Influence of physical activity on serum vitamin D levels in people with multiple sclerosis.

In most cases, multiple sclerosis (MS) patients reduce physical activity with disease progression and many patients are found to be vitamin D deficient. The aim of this study was to explore correlations between daily physical activity in everyday life and 25-hydroxyvitamin-D3 (25(OH)D3) serum levels in mildly disabled patients with an Expanded Disability Status Scale (EDSS) ≤ 4. We analyzed serum 25(OH)D3 levels and recorded daily physical activity (activity duration, number of steps, distance, energy expenditure) using an activity tracker for 14-days in 25 women and 15 men. Participants recorded their daily sunlight exposure time by diary during the study period. We found a positive correlation between physical activity and 25(OH)D3 levels in both, Pearson correlation (r = 0.221) and multivariate regression analysis (β = 0.236), which was stronger than correlation with sunlight exposure time (β = -0.081). EDSS and physical activity were weakly correlated (r = -0.228), but no correlation between EDSS and 25(OH)D3 levels was found (r = -0.077). There were no relevant differences in physical activity (p = 0.803) and 25(OH)D3 concentrations (p = 0.385) between the EDSS groups 0 - 1.5 and 2.0 - 4.0. In conclusion, physical activity has an effect on vitamin D levels independent of sunlight exposure time in people with MS (pwMS) with low-grade disability.

Introduction

Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system whose cause is unknown. Apart from the genetic background a large number of environmental factors have been investigated as potential risk factors for MS or disease activity in people with MS (pwMS) [1]. One of these much-discussed factors is decreased vitamin D levels which are associated with a higher risk of MS diagnosis and disease activity [2]. An inherent challenge with association and correlation is the fact that it does not prove causality and might be explained by confounders [3]. Causality can be shown by testing exposure versus non-exposure, ideally in a prospective randomized setting, which has been done for vitamin D supplementation in pwMS in several trials, none of which could demonstrate an effect on clinical outcomes [4]. Therefore, alternative reasons for the association between MS disease activity and vitamin D levels need to be considered. As vitamin D level is driven by exposure to ultraviolet (UV) B radiation to a large extent [5] and the main source of UV-B is sunlight, one might speculate that low vitamin D levels in pwMS are the consequence of less outdoor activity due to disability.

The aim of this study was to investigate the correlation between objectively measured physical activity in everyday life as well as sunlight exposure time on 25-hydroxyvitamin-D3 (25(OH)D3) serum levels in mildly disabled pwMS in a short prospective observational setting.

Methods

For this prospective, observational study, we recruited forty pwMS at the Department of Neurology in Innsbruck. The inclusion criteria were as follows: 1) diagnosed with MS of any subtype using the diagnostic criteria that were valid at the time of diagnosis [6] with an EDSS score of ≤ 4, defined as being fully ambulatory without the need for any walking aid [7]. 2) no vitamin D supplementation for 38 days immediately before and during the study period (due to the half-life period of approximately two to three weeks [5], it can be expected that after 38 days previous vitamin D supplementation has no effect on 25(OH)D3 serum levels. 3) no vacation during the study period due to a bias in measured physical activity and time spent outdoors.

To ensure comparability of measured 25(OH)D3 serum levels and sunlight exposure time, the whole study was conducted in the time between 15th August and 15th September 2017 beginning the day after blood sampling.

Using an activity tracker (AS80, Beurer Germany—BAT) on the non-dominant wrist, study participants measured their daily step-count, active time and energy expenditure (in calories) in everyday life during a period of 14 days. Steps were registered once thirty steps had been taken in a row with a maximum inactivity time of two seconds in between. Physical activity was defined by number of steps and activity time. BAT was set up individually and anonymized. Sex, age, height and current weight were registered in each device. To determine the average step length, participants had to perform a 7.62 meters’ walk twice. To eliminate bias, the display of the BAT was covered up and participants were instructed not to retrieve data during the study period. BAT stored the measured values in its random-access-memory for 30 days and additionally on the manufacturer’s server in Germany, fulfilling current data safety standards. The data transfer was performed after the two-week study period by using Bluetooth LE® on an Apple® iPhone with operation system iOS 10.3.1.

Further, participants had to keep a log of the time spent outdoors with sunlight exposure each day during the study period and were instructed to avoid extra efforts and unusual activities.

Levels of 25(OH)D3 in serum were measured at the central laboratory of the university hospital Innsbruck using high-performance liquid chromatography (ChromSystems) and serum levels were classified according to the guidelines of the Endocrine Society, where vitamin D deficiency is defined as levels below 50 nmol/l, insufficiency as levels between 50.1–74.9 nmol/l, sufficiency as levels between 75.0–99.9 nmol/l and ideal as levels ≥ 100.0 nmol/l [2].

