The Effects of a Campus Forest-Walking Program on Undergraduate and Graduate Students’ Physical and Psychological Health

We conducted a campus forest-walking program targeting university and graduate students during their lunchtime and examined the physical and psychological effects of the program. We utilized a quasi-experimental design with a control group and a pretest–posttest design. Forty-seven men (M = 25.5 ± 3.8 years) and 52 women (M = 23.3 ± 4.3 years) volunteered to participate (experimental group n = 51, control group n = 48). The intervention group participated in campus forest-walking program once a week for six weeks; they were also asked to walk once a week additionally on an individual basis. Additionally, participants received one lecture on stress management. Post-tests were conducted both just after the program ended and three months after. A chi-square test, t-test, and repeated measures analysis of variance were used to evaluate the effects of the program. Health promoting behaviors (F = 7.27, p = 0.001, ES = 0.27) and parasympathetic nerve activity (F = 3.69, p = 0.027, ES = 0.20) significantly increased and depression (F = 3.15, p = 0.045, ES = 0.18) significantly decreased in the experimental group after the intervention compared to the control group. In conclusion, using the campus walking program to target students during their lunchtime is an efficient strategy to promote their physical and psychological health.


Introduction
One of the most important public health problems today is individuals' lack of physical activity [1]. The World Health Organization has recommended that adults aged 18-64 years should do at least 150 min of moderate-intensity aerobic physical activity or at least 75 min of vigorous-intensity aerobic physical activity or an equivalent combination of moderate-and vigorous-intensity activity throughout the week [2]. The benefits of physical activity include lowering the rates of all-cause mortality, coronary heart disease, high blood pressure, stroke, type-2 diabetes, metabolic syndrome, and depression [3]. measures such as physical activity level; health promoting behavior; and physical, physiological, and psychological biomarkers of health [28,29]. The specific research questions for the study were as follows: Does the six-week campus walk program reduce depression and improve relaxation among college and graduate students? Does this program improve the health promotion behavior and increase physical activity of the students? We will test the following four hypotheses in this study: 1.
The experimental group participating in the program will display a higher score on health promoting behaviors than the control group will. 2.
The experimental group will display a lower depression score than the control group will. 3.
The experimental group will display more parasympathetic nerve activity than the control group will. 4.
The physical health (i.e., bone density, blood lipid profile, and body mass index) of the experimental group will be better than that of the control group.

Study Design and Participants
This was a quasi-experimental study with a control group and a pretest-posttest design. The study occurred from September 2014 to February 2015 (from fall to the end of winter). For the intervention group, the campus forest-walking program was provided for six weeks, and the effects of the program regarding participants' physical and mental health were analyzed. The pre-test took place about a week before the intervention started, the post-test was conducted one week after the final program, and the follow-up test was conducted after another three months.
Participants comprised graduate and undergraduate students from one university in Seoul, South Korea. Participants were recruited by posting notices using the university homepage and sending an e-mail to all students from the university health service center. Students who had medical contraindications to exercise by self-report (e.g., asthma, painful osteoarthritis, or heart conditions) were excluded [30]. One-hundred and eighteen students voluntarily participated in this study. To increase motivation, group assignment (experimental or control) was made per participants' preference ( Figure 1).  This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Institutional Review Board at Seoul National University in Seoul, Republic of Korea (IRB No. 1409/001-001). Before beginning the study, a full explanation about the research purpose, the experimental procedure, and all measurement indices were provided to participants. Those who agreed to participate were selected as study participants who provided full written consent for taking part in the program (experimental group n = 60, control group n = 58). Three of the participants withdrew their participation during the study period. Sixteen participants did not complete the follow-up test because of employment, birth, a heavy schedule, or refusal.

