The Association Between Health-related Quality of Life and Physical Activity in Postmenopausal Korean Women: Findings From the Korea National Health and Nutrition Examination Survey

Background: Physical activity (PA) in postmenopausal women provides some relief of menopausal symptoms and improves health-related quality of life (HRQoL). This study used a cross-sectional database to investigate the association between PA level and HRQoL in postmenopausal Korean women. Methods: This study investigated 8,122 postmenopausal women who participated in the Korea National Health and Nutrition Examination Survey VI to VII (2014–2018). Global Physical Activity Questionnaire (GPAQ) and Euro quality of life-5 dimensions (EQ-5D) questionnaire were used to collect data about the participants’ PA level and HRQoL. The adjusted Odds ratios (aORs) with 95% condence intervals (95% CIs) for each dimensional problem of EQ-5D were derived using multi-variable logistic regression. Mean EQ-5D index scores were compared according to PA level groups. Results: The aORs of each dimension of HRQoL of low PA group versus moderate PA group were 0.67 (95% CI 0.57–0.78) for mobility, 0.71 (95% CI 0.55–0.92) for self-care, 0.64 (95% CI 0.53–0.78) for daily activities, and 0.75 (95% CI 0.66–0.86) for pain/discomfort. Mean EQ–5D scores of the moderate PA (0.927 ± 0.120 vs. 0.884 ± 0.163, P < 0.001) and high PA (0.913 ± 0.144 vs. 0.884 ± 0.163, P = 0.023) groups were signicantly higher than that of the low PA group. Conclusions: PA level was found to be associated with better HRQoL among postmenopausal Korean women. This was especially signicant in participants who were participating with a moderate level of PA. Additional studies are needed to analyze the persistence and long-term effects of PA on HRQoL.


Background
Starting from the mid-forties, most women reach perimenopause and postmenopause as the female sex hormone decreases, causing physical or psychological discomfort [1]. Consequently, various symptoms including anxiety, depression, headache, fatigue, joint pain, and weight gain start appearing [2]. Vasomotor symptoms, such as hot ushes, are one of the most common complaints occurring in about 75% of women during their transition [3][4][5] to menopause. These vasomotor symptoms, which are closely linked to tension, depression, sleeplessness, and general helplessness, have been notably observed to affect the psychological status and result in a decrease in the quality of life [6][7][8][9] .
Hormone replacement treatment (HRT) is currently the most effective treatment for vasomotor and genitourinary syndromes in postmenopausal women. However, a limitation of HRT is that it needs to consider individual differences, treatment goals, age, among other things, in order to maximize effectiveness and safety [10]. Since previous research has suggested that PA could help prevent postmenopausal symptoms [11], exercise has become increasingly utilized for the prevention and treatment of the symptoms. It is now becoming more widely accepted as a signi cant factor in improving the quality of life of these women. Earlier ndings have also supported the usefulness of exercise in improving physical ability and mental stability [12]. Furthermore, previous studies have demonstrated that those who performed regular PA had better HRQoL than those who did not [13].

Page 3/17
A few studies have reported the positive effects of PA on the quality of life in postmenopausal women.
For example, the recommended levels of PA signi cantly improved upon symptom relief and quality of life in postmenopausal women regardless of hormone replacement therapy [14][15][16]. Previous studies had simply analyzed these effects in terms of the presence or absence of simple PA and its effect at a certain level. Furthermore, only few studies have compared the quality of life by classifying PA according to intensity levels. Therefore, this study aimed to examine differences in the quality of life in relation to the level of PA in postmenopausal women included in the Korea National Health and Nutrition Examination Survey.

Study population and data collection
The study was conducted using data extracted from the sixth and seventh Korea National Health and Nutrition Examination Survey (KNHANES VI-VII), which is a nationwide project that investigated Korea's civilian population. KNHANES VI-VII was conducted by the Ministry of Health and Welfare of Korea and the Korea Disease Control and Prevention Agency (KDCA) from 2014 to 2018 and included surveys of health and nutrition. The selection of households and participants for KNHANES VI-VII used a complex, multi-stage, probability-sampling design, allowing the production of nationally representative data. The KNHANES was approved by the institutional review board of KDCA (registration number: 2013-12EXP-03-5C). Written informed consent forms were obtained from the participants. Among 8,122 postmenopausal women who participated in the KNHANES VI-VII, 17 participants who did not have data on HRQoL and subjective health status and 58 participants who had no data regarding their PA were excluded. The selection process for the study population is shown in Fig. 1.

