Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for submitting your manuscript “Midwife-led Home Births in Japan: A 25-year Retrospective Analysis of Care in Accordance with WHO Recommendations Before and After COVID-19” to Health. This manuscript presents a 25-year retrospective case series of planned home births attended by independent midwives in Japan, with two stated aims: (1) to assess alignment of care with selected WHO intrapartum care recommendations; and (2) to compare maternal/neonatal characteristics pre- vs post-COVID-19.

The writing style is solid and engaging.

While I have significant concerns regarding study design, reporting clarity, and statistical analysis, I recognize that research on home birth in Japan is limited in the English-language literature and that this topic is important. I therefore recommend major revision. Substantial clarification, methodological transparency, and reanalysis are required before the manuscript can be considered for publication.

Throughout the manuscript, the authors refer to “observed practice.” However, this study is based on retrospective review of midwifery records rather than direct clinical observation. Therefore, it would be more accurate to use the term “documented practice,” as the findings reflect what was recorded in the charts rather than what was directly observed in real time. This distinction is important, particularly given the acknowledged limitations related to documentation completeness and consistency.

Comments concerning the Introduction and Setting

The manuscript states that the study occurred at “a private midwifery home,” which creates uncertainty about whether births took place in private homes, in a birth center, or in both settings, given that this study analyzed planned home births. The “Setting” section similarly notes that “the study was conducted at a private midwifery home,” while the manuscript repeatedly refers to the sample as “planned home births.” This terminology is confusing and requires clarification. The authors should more clearly describe where the 430 births occurred.

In addition, the manuscript does not indicate how many midwives were involved in providing care across the 25-year study period. It is important to specify whether births were attended primarily by a single midwife or by multiple practitioners, as this has implications for the generalizability of the findings.

More importantly, the inclusion criteria specify “continuity of care by the same independent midwife throughout the perinatal period.” This wording strongly implies one midwife per woman, but it is unclear whether all births were attended by one specific midwife across 25 years or if there were multiple midwives over time, but each client had “the same midwife” assigned. In some countries, intrapartum care involves a team (e.g., primary + second midwife). Was this the case? Or is the “midwifery home” a service distinct from the actual birth location?

This matters because the study’s conclusions: “Independent midwives provided evidence-based…care” and broader implications for Japan depend heavily on whether this reflects one practitioner’s practice style vs a service/center model.

Finally, the authors should address whether birth documentation practices were standardized and consistent over the study period, or whether changes occurred over time. Given the long duration of data collection, variation in record-keeping practices may have influenced data completeness and interpretation and should be discussed.

Comments on the Methods section

The authors report statistically significant differences in maternal age, gestational weight gain, and neonatal birth weight between pre- and post-pandemic groups. However, the post-pandemic sample is small and highly unbalanced relative to the pre-pandemic group. In addition, the large variability in gestational weight gain raises concerns about data distribution and outliers. These factors limit confidence in the robustness of the reported differences and make causal interpretation difficult.

Here are two examples where the authors need to write a justification and perhaps reconsider their analysis:

1. The reported standard deviation for labor duration (342 minutes) is nearly as large as the mean (387 minutes), suggesting substantial variability and possible skewness. In such cases, reporting the median and interquartile range may provide a clearer picture of the typical labor length. Further clarification of the distribution of this variable would strengthen confidence in the use of parametric statistical tests.
2. The reported standard deviation for gestational weight gain exceeds the mean, suggesting substantial variability and possible skewness or outliers. In such cases, the assumption of normality underlying parametric tests should be carefully justified. Reporting median values and distributional characteristics would strengthen confidence in the analysis.

The authors could strengthen the statistical analysis by providing graphical summaries (such as histograms or distribution plots) to show the shape of the data.

Comments on the Discussion

The discussion largely reiterates that practices “align closely with WHO recommendations” and that home birth allows family presence and breastfeeding, then interprets observed pre/post differences as pandemic effects.

Concerns relative to aims:

Aim 1 (WHO alignment):
Given the measurement issues, it is safer to say “documented practices suggest partial alignment with selected recommendations,” rather than “aligns closely,” especially because several WHO domains were excluded and others were not measurable from records.

Aim 2 (pre/post differences):
The discussion attributes changes to pandemic lifestyle/behavior, but the analysis cannot distinguish pandemic effects from: secular trends (e.g., maternal age trend over decades), changes in clientele, changes in recordkeeping, and given that the post-pandemic sample is very small (n=35).

Conclusions are too broad for a single-site dataset:
The manuscript generalizes to “independent midwives in Japan” and policy-level recommendations about expanding access. With unclear provider count and a single “midwifery home,” this needs more caution.

