Round 1

Reviewer 1 Report (New Reviewer)

Comments and Suggestions for Authors

The manuscript addresses a relevant public health issue. However, several aspects of the study design, outcome definition, data presentation, statistical analysis, and interpretation require clarification and revision before the findings can be reliably assessed.

1.This three-month, single-center, hospital-based cross-sectional study included 380 women delivering at a tertiary maternity hospital. Therefore, the findings cannot be considered representative of women in Sudan, or even of the general population of central Sudan. The title, abstract, discussion, and conclusion should be revised to avoid overgeneralization and should clearly state that the results apply only to women delivering at this specific hospital during the study period.

2.The patient numbers reported in the Abstract are internally inconsistent. The authors state that 171 women had HRFB; however, they subsequently reported that 250 women had a single HRFB and 88 had multiple HRFBs. These values ​​sum to 338 and cannot represent subgroups of the 171 women with HRFB. Furthermore, the main Results section reports 52 women with multiple HRFBs, rather than 88. The authors should verify the raw data and correct all related numbers and percentages throughout the manuscript.

3.Systematic random sampling may be appropriate for this study; however, the recruitment procedure is not described in sufficient detail. The sampling interval of five was calculated using delivery numbers from the preceding three months rather than the actual study period. Therefore, the authors should report the total number of deliveries and the number of eligible women between September and December 2023. They should also explain what happened when a selected woman declined participation or was found to be ineligible: whether the next eligible woman was recruited as a replacement or whether the original five-patient interval was maintained.

4.For women without accessible records, the interval appears to have been calculated from the previous live birth to the current delivery, which represents an interbirth interval rather than an interpregnancy interval. The authors should clarify how the estimated conception date was determined, including whether last menstrual period or ultrasound dating was used, whether the duration of the current pregnancy was subtracted, and how previous miscarriages or stillbirths were handled. Recalculation may be required.

5.The operational definition of HRFB is internally inconsistent. The authors initially define HRFB using four components, including maternal age <18 years, but subsequently exclude this component solely because marriage certificates are not legally issued below age 18 in Sudan. Legal marriage regulations do not exclude the occurrence of adolescent pregnancies or births and are not a valid methodological basis for omitting this established HRFB component. Moreover, the manuscript continues to refer to “four risk factors” while listing only three, and uses birth order and parity interchangeably. The authors should provide a single, consistent definition, clarify how current birth order was calculated, and repeat the prevalence and regression analyses if the HRFB classification changes.

6.The Results section and tables contain several serious internal inconsistencies that require a complete data audit. First, the reported residence distribution is incorrect: 225 urban and 55 rural women sum to only 280, and 225/380 corresponds to 59.2%, not 85.5%; Table 2 instead suggests that the urban count may be 325. Second, maternal employment categories appear to be reversed between Tables 1 and 2. Third, the husband's education counts in Table 2 are impossible, as the reported categories within the HRFB group sum to more than the 171 women in that group and are also inconsistent with Table 1. Most importantly, the reported number of women with multiple HRFBs is mathematically incompatible with the frequencies of the individual HRFB components and differs between the Abstract and Results. In addition, the median maternal age is reported as 26.0 years in the Abstract but 26.5 years in the Results, and the confidence interval for the 45% HRFB prevalence is incorrectly presented as “399–50.0%.” The authors should verify the raw dataset and recalculate all frequencies, percentages, confidence intervals, subgroup counts, and tables before the findings can be reliably evaluated.

7.BMI was measured at admission for delivery and therefore reflects gestational weight gain, fetal and placental weight, amniotic fluid, edema, and gestational age rather than pre-pregnancy adiposity. Applying standard WHO adult BMI categories at this time point and interpreting BMI as a determinant of HRFB is therefore problematic and highly susceptible to reverse causality. The authors should substantially revise this analysis and avoid causal conclusions or recommendations regarding obesity and weight management.

8.The Discussion, particularly the section on BMI, is overly speculative and extends beyond the data collected. Explanations involving delayed or early childbearing, multiple pregnancies, family-size pressure, and impaired family planning were not evaluated in this study and should not be presented as explanations for the observed association.

9.The Author Contributions statement does not correspond to the listed authors, as it includes initials such as OMA, KN, and KA that do not appear in the author list. This section should be carefully revised to ensure that all initials accurately match the authors and their actual contributions. In addition, the Funding statement declares “None,” whereas the Acknowledgments section reports financial support from Qassim University (QU-APC-2026). The authors should clarify whether this support was limited to the article processing charge and resolve the inconsistency between these sections.

Author Response

We thank the editor and the reviewers for their valuable comments on the manuscript.  The comments/response have dramatically improved the manuscript.

Reviewers #1

1.This three-month, single-center, hospital-based cross-sectional study included 380 women delivering at a tertiary maternity hospital. Therefore, the findings cannot be considered representative of women in Sudan, or even of the general population of central Sudan. The title, abstract, discussion, and conclusion should be revised to avoid overgeneralization and should clearly state that the results apply only to women delivering at this specific hospital during the study period.

Response

Yes, agreed, and this point has been inserted throughout the manuscript from the title to the conclusion, as suggested.

Comment

2.The patient numbers reported in the Abstract are internally inconsistent. The authors state that 171 women had HRFB; however, they subsequently reported that 250 women had a single HRFB and 88 had multiple HRFBs. These values ​​sum to 338 and cannot represent subgroups of the 171 women with HRFB. Furthermore, the main Results section reports 52 women with multiple HRFBs, rather than 88. The authors should verify the raw data and correct all related numbers and percentages throughout the manuscript.

Response

Yes, agreed, and the numbers and values were checked throughout the manuscript. We apologize for the inconvenience.

Comment

3.Systematic random sampling may be appropriate for this study; however, the recruitment procedure is not described in sufficient detail. The sampling interval of five was calculated using delivery numbers from the preceding three months rather than the actual study period. Therefore, the authors should report the total number of deliveries and the number of eligible women between September and December 2023. They should also explain what happened when a selected woman declined participation or was found to be ineligible: whether the next eligible woman was recruited as a replacement or whether the original five-patient interval was maintained.

Response

Agreed, and these points have been inserted as suggested. Please see lines 122-131.

Comment

4.For women without accessible records, the interval appears to have been calculated from the previous live birth to the current delivery, which represents an interbirth interval rather than an interpregnancy interval. The authors should clarify how the estimated conception date was determined, including whether last menstrual period or ultrasound dating was used, whether the duration of the current pregnancy was subtracted, and how previous miscarriages or stillbirths were handled. Recalculation may be required.

