R0:

Reviewer #1:

To the authors: Respected authors, the theme of your research is highly relevant and timely, particularly given the importance of equity, patient-centered care, and stakeholder engagement in health systems research. This manuscript presents a valuable and comprehensive effort to identify priorities for advancing brain health equity after traumatic brain injury by engaging multiple stakeholder groups. I particularly commend the inclusion of individuals with lived experience and the commitment to co-creation throughout the project. The integration of social determinants, stakeholder perspectives, implementation considerations, and knowledge mobilization strategies reflects a thoughtful and mature approach to equity-oriented research. It is refreshing to see a project where people with lived experience are not merely participants, but meaningful contributors to the development and prioritization process. It is also encouraging to see research that not only identifies challenges but also seeks to advance implementation and develop actionable solutions. The project is ambitious in scope, methodologically detailed, and addresses an important gap in the literature. The work demonstrates substantial effort, coordination, and dedication from the research team. Indeed, the scale of stakeholder engagement, co-creation activities, and implementation planning could easily constitute a full thesis project or multiple standalone publications. I was also encouraged to see that the work has already moved beyond priority setting and toward dissemination and action, with the outputs already generating substantial engagement and visibility. The manuscript is generally well organized and scientifically relevant. However, I have a few concerns and comments that I hope will strengthen an already valuable contribution.

Major Comment 1: My primary concern relates to the characterization of the study as a "global" priority-setting exercise. The findings support priorities among the sampled stakeholders but do not support claims regarding global stakeholder priorities. As the authors note, the participant sample is overwhelmingly concentrated in North America and in high-income countries. According to the manuscript, 88.5% of respondents originated from North America. Consequently, many countries, health systems, cultural contexts, and populations that may experience substantially different barriers to brain health equity are not represented within the stakeholder sample. I recognize that achieving truly global representation is extraordinarily difficult and that no study can realistically capture all perspectives. I also appreciate that the authors acknowledge this limitation and have already initiated efforts to engage underrepresented populations through future grants and collaborations. Nevertheless, given the study's central focus on equity, caution is warranted when presenting these findings as global priorities. Geographic, socioeconomic, cultural, and health system contexts can substantially influence stakeholder priorities. Even within the same country, priorities may differ considerably across regions, communities, and populations. When the ultimate goal is to identify priorities and guide action, these contextual differences become especially important. For this reason, I encourage the authors to consider these options: Reframe the manuscript as a North American study, as the vast majority are from such a region, and even the Knowledge mobilization priorities and implementation tracks identified during Round Robin activity seem to be directed to such a region. Or even limiting to the Canadian context. Reframe the manuscript as a multi-country stakeholder priority-setting study rather than a global study. (which still would be taken with care and not truly appropriate) Adopt more cautious language when discussing global implications and generalizability. Clarify throughout the manuscript that the identified priorities reflect the perspectives of the stakeholder groups represented within this sample. Consider waiting for more data from additional countries, since this initiative has started, as mentioned in the limitations section. Or publishing this study as a preliminary location(s) with the aforementioned suggestions, and then expanding this initiative across a broader range of countries and regions. Engaging with more neurological associations and brain organizations, as well as local leaders across different countries and regions. Such an effort would provide an extraordinary opportunity to compare priorities across settings and strengthen the evidence base needed to support truly global recommendations. In addition, future iterations could consider region-specific analyses or parallel priority-setting exercises conducted within different geographic contexts, followed by a comparative synthesis. This approach would allow for both local relevance and broader cross-context insights, ultimately strengthening the validity and applicability of global recommendations.

Major Comment 2: I appreciate the authors' attention to inclusivity and the deliberate involvement of diverse stakeholder groups, including Indigenous communities and individuals from different racial and ethnic backgrounds. However, the manuscript reports that more than 85% of participants identified as White. This limitation is acknowledged but warrants further discussion given the study's focus on equity and the development of priorities based on such. As highlighted, previous research has demonstrated important differences in TBI incidence, risk factors, access to care, outcomes, and long-term recovery across racialized and marginalized populations. As such, the priorities identified in this study should be interpreted with consideration of the demographic composition of the sample. The discussion and limitations sections could further acknowledge how the underrepresentation of racial and ethnic minority populations may have influenced the priorities identified. Future efforts could benefit from targeted partnerships with advocacy organizations, Indigenous communities, Black communities, Latino communities, hospitals, clinics, and community leaders serving underrepresented populations to ensure broader participation and co-leadership in future iterations of this work. I also believe that the proposed future studies will benefit from collecting a broader range of social identity variables. As well as exploring the intersections between these determinants, some of which have been assessed. Experiences of inequity are often shaped by multiple overlapping social factors rather than a single characteristic in isolation. An intersectional approach may therefore provide a richer understanding of how social determinants influence priorities, access to care, and outcomes following traumatic brain injury.

Additional Comments by Section: 1. Title: While informative, the title may not fully reflect the breadth and complexity of the multi-phase work undertaken and places considerable emphasis on the term "global" despite the acknowledged limitations in representation. 2. Background: No major comments. The background is clear, concise, well-referenced, and effectively establishes the rationale for the study. 3. Methods: The methodology across all three objectives is sound, comprehensive, detailed, and generally well described. These are my suggestions and questions: I particularly appreciated the use of REPRISE and PROGRESS-Plus frameworks. I also suggest considering the SAGER guidelines for reporting sex and gender. And given the survey component, consideration of the CHERRIES checklist may strengthen reporting transparency.

The manuscript states that the survey was designed to maximize accessibility and global reach. If global reach was a primary objective, was translation into additional languages beyond English considered? Additionally, regarding the knowledge dissemination materials described at the end of this objective, the manuscript emphasizes maximizing accessibility and global reach. If the intention is truly global, I would suggest translating materials into more languages beyond English and French. Otherwise, it may be more accurate to describe this as enhancing accessibility within specific linguistic regions rather than globally.

I truly appreciated that multiple stakeholders contributed to authorship and were involved in reviewing the survey prior to dissemination, as well as the a priori effort to recruit respondents across stakeholder groups. However, it was not entirely clear how recruitment targets were determined. Were formal sample size calculations performed? Was thematic saturation considered? How were recruitment targets established for each stakeholder category?

Additional details regarding survey dissemination would be helpful. Through which channels was the survey distributed? Were professional organizations, hospitals, rehabilitation centers, advocacy groups, social media platforms, mailing lists, or community organizations involved? What strategies were used to reach underrepresented populations? This would assist readers in evaluating representativeness and potential selection bias.

1. Results: Overall, the results are clearly presented. Yet a few considerations: I was unsure how to interpret the following sentence: "To further explore the relevance of social parameters for brain health outcomes after TBI, we utilized ML algorithms ... to data extracted from 30 published studies..." I suggest rewording for clarity.

I also suggest rewording the following sentence for clarity: “Taking into account the gaps in existing evidence, survey development and the following objective were intentionally designed to remain broad, enabling comprehensive capture of relevant social equity determinants without restriction.”

Some information currently reported within the Results section appears methodological and may be better placed within the Methods section, or removed if repetitive.

It would be easier to interpret the table if the total weighting reference (5 total) were explicitly stated within the table caption.

I found the framing of some social determinants difficult to interpret. Certain variables appear to be presented as adverse conditions (e.g., housing insecurity), whereas others are presented as positive resources or protections (e.g., access to benefits). Could this have influenced how respondents perceived and prioritized these determinants during survey completion, or could this have affected the interpretation of the resulting weighted scores? For the display of results, for clarity, would it be appropriate to consider a more consistent framing of these variables?

The text following Figure 2 appears somewhat disconnected from the surrounding content. Please verify whether this text belongs within the main manuscript, the figure caption, or supplementary material. This text: “The goal is to provide access and link for people with lived experience to resources across the continuum of care and reducing accessibility barriers. The content areas that were discussed at the Round Robin activity for feasibility and impact, listed in the boxes, are not exhaustive. Complete list of topics is listed in S9 File.” Additionally, the S9 File is already referenced earlier.

Additional figures, given the large number of results and supplementary material, could be beneficial for better visualizing the findings.

1. Discussion: Well written and engaging, however, I encourage the authors to align interpretations more closely with the study's sampling characteristics. In particular, I would avoid statements suggesting that the study demonstrates both global relevance and sensitivity to local context. The findings clearly demonstrate relevance across different stakeholder groups, but the limited geographic and social representation makes it more difficult to support broader conclusions. Another suggestion, if the authors are interested, would be to include examples of similar initiatives, implementation strategies, or policy responses proposed or adopted in other countries. Such examples could provide additional context and help strengthen the proposed calls to action.

2. Limitations: appropriately developed but could potentially be expanded further. Additional limitations that may warrant consideration include: -> Survey availability only in English. ->Limited racial and ethnic diversity within the sample. While this is mentioned briefly, it would benefit from deeper discussion regarding implications for survey design, recruitment, interpretation of findings, and broader consequences. It would also be helpful to outline concrete steps for improving representation in future work. -> Potential selection bias associated with survey dissemination channels. -> Limited consideration of intersectionality across multiple social determinants. -> Virtual participation and Zoom-related accessibility barriers, as well as other potential accessibility limitations. -> For the systematic review component, the authors note that several important social determinants were reported infrequently across the included studies and were therefore excluded from the analysis. The implications of these missing data for subsequent analyses should be discussed further. -> I was also interested in the finding that approximately 88% of survey respondents were female, despite previous evidence suggesting that TBI disproportionately affects males. The authors may wish to discuss possible explanations for this discrepancy.

I encourage the authors to address the comments above and resubmit the manuscript for further evaluation. I congratulate the authors on undertaking such a project. The amount of work, coordination, stakeholder engagement, and dedication required to complete a three-stage study of this nature should not be underestimated. Too often, research stops at describing problems. I appreciate the effort to move beyond, engaging stakeholders throughout the process, co-creating solutions, and beginning to translate findings into practice. It is also encouraging to see that these efforts have already generated substantial engagement and dissemination. While I have raised concerns regarding representation and interpretation, these comments stem from a shared commitment to ensuring that equity-focused research is approached with particular care and reflection. I recognize the challenges involved in conducting this type of work and appreciate the authors' transparency and aim. Overall, this is a thoughtful, well-executed, and important topic that I hope continues to grow through future collaborations and broader stakeholder engagement.

Reviewer #2:

Discussion draws broad conclusions about cross-cultural consistency that are not adequately supported by the data. I suggest tempering the generalizability claims throughout the manuscript or more explicitly qualify them as reflecting primarily North American and European perspectives. Provide more explicit justification for why saturation was claimed across all stakeholder groups, particularly given the small n in the researcher and healthcare provider categories. I would more explicitly discuss how community voice was preserved and weighted during the Round Robin, particularly given that Services and Benefits themes were underrepresented in the team discussion relative to the external survey.

This is an important topic, thank you for your efforts.

Reviewer #3:

This manuscript addresses an important and underexplored issue the integration of social parameters of health into traumatic brain injury (TBI) research, policy, and practice through a multi-stakeholder priority-setting process. There is strong rationale for the study as an equity issue and a timely need to address this. The study combines evidence synthesis, stakeholder engagement, and consensus development within the PROGRESS-Plus framework to develop an equity-oriented knowledge mobilization agenda. The topic aligns well with the aims of PLOS Global Public Health and has the potential to make a meaningful contribution.

The manuscript is thoughtfully conceived and clearly reflects substantial effort from an interdisciplinary team. The commitment to stakeholder engagement, representation from those most affected by but least represented in the research about TBI, and both short- and long-term knowledge mobilization are notable strengths.

However, several methodological and reporting issues currently limit the manuscript's impact. Most importantly, the manuscript is overly ambitious: Addressing all three objectives in one manuscript is too much for the reader to digest. There are insufficient methodological justification and/or scholarly references for several analytic decisions. For example, there is little scholarly justification for why Round Robin approach was selected over established priority-setting approaches such as Delphi Technique or Nominal Group Technique. As another example, there are no scholarly references for thematic analysis of open-ended survey answers.

Major Areas to Address:

1. The manuscript attempts to accomplish too many objectives. It combines evidence synthesis, machine learning meta-research, survey development, stakeholder engagement, qualitative analysis, consensus building, implementation science, and knowledge mobilization. Each component is valuable, but together they reduce the impact of each. Please clarify the primary contribution of the paper.

2. Strengthen mixed-methods integration. Although described as a convergent parallel mixed-methods study, the three objectives function largely as sequential work packages. Consider using joint displays, explicit meta-inferences, and a clearer explanation of how qualitative and quantitative findings informed one another.

3. Expand discussion of sampling implications. Address the implications of English-only recruitment, access to Zoom, organizational recruitment networks, volunteer bias, digital access, and limited representation from low- and middle-income settings in a manuscript that implies global priority setting

Minor Areas to Address: • Clarify the novel contribution beyond previously published systematic reviews and protocol papers (specifically with respect to Objective 1) • Simplify Table 2 and distinguish PROGRESS-Plus domains from individual factors. • Deepen interpretation of why specific priorities emerged during the Round Robin process. • Reduce repetition of "knowledge mobilization" throughout the manuscript and shorten several lengthy paragraphs.