Research method Physical activity (PA) level
The PA level of the participants has been classi ed by the Global Physical Activity Questionnaire (GPAQ) criteria developed by the World Health Organization [17]. In these criteria, a high intensity activity was de ned as vigorous PA that caused tachypnea and tachycardia. Moderate intensity activity was de ned as an average level of PA that caused a mild increase in breathing and heart rate. The PA was based on a self-report questionnaire that contained the number of days the participants performed physical activities in a week and the corresponding minutes of PA per day. The questionnaire results were converted into the metabolic equivalent of task minutes per week (METs). High intensity activity, moderate intensity activity, and walking were converted to 8.0, 4.0, and 4.0 METs. The participants were then assigned to either the low PA group (less than 600 MET min − 1 ), moderate PA group (601-2999 MET min − 1 ), or high PA group (higher than 3000 MET min − 1 ) according to the corresponding levels per week [17].

HRQoL and perceived health status
The KNHANES uses the Korean version of the European quality of life scale-5 dimensions (EQ-5D) questionnaire to assess HRQoL. The EQ-5D consists of ve dimensions of the current health status, including mobility, self-care, daily activities, pain/discomfort, and anxiety/depression. The degree of each dimensional status was rated as "no problem exists," "slight problems exist," or "extreme problems exist." This study used the Korean version of the weighted model of the EQ-5D for its analysis. The reliability and validity of the Korean version of the EQ-5D were evaluated and veri ed by a previous study on the general population, and the reliability level was observed to be moderate (Cohen kappa 0.32-0.64). In addition to the questionnaire with ve dimensions, the EQ-5D index score was also used to produce a single index value that indicated overall health status. In this study, the researchers used the EQ-5D index score, which was calculated and validated by previous studies using the corresponding estimated weighted quality value for Koreans [18]. The perceived health status was inferred based on the following question: "How would you describe your overall health status?" ("very good," "good," "fair," "poor," and "very poor"). The responses "very good," "good," and "fair" were considered to indicate "not poor," while "poor" and "very poor" were considered to indicate "poor." Covariates: sociodemographic information, health status, health behavior Sociodemographic variables (age, marital status, level of education, place of residence, and household income) were collected using a self-reported questionnaire. Age group (below 50, 50 ~ 59, 60 ~ 69, and over 70 years of age), marital status, education (elementary school or lower, middle school, high school, technical college, and college or higher), household income ( rst, second, third, or fourth quartile), and health behaviors (smoking and drinking) were used for the analysis. Regarding the health status, the depressive symptoms and associated diseases (hypertension, diabetes, dyslipidemia, thyroid and coronary artery disease, stroke, and cancer) that were collected using a self-report questionnaire were also used.

Statistics analysis
All participants were divided into three groups (low, moderate, and high PA levels). To compare the basic demographics, anthropometric characteristics, lifestyle factors, and the ve dimensions of quality of life using the EQ-5D of the three groups, continuous variables were presented as estimated mean ± standard errors and subsequently compared using Student's t-test, while the categorical variables were presented as unweighted numbers and estimated percentages and subsequently analyzed using the chi-squared (χ 2 ) test.
Univariate and multivariate logistic regression analyses were performed to evaluate the association between PA and each dimensional status of quality of life. We modi ed the original three categories into two categories by combining the categories "some problems" and "extreme problems," since the number of responses with "extreme problems" for each dimension was insu cient to perform acceptable analyses.
One-way analysis of variance was performed to compare differences in EQ-5D index scores for each group with low, moderate, and high PA plus additional classi cations regarding age and body mass index.
We performed the estimation command svy in the statistical/data analysis program to take the sampling weight into account so that the results could represent the entire Korean population. P < 0.05 was considered to indicate statistical signi cance for all outcomes. All statistical analyses were performed using STATA version 14.0 (STATA Corp., College Station, TX, USA).