Statistical analyses

Statistical analyses were performed with IBM SPSS Statistics 20 (IBM Corp., Armonk, NY, USA). All data were anonymized, and the analyses of outcomes were based on the per-protocol principle. Normality of data was assessed by using the Shapiro-Wilk normality test. All normally distributed variables are shown as mean and standard deviation (SD), while non-parametric variables are presented as median, inter-quartile range (IQR) and minimum to maximum ranges. Correlations between continuous variables were assessed by Pearson’s and Spearman’s correlation test as appropriate. To interpret correlation coefficients Cohen’s standard classification ranges were used (weak 0.10–0.29, moderate 0.30–0.49, strong ≥ 0.50).

A multivariate linear regression analysis was performed to analyze the influence of physical activity and daily sun exposure time (both independent variables) on the 25(OH)D3 serum levels (dependent variable). Due to the small study cohort size, no further independent variables were included in the model to avoid overfitting. Cohen’s f2 was used as a measure of local effect size, classifying f2 ≥ 0.02 as small, f2 ≥ 0.15 as medium and f2 ≥ 0.35 as large effect sizes.

The 25(OH)D3 status and daily activity was compared between two EDSS groups (EDSS 0–1.5 vs EDSS 2.0–4.0), a two-sided t-test. The reasoning was that no impairment can be expected with EDSS scores of ≤ 1.5, whereas EDSS scores of ≥ 2.0 usually reflect physical impairment.

Non-normally distributed variables were compared between two groups with a Mann-Whitney U-test.

Due to the explorative character we did not set a level of statistical significance and consequently, no formal power calculation was performed [8,9].

Ethics

The study was approved by the Ethics Committee of the Medical University of Innsbruck (study number 1088/2017) and all participants signed a written informed consent.

Results

Clinical characteristics of the participants

We included 38 patients (23 women, 15 men) and two participants were excluded. One female participant was lost to follow-up and one female participant was classified as an outlier with respect to her 25(OH)D3 serum level (possible laboratory error and the analysis could not be repeated). All participants had Caucasian skin type. The mean age of all 38 study participants was 39.9 years (± 9.3) and the median EDSS was 2.0. All patients had relapsing remitting disease with a mean annualized relapse rate of 0.46 including self-reported attacks (Table 1). Vitamin D3 serum levels were higher in women (93.7 ± 32.1 nmol/l) compared to men (73.9 ± 19.4 nmol/l, p = 0.045). However, total energy expenditure was lower in women (women 1,680.2 ± 211.1 kcal/day vs. men 2,125.6 ± 123.9 kcal/day, p < 0.001). Sun exposure time did slightly differ between sexes with a median time of 70.4 min/day in women (range 32.5–150.0 min/day) vs. 106.9 min/day in men (range 45.0–276.4 min/day; p = 0.110). Ideal vitamin D levels were detected in 13/38 (34.2%) participants, sufficient levels in 9/38 (23.7%), insufficient levels in 14/38 (36.8%) and deficient levels in 2/38 (5.3%). Only 2 participants had to stop vitamin D supplementation before study entry whereas 38 participants never used such medication.

Table 1

Clinical characteristics of the study participants.