Response

Yes, agreed, and the IPI definition is stated as suggested. Please see lines 157-159

Comment

5.The operational definition of HRFB is internally inconsistent. The authors initially define HRFB using four components, including maternal age <18 years, but subsequently exclude this component solely because marriage certificates are not legally issued below age 18 in Sudan. Legal marriage regulations do not exclude the occurrence of adolescent pregnancies or births and are not a valid methodological basis for omitting this established HRFB component. Moreover, the manuscript continues to refer to “four risk factors” while listing only three, and uses birth order and parity interchangeably. The authors should provide a single, consistent definition, clarify how current birth order was calculated, and repeat the prevalence and regression analyses if the HRFB classification changes.

Response

Yes, agreed, and a consistent definition of HRFB is used with clear three criteria in our study, while other studies used the four criteria (age less than 18). Please see it.

Comment

6.The Results section and tables contain several serious internal inconsistencies that require a complete data audit. First, the reported residence distribution is incorrect: 225 urban and 55 rural women sum to only 280, and 225/380 corresponds to 59.2%, not 85.5%; Table 2 instead suggests that the urban count may be 325. Second, maternal employment categories appear to be reversed between Tables 1 and 2. Third, the husband's education counts in Table 2 are impossible, as the reported categories within the HRFB group sum to more than the 171 women in that group and are also inconsistent with Table 1. Most importantly, the reported number of women with multiple HRFBs is mathematically incompatible with the frequencies of the individual HRFB components and differs between the Abstract and Results. In addition, the median maternal age is reported as 26.0 years in the Abstract but 26.5 years in the Results, and the confidence interval for the 45% HRFB prevalence is incorrectly presented as “399–50.0%.” The authors should verify the raw dataset and recalculate all frequencies, percentages, confidence intervals, subgroup counts, and tables before the findings can be reliably evaluated.

Response

We are deeply sorry; all these were corrected throughout the manuscript. Please see it.

Comment

7.BMI was measured at admission for delivery and therefore reflects gestational weight gain, fetal and placental weight, amniotic fluid, edema, and gestational age rather than pre-pregnancy adiposity. Applying standard WHO adult BMI categories at this time point and interpreting BMI as a determinant of HRFB is therefore problematic and highly susceptible to reverse causality. The authors should substantially revise this analysis and avoid causal conclusions or recommendations regarding obesity and weight management.

Response

Yes, agreed, and this point has been inserted as a limitation, and we have deleted most of the recommendations related to BMI as suggested. Please see them.

Comment

8.The Discussion, particularly the section on BMI, is overly speculative and extends beyond the data collected. Explanations involving delayed or early childbearing, multiple pregnancies, family-size pressure, and impaired family planning were not evaluated in this study and should not be presented as explanations for the observed association.

Response

Yes, agreed and deleted.

Comment

9.The Author Contributions statement does not correspond to the listed authors, as it includes initials such as OMA, KN, and KA that do not appear in the author list. This section should be carefully revised to ensure that all initials accurately match the authors and their actual contributions. In addition, the Funding statement declares “None,” whereas the Acknowledgments section reports financial support from Qassim University (QU-APC-2026). The authors should clarify whether this support was limited to the article processing charge and resolve the inconsistency between these sections.

Response

Okay, the Author Contributions statement is revised.  The funding and acknowledgment statement is required as such from our institution (Qassim University), and this is according to the agreement between the publisher (MDPI) and the university.  

Kind Regards

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Overall Assessment

This manuscript addresses an important and under-studied reproductive health topic in Sudan. The use of hospital-based primary data from a conflict-affected setting is potentially valuable, and the study may contribute useful local evidence regarding high-risk fertility behavior (HRFB). However, several issues currently limit confidence in the reported prevalence estimates and regression findings. The most important concerns involve the operational definition of HRFB, inconsistencies between the Abstract, Results, and tables, the interpretation of BMI as an associated factor, the construction of the multivariable model, and the reporting of the sampling and sample-size procedures. These issues are substantial but, in my view, potentially correctable if the authors conduct a complete audit of the dataset and revise the manuscript accordingly. Therefore, I recommend major revision.

Major Comments

1. The definition of HRFB must be made conceptually and operationally consistent.

The manuscript initially describes HRFB using the conventional components of maternal age <18 years or >35 years, high birth order, and short interpregnancy interval. However, in the Methods section, the authors state that because Sudanese law does not issue a marriage certificate to persons younger than 18 years, only maternal age >35 years was considered high risk. This rationale is not methodologically sufficient. A legal minimum age for marriage does not demonstrate that pregnancies or births do not occur among women younger than 18 years. Moreover, teenage pregnancy is later discussed as a relevant reproductive health concern in Sudan.

The authors should decide whether they are using a previously established HRFB definition or a modified Sudan-specific definition. If the definition is modified, this must be explicitly justified, consistently applied throughout the manuscript, and discussed as a limitation because it affects comparability with previous HRFB studies. The manuscript also repeatedly refers to “four risks,” whereas the operational definition as currently written appears to contain only three categories after exclusion of age <18 years. This needs to be corrected throughout the Abstract, Methods, Results, tables, and Discussion.

2. The primary outcome counts are internally inconsistent and require a complete data audit.

The Abstract reports that 171 women had HRFB, but then states that 250 women had a single HRFB and 88 had multiple HRFBs. These values cannot represent subgroups of 171 women. In the main Results section, the number with multiple HRFBs is reported as 52, which differs from the Abstract.

Because this discrepancy directly concerns the primary outcome, the authors should return to the participant-level dataset, recreate the HRFB classification, and verify all frequencies before revising the text. The manuscript should clearly report: (a) total number with no HRFB, (b) total number with any HRFB, (c) number with exactly one HRFB, (d) number with two or more HRFBs, and (e) the frequency of each individual HRFB component. These numbers must be mutually compatible and identical across the Abstract, Results, and tables.

3. Table 2 contains serious inconsistencies in category frequencies.

Several cell counts in Table 2 do not correspond to the stated group totals. For example, within the HRFB group (n=171), the husband’s education frequencies shown in the table sum to a number greater than the HRFB sample size. The maternal employment rows also appear inconsistent with the totals reported in Table 1, suggesting that labels or values may have been transposed.

This is more than a formatting issue because the same variables are used in the regression analyses. The authors should regenerate Tables 1–3 directly from the verified analysis dataset and check every row and denominator. It would be helpful to have a second investigator independently verify the tables against the statistical output before resubmission.

4. The multivariable regression strategy requires stronger justification.

The manuscript states that variables with a univariable p-value <0.20 were entered into the multivariable logistic regression. Although such screening approaches are sometimes used, selection of confounders should not rely only on univariable statistical significance. Important variables may fail to meet an arbitrary p-value threshold yet remain necessary for adjustment on epidemiological grounds.

The authors should clarify whether variables were selected based on prior literature, a conceptual framework, or statistical screening. A more defensible approach would retain key prespecified covariates that are plausible confounders regardless of univariable significance. The authors should also report the final model-building strategy transparently, including whether interaction terms were considered, how continuous variables were assessed for linearity in the logit, and whether influential observations were evaluated. Given the inconsistencies in Table 2, the multivariable model should be rerun after the dataset is verified.

