E ﬀ ects of a Yoga Program in Reducing Cardiovascular Disease Risk Factors in Workers of Small Workplaces: A Pilot Test

: The purpose of this study was to evaluate the e ﬀ ects of a yoga program provided to workers in small businesses to reduce cardiovascular disease (CVD) risk factors. A nonequivalent control group pre / post-test design was used. The experimental group (n = 31) was assigned a yoga program consisting of yoga postures designed for meditation, strengthening, stretching, and balancing, given twice weekly for 12 weeks. The control group (n = 38) was given no other intervention. The mean age was 48.1 years old in the experimental group and 47.7 years old in the control group. Three trained investigators collected the questionnaires and one month after completing the 12-week yoga program, the same questionnaires that were administered at baseline were again administered. Psychosocial variables (depressive symptoms and job stress), health promotion behaviors, and body measurements (weight and waist circumference) were measured for the program evaluation. The yoga program was e ﬀ ective in improving waist circumference (from 81.8 to 79.2, p < 0.001) and diastolic blood pressure (from 81.0 to 79.1, p = 0.004) as compared to the control group. Furthermore, the experimental group’s job stress score decreased from 1.38 to 1.02, but it was not statistically signiﬁcant ( p = 0.240). A yoga program could be a useful intervention for workers with CVD risk factors, but it was not e ﬀ ective in psychological factors such as job stress and depression. A long-term intervention approach is important to conﬁrm the psychosocial e ﬀ ect. Therefore, future research is needed to investigate the long-term outcomes of such interventions.


Introduction
As of 2018, the number of small enterprises in Korea reached 2,649,967, accounting for 99.8% of all enterprises, and the number of workers in small enterprises accounted for 83.0% (15,818,937) of the entire worker pool [1]. Furthermore, the industrial accident rate in small enterprises is 0.60%, which is higher than the rate of 0.24% in medium-sized or larger enterprises with more than 300 workers [1]. The most prevalent problem among workers is conditions related to physically burdening work (32.3%), followed by lower back pain (31.9%), pneumoconiosis (9.7%), and cardiovascular disease (CVD; 6.8%). Statistics data showed that CVD (39%) was the second cause of mortality for work-related accident death in 2013, but it became the first cause among them in 2018 [1]. According to many previous studies [2][3][4], workers in small workplaces are at high risk of developing cardiovascular disease due to various occupational environmental factors. Firstly, workers in small workplaces suffer from physical fatigue caused by increased physical labor due to incessant physical work and lack of human resources. Moreover, lack of human resources leads to overtime work. For this reason, workers have factors in workers of small enterprises. The purpose of this study was to investigate the effects of a yoga program to reduce CVD risk factors including psychological factors such as job stress and depression provided to workers in small enterprises.
The study examined the effects of a yoga program provided to workers in small businesses to reduce cardiovascular disease (CVD) risk factors, including psychological health behavior factors, and physical measurements.

Study Population
This was a quasi-experimental study with control groups different in number of participants and pre/post-test design that aimed to examine the efficacy of a yoga program in reducing CVD risk factors in workers in small enterprises who have CVD risks. The CVD risk groups were determined based on the CVD risk assessment criteria established by the Korea Occupational Safety and Health Agency (KOSHA) based on the 2003 WHO International Society of Hypertension guidelines (WHO-ISH) ( 1 ≥SBP 140 mmHg or ≥DBP 90 mmHg, 2 age (≥55 years in men, ≥65 years in women), 3 current smoking, 4 high cholesterol ≥240 mg/dL, 5 early occurrence of CVDs in family members in a direct line (<50 years), 6 obesity (≥body mass index 30 kg/m 2 ), physical inactivity, 7 atrial fibrillation) [32]. The participants in the experimental group have had one or more of the above risk factors. Sixty-nine workers in small enterprises based in S and Y city who provided informed consent were enrolled. The experimental group comprised 31 workers of a small audio equipment components manufacturing enterprise located in S city, and the control group comprised 38 workers of three small enterprises which manufactures copper fittings located in Y city, all of which are geographically distant from the experimental group (Control group 1: n = 11, Control group 2: n = 15, Control group 3: n = 12).
Sample size was determined using the G*Power 3.1 program. Based on a test of the means of the two groups, 52 people (26 for each group) were estimated to be required to attain significance α = 0.05, with effect size d = 0.08 and power (1 − β) = 0.80. The effect size was estimated based on the study by Oh and Lee [29], which examined the effects of a yoga program on workers' stress.
Thirty-three participants were selected for the experimental group, and 41 participants who provided informed consents were selected for the control group, in consideration of possible dropouts. After excluding two participants from the experimental group for having a program participation rate of below 80% and three participants from the control group for leaving too many omissions in the questionnaire, a total of 31 and 38 participants from the experimental and control groups, respectively, were included in the final analysis. This study was conducted for six months from March to September 2015.