Intervention
The six-week campus forest-walking program was conducted once a week during lunch. The university campus where the program was implemented is located just northwest of the mountain Gwanack and has many different trees. In addition, there are forest roads and trails near the campus.
The information consisted of lectures on stress management and providing leaflets related to mental and physical health. We sent a text message to promote voluntarily walking once a week. Health leaflets about the effects of forest therapy; the correct walking method; and self-efficacy for walking, stress management, and depression management were provided at the first session of the program. In addition, participants in the experimental group were provided with a wearable activity tracker, Fitbit Zip ® (Fitbit Inc., San Francisco, CA, USA), which allowed them to self-monitor their physical activity ( Figure 2).
The intervention was provided from September to October 2014, once a week during lunchtime or for one hour on Wednesdays at 4 p.m., depending on participants' preferred schedule. Participants in the experimental group walked together in the campus forest at a relaxed pace for about 40 min, with a 10 min rest during the walk, and had a light meal (e.g., sandwich) during the walk.
During the intervention period, we also provided one lecture on stress management. The average group size was 10 (not more than 15). They were also encouraged through a text message to additionally walk at least once a week at their leisure. The control group did not receive leaflets, lectures, or a wearable activity tracker and were asked to follow their routine activity during the study period. The intervention group and control group were asked to participate in pre-, post-, and follow-up tests.

Measurements
Questionnaires were administered to investigate demographic data, health promoting behavior, physical activity level, and depression. Demographic data included age, sex, and academic year.
Health promoting behavior was assessed using the Korean translation of the Health-Promoting Lifestyle Profile II (HPLP-II) [31]. The HPLP-II is a 50-item measure; answers are provided using a four-point Likert scale based on Pender's health promotion model, which contains six subscales: responsibility for health, physical activity, healthy nutrition, social relations, stress management, and spiritual growth. The total scores of the HPLP-II range from 50 to 200, with a higher score indicating a better health-promoting lifestyle. The Cronbach's α for the total instrument was 0.92 and the six subscales ranged from 0.65 to 0.82 [26]. In this study, the Cronbach's α = 0.91 and the six subscales ranged from 0.70 to 0.87.
Physical activity was measured using the International Physical Activity Questionnaire-Short Form, which provides a measure of total physical activity accrued through work and leisure in all settings combined. Participants were asked about frequency and duration of exercise by indicating days and time spent doing vigorous and moderate activities, walking, and sitting in the last seven days. These activity categories may be treated separately to obtain the specific activity patterns or multiplied by their estimated value in Metabolic Equivalent of Tasks (MET) and summed to gain an overall estimate of physical activity in a week.
Total physical activity MET-min/week, which can be calculated as the sum of walking + normal + active MET-minutes/week scores [32], was used for the analysis.
Depression level was measured using the Beck Depression Inventory. It is composed of 21 items and is answered using a four-point (0-3) Likert scale. The total sum of the item scores ranged from 0 to 63, with higher total scores indicating more severe depressive symptoms.
Height and weight were measured with an automatic stadiometer (BSM 370, Inbody Co., Ltd., Seoul, Korea). Body composition was assessed including amount of body fat, body fat percentage, body mass index, skeletal muscle mass, and amount of muscle by body composition analyzer (Inbody 570, Inbody Co.). After 10 min of resting, blood pressure was measured. Bone density was measured at the right calcaneus bone using broadband ultrasound attenuation (Sonost 3000, OsteoSys Co., Ltd., Seoul, Korea). In general, bone density indicates the value with a t-score. "−1 and above" the t-score is normal; "−2.5 to −1" means osteopenia, a condition where the bone density is below normal, which may lead to osteoporosis; and "−2.5 or below" means osteoporosis [33].
Heart Rate Variability (HRV) was measured using a portable electrocardiograph (LXC3203, LAXTHA Inc., Daejeon, Korea). HRV data were obtained at various frequency bands using an HRV software tool (TeleScan, LAXTHA Inc., Daejeon, Korea). After 10 min of resting, HRV was measured. A heart rate monitor using unipolar limb lead electrocardiogram (ECG) recorder was used to collect continuous RR intervals, which were stored on computer for later analysis. It took approximately 5 min to measure this. Among commonly used HRV indices, the low-frequency (LF: 0.04-0.15 Hz) power of the HRV Fourier spectrum has been presumed to reflect some aspects of cardiac sympathetic modulation, and the ratio of LF power to high-frequency (HF: 0.15-0.40 Hz) power (LF/HF ratio) indicates the sympathovagal balance [34]. Parasympathetic nerve activity is high when participants are calm. It is widely used to measure the mediation effect of mind and body relaxation [35].
Blood samples were taken to determine fasting serum total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglyceride. Participants were asked not to eat or drink for 12 h before testing. One laboratory determined all blood biochemistry parameters (Green Cross Laboratories, Gyeonggi-do, Korea).