Overall Assessment:

This manuscript addresses an important gap in equity-oriented traumatic brain injury research and demonstrates impressive stakeholder engagement, and commitment to knowledge mobilization. The integration of evidence synthesis, stakeholder priorities, and implementation planning is innovative and well aligned with the aims of PLOS Global Public Health.

The principal revisions required concern methodological clarity, analytical integration, transparency of reporting, and cautious interpretation of findings. Consider separate manuscripts for Objectives 1, 2, and 3. With these revisions, the manuscript has the potential to make a valuable contribution to the literature on equity-oriented priority setting and knowledge mobilization in traumatic brain injury research.

R0:

Reviewer #1:

The topic addressed in this article is important and, moreover, highly relevant. However, the authors limit themselves to interpreting the data obtained from their quantitative study, neglecting the need to also critically examine it and enrich it with other theoretical contributions. Therefore, for the publication of this highly relevant and necessary text, it is essential to enrich the theoretical dialogue precisely in the conclusions section, which is somewhat sparse and brief. Some elements that could help broaden this dialogue include: 1) the importance of comprehensive sex education, since access to family planning programs is not always sufficient to achieve gender equality and full access to human rights; 2) the right to reproductive health also involves understanding other rights, such as the right to a life free from violence and the connection to concepts like consent. Finally, although these issues are not the primary focus of the study, the conclusions would provide a suitable space to explore in greater depth certain elements that were only briefly mentioned, such as the male partner’s commitment to family planning and the internalization of cultural norms as a barrier to family planning. Presenting the findings is important, however, the production of scientific knowledge also requires a critical discution of their implications and limitations.

Reviewer #2:

Although the data supports the conclusions, is desirable to qualify some arguments. Authors mention relevant concepts that match intercultural and intersectional frameworks, such as equity, gender equity, culturally appropriate; but this sensitivity is unsteady. Authors recognize that qualitative inquiry is needed to explore cultural dynamics and social constraints, as part of the limitations of the article; however, this doesn't qualify suggestions or arguments found on pages 2, 19 and 20, when referring what health authorities should do or when assessing "real progress" (p. 20).

R0:

Reviewer #1:

This application of a qualitative research analysis will likely be of interest to programmatic and policy audiences as well as research teams looking to support programs. Some further information around the toolkit itself and touching on options available for programs with limited resources (if possible) would help. Specifics are outlined below: 1. Introduction – though the introduction states, “We previously developed the Options Assessment Toolkit (OAT) as a decision-support instrument in response to the first major expansion of tools for Plasmodium vivax malaria in more than six decades (3-5)” and points to references, it would be helpful if a brief explanation of what the toolkit includes and is intended to be used would be helpful. 2. Results – similar to the point above – it would be helpful to have some examples of what the OAT is expected to be used for. What are some examples of potential policy adjustments that would be made when a program uses it. 3. Results – the OAT serving as a checklist. It is hard to know how to interpret some of the responses. Did the NMP from Afghanistan mean that it points out what is missing from their available commodities? Or processes? 4. Results – the Vietnam NMP said that it is not that different from what they already do? Is it possible to briefly summarize what the Vietnam program does? 5. Results – lines 215-217 were unclear. 6. Results – lines 225-226 – could OAT be adapted to provide a “next best thing” in the absence of system readiness? Like a decision tree based on the robustness of the system? From the discussion, my guess is probably not, but would be curious if this was a topic in the toolkit development process at least. 7. Results – lines 227-229 – was it discussed that identifying gaps was a useful product in and of itself from using the toolkit? 8. Results – can any more details of the results or responses be shared other than those highlighted? 9. Discussion – great points here. The second paragraph really resonated. 10. Discussion – I agree with what you say in line 315. It is likely that the process of developing the toolkit impacted programmatic operations or at the very least mindsets. Were there questions about policy changes after OAT that may have been indirectly related to the development of OAT? Meaning: were there changes that just ‘happened’ to overlap with the post-implementation or OAT introduction timeline? 11. Discussion – lines 359-361 – this is also a great point and programs having the ‘confidence’ or ability to deviate from simply what is prescribed by WHO or otherwise could results in innovative strategies.

Reviewer #2:

This manuscript by Shrestha et al. addresses an important and timely topic by qualitatively assessing users' experiences with the co-development and early implementation of the Options Assessment Toolkit (OAT), which is intended to support National Malaria Control Programmes (NMCPs) in selecting the most appropriate Plasmodium vivax radical cure strategies. Understanding the utility and practicality of the OAT among its intended users and policymakers across different country contexts (i.e. co-developing and non-co-developing countries) could potentially facilitate the adoption of appropriate P. vivax radical cure policies. Overall, the manuscript is well written and presents valuable insights. However, I have several minor comments for the authors to consider.

Line 66: Typographical error noted ("...co-developed between 2020 and 2022 as...").

Line 91: It is understandable that three countries (Afghanistan, the Solomon Islands, and Vietnam) were involved as the original co-developing countries for the OAT. However, I wonder why only one additional country, Nepal (outside the original co-developing settings and represented by only two participants), was selected from the 22 official member states in the Asia Pacific region. Does the inclusion of only one non-co-developing country provide sufficient basis for an unbiased comparison of "...between co-development-driven ownership and independent engagement with the toolkit" (Lines 92–93)?

Lines 95–96: The authors state that participants held "...senior technical or managerial roles within the NMCPs and were directly involved in malaria policy formulation or advisory processes." Could the authors provide additional information regarding the participants' experience in these positions? For example, how many years had they served in these roles?

Lines 105–107: The study included a total of nine participants from the original co-developing settings and only two participants from a non-co-developing setting. Please clarify how participants were selected. Was purposive sampling employed? Was the sample size enough? Were there any inclusion or exclusion criteria? Did any eligible participants decline to participate?

Lines 261–267: The manuscript states that "Nepal participants described a different perspective." I wonder whether this observation reflects the fact that Nepal was not originally involved in the co-development of the OAT. Consequently, their views may represent a more independent assessment of the toolkit, without any perceived ownership or prior investment in its development.

Line 297: Typographical error noted ("...sstructured...").

General comments:

Could the authors elaborate on what additional steps are required to ensure that the OAT will be widely accepted and adopted as a decision-support tool by NMCPs?

WHO guidelines are developed through extensive consultation with international experts and are based on broad global consensus, whereas the OAT appears to have been developed from a limited number of country settings. Could the authors explain why the OAT has been developed as a standalone decision-support tool rather than being incorporated into existing WHO guidance? Furthermore, how do the authors envision the relationship between the OAT and future WHO recommendations for P. vivax radical cure policies?

R0:

Reviewer #1:

Manuscript Number: PGPH-D-26-01594 Title: Sensitivity and specificity of tuberculosis screening tools in people with diabetes Dear Editor, many thanks for requesting me to review this manuscript by Janrode et al. It is a well-written paper addressing an important public health challenge in a clinical setting with a high dual TB-diabetes burden. Below are my minor comments. 1. For the study, you used a manufacturer’s cut-off for the CAD4 TB score, which might be inappropriate for your population. Using your data, are you able to compute a cut-off ideal for this local population, because AI-aided diagnosis of TB is the future of TB diagnosis in high TB burden settings? 2. In the discussion section, could you include plausible theories or hypotheses to explain the differences in the sensitivity of CXR readings in people with poor vs good glycaemic control and males vs women? 3. In the discussion section, in addition to using AI-aided technology, could you propose additional strategies that can be used in clinical practice or explored in research to improve the diagnosis of TB in people living with diabetes? For example, is there a potential role of haematological parameters such as absolute WBC counts, neutrophil: lymphocyte ratio, and monocyte: lymphocyte ratio? These are simple and cheap parameters that have been shown to predict specific TB treatment outcomes.

Reviewer #2:

PLOS Global Public Health Sensitivity and specificity of tuberculosis screening tools in people with diabetes

Manuscript Number: PGPH-D-26-01594 Article Type: Research Article

REVIEWERS COMMENTS This is an important area of study.

Comment one: The authors could improve the description of the diabetic population phenotype in this study. How many participants had other comorbidites other than HIV? What diabetic medications were the participants on? Especially in the light that some diabetic medications can slow TB progression or precipate it? How many of the patients with HIV had a hsitory of taking TB preventive therapy? What about the socio-economic status? Levels of education? Duration of the diabetes.

Comment two: We appreciate that all participants gave sputum for TB evaluation. How many presented with any cough? A descritpion of the sputum collection or induction process in the methods section would be good.

What were the different cough durations. This population often has respiratory infections, especially viral respiratory infections. The question would arise if we need to evaluate such coughs for TB and the cost implications eventually. Therefore a sub-analysis based on duration would be very insightful.

Comment three: It would also be helpful to also describe the prevalence of the TB symptoms assessed. And also the proportions of the CAD4TB scores below or above 60.

In the combined classification, were all TB symptoms weighted equally?

Comment four: Where was the TB testing performed? At the peripheral Health center TB laboratory or at TB accredited BSL 3 laboratory?

However, the limitations of using a positive Xpert MTB/RIF Ultra test result as the reference standard instead of sputum culture are noted.

Comment five: There were 17.2% participants with previous TB in the study. Why did the authors include the chest X-rays (CXRs) of these participants in the analysis? The results are not surprising and therefore another analysis excluding this population might have better appreciated results.

Comment six: The study adjusted for most of the confounders but omitted occupation which in South Africa is an important risk factor. How many of these participants were miners or have risky occupations? This could affect the CXR findings.

Comment seven: How do these findings align or misalign to finding in other vulnerable populations or to the national TB screening algorithm?

Comment eight: The authors did not discuss all the results despite having many supplementary results depriving the readership of deeper explorations of the findings. And they only referenced 18 studies for such an important study.

R0:

Reviewer #1:

For all the manuscript, correct anti hypersive patient to people with hypertension

1. Abstract: a) Material and Methods:
2. Quantitative data from 397 patient files : It is already a result.Insert it among results.
3. Why Poisson regression and not a linear one?

b) results - there is no characteristics of people living with HTN - add p-value on the regression results: aPR = 7.00; 95% CI: 3.08–15.92, p=

1. Introduction :
2. Line 74: Giving that the HTN could reach 50% among adults in certain places, I would like to propose you to say that there is a discrepancy between urban and rural prevalence of HTN..
3. Line 76 - 77: limited access to care is more global

4. Line 91: specify which guidelines please

5. Materials and Methods

6. Line 97 : No need to precise that is analytical, either you say that this study had 2 components…
7. You stayed mixed methods, precise which one
8. Line 110: outpatients departments or consultations?
9. Line 111: all cadres ??? What do you want to say ? any type ???
10. Line 118: i. All patients with Hypertension who were attended from January 2024 to December 2024 at the selected public district hospitals and have case file and agreeing to be part of the study. ii. All healthcare providers in the public district hospitals who are taking part in the prescribe management of for Hypertensive patients with hypertension and agreeing to be part of the study
11. Line 125: exclusion criteria. dis yiu include Pregnant women??? or Retraction from the study???
12. Line 130 : You did have any minimum sample size???
13. Line 134: I would have prefered that you have a structured questionnaire. qualitative interview are biased
14. Line 165: what is ODK

15. Line 208: what is ATLAS.ti ?-

16. results

17. Figure 1: What were the inegilibility criteria?
18. Line 262: You choose some healthcarer. Why and how did you choose them?
19. Line 272: Pvalue ??? You gave the p-value for the follow-up visits, presents similarly grade &, 2 and 3 hypertension

20. Presentation of barriers is too long 'almost 5 pages). Please try to reduce it

21. discussion:

22. Line 129 : Comparable trends have been observed in other sub-Saharan African settings, including Ethiopia (19.5%), Ghana (30%), and Zimbabwe (35%) (28,30,31), where inadequate provider training and resource limitations were key barriers to adherence. Did these studies proved what are you stating?
23. Line 143 : was also higher in management of patients with Grade 1 and 2. Is it not indicative of therapeutic inerty
24. Line 153: Use the conditional tense. Be less assertive.
25. Line 177 : Please cite Hypertension in Sub-Saharan Africa: Burden, Barriers and Priorities for Improving Treatment Outcomes Doi: 10.1161/CIRCRESAHA.124.323889

R1:

Reviewer #2:

The manuscript can be improved as follows:

1. The introduction contains useful points but is too long. It can be shortened and made to focus more on the problem that the manuscript attempts to answer, which is adherence to standard treatment guidelines.
2. Data analysis (P9, line 212)- The authors justify the use of the adherence cutoffs (e.g. 65% is good) based on prior studies that used the same cutoff. However this is a very low cutoff and there are data to show that this level of adherence would negatively affect patient outcomes. A better method should be considered or stronger justification for using these cutoffs.
3. Results a. Line 279- please avoid using causal language in describing association between presence of comorbidities and receipt of guideline adherent care b. Resource barriers P19- is it possible to quantify the resource barriers e.g. by systematically reporting level of availability of essential drugs and equipment in the participating hospitals? There are tools such as the WHO SARA tool that can be used to do this

R0:

Reviewer #1:

Definitely a timely article - contributing to the evidence base around HCV self-testing - an important further step in HCV elimination methods.

Overall, I thought the paper to be well-written and close to publication. However, there were some areas that I thought needed either amendment or possible additional analysis. In particular, I have some real concerns about the description of acceptability analysis.

