Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I have finished my review of the manuscript entitled “Determinants of Vitamin D Status: An Analysis in a Primary Care Setting in Lithuania of Age, Gender and Seasonality.”

The manuscript presents a large retrospective cross-sectional analysis of serum 25(OH)D concentrations in more than 14,000 individuals from a primary care setting in Lithuania. The topic is clinically relevant, particularly given the high prevalence of vitamin D insufficiency in northern European populations and the scarcity of recent large-scale Lithuanian data.

The manuscript repeatedly refers to “population prevalence,” but the cohort cannot be considered representative of the Lithuanian general population. The conclusions should therefore be substantially tempered and delimited

In regression, several major determinants of vitamin D status are unavailable or not adjusted for

and the liitation limitation critically restricts causal interpretation of the observed associations.

The statement that the study does not qualify as biomedical research because it “does not aim to test a new scientific hypothesis” is unusual and potentially inappropriate for an international biomedical journal., because even retrospective anonymized human laboratory studies commonly undergo review. The current wording may raise ethical concerns.

The discussion repeatedly attributes high vitamin D levels in infants and young children to supplementation and prophylaxis practices. However, supplementation data were not collected, therefore, these interpretations are plausible but speculative and should be explicitly framed as hypotheses rather than conclusions.

Although the manuscript includes multivariable logistic regression, the analytical framework remains relatively limited considering the very large sample size, additional analyses that would strengthen the manuscript include interaction analyses (season × age, sex × season), the spline/nonlinear modeling for age, if possible a sensitivity analyses or adjusted models stratified by pediatric/adult populations.

Figure formatting should be revised carefully.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is a study of vitamin D status of individuals who have had 25(OH)-vitamin D (25(OH)D) measured in their first blood measurement in one centre in Lithuania in 2025. The authors have assessed vitamin D status according to age, gender and season by the use of CEEE Consensus categorisation. There have been other such studies in Lithuania, which the authors have described, but this study employs a sizeable cohort, and the data has the potential to offer further insight into prevalence of vitamin D deficiency in that country.

The Introduction describes the background well. Apart from two references, which I will outline below, all other references are appropriate and helpful. One of the limitations of the study, which the authors have not highlighted, is the potential that a cohort of patients or individuals from whom blood has been collected for 25(OH)D analysis might not be a good representative of the population in general. Furthermore, there are a number of important issues regarding the statistics and the presentation of the work, which need to be addressed.

Data analysis and presentations

1. The authors say that the distribution of data has been assessed visually and/or by KDE (lines 148-9). This is insufficient, and a proper statistical assessment of normality needs to be made. For a visual assessment, a Q-Q plot should be used, but preferably Kolmogorov-Smirno, Shapiro-Wilk or Anderson-Darling statistics should be used to assess whether the data are normally distributed, and P values should be given.

2. As the vitamin D data are clearly non-Gaussian, only non-parametric statistics should be used. However, the authors have used both parametric and non-parametric statistics, and given means (with SD and SEM) as well as medians (with IQR). For almost all the presentations and data analysese, it seems that mean values have been used. This is unacceptable. All data analyses, presentations and relevant text should be amended accordingly.

3. The age distribution, which has been shown as a population pyramid in Figure 2, also appears to be non-Gaussian. It also seems to be bimodal for males. Did the Kernel density assessment show bimodality?

4. The data in Figure 6 (mislabelled Figure 4 - line 268) show a trend towards higher 25(OH)D concentrations with age beyond 18 years. The authors allude to this in the Discussions (lines 397-8). However, the data in Table 9 suggest that the prevalence of vit D deficiency increases with age. This discrepancy seems to be due to the fact that children have a better vit D status that adults. Another reason may be the skewness of the 25(OH)D data, and the use of mean values in Fig 6 instead of medians.

5. Similarly, according to Figure 6 (mislabelled Figure 4), adult men have lower values compared with adult women, but the data in Table 9 suggests that women are more at risk of deficiency. This might also be due to the fact that male children, the population of which is large, have a better vitd D status that females children. Again, the use of means instead of medians may have influenced the findings.

6. For reasons mentioned in items 4 and 5 above, it would be good to separate children from adults for analyses according to age and sex.

7. Mention is made of differences between groups, but no P value is given, e.g. lines 221-2 etc.

8. In lines 182-3, the authors describe age distribution, and refer to Figure 1. But Figure 1 depicts is 25(OH)D distribution.

9. The legend of Figure 4 offers no information on what the values and error bars represent. They appear to be mean and either SD or SEM. But the text (lines 222-4) says that the values are median, in which case the error bars do not make any sense.

