T, Schuch FB, Golubic R, Kelly P, Khan S, Utukuri M, Laird Y, Mok A, Smith A, Tainio M, Brage S, Woodcock J. Association Between Physical Activity and Risk of Depression: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2022;79:550–9. doi: 10.1001/jamapsychiatry.2022.0609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Schuch FB, Vancampfort D, Richards J, Rosenbaum S, Ward PB, Stubbs B. Exercise as a treatment for depression: A meta-analysis adjusting for publication bias. J Psychiatr Res. 2016;77:42–51. doi: 10.1016/j.jpsychires.2016.02.023. [DOI] [PubMed] [Google Scholar]
- 13.Vandelanotte C. A meta-meta-analysis of the effect of physical activity on depression and anxiety in non-clinical adult populations. Health Psychol Rev. 2015;9:366–78. doi: 10.1080/17437199.2015.1022901. [DOI] [PubMed] [Google Scholar]
- 14.Xu Z, Zheng X, Ding H, Zhang D, Cheung PM, Yang Z, Tam KW, Zhou W, Chan DC, Wang W, Wong SY. The Effect of Walking on Depressive and Anxiety Symptoms: Systematic Review and Meta-Analysis. JMIR Public Health Surveill. 2024;10:e48355. doi: 10.2196/48355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ekelund U, Tarp J, Steene-Johannessen J, Hansen BH, Jefferis B, Fagerland MW, Whincup P, Diaz KM, Hooker SP, Chernofsky A, Larson MG, Spartano N, Vasan RS, Dohrn IM, Hagströmer M, Edwardson C, Yates T, Shiroma E, Anderssen SA, Lee IM. Dose-response associations between accelerometer measured physical activity and sedentary time and all-cause mortality. BMJ. 2019;366:l4570. doi: 10.1136/bmj.l4570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kim J, Han M, Lee Y. Trends in adherence to physical activity guidelines among elderly adults in South Korea (2011-2020) J Aging Phys Activ. 2020;32:461–73. doi: 10.1123/japa.2023-0341. [DOI] [PubMed] [Google Scholar]
- 17.Han C, Jo SA, Kwak JH, Pae CU, Steffens D, Jo I, Park MH. Validation of the Patient Health Questionnaire-9 Korean version in the elderly population: the Ansan Geriatric study. Compr Psychiatry. 2008;49:218–23. doi: 10.1016/j.comppsych.2007.08.006. [DOI] [PubMed] [Google Scholar]
- 18.Ainsworth BE, Haskell WL, Leon AS, Jacobs DR, Montoye HJ, Sallis JF, Paffenbarger Jr RS. Compendium of physical activities: Classification of energy costs of human physical activities. Med Sci Sports Exerc. 2011;43:1575–81. doi: 10.1249/00005768-199301000-00011. [DOI] [PubMed] [Google Scholar]
- 19.Segura-Jiménez V, Munguía-Izquierdo D, Camiletti-Moirón D, Alvarez-Gallardo IC, Ortega FB, Ruiz JR, Delgado-Fernández M. Comparison of the International Physical Activity Questionnaire (IPAQ) with a multi-sensor armband accelerometer in women with fibromyalgia: the al-Ándalus project. Clin Exp Rheumatol. 2013;31:S94–101. [PubMed] [Google Scholar]
- 20.Albert PR. Why is depression more prevalent in women? J Psychiatry Neurosci. 2015;40:219–21. doi: 10.1503/jpn.150205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Holt-Lunstad J, Smith T, Layton B. Social isolation and mortality: A meta-analysis. PLoS Med. 2010;7:e1000316. doi: 10.1371/journal.pmed.1000316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lim YM, Baek J, Lee S, Kim JS. Association between loneliness and depression among community-dwelling older women living alone in South Korea: the mediating effects of subjective physical health, resilience, and social support. Int J Environ Res Public Health. 2022;19:9246. doi: 10.3390/ijerph19159246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lorant V, Deliège D, Eaton W, Robert A, Philippot P, Ansseau M. Socioeconomic inequalities in depression: a meta-analysis. Am J Epidemiol. 2003;157:98–112. doi: 10.1093/aje/kwf182. [DOI] [PubMed] [Google Scholar]
- 24.Jespersen A, Madden RA, Whalley HC, Reynolds RM, Lawrie SM, McIntosh AM, Iveson MH. Socioeconomic status and depression-a systematic review. Epidemiol Rev. 2025;47 doi: 10.1093/epirev/mxaf011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.McCurdy LE, Winterbottom KE, Mehta SS, Roberts JR. Using nature and outdoor activity to improve children’s health. Curr Probl Pediatr Adolesc Health Care. 2010;40:102–17. doi: 10.1016/j.cppeds.2010.02.003. [DOI] [PubMed] [Google Scholar]
- 26.Hagen BNM, Albright A, Sargeant J, Winder CB, Harper SL, O’Sullivan TL, Jones-Bitton A. Research trends in farmers’ mental health: A scoping review of mental health outcomes and interventions among farming populations worldwide. PLoS One. 2019;14:e0225661. doi: 10.1371/journal.pone.0225661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Warburton DER, Bredin SSD. Health benefits of physical activity: a systematic review of current systematic reviews. Curr Opin Cardiol. 2017;32:541–56. doi: 10.1097/HCO.0000000000000437. [DOI] [PubMed] [Google Scholar]
- 28.OECD . Healthy Aging Policy Report. Paris: OECD Publishing; 2023. [Google Scholar]
- 29.Huxhold O, Miche M, Schüz B. Benefits of having friends in older ages: differential effects of informal social activities on well-being in middle-aged and older adults. J Gerontol B Psychol Sci Soc Sci. 2014;69:366–75. doi: 10.1093/geronb/gbt029. [DOI] [PubMed] [Google Scholar]