Abstract:

• Clarify that use of HCVST is referring only to antibody testing (this applies to the paper throughout)
• I would clarify that analysis is only based on 1,995 valid participants.

Introduction:

• Perhaps just a quick sentence describing the nature of Nigeria's epidemic - i.e. is it generalised of population specific?

Methods:

• IMPORTANT: Needs to be more information about participant recruitment. The paper at times implies that the large sample size equates to high acceptability - but there is no information about the refusal rate. This would actually be a critical aspect of determining acceptability of self-testing. If you recruited 2,000 participants - but an additional 2,000 refused to participate (for example) - this would suggest self-testing is acceptable to only 50% of clients.

• IMPORTANT: Very little in methods about your qualitative interviewing - this needs expansion.

• IMPORTANT: There is very little satisfactory information about the acceptability questionnaire - and it's subsequent transformation into an acceptability score using factor analysis. Currently, it reads as if the audience is simply supposed to trust the authors that all methods were satisfactory, and that - therefore - results are valid. You don't specify which questions were included in factor analysis, or if it was appropriate to score each item similarly or to aggregate scores. You also don't indicate if the questions were a previously developed tool or something developed specifically for the study - or that it's even available in the supplementary data. This almost feels like a separate paper, so that your method of transforming the acceptability survey can be more properly peer reviewed.

• IMPORTANT: For you acceptability analysis, how did you determine that 10% equals 'unacceptable'? Shouldn't people with only a 20% acceptability score be considered broadly non-accepting?? Also, considering all the noted limitations and collinearity issues with site - does it even make sense to compare by site?

• How did you determine the 2,000 sample size?

• The selection criteria is difficult to follow and should be clarified - perhaps with dot points?

• Specify that "blood-based" testing means finger prick?

• Again, need to clarify that testing is antibody, correct? Presuming also that RA information discussed testing procedures – i.e. what happened if the ST was positive and it’s interpretation?

• How many of each test did you have? Why did you not have enough tests so that any participant could choose whichever test they wanted for the entirety of the study period?

• Presuming drug abstention is not a requirement to treatment initiation?

• For the exposure variables - is 'education' referring to attendance or completion?

• You explain later in the paper that participants could be part of multiple KP categories - but probably good to explain in the "exposure variables" section.

Results:

• Non-considering possible refusal rates - your results suggest that of 2000 participants, essentially 100% completed the self-test? This is a remarkable outcome – particularly for the 849 participants who were unobserved!

• "Five participants 255 took HCVST kits home but did not return to complete the endline survey, therefore no demographic or clinical information was recorded for those participants, and they were excluded from the analysis." Why wasn't demographic information collected at baseline, as would be usual?

• Suggest re-doing Figure 2 - removing the column for those 'tested' as this is inherently 100% and that every subsequent column is small by comparison. Add percentages to columns. Could consider comparing clinic types.

• Following on from methods comments regarding the acceptability data - you don't specify the number of participants falling within the purported "least 10%". The sentence describing qual data “Acceptability appeared to be more associated with clients characteristics than facility type…” - also suggests there are issues with your acceptability analysis methods.

• Did you consider differences across test type? In particular those who elected to conduct unobserved self-testing? What were the characteristics of these individuals?

• Sentence on page 18 “Observations about test type choice are thus descriptive only, not inferential” should probably be in the methods – following discussion of how tests ran out.

Discussion:

• How much do you think the demographic differences between facility-type clients impacted outcomes? For example – OSS clients are much younger and more males.

• OSS clients used more ‘blood’ tests. Did this have any impact on acceptability?

• The statement about PWID ‘surprise’ may be misplaced. How many PWID did you approach, who refused to participate? How does the number recruited compare against the OSS client list? Also, it’s 28% (PWID) compared to (31% MSM) – are you surprised about MSM?

• “Additionally, we did not assess whether the acceptability scale functioned equivalently across ART and OSS settings, so observed differences should be interpreted with that caveat.” - what does this statement mean? Is this another potential flaw of the acceptability scale?

• I'd include explanation of the SVR12 limitation somewhere higher in the paper.

AE:

This is an important report on controlling viral hepatitis and, in particular, hepatitis C (HCV) in Africa. The report is well written and is also a good example of how to conduct feasibility studies. The authors managed to set a clear and moving background. However, there are a few issues that must be addressed.

Issues: 1. Please indicate how the 2000 sample size was determined.

1. Line 132 is a bit unclear. Can it be rewritten?

2. About the acceptability, it needs to be recognized that this instrument has never been tested in this population before. The EFA and CFA results are good, but it is not a guarantee that this will work forever.

The continuous acceptability score is used/analysed in two ways. One is the study of distribution using histograms and boxplots. The other one is through a dichotomization of the acceptability and use it as an outcome for the binary logistic regression. • For the study of distribution, as in figure 3, I would suggest a cumulative counts plot. On such plot, you could put counts on the left y-axis, and percentages on the right y-axis. Heights are easier to read than histogram areas. • About the “least 10% of acceptability” concept. Can we call this the lowest decile acceptability score? • Figure 4 is successfully to show the distribution of the acceptability score for a few key variables. However, because it is based on quartiles, it fails to tell us the proportion of observations below the line of the overall lowest decile. I suggest adding a table with such proportions.

1. About the cascade. Table 2 is accurate and quite informative. Figure 2 is a bit deceptive, because it suggests that the denominators for each step are accurate [which is not true for the last step, for example]. So I suggest i) to keep table 2 and ii) compute accurate proportions (in percentages) and add 95% confidence intervals for the cascade plot.

R1:

Reviewer #1:

• Authors have responded to my comments

• Given the noted limitation of not being able to demonstrate comprehensive acceptability (without refusal rates) – it’s advised to further review the paper for instances of potential overstatement of findings. For example, on page 20. “…while both…facilities demonstrated high engagement…” – 1,000 clients is definitely a lot, but without the context of how many overall clinic clients there are, and how many may have refused, it would be prudent to temper the statements slightly.

• Importantly, the additional information provided by the authors about their bespoke acceptability measure gives me greater reservations – considering it has no prior validity testing. This is not addressed in the limitations and definitely should be.

R2:

Reviewer #1:

Thank you to the authors for addressing my comments.

R0:

Reviewer #1:

the authors have done appreciable work in collecting data on the health care issues wrt to Rabies prevention in Tanzania. I have the following observations. Some more statistics ie tables etc may be added to the article to justify the results and discussion. The tables should then be discussed in detail with conclusion pointing out the steps needed to be taken to improve the prevalent chaos in their health care system. costs per patient could be estimated inclusive of loss of wages etc. this will further improve the impact of the article. I couldnt open the supplementary GITHUB site for more results as it seemed to be a paid site.

Reviewer #2:

REVIEW of the article "Access to Rabies Post-Exposure Prophylaxis in Tanzania: A Mixed-Methods and Theoretically Informed Study to Inform Policy and Practice"

This article is a comprehensive study addressing a critical global public health issue: access to rabies post-exposure prophylaxis (PEP) in Tanzania. The study adheres strictly to scientific research principles and is highly relevant, clearly structured, and methodologically rigorous. The use of mixed methods (quantitative analysis of large-scale data from the Integrated Bite Case Management (IBCM) platform and qualitative analysis of healthcare worker communications) within a theoretical model of access to healthcare is a key strength of the study, allowing us not only to document the facts but also to understand the mechanisms and context of barriers. This article makes a significant contribution to both the academic understanding of health systems issues in low- and middle-income countries and to intervention planning, particularly in light of the launch of funding for PEP programs by GAVI (the Vaccine Alliance).

The relevance of this study is undeniable. Rabies remains a deadly, yet 100% preventable, disease that claims tens of thousands of lives annually, primarily among the poorest rural populations in Africa and Asia. The announcement of the "Zero by 30" goal and GAVI's decision to support the procurement of rabies vaccines create a historic window of opportunity. At this moment, field-based evidence is crucial, demonstrating that the mere availability of vaccines in a country is insufficient. This study brings this issue to the forefront by systematically identifying a cascade of interconnected barriers.

The scientific novelty lies in the following:

Application and development of the theoretical model. The authors not only utilize Penchansky and Thomas's classic model (five dimensions of access), but also empirically substantiate and add a sixth dimension—"Appropriateness." This crucial conceptual addition highlights provider competence as a separate, critical access factor, as clearly demonstrated by data on the inappropriate use of RIG or tetanus toxoid.

Depth and Richness of Data. The combination of data on over 10,000 patients with an analysis of real-life healthcare worker conversations and concerns (>3,000 messages) provides a unique, comprehensive picture of the situation. This allows us to move beyond dry statistics ("5% experienced interruptions") to understanding the human tragedies and professional dilemmas behind these figures.

Documenting Emerging Threats. The study documents the emergence of a new, dangerous practice—the use of rabies immunoglobulin (RIG, Equirab) as a substitute for vaccination, which has resulted in fatalities. This is a crucial observation for surveillance systems and regulatory agencies.

The study's methodological framework is exemplary for public health research.

Design: Mixed methods are entirely justified for studying the complex phenomenon of access. The quantitative component objectively captures the scale of the problem (distances, completion rates, interruptions), while the qualitative component reveals the causes, context, and subjective experiences.

Theoretical Framework: The use of a structured conceptual framework (expanded healthcare access framework) lends systematicity to the analysis, prevents fragmented findings, and allows for the classification of barriers into clear categories, which is useful for developing targeted interventions.

Triangulation: The authors consciously employ triangulation, systematically comparing findings from quantitative and qualitative data. This enhances the validity and reliability of the results, allowing for identification of areas where the data complement or corroborate each other.

Ethics: The study has all necessary ethical approvals, and the process for obtaining informed consent from participants (healthcare workers) is described.

Despite its undeniable strengths, the study has a number of limitations, some of which the authors honestly and thoroughly describe in the relevant section. This enhances the credibility of the study.

The absence of a patient voice. This is the main methodological limitation, noted by the authors themselves. The study brilliantly analyzes the system from the service provider's perspective, but the "measure" of acceptability—that is, how much patients trust the service, how comfortable they are, and whether the services meet their cultural expectations—remains unexplored. This is an important gap, as cultural beliefs and stigma can be a significant barrier.

Sample bias. The data only cover those who reached official medical facilities included in the KUSU. The most vulnerable groups—those who immediately turned to traditional healers—didn't have the means to even pay for the first visit.

Reviewer #3:

The authors employed a mixed-methods approach, integrating quantitative data from the Integrated Bite Case Management platform with qualitative data from hotline calls and peer-support chats among health and veterinary workers, to identify barriers and facilitators affecting bite victims’ access to post-exposure prophylaxis across four regions of Tanzania. This work addresses a critical and timely public health issue, as inadequate knowledge of post-bite treatment and constraints in access to rabies biologics remain major contributors to preventable rabies deaths in dog-endemic regions, including Tanzania. I have just one major concern with the manuscript in its current version. While the authors correctly explain early in the Introduction (line 69 and Fig. 2) that post-exposure prophylaxis (PEP) comprises of immediate wound washing, a series of rabies vaccinations, and administration of rabies immunoglobulin (RIG) for previously unvaccinated patients with severe (Category III) exposures, this framework is not consistently or adequately reflected in the subsequent analysis and interpretation. Specifically, across four of the five dimensions of the expanded healthcare access framework, the discussion focuses exclusively on rabies vaccines, with little to no consideration of rabies immunoglobulin (RIG). For patients with severe (Category III) exposures, RIG is arguably the most critical component of PEP, and its omission substantially weakens the analysis. Issues related to availability, accessibility, and affordability should explicitly address RIG, as barriers to RIG access often differ from—and are more severe than—those associated with vaccines. Without clearly incorporating RIG into these dimensions, the manuscript provides an incomplete assessment of the factors influencing access to life-saving PEP. I understand the challenges surrounding RIG access, particularly as the authors allude to in line 363, where they note that the purchase of RIG had not previously been seen in Tanzania. Given this context, the implications of RIG unavailability should be discussed explicitly and in greater depth in the Discussion section. Doing so would help clarify how historical absence, procurement barriers, cost, cold-chain requirements, and limited clinician familiarity contribute to ongoing and future challenges in delivering complete PEP. A clearer, more focused discussion of RIG would strengthen the manuscript by providing a more realistic and policy-relevant assessment of barriers to comprehensive rabies prevention in Tanzania.

Academic Editor:

I would like to sincerely apologise for the delay you have incurred with your submission. It has been exceptionally difficult to secure reviewers to evaluate your study. We have now received three completed reviews; the comments are available below. The reviewers have raised significant scientific concerns about the study that need to be addressed in a revision.

R1:

Reviewer #1:

good article

Reviewer #2:

This article's conclusions and table discuss the need to decentralize PEP services to priority facilities.

I suggest the authors provide more specific criteria or models. For example: "In each ward with a population greater than X and/or located more than Y km from an existing PEP site, at least one medical facility (dispensary or health center) should be designated for rabies storage and vaccination. The selection of these facilities should be based on accessibility to the public (24/7) and the availability of trained personnel." This transforms a general recommendation into a measurable standard.

Reviewer #3:

Thank you for addressing reviewer comments from the previous round.

R2:

Reviewer #1:

All comments have been addressed.