	total	female	male
sex
n (%)	38 (100.0)	23 (60.5)	15 (39.5)
age (years)
M ± SD (range)	39.9 ± 9.3 (23.0 - 63.0)	42.0 ± 9.7 (25.0 - 63.0)	36.7 ± 7.9 (23.0 - 49.0)
clinical phenotype
n (%)	RRMS: 38 (100.0)
EDSS
median (IQR, range)	2.0 (1.5, 0 - 4)	2.0 (1.5, 0 - 4)	1.5 (1.0, 0 - 3)
disease duration (months)
M ± SD (range)	131.4 ± 96.3 (9 - 352)	175.6 ± 92.9 (9 - 352)	63.5 ± 52.8 (11 - 168)
annualized relapse rate*
M ± SD (range)	0.46 ± 0.48 (0.00–2.67)	0.41 ± 0.57 (0.00–2.67)	0.53 ± 0.30 (0.00–1.09)
25(OH)D3 (nmol/l)
M ± SD (range)	86.1 ± 28.9 (44.0 - 173.0)	93.7 ± 32.1 (44.0 - 173.0)	73.9 ± 19.4 (52.0 - 104.0)
activity/day (min)
M ± SD (range)	57.4 ± 22.3 (16.0 - 120.0)	57.0 ± 23.4 (16.0 - 120.0)	59.1 ± 21.6 (26.0 - 113.0)
steps/day
M ± SD (range)	6,863.3 ± 2,592.0 (2,082.9 - 14,656.0)	6,862.9 ± 2,785.9 (2,082.9 - 14,656.0)	7,003.8 ± 2,381.5 (3,293.4 - 13,060.7)
distance/day (km)
median (IQR, range)	4.2 (3.0, 1.3 - 9.9)	4.0 (2.9, 1.3 - 9.7)	5.5 (2.9, 2.6 - 9.9)
total calorie consumption/day (kcal)
M ± SD (range)	1,852.1 ± 281.0 (1,292.1 - 2,373.8)	1,680.2 ± 211.1 (1,292.1 - 2,083.9)	2,125.6 ± 123.9 (1,944.0 - 2,373.8)
sun exposure time/day (min)
median (IQR, range)	72.9 (71.8, 32.5 - 276.4)	70.4 (60.3, 32.5 - 150.0)	106.9 (65.1, 45.0 - 276.4)
fish consumption/week (g)
median (IQR, range)	150.0 (150.0, 0 - 700.0)	75.0 (150.0, 0 - 400.0)	200 (106.0, 0 - 700.0)
BMI (kg/m2)
M ± SD (range)	23.8 ± 4.2 (16.5 - 34.5)	23.5 ± 4.8 (16.5 - 34.5)	24.3 ± 3.3 (18.4 - 31.1)

Abbreviations: IQR, interquartile range; M ± SD, mean ± standard deviation; n, number; EDSS, expanded disability status scale; 25(OH)D3, 25-hydroxyvitamin-D3; BMI, Body mass index

* up to 5 years before study entry depending on disease duration, includes self-reported relapses.

Correlation between EDSS and physical activity

There was a weak negative correlation between EDSS and activity (r = -0.228) and no correlation was found between EDSS and 25(OH)D3 (r = -0.077). EDSS groups 0–1.5 (n = 17) and 2.0–4.0 (n = 21) did not differ regarding 25(OH)D3 levels (90.7 ± 36.1 nmol/l vs. 82.4 ± 21.8 nmol/l, p = 0.385) nor for daily activity time (58.4 ± 13.6 min vs. 56.6 ± 27.8 min, p = 0.803).

Correlation between physical activity and 25(OH)D3 levels

A weak correlation between activity and 25(OH)D3 levels was found (r = 0.221; Fig 1 panel A). Further, a weak negative correlation was found between body mass index (BMI) and 25(OH)D3 (r = -0.282) and a moderate negative correlation between body weight and 25(OH)D3 (r = -0.403).

Fig 1

Scatter plots and linear regression lines for vitamin D concentrations versus activity (r = 0.221; panel A) and sun exposure time (r = -0.002; panel B).

Sun exposure time and 25(OH)D3 levels were not correlated (r = -0.002), possibly due to a ceiling effect as discussed below (Fig 1 panel B). However, many other factors might play a role such as skin type, genetic make-up, clothing, use of sun blockers, age, and altitude. Performed steps and daily activity were strongly correlated (r = 0.989), and there was a moderate correlation between walked distance and total energy expenditure (r = 0.446).

Regression analyses

The influence of physical activity and daily sun exposure time on 25(OH)D3 serum levels was analyzed with a multivariate linear regression model: Combined, both variables (physical activity and sun exposure time) had a weak influence on 25(OH)D3 levels (R2 = 0.055, p = 0.383). Furthermore, local effect size was classified as being small, using Cohen’s f2 (f2 = 0.058).

Importantly, the standardized coefficient (β) indicated that physical activity has a stronger influence on 25(OH)D3 levels than daily sun exposure time (β = 0.236, p = 0.174 vs. β = -0.081, p = 0.638, Table 2). No interference between physical activity and sun exposure time in multivariate linear regression analyses could be detected.

Table 2

Multivariate and univariate linear regression analyses.

Multivariate linear regression (influence of sun exposure and physical activity on 25(OH)D3 levels)	Univariate linear regression (influence of sun exposure or physical activity on 25(OH)D3 levels)
R2 = 0.055, p = 0.383
Sun exposure: β = -0.081, p = 0.638	Sun exposure: β = -0.035, p = 0.837
Physical activity: β = 0.236, p = 0.174	Physical activity: β = 0.221, p = 0.183

Discussion

We investigated the effect of objectively measured physical activity on 25(OH)D3 serum levels in pwMS who had EDSS scores ≤ 4 and observed a correlation between daily activity time and 25(OH)D3 serum levels independent of sun exposure time. This is in line with a previous study, also finding a trend towards higher 25(OH)D3 serum levels in more active pwMS with an EDSS ≤ 3.5 evaluating the activity status by questionnaire and measurement of cardiorespiratory fitness using spiroergometry, an objective measure of physical performance [10]. The authors assumed that higher vitamin D levels in active people were driven through longer sunlight exposure [10] because the majority of the study population reported longer outdoor rather than indoor activity. In contrast, we found no association between sunlight exposure time and vitamin D serum levels. This may be explained by a ceiling effect of serum 25(OH)D3 levels after cumulative UVB radiation time in summer but not by previous vitamin D supplementation as the vast majority did not take any such medication. The steady state between vitamin D production and degradation is already reached within 20 min of UV-radiation exposure in Caucasians and with further UVB exposure previtamin D3 is photoisomerized to the biologically inactive isomers lumisterol3 and tachysterol3 [11,12]. The ceiling concentration of serum 25(OH)D3 was obtained at levels of 55–80 nmol/l through phototherapy [12] comparable to our study participants with an average level of 86 nmol/l. The slight difference may be explained by use of different assays [13]. There is a number of other possible confounders that may influence vitamin D concentrations, such as skin type, genetic make-up, clothing, use of sunscreen, age and altitude [14,15]. It is hard to determine how much these factors contributed to the results which were not included or not recorded to avoid overfitting of regression analyses limiting interpretation of our observations. Similar to our findings others observed a correlation between activity and plasma 25(OH)D3 concentrations independent of sun exposure time or outdoor exercise [14,16] indicating that physical activity can raise vitamin D levels independent of sunlight exposure. How vitamin D levels are driven by physical activity remains a matter of debate. There is some evidence that parathyroid hormone, which is stimulated by exercise, activates renal calcitriol synthesis [17]. Also, the exercise-induced decrease of serum phosphate (a vitamin D inhibitor) might lead to an increase of vitamin D levels [18].

In some studies, higher vitamin D levels were attributed to activity during sunlight exposure, however, in most of these investigations physical activity independent of sunlight exposure was not accounted for. In the present study, pwMS documented 1.2 hours of sunlight exposure per day on average, which is similar to another study reporting 1.4 hours, also using a diary [19].

Our findings of high BMI and higher body weight being negatively correlated with vitamin D levels are consistent with other studies [20,21]. This can be partially explained by the dilution of intracutaneously synthesized or ingested vitamin D in adipose tissue [22] and by less UVB radiation exposure because of less participation in outdoor activities [23]. Furthermore, solid evidence suggests that high BMI levels, particularly high levels of adipose tissue induce inflammation [23]. This ongoing inflammation processes in pwMS may consume vitamin D and hence, support our findings [23,24].

Past studies addressing the daily number of steps in pwMS are inconsistent with conflicting results even in patients on a similar level of disability, presumably because different tracking devices were used. This led to a wide range of documented average step counts, starting from 5,903 ± 3,185 up to 10,243 ± 3,817 steps per day among fully ambulatory pwMS [25,26]. In our study cohort we observed an average step count (6,863 ± 2,592) at the lower end of the range described above. Likewise, the total energy expenditure in our study cohort was comparably low. The BAT device probably underestimates true physical activity as it starts counting steps only after a minimum of thirty subsequent steps allowing a maximum interval of two seconds between each step. Despite these drawbacks we selected BAT over other tracking devices due to the internal memory and battery capacities. However, because all participants used the same device comparability within our study cohort is given. We are aware of further limitations, particularly the relatively small sample size because of the explorative purpose of the study and the weak to moderate effect sizes. However, our results might be considered in future interventional vitamin D trials, not only in MS, by including activity as a covariate.

Although the sample size was low due to the explorative setting, the present study indicates that physical activity correlates with 25(OH)D3 serum levels in ambulatory pwMS. Previous studies could demonstrate a positive impact of physical activity on various clinical outcomes in pwMS [27,28]. Apart from that, physical activity might have an effect on vitamin D serum levels, whereas vitamin D supplementation can increase 25(OH)D3 serum levels but failed to show an effect on primary clinical outcomes in a series of interventional trials [4,29,30].

(XLSX)

Click here for additional data file.

21 Apr 2020

PONE-D-20-10627

Influence of physical activity on serum vitamin D levels in people with multiple sclerosis

PLOS ONE

Dear Prof. Deisenhammer,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please take note of the comments for the reviewer. For this to be considered there needs an extensive acknowledgement of the limitations of the study.

We would appreciate receiving your revised manuscript by Jun 05 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Sreeram V. Ramagopalan

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements:

1.    Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Main summary

The authors conducted a 14-days cohort-study among N=38 people with MS, with an EDSS score<=4. The aim was to assess correlations between 25(OH)D an both physical activity as well as sunlight exposure. Physical activity was measured with an electronic activity tracker, outdoor sunlight exposure with a diary. After bloodsampling, patients were followed for 14 days.

The authors report a weak negative correlation between EDSS score and acitivity (R=-0.228) but not 25(OH)D (R=-0.077).

A weak correlation between 25(OH)D levels and activity (R=0.221) but not sun exposure (R=-0.002). In a linear regression model, sunlight exposure showed in the univariate and multivariate model a less steep beta when compared to physical activity.

The authors conclude they observed no association between sunlight exposure time and serum 25(OH)D levels. They conclude that physical activity has an independent effect on MS. They conclude their paper stating that patients should be encourages to practice physical exercise rather than taking vitamin D supplements.

General comments

The authors address a relevant question, assessing the association between MS disease outcomes and vitamin D levels. However, their study is flawed by a small sample size, methodological shortcomings ignoring this small sample size, inaccurate measurements of endpoints, and an overinterpretation of own data based on earlies studies.

Major points

The authors refer to a very provocative paper of McShane et al. (Am stat 2019), justifying their decision not to set a significance level. There are several points to make regarding this statement:

1. McShane et al. argue that, in biological sciences, effects observed are usually small and variable, and that therefore a significance level of 0.05 does not per se reflect the validity of an association. The point they want to make therefore concerns mostly effect size and not sample size. In their paper, the authors reason that authors should decide whether their design (including sample size) would be accurate enough to detect an effect they want to measure. Leaving a P-value threshold is proposed as a cure for detecting small yet biologically relevant associations among small but biologically irrelevant findings, but not as a cure for underpowered studies. The decision of the authors to report in an underpowered sample correlation coefficients without adhering to any significance level is therefore not supported by the cited paper.

2. The paper by McShane et al reasons that any form of threshold-setting is not favorable (including Cohen’s classification of correlation ranges), but rather advocates that reviewers assess whether study design and data quality are robust enough to substantiate conclusions drawn including the association with prior evidence. The authors do not show scatterplots which allow the readers to assess the distribution of datapoints (including the proposed linearity of correlations), and to decide whether these datapoints convincingly support the claim of (absence of) an effect the authors propose.

3. The authors state that no power-analysis for the assumptions to be tested could be provided since there were no prior data, yet refer to 2 earlier papers in which a correlation between activity and plasma 25(OH)D concentrations independent of time spend outside was performed (Touvier et al., J Invest Dermatol 2015; Brock et al., JSBMB 2010). These data could have been used to provide estimates.

The authors refer in a variable terminology to the sunlight exposure-correlate they measured. The most accurate is probably ‘diary of time spent outdoors’ but also sunlight exposure and UVB exposure have been mentioned. The inducer of vitamin D3 photosynthesis in the skin is UVB light. Sophisticated UVB light measuring devices have been used in previous studies in MS (work of the van der Mei group, Tasmania). Besides time spent outside and BMI, UVB exposure is determined by skin type, genetic constitution, clothing, wearing sunblock, age, and altitude. In their sample of N=38, there are too many confounders to be able to conclude that no association between sunlight exposure and 25(OH)D levels is present.

The advice in the final paragraph of the paper is not in any way supported by the data which the authors collected. None of the clinical trials performed thus far has been designed to detect any effect on EDSS-score or walking ability as primary endpoint, but rather inflammatory disease activity (such as relapses or MRI-activity).

Minor points

Since the study was conducted in 2 different cities, differences between including sites should preferably be excluded (more or less sunny during sampling period, local UVB-levels are available from public sources).

The authors included a heterogeneous cohort, regarding age. The authors should also disclose characteristics as disease duration, clinical phenotype, and relapse/ MRI disease activity to provide a feeling of the patient-type included.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

6 May 2020

Response to reviewer #1

Reviewer #1: Main summary

The authors conducted a 14-days cohort-study among N=38 people with MS, with an EDSS score<=4. The aim was to assess correlations between 25(OH)D an both physical activity as well as sunlight exposure. Physical activity was measured with an electronic activity tracker, outdoor sunlight exposure with a diary. After bloodsampling, patients were followed for 14 days.

The authors report a weak negative correlation between EDSS score and acitivity (R=-0.228) but not 25(OH)D (R=-0.077).

A weak correlation between 25(OH)D levels and activity (R=0.221) but not sun exposure (R=-0.002). In a linear regression model, sunlight exposure showed in the univariate and multivariate model a less steep beta when compared to physical activity.

