Feasibility Study of Community-Based Training for Musculoskeletal Health Promotion

Background: To minimize fracture risk, multimodal training regimens are recommended. However, their effectiveness in community settings remains uncertain. This study evaluated the feasibility of 19-weeks of multimodal training in a local community center with emphasis on musculoskeletal health in postmenopausal women. Methods: In a controlled trial, 28 postmenopausal women (53-68-years-old) were assigned to a multimodal training group (MMT, n=15) or a control group (CON, n=13). The training consisted of high- and odd-impact, resistance and balance-coordination training 1-2 hours weekly. The outcomes were attendance rate, regional and total bone mineral density (BMD), bone mineral content (BMC), bone turnover markers (BTM), body composition, functional muscle strength and power, and dynamic balance. All were determined at baseline and after 19 weeks of training. BTM was assessed after three weeks. Results: Overall, 22(79%) participants (MMT, n=9; CON, n=13) completed the study, and the mean attendance rate for MMT was 65.5% of the maximum sessions (2) offered. Only right trochanter BMD increased (p<0.05) by 1.0±1.1% in MMT, which was higher(p<0.05) than CON. While whole-body BMC was not changed at 19 weeks from baseline in MMT, it decreased (p<0.05) in CON resulting in a significant difference (p<0.05) in whole-body BMC delta values between the two groups. Compared to baseline, body fat percentage(%BF), fat mass(FM), and visceral adipose tissue (VAT)-mass and -volume were decreased (p<0.01) in MMT, and were larger (p<0.05) than CON. No significant changes were observed in BTM, muscle strength and power, and dynamic balance after 19 weeks. Conclusions: Nineteen weeks of multimodal training 1-2 hours per week in a local community had a health-enhancing effect on %BF, FM, and VAT, whereas the musculoskeletal health impact was modest. We hypothesize that the reason might be too low training volume and frequency and supposedly too low musculoskeletal training intensity for some participants. Registration: ClinicalTrials.gov NCT05164679 (21/12/2021).


Introduction
Osteoporosis is a growing public health concern.The disease is characterized by impaired bone strength due to the reduction of bone mass and impairment of the micro-architecture of the bone, which increase the risk of bone fracture. 1It is estimated that 200 million females are affected worldwide and causes more than 8.9 million fractures per year. 2 Falls in osteoporotic patients are linked to high morbidity and mortality. 3Therefore, fall prevention strategies while improving the bone strength aimed at the high-risk population are needed. 4e way to prevent osteoporosis can be through engaging in evidence-based training to improve bone mineral density (BMD) and reduce age-related bone loss, 5 such as high-impact and resistance training.Furthermore, balance and coordination training can minimize the risk of falls and fractures, which is relevant to the high-risk population. 4Studies have shown that high-impact exercises, which produce high vertical ground reaction forces with a high force development rate, 6 are beneficial for bone structure and mineralization in children, 7 adults, 8 and the elderly. 9In a previous study of the acute osteogenic response to high-impact jumping in postmenopausal women, 10 stimulation of bone formation without any increase in bone resorption after jumping was reported.Moreover, the latter study showed a dose-response relationship between vertical as well as combined three-axes ground reaction forces (GRF) and the acute procollagen type I amino-terminal propeptide (P1NP)-response after countermovement jumps.
In addition, resistance exercise, which induces various muscle loads applied to the bone, 11 is reported to be a relevant osteogenic training modality for maintaining or increasing bone density and mass in older people. 12Besides, to prevent falls, balance and coordination exercises are crucial to be included as they improve static and dynamic balance. 13herefore, it is hypothesized in the present study that training (which combines evidence-based exercise) as health promotion will have an osteogenic effect on bone mass and bone strength that may counteract the normal age-related bone loss in postmenopausal women and improve the dynamic balance as well as muscle strength and power, which may prevent falls and reduce the fracture risk.
Although the association between exercise training, bone loss, and falls prevention in older adults appears to be established in intervention studies, 14 only a few studies have examined the effects of combined training on musculoskeletal health in postmenopausal women, 9,[15][16][17][18][19][20][21] and the results are varied.Therefore, there is still a need for safe, reliable, affordable, and evidence-based training programs that can be introduced, embraced, and sustained in high-risk populations, such as postmenopausal women.
The present study was performed in a local community setting.The aim was to investigate the feasibility and the healthpromoting musculoskeletal effects of 19wk multimodal training (high-and odd-impact, resistance, balance, and coordination training) for postmenopausal women.The primary outcomes were bone mass and BTM, while body composition, dynamic balance, muscle strength and power were the secondary outcomes.