Statistics
Statistical analyses were performed using a Windows-Based Statistical Package version 22.0 software (SPSS, Chicago, IL, USA). Descriptive statistics comprised mean, standard deviation, frequency, and percentage to present demographic information and outcome variables. An independent t-test and χ 2 test were conducted to test homogeneity at baseline between the experimental and control groups. For this study, a repeated-measures analysis of variance (rmANOVA) was used to analyze the effects of intervention directly after the program and three months later. All statistical tests were two-tailed and a p-value < 0.05 was considered statistically significant.

Homogeneity Test of the Experimental and Control Group
Participants' mean age was 24.3 ± 4.19 years. A homogeneity test of the general characteristics between the experimental and control groups in pretest showed no significant difference (Table 1).

Effects of the Intervention on Outcome Measures
To analyze the effect of the intervention, a rmANOVA and Mauchly's test of sphericity were used. Mauchly's test for sphericity, a test of homogeneity of variance, tests the null hypothesis (p > 0.05) that the differences between variances comparing all possible pairs of groups are equal. In those cases, when the null hypothesis is not true (e.g., high-density lipoprotein (HDL), low-density lipoprotein (LDL), Bone density, body mass index (BMI), low frequency/high frequency (LF/HF) ratio), we used a multivariate analysis of variance (MANOVA) [36].
The interaction effect of group (experimental and control) and time (pre-intervention, post-intervention, and three-months after the intervention) were tested. Mean scores of physical activity level and health promoting behavior for experimental and control groups are presented in Table 2 and Figure 3. The results of the repeated measure ANOVA showed that there was a significant group × time effect for the health promoting behavior (F = 7.27, p = 0.001, ES = 0.27). Among the six subscales of health promoting behavior, there were significant group × time effects for physical activity (F = 5.91, p = 0.003, ES = 0.25), healthy nutrition (F = 3.64, p = 0.028, ES = 0.19), stress management (F = 3.32, p = 0.038, ES = 0.18), and spiritual growth (F = 3.14, p = 0.045, ES = 0.18). There was a significant difference in health promoting behavior in the time by group interaction effect; however, no interaction effect was found for physical activity level.   Figure 4).