I recommend that the authors tighten their claims to better reflect the study design. This is a single-site, retrospective case series based on documented care processes. Given the small post-pandemic sample size, the data are insufficient to support strong comparative conclusions or to generate robust hypotheses regarding pre- and post-pandemic differences.

The Limitations

While the authors acknowledge some important limitations, this section remains incomplete and requires substantial strengthening to reflect the methodological constraints of the study more accurately.

I mentioned this above and will mention it again in terms of limitations. Although the authors state that the study was conducted at a “single midwifery home,” they do not clarify whether care was provided by one midwife or by multiple practitioners over the 25-year period. This is a major limitation that is not adequately addressed. If the majority of births were attended by a single practitioner, the findings may primarily reflect individual practice style rather than a generalizable model of independent midwifery care. This distinction is critical for interpreting claims about “midwife-led home births in Japan” and should be explicitly acknowledged.

Also, the authors note that the study relied on midwifery records, but they do not fully engage with the implications of this reliance. Retrospective record-based studies are subject to documentation bias, changes in recording practices over time, and incomplete or inconsistent data capture, especially across a 25-year period. These factors may have influenced the apparent frequency of clinical practices (e.g., skin-to-skin contact, breastfeeding initiation, labor monitoring) and should be discussed more explicitly. Missing data are briefly mentioned but not contextualized in relation to potential systematic bias.

Further, the limitations section does not sufficiently address the restricted scope of the analytic sample. The exclusion of maternal and neonatal transfer cases likely leads to an overrepresentation of uncomplicated births and may bias outcome estimates in a favorable direction. This selection bias is particularly relevant given the study’s focus on safety and alignment with WHO recommendations and should be acknowledged as a major limitation.

Lastly, although the authors mention the absence of qualitative data, they frame this primarily as a gap in understanding women’s subjective experiences. This limitation should be expanded to note that the study cannot adequately assess key WHO constructs such as respectful care, informed choice, and perceived autonomy using record data alone. Consequently, conclusions about “positive childbirth experience” remain indirect and incomplete.

The “Future Directions” section would benefit from closer alignment with these methodological limitations. In addition to recommending qualitative research and improved documentation, the authors should consider advocating for:

• multi-site studies involving diverse geographic and organizational contexts,
• explicit reporting of provider characteristics and care team composition,
• prospective data collection with standardized measures, and
• inclusion of transfer cases and comparison groups where feasible.

Overall, the limitations section currently underrepresents several major sources of bias and uncertainty. A more transparent and comprehensive discussion is needed to ensure that readers can appropriately interpret the study’s findings and their applicability.

Author Response

We sincerely appreciated the reviewer’s thorough and constructive feedback. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

General Assessment

The manuscript presents a valuable longitudinal retrospective analysis of planned home births attended by independent midwives in Japan, with particular emphasis on alignment with the 2018 intrapartum care recommendations of the World Health Organization (WHO). In the context of the extremely low prevalence of home births in Japan (0.1%), this study provides rare and relevant empirical insight into a model of care that remains insufficiently explored in the Japanese literature.

 

Methodology

A retrospective case series (n = 400) represents an appropriate design for descriptive and comparative analysis of practices and outcomes within a single center.

Methodological limitations:

1. The extended 25-year time frame strengthens the value of the findings; however, it simultaneously raises questions regarding the consistency and homogeneity of documentation over time.
2. There is no analysis of perinatal morbidity using standardized indicators (e.g., NICU admission, neonatal acidosis).
3. The comparison with WHO recommendations appears partially interpretative and dependent on the availability and completeness of documentation. Although the authors state that 23 out of 33 WHO recommendations were analyzed and clearly explain why 10 were excluded, it remains unclear whether objective operational indicators were defined for the remaining 23 recommendations, or whether compliance assessment was primarily descriptive. The absence of clearly operationalized criteria reduces the reproducibility and transparency of the analysis.

Recommendation:

Consider conducting regression analyses (e.g., linear or logistic regression) to control for potential confounders such as maternal age, BMI, and parity.

Results

4. The reported blood loss ≥500 mL in 14.1% of cases requires further interpretation. Is this rate clinically relevant according to Japanese obstetric standards? Blood loss ≥500 mL occurs in approximately 5–15% of vaginal births. Importantly, how was blood loss measured in the context of home births?
5. Only 52.9% of mothers experienced skin-to-skin contact with their newborns despite high Apgar scores. This finding warrants clarification.
6. The mean duration of labor was 387.4 minutes; however, no data on parity are provided, which is crucial for interpreting labor duration and perineal outcomes.
7. The interpretation of pre- and post-COVID-19 differences is based on international research regarding lifestyle changes during the pandemic. However, these conclusions remain speculative because no data were collected on maternal diet, physical activity, or psychosocial stress, even though the women received continuous care from the same midwife throughout pregnancy. Furthermore, there was no longer-term follow-up of maternal psychological well-being, which limits the strength of conclusions regarding positive childbirth experiences.