The authors conclude they observed no association between sunlight exposure time and serum 25(OH)D levels. They conclude that physical activity has an independent effect on MS. They conclude their paper stating that patients should be encourages to practice physical exercise rather than taking vitamin D supplements.

Response: Thanks for the concise summary.

General comments

The authors address a relevant question, assessing the association between MS disease outcomes and vitamin D levels. However, their study is flawed by a small sample size, methodological shortcomings ignoring this small sample size, inaccurate measurements of endpoints, and an overinterpretation of own data based on earlies studies.

Response: Points well taken. We address all these issues below.

Major points

The authors refer to a very provocative paper of McShane et al. (Am stat 2019), justifying their decision not to set a significance level. There are several points to make regarding this statement:

Response: We agree that there is a controversial discussion about using statistical significance. This reference has been used in lieu of a long list of other possible references addressing the same issue, but for space saving reasons decided against it. However, we have added another reference supporting the thought of not setting a level of significance because of frequent misinterpretation, particularly with respect to true versus no effect. Also, we don’t feel that paper by McShane et al is too provocative. The debate dates back several decades (e.g. Rozenboom WW, The Fallacy of the Null Hypothesis Significance Test, Psychological Bulletin, 1960; 57, 416–428) and there is a long list of scientific journals discouraging authors to use significance levels and even reporting p-values. In the instructions of GraphPad’s Prism, a very popular statistical program, it says “The entire construct of 'hypothesis testing' leading to a conclusion that a result is or is not 'statistically significant' makes sense in situations where you must make a firm decision based on the results of one P value. While this situation occurs in quality control and maybe with clinical trials, it rarely occurs with basic research.” (https://www.graphpad.com/guides/prism/6/statistics/statistical_significance_in_science.htm). Accordingly, we don’t see an urgent need to come to a firm and final decision. We rather see it as another piece of evidence supporting the positive effects of physical activity in MS.

1. McShane et al. argue that, in biological sciences, effects observed are usually small and variable, and that therefore a significance level of 0.05 does not per se reflect the validity of an association. The point they want to make therefore concerns mostly effect size and not sample size. In their paper, the authors reason that authors should decide whether their design (including sample size) would be accurate enough to detect an effect they want to measure. Leaving a P-value threshold is proposed as a cure for detecting small yet biologically relevant associations among small but biologically irrelevant findings, but not as a cure for underpowered studies. The decision of the authors to report in an underpowered sample correlation coefficients without adhering to any significance level is therefore not supported by the cited paper.

Response: We agree with the reviewer. It was not our intention to avoid a level of significance BECAUSE of lack of power. As the alpha error is one of the input parameters for a formal power calculation it could not be done. The wording in the statistical methods section was indeed unfortunate and we now made clear statements in this regard.

2. The paper by McShane et al reasons that any form of threshold-setting is not favorable (including Cohen’s classification of correlation ranges), but rather advocates that reviewers assess whether study design and data quality are robust enough to substantiate conclusions drawn including the association with prior evidence. The authors do not show scatterplots which allow the readers to assess the distribution of datapoints (including the proposed linearity of correlations), and to decide whether these datapoints convincingly support the claim of (absence of) an effect the authors propose.

Response: Good point. We have added scatterplots for better visualisation of the data.

3. The authors state that no power-analysis for the assumptions to be tested could be provided since there were no prior data, yet refer to 2 earlier papers in which a correlation between activity and plasma 25(OH)D concentrations independent of time spend outside was performed (Touvier et al., J Invest Dermatol 2015; Brock et al., JSBMB 2010). These data could have been used to provide estimates.

Response: We agree that it looks like an apparent discrepancy. As pointed out above, we have now changed the wording and reasoning regarding power.

4. The authors refer in a variable terminology to the sunlight exposure-correlate they measured. The most accurate is probably ‘diary of time spent outdoors’ but also sunlight exposure and UVB exposure have been mentioned. The inducer of vitamin D3 photosynthesis in the skin is UVB light. Sophisticated UVB light measuring devices have been used in previous studies in MS (work of the van der Mei group, Tasmania). Besides time spent outside and BMI, UVB exposure is determined by skin type, genetic constitution, clothing, wearing sunblock, age, and altitude. In their sample of N=38, there are too many confounders to be able to conclude that no association between sunlight exposure and 25(OH)D levels is present.