Development of a Yoga Program
The yoga program was developed based on a systematic literature review performed by one nursing professor and two researchers about interventions that lower CVD in workers, with cooperation from a professional yoga instructor. Figure 1 shows a flow chart of the yoga program intervention procedure. It provides the contents of the yoga program which was structured in three stages: warm-up, yoga poses, and cool-down through breathing and meditation. The first stage, the warm-up, comprised yoga breathing (one minute), meditation (two minutes), and simple stretching exercises (two minutes) for a total of five minutes. The second stage, the yoga pose stage, was designed to proceed for 20 min. During this stage, 11 yoga poses (sun salutation, triangle, cobra, grasshopper, alligator, chair, fish, kneeling, cat, spinal twist, extended puppy) were performed to relax the mind, strengthen muscles, and increase flexibility and balance. The third stage, the cool-down stage, comprised two minutes of yoga breathing and three minutes of meditation to focus the mind on oneself, to promote relaxation and reduce stress ( Figure 1).

Baseline Measurement in Experimental and Control Groups
Prior to beginning the study, researchers selected experimental and control groups. Then, the investigator individually met with the owner or health manager of each small enterprise to obtain informed consent and request cooperation. Three trained investigators visited each enterprise in March 2015 to administer the questionnaires. The investigators explained to the workers about study aims, confidentiality, the use of collected data strictly for study purposes, and the contents of the yoga program, and obtained signed consent forms.
Structured questionnaires were administered only to those who provided informed consent to investigate workers' depressive symptoms, job stress, and health-promoting behaviors. It took about 15-20 min for each participant to complete the questionnaires. Height and weight were measured using an electronic measuring scale, and the body mass index (BMI) was computed. Waist circumference was measured around the ilium using a measuring tape. Blood pressure was measured using a manual sphygmomanometer after a 15-min resting period.

Experimental Treatment: Yoga Program
The yoga program, intended to lower job stress and CVD risks, was administered by a professional yoga instructor twice a week for 12 weeks (24 sessions) from 28 April 2015 to 26 August 2015. Each session was performed for 30 min, from 7:50 a.m. to 8:20 a.m., before work start. Yoga took place in a large auditorium, and no dropouts occurred during the 12-week yoga program. In contrast to the yoga group, the participants in the control groups were simply instructed to maintain their normal life without participating in regular physical activity programs.

Post-Intervention Measurement in Experimental and Control Groups
On 21 September 2015 (one month after completing the 12-week yoga program), the same questionnaires that were administered and the weight, waist circumference, and blood pressure that were measured at baseline were again administered and measured for both the experimental and control groups.

Ethical Considerations
This study was approved by the Institutional Review Board of K University Hospital (IRB No. KMC IRB 1443-06). Prior to recruiting the participants, the investigators visited the enterprises of interest to obtain informed consent and request cooperation from the owner and health managers. Before beginning the program, signed informed consent was obtained from all participants. The consent form included a description and contact information for the investigator, aims and method of the study, and statements noting that personal information would not be exposed and that the participants could withdraw at will at any point in the study. Anonymity of the collected data was ensured by using individualized allocation numbers (ANs), and all collected data were stored in a locked safe. A small gift was given to both the experimental and control groups for participating in the study after the study was completed, and a leaflet describing methods to reduce job stress and prevent and manage CVD was also provided to the control group at no cost after the study was concluded.