5. BMI measured at delivery should not be interpreted as a straightforward determinant of HRFB.

Weight and height were measured at admission for delivery, and the manuscript correctly acknowledges that this BMI includes gestational weight gain. This creates an important temporality problem. HRFB incorporates prior reproductive history, including parity and interpregnancy interval, whereas BMI was measured at the end of the current pregnancy. It is therefore not possible to determine whether higher BMI preceded the fertility pattern, resulted partly from repeated pregnancies, or both.

The authors should consistently describe BMI as an “associated factor” rather than a determinant or causal predictor. Statements suggesting that weight management would reduce HRFB should be substantially moderated unless supported by longitudinal evidence. The possibility of reverse causality is already recognized in the Discussion and should also shape the Abstract and Conclusion.

6. The sampling description is contradictory and should be rewritten.

The manuscript states that all women who gave birth during the study period were invited to participate, but later describes systematic random sampling in which every fifth eligible woman was approached. These are different sampling procedures.

The authors should provide a single, precise description of recruitment. Please clarify how the random starting point was selected, whether every fifth admission or every fifth eligible admission was sampled, how women who were ineligible or declined participation were handled within the sampling sequence, and whether replacement participants were selected. The statement that the procedure yielded a “highly representative” sample should also be softened because this was a single tertiary hospital and representativeness is limited by the facility-based design.

7. The sample-size calculation is not sufficiently reproducible.

The authors report use of a single-population proportion formula, a presumed prevalence of 50%, and a 95% confidence level, resulting in n=380. However, the margin of error/precision is not clearly stated. The same paragraph also introduces an assumption that inadequate ANC doubles the risk of HRFB, which appears to relate to an exposure–outcome comparison rather than a simple prevalence calculation.

Please provide the exact sample-size formula and all assumptions, including expected prevalence, confidence level, precision, design effect, and anticipated non-response. If a separate calculation was conducted to detect an association between ANC and HRFB, this should be reported separately and the larger required sample size should be identified.

8. The interpretation of the cross-sectional findings should be more cautious.

The manuscript frequently uses language such as “determinants,” “drivers,” and intervention-oriented statements implying that changing education or BMI would reduce HRFB. A cross-sectional design can identify associations but cannot establish temporal or causal relationships. This is particularly important for BMI, but it also applies to other exposures.

Please revise the title, Abstract, Results, Discussion, and Conclusion to use association-based terminology consistently. The Discussion should distinguish clearly between what the study observed and what is hypothesized based on previous literature. Speculative explanations should be shortened unless they are directly supported by the present data.

9. Generalizability is narrower than the manuscript sometimes implies.

This was a single-center study conducted at a tertiary maternity hospital during an active conflict. Women who reached this hospital may differ substantially from women who delivered elsewhere or were unable to access institutional care. The conflict may also have changed referral patterns, travel, healthcare access, and the clinical profile of women presenting for delivery.

The authors appropriately mention these issues in the Limitations section, but the wording elsewhere should be aligned with this limitation. Statements such as “prevalence among women in central Sudan” may overstate population representativeness. The conclusions should instead refer to women delivering at the study hospital or to the hospital-based sample unless population-based inference can be justified.

10. The Discussion should be reorganized to focus more directly on the study findings.

The Discussion is relatively long and at times moves from the observed associations to broad explanations involving sociocultural norms, contraception, obesity, psychological mechanisms, and reproductive decision-making. Some of these interpretations are plausible, but they are not measured in the present study.

I suggest restructuring the Discussion around four clear components: (1) prevalence of HRFB and comparison with regional literature; (2) maternal education; (3) BMI, with strong emphasis on temporality and reverse causation; and (4) non-significant findings such as ANC, residence, husband’s education, and employment. The final paragraphs should then summarize strengths, limitations, and implications. This would reduce repetition and improve the scientific focus.

11. Several reporting and manuscript-integrity details must be corrected.

The manuscript contains additional inconsistencies that should be addressed before publication. The Author Contributions section includes initials that do not clearly correspond to the authors listed on the title page. The manuscript states “Funding: None,” while the Acknowledgments section mentions financial support from Qassim University. If this support covered only the article processing charge, this should be stated clearly and consistently in the appropriate section.

The reported confidence interval for HRFB prevalence also appears to contain a typographical error (“399–50.0%”). Please conduct a complete consistency check of author initials, funding statements, percentages, confidence intervals, abbreviations, table labels, and numerical values across all sections.

12. STROBE reporting should be strengthened.

The manuscript states that STROBE guidelines were strictly followed. However, several items would benefit from fuller reporting, including the exact participant flow, handling of refusals within the systematic sampling scheme, rationale for categorized continuous variables, missing-data procedures, and the basis for confounder selection.

Please review the manuscript against the STROBE checklist item by item and revise accordingly. If a completed STROBE checklist is required by the journal, the authors should provide one at resubmission.

Minor Comments

• Please correct typographical errors and spacing problems throughout the manuscript, including inconsistent spacing before citations and around parentheses.
• Use one term consistently for interpregnancy interval (IPI) and ensure that the abbreviation is defined once and used consistently thereafter.
• Avoid alternating between “high-risk fertility behavior” and “high-risk births” when these terms are not conceptually identical.
• Clarify whether parity and birth order are used interchangeably. If not, define each variable precisely.
• For continuous BMI in logistic regression, state clearly that the odds ratio represents the change in odds per 1 kg/m² increase. This note is already provided in Table 3 and should be reflected consistently in the text.
• Consider reporting exact 95% confidence intervals for the overall HRFB prevalence and for important subgroup estimates.
• The phrase “100% completion rate” should be reconciled with the statement that 10 women declined participation. Presumably, the authors mean there were no missing values among the 380 enrolled participants.
• Please verify the terminology used for ANC. The WHO model refers to antenatal contacts rather than merely visits; the manuscript should use terminology consistent with the source cited.
• The Conclusion should be shortened and limited to findings supported by the study design. Policy recommendations should be framed cautiously.
• English-language editing is recommended after the scientific revisions are completed.

 

Author Response

We thank the editor and the reviewers for their valuable comments on the manuscript.  The comments/response have dramatically improved the manuscript.

Major Comments

1. The definition of HRFB must be made conceptually and operationally consistent.

The manuscript initially describes HRFB using the conventional components of maternal age <18 years or >35 years, high birth order, and short interpregnancy interval. However, in the Methods section, the authors state that because Sudanese law does not issue a marriage certificate to persons younger than 18 years, only maternal age >35 years was considered high risk. This rationale is not methodologically sufficient. A legal minimum age for marriage does not demonstrate that pregnancies or births do not occur among women younger than 18 years. Moreover, teenage pregnancy is later discussed as a relevant reproductive health concern in Sudan.

The authors should decide whether they are using a previously established HRFB definition or a modified Sudan-specific definition. If the definition is modified, this must be explicitly justified, consistently applied throughout the manuscript, and discussed as a limitation because it affects comparability with previous HRFB studies. The manuscript also repeatedly refers to “four risks,” whereas the operational definition as currently written appears to contain only three categories after exclusion of age <18 years. This needs to be corrected throughout the Abstract, Methods, Results, tables, and Discussion.

Response

Yes, agreed, and it has been corrected throughout the manuscript as suggested.

 

 

Comment

1. The primary outcome counts are internally inconsistent and require a complete data audit.

The Abstract reports that 171 women had HRFB, but then states that 250 women had a single HRFB and 88 had multiple HRFBs. These values cannot represent subgroups of 171 women. In the main Results section, the number with multiple HRFBs is reported as 52, which differs from the Abstract.

Because this discrepancy directly concerns the primary outcome, the authors should return to the participant-level dataset, recreate the HRFB classification, and verify all frequencies before revising the text. The manuscript should clearly report: (a) total number with no HRFB, (b) total number with any HRFB, (c) number with exactly one HRFB, (d) number with two or more HRFBs, and (e) the frequency of each individual HRFB component. These numbers must be mutually compatible and identical across the Abstract, Results, and tables.

Response

Yes, agreed, and this is corrected throughout the manuscript as suggested.

Comment

1. Table 2 contains serious inconsistencies in category frequencies.

Several cell counts in Table 2 do not correspond to the stated group totals. For example, within the HRFB group (n=171), the husband’s education frequencies shown in the table sum to a number greater than the HRFB sample size. The maternal employment rows also appear inconsistent with the totals reported in Table 1, suggesting that labels or values may have been transposed.

This is more than a formatting issue because the same variables are used in the regression analyses. The authors should regenerate Tables 1–3 directly from the verified analysis dataset and check every row and denominator. It would be helpful to have a second investigator independently verify the tables against the statistical output before resubmission.

Response

Yes, agreed, and this is corrected throughout the manuscript as suggested.

Comment

1. The multivariable regression strategy requires stronger justification.

The manuscript states that variables with a univariable p-value <0.20 were entered into the multivariable logistic regression. Although such screening approaches are sometimes used, selection of confounders should not rely only on univariable statistical significance. Important variables may fail to meet an arbitrary p-value threshold yet remain necessary for adjustment on epidemiological grounds.

The authors should clarify whether variables were selected based on prior literature, a conceptual framework, or statistical screening. A more defensible approach would retain key prespecified covariates that are plausible confounders regardless of univariable significance. The authors should also report the final model-building strategy transparently, including whether interaction terms were considered, how continuous variables were assessed for linearity in the logit, and whether influential observations were evaluated. Given the inconsistencies in Table 2, the multivariable model should be rerun after the dataset is verified.

Response

Yes, agreed, and the details of the statistics and modeling with covariates are shown. Please see statistics.

Comment

1. BMI measured at delivery should not be interpreted as a straightforward determinant of HRFB.

Weight and height were measured at admission for delivery, and the manuscript correctly acknowledges that this BMI includes gestational weight gain. This creates an important temporality problem. HRFB incorporates prior reproductive history, including parity and interpregnancy interval, whereas BMI was measured at the end of the current pregnancy. It is therefore not possible to determine whether higher BMI preceded the fertility pattern, resulted partly from repeated pregnancies, or both.

The authors should consistently describe BMI as an “associated factor” rather than a determinant or causal predictor. Statements suggesting that weight management would reduce HRFB should be substantially moderated unless supported by longitudinal evidence. The possibility of reverse causality is already recognized in the Discussion and should also shape the Abstract and Conclusion.

Response

Yes, agreed, and this point has been inserted as a limitation, and we have deleted most of the recommendations related to BMI as suggested. Please see them.

Comment

1. The sampling description is contradictory and should be rewritten.

The manuscript states that all women who gave birth during the study period were invited to participate, but later describes systematic random sampling in which every fifth eligible woman was approached. These are different sampling procedures.

The authors should provide a single, precise description of recruitment. Please clarify how the random starting point was selected, whether every fifth admission or every fifth eligible admission was sampled, how women who were ineligible or declined participation were handled within the sampling sequence, and whether replacement participants were selected. The statement that the procedure yielded a “highly representative” sample should also be softened because this was a single tertiary hospital and representativeness is limited by the facility-based design.

Response

Yes, agreed, and the sampling was edited as suggested. Please see it

Comment

 

1. The sample-size calculation is not sufficiently reproducible.

The authors report use of a single-population proportion formula, a presumed prevalence of 50%, and a 95% confidence level, resulting in n=380. However, the margin of error/precision is not clearly stated. The same paragraph also introduces an assumption that inadequate ANC doubles the risk of HRFB, which appears to relate to an exposure–outcome comparison rather than a simple prevalence calculation.

Please provide the exact sample-size formula and all assumptions, including expected prevalence, confidence level, precision, design effect, and anticipated non-response. If a separate calculation was conducted to detect an association between ANC and HRFB, this should be reported separately and the larger required sample size should be identified.

Response

Yes, agreed, and more details are given. Please see lines 140147.

Comment

1. The interpretation of the cross-sectional findings should be more cautious.

The manuscript frequently uses language such as “determinants,” “drivers,” and intervention-oriented statements implying that changing education or BMI would reduce HRFB. A cross-sectional design can identify associations but cannot establish temporal or causal relationships. This is particularly important for BMI, but it also applies to other exposures.

Please revise the title, Abstract, Results, Discussion, and Conclusion to use association-based terminology consistently. The Discussion should distinguish clearly between what the study observed and what is hypothesized based on previous literature. Speculative explanations should be shortened unless they are directly supported by the present data.

Response

Yes, agreed the manuscript was revised thoroughly for these terms. Please see it.

Comment

1. Generalizability is narrower than the manuscript sometimes implies.

This was a single-center study conducted at a tertiary maternity hospital during an active conflict. Women who reached this hospital may differ substantially from women who delivered elsewhere or were unable to access institutional care. The conflict may also have changed referral patterns, travel, healthcare access, and the clinical profile of women presenting for delivery.

The authors appropriately mention these issues in the Limitations section, but the wording elsewhere should be aligned with this limitation. Statements such as “prevalence among women in central Sudan” may overstate population representativeness. The conclusions should instead refer to women delivering at the study hospital or to the hospital-based sample unless population-based inference can be justified.

Response

Yes, agreed and it has been added as suggested.