Discussion
We evaluated the effects of an on-campus, forest-walking program on the physical and psychological health of undergraduate and graduate students. Pre-, post-, and follow-up variables were measured to compare an experimental group and a control group.
Participants in the intervention group had a higher score on health promoting behavior than the control group did, especially physical activity, health nutrition, stress management, and spiritual growth scores among the six subscales. This is consistent with the effect of resistance band exercise (low intensity exercise) among nursing students [37]. Although the campus forest-walking program was a low intensity activity for six weeks, it provided the opportunity to be motivated by periodic group walking, self-monitoring, and concentrating on health. It is meaningful that this intervention positively affected the health-promotion behavior of participants-including a healthy diet, physical activity, stress management, and spiritual growth-because this will subsequently lead to better lifestyle. The pleasant experience of forest walking increased participants' interest in health care, and the stress management education may have influenced this positive effect.
In this study, the campus forest-walking program had no significant effect on physical activity level. Moreover, physical activity level decreased in both groups at the three-month follow-up, and significant differences were only shown for the time variable. The intervention program was provided in the autumn (September to November 2014), and the follow-up test was conducted at the end of winter (February 2015). The wintry weather could affect the level of physical activity for both groups. Therefore, the decrease in PA may have been affected by the timing of the program. Moreover, because winter has fewer sunlight hours and disadvantageous weather conditions, physical activity may be less convenient than it is in the summer [38]. Walking programs are known to be effective in promoting physical activity. Therefore, we can consider various strategies for future research: interventions tailored to people's needs; targeting the most sedentary or the most motivated to change; and intervening at either the individual, household, or group level [39].
The effectiveness on psychological health, which was shown with higher levels of parasympathetic nerve activation and decreased depression, is the most noteworthy aspect of the forest-walking intervention. The results of increased parasympathetic activity in the experimental group are consistent with previous studies of physiological responses to the forest environment, suggesting that walking in forests may provide health benefits of relaxation [35,40,41]. Numerous forest-walking studies have shown the effects on relieving negative emotions such as anxiety and depression [17,18,[42][43][44]. Walking in the forest has also been shown to be more effective on mental health than walking in the city due to less environmental stress [45]. Young adults who walked for 50 min in nature after a stressful situation showed reduced stress and anger and increased positive affect [46]. Our results add evidence to the known positive effects of forest walking in increasing mental relaxation and reducing depression. Therefore, forest walking could be a simple, accessible and low-cost strategy to improve physical and mental health.
We acknowledge the limitations of this study. The limitations include the lack of randomization, which may lead to selection bias. In addition, participants chose to take part either in the intervention or control groups: attitudes, motivation, and personal factors might have caused a selection bias. Participants were highly motivated and were willing to volunteer to be in the experimental group. People who are highly motivated to change their behavior may be more inclined to take part in an intervention and more likely to benefit from the intervention. Although the two groups did not seem to differ significantly at baseline in relation to the measures, we did not control for personal orientation or motivational processes; therefore, the findings may not be directly generalizable to all college students. In addition, the intervention group was required to exercise on their own once a week. Although the physical activity level was measured with a Fitbit, the Fitbit data were not evaluated. Additionally, we did not know if within the intervention group some students engaged in self-walking more than others. Social factors related to performing an activity with others could have played a role. Moreover, small intervention groups were selected according to individuals' schedules. While walking, participants could see and 'feel' the forest without talking to each other or looking at a cell phone; however, we cannot rule out the possibility that the nature of the program and meeting new people could have affected the outcome. Finally, because the walking program was offered to participants once per week, the frequency may be different from other intervention studies, which makes comparisons between studies difficult.
Despite these limitations, our study has several notable strengths including the use of the campus forest, which is easily accessible to students, and performing the program at lunch time, which is convenient. In addition, the physiological and psychological results of this experiment provide evidence for the physiological and psychological benefits of campus forests. Walking in the campus forest reduced depression by increasing parasympathetic activity. In addition, we examined participants at a three-month follow-up and included multiple outcome variables such as physiological measures and self-reported measures.

Conclusions
This study showed that a six-week, on-campus, forest-walking program produced beneficial changes in the physical and psychological health of undergraduate and graduate students. The intervention significantly increased participants' health promoting behavior and parasympathetic nerve activity compared to the control group, as well as significantly decreasing depression.
Further research is needed to examine changes in exercise frequency and duration. A study with a larger sample size is required to address the generalizability of the results. The random assignment of participants and manipulating potential confounding factors of the intervention are also needed. However, this study suggests a campus forest-walking program as a strategy to promote students' physical and mental health, thereby contributing to the health promotion and a healthy culture among university and graduate students. Therefore, students should be encouraged to take advantage of the forest during their lunch hour as it may positively affect their quality of life.
writing the paper. Min Kyung Song contributed towards the intervention, data collection, and writing the paper. All authors read and approved the final manuscript.