Results
Baseline characteristics of the study participants Table 1 depicts the descriptive characteristics of the study's participants strati ed according to the PA levels. Among 8122 participants, 5762 women (70.94%) were assigned to the low PA group, 2148 women (26.45%) to the moderate PA group, and 212 people (2.61%) to the high PA group. The majority had normal body mass index (BMI), were married, lived in cities, were in the middle-third in terms of household income, did not smoke nor drink, and did not report depressive symptoms (Table 1). Among the sociodemographic pro les, health status, and health behaviors in different PA level groups, all factors were found to show signi cant differences, except for smoking and depression. Comparison of HRQoL and perceived health status according to PA levels Table 2 shows the distribution of the HRQoL, EQ-5D index scores, and perceived health status based on the level of PA. Signi cant differences were observed between the moderate and high PA groups and the low PA group in all ve dimensions of EQ-5D: exercise ability, self-management, daily activities, pain/discomfort, and anxiety/depression. Furthermore, the mean EQ-5D scores of the moderate PA (0.927 ± 0.120 vs 0.884 ± 0.163, P < 0.001) and high PA (0.913 ± 0.144 vs 0.884 ± 0.163, P = 0.023) group were signi cantly higher than those of the low PA group (Table 2). Similarly, the perceived health conditions was associated with a signi cant difference in the proportion of negative answers between the low, moderate, and high PA groups. Relationship between PA level and health-related quality of life (HRQoL) Table 3 presents the odds ratios (ORs) and 95% con dence intervals (CIs) for poor HRQoL and perceived health status on all 5 EQ-5D dimensions (mobility, self-care, daily activities, pain/discomfort, and anxiety/depression) based on the PA as the independent variable. From the crude logistic analyses, a signi cant tendency for the cross-ratio regarding low quality of life to be reduced in the moderate and high PA group was con rmed in all ve dimensions of EQ-5D as compared to that in the low PA group. Similarly, in the ORs for poorly-perceived health status, the moderate and high PA groups showed a reduced cross-ratio in comparison with the low PA group.  Figure 3 shows the EQ-5D index values according to the level of PA in three groups: low, normal, and high body mass index (BMI). Compared to the low PA group, the moderate PA group had signi cantly higher EQ-5D index in the normal (0.899 ± 0.151 vs 0.936 ± 0.110, P < 0.001) and high BMI groups (0.863 ± 0.172 vs 0.908 ± 0.132, P < 0.001).