Comment

1. The Discussion should be reorganized to focus more directly on the study findings.

The Discussion is relatively long and at times moves from the observed associations to broad explanations involving sociocultural norms, contraception, obesity, psychological mechanisms, and reproductive decision-making. Some of these interpretations are plausible, but they are not measured in the present study.

I suggest restructuring the Discussion around four clear components: (1) prevalence of HRFB and comparison with regional literature; (2) maternal education; (3) BMI, with strong emphasis on temporality and reverse causation; and (4) non-significant findings such as ANC, residence, husband’s education, and employment. The final paragraphs should then summarize strengths, limitations, and implications. This would reduce repetition and improve the scientific focus.

Response

Yes, agreed, and it has been restructured as suggested.

Comment

1. Several reporting and manuscript-integrity details must be corrected.

The manuscript contains additional inconsistencies that should be addressed before publication. The Author Contributions section includes initials that do not clearly correspond to the authors listed on the title page. The manuscript states “Funding: None,” while the Acknowledgments section mentions financial support from Qassim University. If this support covered only the article processing charge, this should be stated clearly and consistently in the appropriate section.

The reported confidence interval for HRFB prevalence also appears to contain a typographical error (“399–50.0%”). Please conduct a complete consistency check of author initials, funding statements, percentages, confidence intervals, abbreviations, table labels, and numerical values across all sections.

Response

Yes, agreed, and it has been corrected.

Comment

1. STROBE reporting should be strengthened.

The manuscript states that STROBE guidelines were strictly followed. However, several items would benefit from fuller reporting, including the exact participant flow, handling of refusals within the systematic sampling scheme, rationale for categorized continuous variables, missing-data procedures, and the basis for confounder selection.

Please review the manuscript against the STROBE checklist item by item and revise accordingly. If a completed STROBE checklist is required by the journal, the authors should provide one at resubmission.

Response

Yes, agreed, and it has been used throughout.

Comment

Minor Comments

Please correct typographical errors and spacing problems throughout the manuscript, including inconsistent spacing before citations and around parentheses.

Use one term consistently for interpregnancy interval (IPI) and ensure that the abbreviation is defined once and used consistently thereafter.

Avoid alternating between “high-risk fertility behavior” and “high-risk births” when these terms are not conceptually identical.

Clarify whether parity and birth order are used interchangeably. If not, define each variable precisely.

For continuous BMI in logistic regression, state clearly that the odds ratio represents the change in odds per 1 kg/m² increase. This note is already provided in Table 3 and should be reflected consistently in the text.

Consider reporting exact 95% confidence intervals for the overall HRFB prevalence and for important subgroup estimates.

The phrase “100% completion rate” should be reconciled with the statement that 10 women declined participation. Presumably, the authors mean there were no missing values among the 380 enrolled participants.

Please verify the terminology used for ANC. The WHO model refers to antenatal contacts rather than merely visits; the manuscript should use terminology consistent with the source cited.

The Conclusion should be shortened and limited to findings supported by the study design. Policy recommendations should be framed cautiously.

English-language editing is recommended after the scientific revisions are completed.

Response

The typographical errors and spacing problems throughout the manuscript are corrected.

One term is consistently used for interpregnancy interval (IPI).

“high-risk fertility behavior is used throughout the manuscript as requested.

Parity is used throughout the work as suggested.

For continuous BMI in logistic regression, it is clear that the odds ratio represents the change in odds per 1 kg/m² increase. t. Please see line 258.

The exact 95% confidence intervals for the overall HRFB prevalence and for important subgroup estimates are added as requested. Please see line 258,

Yes, we mean no missing data. This has been corrected. Please see line 164

Yes, agreed, and contacts are used throughout the work, and references are inserted.

The Conclusion is shortened and limited to findings supported by the study design. Policy recommendations are framed cautiously.

Kind Regards

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

Major comments

1. Definition of high-risk fertility behavior (HRFB)
   The definition of HRFB is inconsistent throughout the manuscript. The Methods section refers to “four risks,” but only three risk factors are ultimately included in the analysis (maternal age >35 years, parity >4, and interpregnancy interval <24 months). In addition, women aged <18 years are mentioned in the general definition but subsequently excluded from the study definition. Please provide one clear and consistent definition of HRFB and apply it consistently throughout the Abstract, Introduction, Methods, Results, and Discussion.
2. Inconsistencies in the reported HRFB numbers
   There is a major numerical inconsistency in the Abstract. The manuscript reports 171 women (45.0%) with HRFB, but subsequently states that 250 women had a single HRFB and 88 had multiple HRFBs, which is incompatible with the total number of women with HRFB. The Results section reports 52 women (13.7%) with multiple HRFBs. Please carefully verify and correct all HRFB numbers throughout the manuscript. The reported 95% CI for the prevalence (39.9–50.0%) should also be checked and corrected.
3. Clarification of the age-related HRFB criterion
   The manuscript initially defines HRFB as including maternal age <18 or >35 years; however, the Methods section states that only age >35 years was considered a high-risk criterion because of the legal context in Sudan. Moreover, no results are presented for women aged <18 years. The authors should clearly explain the final age criterion used in the analysis and ensure consistency across all sections.
4. Interpretation of BMI
   BMI was measured at delivery and therefore reflects late-pregnancy BMI rather than pre-pregnancy BMI. Since parity and short interpregnancy intervals are components of HRFB and may themselves be related to maternal weight, the observed association between BMI and HRFB should not be interpreted causally. The Discussion should use cautious language and consistently refer to an “association” rather than suggesting that high BMI contributes to or causes HRFB.
5. Sampling and representativeness
   The statement that the systematic sampling approach “capture[d] a highly representative sample” appears too strong for a single-center, hospital-based study. The sample may not be representative of the general population of Sudan or even of all women in Gezira State. Please revise this statement and describe the sampling procedure more cautiously, while acknowledging the potential for selection bias.
6. Numerical inconsistencies in Tables 1 and 2
   Several numerical inconsistencies require verification against the original dataset. For example, in Table 1, the numbers reported for urban and rural residence (225 and 55) sum to 280 rather than 380, and 225/380 does not correspond to 85.5%. In Table 2, the frequencies reported for husband’s education also appear inconsistent with the denominators and percentages. Please carefully cross-check all tables against the original database and statistical output before resubmission

Author Response

We thank the editor and the reviewers for their valuable comments on the manuscript.  The comments/response have dramatically improved the manuscript.

Comment

Major comments

1. Definition of high-risk fertility behavior (HRFB)
   The definition of HRFB is inconsistent throughout the manuscript. The Methods section refers to “four risks,” but only three risk factors are ultimately included in the analysis (maternal age >35 years, parity >4, and interpregnancy interval <24 months). In addition, women aged <18 years are mentioned in the general definition but subsequently excluded from the study definition. Please provide one clear and consistent definition of HRFB and apply it consistently throughout the Abstract, Introduction, Methods, Results, and Discussion.

Response

Response

Yes, agreed, and a consistent definition of HRFB is used with three clear criteria in our study, while other studies used four criteria (age less than 18). Please see it.

Comment

1. Inconsistencies in the reported HRFB numbers
   There is a major numerical inconsistency in the Abstract. The manuscript reports 171 women (45.0%) with HRFB, but subsequently states that 250 women had a single HRFB and 88 had multiple HRFBs, which is incompatible with the total number of women with HRFB. The Results section reports 52 women (13.7%) with multiple HRFBs. Please carefully verify and correct all HRFB numbers throughout the manuscript. The reported 95% CI for the prevalence (39.9–50.0%) should also be checked and corrected.

Response

Yes, agreed, and the numbers and values were checked throughout the manuscript. We apologize for the inconvenience.

 

1. Clarification of the age-related HRFB criterion
   The manuscript initially defines HRFB as including maternal age <18 or >35 years; however, the Methods section states that only age >35 years was considered a high-risk criterion because of the legal context in Sudan. Moreover, no results are presented for women aged <18 years. The authors should clearly explain the final age criterion used in the analysis and ensure consistency across all sections.

Response

Yes, agreed, and a consistent definition of HRFB is used with three clear criteria in our study, while other studies used four criteria (age less than 18). Please see it.

Comment

1. Interpretation of BMI
   BMI was measured at delivery and therefore reflects late-pregnancy BMI rather than pre-pregnancy BMI. Since parity and short interpregnancy intervals are components of HRFB and may themselves be related to maternal weight, the observed association between BMI and HRFB should not be interpreted causally. The Discussion should use cautious language and consistently refer to an “association” rather than suggesting that high BMI contributes to or causes HRFB.

Response

Yes, agreed, and this point has been inserted as a limitation, and we have deleted most of the recommendations related to BMI as suggested. Please see them.

 

Comment

1. Sampling and representativeness
   The statement that the systematic sampling approach “capture[d] a highly representative sample” appears too strong for a single-center, hospital-based study. The sample may not be representative of the general population of Sudan or even of all women in Gezira State. Please revise this statement and describe the sampling procedure more cautiously, while acknowledging the potential for selection bias.

Response

Yes, agreed and corrected.

Comment

1. Numerical inconsistencies in Tables 1 and 2
   Several numerical inconsistencies require verification against the original dataset. For example, in Table 1, the numbers reported for urban and rural residence (225 and 55) sum to 280 rather than 380, and 225/380 does not correspond to 85.5%. In Table 2, the frequencies reported for husband’s education also appear inconsistent with the denominators and percentages. Please carefully cross-check all tables against the original database and statistical output before resubmission

Response

Yes, agreed, and this has been corrected. My deep apologies for the inconvenience.

Kind Regards

 

Round 2

Reviewer 1 Report (New Reviewer)

Comments and Suggestions for Authors

Thank you to the authors for their careful revisions and for addressing the previous comments. The manuscript has been substantially improved. Only one point remains that requires clarification:

1.Most of the numerical inconsistencies identified in the previous review have been corrected. However, a substantial discrepancy remains for residence. Table 1 reports 225 urban and 155 rural participants, whereas Table 2 reports 143+182=325 urban and 28+27=55 rural participants. Please verify the raw data, correct the residence frequencies consistently throughout the manuscript, and confirm/recalculate the corresponding odds ratio and confidence interval in Table 2 as necessary.

Author Response

We thank the editor and reviewers for their valuable comments on the manuscript. The comments/response have dramatically improved the manuscript.

Reviewers 1

1.Most of the numerical inconsistencies identified in the previous review have been corrected. However, a substantial discrepancy remains for residence. Table 1 reports 225 urban and 155 rural participants, whereas Table 2 reports 143+182=325 urban and 28+27=55 rural participants. Please verify the raw data, correct the residence frequencies consistently throughout the manuscript, and confirm/recalculate the corresponding odds ratio and confidence interval in Table 2 as necessary.

Response

Okay, thank you so much for this help, and we do appreciate your time and effort. My deepest apologies for this inconvenience.  The discrepancy for residence is corrected. Please see it now.

Kind Regards

 

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Comments to the Authors

The revised manuscript has been substantially improved, and most of the previous concerns have been adequately addressed. The topic is relevant and the manuscript is close to publication; however, a few points still require clarification.

First, BMI measured at delivery reflects both maternal adiposity and gestational weight gain and should not be interpreted as pre-pregnancy BMI. This qualification should be stated consistently in the Results and Discussion.

Second, excluding maternal age <18 years from the conventional HRFB definition on the basis of Sudanese marriage law may underestimate HRFB and limit comparability with international studies. This should be acknowledged more clearly as a methodological limitation, particularly because teenage pregnancy may still occur despite legal restrictions.

Please also correct several reporting inconsistencies. Residence totals differ between Tables 1 and 2 (225/155 versus 325/55), and the stated response rate of 97.4% is inconsistent with 380 participants among 402 women approached (94.5%); 97.4% applies only after excluding ineligible women. In addition, an AOR of 1.08 should be described as an 8% increase in the odds of HRFB per 1 kg/m2 increase in BMI, rather than as a '0.08 increased AOR.'

Ethical Considerations

Ethical approval and informed consent are reported. Nevertheless, the authors should clarify whether an adapted consent procedure was used for women with limited literacy. The exclusion of women with physical disabilities that prevented anthropometric assessment should also be justified to address potential selection inequity. Finally, because one author is an associate editor of the journal, the manuscript or editorial record should confirm that independent editorial handling and peer review were ensured.

Final Recommendation

A final language and formatting check is recommended to correct minor grammatical and typographical inconsistencies. Overall, the manuscript is suitable for publication after revision.

Author Response

We thank the editor and reviewers for their valuable comments on the manuscript. The comments/response have dramatically improved the manuscript.

Comment

First, BMI measured at delivery reflects both maternal adiposity and gestational weight gain and should not be interpreted as pre-pregnancy BMI. This qualification should be stated consistently in the Results and Discussion.

Response

Yes, agreed, and it has been mentioned in the Results and Discussion as suggested. Please see lines 236-239, 326-328.

Comment

Second, excluding maternal age <18 years from the conventional HRFB definition on the basis of Sudanese marriage law may underestimate HRFB and limit comparability with international studies. This should be acknowledged more clearly as a methodological limitation, particularly because teenage pregnancy may still occur despite legal restrictions.

Response

Yes, agreed, and I inserted it as a limitation as suggested. Please see lines 380-384.

Comment

Please also correct several reporting inconsistencies. Residence totals differ between Tables 1 and 2 (225/155 versus 325/55), and the stated response rate of 97.4% is inconsistent with 380 participants among 402 women approached (94.5%); 97.4% applies only after excluding ineligible women. In addition, an AOR of 1.08 should be described as an 8% increase in the odds of HRFB per 1 kg/m2 increase in BMI, rather than as a '0.08 increased AOR.'

