Sustainability of Hiking in Combination with Coaching in Cardiorespiratory Fitness and Quality of Life

Although strong evidence shows that physical inactivity and sedentary behavior are associated with many negative health outcomes, inactive lifestyles are still increasing. Consequently, new approaches must be developed to increase adherence to an active lifestyle and hence a longer life. Green exercise and health coaching could be effective ways to induce long-lasting lifestyle changes geared towards more physical activity. In this randomized controlled trial, we investigated the effects of mountain hiking and psychological coaching on adults with a sedentary lifestyle. The coaching group (n = 26) participated in a 7-day guided hiking program with three personal coaching sessions, whereas the hiking group (n = 32) received no coaching. The effects on aerobic capacity, spirometry and quality of life were assessed at baseline (day 0), after the intervention week (day 7) and after 80 days. Fully nonparametric statistical analysis revealed a gender-based effect for aerobic capacity—the female participants of the coaching group showed a greater improvement (p = 0.03) than the hiking group. No significant effects were found for spirometry. Quality of life parameters improved in both groups. In conclusion, both green exercise and health coaching are capable of inducing improvements in health-related quality of life and cardiorespiratory fitness. No superior effects of health coaching were found.


Introduction
Although strong evidence shows that physical inactivity and sedentary behavior are associated with many adverse health effects, the "global pandemic" of inactive lifestyles is still on the rise [1,2]. Sedentary behavior is generally defined as any waking behavior with an energy expenditure ≤ 1.5 METs in a sitting or reclining posture, whereas physical inactivity is characterized by the lack of sufficient moderate-to vigorous-intensity physical activity [3]. Sedentary behavior can therefore be seen as the lowest edge of the physical activity spectrum. An inactive lifestyle is associated with many adverse health effects including increased risk of coronary heart diseases, hypertension, type 2-diabetes, as well as cancer and reduced life expectancy [1,[4][5][6][7]. Furthermore, sedentary behavior increases markers associated with inflammation, the risk of obesity, depression, musculoskeletal diseases and osteoporosis risk for women [8][9][10][11][12]. Physical inactivity and sedentary behavior cause not only morbidity and mortality, but also create a major economic burden, especially in high-income countries [13].
A well-known factor associated with physical inactivity is urbanization, which is rapidly increasing worldwide. In 1970, only 36.6% of the world's population lived in urban settlements. This number had risen to 55.3% in 2018 and is expected to reach 60.4% by to an improvement of physical and mental health and may enhance the quality of life [40]. Most studies using health coaching concern chronic diseases such as diabetes type 2 [39,41]. So, there is a lack of research on the impact of coaching in people with predominantly sedentary lifestyles and the influence on health parameters, especially in terms of gender and age [42].
From the aspect that sedentary behavior at work is not sufficiently compensated during leisure time [43], innovative approaches to promote the enjoyment of physical activity and thus encourage adherence to an active lifestyle supporting a healthy lifespan must be developed. Health coaching in combination with supervised green exercise could initiate long-lasting lifestyle changes. The aim of the presented HICO study was to investigate the effects of a 7-day intervention with green exercise and health coaching on cardiorespiratory fitness and quality of life of sedentary couples. To determine improvements in cardiorespiratory fitness, aerobic capacity and spirometry parameters were measured. For mental health enhancement, questionnaires EQ-5D-5L and SF-36 were provided. Hence, the following hypotheses are approached: Hypothesis 1 (H1). The combination of hiking and coaching improves the cardiorespiratory fitness after 80 days more sustainably than hiking without coaching.
Hypothesis 2 (H2). The combination of hiking and coaching improves the cardiorespiratory fitness after seven days more than hiking without coaching.

Hypothesis 3 (H3).
The combination of hiking and coaching improves the quality of life more than hiking without coaching.

