Comorbidities and Pregnancy-Related Risk Factors in Patients with Severe Maternal Morbidity: Application of a Validated Obstetrical Comorbidity Scoring System to a Surveillance-Identified Population

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

It is well designed and interesting article. I suggest some corrections:

The use of “Leonard Index” is appropriate, but for readers unfamiliar with the original study, a brief summary or visual (e.g., in an appendix or figure) of the 27 conditions and their weights would be beneficial. Table A1 could be expanded or referenced earlier.

P-values are provided but effect sizes or confidence intervals for key comparisons (e.g., comorbidity scores across SMM causes) would improve interpretability.

Consider presenting correlation thresholds (e.g., low/moderate/high) when interpreting Table 3 and Table 4 to aid readers less familiar with tetrachoric correlation coefficients.

Expand briefly on why comorbidity scores were not significantly different for ICU admission or inter-hospital transfer. These findings are unexpected and deserve further discussion.

The rationale behind the inclusion of some additional risk factors (e.g., nulliparity, prior hemorrhage) could be better justified with citations or theoretical context.

 

 

Author Response

Comment 1: The use of “Leonard Index” is appropriate, but for readers unfamiliar with the original study, a brief summary or visual (e.g., in an appendix or figure) of the 27 conditions and their weights would be beneficial. Table A1 could be expanded or referenced earlier.

Response 1: Thank you for this suggestion. We have added a reference to table A1 in the introduction section and have expanded the description of the index in the introduction section as well.

 

Comment 2: P-values are provided but effect sizes or confidence intervals for key comparisons (e.g., comorbidity scores across SMM causes) would improve interpretability.

Response 2: Thank you for this suggestion. We have added confidence intervals to Figure 1 and the associated text in the results section.  

 

Comment 3: Consider presenting correlation thresholds (e.g., low/moderate/high) when interpreting Table 3 and Table 4 to aid readers less familiar with tetrachoric correlation coefficients.

Response 3: Thank you for this suggestion. We have added correlation coefficient thresholds to the methods section to improve interpretability.

 

Comment 4: Expand briefly on why comorbidity scores were not significantly different for ICU admission or inter-hospital transfer. These findings are unexpected and deserve further discussion.

Response 4: Thank you for this suggestion. We have added additional information in the discussion section to explain why these unexpected findings may have occurred.

 

Comment 5: The rationale behind the inclusion of some additional risk factors (e.g., nulliparity, prior hemorrhage) could be better justified with citations or theoretical context.

Response 5: Thank you for this suggestion. We have added additional citations to support inclusion of these risk factors.

Reviewer 2 Report

Comments and Suggestions for Authors Dear Authors, Congratulations on your manuscript, which is well-structured and clearly written. The case series is extensive and well standardized. With this study, you demonstrate that the Leonard Index—an important tool for comparing SMM across centers—can also be applied to this surveillance-identified population, both for delivery-related events and for those occurring before and after childbirth. This finding is particularly relevant in view of the potential future expansion of the Maryland cohort. Moreover, you have listed additional risk factors not included in the Leonard Index but already described in the literature, which may influence the incidence of SMM. These additional risk factors, along with those already part of the Leonard Index, could serve as a basis for developing tools or risk calculators to help identify pregnancies at higher risk of severe adverse events. Please find below two minor comments:

• In Table 1, the percentage reported for "injury" appears to be incorrect.
• Although likely to be low, have you assessed the prevalence of the additional risk factors (e.g., macrosomia, FGR, etc.) in a cohort without SMM?

Best wishes for the publication of your study.

Author Response

Comment 1: In Table 1, the percentage reported for "injury" appears to be incorrect.

Response 1: Thank you for catching this error. The number with injury should be 10, not 1 and has been corrected in table 1.

 

Comment 2: Although likely to be low, have you assessed the prevalence of the additional risk factors (e.g., macrosomia, FGR, etc.) in a cohort without SMM?

Response 2: Thank you for this suggestion. We are unable to assess the prevalence of these conditions in the general population because surveillance data only includes information about patients with SMM. We have added this as a study limitation in the discussion section.

Reviewer 3 Report

Comments and Suggestions for Authors

Summary and general comments:
This manuscript investigates the application of an obstetrical comorbidity index to a surveillance-identified population of patients with severe maternal morbidity (SMM). Using detailed case review data from 978 SMM events occurring between 2020 and 2024, the study evaluates how comorbidity scores vary by cause of morbidity, hospital level, and adverse outcomes, while also identifying additional maternal, fetal, and social risk factors for consideration in future indices. The work is timely and highly relevant given ongoing efforts to refine risk-adjustment tools for maternal health outcomes. The paper underscores both the utility and the limitations of applying validated indices across different data sources. However, certain interpretive cautions and improvements in presentation would strengthen the manuscript and enhance its contribution to the field.

Specific comments:
-Introduction:
The introduction effectively situates the problem but could more explicitly state the knowledge gap addressed (i.e., the limited external validation of comorbidity indices in surveillance datasets versus administrative datasets.)
-Methods:
Please clarify whether data abstraction procedures were standardized across hospitals and whether inter-rater reliability was assessed. This is important for validity.
The rationale for choosing chi-square tests in some comparisons (rather than logistic regression adjusting for confounders) should be briefly discussed.
-Results:
Tables are comprehensive but may overwhelm the reader; highlighting the most clinically relevant comparisons in the text would improve readability.
-Discussion:
While the authors note that comorbidity scores vary by cause of SMM, more reflection is needed on the implications for applying the index in quality improvement versus research contexts.
Limitations are acknowledged, but the potential impact of missing data from the six non-participating hospitals should be quantified or at least discussed more concretely.
The term “late postpartum” should be consistently defined across sections.
-Overall Contribution:
The manuscript provides a strong application of a validated comorbidity index in a novel dataset, and the identification of additional candidate risk factors advances the field. With additional methodological clarity and sharper discussion, this study could have a substantial impact on maternal health risk stratification research. However, some changes must be adjusted before the manuscript is ready for publication.

Author Response

Comment 1: The introduction effectively situates the problem but could more explicitly state the knowledge gap addressed (i.e., the limited external validation of comorbidity indices in surveillance datasets versus administrative datasets.)

Response 1: Thank you for this suggestion. We have added additional information to the introduction to clarify the limited application of comorbidity indices to SMM surveillance data in prior research.

 

Comment 2: Please clarify whether data abstraction procedures were standardized across hospitals and whether inter-rater reliability was assessed. This is important for validity.

Response 2: Thank you for this question. Data abstraction is standardized, and we have now clarified this in the introduction section. We have not assessed inter-rater reliability, which is now included as a limitation in the discussion section.

 

Comment 3: The rationale for choosing chi-square tests in some comparisons (rather than logistic regression adjusting for confounders) should be briefly discussed.

Response 3: We are using chi-square tests to examine key associations in an attempt to validate the use of an obstetric comorbidity index in a high-risk population. We were not aiming to predict an outcome based on one or more independent variables and this is why we do not use regression modeling. We can include this justification in the manuscript if the editor and the reviewers consider it necessary.

 

Comment 4: Tables are comprehensive but may overwhelm the reader; highlighting the most clinically relevant comparisons in the text would improve readability.

Response 4: We consider all shown comparisons to be important and would like to keep the tables and text as they are. If the editor or reviewers want us to highlight specific findings, we can do so. Please let us know — thank you.

 

Comment 5: While the authors note that comorbidity scores vary by cause of SMM, more reflection is needed on the implications for applying the index in quality improvement versus research contexts.

Response 5: Thank you for this suggestion, we have added additional detail to the conclusion regarding the implications of comorbidity scores for quality improvement.

 

Comment 6: Limitations are acknowledged, but the potential impact of missing data from the six non-participating hospitals should be quantified or at least discussed more concretely.

Comment 6: Thank you for this comment we have added information to the limitations section to clarify that hospitals contributing data make up >90% of Maryland births.

 

Comment 7: The term “late postpartum” should be consistently defined across sections.

Response 7: Thank you for this comment, we have revised the headings of table 2 to clarify the designated time periods. Late postpartum refers to >8 hours after delivery throughout.

 

Overall Contribution:
The manuscript provides a strong application of a validated comorbidity index in a novel dataset, and the identification of additional candidate risk factors advances the field. With additional methodological clarity and sharper discussion, this study could have a substantial impact on maternal health risk stratification research. However, some changes must be adjusted before the manuscript is ready for publication.

Response: Thank you for the comment.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The corrections made have substantially improved the manuscript.