REVISED Amendments from Version 2
In response to reviewer feedback and to improve the clarity and focus of our manuscript, several significant revisions were made from version 2 to version 3.
First, the title was changed from "A feasibility study of training in a local community aimed upon health promotion with special emphasis on musculoskeletal health effects" to "Feasibility of Community-Based Training for Musculoskeletal Health Promotion."This revision reflects a more focused and concise articulation of the research scope, eliminating redundant elements and emphasizing the specific method of intervention and the targeted health outcome.
Regarding the content, following reviewer suggestions, relevant tables (Table 2, 3) were revised to include additional statistical details, specifically F-statistics, p-values, and partial eta squared values for the interaction effects between group and time, enhancing the rigor of the analysis.
Additionally, exercise details were expanded, as recommended by the reviewers, to provide a more comprehensive understanding of the interventions used, which better supports the study's conclusions.
We also removed Figure 3 to avoid redundancy, as the results depicted were already thoroughly presented in Table 4.This change helps streamline the presentation of results and reduces unnecessary duplication of information.
These revisions aim to address the reviewers' comments, improve the overall clarity, focus, and scientific rigor of the manuscript, making it more accessible and informative for the readers.
Any further responses from the reviewers can be found at the end of the article

Study design
This is a feasibility study in a local community with a non-randomized controlled trial approach.The multimodal training was offered to postmenopausal healthy women in a local community.The outcomes for the training group (MMT) were compared with a sedentary control group (CON).The primary outcomes were musculoskeletal health, namely BMD, bone mass, and bone turnover markers (BTM), while the secondary outcomes were attendance rate, dropout, body composition including body weight and height, vertical jump height, and dynamic balance.Training was offered twice weekly for one hour, and the average attendance rate of MMT participants over the 19wk had to be >1 hour weekly.To collect baseline (BASE) data, all participants showed up in the laboratory on three occasions prior to the intervention (a description of the specific methods used is given in later sections).On the first visit, resting blood samples were collected to evaluate the concentration of BTM, body weight, height, body composition including BMD and bone mass were measured.On the second visit, a dynamic balance test (four square step test) and a functional muscle strength and power test (jump-and-reach test) were performed.On the third visit, the aerobic capacity was evaluated by assessment of VO 2 -max.After three weeks (3wk) of training, resting blood samples were additionally collected.After 19wk of training, all assessments were repeated and compared with baseline and MMT and CON were compared.
Blood sample collection, DXA-scanning, and VO 2 -max tests were carried out at the Department of Nutrition, Exercise and Sports, University of Copenhagen, Denmark.The testing of vertical jump height, as well as dynamic balance, were performed at the health promotion training initiative in the Copenhagen area, Denmark.The blood samples were analyzed at the Department of Clinical Biochemistry, Rigshospitalet Glostrup Hospital, Denmark.
The trial was registered retrospectively at ClinicalTrials.gov and published on December 21 st , 2021 (NCT05164679).As registration at ClinicalTrials.gov has until recently not been a part of the general research policy in our team, the trial was registered after completion of the study.