10. Section 3.2 heading (line 191): The title does not quite fit the contents of the section. Importantly, this section does not describe how vitamin D status changes with age above 18 years of age.

11. In Table 9, January's data are missing. Is that because "Risk change" has been compared against January? If so, this is somewhat odd.

12. Could the fact that there were comparatively more older females than older males have influenced the finding that one gender had a higher or a lower 25(OH)D concentrations than the other gender?

13. Many table or figure legends need to be more informative. For example, it is unclear what the values in Table 8 or Figure 7 are. Are they mean values or medians?

14. There is a lot of repetition in the manuscript. Only a few examples are given below, but there are many more.
- The aims of the study have been repeated in the Materials and Methods (lines 105-6).
- The number of blood specimens used appears in the Methods, but has also been mentioned in the Introduction.
- The fact that CEEES Consensus 2022 has been used appears in both the Introduction and the Methods.
- The Lithuanian latitude has been repeated in Discussions.

15. Many figures and tables describe the same data, and either one or the other should be omitted. The details of these are below. It may be preferable to keep the figures, and present the tables with the actual values as Supplements.
- The data in Tables 2 and 3 should be added to Table 1 to consolidate them all.
- Table 4 and Figure 3 describe exactly the same data, and one of them should be omitted.
- Table 5 and Figure 4 also describe the same data, and one should be omitted. If the data are normally distributed and Figure 4 is used, then the legend should describe what the bars depict.
- Table 8 and Figure 7 also display the same data, and one of them should be omitted.
- Table 9 and Figure 8 shows the same data, and one of them should be omitted.
- Table 6 and Figure 5 show the same data. Although table 6 gives the absolute numbers, it is really the percentages that matter, and therefore the table can be omitted. Only if the authors fee strongly about stating the absolute numbers, then can be stated in the text. Arguably, absolute numbers do not really mean anything as they depend on the size of the population studies.

16. The Discussions is too long, and needs to be shortended. Granular data from other studies would not be needed unless they are important in discussing the findings of this study. Focus should be more on other Lithuanian studies, as the practice of vitamin D supplementation may be different in other countries, as the authors themselves allude to this.

The Discussions needs to give a general overview of the prevalence of vitamin D deficiency observed in the study first, and discuss further findings of the study.

Importantly, discrepancies between the findings of this study and other, especially Lithuanian, studies should be briefly discussed (lines 340-2, 397-9, 444-6).

17. Should the authors wish to shorten other sections of the manuscript, some unnecessary explanations may be removed. Examples:
- Lines 249-52, where it is obvious that 68.9 is 2 nmol/L higher than 66.9, and that both fall in the insufficiency range as described previously.
- Lines 257-9: Only median (IQR) for male and female subjects can be given. Again, mention of differences should be made only if they are statistically significant (and P values given).
- Seasonal variations have been known for decades, but has been described in long detail in the last paragraph of the Discussions.

18. The paragraph of lines 261-6 should appear after the paragraph that follows it, as the description of quartiles is given only in the subsequent paragraph. Therefore, it is not immediately clear that Q4 and Q2 refer to age quartiles. Furthermore, in line 261, it is unclear that "highest levels" refers to 25(OH)D concentrations.

19.
- Please use 'vitamin D' when describing the status, but please use '25(OH)D' (or an equivalent word) when referring to measured serum concentrations.
- The sentence in lines 94-5 belongs to the Discussions.
- There is no reference to Table 8 in the text.
- Section 3.5 title (line 291): A more informative title is required.
- Unless there is a National or adopted consensus for use of the term "elevated" vitamin D, should hypervitaminosis D be used instead of "elevated levels"? The authors use hypervitaminosis [D] in the key section of Figure 5, so it seems that they might be happy with the use of the term.

20. Please consider replacing references 3 and 5 with one or more of the following:
- Giustina A, di Filippo L, Aleksova A, et al. Consensus statement on vitamin D role in metabolic health. Metabolism. 2026;176 :156484. doi: 10.1016/j.metabol.2025.156484 (PMID: 41435994)
- Dalamaga M, Emfietzoglou R, Petropoulou D, et al. Vitamin D and Health Outcomes: State-of-the-Art Review of Triangulated Evidence and Ongoing Controversies. Curr Nutr Rep. 2026; 15: 26. doi: 10.1007/s13668-026-00748-2. (PMID: 41849024)
- Młynarska E, Lisińska W, Hossa K, et al. Vitamin D and Chronic Disorders: A Review of Metabolic and Cardiovascular Diseases. Pharmaceuticals 2025; 18: 1467. doi: 10.3390/ph18101467. (PMID: 41155582)
- Deepika, Kumari A, Singh S, et al. Vitamin D: recent advances, associated factors, and its role in combating non-communicable diseases. NPJ Sci Food. 2025; 9: 100. doi: 10.1038/s41538-025-00460-5. (PMID: 40514375)