Depressive Symptoms
Depressive symptoms were measured with the Korean version of the Center for Epidemiologic Studies Depression (CES-D) Scale, which was originally developed by Radloff [33] and translated and validated by Cho and Kim [34]. A higher score reflects more severe depressive symptoms. The reliability (Cronbach's α) of the Korean version of the CES-D was high (0.90) at the time of the development, but the Cronbach's α of the scale in this study was 0.67 at baseline, and 0.72 at post-intervention.

Job Stress
Job stress was measured using the Korean version of Effort-Reward Imbalance (ERI), which was originally developed by Siegrist et al. [35] and adapted by Eum et al. [36]. The questionnaire is based on the effort-reward imbalance model and consists of 17 items (6 items on effort and 11 items on reward). Each item is rated on a five-point scale from 1 to 5. Job stress is assessed using the ER ratio, obtained by dividing the total effort score by the total reward score. Specifically, the ER ratio is calculated by first adjusting the total reward score with a correction factor, 6/11 (=0.545) to transform the total scores of effort and reward to 1 (as there are 6 and 11 items in each of the subscales, respectively), and by computing the ratio with the transformed reward score as the denominator and the effort score as the numerator. An ER ratio less than 1 indicated low-level job stress, and an ER ratio greater than 1 indicated the presence of high-level job stress. The reliability (Cronbach's α) of the scale in previous studies [35,37] was 0.68 and 0.71, respectively, and that in this study was 0.90 at baseline point and 0.76 at post-intervention.

Health-Promoting Behaviors
Health-promoting behaviors were measured using the Health Promoting Lifestyle Profile II (HPLP II) developed by Walker, Shechrist, and Pender [38] and adapted by Hwang [39]. Each item is rated on a four-point scale: 1 for "not at all", 2 for "sometimes", 3 for "frequently", and 4 for "regularly." The HPLP II consists of 52 items, and the higher the score, the higher the level of health-promoting behavior. The reliability (Cronbach's α) of the overall scale at the time of the development was 0.94. The Cronbach's α values in the present study at baseline (post-intervention) was 0.93 (0.88).

Data Analysis
The data were analyzed as follows, using SPSS/WIN 22.0 software. First, the participants' general and disease-related characteristics were analyzed, with results reported as mean and standard deviation or number and percent.
Second, the homogeneity of general characteristics between the experimental and control groups was analyzed with Chi-square tests and F-tests.
Third, the effects of the yoga program intended to reduce the CVD risk factors in workers of small enterprises were assessed using two-way repeated measure ANOVA.

Homogeneity Testing
A total of 69 participants were enrolled, with 31 in the experimental group and 38 in the control group. There were no significant differences between the two groups in gender (p = 0.354), educational level (p = 0.235), marital status (p = 0.437), monthly income (p = 0.601), daily working hours (p = 0.707), or age (p = 0.845), confirming homogeneity between experimental group and control group (Table 1).

Psychosocial CVD Risk Factor
In the control group, job stress was 1.09 ± 0.32 before applying the yoga program, and there was no change to 1.09 ± 0.30 after applying the program. In the experimental group, it was confirmed that it was 1.38 ± 1.80 before applying the yoga program and decreased to 1.02 ± 0.24 after applying the program. However, there was no statistically significant difference in the reduction of job stress between groups at each measurement point in the effect verification (p = 0.240). The depressive symptom score did not change significantly before and after the yoga program between groups, but in the effectiveness verification, there was a statistically significant difference in job stress reduction between groups at each measurement point (p = 0.008). In the case of health promotion behavior, it was confirmed that scores increased after applying the yoga program compared to before applying the yoga program in both the experimental group (109.13 ± 25.94 to 111.35 ± 15.56) and the control group (97.26 ± 22.85 to 98.82 ± 20.41). The difference score of the experimental group (−2.23 ± 3.08) was greater than that in the control group (−1.55 ± 1.22), and it was statistically significant between groups. However, in the effect verification between groups at each measurement point, there was no statistically significant difference in the decrease in health promotion behavior scores (p = 0.828) ( Table 2).