Discussion
The current ndings suggest signi cant differences in the quality of life between the moderate and high PA groups and the low PA group in postmenopausal women; the EQ-5D index score also showed a signi cant difference. In particular, compared to the low PA group, the moderate PA group showed a signi cantly higher quality of life index in the age groups above 60 with normal or high BMI. Although a number of previous studies have reported the positive aspects of PA in postmenopausal women, no study has compared the quality of life according to the level of PA in large groups to the best of our knowledge. Therefore, this study was the rst attempt in analyzing the HRQoL and perceived health status in different PA level groups involving 8,122 postmenopausal women who participated in the sixth and seventh edition of the National Health and Nutrition Survey (2014-2018).
Postmenopausal women may experience reduced quality of life due to physical, psychological, and social issues. In particular, PA subsequently decreases due to weight gain and decadence of the musculoskeletal system. However, it must be noted that PA level, along with the quality of life, can still be increased through regular and well-controlled exercise [19]. The current study also found signi cant differences between the moderate and high PA groups and the low PA group in all ve dimensions of the quality of life, namely exercise ability, self-management, daily activities, pain/discomfort, and anxiety/depression on top of the EQ-5D index score. In particular, when compared to the low PA group, the moderate PA group showed signi cantly higher quality of life index in the age groups above 60 with normal and high BMI. This is consistent with the meta-analysis reported by Pucci GC et al. [20] who revealed a higher quality of life in the high PA level elderly groups when compared to low PA level elderly groups. Similarly, a positive correlation between elevated PA and high quality of life in groups with underlying diseases such as diabetes and cancer [21,22] has been previously reported. Studies investigating the relationship between menopause and HRQoL in middle-aged women revealed that postmenopausal women had lower EQ-5D index values than pre-menopausal women, which suggests that the deterioration of physical function happening mostly within ve years of menopause is a major factor for this difference [23]. However, according to a randomized control study by Moriyama CK et al. [15], a signi cant difference existed in the changes in physical function in the group with moderate PA levels regardless of undergoing hormone replacement therapy. Furthermore, PA was observed to improve the quality of life along with the alleviation of menopausal symptoms.
The effects of exercise and PA in postmenopausal women on their health outcomes have been reported in several studies; e.g., exercise can lower the risk of cardiovascular disease, prevent osteoporosis, and lower BMI and blood pressure [24][25][26][27]. Exercise can also alleviate hot ushes and menopausal symptoms since PA has been attributed to increased blood beta-endorphins, which is known to decrease after menopause [28], resulting in the relief of vasomotor symptoms [29]. The current ndings are also consistent with previous results. Furthermore, the current study found that moderate PA showed a more signi cant difference than high PA when compared to low PA. In previous studies investigating the level of PA and the quality of life, the group that performed moderate or high intensity (23METs/hour/week) PA levels showed signi cantly higher quality of life than those who did not [16]. Morimoto T, et al. notably reported that a higher PA level would lead to positive effects on the quality of life in both men and women and, in particular, the highest intensity of PA showed a signi cant difference in various dimensions of the quality of life in women [14].
The change in PA itself was not observed to have a direct relation to the alleviation of vasomotor or psychological symptoms [30]. However, the weight loss that resulted from increased PA was correlated with the alleviation of overall symptoms of menopause. Thus, it is believed that a moderate level of PA may have led to weight loss and an increased awareness of health levels, which could have in uenced the results. In particular, in the comparative analysis of EQ-5D according to BMI, only the moderate PA group had a signi cant difference compared to low PA group in the normal and high BMI groups. This was not found in the low BMI group, which suggests that factors such as varying weight loss due to difference in PA levels may be related to the observed result. This study did not include analysis of the weight loss among the three comparison groups. Thus, further studies involving additional comparative analysis on this topic would be useful.
It has been previously reported that a clear association was found between menopausal symptoms, including vasomotor symptoms, and the resulting low quality of life [8]. Therefore, it can be inferred that the alleviation of menopausal symptoms with the increase in beta-endorphin levels through PA or weight loss improves the quality of life. Additionally, the psychological and social effects of PA, which improves the quality of life, should also be considered. The EQ-5D evaluation dimensions also re ect certain psychological and social factors, such as self-management and anxiety/depression. The corresponding factors resulting from the PA level, including social connectedness and reinforcement of self-esteem must also be considered. On the other hand, in the comparison by age groups, no signi cant difference according to the PA level in the age groups under 60 years of age was found. The researchers believe that the PA level in those groups produced a different effect on the quality of life because, unlike the age group over 60 years of age, the postmenopausal symptoms and durations differed depending on the duration after menopause.
Our study had several limitations. First, the cross-sectional study design prevented the explanation of the clear causal relationship between the level of PA and quality of life. Second, since the quanti cation of PA level was based on a self-report survey, it may be more or less inaccurate. In particular, recall bias could not be completely ruled out. Lastly, re ecting on the persistence of PA was di cult due to the analysis of the PA level being only conducted at the time of the survey. Analyzing the persistence and long-term effects of PA through subsequent studies is highly recommended. Despite these limitations, our study was more extensive than many previous studies, as this study used a nationally representative study population. Furthermore, a greater number of sociodemographic factors, including residence, household income, and marital status, were taken into account as opposed to previous studies. Finally, this was the rst attempt to consider PA levels in comparing the quality of life in postmenopausal women.

Conclusion
This study con rmed that differences exist in HRQoL and perceived health status among the low, moderate, and high PA groups in postmenopausal Korean women. In particular, moderate PA was more e cient in improving the quality of life as compared to high PA. This suggests that PA may help women who suffer from postmenopausal symptoms and low quality of life by alleviating their symptoms and improving their quality of life. This approach is different from conventional hormone replacement therapy. In particular, the researchers believe that this study will become a useful guideline in improving the actual quality of life in postmenopausal women, especially in that it shared the concept as well as laid out concrete action plans involving moderate PA for the future.