Response

. My deepest apologies for this inconvenience.  The discrepancy for residence is corrected. Please see it now.

The point” 380 participants among 402 women approached (94.5%); 97.4% applies only after excluding ineligible women” is corrected. Please see line 227.

Comment

The “ In addition, an AOR of 1.08 should be described as an 8% increase in the odds of HRFB per 1 kg/m2 increase in BMI, rather than as a '0.08 increased AOR.”  Is corrected. Response

Yes, agreed and corrected. Please see line 258.

Comment

Ethical Considerations

Ethical approval and informed consent are reported. Nevertheless, the authors should clarify whether an adapted consent procedure was used for women with limited literacy. The exclusion of women with physical disabilities that prevented anthropometric assessment should also be justified to address potential selection inequity. Finally, because one author is an associate editor of the journal, the manuscript or editorial record should confirm that independent editorial handling and peer review were ensured.

Response

The “ an adapted consent procedure was used for women with limited literacy” is inserted as suggested. Please see it

The “ exclusion of women with physical disabilities that prevented anthropometric assessment should also be justified to address potential selection inequity” is inserted as suggested.

The “one author is an associate editor of the journal, the manuscript or editorial record should confirm that independent editorial handling and peer review were ensured” is removed because none of the authors is on the editorial board.

 

Comment

Final Recommendation

A final language and formatting check is recommended to correct minor grammatical and typographical inconsistencies. Overall, the manuscript is suitable for publication after revision.

Response

Yes, agreed, and it has been revised.

Kind Regards

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors, The manuscript is not very interesting, it explores already known topics, but it is well-written and therefore requires minor revisions.

Here are some suggestions.

1) Please include the study period in the abstract.
2) Please include MeshTerms in keywords if possible.
3) Please implement section 2.1; understanding the context is essential for an international reader.
4) Explain the sampling method in more detail; it is unclear.
5) In section 2.7, write "For previous studies [2]..." but cite only one study.
6) Reformat the tables to be clearer.
7) In discussions, try to also clarify the mental mechanisms that increased HRFB.
8) The conclusions should be implemented by providing guidance to policymakers on how to improve the situation.

Kind regards.

Author Response

Reviewers #2

Dear Authors, The manuscript is not very interesting, it explores already known topics, but it is well-written and therefore requires minor revisions.

Here are some suggestions.

1) Please include the study period in the abstract.
2) Please include MeshTerms in keywords if possible.
3) Please implement section 2.1; understanding the context is essential for an international reader.
4) Explain the sampling method in more detail; it is unclear.
5) In section 2.7, write "For previous studies [2]..." but cite only one study.
6) Reformat the tables to be clearer.
7) In discussions, try to also clarify the mental mechanisms that increased HRFB.
8) The conclusions should be implemented by providing guidance to policymakers on how to improve the situation.

Kind regards.

Response

1) The study period is included in the abstract as suggested. Please see lines 16-17

2) MeshTerms are included in the keywords. Please see them

3) Section 2.1 is implemented as suggested.

4) The sampling method is included in more detail as suggested. Please see lines 111-134.

5) In section 2.7, "For previous studies,” more references are cited as requested.

6) Tables are reformatted to be clearer.

7) Mental mechanism is added as suggested. Please see lines 350=353

8) The conclusion is edited.

Kind regards.

 



We thank the editor and reviewers for their valuable comments on the manuscript. The comments/responses have dramatically improved the manuscript

Reviewer 2 Report

Comments and Suggestions for Authors

• HRFB was defined using maternal age, parity, and interpregnancy interval (IPI). However, the majority of HRFB cases appear to be driven by high parity and short IPI. The authors report that higher BMI is associated with HRFB, yet they also acknowledge that multiparity and repeated pregnancies may contribute to increased BMI. Therefore, the direction of the association remains unclear, and reverse causality cannot be excluded. This issue should be discussed more explicitly.
• The prevalence of HRFB was extremely high (88.9%), with only 42 women classified as non-HRFB. Such an imbalance raises concerns regarding the stability of the logistic regression model and the interpretation of the reported odds ratios. Please provide additional information on how model stability was assessed.
• BMI was measured at delivery rather than before pregnancy. Could gestational weight gain have confounded the observed association between BMI and HRFB? Please justify the use of delivery BMI and discuss its implications for interpretation.
• The Methods section states that missing data were handled using regression-based imputation. However, the manuscript also reports a 100% completion rate and no missing values. Please clarify this apparent inconsistency.
• The Results section reports a Nagelkerke R² value of 126. This appears to be a reporting error, as R² values cannot exceed 1. Please verify and correct this value.
• Only 42 women were classified as non-HRFB, yet several variables (including BMI, education, and employment) were included in the multivariable model. Given the limited number of outcome events in the reference group, concerns remain regarding model robustness and potential overfitting.
• A substantial proportion of the cited literature originates from the same research group (e.g., References 19, 20, 21, 35, 39, and 40). While these references are relevant, the authors may consider balancing the discussion with a broader range of external evidence.
• There appears to be an inconsistency between Table 1 and the Results section. Maternal age is reported as 25.8 years in Table 1 but 30.0 years in the Results. Please verify and correct these values.
• A similar discrepancy is present for BMI. Table 1 reports a median BMI of 32.6 kg/m², whereas the Results section reports 25.8 kg/m². Please clarify and correct this inconsistency.

Author Response

We thank the editor and reviewers for their valuable comments on the manuscript. The comments/responses have dramatically improved the manuscript.

Reviewer’s# 1

HRFB was defined using maternal age, parity, and interpregnancy interval (IPI). However, the majority of HRFB cases appear to be driven by high parity and short IPI. The authors report that higher BMI is associated with HRFB, yet they also acknowledge that multiparity and repeated pregnancies may contribute to increased BMI. Therefore, the direction of the association remains unclear, and reverse causality cannot be excluded. This issue should be discussed more explicitly.

Response

Yes, this point and the direction of the association are discussed in more detail. Please see lines 302-331.

Comment

The prevalence of HRFB was extremely high (88.9%), with only 42 women classified as non-HRFB. Such an imbalance raises concerns regarding the stability of the logistic regression model and the interpretation of the reported odds ratios. Please provide additional information on how model stability was assessed.

Response

Yes, agreed, the model stability is tested and mentioned. Please see lines 211-214, 384-386, 392-394.

Comment

BMI was measured at delivery rather than before pregnancy. Could gestational weight gain have confounded the observed association between BMI and HRFB? Please justify the use of delivery BMI and discuss its implications for interpretation.

Response

Yes, agreed, and this point “pre-pregnancy BMI is discussed. Please see Lines 326-329.