R0:

Reviewer #1:

This paper examines factors associated with Shigella-attributed diarrhea among children aged 6–35 months in Malawi, including a novel assessment of seasonal effect modification. The analyses are technically rigorous and appropriately applied to the observational dataset, and the findings provide valuable evidence to guide targeted interventions, including the forthcoming Shigella vaccine rollout. I recommend publication pending a few minor revisions noted below.

• The introduction explicitly situates the burden of Shigella in an economic context, but are there any other lenses, perhaps more human-centric, through which we can think about the implications of this burden? • Please include references for the categorization methods of diarrhea severity and WASH in the “Predictor Variables” section of the Methods. • How did you approach producing age group bins for this study? Was it data driven or decided a priori based on some contextual motivation? • Please add a statement justifying Poisson regression as the chosen analytic method. • Given that some samples were tested by culture, some by qPCR, and some by both it would be beneficial to add more clarification in the methods about the different testing procedures and results classification. For the samples tested by both methods, what happened if one result was positive and the other negative? In the discussion you state “Notably, 43% of qPCR-positive cases were also culture-positive, supporting the clinical relevance of the qPCR-detected cases”, however only 43% overlap between the two methods actually seems quite low – can you point to any other studies that have looked at this? • I believe when using generalized estimating equations (GEE) to account for clustering it is standard to report the number of clusters and distribution of cluster size. • These analyses rely on an assumption of missing data at random/completely at random, however the complete case analysis conducted excludes 32% of children with missing vaccination data. Please elaborate on this missingness in the limitations beyond reduction in sample size to include the potential bias introduced if the data is not in fact missing randomly. Alternatively, it may be worthwhile to consider inclusion of the observations with unknown vaccination status as a third category – which may still have relevant interpretation given the reality of often not knowing children’s vaccination status when designing interventions. • Was there any consideration of prior antibiotic use among patients reporting to the clinic with diarrhea? Please elaborate how this may or may not influence these results (perhaps in the limitations). • Please standardize spelling of “enrollment” throughout the manuscript (sometimes one vs two ls). • In Table 1 the Wasting “None” group percentage needs a decimal instead of a comma.

Reviewer #2:

The findings are interesting but it need through revision considering the following critical points. • Clearly mention the inclusion and exclusion criteria for selection of patients in the current study? • What was the limitation of the study? • Shigella were isolated and identified using culturing. What was the specie distribution of Shigella? • Mention the duration of the study (months/years) in the abstract. • Briefly describe how the culture and qPCR were used for detection of Shigella. Is Shigella DNA directly detected in fecal sample or it is detected from culture? • Define the criteria of Household drinking water source categorization: Improved, Unimproved?? • The abbreviations used in the tables should be defined in the table’s foot note.

Academic Editor:

Two reviewers have evaluated your manuscript and provided their comments below. In particular, please provide more detail in the methods section on the microbiological methods for Shigella detection and add information on any missing critical variables e.g. in the table footnotes. In addition and given the high missingness for vaccination status, conducting a sensitivity analysis for the multivariable models while excluding vaccination would aid with the interpretation of the study findings.

R0:

Reviewer #1:

Thank you for the opportunity to review this manuscript examining determinants of measles vaccine (MR1–MR2) dropout in an urban slum setting. The topic is highly relevant to measles elimination efforts, particularly in vulnerable urban populations where service continuity remains challenging. The study addresses an important operational question and employs an appropriate case–control design. However, several aspects of the manuscript require clarification and strengthening before it can be considered for publication.

Introduction 1. The Introduction does not crisply say what is unknown about measles dropout in the Dhaka slum context. You need a single paragraph that: (a) identifies the knowledge gap, (b) explains why Korail/slum populations require focused study, and (c) states the study’s unique contribution. 2. Terminology and consistency. Use consistent vaccine labels (Penta1, MCV1, MR1, MR2). Early in the Introduction define what you mean by “drop-out” (the operational definition appears later in Methods — bring a short definition into the Introduction). 3. Some background facts belong in Methods/Results. E.g., the specific urban slum population size and annual vaccination target are Methods details and should not be in the Introduction. 4. At present the Introduction jumps between global stats, EPI history, WHO recommendations, and local data without clear transitions. Reorder so each paragraph follows logically (global → regional → national → evidence on determinants → gap in urban slums → study aim). 5. Missing justification of novelty. Say explicitly why this study adds new evidence (e.g., limited studies in Dhaka slums, few case–control analyses of MR1–MR2 drop-out in highly mobile urban slum populations, service-delivery factors not well quantified in Dhaka). This is essential for reviewers. 6. The intro cites Ethiopia, Somalia, Pakistan; add (or at least state you reviewed) South-Asia or Bangladesh-specific evidence on urban slums and drop-out. If those studies are sparse, say that explicitly — that is the gap.

Methods Section 1. Please provide a precise definition of dropout, including the age/time cutoff for MR2 completion and whether delayed vaccination was considered acceptable. 2. It should be explicitly stated that both groups were drawn from the same source population using identical eligibility criteria to minimize selection bias. 3. Provide more detail on the sampling frame and recruitment process. Specify whether participants were identified through EPI registers, household listings, or community census data, and clarify whether random or convenience sampling was used. 4. Expand the description of the sample size calculation. Include the assumed effect size (odds ratio), exposure prevalence among controls, alpha level, statistical power, and the formula or software used. 5. Clearly define how key variables (e.g., maternal education, household income, waiting time, ANC/PNC visits) were categorized and justify the chosen cutoffs. 6. Clarify the source and validation of vaccination data. Indicate whether vaccination status was verified through vaccination cards, caregiver recall, or both, and explain how discrepancies were handled. 7. Provide more detail on bias control measures. Describe steps taken to minimize selection bias, recall bias, and information bias, including interviewer training and standardization procedures. 8. Strengthen the description of data management and quality control. Indicate whether data were double-entered, validated, and which statistical software (including version) was used. 9. Clarify ethical procedures. Provide the name of the approving ethics committee, approval number, and details on how informed consent was obtained and documented.

Results Section 1. Strengthen the presentation of multivariable findings. Adjusted odds ratios (AORs), 95% confidence intervals, and p-values should be consistently reported in both tables and narrative text. Emphasize adjusted results over crude associations to avoid overinterpretation of unadjusted findings. 2. Clarify reference categories in tables. Tables presenting logistic regression results should clearly indicate the reference group for each categorical variable. This is essential for accurate interpretation of odds ratios. 3. Improve consistency between tables and text. Ensure that all key numerical results mentioned in the text match those in the tables (including decimal places). Avoid repeating entire tables in narrative form; instead, summarize the most important findings.

Discussion and Conclusion 1. Strengthen the synthesis between findings and existing literature. While prior studies are cited, the discussion would benefit from more critical comparison. Clearly indicate whether your findings confirm, contradict, or extend previous evidence, particularly in South Asian or urban slum contexts. 2. Deepen interpretation beyond statistical significance. Move beyond reporting that associations were significant and elaborate on potential mechanisms (e.g., health system barriers, caregiver perceptions, structural inequities) that may explain the observed relationships. 3. Expand the discussion of public health and programmatic implications. The manuscript would benefit from clearer operational recommendations. For example, explain how EPI managers or urban health planners could translate these findings into targeted interventions. 4. Explicitly discuss potential biases (selection bias, recall bias, residual confounding), the inherent limitations of the case–control design, and issues of generalizability beyond Korail slum. 5. The discussion should more explicitly state what new knowledge this study adds compared to prior research—particularly regarding urban informal settlements and MR1–MR2 dropout.

Reviewer #2:

Beyond the reduced confirmed incidence of measles from 2019 (31.4) to 2022 (1.34 per million), at national level needs to include the measles case- based surveillance sensitivity indicators if they are met (Non measles febrile rash illness rate and % specimen collected from suspected cases ) Measles case-based surveillance indicators if achieved or not need to be described for the period of the study 15/10/2023 to 30/04/2024 for Dahka

Reviewer #3:

In this manuscript, Alam et al. report a study investigating factors associated with measles vaccination drop-out in children aged 12-23 months in urban areas in Bangladesh. It is quite interesting that the author found that maternal occupation, birth order, and waiting time for vaccination were among factors that were significantly associated with vaccination drop-out. Overall, the manuscript is modestly presented. My comments are summarized below. 1. What are MCV1, MCV2? 2. What is included in a pentavalent dose? 3. The statement in lines 53-54 needs to be supported by a reference. Particularly, to better protect children against measles, a herd immunity level of 95% is needed. And if the statement remains correct, what is the problem when the drop-out rate in Bangladesh is 3.4-5.5%? 4. The differences in vaccine coverage between urban and rural areas (lines 61-62) are significant? And is it due to vaccination drop-out? 5. Paragraphs in lines 53-60 and 63-70 are duplicated. 6. I think the introduction needs to be revised. It remains unclear why this study is conducted. 7. If possible, please make the questionnaire available for review and describe how the questionnaire is evaluated. 8. Some language errors in Table 4. 9. The conclusion of the manuscript introduces new information (measles elimination in Bangladesh), but it seems likely irrelevant.

R0:

Reviewer #1:

In the abstract: - please use the background, methods, results, conclusions subheadings. Indicate more clearly that this is not the results of a trial, but an examination of baseline data among trial participants for a different purpose. It is unclear whether there are 12,000 participants who provided data, or if only 1,600 provided data for analysis? How many patients in the V-CaN hybrid type 2 implementation effectiveness trial? What is the main aim of this trial and when will that be reported? "We have also created a best-fit regression model for risk factors as outcome variables" What are the outcome variables? predictors? Please lead the reader through to understand why this was done. - it is not clear what these abbreviations signify: RKSK, RBSK, NP-NCD, NFHS. -was the prevalence of smoking only 0.3%?

line 34: don't use DM without explaining it. line 35 HTN? is this the same definition of high blood pressure mentioned in line 39? If so, use same terms. Why higher percentages of high blood pressure in 10-12 year olds compared to 13-19 year olds?

line 93: was the data cleaned using R studio? How?

line 109: (perhaps give context to 600 met-minutes - i.e. ~ an hour of running, for example, if appropriate.

line 120: Capitals for American Academy of Pediatrics.

Table 1: is Other backward class the best descriptor?

Could there be a figure to show the predictors of each outcome?

line 227: should be 0.50

line 227: compared to a farmer and laborer

Instead of gender, should it be sex?

Were the blood samples only tested for blood glucose status?

Reviewer #2:

Overall relevance and significance

The topic is highly relevant, as studying NCD risk factors among adolescents is important because this age group is particularly vulnerable to commercial determinants of health.

Adolescence represents a critical window for early prevention, making the study timely and valuable.

1. Clarification on participant flow

Please clarify what happened to participants who consented for testing but were not included.

You mention that ~11,000 agreed, but the final sample was 1,600.

A clear flow diagram (screened → consented → included → analyzed) with reasons for exclusion would improve transparency and reduce potential selection bias concerns.

1. Regression modelling approach

In the prediction models, consider including only biologically plausible predictors for each outcome.

For example, for stress, variables such as psychosocial factors may be more appropriate.

For salt intake, predictors like dietary patterns or processed food consumption would be more relevant.

Some associations (e.g., smoking with higher salt intake, or stress with vegetable intake) may not be conceptually necessary and could introduce overadjustment or model instability.

1. Variable selection strategy

Instead of stepwise methods, an enter (forced entry) method based on conceptual framework is preferable.

This helps avoid data-driven selection and preserves meaningful associations grounded in theory.

1. Interpretation of findings

The discussion should expand on the biological and behavioral plausibility of key associations.

For example, explain why salt intake might relate to physical activity (e.g., dietary patterns, sports drinks, higher caloric intake).

Similar explanatory reasoning with supporting literature should be provided for other significant findings.

1. Strengthening the discussion

Link findings more explicitly to existing evidence and mechanisms, not just statistical significance.

Highlight how results inform adolescent NCD prevention strategies or policy implications.

Academic Editor:

Thank you for this timely paper. My comments: 1. The Introduction is well-written in terms of presenting the burden of NCDs and citing national datasets such as NFHS and CNNS. However, it may be helpful to strengthen the rationale slightly to more clearly justify the need for conducting this large baseline assessment.

1. At present, the paragraphs primarily describe existing prevalence data. It would add clarity if a few sentences explicitly explained how this study differs from or adds to existing datasets, particularly given that CNNS and other national surveys have already reported adolescent risk factors. Clarifying what specific gap this study addresses (for example, extension of the age range to 30 years, district-level HDSS-based profiling, integration of behavioral and biochemical measures, or linkage to the implementation arm of the V-CaN trial) would strengthen the justification.

2. Introduce your study clearly and boldly, early introduce your protocol paper and then it may be useful to state clearly that this manuscript presents baseline findings intended to inform and tailor the implementation component of the hybrid trial. Making this linkage more explicit would help readers better appreciate the added value and purpose of the baseline survey. Otherwise, readers may be high with the thought that this study would give us something from A hybrid type 2 implementation effectiveness trial, whereas, in current form it is just a cross-sectional study.