Study Design and Settings
We performed a randomized, controlled trial (HICO Study, https://doi.org/10.1186/ ISRCTN25562081 (accessed on 2 December 2021)) to investigate the combined effects of coaching and moderate mountain hiking on the cardiorespiratory fitness of couples with a sedentary lifestyle. In the HICO Study, two intervention groups (hiking and coaching) and one non-intervention control group were included. Only the two intervention groups were finally analyzed due to a high dropout rate and recruitment problems in the control group. Therefore, this work is focused on the comparison of the hiking and coaching group. The allocation ratio for all groups was set at an equal sample size. The hiking group participated in a 7-day mountain hiking program and the coaching group additionally received several coaching sessions with a psychologist. The study protocol was approved by the Ethics Committee of Salzburg (415-E/1488/2-2012) and the study was conducted in Pinzgau (Salzburg Land, Austria) between June and September 2012. Follow-up examinations took place in Salzburg (Austria) between October and December 2012.

Participants
Eligible participants were couples with a sedentary lifestyle. Participants were recruited all over Austria and Germany through advertisements in newspapers and communication via webpage (http://gesund-umdenken.com/was-wir-bieten/klinische-studie/ index.html) between May and June 2012. Written informed consent was obtained from all participants. Inclusion criteria were age 22-54 years and a sedentary lifestyle, which means a maximum energy expenditure of 1.0-1.5 metabolic equivalents per day and the physical ability to participate in moderate hiking tours. The questionnaire "Assessment of the Physical Activity Level with two Questions" by Johansson and Westerterp [44] was used as a measure of a sedentary lifestyle. As no official German translation exists, the questionnaire was translated by the authors themselves. Only people scoring ≤ 1.6 were included. Exclusion criteria were as follows: non-sedentary lifestyle (Score > 1.6), cardiovascular diseases, severe hypertension (≥level 3), antihypertensive medication, pulmonary dysfunction, uncontrolled metabolic diseases (e.g., diabetes), malignant neoplastic diseases, orthopedic diseases, acute pain, active infectious diseases and pregnancy. Exclusion criteria were screened during the recruitment process using a short survey.

Intervention
The study was carried out as part of a 7-day vacation in four regions in Pinzgau (Salzburg Land). The hiking and coaching group participated in an identical mountain hiking exercise program. All participants completed five hiking tours between Sunday and Friday (no hiking on Wednesday) with a daily difference in altitude of at least 600 m. The hiking tours were carried out in comparable mountain massifs, including the High Tauern, Northern Limestone Alps, Berchtesgadener Alps and Kitzbühel Alps. Regarding the technical classification of the trails, mainly "blue" and short stages "red" paths were completed during the hiking; these mean easy and moderately difficult, sometimes also narrow and steep trails, which have no or hardly any areas with a risk of falling. In addition, the coaching group received three individual psychological coaching sessions with a psychologist owning a university diploma in health psychology. All coaching sessions were performed in a quiet and pleasant ambiance at the hotels where all participants stayed overnight. Each coaching session lasted approximately 1.5 h. A follow-up visit was scheduled 80 days after the first intervention. The coaching group received another individual psychological coaching session at the follow-up meeting. All medical examinations were performed by members of the Institute of Ecomedicine and of the Institute of Physiology and Pathophysiology from the Paracelsus Medical University of Salzburg respectively.

Data Collection and Outcomes
Data were anonymized using four-digit-IDs. The overall trial start was on 1 January 2012. After a preliminary phase, the recruitment of subjects took place from 7 May 2012, followed by randomization. The intervention phase started on 30 June 2012. The study ended with a follow-up phase and data analysis on 1 February 2013. Medical examinations at baseline (T0; day 0) and after the intervention phase (T1; day 7) were performed in mobile lab setups in the participants' accommodations. Follow-up examinations (day 80; T2) were completed at the Paracelsus Medical University in Salzburg, Austria. Questionnaires were handed out for completion at baseline (T0, day 0), after the intervention phase (T1, day 7) and at the follow-up meeting (T2, day 80). An overview is given in Figure 1. The assessments were conducted, with a warm-up before, outdoors in the summertime at the same time of the day.

One-Mile Walking Test
V O 2 max [mL kg −1 min −1 ] describes the maximum rate of oxygen consumption during exercise. The 1-mile walking test is a validated and economical method to estimate the V O 2 max indirectly [45]. In preparation for the 1-mile walking test data was collected on weight, age and gender of every person. The values of V O 2 max are different for gender and age. Men have a higher maximum rate of oxygen consumption than women with a comparable fitness level [46,47]. The participants were asked to walk 1 mile (1.6 km) as fast as possible. At the end of the walk heart rate and oxygen saturation were measured. The estimated V O 2 max was calculated using the equation by Kline et al.

One-Mile Walking Test
[mL kg −1 min −1 ] describes the maximum rate of oxygen consumption dur exercise. The 1-mile walking test is a validated and economical method to estimate indirectly [45]. In preparation for the 1-mile walking test data was collected weight, age and gender of every person. The values of are different for gen and age. Men have a higher maximum rate of oxygen consumption than women wit comparable fitness level [46,47]. The participants were asked to walk 1 mile (1.6 km) fast as possible. At the end of the walk heart rate and oxygen saturation were measur The estimated was calculated using the equation by Kline et al. [45]: 6.965 (0.0091 × WT) − (0.0257 × AGE) + (0.5955 × SEX) − (0.220 × T1) − (0.0115 × HR); WT = wei in pounds; AGE= Age in years; SEX = Gender: 0 = female, 1= male; T1 = time for 1 mile minutes; HR = heart rate.

Questionnaires for Health-Related Quality of Life
TheEuroQOL-5 Dimension Questionnaire version (EQ-5D-5L) involves five dim sions with five levels each. The answers for all dimensions yield a five-digit number t describes the participant's health status adapted for each country. EQ-5D-5L also inclu a visual analogue scale (VAS) which records the participant's self-rated health st [48,49]. The 36-Item Short Form Health Survey (SF-36) comprises 36 items which can subsumed into eight concepts: physical functioning, bodily pain, role limitations due physical health problems, role limitations due to personal or emotional problems, em tional well-being, social functioning, energy/fatigue, and general health perceptions. T scores of each concept range from 0 to 100. The lower the score in a concept, the m limited the participant is [50].

Questionnaires for Health-Related Quality of Life
TheEuroQOL-5 Dimension Questionnaire version (EQ-5D-5L) involves five dimensions with five levels each. The answers for all dimensions yield a five-digit number that describes the participant's health status adapted for each country. EQ-5D-5L also includes a visual analogue scale (VAS) which records the participant's self-rated health state [48,49]. The 36-Item Short Form Health Survey (SF-36) comprises 36 items which can be subsumed into eight concepts: physical functioning, bodily pain, role limitations due to physical health problems, role limitations due to personal or emotional problems, emotional wellbeing, social functioning, energy/fatigue, and general health perceptions. The scores of each concept range from 0 to 100. The lower the score in a concept, the more limited the participant is [50].

Coaching
The coaching group received three single coaching sessions of 90 min at days 3, 5, 7 and 80. The coaching sessions were standardized and performed by four certified coaches (psychologists, University of Salzburg) with a university diploma in coaching.
Due to the short intervention time, participants could not receive more coaching sessions as in the usual psychological coaching process. Based on the grow model [52], participants were supported in their goal settings, in checking the actual-/desired condition (reality), in choosing the best and appropriate options, and in beginning the first concrete action steps (will power, what, when, who).
Session 1: The first session aimed to create an optimal basis for coaching by focusing on the coach-client relationship and creating an acceptance of the method. Each client had the opportunity to set three individual, relevant goals on the subject of health. These goals were roughly classified in the areas of sport, nutrition and stress. The goals were then accurately reflected and operationalized to reduce their abstraction and to increase target specificity and clarity. The final part of the first session was the discussion of the test results of the burnout screening scales [53].
Session 2: In the second session, the results of two stress-related questionnaires were discussed. With the help of the trier inventory for chronic stress [54] and the stress processing questionnaire [55] specific stressors of the participants were identified. These results were discussed in detail in the further course of the session in coordination with the client's goals and roughly analyzed for strengths, weaknesses, opportunities, threats in the sense of the SWOT model. Session 3: In the third session, the main goal of the client was concretized by intensively discussing the first steps to be taken for the time after the coaching. Clear, measurable and verifiable goals and sub-goals were set by checking them for specificity, measurability, acceptance, feasibility and scheduling on the basis of the SMART criteria. During the preparation of the implementation plan, milestones were also created, other people from the client's environment were integrated (for example, as initiates in the plans, as feedback providers or as controllers) and a relapse prophylaxis was set up in the event of failure.

Randomization and Sample Size
Randomization was performed with the "Random Allocation Software" (Isfahan, Iran) program with a block randomization protocol [56]. Recruitment of eligible participants, randomization and assignment to treatments were performed by Arnulf Hartl. No a priori sample size calculation was performed.

Statistical Analysis
In an intention-to-treat analysis, all statistical analyses were entered into the R-GNU software environment (General Public License, R Foundation for Statistical Computing, Vienna, Austria, Version 4.0.2). Variables reported in tables were presented as means and their standard deviation, as far as not stated otherwise. Missing values were replaced by two methods: LOCF (last outcome/observation carried forward) if data was missing on day 7 or day 80 by random and NOCB (next outcome/observation carried backward) for missing data on day 0. For all tests, a significance level of 5% probability was set. As the data were not normally distributed, longitudinal data analysis was performed with the nparLD-package [57]. This package offers ANOVA-type statistics for nonparametric longitudinal data analysis. Within the F1-LD-F1-model from the nparLD-package, group (hiking, coaching) was defined as whole-plot-factor and time (T0, T1 and T2) as sub-plotfactor. The F1-LD-F1 model provides an ANOVA-type statistic for group, time and the interaction of group and time (group × time). In case of significant main effects for time or treatment, post hoc tests were applied for a comparison of T0 and T1, respectively, T0 and T2 with another F1-LD-F1-model. Post hoc tests were amended for multiple testing by the Bonferroni-Holm method.
Next to the ANOVA-type statistic, the F1-LD-F1 model offers relative treatment effects (RTE) as a unitless measure of effect size. The RTE reaches values between 0 and 1 and can be interpreted as follows: An RTE of 0.25 for a certain subgroup means that the probability of a randomly chosen person from this subgroup to score higher than a randomly chosen person from the entire dataset is estimated to be 25%. On the other hand, the probability that a randomly chosen person from this subgroup scores lower than a randomly chosen person from the entire dataset is estimated to be 75%. An RTE equal to 0.50 means no tendency for a higher or lower score in any subgroup.

Sample Size Simulation
In addition, we performed a post hoc sample size calculation with the primary outcome aerobic capacity. In order to meet the requirements of modern statistical approaches, the sample size calculation was performed based on a bootstrap simulation for F1-LD-F1 models and ANOVA [58]. Within the bootstrap simulation, the group size of n = 20 to n = 70 was varied by steps of ten for each group with random values from the corresponding group (initial seed was set at 1). The statistical power can be estimated for each group size by the percentage of significant (p-value < 0.05) counts.

Study Participants and Baseline Characteristics
Out of 90 eligible people, 28 were enrolled for the coaching group and 36 people for the hiking group. In total, 26 people were excluded because of personal reasons or because they did not meet the inclusion criteria. Two participants from the coaching and four from the hiking group declined to participate because of personal reasons. For the statistical analysis, 26 participants of the coaching group and 32 participants of the hiking group were included ( Figure 2). All participants tolerated the hiking and coaching program well. No harm or unintended effects were observed. come aerobic capacity. In order to meet the requirements of modern statistical a the sample size calculation was performed based on a bootstrap simulation fo models and ANOVA [58]. Within the bootstrap simulation, the group size of n 70 was varied by steps of ten for each group with random values from the corr group (initial seed was set at 1). The statistical power can be estimated for each by the percentage of significant (p-value < 0.05) counts.

Study Participants and Baseline Characteristics
Out of 90 eligible people, 28 were enrolled for the coaching group and 36 the hiking group. In total, 26 people were excluded because of personal reasons they did not meet the inclusion criteria. Two participants from the coaching and the hiking group declined to participate because of personal reasons. For the analysis, 26 participants of the coaching group and 32 participants of the hik were included (Figure 2). All participants tolerated the hiking and coaching pro No harm or unintended effects were observed.  Baseline characteristics show no relevant differences between the study groups (Tables 1 and 2), except for age, V O 2 max and two variables of the SF-36 questionnaire. The coaching group is significantly older than the hiking group (t (55.1) = −2.92, p = 0.01) and the hiking group also shows significantly higher scores for V O 2 max than the coaching group. The values of physical functioning and bodily pain are significantly higher in the hiking group than in the coaching group (physical functioning: W = 539.5, p = 0.03; bodily pain: W = 571.5, p = 0.01). Descriptive statistics over all time points are summarized in Tables A1 and A2 in Appendix A.

One-Mile Walking Test
For V O 2 max of the 1-mile walking the F1-LD-F1-model provided a significant main effect for time (Table 3) but post hoc tests did not show any interaction effects at single time points. Because V O 2 max depends on gender, women and men were analyzed separately by the F1-LD-F1 models. For men, a significant time effect was found but post hoc tests did not yield any significant effects. The RTEs indicate a parallel development of both groups, whereas the coaching group is generally characterized by lower levels for V O 2 max (Table 3, Figure 3). For women, a significant main effect was also found for time, but post hoc tests revealed a significant interaction effect on day 80. The RTEs indicate a greater improvement of the coaching group between day 7 and day 80 (Table 3, Figure 3).

EQ-5D-5L
The F1-LD-F1-model for the visual analogue scale of EQ-5D-5L revealed a significant main effect for time but post hoc tests did not show any interaction effects at the single time points. Both the hiking (83.4 ± 9.71 vs. 86.88 ± 9.31%) and the coaching group (80.38 ± 15.09 vs. 87.31 ± 7.78%) rated their health status better. The analysis of the EQ-5D-5L

EQ-5D-5L
The F1-LD-F1-model for the visual analogue scale of EQ-5D-5L revealed a significant main effect for time but post hoc tests did not show any interaction effects at the single time points. Both the hiking (83.4 ± 9.71 vs. 86.88 ± 9.31%) and the coaching group (80.38 ± 15.09 vs. 87.31 ± 7.78%) rated their health status better. The analysis of the EQ-5D-5L index showed no significant main effects for treatment, time, or interaction (Table 4).

SF-36
The F1-LD-F1-model for sum scores of the SF-36 questionnaire revealed a significant effect for time in general health, role emotional, physical dimension, mental dimension, and total score, indicating an increase in both groups. Post hoc tests did not show any interaction effects at the single time points. Effects in treatment were significant in physical functioning and bodily pain. For bodily pain, RTEs show higher improvement of the coaching group with time (Table 4).

Sample Size Simulation
The post hoc sample size simulation yielded high differences between the two statistical models (F1-LD-F1 and ANOVA) for sample sizes ≤ 50. (Figure 4, Table A3). The F1-LD-F1 model reaches with n = 50 already an acceptable power of 1 − β = 0.97 whereas the estimated power for the ANOVA lies by 1 − β = 0.90. Similar results for both statistical models are evident from a sample size ≥ 60 per group. Hence, in sample sizes n ≥ 60 per group an acceptable estimated power of 1 − β ≥ 0.94 can be expected for both statistical models.
The post hoc sample size simulation yielded high differences between the two statistical models (F1-LD-F1 and ANOVA) for sample sizes ≤ 50. (Figure 4, Table A3). The F1-LD-F1 model reaches with n = 50 already an acceptable power of 1 − β = 0.97 whereas the estimated power for the ANOVA lies by 1 − β = 0.90. Similar results for both statistical models are evident from a sample size ≥ 60 per group. Hence, in sample sizes n ≥ 60 per group an acceptable estimated power of 1 − β ≥ 0.94 can be expected for both statistical models.

Discussion
A wide range of adverse health effects is associated with sedentary behavior and physical inactivity. Urbanization and lifestyle changes promote an inactive lifestyle, leading to a global increase in chronic diseases [17,59,60]. Therefore, new concepts are urgently needed to bring people back to an active and healthy lifestyle. Health coaching together with moderate mountain hiking as a sport that needs little equipment and personal skills could be instrumentalized to induce a more active lifestyle in the working population. The aim of the presented randomized, controlled trial is to examine the effects of moderate green exercise in form of mountain hiking and health coaching and on the cardiorespiratory fitness and quality of life of couples with a sedentary lifestyle.
A valid parameter to evaluate cardiorespiratory fitness is spirometry. Within this population of sedentary couples, no significant changes were found for any spirometry parameter, neither as a short-term effect nor a long-term effect. Both groups start with a good lung function and keep this level throughout the intervention and post-treatment phases. As the baseline levels are already above average and the intervention duration is rather short, no relevant changes can be expected either way. Another well-established indicator of cardiorespiratory fitness is aerobic capacity. The direct measurement of aerobic capacity is very time and cost-intensive because of the need for trained staff and technical equipment [61]. For this reason, we performed a less expensive but well-validated method instead: the one-mile walking test. In this submaximal exercise test, the participants are asked to walk one mile as fast as possible [45]. The estimated aerobic capacity was analyzed separately for men and women. Although a significant baseline difference for aerobic capacity was found between the intervention groups, the statistical analysis revealed a significant time effect. The hiking group (9.89 ± 6.35 L/kg * min) starts with a

Discussion
A wide range of adverse health effects is associated with sedentary behavior and physical inactivity. Urbanization and lifestyle changes promote an inactive lifestyle, leading to a global increase in chronic diseases [17,59,60]. Therefore, new concepts are urgently needed to bring people back to an active and healthy lifestyle. Health coaching together with moderate mountain hiking as a sport that needs little equipment and personal skills could be instrumentalized to induce a more active lifestyle in the working population. The aim of the presented randomized, controlled trial is to examine the effects of moderate green exercise in form of mountain hiking and health coaching and on the cardiorespiratory fitness and quality of life of couples with a sedentary lifestyle.
A valid parameter to evaluate cardiorespiratory fitness is spirometry. Within this population of sedentary couples, no significant changes were found for any spirometry parameter, neither as a short-term effect nor a long-term effect. Both groups start with a good lung function and keep this level throughout the intervention and post-treatment phases. As the baseline levels are already above average and the intervention duration is rather short, no relevant changes can be expected either way. Another well-established indicator of cardiorespiratory fitness is aerobic capacity. The direct measurement of aerobic capacity is very time and cost-intensive because of the need for trained staff and technical equipment [61]. For this reason, we performed a less expensive but well-validated method instead: the one-mile walking test. In this submaximal exercise test, the participants are asked to walk one mile as fast as possible [45]. The estimated aerobic capacity was analyzed separately for men and women. Although a significant baseline difference for aerobic capacity was found between the intervention groups, the statistical analysis revealed a significant time effect. The hiking group (9.89 ± 6.35 L/kg × min) starts with a higher aerobic capacity in comparison to the coaching group (44.04 ± 8.94 L/kg × min). During the 7-day intervention, the aerobic capacity is improved in both groups. However, the relative treatment effects indicate a stronger increase in the hiking group, which could be explained by the lower baseline values. Within the female subgroup, a significant time effect was detected, indicating a slight decrease of aerobic capacity in the hiking group during the 7-day intervention, followed by an increase during the post-treatment phase. In contrast, the aerobic capacity of the female participants of the coaching groups improves their aerobic capacity already during the intervention period and shows a clear increase during the post-treatment phase. This different development in the female subgroup is reflected by a significant interaction effect at day 80 (treat × time p = 0.03). Although the changes in aerobic capacity occur in a minimal amount, it reveals a possible gender effect: females seem to be receptive to the health coaching approach.
Looking at Hypothesis 1-The combination of hiking and coaching improves the cardiorespiratory fitness more sustainably than hiking without coaching this gender aspect must be considered. Women seem to have a better relation to their feelings and impulses and also tend to attribute their "wrong" behavior to internal causes due to a lack of knowledge and skills [62]. Furthermore, women accept the activities and health recommendations of the coach more than men do, as Linning et al. [63] show in their study on the promotion of fitness and health in employees. Thus, it is not unexpected that women rate their personal coaching process outcome more positively than men. Women seem to be more able to establish good working relations with the coach which also had a positive impact on the evaluation of coaching effectiveness [64]. Other studies also found significant differences in gender because of the interpersonal variation in how people participate in and progress through a health coaching program [65]. Apart from this gender-based coaching effect, further studies should also include the neurological aspects of (green) as Mason et al. [66] show that a diet and exercise intervention with physical activity training can reduce reward-driven eating and, consequently, promote weight loss.
Within this study population of sedentary couples, no evidence was found to support Hypothesis 2-Combination of hiking and coaching improves cardiorespiratory fitness more than hiking without coaching. Although the aerobic capacity improves in the male subgroup during the 7-day intervention no superior effects were found for the coaching group. Furthermore, the changes in aerobic capacity are rather small. The baseline levels of spirometry are above average, and the aerobic capacity is also in a normal range. This leaves little space for improvements. However, the questionnaire by Johannsson and Westerterp [45] should be critically evaluated as an inclusion criterion. In addition, no official German translation exists and the translation by the authors themselves may create a bias.
Besides cardiorespiratory fitness, health-related quality of life is an important patientcentered outcome. The SF-36 and the EQ-5D-5L were used for the measurement of the quality of life. The EQ-5D-5L index clearly shows that both groups rate their health status at baseline as already very good. Considering that the maximum score in the EQ-5D-5L index is 1, the average score of both groups of 0.97 leaves almost no room for improvement. However, a significant time effect (time p = 0.01) can be observed for the visual analogue scale, indicating a comparable improvement in both groups. The SF-36 questionnaire revealed significant changes over time without any relevant group or interaction effects. Significant time effects were observed for General Health (p = 0.01), Role Emotional (p = 0.01), Physical Dimension (p = 0.04), Mental Dimension (p = 0.03) and Total Score (p = 0.03), all indicating comparable improvements in both groups during the 7-day intervention. For the Physical Functioning and Physical Pain subscales, there is a significant group effect, which can be attributed to significant differences between the groups at baseline. For health-related quality of life, no indicators were found to support Hypothesis 3-The combination of hiking and coaching improves the quality of life more than hiking without coaching. Slight improvements in health-related quality of life can be observed in both groups, without any superior effect of coaching.
Further research is needed to evaluate the effects of health coaching on improving cardiorespiratory fitness, as our results are limited to a highly functioning sedentary population. Furthermore, the results need to be discussed in the context of the small sample size. As mentioned in the methodology, only two intervention groups were evaluated due to a high dropout rate and recruitment problems in the control group. However, since we have two randomly assigned intervention groups, the study design of a randomized, controlled trial remains. To keep the sample size as high as possible, missing values were reconstructed by the Last Observation Carried Forward Method (LOCF) and the Next Observation Carried Backward Method (NOCB), respectively. Hence, baseline values on day 0 and values on day 7 and day 80 were reconstructed, which might lead to biases in both the short-term and long-term effects due to the small sample size. Another bias within this study is the coaching itself-not only sympathy but also age and gender of the coaches could influence the impact of coaching and the effect of achieving personal goals [57]. However, the presented data shows the feasibility of such approaches. Furthermore, we performed a post hoc sample size simulation to provide a data-based sample size estimation for further studies. The sample size simulation was performed for aerobic capacity with both nonparametric (F1-LD-F1) and parametric models (ANOVA). A sample size of n ≥ 60 people should be reached for such study designs to obtain an estimated power of 1 − β ≥ 0.94.

Conclusions
Regular exercise in nature can help reduce the risk of cardiovascular diseases, which appear to be a common health problem with a sedentary lifestyle. Mountain hiking and mountain hiking in combination with health coaching are capable of inducing improvements in health-related quality of life and cardiorespiratory fitness. No superior effects of health coaching were found. In further studies, a sample size of n ≥ 60 must be achieved in order to gain an acceptable statistical power.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Acknowledgments:
We thank Penelope Hahne for the logistical support, Eva Traut-Mattausch for the project support, Sonja Zankl for the operative project management and Julian Wiedenhaus for the coordination of the coaching. We also want to thank all coaches and the participants.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

ANOVA
Analysis of variance ATS/ERS American thoracic society/ European respiratory society EQ-5D-5L EuroQol health survey with 5 dimensions and 5 levels FEV1 Forced Results from the post hoc sample size simulation for F1-LD-F1 models and ANOVA with group, time and group × time interaction effects.