Participants
Healthy, sedentary postmenopausal women aged below 70 years were eligible to participate in the present study.Inclusion criteria were non-smoker and body mass index (BMI) <30 kg/m 2 .Exclusion criteria were: T-score < -3 SD in the lumbar spine or hip; Z-score > 1.5 SD (high age-related BMD); use of hormone therapy, medical treatment, or supplements that affect bone metabolism; previous or current medical condition affecting bone health; engagement in regular and systematic weight-bearing training or strength training during the preceding two years (to distinguish active from sedentary).The engagement in regular, systematic training here means engage minimum at least one time a week and has a goal of the training.
Initially, twenty women were recruited to the training via advertisements online and an in a local newspaper, of which 19 showed up for pre-testing (Figure 1).After a medical examination, one participant was excluded due to low BMD (T-score < -3 SD), and two were excluded due to high BMI (≥30 kg/m 2 ).In addition, one woman refrained from participating in the training, and thus, 15 participants were recruited to MMT.CON consisted of 13 age-matched sedentary postmenopausal women (12 from a previous study (not a long time before the present study) plus the woman who refrained from participating in the training) which recruited with the same procedure, via advertisement online and a local newspaper.Participants' baseline characteristics, including maximum oxygen consumption (VO 2 -max), are shown in Table 1.
Every participant was fully informed in writing and verbally before giving her written consent to the procedures and potential discomfort associated with the study.The study was conducted in accordance with the Declaration of Helsinki and approved by the local ethics committee of the Capital Region of Denmark, H-18044190.

Training program
The present feasibility study evaluated the osteogenic impact of a training concept already offered by a health promotion initiative, "Knoglestaerk" ("Bone strong"), in a local community.
According to the initiative, the training was offered as "evidence-based bone training" aimed at enhancing musculoskeletal health.It was carried out 2 Â 60 minutes weekly and included three main components: 1) High-and odd-impact exercise, including multi-directional games; 2) Progressive resistance training (PRT); and 3) Balance and coordination training.
The participants engaged in recurrent gymnastic activities and small games designed to impose osteogenic and varied strain on bones, primarily targeting the legs and arms.The training included various jumping exercises, such as countermovement jumps, jumps from the floor up onto a bench, and from a bench down to the floor.Other exercises involved quick walking up and down on gymnastics equipment and sprinting over short distances in different directions, often integrated into small games to promote interaction between participants.
To ensure a high strain rate and strain magnitude necessary for stimulating bone formation, the training incorporated exercises like unilateral jumps from side to side, random unilateral jumps, and jumps in a square.The intensity of these exercises varied from low (jogging and brisk walking) to medium (catch games, dodgeball, jumps, and leaps) and high intensity (box jumps with drop landing, plyometric jumps, and sprints with a sudden stop).Each training session included at least 40 medium-to-high-intensity impacts to maximize the osteogenic stimulus.
In addition to these impact exercises, the PRT component consisted primarily of exercises using participants' own body weight, supplemented with medicine balls, resistance bands, and sandbags weighing 5-15 kg.The resistance training started with 15-20RM and gradually progressed to 6-12RM, with 2-3 sets per exercise.Exercises included back extensions, air squats, wall balls, rows with resistance bands, lunges, jumping lunges, and sprints with partner resistance.These exercises were performed intermittently and varied to prevent the desensitization of the skeleton that is often observed with repetitive stimuli, thereby enhancing the osteogenic response.
The overall training approach was designed to provide a comprehensive osteogenic stimulus through a combination of high-impact, odd-impact, and resistance exercises, all delivered in an engaging, community-based format.

Attendance and dropout
In the present study, the feasibility was measured by the attendance rate.The attendance rate range, first and last four weeks reported.

BMD and body composition
To test the BMD exclusion criteria at baseline as well as evaluate bone adaptation in proximal femur (PF), lumbar spine (LS), and whole-body (WB) after 19wk of training, BMD (g/cm 2 ) were assessed by Dual-energy X-ray Absorptiometry scanning (iDXA, Lunar Corporation, Madison, Wisconsin, USA) according to standard procedures.The regions of interests were determined by the encore software (encore software version 14.10.022,GE Medical Systems, Madison, United States).In addition, body composition was evaluated by the whole-body scan: body weight (BW, kg), BMI (kg/m 2 ), body fat percentage (%BF, %), total fat mass (FM, g), total lean body mass (LBM, g), visceral adipose tissue (VAT) mass (g) and volume (m 3 ), and total bone mineral content (BMC, g).The participants were asked to remove metal objects and empty their bladder prior to scanning.

Blood sampling & Biochemical analyses
The plasma concentration of BTM at BASE, after 3wk, and after 19wk of training were measured.After an overnight fast and without any vigorous activities in the previous 48 hours, participants showed up in the laboratory in the early morning.Blood samples were collected from the antecubital vein with a butterfly needle, then transferred to EDTA tubes and centrifuged immediately.The plasma fractions were put on dry ice.Eighteen ml of blood were taken from each participant per test day.Following each test, a sample was placed at -80°C for future P1NP, OC, and CTX analysis, which were assessed by the Chemiluminescence method using a fully automated immunoassay system (iSYS, Immunodiagnostic Systems Ltd., Bolton, England).The assay performance expressed as inter-run variation coefficients were 8% for P1NP, 9% for OC, and 10% for CTX.

Training status
To estimate the participants' general training status, a progressive test of maximal oxygen uptake (VO 2 -max) (ml/kg/min) was performed on an electronic bicycle ergometer (Monark 839E, Monark Exercise AB, Vansbro, Sweden) according to standard procedures.A breath-by-breath gas online analyzing system (Jaeger Oxycon Pro, VIASYS Healthcare, Höchberg, Germany) was connected to the participant, and a direct VO 2 -max measurement was conducted.

Dynamic balance
The "four square step test" (FSST) was performed to evaluate the dynamic balance 22 in MMT.The test requires a person to move systematically forward, sideward, backward, and sideward again over four narrow sticks circa 2.5 cm in diameter and 100 cm in length placed on the floor like a cross.The participants were requested to complete the sequence as fast as possible without touching the sticks.Facing the cross, the participants started in the lower left quadrant by stepping forward over the stick into the next left quadrant, stepped sideward over the stick again into the upper right quadrant, and stepped backward over the stick into the lower right quadrant, then stepped sideward over the stick into the left quadrant where they started.Thus, they were moving in a clockwise direction.The floor in each square must be contacted by both feet, and the participants were asked to face forward for the whole sequence.However, the participants were allowed to turn to step into the next square if needed.They started and finished in the same square, and the score was given in seconds (sec).The test was done twice, and the fastest time was used as the FSST score.

Functional muscle strength and power
The functional muscle strength and power were evaluated in MMT by a jump-and-reach test (vertical jump height) (Vertec Sports Imports, Hilliard, OH).First, the participants were taught how to perform a countermovement jump and how to displace the vanes that was placed above reaching height on a vertical stand.After thoroughly instruction and familiarization, the jump-and-reach test was performed.The maximum vertical jump height (cm) was determined by the difference between the participant standing reach height and the highest displaced vane.The greatest value out of three trials was taken as the result.between group and time were evaluated using F-statistics, and effect sizes were calculated using eta squared (η 2 ) to provide a more nuanced understanding of the intervention's impact.A p-value < 0.05 was considered significant.Unless otherwise stated, values are given as mean AE standard error (SE).

Participant characteristics
During the study, two participants in the MMT group dropped out due to personal reasons, and four participants did not meet the mean attendance criterion of at least one session weekly (Figure 1).Therefore, 9 participants from the MMT group and 13 from the CON group were included in the final analysis.At baseline, there were significant differences between MMT and CON for BW, BMI, FM, left and mean femur BMD values, OC, and CTX (Table 1).

Attendance and dropout
As described in the participants section, two persons in MMT dropped out due to personal reasons, and four participants did not meet the mean attendance criterion of at least 1 session weekly (>19 sessions) and were subsequently excluded in the data analyses.The mean attendance rate for the remaining MMT participants (9 women) was 24.2 sessions (65.5%) of the maximum number of sessions offered (37 sessions), with individual attendance rates ranging from 20 to 30 sessions (54.1-81.1%).Attendance decreased over time, from 78% (6.2 out of 8 sessions) in the first four-week period to 64% (5.1 out of 8 sessions) in the last four-week period.
The attendance rate range (54.1-81.1%) in the present study showed that there is inter-individual variability.The participants attended the training more in the first than the last four-week.

BMD and body composition
After 19wk of training, right trochanter BMD (g/cm 2 ) increased significantly from baseline for MMT (1.0%, p=0.03), which was significantly different from CON (p=0.03)(Table 2), but there was no significant BMD change in any other region.Whole-body BMC decreased significantly (0.7%, p=0.016) within CON (Figure 2).

BTM
The plasma concentrations of P1NP, OC, and CTX are shown in Table 4.There was no significant within-or betweengroup difference observed at 3wk or 19wk compared to baseline for all markers.

FSST & Jump-and-reach test
There were no significant changes in both FSST and jump-and-reach test at 19-week compared to baseline for MMT (Table 5).However, there was a trend towards faster FSST times at 19-week (p=0.06).

Discussion
The present study aimed at evaluating the feasibility and whether 19wk of multimodal training offered as musculoskeletal health promotion in a local community had an effect on musculoskeletal health, body composition, dynamic balance, and functional muscle strength in postmenopausal women.The main finding from this study was that the multimodal training program effectively improved body composition, particularly by significantly reducing body fat percentage, total fat mass, and visceral adipose tissue (VAT) in postmenopausal women.Additionally, the training program led to a significant increase in BMD at the right trochanter, indicating localized benefits for bone health.
This study demonstrated that the right trochanter BMD significantly increased by 1.0% in the MMT group, supported by an F-statistic of 5.814 and an eta squared value of 0.225, indicating a medium effect size for the interaction between the training and time on BMD.This suggests that the targeted exercises were particularly effective in enhancing bone density in this specific region.The lack of significant changes in other BMD regions may be attributed to insufficient training frequency and intensity, as the average attendance did not meet the recommended guidelines for osteogenic training.
According to Daly and Giangregorio, 23 the minimum frequency of resistance training is at least two days per week and for weight-bearing impact exercise 4-7 times per week, which the present study not met as no participant fulfilled 37 times in total for 19wk of training.
The whole-body BMC was not changed in MMT, whereas it was significantly decreased (p=0.02) by 0.7% in CON, which was significantly (p=0.04)different from MMT (Table 3).Thus, it seems from the present study that the training at "Knoglestaerk" maintained total bone mass, even with a required minimum training attendance of just one hour weekly.The loss of whole-body BMC (-0.7%, p=0.016) in CON group is consistent with Gallagher et al., 24 who investigated the effect of age and menopause status on bone.The finding that the change in BMC differed between MMT and CON is not in line with Mosti et al., 16 who investigated the impact of three sessions of squat exercise maximal strength training in postmenopausal women with osteoporosis or osteopenia (mean age: 61.9 AE 5.0 y) for 12 weeks.Moreover, they found an increase in lumbar spine BMC not whole-body.Thus, they found an increase of 2.9 AE 2.8% (p=0.01)LS BMC from baseline in training group, and this change was higher (p=0.03)than in control group.
The significant reductions in %BF and FM, as shown in Table 3, are supported by F-statistics of 4.410 (p=0.049,η 2 =0.181) and 6.232 (p=0.021,η 2 =0.238) respectively, indicating moderate to large effects.These findings align with previous research, such as Daly et al., 25 who investigated the effect of one year supervised and structured multicomponent training program in community-dwelling independently living men (27%) and women (73%) (mean age: 67.4 y, 60-86 years).They found that the FM was significantly decreased (p<0.001) by 3.2%, which is lower than the present study (6.0%, p=0.01).The higher reduction in the present study compared to Daly et al. 25 study might be due to the younger participant (62.4 AE 3.8 y) which has a higher metabolic rate than the older one (66 -94 y) 26 , that could influence the change of body fat.The reduction in %BF and FM is in line with Nielsen et al. 27 as well, who investigated the feasibility and physiological health effects of 15-week training performed minimum once a week for sedentary elderly.They found that the total FM was significantly decreased (-2.0 kg, p=0.01) and the %BF as well (-1.6%,p=0.01).
As the VAT assessment by iDXA is a fairly new functionality, only very few studies report results for this outcome, and it seems that none has studied training effects on VAT for healthy postmenopausal women.Moreover, no normal reference intervals for specific populations have been established yet.The present finding of a 11.3% reduction in VAT seems in line with the Yen et al. study, 28 which reported a 6.3% decrease in VAT after eight weeks of aerobic and resistance exercise intervention.However, the study participants were a group of male and female patients receiving chemotherapy for head and neck cancer (mean age: 52 AE 10.7 y).As reported by Yen et al. 28 the lower response than the present study (-6.3% vs. -11.3%)may be due to the disease or the shorter duration (8wk vs. 19wk), even though their intervention more frequent (3 days/week vs. one day/week). 29Moreover, it could be hypothesized that the participants in Yen et al. 28 study had a lower %BF and VAT at baseline, therefore, it is more difficult to induce a training response.Nevertheless, more studies still needed on healthy postmenopausal women.
Despite the favorable result, as evidenced by the significant F-statistics and effect sizes reported, the present study has several limitations.The differences in baseline value and the selection of the participants were not randomized, these could lead to bias.The duration was short (<6 months) according to the guide to the optimal prescription, preferably 12 to 24 months. 30The dietary intakes were not recorded.Thus, the changes that occur in the body composition could not be known whether entirely due to the training or dietary intake of participants.Future studies should consider taking the dietary intake history and monitoring habitual daily physical activity, which may add an explanation regarding the changes specifically in body composition in the present study.Another thing to note is that the Knoglestaerk health promotion program is a real-life setting and thus the training is not as well controlled as in a laboratory setting.Since neither heart rate (HR) nor accelerometer measurements were recorded during exercise, it cannot be ruled out that the individual training intensity varied among the participants.Hence, the use of HR monitor and accelerometer would improve the control and the standardization of the training and thus reduce the variation in the individual response and the reliability of the results.Recent studies 18,20 have shown that exercise intensity is positively associated with bone changes, and that low-intensity exercise is ineffective.In addition, we did not carry out a priori power analyses.The power calculations were performed after the studies had been conducted (provided as Extended data 31 ), and the statistical power of the present study seems to be underpowered.Therefore, a larger sample size would have a positive effect on the effect sizes due to a reduction in standard error and thus an improvement of the statistical power.The effect size of the intervention was varied, with the greatest effect on VAT-volume & mass, total BMC, weight, total fat mass and body fat percentage.On the contrary, the intervention has smaller effect on general and specified BMDs.The given amount of time (19wk) and specified intervention (multimodal training group) had greater effect on fat changes rather than bone changes.The interesting part was the total BMC that affected by the intervention was not followed by BMD's changing.

Conclusions
The results of the present study indicate that while 19 weeks of multimodal musculoskeletal training as offered by the Knoglestaerk health promotion training initiative is feasible for postmenopausal women by inducing health-enhancing effects on body composition, however, due to attendance compliance, the training did only elicit minor significant improvements in BMD, thus an increase was found in right trochanter BMD and whole-body BMC.No changes in dynamic balance or functional muscle strength were observed in the present study.To promote musculoskeletal improvements for all participants, better control of the individual musculoskeletal training intensity, frequency, and attendance rate is highly recommended.I have reservations about the statistical methods used or rather about presentation and interpretation of their outcomes.For an experiment, the most important result is the interaction of a fixed factor (Group) and a repeated factor (please report F, p and eta squared).If this interaction occurs, one can talk about a different response in the groups.This should not necessarily focus on the results of post-hoc tests.However, the authors focused on within-group and between-group differences which is not as important as interaction.

Extended data
As I had some time I recalculated the results again and attached them in an excel file please refer to this link: https://f1000research.s3.amazonaws.com/linked/651424.A_Feasibility_Study_of_Training_Dataset_ver_1_1_.xlsxWhen the variables involved left and right, I expanded the analysis to include a second factor repeated SIDE.I didn't check the normality of the distributions, but it should probably be done.It would be worthwhile to devote some space to the effect sizes (eta squared).In some cases, the interactions were close to significance with relatively large effect sizes.Unfortunately, due to low statistical power (group sizes), such results have low reliability but at least there is something to discuss.I would also ask the authors to expand on the description of the exercises used.Based on the description, it would not be possible to reproduce the intervention exactly.

Specific comments
In Table 2, the p-value evaluates the group effect, which is not valid in this case.I would rather put F, p and eta2 for the interaction in the last column.Asterisks could be used to indicate any differences between groups in individual measurements (different than control).
The same goes for Table 3

Summing up
Of course, the article has irremovable flaws related to the allocation into groups and their size, but if the practical aspect of the results obtained is taken into account, the value of the study increases.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound?Yes

Are sufficient details of methods and analysis provided to allow replication by others? Yes
If applicable, is the statistical analysis and its interpretation appropriate?Yes Are all the source data underlying the results available to ensure full reproducibility?Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Orthopedics, sports medicine I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
Statistical power should be calculated.

3.
The exercise program is not strong enough to have physiological impact to the health.

4.
The duration of exercises is not long enough.

5.
The study is required to do prospective study.The assignment to exercise and control group should be randomized.

Is the work clearly and accurately presented and does it cite the current literature? No
Is the study design appropriate and is the work technically sound?changes in this short period since, in our previous study, there were acute responses in BTM.Thank you for your suggestions.
Is the work clearly and accurately presented and does it cite the current literature?

○
We have cited the current literature Is the study design appropriate and is the work technically sound?

○
We have added, "This is a feasibility study in a local community with a non-randomized controlled trial approach."Are sufficient details of methods and analysis provided to allow replication by others?

○
We think it is sufficient since the study method and the training program were described clearly.Are all the source data underlying the results available to ensure full reproducibility?

○
We share the raw data in the 'Extended data' section.

Competing Interests:
The authors have no professional relationships with companies or manufacturers that might benefit from this study.The results do not constitute an endorsement of any product.

Introduction
Paragraph (P) 1: "is it estimated that 200 million females…".This number actually includes males and females.P2: "One way to prevent… age-related bone loss".This is a very old reference.Since 1987 a large number of exercise trials for bone have been conducted.I suggest including a more recent reference.
P3 & 4: In para 3 you hypothesize that "evidence-based multimodal training will have an osteogenic effect and in para 4 you state that there is a need for "evidence-based trainings".This is contradicting and I don't think you can call this MMT "evidence based" P4: "Only few studies have examined…".There has been a large number of studies that have tested combined impact, resistance and balance training interventions in recent years which showed safety and efficacy. 1,2,3,4These should be cited here plus the existing gap in the literature identified.P5: "The aim was to investigate the feasibility…" Feasibility is mentioned in the title, the aims and discussion (e.g.discussion 1 st sentence and conclusion 1 st sentence), however, it is unclear how feasibility was measured.Feasibility is also not listed as an endpoint (e.g.clinicaltrials.gov).Please clarify.

Methods
General: please clarify which outcomes were primary and which ones were secondary.P1: Please explain how participants were recruited, particularly the sedentary control group.P1: The duration of 19 weeks is inappropriate to measure changes in bone mineral density.A minimum of 6-8 months is needed to detect changes in bone.P3: "to ensure transparency" The trial was registered close to manuscript submission data, which does not ensure transparency.I suggest deleting this part of the sentence.P4: How was sedentary defined?Was there a minimum age set?
○ How was postmenopausal defined?(e.g.how many years post menopause?)

○
How was regular, systematic training defined?
○ P5: The information on the number of participants and drop outs should be moved to the results section.P5: "…12 from a previous study...".Please describe the control group somewhere.P8: How many sets and repetitions were performed of each exercise, and how did you ensure the resistance training was progressive?P9: Can you include information on the short-term measurement variability of BMD and body composition measures in your lab? P12/13: Why were FSST and jump-and-reach tests not performed in the control group?P14: Given the large number of outcomes, adjustment for multiple comparison is strongly recommended.

Results
P1: There are significant baseline differences between groups which is of great concern.How did you address this, were analyses adjusted for baseline values?P4: "… LS and total femur… tended to increase".Reporting of trends is controversial; I suggest removing this sentence.

Discussion
Rather than comparing present findings to individual studies, I would recommend reviewing the large body of systematic reviews and meta-analyses on the topic.Many of those have examined the influence of training parameters (e.g.intensity, volume and frequency. 3,5) Interpretation of your findings in context of those findings would be helpful.For example, recent studies have shown that exercise intensity is positively associated with bone changes, and that low-intensity exercise is ineffective.
There are a number of limitations of the present study that have not been addressed in the discussion.E.g. bias related to baseline differences, not randomised, only short term (< 6 months), generalisability.
You state that no power calculation was undertaken. 1) How was the sample size determined?2) How was a posteriori power calculated and what were the results?
I would like to see a section about clinical significance of the findings.

Conclusion
"… inducing health-enhancing effects".In light of the significant baseline differences, other study limitations and the fact that only body composition variables (which were significantly different between groups at BL) improved, I think this is overstating the findings of the study.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com Figure 1.CONSORT flow diagram.

Table 1 .
Baseline characteristics of the participants.
Values are given as means AE SD.BMI = body mass index, BMD = bone mineral density, BTM = bone turnover marker.*Significant difference between groups (p < 0.05).Unpaired T-test.

Table 3 .
Body composition values (Mean MMT = Multimodal training group, CON = Control group, BMI = Body Mass Index.a Geometric mean.b General Linear Model + post hoc test.*Significant difference between groups (p < 0.05).# Significant difference within-group (p < 0.05).

Table 5 .
Dynamic balance and functional muscle strength result for multimodal training group (n=9).Paired t-test.
designed as a controlled trial.In the end, only 9 subjects in the experimental group and 13 in the control group completed participation in exercise and measurement.I like the idea itself very much.Evaluation of the effectiveness of the exercise program in the real conditions of the local community shows the practical usefulness of similar initiatives.The feasibility of the established program turned out to be moderately satisfactory.The effects of the program have not been fully proven.This had to do primarily with the relatively small size of the groups.I checked with G*Power that the minimum sizes to detect a moderate effect (eta squared>0.06)at80%statistical power are at least 17 people in each group.When planning the experiment, a correspondingly larger number of subjects should have been taken into account (e.g.22 and 22).I understand that for various reasons it was not possible.It was unfortunate that the groups differed significantly in the base line parameters (especially in age, bone turnover markers.Of course it cannot be changed. Figshare: Working title: A Feasibility Study of Training in a Local Community Aimed Upon Health Promotion with Special Emphasis on Musculoskeletal Health Effects.Extended data -Statistical power of the paired t-tests for body composition, regional BMD, and bone turnover markers.https://doi.org/10.6084/m9.figshare.16611655.v1. 31Figshare: Working title: A Feasibility Study of Training in a Local Community Aimed Upon Health Promotion with Special Emphasis on Musculoskeletal Health Effects.Extended data -Effect size of the Training Program.https://doi.org/10.6084/m9.figshare.16611787.v1. 33Reporting guidelines Figshare: CONSORT checklist, extension for Pilot and Feasibility Trials for study "A Feasibility Study of Training in a Local Community Aimed Upon Health Promotion with Special Emphasis on Musculoskeletal Health Effects".https:// doi.org/10.6084/m9.figshare.16610341.v1. 34Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).
. Why does the designation 'a-geometric mean' at body weight appear (what's the reason)?Body composition parameters are shown twice in Table 1 and Table 3. Aren't the results shown in Figure 3 duplicated in Table 4? Let's stick to the rule that each result is presented once in one of three ways: in text, in a table or on a graph.

Table 2 :
Please explain why you are reporting mean femur BMD in addition to right and left femur values.Table 2/3: Your findings show a decrease in whole body BMC and an increase in BMD.How would you explain these differences.