In Methods 1. I would suggest clearly adding standard sub-headings for easy reading. 2. Given the number of variables assessed in this study, it may be helpful to consider presenting a consolidated table in the Methods section summarizing the operational definitions and measurement approaches used. This could include definitions of outcome variables, biochemical assessment methods (including laboratory techniques and cut-offs), procedures for estimating salt intake in grams per person per day, and the method used to assess fruit and vegetable servings and physical activity. 3. 10-20 and 20-30 years: rationale for these categories. consider aligning with The World Health Organization (WHO) or similar body, that provides specific definitions for different age groups within the young population. According to WHO, adolescents are defined as individuals between 10 and 19 years of age, young people encompass those in the age group of 10-24, and youth are categorized as those aged 15-24. Currently we are overlapping.

Results 1. Table 3: Association of stress with socio-demographic and other behavioral characteristics among 10-30 years old persons. Why are we presenting this as a separate table, as it was less talked about while building the rationale. Need more clarity on this. 2. table 4: only talks about Smokeless tobacco. rationale? 3. while doing regression, please recheck multicollinearity, as currently there are so many variables, that do not fit in causal pathways. example: role of marital status while doing regression for salt and sugar. Tables in general are very cluttered, that affects readability and making inference.

Discussion 1. Please add 4 key messages from your study, that would help u during the intervention or would inform policy, and some recommendations at the end of the discussion.

R0:

Reviewer #1:

Minor comments: 1. In lines 172 and 179, PCV13 is incorrectly written as “VCP-13” and “VPC-13” and should be corrected.

1. The abbreviations PCV-13 and PCV13 are both used in the manuscript. Authors should be consistent and use a single form.

2. The title of Figure 2 contains an error, it reads “stautt” instead of “status”.

3. The WHO reference used for the pneumonia case definition links to a French-language document. Authors should change this to a link to an English version of the document.

4. The endpoint of the Cox analysis is time to hospital discharge; however, the authors use different terms to describe this outcome (e.g. “recovery”, “released cured”, and “discharge”), consistent terminology for time to hospital discharge would improve clarity.

5. The study does not capture information on siblings and their vaccination status. As vaccinated siblings may provide indirect protection (on a household-level), this unmeasured factor could influence disease severity and time to hospital discharge and should be mentioned as a limitation.

Major comments: 7. Although a dose–response analysis is performed in a secondary linear regression analysis, the primary Cox model groups children with one or two PCV13 doses together with unvaccinated children. As children with one or two doses are likely to have some level of protection, this grouping may make the effect of full vaccination appear smaller. The authors may consider a sensitivity analysis using separate vaccination dose categories in the Cox model.

1. Please clarify whether repeat hospitalizations of the same child were possible during the study period and, if so, how this was handled in the analysis, repeat hospitalizations are often not random and could influence the estimated effect of vaccination.

2. As the study includes children from 3 different hospitals, it would be helpful to clarify whether admission practices and discharge criteria were comparable across sites, and whether differences were considered in the analysis, as these could influence time to hospital discharge.

R0:

Reviewer #1:

What are “cover clothes “ ? It’s interesting that the majority state that they would take their child to a clinic for symptoms of scabies . Does this include taking to children to traditional healers as well as a government or private medical clinic? I am trying to reconcile this finding with the section that shows support for traditional treatments rather than orthodox treatments. Likewise does this question “Seeking medical help for scabies is unnecessary” include traditional healers ? “Scabies treatment is too expensive for most families in our community” . It would be informative to explain this further in the discussion – roughly how much does the treatment cost in this setting It is interesting that mothers achieving a higher level of education were also amongst those with poorer practices. Have you any explanation for this ? Is there a role for the “expert” patient in prevention ?

Reviewer #2:

The manuscript meets the publication criteria for PLOS Global Public Health. It contributes to existing knowledge gaps, and the results align with and support the stated research objectives.

Regarding the methodology, the author presents the overall approach clearly; however, a few areas require clarification:

1. In the Materials and Methods of the manuscript, the author provided that all mothers of under-five children who consented to participate were included in the study. However, the sampling technique used is not clearly described. It is unclear whether the author employed simple random sampling, purposive sampling, or another approach. Clarifying the sampling method is important for understanding the representativeness of the study population and the generalizability of the findings. So, there is a need for clarity.

2. In the manuscript, under study variables, it states that participants 0–5 were classified as having poor knowledge and those scoring 6–9 as having good knowledge. However, the results table presents only item-level frequencies and percentages for the nine knowledge questions. While this shows awareness of individual items, it does not allow readers to determine overall knowledge levels. The table does not reflect the good/poor knowledge classification described in the methods, making it difficult to assess the overall knowledge gap on scabies. I recommend revising the table or adding a separate summary table that reports the proportion of participants with good versus poor knowledge to ensure clarity and consistency between the analysis and the results. I suggest the author extends this modification to attitude and practice.

For the statistical analysis section, it clearly presents frequencies and percentages and attempts to identify socio-demographic variables associated with KAP. However, additional clarification is needed regarding the analysis of the attitude score. In the manuscript, under the study variable, it states that attitude was measured using a 14-item Likert scale, with total scores of 0–8 categorized as negative attitude and 9–14 as positive attitude. Since Likert responses are non-dichotomous, it is unclear how individual item responses were coded to produce these totals. Clarifying the scoring method used to convert the Likert responses into these two attitude categories is necessary, as it may have influenced the study’s findings.

For the availability data underlying the findings in the manuscript, they were fully available and easy to follow.

Reviewer #3:

This paper examines the knowledge, attitudes, and preventative practices toward scabies infections among mothers (N = 320) of children under the age of five in Ibadan, Nigeria. The findings are nuanced and show that knowledge about the infection is mixed, and practices are generally poor. The results will be useful to conduct targeted educational campaigns to improve mothers’ knowledge, attitudes, and practices toward scabies infections in the region. Overall, I feel positive about this paper, but I have several issues that need to be addressed before publication.

Major points: - Attitudes scale: what is the meaning of “traditional” vs. “orthodox” treatments? Do participants know the difference? To me, they sound the same. - The Attitudes question about the effectiveness of traditional vs. orthodox treatments seems rather a knowledge question because it seems to have a clear True/False answer. - “Once a child has had scabies, they become immune to future infections.” This is a knowledge question with a clear True/False answer, not an Attitudes question. - Why are the parts of the questionnaire in the Appendix E and F not reported in the Results? The questions look interesting.

Minor points: English: - In the title (and throughout the paper): write “attitudes” instead of “attitude.” - Throughout the paper, instead of “Majority of the” write “Most.” - Whenever you report results, you need to put the numbers and percentages in parentheses. For example, “A large number of respondents (254, or 79.3%),” and so on.

Regression tables: - Instead of including the baseline category under the list of coefficients with a 0 next to it, remove each baseline category and instead write a note at the bottom of the table where you indicate the baseline category for each variable.

Academic Editor:

Thank you for sharing this interesting manuscript. Kindly carefully review suggestions from our reviewers. One of the main issues is the English and academic writing that should be revised throughout the manuscript. In addition, please provide point-by-point information to indicate your revisions per individual comments, including line numbers and pages.

We hope to receive your revision soon.

R0:

Reviewer #1:

Methodology: study design & datasource, heading overlap with data source and extraction. You may consider differentiating the study design+population from the data source+data extraction to improve clarity and avoid repetition.

The covariates are presented in a table. Consider adding a concise paragraph describing key variables and their categorization. Perhaps detailed coding and operational definitions should be presented in a supplementary table.

The analysis plan refers to Objective 2 and Objective 3 however, the study objectives are not stated anywhere earlier in the manuscript. They should be explicitly stated to improve coherence

Also, this contradicts “shared frailty at the individual level”. Please clarify the intended clustering structure. Did you model with the shared or Individual-level frailty?

Reviewer #2:

This manuscript addresses an important and under-researched area by examining long-term trends in attrition from HIV care and associated predictors among adolescent girls and young women (AGYW) with non-viral load suppression in Tanzania. The use of routine programmatic data is a major strength and enhances the relevance of the findings for the national program. The objectives are clear, and the results provide insights into patterns of disengagement from care in this vulnerable population. However, the manuscript would benefit from clearer operational definitions of key outcomes particularly inconsistent viral load suppression. Further discussion linking the findings to existing regional and global literature, as well as clearer description of programmatic and policy implications, would strengthen the paper. I recommend minor revisions to improve clarity and consistency.

R1:

Reviewer #1:

All comments have been addressed

Reviewer #2:

General review of the revised manuscript I have reviewed the revised manuscript and found that the authors have addressed all the comments; however, a few minor edits are still required before publication. Please see the table below.

Section: Introduction <br /> Comment: Correct spelling UANIDS <br /> Line number: 58

Section: Methods <br /> Comment: It is essential that the term inconsistent viral load suppression needs to be clearly defined in the manuscript. Note that two viral load results ≥1000 copies/ml at least six months after ART initiation is the standard definition for non-viral suppression or ART virological failure.<br /> Ideally, inconsistent viral load suppression would have been a situation whereby PLHIV achieves an undetectable viral load at least 6 months after ART initiation, but subsequently experience detectable viral load result/s ≥1000 copies/ml. To avoid confusion, refer to this sentence in conclusion: Notably, individuals with an initial unsuppressed viral load were more likely to disengage from care.

Section: Results <br /> Comment: Table 1: First line and second line not defined. Not every reader knows what Tanzanian ART regimens. Line number: 170-171

Section: Discussion Comment: A citation is not inserted <br /> Line number: 253

Section: Discussion Comment: Correct the word form to read from Line number: 254

R0:

Reviewer #1:

The article is relevant and timely. Please use COREQ guidelines/other community guidelines for qualitative research listed on the EQUATOR network to improve the rigour and flow of the article. Additional suggestions for your consideration are listed below. 1. The rationale for the article as well as research question needs to be made more explicit. Additionally, the novelty of the article is relatively unclear. 2. The use of a framework devised for infectious diseases needs to be examined- why was this framework chosen? 3. Line 52-54: It may be useful to examine the cause of death in context of age; additionally, the later use of terms like young-old-middle aged are ambiguous later in the article, please provide an operational definition. 4. Line 54-57: The reason for prioritising lived experience is unclear. The logical flow establishing the rationale for this inquiry needs strengthening. 5. The linkages of NCDs with urbanisation need more citation and can be explored better. The authors may consider reassessing the introduction section which seems to appear definitive without providing a research gap. 6. Why is this research needed? What makes it relevant for global health? This needs to be elucidated in the text. 7. The use of the word "drug peddler" should be in quotes at least at first use, to ensure that it is seen in its local context. 8. The narrative findings are rich and well presented- the global health impact of these findings/future paths based on these findings need to be be made explicit.

Reviewer #2:

This manuscript is well-written and engaging, and I would like to commend the authors for presenting such rich and dense information.

Here are my comments, which I believe will enhance the content of this manuscript.

1. Page 2, line 27 (abstract section): The term "oppression" appears quite strong, and I do not find any evidence supporting it in the results section. Could the authors clarify what they mean by "oppression"? Additionally, the sentence “as gender norms and experiences of masculinities and femininities reflected the division of roles and access to resources by men and women, which in turn shaped their ability to seek early and better healthcare interventions” seems to contradict the notion of oppression if participants are able to seek early and improved healthcare interventions.

2. Page 6, line 133: typo: a dot is in the middle of the sentence. Consider removing.

3. Page 7 on selection of study participants: How did researchers identify participants based on diabetes and hypertension criteria, given that these medical conditions require a confirmed diagnosis from health professionals? Additionally, were any participants identified using a snowball sampling technique?

4. It is mentioned that this study was conducted as part of the broad ARISE project. Can the authors clarify whether the study participants were enrolled into the main project and if so, to what extent this initial involvement of the main study had impacted the data collection?

5. An important section on data collection is missing. I would like authors to include this in order to explain what specific data were collected at each round, why was it necessary to split data collection over different timepoints, where and in which language were data collected, how long lasted interviews on average, and who collected the data, any bias that may result from the way the study participants perceived the data collection team during fieldwork?

6. The data analysis section requires further clarification. It would be beneficial if the authors could elaborate on their data coding strategy. It appears that a deductive approach was employed, yet they did not explain how the coding frame was initially developed before its application to the data, nor did they specify who conducted these procedures. Additionally, information regarding the software used to facilitate the data analysis is missing.

7. The use of relevant qualitative data reporting guideline is missing. This is important to include as it will help ensure that ensure that the methods and findings are clearly and transparently communicated in the manuscript.

8. In relation to adapting the framework for Infectious Diseases of Poverty Research, it would be beneficial if the authors could clarify the specific adaptations made to the framework and how it was applied during data analysis.

9. They mentioned that participants who faced distress during the interviews were allowed to grieve and were consoled. How many participants developed such condition during interviews at which round of data collection, and to what extent did this impact on data collected?

10. This study seeks to explore the lived experiences of individuals affected by NCDs in informal settlements. However, the findings presented in the results section, particularly the initial part titled "lived experience of NCD," fall short of adequately conveying the participants' current experiences with these health conditions. Firstly, the authors include excerpts from three informants but fail to demonstrate how these accounts are representative of all study participants. Secondly, the discussion of informants' lives prior to developing their current health conditions renders the information less pertinent to the study's objectives. It would be beneficial if the authors could focus on participants current lives to discuss the intersectionality of NCDs with social, economic, and gender factors within the participants' living environments, and how these elements contribute to shaping the management of their health conditions. This way of presenting the findings restricts our understanding, as the same observation is reiterated throughout the discussion section.