Physical CVD Risk Factor
After the yoga intervention, the experimental group showed a significant reduction of waist circumference from 81.87 ± 8.01 cm to 79.21 ± 7.92 cm (p < 0.001), but the score of the control group was not changed. Also, there was statistically significant difference in the decrease of waist circumference scores in the effect verification between groups at each measurement point (p < 0.001). Furthermore, diastolic blood pressure decreased after applying the yoga program in the experimental group (81.04 to 79.16), but there was no change in the control group. Also, there was a statistically significant difference in the decrease of diastolic blood pressure at each measurement point between groups (p = 0.004). On the contrary, there were no significant changes in waist circumference or diastolic blood pressure in the control group. In addition, in the case of body weight or systolic blood pressure, there was no significant change in the experimental group and the control group when comparing before and after applying the yoga program. As a result, the yoga program was found to be effective in reducing waist circumference and diastolic blood pressure (p = 0.004) ( Table 2).

Discussion
This study applied a 12-week (24 sessions) yoga program (warm-up, yoga poses, cool-down) to workers in small enterprises-a pool of workers who are particularly vulnerable in terms of organized health management due to a lack of on-site health managers in small enterprises-in an attempt to lower CVD risk factors including psychosocial factors such as job stress, depression, and health promotion behaviors.
The experimental group who underwent the yoga intervention showed a decrease in job stress score after the intervention, even though it was not statistically significant. This is a similar result to that of a previous study showing a reduction in job stress in the experimental group after providing a yoga program to office workers [28]. In the previous study, the yoga program was designed as a 10-week, 20-session program with two 60-min sessions per week [28]. That program was similar in content to our program, which comprised warm-up, yoga posture, breathing, and meditation, and both programs led to reductions in job stress, though the duration of the sessions differed. A previous study found that yoga programs based on meditation and breathing have positive effects on social and emotional changes, namely providing psychological stability and reducing job stress, by activating the parasympathetic nerves and relaxing tense muscles [26]. Even though the score of job stress was not significantly decreased in this study, we could speculate that our yoga program could possibly lead to positive outcomes in terms of job stress in a future study. Also, we thought that a yoga program tailored to the working patterns of the workers, set to begin before work in the morning, could help relax the workers' minds and invigorate them, increasing their happiness and subsequently reducing job stress.
Depression in workers has been identified as a predictor of "presenteeism", which refers to work and productivity loss, increase of health management expenses, and indirect productivity loss induced by working with a health problem [40,41]. The experimental group that underwent the yoga intervention did not show significant changes in depressive symptom scores after the intervention. In other words, the yoga intervention was found to have no impact on the depression score in this study, which is largely similar to the findings of a previous report stating that a 16-week yoga program did not significantly decrease depression levels-though depression levels dropped at certain time points-in medical students [42]. These findings call for additional studies with longer follow-ups, because it is difficult to establish the effects of interventions on social and emotional factors, such as depression, in the short term [43]. Longer yoga programs are expected to induce significant changes. In contrast to our findings, there was a study reporting that a 12-week yoga program significantly decreased depression scores in women with post-traumatic stress disorders [44]. Such differences in results may be attributable to the limitations of our sample; we speculate that the yoga intervention might have led to significant changes of depression scores if we had screened out the subjects with high depression scores and applied the yoga intervention to this subset of subjects. In the Western countries, the criteria for probable and definite depression are generally 16 and 25 points, respectively. In a depression epidemiological survey on South Koreans by Cho and Kim in 1993 [34], investigators suggested using 21 points as the baseline cut-off for screening for suspected depression and 25 points for screening for definite depression. CES-D scores at baseline in the experimental and control groups were 19.44 and 21.36, respectively. Compared with the level of CES-D (23.2 ± 11.62) of a previous study conducted to find out affecting factors on depression among female workers in small-sized workplaces [45], the level of depressive symptoms of participants in the experimental group in this study was relatively low. Also, these depressive symptoms' level in the study population belonged to the normal range. Thus, replication studies using a sample of participants with high depression scores are needed to investigate whether yoga interventions have a positive effect on workers' depression.