Response: Thanks for these important points. We have changed the wording regarding outdoor time uniformly throughout the manuscript to sun exposure time. There is absolutely no doubt that UVB light is the inducer of vitamin D3. It is just that we couldn’t observe a correlation between sunlight exposure time and vitamin D levels in the setting of the study, very likely because of the seasonal ceiling effect not allowing a further increase. We have addressed this issue in the discussion and added a corresponding statement in the results section. We are aware of the many possible confounders but did not include those in the multivariate model in order to avoid overfitting.

5. The advice in the final paragraph of the paper is not in any way supported by the data which the authors collected. None of the clinical trials performed thus far has been designed to detect any effect on EDSS-score or walking ability as primary endpoint, but rather inflammatory disease activity (such as relapses or MRI-activity).

Response: We agree that the advice in the last paragraph is not fully supported by our own data. We changed the wording accordingly and removed the heading “conclusion” to allow a broader discussion at the end of the manuscript.

True, the interventional clinical trials examining the effects of vitamin D in MS patients used various primary endpoints with relapse rates most frequently reported. Relapse rates are still the primary target in RCTs for relapsing MS and therefore also valid as primary outcome in vitamin D studies involving relapsing MS patients. However, there were a few studies using EDSS or MRI as primary endpoints, as reviewed in: Cochrane Database Syst Rev. 2018 Sep; 2018(9): CD008422. As outlined in the manuscript, none of these studies met the primary endpoint which is why we made a strong statement in favour of physical activity.

Minor points

Since the study was conducted in 2 different cities, differences between including sites should preferably be excluded (more or less sunny during sampling period, local UVB-levels are available from public sources).

Response: This study was not done in 2 different cities. Maybe the impression came up because two of the co-authors moved from Innsbruck to Vienna and are now using their current affiliation. The study was exclusively performed at the Medical University of Innsbruck.

The authors included a heterogeneous cohort, regarding age. The authors should also disclose characteristics as disease duration, clinical phenotype, and relapse/ MRI disease activity to provide a feeling of the patient-type included.

Response: We added this information to table 1 and the results section except for MRI activity. The latter was not part of the study protocol and we only perform MRI on particular request (e.g. unexpected events, change or end of treatment, etc) which is why we don’t have systemic data in our population.

Also, we discovered a minor mistake of the calculation of distance/day not affecting the overall results and conclusions, and changed the data in the table accordingly.

Submitted filename: Review PLOS one response.docx

Click here for additional data file.

18 May 2020

PONE-D-20-10627R1

Influence of physical activity on serum vitamin D levels in people with multiple sclerosis

PLOS ONE

Dear Prof. Deisenhammer,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR:

In order to accept the manuscript the limitations need to be more carefully spelt out. This study is explorative, not hypothesis testing and no conclusions can be made.

==============================

We would appreciate receiving your revised manuscript by Jul 02 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Sreeram V. Ramagopalan

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors submitted a revision of their manuscript, in which some minor textual changes were made, a figure was added, and the table was updated. The addition of the scatterplots is very helpful, the addition of clinical data is also very helpful to disclose that their cohort is not only very small but also very heterogeneous in many aspects.

Three major points remain:

1. The issue of not setting a significance level has not been solved. The authors correctly refer to other publications raising the discussion on the use of P-values, and reveal the instructions of a popular statistical program apparently provide some guidance. The severe lack of power of their dataset still receives a very limited role in the discussion of the validity of their results. The authors should at least be very careful in their formulation of study design and of conclusions they can and cannot draw from their data. This is most critical in their abstract:

*”The aim […] was to correlate” is less accurate than “The aim […] to explore correlations between”

*”In conclusion, physical activity has an independent effect on vitamin D levels” is less accurate than “in conclusion, physical activity has an effect on vitamin D levels independent of sunlight exposure time” since many other confounders were not assessed.

2. In their explanation of a lack of correlation between 25(OH)D and sun exposure time in the manuscript, the authors completely ignore that they have neglected a lot of potential very relevant confounders (see my first comments) and their proxy estimate of UVB light exposure may be very inaccurate. Explaining the lack of correlation only as ceiling effect is not valid. The authors should include a statement on this both at line 146 (were hey attribute the lack of any correlation to ceiling effect) and in the discussion section.

3. I still have problems with the revised last line of their discussion: “Therefore, there is a dual effect of physical exercise on both, vitamin D serum levels and clinical improvement, whereas vitamin D supplementation can increase 25(OH)D3 serum levels but failed to show a clinical effect in a series of interventional trials.” This statement is still not supported by their data, further downtuning would be appropriate.

First, the authors did not show there is an effect of exercise on 25(OH)D levels in their cohort, but rather that there is a weak (not talking about significance) correlation between physical activity and 25(OH)D levels in their small cohort.

Second, the authors refer to a meta-analysis from 2018 on vitamin D supplementation studies, but do ignore the two largest and most recently published trials on vitamin D supplementation in MS (Camu et al., N3 2018; Hupperts et al., Neurology 2018). The authors are correct on the observation that these studies did not make their primary endpoints. However in secondary endpoints, and subgroup analysis, potential relevant signals were noted in relapse rate and MRI outcomes. The notion on MRI endpoints is further in line with an earlier small study by Soilu-Hanninen et al., JNNP 2012.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

22 May 2020

Response to reviewer:

The authors submitted a revision of their manuscript, in which some minor textual changes were made, a figure was added, and the table was updated. The addition of the scatterplots is very helpful, the addition of clinical data is also very helpful to disclose that their cohort is not only very small but also very heterogeneous in many aspects.

Response: Thanks for carefully evaluating the revised manuscript.

1. The issue of not setting a significance level has not been solved. The authors correctly refer to other publications raising the discussion on the use of P-values, and reveal the instructions of a popular statistical program apparently provide some guidance. The severe lack of power of their dataset still receives a very limited role in the discussion of the validity of their results. The authors should at least be very careful in their formulation of study design and of conclusions they can and cannot draw from their data. This is most critical in their abstract:

*”The aim […] was to correlate” is less accurate than “The aim […] to explore correlations between”

*”In conclusion, physical activity has an independent effect on vitamin D levels” is less accurate than “in conclusion, physical activity has an effect on vitamin D levels independent of sunlight exposure time” since many other confounders were not assessed.

Response: We agree that the wording should be more careful and adopted the suggestions accordingly.

2. In their explanation of a lack of correlation between 25(OH)D and sun exposure time in the manuscript, the authors completely ignore that they have neglected a lot of potential very relevant confounders (see my first comments) and their proxy estimate of UVB light exposure may be very inaccurate. Explaining the lack of correlation only as ceiling effect is not valid. The authors should include a statement on this both at line 146 (were hey attribute the lack of any correlation to ceiling effect) and in the discussion section.

Response: We acknowledge the issue of confounders and have now added statements in the results and discussion sections. It is not that we ignored that fact, but it is hard to estimate how much of an effect those factors had.

3. I still have problems with the revised last line of their discussion: “Therefore, there is a dual effect of physical exercise on both, vitamin D serum levels and clinical improvement, whereas vitamin D supplementation can increase 25(OH)D3 serum levels but failed to show a clinical effect in a series of interventional trials.” This statement is still not supported by their data, further downtuning would be appropriate.

First, the authors did not show there is an effect of exercise on 25(OH)D levels in their cohort, but rather that there is a weak (not talking about significance) correlation between physical activity and 25(OH)D levels in their small cohort.

Response: True, we did not show an effect in terms of an interventional trial. The term effect is frequently used in context with effect size, strength of correlation or association, etc. However, we have changed the wording to correlation rather than effect.

Second, the authors refer to a meta-analysis from 2018 on vitamin D supplementation studies, but do ignore the two largest and most recently published trials on vitamin D supplementation in MS (Camu et al., N3 2018; Hupperts et al., Neurology 2018). The authors are correct on the observation that these studies did not make their primary endpoints. However, in secondary endpoints, and subgroup analysis, potential relevant signals were noted in relapse rate and MRI outcomes. The notion on MRI endpoints is further in line with an earlier small study by Soilu-Hanninen et al., JNNP 2012.

Response: We have added the 2 recently published studies by Camu et al and Hupperts et al. and changed the wording to “no effect on primary clinical outcomes”. As the reviewer rightly observes, the primary endpoints were not met in any of the interventional trials so far. We feel that evidence from primary outcomes outweigh secondary non-clinical outcomes and post-hoc subgroup analyses. A 5% significance level, which is most commonly used, by definition results in one out of 20 false positive results. The more post-hoc analyses one performs the higher the likelihood of such a false positive outcome. Altogether, 14 interventional trials (12 summarized in the Cochrane review, reference 29, and the 2 recent trials mentioned above) failed to meet the primary outcomes.

Submitted filename: Response to Revision 1.docx

Click here for additional data file.

26 May 2020

Influence of physical activity on serum vitamin D levels in people with multiple sclerosis

PONE-D-20-10627R2

Dear Dr. Deisenhammer,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Sreeram V. Ramagopalan

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

1 Jun 2020

PONE-D-20-10627R2

Influence of physical activity on serum vitamin D levels in people with multiple sclerosis

Dear Dr. Deisenhammer:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Sreeram V. Ramagopalan

Academic Editor

PLOS ONE