11. In the context of chronic disease, it can be assumed that participants with diabetes, hypertension, or comorbidities might encounter similar yet distinct challenges. It would be beneficial for the authors to discuss the findings across these cases, highlighting both similarities and differences, and to establish a hierarchy within the data.

12. The use of terms like "people," "women," or "men" in the results section can be misleading, as they might suggest that the findings apply to the general population. Instead, it is advisable to use terms such as "informants," "study participants," or "men" or "women in this study" to ensure that the information reported is limited to those individuals who participated in this data collection.

R0:

Reviewer #1:

I reviewed the manuscript titled ‘Awareness of anthrax disease and the knowledge of its transmission and symptoms identification: a cross sectional study among butchers in Ile-Ife’ and below are my observations: Abstract The authors were not consistent in the use of tenses. They should stick to the use of past tenses or continuous tenses. In line 43 they mentioned ‘were not aware’ but in 45 used ‘are aware’. Aware ‘of’ is more preferable than aware ‘about’. They were also not consistent in the use of singular or plural words. For instance, in line 46, they referred to ‘animal’ and then ‘animals’. They can stick to the former as a collective noun or the latter as plural. The authors’ conclusion that there was poor awareness is not consistent with the fact that there was no mention of scoring the questionnaire in the methods. In addition, the lack of mention of knowledge which is a part of the title in the abstract leaves a major void. Knowledge should be captured both in the methods and results. Otherwise, it should be deleted from the title. Introduction Remove the space between the punctuation and the citation in square bracket see lines 63, 79 and 112, for instance. Anthrax is a disease (not an agent) caused by Bacillus anthracis not ‘Bacillus Anthrax’. Line 75: Change ‘buy’ with ‘but’. Authors should stick to ‘human or ‘humans’ (Lines 74 and 76). The transition from one paragraph to the other is not smooth. For instance, lines 79 and 80 do not have any connection. In fact paragraph 4 has no place in its current position in the background of the study. The first sentence of paragraph 5 makes no sense and should be recast. The statement that the first case of anthrax in Nigeria was in 2023 is a lie! The authors should make use of the correct words: For instance, ‘body openings’ should be referred to as ‘body orifices’. Line 104: The authors seem to have changed from reporting a study to writing a proposal. Lines 105 – 108 require referencing. Lines 111 and 112: Change ‘has’ to ‘have’. Line 118: ‘Osun - State, South-Western’, should be written ‘Osun State, southwestern’. Line 121: Change ‘it's’ to ‘its’. Line 125: Change ‘accommodates’ to ‘accommodate’. The authors should draw a map of the area being described. Line 134: Change ‘accommodates’ to ‘accommodate’. The inclusion and exclusion criteria are confusing: ‘Butchers who were standing were included’. Does this mean that those butchers who were sitting were excluded? ‘Butchers who were sick were excluded’, therefore only the healthy were included. How did the authors identify healthy and sick butchers? What was the reason for the chosen criteria? Lines 140 - 149: The fact that the authors are contemplating calculating sample size and recruiting respondents for a study that is concluded is absurd. Lines 158 and 159: Change ‘was’ to ‘were’. Line 161: Which random sampling did the authors use to select the butchers? They should be specific in explaining how the sampling method used was applied in the work. The sampling was not well explained. The authors claimed to have used multi-stage sampling technique. What was the sampling frame in each stage and how were they prepared or accessed? The authors did not explain whether and how the questionnaire was validity and reliability tested. Was there a pilot study and how was it conducted? Without these, the instrument is not valid for a meaningful research. How many sections did the instrument contain and how many questions were there in each section? Was the questionnaire scored and how was it scored? Without scoring the questionnaire, the authors cannot determine the levels of awareness or knowledge. Results Line 167: Change ‘questionnaires’ to ‘questionnaire’ because only one questionnaire was used for the study. Line 174: Change ‘rare’ to ‘rear’. The authors claimed to have worked with only commercial butchers. How come most of them have become cattle rearers? Table 1: Change ‘live stock’ to ‘livestock’. Line 180 – 181. Recast the sentence. Discussion The authors seem not to understand that there is a difference between knowledge and awareness, Therefore, it may be worthwhile to define both words in the work. The authors’ claims on knowledge level may not be correct without scoring knowledge to determine the level among the participants. Comparisons made with some author, example Cadmus et al. (2024) do not rhyme.

Limitations Secondary education is not low level of literacy. Therefore, an area that has more than 70% of the population that attained secondary education cannot be regarded as having low level of literacy.

Reviewer #2:

The manuscript “Awareness of Anthrax Disease and the Knowledge of Its Transmission and Symptoms Identification: A Cross-Sectional Study Among Butchers in Ile-Ife” addresses an important public health issue, the awareness and knowledge of anthrax among butchers, that represents a high-risk occupational category in Nigeria. The topic is relevant, timely, also including a One Health perspective. The study provides useful descriptive data that could inform public health education and prevention strategies. The authors present clear findings about the low awareness of anthrax among butchers and the implications for public health. The study is quite good in methodology and analysis, but there are different areas for improvement, before the manuscript could be considered for publication. As regards the study design, it would be appropriate to clarify the sampling frame, randomization process, and justify the adequacy of the final sample size. How were markets and butchers selected? Was any sampling frame used? No basic inferential analyses (e.g., chi-square tests) were performed to identify factors associated with awareness and knowledge. Could you please, add this kind of analyses? Despite collecting rich demographic data and having explicit research objectives to assess awareness, no inferential analyses examine whether education, income, or animal sourcing practices associate with knowledge levels. Chi-square tests and logistic regression would identify high-risk subgroups requiring targeted interventions, fulfilling the study's stated aims and enhancing public health utility beyond simple prevalence reporting The results could be enriched by analysing possible relationships between the respondents' demographics (e.g., education, age) and their awareness or knowledge of anthrax. This would help contextualize the findings. The discussion could be strengthened by addressing any potential biases in the study (e.g., recall bias due to the use of self-reported data). The paper could explore in greater depth why the awareness was low (e.g., cultural factors, lack of effective public health messaging). While the conclusion addresses the need for improved awareness, it could be more specific about what measures should be taken. For example, the authors could suggest specific communication strategies, community-based interventions, or policy changes. The figures could benefit from more detailed captions that explain what each figure illustrates beyond just showing the data. Furthermore, the manuscript contains grammatical errors and spelling mistakes throughout such as “symtoms instead of symptoms in the short title” “Bacillus Anthrax (line 67)” instead of Bacillus anthracis, “unclothed blood (line 83)” instead of “unclotted blood”. Unclotted is better than un-clotted. Line 217 4got? Correct. Write B. anthracis in italics all over the manuscript. Some sentences are unclear or repetitive, especially in the Background and Discussion sections, so a professional English language editing is recommended.

Academic Editor:

The reviewers have raised a number of concerns that need attention. In particular, they request additional information on methodological aspects of the study, additional analysis, a clearly defined research question, and improvements to the quality of the language used.

R0:

Reviewer #1:

The article “Profiling Zero-Dose Measles-Rubella Children in Zambia: Insights from the 2024 Post-Campaign Coverage Survey” addresses an urgent global health issue aligned with IA2030 and Gavi’s zero-dose priorities. The title is concise, descriptive, and fits the scope of PLOS Global Public Health (PGPH).

The study employs a cross-sectional, two-stage cluster survey following WHO guidelines, with robust sample size (n=8,634) and weighting for representativeness. Statistical analyses—survey-weighted logistic regression and confidence intervals—are appropriate. Ethical standards and data quality controls are well-documented. However, heavy reliance on caregiver recall (88.3%) introduces recall bias, and the absence of district-level disaggregation limits local applicability. The manuscript’s use of WHO standards and analytical transparency strengthens credibility.

It provides novel national evidence on MR zero-dose prevalence and systemic immunization failures in Zambia, filling a gap between administrative and survey estimates. The identification of access and awareness barriers (e.g., 42.6% unaware of campaigns) adds actionable insights for health The article follows a clear IMRaD structure with strong coherence between results, discussion, and policy recommendations. Figures and tables are informative, though data visualization could be simplified for readability. Language is clear and professional, though some sections (e.g., policy implications) could be condensed to reduce redundancy.

No conflicts of interest or funding bias reported. Data availability upon request aligns with journal policy, though full open-access data would enhance transparency. Overall, the manuscript is methodologically sound, policy-relevant, and well-aligned with PGPH’s thematic focus on equity and immunization coverage. Minor revisions are recommended—clarify recall bias mitigation, improve data visualization, and emphasize data accessibility. With these revisions, it is highly suitable for publication in PLOS Global Public Health.

Reviewer #2:

1. Is this a national wide surgvey? Please explain.

2. Does this survey cover whole population? or part of population? What is the percentage of coverage?

3. The survey is procpective study. How does it happen to miss the data? Please expalin.

4. The ststistical part need more elaboration considering the variables.

5. In discussion section, avoid bullet. Avoid the policy implecation rather right as paragraph.

6. Rewrite the conclusion. Avoid frequency, percent, only mentiont he fidnings in relation to the objective of the study.

Reviewer #3:

In the current era of changing global and public health landscape, this manuscript is very timely in helping Zambia to improve vaccination coverage and address the inequities that exacerbates children to miss vaccinations. The manuscript is nearly perfect for publication with exception of few editorial areas which I request the authors to work on before the manuscript gets published. The areas are highlighted below:

ABSTRACT Background  Line 22: I suggest to add the abbreviation MR in brackets after “Rubella”.  Line 23: I suggest to replace “and” with “can” between communities & sustain. Methods  Line 27: suggest to insert “was conducted from” before 27th & replace “-” with “to” between 2024 and 16th. Conclusions  Line 42: I suggest to write “RI” in its long form and the abbreviation in brackets.  Line 44: I suggest to edit “IA2030” to be “Immunization Agenda 2030”.

INTRODUCTION  Line 86: I suggest to insert the abbreviation “RI” in brackets after “immunisation” before “performance”.

METHODS Study Design  Line 97: I suggest to replace “Post-Campaign Coverage Survey (PCCS)” with its abbreviation PCCS.  Lines 98-99: I suggest to replace “Measles–Rubella (MR) Supplementary Immunisation Activity (SIA)” with the abbreviations “MR – SIA.

RESULTS Zero-Dose Prevalence  Line 172: I suggest to write “DPT” in its long form and the abbreviation in brackets.

DISCUSSION  Line 262: I suggest to replace “routine immunisation” with the abbreviation “RI”.

CONCLUSION  Line 325: I suggest to replace “routine immunisation” with the abbreviation “RI”.

R0:

Reviewer #1:

This sub study was nested in a factorial randomized controlled trial (RCT) in women aged 18–30 years. Participants included in this study were randomized to receive either a preconception intervention package or routine care until early childhood. The design strategy involved a reasonable sample size justification to show superiority. The sample needed for the study objectives was well justified with power considerations. However, the investigators do note that the sample size, while adequate for detecting moderate effect sizes, may have been insufficient to identify smaller but clinically meaningful differences. The descriptives are informative as seen in Tables 1 and 2.

1. Please define IQR in the footnote of Table 2 or put a descriptive section in the ‘Analysis Plan’ paragraph.

Generalized linear models (GLMs) with a Gaussian family and identity link function were used to estimate mean differences in CRP, AGP, IGF-1, and IGFBP3 concentrations. To estimate risk ratios for inflammatory status between infants in the intervention and routine care groups, GLMs with a binomial family and log link function were employed. Final models were adjusted for place of birth. There are several considerations needing clarification.

There are four endpoints. Therefore,

1. Some consideration of multiple comparison p-value adjustment should have been discussed.

Also, with respect to model content,

1. Exactly how was adjustment by birthplace incorporated into the models?

The overall conclusions follow from the analyses performed and results seen in Table 3. The strengths and limitations are reasonably described in the ‘Discussion’ section. As an added point, however,

4.There is a gap between the manuscript text and the supplement supporting information proposal Version 2.0. Was there any attempt to explore the mediation analysis discussed in that proposal?

Reviewer #2:

1. Overall Assessment This study reports a well-designed randomized controlled trial. It investigates the impact of an integrated intervention on infant biomarkers related to inflammation and growth like CRP, AGP, IGF-1, IGFBP3. The research addresses a significant question in maternal and child health. However, the discussion sections can be improved with detailed explanation on biological plausibility. Also, the implications of this study can be broadly elaborated.
2. Originality and Relevance The research topic appears to be original and highly relevant. The novelty in this study is integrated interventions across different stages right from preconception to 2 years of early child development. The intervention is policy-relevant and aligns well as per Goal-2 and Goal-4 of SDG-2030. The concept is innovative and similar integrated frameworks are reported in the literature. The specific distinct approach of this study needs to be articulated.
3. Scientific Rigor and Methodology This randomized controlled design follows standard protocols and manuscript is well-aligned as per CONSORT guidelines. Please elaborate on randomization process, blinding, and control of confounders. The sample size calculations appear to be powered for anthropometric assessments. For biomarker outcomes, sample size calculations need to be refined/justified.
4. Results and Interpretation The results of this study report no significant differences in biomarkers between intervention and control groups. The null findings can be discussed with possible biological explanations like timing of assessment, nutritional variability, breastfeeding. Subgroup analysis by maternal or infant characteristics can be helpful.
5. Discussion and Implications There is a scope to elaborate the discussion section by linking the pathways of maternal interventions with infant biomarker responses. Implications of this study for public health, including integration into maternal and child health programs, can be discussed highlighting the need for long-term follow-up.
6. Presentation and Clarity The manuscript is well-written and well-organized as per required guidelines. However, most of the references are quite older and references from 2022 onwards are missing. More recent Citations can be included from year 2023-2025.
7. Ethical and Data Considerations All the ethical procedures are described clearly including IEC and CTRI. Data availability through Open Access links is provided.
8. Conclusion and Recommendation This well-executed trial can be good evidence for understanding the biological outcomes of integrated maternal-child interventions.

Recommendation: Minor Revision.

Reviewer #3:

This study is a secondary analysis of the WINGS factorial randomized controlled trial evaluating the effects of a multidomain, integrated intervention delivered from preconception through early childhood on infant biomarkers of inflammation and growth (CRP, AGP, IGF-1, IGFBP3) at 3 and 6 months of age. This study links the integrated intervention to specific changes in inflammatory and growth-related biomarkers like CRP, AGP, IGF-1 and IGFBP3. The study addressed the biologically relevant and policy-important question related to early-life interventions in low-resource settings The findings indicate no significant differences in these biomarkers between the intervention and control groups, except for a transient decrease in IGFBP3 at 3 months, which was not sustained at 6 months. The authors conclude that while the intervention improved growth outcomes in the parent trial, it did not significantly influence early-life inflammation or IGF axis biomarkers. The manuscript is well-written, clearly articulated and follows the required CONSORT Guidelines. Major Comments 1. Rationale and Framing • Biological rationale connecting integrated maternal–child interventions (nutrition, WASH, psychosocial care) with the specific biomarkers studied (CRP, AGP, IGF-1, IGFBP3), needs clarity • Clarify why these markers and 3- and 6-month time points were selected, especially since primary growth outcomes were reported at 24 months in the main WINGS paper. • A concise conceptual model or figure showing hypothesized pathways could help readers follow the mechanistic logic. 2. Study Power and Sample • The power calculation is based on CRP only. Please justify the adequacy of the sample size for detecting meaningful differences in IGF-1 and IGFBP3, given their biological variability in infancy. • Power calculations are based on LAZ outcomes from the primary WINGS study rather than biomarker data. This needs justification. 3. Statistical Analysis and results • Tables 2 and 3 could be simplified to highlight group comparisons more effectively. • Adjustment only for the place of delivery seems limited. • The author may consider other covariates, such as mothers’ BMI, socioeconomic indicators, or exposure to infections, in the analysis. In case they are intentionally excluded from the analysis, explain their exclusion. • It would be useful to include effect size interpretation (e.g., percentage change or standardized mean difference) to better convey the biological relevance of null findings. 4. Interpretation of Findings • However, cautious interpretation of the null findings is needed. Aspects such as biological plausibility, contextual limitations, and future implications for longitudinal research require further elaboration. • The discussion acknowledges the absence of significant effects, but can be deepened if the authors discuss the following issues o Address low baseline inflammation as a potential ceiling effect. o Note that intervention effects might appear later in life (after 6 months). o Acknowledge that non-inflammatory mechanisms (caregiving, infection prevention, psychosocial stimulation) might explain the positive growth outcomes in the primary trial. • Expand the comparison with similar trials—such as SHINE (Zimbabwe), ELICIT (Tanzania), and MAL-ED studies—that examined inflammation and growth factor pathways. • The trial was conducted in a single urban Indian setting, which limits extrapolation to rural or diverse socioeconomic contexts. The discussion should acknowledge this limitation more explicitly and suggest strategies for replication in varied environments. 5. Policy and Program Implications • The conclusion is based on the non-significant findings of biomarkers. Whereas the short duration of biomarker assessment may oversimplify complex biological processes. More elaborate discussion is needed on possible confounders like infections, duration, and type of breastfeeding.

Minor Comments 1. Abstract: Conclude with a stronger statement about contribution: e.g., “These findings add to the understanding of biological mechanisms underlying integrated early-life interventions in LMICs.” 2. Tables: Present only adjusted results in the main text; unadjusted data may be submitted as supplementary files. Ensure all tables include units (mg/L, ng/mL) and consistent decimal formatting. 3. CONSORT Diagram: Please include the number of exclusions, losses to follow-up, and reasons for non-participation in Figure 1 for transparency. 4. Discussion: Add a short note acknowledging that biomarker variability in early infancy is high and may obscure subtle intervention effects. 5. References: Consider citing more recent literature (published within the last 3 years) that links microbiome–inflammation–growth relationships in infants. 6. Language and Formatting: Ensure consistency in abbreviations (e.g., IGFBP3 vs IGF-BP3). Use consistent phrasing for “preconception, pregnancy, and early childhood interventions, growth-related biomarkers, and growth factor profiles” throughout.

Overall Recommendations: Minor–to–Moderate Revision This is a robust, well-implemented study addressing an important mechanistic question within global child health. Although the results are null, they offer valuable insights into early-life biology and integrated program evaluation. Strengthening the biological framing, contextual discussion, and presentation of adjusted analyses will substantially enhance the manuscript’s impact and readability.

Guest Editor:

Thank you for your submission to PLOS Global Public Health. I appreciate the time and effort invested in conducting the study and preparing the submission.

After careful review and evaluation of the manuscript, we are sorry to report that we are unable to accept your manuscript for publication in PLOS Global Public Health, in its current form. While a few considerations factored this decision, our concern were the lack of novelty and too preliminary of results to guide clinical practice.

We recognize the importance of your work and its contribution to the field and feel that substantial revision of the above stated concerns might aid in curating the manuscript to meet the editorial scope of PLOS journals.

We sincerely appreciate your interest and hope that you will consider submitting your future works to PLOS Global Public Health.

R0:

Reviewer #1: Manuscript as reviewed meets PLOS Global Public Health publication requirements, the author(s) clearly presented the study background, methods, results, discussions and conclusion. My comments and revision request are minor formatting and suggested input. No ethics concerns at this time. Reviewer #2: This is a well-written paper with clear methodology. From the perspective of data science applied to public health, this manuscript does a great job of clearly discussing and defining its methodology, which are all the current best practices. Correcting for class imbalance was a good choice, given the low prevalence of EC in the survey population. The use of SMOTE on the training set only ensured minimal data leakage, and is the current best practice. Using such a large variety of machine learning models creates a challenge in describing each model well enough within one manuscript, and the author did a good job of balancing that challenge.

I only have a few minor suggestions toc clarify the methodology of the manuscript:

Please specify upfront how many observations were used in training and testing, and specify how many positive EC outcomes were included in the testing set. With such a low prevalence of a positive outcome in a relatively small set of observations, it is worth mentioning that there are perhaps only 10-20 positive outcomes being predicted in the test set. In the absence of weighting, it may be that characteristics of those few positive outcomes in test set are biasing the predictors, and this is worth mentioning.

Please discuss how the initial 38 variables were selected from the survey. If there was an initial expert judgment on inclusion into the variable set for feature selection, that should be mentioned.

Cluster design was mentioned in the PMA survey. This indicates that the survey includes survey weights of some kind. Please discuss whether those weights were addressed in the machine learning methods, or defend why they were not included in the model design. Survey weights can be included in machine learning models to make the predictors more representative of the population of interest.

In the discussion, please discuss the impact of low precision, where there were many false positives compared to true positives. While it is mentioned, there are consequences (e.g., loss of trust) for low precision prediction models in public health, and this characteristic of the findings could be discussed more.

Consider including a SHAP dependance plot, because potential interactions are discussed (e.g., knowledge and ad exposure) without showing evidence. A SHAP Dependence plot could take care of this.

Consider explicitly discussing the limitation of cross-sectional survey data used for prediction, where proxies were used in place of quantitative evidence (e.g., exposure to ads to proxy perceptions).

Overall, great work, timely, and well constructed. Reviewer #3: SEE word document attached with clear table

Manuscript Number: PGPH-D-25-01837 Review report

This manuscript demonstrates a significant strength in its application of advanced machine learning and Explainable AI (XAI) to address the critical public health challenge of low emergency contraceptive (EC) use in Ethiopia. By rigorously testing multiple models and using SMOTE to handle severe class imbalance, it identifies key modifiable predictors like primarily EC awareness and media exposure rather than static socioeconomic factors. The use of SHAP values transforms complex model outputs into actionable insights, revealing that knowledge gaps are the primary barrier. This approach provides a powerful, data-driven blueprint for designing targeted interventions, such as tailored media campaigns and improved health counselling, to effectively increase EC uptake and reduce unintended pregnancies. However, the following points may need to be considered, so as to improve the quality of the paper.

Topic/ subtopic Issue Suggestions Title: Predicting Utilization of Emergency Contraceptive Usage in Ethiopia and Identifying Its Predictors Using Machine Learning Redundancy. "Utilization" and "Usage" mean the same thing. Predicting the Utilization of Emergency Contraception in Ethiopia and Identifying Its Predictors Using Machine Learning. Affiliation Inconsistent institution name. on page 1 says "College of Medicine Health Science" while first page of manuscript is "College of Health Science". Use consistent affiliation name Abstract "Traditional analyses have struggled to identify complex predictors." For flow, consider: Traditional statistical analyses have struggled to… Abstract "with SMOTE used to address class imbalance" – Grammar: This is a dependent clause. It should be connected to the previous sentence. ..., and the SMOTE was used to address class imbalance. Abstract "Findings highlight that knowledge gaps, not poverty or access, are key barriers to EC use." – Clarity: "access" is vague. Be more specific. ...not poverty or physical access barriers, are key. Introduction Page 3: "moderate’s" Change to moderates ("the way the education level moderate’s religion-based stigma"). Introduction "drives excessive maternal mortality rates of over 500 deaths per 100,000 live births, drives poverty cycles, constrains girls' and women's educational and economic opportunities, and overwhelms poor healthcare infrastructures." – The word "drives" is used twice in close succession. ...contributes to high maternal mortality rates of over 500 deaths per 100,000 live births, perpetuates cycles of poverty, constrains... Introduction "is a central preventive intervention" is a crucial preventive intervention Introduction "the use of EC remains embarrassingly low" "Embarrassingly" is subjective and informal. ...remains critically low. Introduction "tempts women to shun services" Word choice not good. ...pressures women to shun services. Introduction "woefully underserved" Informal. ...significantly underserved. Introduction "yield the predictive resolution necessary" "Resolution" unusual in this context. ...yield the predictive accuracy necessary Introduction "vastness tests for fairness" – Phrase is unclear and likely an error. Correct the phrase to clarity Methods Data Source & Inclusion Criteria: The criteria for selecting the 2,334 women from the larger PMA sample of 8,943 are not explicitly stated. Was it a complete case analysis? This needs clarification as it affects the generalizability of the findings. Clarify if sampling was done or it was a complete case study Methods "The dataset demonstrates low overall missing data prevalence" –"Prevalence" is for diseases outbreaks. The missing data were minimal overall; Methods "offering robust classifier building while preserving real performance measurement." ...facilitating the development of robust classifiers while preserving a realistic assessment of performance. Results "nailing 17 true positives" Informal word choice. ...correctly identifying 17 true positives... Results "It manages this recall strength at the expense of precision, though, which sits at approximately 11%." – "Sits at" is informal. It achieves this high recall at the expense of precision, which was approximately 11%. Results "The most influential positive feature was “heard_emergency”, indicating awareness of emergency services has the greatest influence..." add which . The most influential positive feature was “heard_emergency”, which indicates that awareness of emergency contraception has the greatest influence... Results "This resonates with core assumptions of health behavior theories like the Health Belief Model, which posit perceived knowledge as a harbinger of action." "Harbinger" misused. ...which posit knowledge as a prerequisite for action. Results Page 18: "radio-implemented" Change to radio-delivered or radio-based. Results "Even positive, this reflects continued systemic disincentives documented elsewhere" – Unclear Even not a correct word. Although positively associated, this factor reflects... Results "all the sources of blunting the effect of being in contact with the health system." Grammatically incorrect and unclear. ...all of which blunt the effect of health system contact. Results "One of the thoughtful discoveries of SHAP values was the sizeable negative impact" "Thoughtful" incorrect. A notable discovery from the SHAP analysis was. Results "Isolated use of SMOTE in the training set" – "Isolated" wrong word. Applying SMOTE exclusively to the training set Results "It shifted the ML model from being a prediction device to an analysis tool, not just deciding which features were significant, but the size and sign of their effects, and significantly, potential interactions" Not clear because of parallel verbs. It transformed the ML model from a prediction device into an analytical tool, revealing not only which features were significant but also the magnitude and direction of their effects, as well as potential interactions. Results "Simulation by counterfactual SHAP analysis suggests a hypothetical 30% increase in EC knowledge might boost utilization by approximately 12.7%, a valuable public health gain." The sentence needs clearer explanation. Counterfactual simulation using SHAP values (e.g., calculating the mean impact of increasing the "heard_emergency" feature value) suggested that a 30% increase in EC knowledge could potentially increase utilization by approximately 12.7%, representing a valuable public health gain. Results "Geographic ML modeling over the geographic data would also potentially be able to further optimize resource deployment" Repetition: "Geographic" used twice. Rewrite the sentence for clarity Results "the implied vulnerability evidenced by the 'forced pregnancy' variable (despite missing data concerns) underscore" Not clear as the subject-verb disagreement. .use the word..underscores. Methods Model Selection Justification: The list of eight algorithms is comprehensive, but justification for simpler models like Naive Bayes is weak. Justify the inclusion of Naïve Bayes. Is it possible because they were included as benchmarks. Methods Evaluation Metrics: AUC-ROC emphasized, but for imbalanced problems F1-Score or Precision-Recall AUC may be better. Also consider using F1-Score or Precision as the data is not balanced or Justify the use of AUC-ROC Methods Model Performance Presentation: Logistic Regression focus unclear since Gradient Boosting achieved higher AUC-ROC (0.85). Consider Gradient Boosting as it achieved AUC-ROC 0.85 OR Explain rationale (e.g., performance vs. interpretability). Results Confusion Matrix Analysis (Figure 3): Issue: The analysis states precision is "approximately 11%." Based on the described confusion matrix (TP=17, FP=138), precision is 17 / (17+138) = 11.0%. This is a critical weakness of the model that deserves more emphasis. It means ~89% of the people predicted to be EC users were actually non-users. This has huge implications for the cost and efficiency of any intervention based on this model Discuss this trade-off explicitly: "The model's high recall (85%) comes at the cost of low precision (11%), resulting in a high false positive rate. This suggests the model is well-suited as a screening tool where identifying most true cases is prioritized over resource efficiency, but would require secondary screening or low-cost interventions to target the large number of false positives." Discussion Addressing Limitations More Forcefully: Underreporting of EC likely major issue. Add: "A key limitation is the potential for significant underreporting of EC use due to social desirability bias and stigma..." Conclusion "myth-busting" Word choice is Informal. myth-dispelling Conclusion "stock guarantees of EC" Not clear Consider write as guaranteed EC stock availability Conclusion "This research provides an ethical and evidence-based blueprint to accelerate gains in reducing maternal mortality and advancing reproductive autonomy in Ethiopia and similar settings." – Awkward phrasing. .Conside rephrasing as ..blueprint to reduce maternal mortality and advance... Reviewer #4: This manuscript applies machine learning (ML) and explainable AI (XAI) methods to predict emergency contraceptive (EC) use among women in Ethiopia, using data from the 2023 PMA survey. The authors compare eight algorithms, address severe class imbalance with SMOTE, and use SHAP values to interpret predictors. They find that awareness of EC is the strongest predictor, followed by media exposure and health facility discussions, while demographic variables show limited predictive value.

However, the results as currently presented are unreliable. Major inconsistencies in reported performance metrics (e.g., contradictory precision values, implausible Naive Bayes results, inflated accuracy) call into question the validity of the analyses. In addition, the small number of EC users makes the modeling unstable, and subgroup analyses are not feasible with this dataset. These issues, combined with over-interpretation of SHAP as causal, limit both the methodological credibility and substantive contribution of the paper.

Contradictory precision results The performance metrics are inconsistent. Table 4 shows Logistic Regression with SMOTE achieving precision = 0.72 and recall = 0.85, yet the confusion matrix description reports precision at only ~11%. These cannot both be correct. This discrepancy raises questions about the accuracy of the reported results and must be clarified.

Inflated accuracy The reported accuracy of 0.95 for Logistic Regression with SMOTE appears implausibly high given the extreme class imbalance (4.4% EC use). Accuracy is not an informative measure in this context, and such values raise concerns about potential data leakage or overly optimistic validation. The authors should confirm that the outcome variable or proxy features were not inadvertently included in the predictors.

Over-interpretation of SHAP The SHAP analysis is framed in causal terms (e.g., a 30% increase in knowledge leading to a 12.7% increase in use). SHAP values describe associations within the model, not causal effects. The manuscript should temper these statements and present SHAP findings as indicators of relative predictive importance, not intervention outcomes.

Implausible Naive Bayes results Naive Bayes is reported as having accuracy of only 0.06 pre-SMOTE. Given that 95% of the sample did not use EC, even a trivial majority-class classifier would achieve ~95% accuracy. Such a result suggests an error in coding or reporting that must be checked.

Small minority class vs. model complexity Only 103 EC users were present in the dataset. Training and tuning eight algorithms with hyperparameter searches on such a small minority class risks overfitting and unstable results, even with SMOTE. This limitation should be acknowledged explicitly, with emphasis on the need for validation on independent samples.

Subgroup analysis claims The manuscript claims fairness testing across subgroups (rural/urban, religion, age), but no results are presented. With so few EC users, subgroup analyses would be underpowered and unreliable. It would be more appropriate to note this limitation rather than imply subgroup robustness.

Causality Issue The manuscript repeatedly interprets predictive associations as though they were causal effects. For example, SHAP values are used to suggest that increasing knowledge by 30% would increase EC use by 12.7%. Since the data are cross-sectional and observational, such statements are not justified. Machine learning models in this setting can identify predictive patterns, but they cannot establish causal relationships between predictors and outcomes. This overreach is particularly concerning because it could mislead policymakers or practitioners into believing the study provides evidence of causal effects. Reviewer #5: Summary This study investigates the underuse of emergency contraception in Ethiopia using a machine learning framework. Strengths include the application of multiple algorithms, careful handling of class imbalance, and the use of Explainable AI to interpret model outputs. The paper is generally well-structured, and the methodological workflow is presented clearly. At the same time, the results are presented in a way that overstates the model’s practical utility while giving insufficient attention to the precision–recall trade-off. The manuscript should be revised to consistently acknowledge the low precision across the abstract, results, and discussion, and to provide a clear justification for the relevance of a high-recall, low-precision model in this public health context. The limitation posed by the small number of positive cases in the validation set should also be explicitly discussed. Addressing these points is necessary to strengthen the scientific validity of the work. Specific comments 1. Title; It should be shortened to remove redundancy since Utilization and Usage mean the same thing 2. Abstract. I think something key was missed. The aurthors state a recall of 0.85 without mentioning the precision. I see that (Figure 3, page 20) show that the precision is approximately 11%. My understanding of this that for every 100 women the model flags as likely EC non-users who need intervention, 89 of them are false alarms. An abstract must present a balanced view of performance. 3. Methods (About the data): A sample size of 2,334 with a 4.4% prevalence means you only have ~103 positive cases (EC users). After an 80/20 train-test split, your test set contains only ~21 positive cases. This number is critically small and raises serious questions about the stability and generalizability of your reported performance metrics. A different random split could yield vastly different results. I suggest that such a major limitation is addressed upfront in the limitations section and acknowledged in the methods section. 4. Data balancing; I like the write up of this section 5. Evaluation Metrics; The text states the test set has 18.7% EC users, but the abstract and data balancing section state the overall prevalence is 4.4%. Please clarify this discrepancy. Is 18.7% a typo? Or did the stratified split result in a test set with a much higher prevalence than the overall dataset? This needs to be consistent. Could you also add the precision-recall plots, since you state that they were tracked. 6. Results: - In Table 4, the columns are F1 and Score. This seems like a typo. It should likely be a single column: F1 Score. Please correct. - Lastly, i think it would be good to acknowledge the weaknesses of SMOTE Reviewer #6: The title of the article is: Predicting Utilization of Emergency Contraceptive Usage in Ethiopia and Identifying Its Predictors Using Machine Learning. The author explains that traditional analyses have struggled to identify complex predictors and therefore they used machine learning (ML) and Explainable AI (XAI) to improve the prediction and interpretability of Emergency Contraceptive (EC) use. The paper can be published with the following corrections and some are extremely important. In particular methodological perspectives. Category Authors Contribution Comments Objectives The primary objectives are twofold:

one, to predict the likelihood of EC use with far greater accuracy than conventional regression techniques;

two, to identify the key modifiable socio-behavioural predictors e.g., self-efficacy, mass media exposure, provider perception, and women's autonomy through XAI methods like SHAP values to yield interpretability and actionable insights. First objective can be modified. Far greater is a vague statement. Measuring accuracy is an indicator of choosing between models but conventional regression techniques why has a problem in this study should focus on that.

Second objective seems motivation of the study. This objective should be written in clear sentence. Identify predictors to yield interpretability and actionable insights are subjective things. These objective seems ambiguous.

Methodological view Page 5: Methodologically, it represents a new contribution by rigorously testing the performance of eight alternative ML classifiers and developing an optimized analytical pipeline specifically designed to handle skewed healthcare datasets prevalent in rare outcomes like EC use

Theoretically, it applies the Socio-Ecological Model (SEM) framework to hierarchically analyze predictors at levels of individual (knowledge, attitudes), interpersonal (partner communication, family influence), community (stigma norms, access), and policy (health system factors) providing an integrated explanation for the interrelating influences on EC behavior. It is not methodological contribution.

Moreover, author mentioned theoretical contribution. However, it is just exploratory of the data.

Methodology In page 4: In contrast to conventional statistical approaches, ML algorithms, such as random forests, gradient boosting machines (e.g., XGBoost), and neural networks, can particularly identify complex, high-dimensional patterns within diverse data sets, properly manage missing data, and produce personalized risk predictions with improved accuracy Author mentioned several times about conventional statistical technique. However, in the report author directly reported the model performance of ML. My suggestion is to first run the analysis using traditional or conventional methods and then compare with ML techniques. This is very important. Outcome Variable Page 8: The outcome of interest is EC Usage, a binary measure of whether emergency contraception was used in the last 12 months. This is the dependent variable for analysis. Redundant as at the beginning you mentioned outcome of interest is….. Missing data For handling missingness in our data, a stratified approach based on missingness mechanisms and rates was followed and so on……….. The author used many approaches and it is difficult to keep track. So it is better to explain step by step and pros and cons of each process. Moreover, explain why this approach is best in this study Variables Page 12

Lots of category under one variable. Some category has very few observations. Justify the necessity. May be we can also show some cross-tabulation analysis result and report the p-value. Research Gap Page 19: The research goes beyond the correlational limitations of previous studies by utilizing predictive analytics to identify the modifiable factors and approximate their hypothetical effects What do you mean by correlational limitations? Moreover, over the report the previous studies were not mentioned in comparison to the authors current approaches. Sa add some recent references and explain the research gap. The Machine learning techniques are not new. So it is required to mention how those machine learning helps in your study as a novelty. All over the report there is a missing of synchronization and coherence of sentences. Moreover, the references, table titles etc are not space maintained. Abstract 1. SMOTE and SHAP 2. Conversely, recent reproductive events such as unintended pregnancy were linked to non-use. Static demographic factors showed poor predictive value. Findings highlight that knowledge gaps, not poverty or access, are key barriers to EC use. Tailored media campaigns and routine health counseling could enhance EC uptake. ML and XAI offer powerful tools for guiding targeted reproductive health interventions. 1. Did not mention what it is?

1. The message of these sentences are not coherent. I think author can check the whole paper from an English native reviewer.

R1:

Reviewer #4: I appreciate the authors' thoughtful revisions and detailed responses. Several of my earlier comments were addressed—specifically, the correction of Naive Bayes reporting errors, improved acknowledgment of sample size limitations, and removal of unsupported subgroup analyses. These are welcome improvements. However, key concerns about the internal consistency of results, causal interpretation of SHAP analyses, and overextension of policy recommendations remain unresolved.

First, while the outdated "11% precision" text has been removed, the confusion matrix values (TP=102, FP=180, FN=18) still do not correspond to the reported performance metrics. With these numbers, precision would equal roughly 0.36, not the 0.72 cited in Table 4. This suggests an ongoing internal inconsistency between the descriptive counts and the summary metrics. The lack of alignment raises continuing doubts about the reliability of the reported model performance.

Second, the manuscript still places heavy emphasis on accuracy values approaching 0.92–0.95 despite a highly imbalanced outcome (4.4% EC use). Although the authors state that AUC-ROC and recall were prioritized, the presentation continues to foreground accuracy, which is misleading in this context. No calibration or uncertainty measures (e.g., Brier score, calibration curve) have been added, leaving the reader without a sense of how well the predicted probabilities reflect actual risk.

Third, although the authors softened their language, the interpretation of SHAP values remains quasi-causal. The new statement—"counterfactual simulation using SHAP values … suggested that a 30% increase in EC knowledge could potentially increase utilization by approximately 12.7%", still presents SHAP outputs as if they represent real-world intervention effects. SHAP analysis identifies predictive associations within a model; it does not estimate the causal impact of changing a feature in the population. Likewise, subsequent phrases such as “integrating a predictive risk-scoring tool can help identify women at high risk” and “geographic machine learning modeling can optimize resource deployment” continue to frame the model as a validated operational tool. These remain prescriptive policy claims that move beyond what a cross-sectional, unvalidated predictive study can substantiate.

Finally, while the tone of the manuscript has improved, the discussion still reads as policy advocacy rather than analytical interpretation. Phrases like "representing a valuable public health gain”" and "can help optimize resource deployment" give the impression of proven effectiveness rather than exploratory modeling. A clearer distinction between predictive insights and causal or operational evidence is necessary for the study to maintain methodological integrity.