Usually, small enterprises lack an on-site health manager to provide systematic health management for workers, rendering it difficult to apply long-term health promotion programs and consequently hindering efforts to improve workers' health. We speculate that these factors would have contributed to the lack of significant health-promoting behaviors among the workers who underwent our yoga program. Studies on health promotion programs based on participatory action-research are underway in the field of industrial health research to ensure easy and long-term access [2]. This is a bottom-up type of intervention, in which participants proactively discover their health problems, and find and apply solutions with the help of researchers, as opposed to a top-down intervention in which the researcher identifies the problem, plans and designs a program to solve it, and applies the program to the participants. This type of intervention would be appropriate for small enterprises that lack on-site occupational health managers [2,46]. In fact, at least one study verified the effects of a participatory action-oriented training (PAOT)-type of intervention in lowering cholesterol, fasting glucose, body weight, and blood pressure while improving health-promoting behaviors in workers with a CVD risk [47]. The authors pointed out that many health promotion projects conducted in the worksites do not have any practical health benefits for workers because they are relatively short-term and passive, and also argued that methods of applying these health promotion interventions should be changed to address such limitations [2,46]. In addition, numerous studies emphasize the importance of community competency-strengthening projects as a strategy to ensure that enterprises continue to implement health promotion programs on their own even after the conclusion of the study. A previous study reported that taking a community competency-strengthening approach developed leadership and formed partnerships, which improved the self-regulation of health management, ultimately improving each individual participant's health-related quality of life and health promotion behaviors [2,47]. Based on these outcomes, we expect that efforts by small enterprises to foster health leaders for workers as a strategy to strengthen community's competence would also serve as an effective strategy for them to continue implementing health promotion programs on their own, and doing so would ultimately facilitate workers' health-promoting behaviors and improve their health-related quality of life. In light of these findings and suggestions, we are planning to develop and assess the effects of a PAOT-based health promotion program and competence-strengthening projects, such as training health leaders, as a strategy to lower CVD risks in workers of small enterprises. Through these studies, we will develop health promotion program strategies tailored to small workplaces that have low access to health promotion programs and generate long-term effects, contributing to developing policies to reduce CVD in workers of small enterprises. Furthermore, a previous study asserts that assessing the perceptions of the participants is important to improve the quality of clinical research [48]. Unfortunately, we did not collect data on participants' perceptions of this yoga program. Therefore, we propose to investigate the research participants' perception on the program in the follow-up study and use it in the intervention study to improve the quality of the research.
According to a meta-analysis of 61 studies investigating the relationship between blood pressure and mortality [44], a reduction of 2 mmHg of systolic blood pressure reduces mortality caused by ischemic heart disease by 7% and mortality caused by stroke by 10%. We verified that the yoga program developed in this study reduced diastolic blood pressure in workers of small enterprises, who are at higher risk of CVD than are those of medium-sized or larger enterprises, thereby confirming that this program is effective in lowering blood pressure, a critical physiological index related to CVD risks. Furthermore, physical measurement indexes such as waist circumference were significantly decreased in this study. Previous studies asserted that abdominal obesity has been identified as an important risk factor for metabolic syndrome and CVD [49,50]. Also, the other studies have shown that waist circumference was reduced relatively more in women than men [49,51]. Furthermore, Siu and colleges presented that one year of regular yoga training was effective on the improvement of waist circumference in middle-aged and older adults with metabolic syndromes [52]. Although it was analyzed that there was no gender difference between the experimental and the control group in the homogeneity test of this study, the ratio of women in the experimental group was relatively higher than that of the control group. Also, the average age of the participants in this study was 48.15 in the experimental group and 49.46 in the control group, which is considered to be a middle-aged group. Therefore, it was considered that the yoga program was effective in reducing waist circumferences in this study. In order to more reliably confirm the effect of yoga on the reduction of waist circumference, additional studies on workers considering age and gender are needed.
Our study has several limitations. First, sample size was relatively small (experimental group, n = 31; control = 38). Second, this study did not consider the level of individual capacity for the yoga program. Third, the study period was short to expect psychological effects such as depression and job stress. Therefore, a longer period of interventional study is needed in the future.

Conclusions
This study investigated whether a 12-week yoga program is effective in lowering psychological and physical measurements related to CVD risks in workers of small enterprises by using a nonequivalent control group design. The findings showed that the yoga program was effective in reducing physical measurements such as waist circumference and diastolic blood pressure in workers of small enterprises. However, this program could be an effective and efficient health program for the blue-collar workers who work in small-sized workplaces. Therefore, these findings provide evidence that a yoga program may serve as a new exercise strategy for feasibility among Korean workers who work in small-sized workplaces.