Comment

The Methods section states that missing data were handled using regression-based imputation. However, the manuscript also reports a 100% completion rate and no missing values. Please clarify this apparent inconsistency.

Response

Yes, agreed and the discrepancy has been edited. Please see it

Comment

The Results section reports a Nagelkerke R² value of 126. This appears to be a reporting error, as R² values cannot exceed 1. Please verify and correct this value.

Response

Yes,  agreed it was typo error and corrected. Please see line 248.

Comment

Only 42 women were classified as non-HRFB, yet several variables (including BMI, education, and employment) were included in the multivariable model. Given the limited number of outcome events in the reference group, concerns remain regarding model robustness and potential overfitting.

Response

Yes, agreed, and this point has been inserted. Please see lines 384-386.

Comment

A substantial proportion of the cited literature originates from the same research group (e.g., References 19, 20, 21, 35, 39, and 40). While these references are relevant, the authors may consider balancing the discussion with a broader range of external evidence.

Response

Yes, we added some new references to have a balance.

Comment

There appears to be an inconsistency between Table 1 and the Results section. Maternal age is reported as 25.8 years in Table 1 but 30.0 years in the Results. Please verify and correct these values.

A similar discrepancy is present for BMI. Table 1 reports a median BMI of 32.6 kg/m², whereas the Results section reports 25.8 kg/m². Please clarify and correct this inconsistency.

Response

Yes,  agreed and all corrected.

Kind Regards



 

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The main shortcoming of this paper is the lack of any information on ethnicity. Gezira State has three main ethnic divisions: Riverian Arabs; Western Sudanese; and West Africans. These three groups differ in ways that are relevant to HRFB:

“… each group follows a different pattern of demographic behavior. (1) The Riverian Arabs utilize the education system and their social networks to institutionalize their position of dominance. In this group, high fertility levels are associated with education and with low extra-domestic employment of women. (2) The Western Sudanese exhibit a more conventional demographic behavior, where a trade-off between the quantity and quality of children is made necessary by the limited access to education and social mobility. (3) The West Africans exhibit high levels of fertility and of women's extra-domestic employment, as well as, relatively, widespread polygamous marriages. These people rely solely on their labor, and opportunities for education and social mobility are lacking.” (Omer, 1991)

I suspect that much of the noise in the data, such as the weak association between HRFB and low maternal education, is due to the inability to control for ethnicity. Also, by understanding the relationships between ethnicity and HRFB, it would be easier to prepare better targeted interventions to improve maternal and perinatal outcomes.

Other point:

The sentence running from line 166 to line 171 is difficult to read. I would replace it with the following text:

These factors included residence (urban or rural), body mass index (BMI), couple’s educational level, and mother’s occupation (housewife or employed). Couple’s educational level had two possible values: (1)  less than secondary, i.e., participants who had no formal education or had only completed primary/intermediate schooling; and (2) secondary and above, i.e., participants who had completed secondary school (high school) or attained higher education degrees (university or postgraduate).

Reference

Omer, A. B. (1991). Ethnicity, women's employment, and fertility: A case study of the Gezira area in central Sudan. University of Michigan. https://deepblue.lib.umich.edu/items/b40d9163-816b-4282-8966-6aeb106a3445

Author Response

Reviewers # 3

The main shortcoming of this paper is the lack of any information on ethnicity. Gezira State has three main ethnic divisions: Riverian Arabs; Western Sudanese; and West Africans. These three groups differ in ways that are relevant to HRFB:

“… each group follows a different pattern of demographic behavior. (1) The Riverian Arabs utilize the education system and their social networks to institutionalize their position of dominance. In this group, high fertility levels are associated with education and with low extra-domestic employment of women. (2) The Western Sudanese exhibit a more conventional demographic behavior, where a trade-off between the quantity and quality of children is made necessary by the limited access to education and social mobility. (3) The West Africans exhibit high levels of fertility and of women's extra-domestic employment, as well as, relatively, widespread polygamous marriages. These people rely solely on their labor, and opportunities for education and social mobility are lacking.” (Omer, 1991)

I suspect that much of the noise in the data, such as the weak association between HRFB and low maternal education, is due to the inability to control for ethnicity. Also, by understanding the relationships between ethnicity and HRFB, it would be easier to prepare better targeted interventions to improve maternal and perinatal outcomes.

Other point:

The sentence running from line 166 to line 171 is difficult to read. I would replace it with the following text:

These factors included residence (urban or rural), body mass index (BMI), couple’s educational level, and mother’s occupation (housewife or employed). Couple’s educational level had two possible values: (1) less than secondary, i.e., participants who had no formal education or had only completed primary/intermediate schooling; and (2) secondary and above, i.e., participants who had completed secondary school (high school) or attained higher education degrees (university or postgraduate).

Reference

Omer, A. B. (1991). Ethnicity, women's employment, and fertility: A case study of the Gezira area in central Sudan. University of Michigan. https://deepblue.lib.umich.edu/items/b40d9163-816b-4282-8966-6aeb106a3445

 Response

The point “The main shortcoming of this paper is the lack of any information on ethnicity. Gezira State has three main ethnic divisions: Riverian Arabs; Western Sudanese; and West Africans. These three groups differ in ways that are relevant to HRFB:” is inserted in the limitation.

Please see lines 90-95.

The “The sentence running from line 166 to line 171 is difficult to read. I would replace it with the following text” is replaced as suggested. Please see lines 181-187.

Kind Regards

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript has improved after revision, and most of the reviewer comments have been addressed. However, some important methodological concerns remain. First, the multivariable logistic regression model may still be affected by overfitting because of the highly unbalanced outcome (338 vs. 42). Reporting the VIF and Hosmer–Lemeshow test does not fully address this issue. Additional evidence for model stability (e.g., EPV justification, penalized logistic regression, or internal validation) would strengthen the findings. Second, BMI was measured at delivery rather than before pregnancy. Although the authors discussed this limitation, delivery BMI may have been influenced by gestational weight gain and repeated pregnancies. Therefore, the association between BMI and HRFB should be interpreted with caution, and causality cannot be established because of the cross-sectional study design.

Author Response

We thank the editor and reviewers for their valuable comments on the manuscript. The comments/responses have dramatically improved the manuscript.

Comment

Additional evidence for model stability (e.g., EPV justification, penalized logistic regression, or internal validation) would strengthen the findings

Response

Yes, agreed, and we have added these points. Please see lines 219-224, 251-253, 391-392.

Comment

Although the authors discussed this limitation, delivery BMI may have been influenced by gestational weight gain and repeated pregnancies. Therefore, the association between BMI and HRFB should be interpreted with caution, and causality cannot be established because of the cross-sectional study design.

Response

Yes,  agreed and added. Please see lines 394-398.

Kind Regards