Recommendation:

Clearly distinguish between hypotheses and evidence-based conclusions.

Discussion

The discussion is well structured and appropriately contextualized within the Japanese healthcare system. The argument that midwife-led continuity models may enhance system resilience during crises such as pandemics is theoretically convincing; however, it requires additional comparative research.

8. The manuscript notes that institutional births may restrict the presence of multiple family members (siblings, grandparents), yet in the results section family presence is reported only generally, without specifying the number or family background of those present. Greater precision would strengthen this argument.

Limitations

9. While the authors appropriately acknowledge key limitations, the following should also be explicitly addressed: potential selection bias (more motivated and possibly more educated women may choose home birth); potential reporting bias; lack of long-term neonatal outcomes.

Structural and Language Issues

A thorough linguistic revision is required, including harmonization of professional terminology, correction of inconsistencies,

Comments for author File: Comments.pdf

Comments on the Quality of English Language

A thorough linguistic revision is required, including harmonization of professional terminology, correction of inconsistencies, and reduction of repetition

Author Response

We sincerely appreciated the reviewer’s thoughtful and constructive evaluation. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Clearer framing of hypotheses versus evidence-based conclusions and correction of typographical errors will improve readability and precision.

Comments on the Quality of English Language

A thorough linguistic revision is required, including harmonization of professional terminology, correction of inconsistencies, and reduction of repetition

Author Response

Response to Reviewer 2

We sincerely thank Reviewer 2 for the careful reading of our manuscript and for providing constructive comments that helped us improve the clarity and rigor of the paper. Below, we address each point in detail.

 

1. Quality of English Language⇒The English could be improved to more clearly express the research.

Response: We appreciate this important suggestion. The manuscript has now been checked by a professional English language editor. Please find the editing certificate attached.

 

2. Is the research design appropriate? ⇒Can be improved.

Response: We agree that additional clarification was required. We have now expanded the Methods section to better describe the retrospective study design, the evaluation procedures, and the limitations inherent to the record-based data. These additions have strengthened the transparency and appropriateness of the research design.

 

3. Are the methods adequately described? ⇒Can be improved.

Response: We have now added methodological details regarding:

・how the WHO recommendations were operationalized using the available documentation,

・how any discrepancies between reviewers were resolved,

・how the missing data were handled

To further enhance methodological transparency, we have added a Supplementary Table that lists all WHO intrapartum care recommendations and identifies the 23 items that correspond to variables documented in the midwifery records used in this study. This table also clarifies how each recommendation was operationalized for retrospective assessment. We hope this addition addresses the reviewer’s concern regarding the clarity and completeness of the methodological description.

 

4. Are the results clearly presented?　⇒Can be improved.

Response: We have now revised the Results section to improve the readability and flow. While we did not add any new numerical summaries, we have clarified the descriptions of existing findings and improved the structure of the narrative to enhance comprehension.

 

5. Are the conclusions supported by the results?　⇒Can be improved.

Response: 

We have refined the Discussion to ensure that all interpretations remain within the limits of the descriptive data. In particular, statements regarding positive childbirth experiences were adjusted to emphasize the alignment with WHO principles, rather than implying the direct measurement of women’s subjective experiences.

 

 

6. Comments and Suggestions for Authors

⇒Clearer framing of hypotheses versus evidence-based conclusions and correction of typographical errors will improve readability and precision.

Response: Thank you very much for this insightful comment.　We would like to clarify that the present study is a retrospective descriptive analysis based on midwifery records, and therefore it was not designed as a hypothesis‑testing study. For this reason, we did not formulate or test specific hypotheses in the Introduction.　However, we fully agree with the reviewer that conclusions must be clearly grounded in the evidence presented. In response, we revised the Discussion to ensure that all interpretations are explicitly linked to the descriptive findings, and we adjusted wording that could be interpreted as implying causal inference or hypothesis testing. We also corrected typographical inconsistencies and improved clarity throughout the manuscript.　We hope these revisions address the reviewer’s concerns regarding readability and precision.

 

7. Comments on the Quality of English Language

⇒A thorough linguistic revision is required, including harmonization of professional terminology, correction of inconsistencies, and reduction of repetition

Response：Thank you very much for your valuable comments and suggestions.　We appreciated your careful review of our manuscript.

・We revised the manuscript to improve the clarity and precision of our conclusions, ensuring that they were fully supported by the presented evidence.

・We also corrected typographical errors and made additional revisions to enhance overall readability.

・Regarding the English language quality, we conducted a thorough linguistic revision, including the harmonization of terminology, correction of inconsistencies, and reduction of repetition.

We sincerely appreciated your constructive feedback and incorporated these improvements in the revised version.

Author Response File: Author Response.pdf