 

Comments on the Quality of English Language

Some parts of the text can benefit from a better use of English. A few examples are given below.
- Lines 294, 296, 298-9: Gender is not a 'risk factor' per se. Rather, lack of sun exposure or vitamin D supplementation is a risk factor that may be more relevant in a certain gender or age group. Also, age and months of year do not have any 'protective effects'. Please describe these in the context of prevalence or risk rather than "protective effect".
- Lines 48-50: ".... healthy adults younger than 75 years require no routine screening for 25(OH)D in the general population and empiric vitamin D supplementation". This can mean that they require no empiric supplementation.
- Lines 91-2: "mean 25(OH)D levels during the warm season being approximately two-fold higher than during cold season". This means that if the concentration was X, a two-fold higher concentration would be X+2X, i,e, 3X. But I think the authors mean the concentration in the warm season are approximately twice the concentration during the cold season'.
- Line 292: "a multivariate logistic regression models were created"
- Etc

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Authors have improved the manuscript significantly, but there are still aspects to be addressed.

This paragraph needs to be revised: "Our study does not satisfy the first criterion, as it does not aim to test a new scientific hypothesis or create novel theoretical knowledge..."

Therefore, claiming that it is not biomedical research because it does not test a scientific hypothesis is not convincing.

The authors may be correct that Lithuanian regulations exempt anonymized retrospective datasets from ethics review

the authors should ensure that conclusions do not imply national representativeness beyond the studied clinic population.

Comments on the Quality of English Language

authors often use qualifying language ("may", "hypothetically", "could"), which makes these acceptable, but a few sentences could still be shortened.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this revised manuscript, the authors have addressed a number of points that had been raised previously. The manuscript has improved significantly, but, unfortunately, attention has not been paid to a few important issues that have been missed.  These relate again to the analyses and presentations based on parametric statistics.  They have been detailed below.

- Both Figures 5 and 6 are exactly the same as those in the previous version of the manuscript, in which the values represented mean 25(OH)-vit D concentrations.  Again, the skewness of the data, which the authors themselves allude to in lines 354-8 and lines 372-4, makes the use of mean values inappropriate.  These figures should be replaced with ones that represent medians and IQRs.
The legends to Figures 5 should then describe the fact that the values are medians and what the shaded areas represent.
In Figure 6, the y-axis should mention that it represents serum 25(OH)-vit D.  The legend to the figure should also describe the fact that the values are medians, F and M represent females and males respectively, and 1-4 represent age quartiles.
The text in the Results and Discussions should then be reviewed and revised accordingly in order to describe the findings based on medians and IQRs.

- Lines 142-4: References to mean and SD are made.

- Lines 238-9: A reference to SD is made.

 

A few other points, which are relatively minor and are highlighted below, also need to be mentioned.

- In lines 53-8, the authors outline the way in which they have categorised 25(OH)-vit D concentrations, but this appears before any information on the study has been given.  Therefore, the sentence in those lines is misplaced.  In fact, it can be deleted from the Introduction, as the classification is described in the Methods section (lines 29-34).

- There are discrepancies between the categorisation of high 25(OH)-vit D in lines 133-4 and in Figure 4. (In a number of guidelines, a 25(OH)-vit D of >375 nmol/L is regarded as toxic.)
If the classification in Figure 4 is amended, the related text (lines 215-26) will also need to be amended accordingly.
Also in Figure 4, "Clinical status" in the colour-code key should be changed to "Vitamin D status".  The key gives the related 25(OH)-vit D concentrations, but the units are missing.

- The reference to Figure 5 (line 246) is incorrect, and it should state Figure 6. This would then make reference to Figure 6 appear before reference to Figure 5.  Therefore, a rearrangement of the text in that section is necessary.

- Line 249: Please state "The oldest cohort (Q4) exhibited the highest levels 'of 25(OH)-vit D', with a median of ....".

- Line 304: The sentence is incomplete.

- Lines 293 and 298: Correct hyphenation is needed, i.e. 'well-established', 'age-stratified'.

- Lines 260, 264 and 315: Please state that the values are median concentrations.

- Line 338: It is unclear what the "highest baseline 25(OH)D" means.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf