Datasets:

JerMa88
/

BMC_Medicine_Peer_Reviews

title stringlengths 32 37	abstract stringclasses 1 value	full_text stringlengths 117 61.6k	review stringlengths 117 61.6k	source stringclasses 1 value	review_src stringlengths 37 41	criteria_count dict	dimension_scores dict	overall_score float64 0 5	conversations listlengths 2 2	year int64 2k 2.03k
Review for 10.1186/s12916-025-03982-7	Reviewed - bmc_medicine	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 0 Date: 17 Oct 2024 Reviewer's report: I read with interest the descriptive study, "Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups" by Zhang et al. The research team performed ITS1 sequencing of pharyngeal swab samples from PH Groups 1, 3, and 4 along with a group with lung disease without PH. They also distinguished patients based on whether they had recent corticosteroid and/or antibiotic exposure. They identified a distinct mycobiome profile between control and PH patients (and controlled for corticosteroid/antibiotic exposure) and between Group 3 PH and lung disease patients. The novelty of the study is that the mycobiome in understudied in pulmonary hypertension and all disease states. I recommend completion of the following additional analyses and modifications to improve the impact of the study: 1. The authors should complete ROC analysis to determine whether the pharyngeal mycobiome can distinguish the PH Group types as the pathophysiology of each of these PH types is different (Group 1 vs 3, Group 1 vs 4, Group 3 vs 4). 2. Ideally, the authors should have integrated metagenomic/microbiome data with the mycobiome data as alterations in the microbiome would have changed the entire micro/ mycobiome. 3. Additional transparency regarding the characteristics of the PH patients by PH type should be included in the characteristics table such has the average hemodynamics (mPAP, PVR, CO/ CI) and all therapies that each of the PH groups were on (pulmonary vasodilators, inhaled treprostinil, cortisteroid/antibiotics, anticoagulation, etc.) - not just clustering all data within all of PH. 4. It is unclear to me why any Group 1 or 4 PH patients would have taken corticosteroids and/ or antibiotics (recent infection?). Including the reasons why each of the Group 1 and 4 patients are on corticosteroids/antibiotics in the supplement would be beneficial. 5. Please discuss in the discussion why Group 2 PH patients were not included in the study as it is a more prevalent PH type. 6. Why was the mPAP subgroup analysis/breakdown not based on typical cutoffs for mild (mPAP 20-34), moderate (mPAP 35-44), and severe (mPAP >/= 45 mmHg).	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 0 Date: 17 Oct 2024 Reviewer's report: I read with interest the descriptive study, "Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups" by Zhang et al. The research team performed ITS1 sequencing of pharyngeal swab samples from PH Groups 1, 3, and 4 along with a group with lung disease without PH. They also distinguished patients based on whether they had recent corticosteroid and/or antibiotic exposure. They identified a distinct mycobiome profile between control and PH patients (and controlled for corticosteroid/antibiotic exposure) and between Group 3 PH and lung disease patients. The novelty of the study is that the mycobiome in understudied in pulmonary hypertension and all disease states. I recommend completion of the following additional analyses and modifications to improve the impact of the study: 1. The authors should complete ROC analysis to determine whether the pharyngeal mycobiome can distinguish the PH Group types as the pathophysiology of each of these PH types is different (Group 1 vs 3, Group 1 vs 4, Group 3 vs 4). 2. Ideally, the authors should have integrated metagenomic/microbiome data with the mycobiome data as alterations in the microbiome would have changed the entire micro/ mycobiome. 3. Additional transparency regarding the characteristics of the PH patients by PH type should be included in the characteristics table such has the average hemodynamics (mPAP, PVR, CO/ CI) and all therapies that each of the PH groups were on (pulmonary vasodilators, inhaled treprostinil, cortisteroid/antibiotics, anticoagulation, etc.) - not just clustering all data within all of PH. 4. It is unclear to me why any Group 1 or 4 PH patients would have taken corticosteroids and/ or antibiotics (recent infection?). Including the reasons why each of the Group 1 and 4 patients are on corticosteroids/antibiotics in the supplement would be beneficial. 5. Please discuss in the discussion why Group 2 PH patients were not included in the study as it is a more prevalent PH type. 6. Why was the mPAP subgroup analysis/breakdown not based on typical cutoffs for mild (mPAP 20-34), moderate (mPAP 35-44), and severe (mPAP >/= 45 mmHg).	BMC	12916_2025_3982_ReviewerReport_V0_R1.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 1, "materials_and_methods": 12, "praise": 0, "presentation_and_reporting": 2, "results_and_discussion": 1, "suggestion_and_solution": 6, "total": 19 }	{ "criticism": 0.05263157894736842, "example": 0, "importance_and_relevance": 0.05263157894736842, "materials_and_methods": 0.631578947368421, "praise": 0, "presentation_and_reporting": 0.10526315789473684, "results_and_discussion": 0.05263157894736842, "suggestion_and_solution": 0.3157894736842105 }	1.210526	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and\nSubgroups\nVersion: 0 Date: 17 Oct 2024\nReviewer's report:\nI read with interest the descriptive study, \"Distinct Airway Mycobiome Signature in Patients\nwith Pulmonary Hypertension and Subgroups\" by Zhang et al. The research team performed\nITS1 sequencing of pharyngeal swab samples from PH Groups 1, 3, and 4 along with a group\nwith lung disease without PH. They also distinguished patients based on whether they had\nrecent corticosteroid and/or antibiotic exposure. They identified a distinct mycobiome profile\nbetween control and PH patients (and controlled for corticosteroid/antibiotic exposure) and\nbetween Group 3 PH and lung disease patients. The novelty of the study is that the mycobiome\nin understudied in pulmonary hypertension and all disease states. I recommend completion of\nthe following additional analyses and modifications to improve the impact of the study:\n1. The authors should complete ROC analysis to determine whether the pharyngeal mycobiome\ncan distinguish the PH Group types as the pathophysiology of each of these PH types is different\n(Group 1 vs 3, Group 1 vs 4, Group 3 vs 4).\n2. Ideally, the authors should have integrated metagenomic/microbiome data with the\nmycobiome data as alterations in the microbiome would have changed the entire micro/\nmycobiome.\n3. Additional transparency regarding the characteristics of the PH patients by PH type should\nbe included in the characteristics table such has the average hemodynamics (mPAP, PVR, CO/\nCI) and all therapies that each of the PH groups were on (pulmonary vasodilators, inhaled\ntreprostinil, cortisteroid/antibiotics, anticoagulation, etc.) - not just clustering all data within all\nof PH.\n4. It is unclear to me why any Group 1 or 4 PH patients would have taken corticosteroids and/\nor antibiotics (recent infection?). Including the reasons why each of the Group 1 and 4 patients\nare on corticosteroids/antibiotics in the supplement would be beneficial.\n5. Please discuss in the discussion why Group 2 PH patients were not included in the study as\nit is a more prevalent PH type.\n6. Why was the mPAP subgroup analysis/breakdown not based on typical cutoffs for mild\n(mPAP 20-34), moderate (mPAP 35-44), and severe (mPAP >/= 45 mmHg).", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03982-7	Reviewed - bmc_medicine	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 0 Date: 15 Nov 2024 Reviewer's report: I am reviewing a manuscript titled "Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups" submitted for publication by Dr. Zhang and colleagues. Overall, the manuscript is well written and well described. The authors put in detailed work in their publication. There are some issues I would like to see addressed: 1. There is a lot of information that is not clearly connected in the manuscript. The message and hypothesis needs to be tightened up and made more cohesive. As it stands, the manuscript has interesting conclusions and observations, but perhaps tying it to more cause would actually help the manuscript give more cohesion. For example, describing or using two different classifications in the manuscript between clinical groups and WHO classifications is difficult to understand the choice of authors. I suggest that the authors choose one and then put the other groups in the supplementary figure. 2. The details surrounding antibiotic use should be better described e.g, what antibiotics were the participants exposed to. And the authors should consider explaining in detail how the mycobiome may be affected by antibiotics which should not directly affect this form of -oMics. 3. Finally, One important aspect is that pulmonary hypertension is a symptom of larger diagnoses, for example, COPD and cor pulmonale - could the authors do subanalyses associated with diagnoses instead of classes and WHO groups? I will be using the word document page numbers and leftmost line numbers. Abstract: Page 2, Line 61 "Dramatically opposite patterns of the two major fungal phyla, Ascomycota and Basidiomycota..." I suggest tempering down this statement, which borders on being an over- statement. Background: Page 5, Line 96 "Although PH is mainly a cardiopulmonary disease, recent study has proposed that it could be considered a systemic disease affecting multiple organs, such as the lungs, heart, central nervous system, bone marrow, and gut. " I think there's some issues here is pulmonary hypertension is not a disease in it of itself, it is rather a symptom or a syndrome associated with multiple other diseased. I would carefully re-write this sentences to reflect this. Page 7 Line 123 "Extended recruitment procedure and methods are provided in the online Supplementary Material." In regards to the recruitment, can the authors provide a consort diagram of people screened for recruitment. I would like to see the number of patients who were excluded. Moreover, if all patients/participants received a RHC, could the authors also discuss if the RHC was to make the diagnosis or/and if it includes a table of the results from the RHC. Supplemental Material Page 2, Line 28 "The PH group was further divided into the subgroups of non-use of antibiotics and/or corticosteroids (non-ANT/CORT) and use of ANT, CORT, or ANT+CORT within 1 mont" Can the authors explain why this was done? This is a good category to analyze, but I would consider the PH WHO categories more clinical relevant. Supplemental Material Page 2, Line 30 "The patients were diagnosed with chronic obstructive pulmonary disease, bronchiectasis, interstitial lung disease, and other relevant conditions. All of the patients experienced dyspnea, but their echocardiographic probability of PH was low. " I'm surprised that the pH probability was low - can the authors also include the these results in a table for review? Supplemental Material Page 2, Line 34 "The healthy-control (CON) group was free of respiratory diseases and other serious diseases and had no oral and respiratory infections within 1 month. " This wasn't clear, but did these patients/controls also receive RHC? Or was there just for samples? Supplemental Material: Can the authors include IRB protocol numbers and Patient assurances for each of their sites? While the authors include multiple hospitals, they do not include the IRB protocol approvals - please include these assurances. Supplemental Material, Page 2, Line 64 "The extracted fungal genome DNA from positively detected pharyngeal swab sample and pharyngeal swab blank control and buffer blank control were verified by agarose electrophoresis, as shown in Supplementary Figure S2b." Were any control samples were obtained? Supplemental Material, Page 7, Line 137 "Figure S2: Quality control for pharyngeal swab samples a: Bar graphs showing the averaged concentration of PCR amplified DNA products and assigned " For Figure S2, please include all points for Panels A, for Panel C please also include averages of control and PH - these are difficult to interpret. Page 8, Line 148 "The #-diversity was performed using mothur (version v. 1.30.2, https:// mothur.org/wiki/calculators/), with the evaluation index applied at a similarity level of 97% for OTUs (0.97). The #-diversity distance matrix is computed by Qiime (2020.2.0, http://qiime.org/ install/index.html)." I am wondering why the authors decided to use two different pipelines to calculate alpha and beta diversity. I would suggest using one, rather than another. Stastical anlaysis Page 7 to 9 I am wondering if this would be better moved into the supplementary and the discussion about patient recruitment moved to the paper methods. Results: Page 10, Line 177 "As shown in Figure 1a, #-diversity indicated no statistical difference in mycobiome richness (Sobs index and Ace index), while significantly lower diversity (indicated by decreased Shannon index and increased Simpson index) was observed in patients with PH than in CON. For #-diversity, both non-metric multidimensional scaling (NMDS) using the Bray-Curtis distance algorithm (Figure 1b) and Partial Least Squares Discriminant Analysis (PLS-DA) (Figure 1c)" The p-values should be expressed on each comparison, even if there are p=ns. The Beta-diversity comparisons in Panel B and Panel C are possibly significant. Page 10, Line 187 "At the genus level, Candida, Cutaneotrichosporon, Apiotrichum, and Sarocladium were the most abundant genera with the highest proportion in the PH group, while Candida, Fusarium, Cutaneotrichosporon, and Cladosporium in the CON group (Figure 1e). " Can the authors reconcile this with candida found in COVID19 and ILD. Page 12, Line 222 "As seen in Figure 2b, divergent fungal diversity was found by both #- diversity (Shannon index) and #-diversity (NMDS) analyses among three subgroups, with genera Holtermanniella, Trametes, Itersonilia, Psathyrella, and Geotrichum showing dramatic enrichment in IPAH but not in CHD-PAH or CTD-PAH." In Panels 2a, 2b, and 2c - needs p- values to be placed on the figure. Page 13, Line 259 "...PH subgroups based on the mPAP levels. Simultaneously, Candida increased dramatically in the PH-low and PH-medium groups (Figure 2d)." As it's presented, the role of candida is very interesting, have the authors considered the role of candida in other diseases and how it may relate to pulmonary hypertension? Page 14, Line 265 "ANT/CORT (Supplementary Figure S7b), and ANT (Supplementary Figure S7c), and CORT (Supplementary Figure S7d), and ANT+CORT groups (Supplementary Figure S7e), using Random Forest analysis." Can the authors also include the type of antibiotics that the patients were exposed to? Could you include this as a table? Page 14 Line 280 "he abundance of Ascomycota exhibited a gradual increase in non-ANT/ CORT, ANT, CORT, and ANT+CORT subgroups, whereas the abundance of Basidiomycota showed a gradual decline, and only ANT+CORT group and CON group had statistical difference (Figure 3d). " Is this due to relative abundance diagnosis, can the authors pull out read counts to see if this pattern is still true? Page 15, Line 289 ". LEfSe multilevel species difference discriminant analysis also indicated similar fungal genera between the CON and non-ANT/CORT (Figure 4a), or ANT (Figure 4b), or CORT (Figure 4c), or ANT+CORT (Figure 4d) groups. Moreover, the LEfSe multi- level species difference discriminant analysis also suggested similar and distinct genera between the non-ANT/CORT and ANT (Figure 4e), or CORT (Figure 4f), or ANT+CORT (Figure 4g) groups." Can the authors also use different orthogonal approaches such as DESEq2 or EdgeR to calculate other differential enrichment. Page 17, Line 337 "White blood cells and cardiac index were excluded, and the correlation between the remaining 30 clinical factors and different mycobiome taxa (phyla and genera) was analyzed" Could the authors explain why this was done? It is not clear that this is actually helpful or why these were excluded. Discussion: Page 20, Line 380 "Blockage of the metabolite, trimethylamine N-oxide, which is derived from the intestinal microbial flora and has been proven to be enriched in patients with PH..." This seems as though it's a non-sequitur - is there anything that ties the trimethylamine N-oxide to the mycobiome? I suggest this introduction is tied closer to the mycobiome. Page 21, Line 404 "Notably, through multiple comparisons, we identified stable and consistent mycobiome signatures in PH (and subgroups) compared to CON without being influenced by ANT, CORT, or both, characterized by specific enrichment of the fungal genera Purpureocillium, Issatchenkia, and Cyberlindnera and a decrease in the fungal genera Peroneutypa, Simplicillium, and Metarhizium in PH (and subgroups) compared to CON." I am not entirely sure that the authors performed multiple comparisons per se, I understand there are multiple comparisons, but the authors should consider also applying multiple comparisons to their analysis. LEFSE does not put in adjustment for multiple comparisons, but would be great if they could do this in EdgeR. Page 21, Line 411 "Moreover, a specific correlation pattern was observed between the airway mycobiome signature and clinical parameters of PH, strongly indicating a potential association between the altered airway mycobiome and disease progression of PH" I would recommend that while this is the hypothesis that the authors are testing, the presentiation of this is difficult to understand. I suggest that the authors consider significant re-organization of their work - for example - the redundancy of clinical groups and WHO groups does not help discern the message and hypothesis of the patients/particiapnts with PH. Especially, when clinicians tend to think about pulmonary hypertension as a syndrome or symptom of a larger issue - thus - I suggest to either discuss among authors which Ph groups you would like to use, but I think the use of both clinical groups and WHO groups confounds the conclusions. I suggest picking one and moving the other analysis to the supplementary. Page 21, Line 418 "Antibiotics modulate host microbiome rapidly and remarkably. 33,34 Ng et al. reported that antibiotics significantly reshape the gut microbiota composition, and postantibiotic recovery depends on multiple factors, including the host diet, community context, and environmental reservoirs. 35 Ward et al. showed that antibiotics can alter the #-diversity and #-diversity of lung mycobiome. 36 Corticosteroids are immunosuppressive agents that affect host microbiomes. In a meta-analysis of patients with COPD, asthma, and chronic rhinosinusitis, Hartmann et al. demonstrated that corticosteroids significantly affected the airway microbiome composition. 37 Our study showed similar results: the #-diversity, #-diversity, and mycobiome composition were remarkably reduced in the ANT, CORT, or ANT+CORT groups versus the non-ANT/CORT group. " I think that the authors should consider explaining how antibiotics which do not affect the mycobiome but the microbiome, how the use of antibiotics may be associated with mycobiome changes. There is a logical leap between the mycobiome and microbiome, the authors may consider bridging this gap during the discussion. Corticosteroids is much more understandable, but needs to be contextualized into the change associated with possible immune suppression. Page 23, Line 450 "A recent study reported that P. lilacinum might induce fetal fungal pneumonia in a green tree Python by excluding other potential pathogens." Not sure if this is actually helpful. Minor Page 18, Line 369 "animal pathogens,s " there is a typo here.	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 0 Date: 15 Nov 2024 Reviewer's report: I am reviewing a manuscript titled "Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups" submitted for publication by Dr. Zhang and colleagues. Overall, the manuscript is well written and well described. The authors put in detailed work in their publication. There are some issues I would like to see addressed: 1. There is a lot of information that is not clearly connected in the manuscript. The message and hypothesis needs to be tightened up and made more cohesive. As it stands, the manuscript has interesting conclusions and observations, but perhaps tying it to more cause would actually help the manuscript give more cohesion. For example, describing or using two different classifications in the manuscript between clinical groups and WHO classifications is difficult to understand the choice of authors. I suggest that the authors choose one and then put the other groups in the supplementary figure. 2. The details surrounding antibiotic use should be better described e.g, what antibiotics were the participants exposed to. And the authors should consider explaining in detail how the mycobiome may be affected by antibiotics which should not directly affect this form of -oMics. 3. Finally, One important aspect is that pulmonary hypertension is a symptom of larger diagnoses, for example, COPD and cor pulmonale - could the authors do subanalyses associated with diagnoses instead of classes and WHO groups? I will be using the word document page numbers and leftmost line numbers. Abstract: Page 2, Line 61 "Dramatically opposite patterns of the two major fungal phyla, Ascomycota and Basidiomycota..." I suggest tempering down this statement, which borders on being an over- statement. Background: Page 5, Line 96 "Although PH is mainly a cardiopulmonary disease, recent study has proposed that it could be considered a systemic disease affecting multiple organs, such as the lungs, heart, central nervous system, bone marrow, and gut. " I think there's some issues here is pulmonary hypertension is not a disease in it of itself, it is rather a symptom or a syndrome associated with multiple other diseased. I would carefully re-write this sentences to reflect this. Page 7 Line 123 "Extended recruitment procedure and methods are provided in the online Supplementary Material." In regards to the recruitment, can the authors provide a consort diagram of people screened for recruitment. I would like to see the number of patients who were excluded. Moreover, if all patients/participants received a RHC, could the authors also discuss if the RHC was to make the diagnosis or/and if it includes a table of the results from the RHC. Supplemental Material Page 2, Line 28 "The PH group was further divided into the subgroups of non-use of antibiotics and/or corticosteroids (non-ANT/CORT) and use of ANT, CORT, or ANT+CORT within 1 mont" Can the authors explain why this was done? This is a good category to analyze, but I would consider the PH WHO categories more clinical relevant. Supplemental Material Page 2, Line 30 "The patients were diagnosed with chronic obstructive pulmonary disease, bronchiectasis, interstitial lung disease, and other relevant conditions. All of the patients experienced dyspnea, but their echocardiographic probability of PH was low. " I'm surprised that the pH probability was low - can the authors also include the these results in a table for review? Supplemental Material Page 2, Line 34 "The healthy-control (CON) group was free of respiratory diseases and other serious diseases and had no oral and respiratory infections within 1 month. " This wasn't clear, but did these patients/controls also receive RHC? Or was there just for samples? Supplemental Material: Can the authors include IRB protocol numbers and Patient assurances for each of their sites? While the authors include multiple hospitals, they do not include the IRB protocol approvals - please include these assurances. Supplemental Material, Page 2, Line 64 "The extracted fungal genome DNA from positively detected pharyngeal swab sample and pharyngeal swab blank control and buffer blank control were verified by agarose electrophoresis, as shown in Supplementary Figure S2b." Were any control samples were obtained? Supplemental Material, Page 7, Line 137 "Figure S2: Quality control for pharyngeal swab samples a: Bar graphs showing the averaged concentration of PCR amplified DNA products and assigned " For Figure S2, please include all points for Panels A, for Panel C please also include averages of control and PH - these are difficult to interpret. Page 8, Line 148 "The #-diversity was performed using mothur (version v. 1.30.2, https:// mothur.org/wiki/calculators/), with the evaluation index applied at a similarity level of 97% for OTUs (0.97). The #-diversity distance matrix is computed by Qiime (2020.2.0, http://qiime.org/ install/index.html)." I am wondering why the authors decided to use two different pipelines to calculate alpha and beta diversity. I would suggest using one, rather than another. Stastical anlaysis Page 7 to 9 I am wondering if this would be better moved into the supplementary and the discussion about patient recruitment moved to the paper methods. Results: Page 10, Line 177 "As shown in Figure 1a, #-diversity indicated no statistical difference in mycobiome richness (Sobs index and Ace index), while significantly lower diversity (indicated by decreased Shannon index and increased Simpson index) was observed in patients with PH than in CON. For #-diversity, both non-metric multidimensional scaling (NMDS) using the Bray-Curtis distance algorithm (Figure 1b) and Partial Least Squares Discriminant Analysis (PLS-DA) (Figure 1c)" The p-values should be expressed on each comparison, even if there are p=ns. The Beta-diversity comparisons in Panel B and Panel C are possibly significant. Page 10, Line 187 "At the genus level, Candida, Cutaneotrichosporon, Apiotrichum, and Sarocladium were the most abundant genera with the highest proportion in the PH group, while Candida, Fusarium, Cutaneotrichosporon, and Cladosporium in the CON group (Figure 1e). " Can the authors reconcile this with candida found in COVID19 and ILD. Page 12, Line 222 "As seen in Figure 2b, divergent fungal diversity was found by both #- diversity (Shannon index) and #-diversity (NMDS) analyses among three subgroups, with genera Holtermanniella, Trametes, Itersonilia, Psathyrella, and Geotrichum showing dramatic enrichment in IPAH but not in CHD-PAH or CTD-PAH." In Panels 2a, 2b, and 2c - needs p- values to be placed on the figure. Page 13, Line 259 "...PH subgroups based on the mPAP levels. Simultaneously, Candida increased dramatically in the PH-low and PH-medium groups (Figure 2d)." As it's presented, the role of candida is very interesting, have the authors considered the role of candida in other diseases and how it may relate to pulmonary hypertension? Page 14, Line 265 "ANT/CORT (Supplementary Figure S7b), and ANT (Supplementary Figure S7c), and CORT (Supplementary Figure S7d), and ANT+CORT groups (Supplementary Figure S7e), using Random Forest analysis." Can the authors also include the type of antibiotics that the patients were exposed to? Could you include this as a table? Page 14 Line 280 "he abundance of Ascomycota exhibited a gradual increase in non-ANT/ CORT, ANT, CORT, and ANT+CORT subgroups, whereas the abundance of Basidiomycota showed a gradual decline, and only ANT+CORT group and CON group had statistical difference (Figure 3d). " Is this due to relative abundance diagnosis, can the authors pull out read counts to see if this pattern is still true? Page 15, Line 289 ". LEfSe multilevel species difference discriminant analysis also indicated similar fungal genera between the CON and non-ANT/CORT (Figure 4a), or ANT (Figure 4b), or CORT (Figure 4c), or ANT+CORT (Figure 4d) groups. Moreover, the LEfSe multi- level species difference discriminant analysis also suggested similar and distinct genera between the non-ANT/CORT and ANT (Figure 4e), or CORT (Figure 4f), or ANT+CORT (Figure 4g) groups." Can the authors also use different orthogonal approaches such as DESEq2 or EdgeR to calculate other differential enrichment. Page 17, Line 337 "White blood cells and cardiac index were excluded, and the correlation between the remaining 30 clinical factors and different mycobiome taxa (phyla and genera) was analyzed" Could the authors explain why this was done? It is not clear that this is actually helpful or why these were excluded. Discussion: Page 20, Line 380 "Blockage of the metabolite, trimethylamine N-oxide, which is derived from the intestinal microbial flora and has been proven to be enriched in patients with PH..." This seems as though it's a non-sequitur - is there anything that ties the trimethylamine N-oxide to the mycobiome? I suggest this introduction is tied closer to the mycobiome. Page 21, Line 404 "Notably, through multiple comparisons, we identified stable and consistent mycobiome signatures in PH (and subgroups) compared to CON without being influenced by ANT, CORT, or both, characterized by specific enrichment of the fungal genera Purpureocillium, Issatchenkia, and Cyberlindnera and a decrease in the fungal genera Peroneutypa, Simplicillium, and Metarhizium in PH (and subgroups) compared to CON." I am not entirely sure that the authors performed multiple comparisons per se, I understand there are multiple comparisons, but the authors should consider also applying multiple comparisons to their analysis. LEFSE does not put in adjustment for multiple comparisons, but would be great if they could do this in EdgeR. Page 21, Line 411 "Moreover, a specific correlation pattern was observed between the airway mycobiome signature and clinical parameters of PH, strongly indicating a potential association between the altered airway mycobiome and disease progression of PH" I would recommend that while this is the hypothesis that the authors are testing, the presentiation of this is difficult to understand. I suggest that the authors consider significant re-organization of their work - for example - the redundancy of clinical groups and WHO groups does not help discern the message and hypothesis of the patients/particiapnts with PH. Especially, when clinicians tend to think about pulmonary hypertension as a syndrome or symptom of a larger issue - thus - I suggest to either discuss among authors which Ph groups you would like to use, but I think the use of both clinical groups and WHO groups confounds the conclusions. I suggest picking one and moving the other analysis to the supplementary. Page 21, Line 418 "Antibiotics modulate host microbiome rapidly and remarkably. 33,34 Ng et al. reported that antibiotics significantly reshape the gut microbiota composition, and postantibiotic recovery depends on multiple factors, including the host diet, community context, and environmental reservoirs. 35 Ward et al. showed that antibiotics can alter the #-diversity and #-diversity of lung mycobiome. 36 Corticosteroids are immunosuppressive agents that affect host microbiomes. In a meta-analysis of patients with COPD, asthma, and chronic rhinosinusitis, Hartmann et al. demonstrated that corticosteroids significantly affected the airway microbiome composition. 37 Our study showed similar results: the #-diversity, #-diversity, and mycobiome composition were remarkably reduced in the ANT, CORT, or ANT+CORT groups versus the non-ANT/CORT group. " I think that the authors should consider explaining how antibiotics which do not affect the mycobiome but the microbiome, how the use of antibiotics may be associated with mycobiome changes. There is a logical leap between the mycobiome and microbiome, the authors may consider bridging this gap during the discussion. Corticosteroids is much more understandable, but needs to be contextualized into the change associated with possible immune suppression. Page 23, Line 450 "A recent study reported that P. lilacinum might induce fetal fungal pneumonia in a green tree Python by excluding other potential pathogens." Not sure if this is actually helpful. Minor Page 18, Line 369 "animal pathogens,s " there is a typo here.	BMC	12916_2025_3982_ReviewerReport_V0_R3.pdf	{ "criticism": 9, "example": 30, "importance_and_relevance": 2, "materials_and_methods": 47, "praise": 5, "presentation_and_reporting": 19, "results_and_discussion": 11, "suggestion_and_solution": 29, "total": 86 }	{ "criticism": 0.10465116279069768, "example": 0.3488372093023256, "importance_and_relevance": 0.023255813953488372, "materials_and_methods": 0.5465116279069767, "praise": 0.05813953488372093, "presentation_and_reporting": 0.22093023255813954, "results_and_discussion": 0.12790697674418605, "suggestion_and_solution": 0.3372093023255814 }	1.767442	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and\nSubgroups\nVersion: 0 Date: 15 Nov 2024\nReviewer's report:\nI am reviewing a manuscript titled \"Distinct Airway Mycobiome Signature in Patients\nwith Pulmonary Hypertension and Subgroups\" submitted for publication by Dr. Zhang and\ncolleagues. Overall, the manuscript is well written and well described. The authors put in detailed\nwork in their publication. There are some issues I would like to see addressed:\n1. There is a lot of information that is not clearly connected in the manuscript. The message\nand hypothesis needs to be tightened up and made more cohesive. As it stands, the manuscript\nhas interesting conclusions and observations, but perhaps tying it to more cause would actually\nhelp the manuscript give more cohesion. For example, describing or using two different\nclassifications in the manuscript between clinical groups and WHO classifications is difficult to\nunderstand the choice of authors. I suggest that the authors choose one and then put the other\ngroups in the supplementary figure.\n2. The details surrounding antibiotic use should be better described e.g, what antibiotics were\nthe participants exposed to. And the authors should consider explaining in detail how the\nmycobiome may be affected by antibiotics which should not directly affect this form of -oMics.\n3. Finally, One important aspect is that pulmonary hypertension is a symptom of larger\ndiagnoses, for example, COPD and cor pulmonale - could the authors do subanalyses associated\nwith diagnoses instead of classes and WHO groups?\nI will be using the word document page numbers and leftmost line numbers.\nAbstract:\nPage 2, Line 61 \"Dramatically opposite patterns of the two major fungal phyla, Ascomycota and\nBasidiomycota...\" I suggest tempering down this statement, which borders on being an over-\nstatement.\nBackground:\nPage 5, Line 96 \"Although PH is mainly a cardiopulmonary disease, recent study has proposed\nthat it could be considered a systemic disease affecting multiple organs, such as the lungs, heart,\ncentral nervous system, bone marrow, and gut. \" I think there's some issues here is pulmonary\nhypertension is not a disease in it of itself, it is rather a symptom or a syndrome associated with\nmultiple other diseased. I would carefully re-write this sentences to reflect this.\nPage 7 Line 123 \"Extended recruitment procedure and methods are provided in the online\nSupplementary Material.\" In regards to the recruitment, can the authors provide a consort\ndiagram of people screened for recruitment. I would like to see the number of patients who were\nexcluded. Moreover, if all patients/participants received a RHC, could the authors also discuss\nif the RHC was to make the diagnosis or/and if it includes a table of the results from the RHC.\nSupplemental Material Page 2, Line 28 \"The PH group was further divided into the subgroups\nof non-use of antibiotics and/or corticosteroids (non-ANT/CORT) and use of ANT, CORT, or\nANT+CORT within 1 mont\" Can the authors explain why this was done? This is a good category\nto analyze, but I would consider the PH WHO categories more clinical relevant.\nSupplemental Material Page 2, Line 30 \"The patients were diagnosed with chronic obstructive\npulmonary disease, bronchiectasis, interstitial lung disease, and other relevant conditions. All\nof the patients experienced dyspnea, but their echocardiographic probability of PH was low. \"\nI'm surprised that the pH probability was low - can the authors also include the these results in\na table for review?\nSupplemental Material Page 2, Line 34 \"The healthy-control (CON) group was free of\nrespiratory diseases and other serious diseases and had no oral and respiratory infections within\n1 month. \" This wasn't clear, but did these patients/controls also receive RHC? Or was there\njust for samples?\nSupplemental Material: Can the authors include IRB protocol numbers and Patient assurances\nfor each of their sites? While the authors include multiple hospitals, they do not include the IRB\nprotocol approvals - please include these assurances.\nSupplemental Material, Page 2, Line 64 \"The extracted fungal genome DNA from positively\ndetected pharyngeal swab sample and pharyngeal swab blank control and buffer blank control\nwere verified by agarose electrophoresis, as shown in Supplementary Figure S2b.\" Were any\ncontrol samples were obtained?\nSupplemental Material, Page 7, Line 137 \"Figure S2: Quality control for pharyngeal swab\nsamples\na: Bar graphs showing the averaged concentration of PCR amplified DNA products and assigned\n\" For Figure S2, please include all points for Panels A, for Panel C please also include averages\nof control and PH - these are difficult to interpret.\nPage 8, Line 148 \"The #-diversity was performed using mothur (version v. 1.30.2, https://\nmothur.org/wiki/calculators/), with the evaluation index applied at a similarity level of 97% for\nOTUs (0.97). The #-diversity distance matrix is computed by Qiime (2020.2.0, http://qiime.org/\ninstall/index.html).\" I am wondering why the authors decided to use two different pipelines to\ncalculate alpha and beta diversity. I would suggest using one, rather than another.\nStastical anlaysis Page 7 to 9 I am wondering if this would be better moved into the\nsupplementary and the discussion about patient recruitment moved to the paper methods.\nResults:\nPage 10, Line 177 \"As shown in Figure 1a, #-diversity indicated no statistical difference in\nmycobiome richness (Sobs index and Ace index), while significantly lower diversity (indicated\nby decreased Shannon index and increased Simpson index) was observed in patients with PH\nthan in CON. For #-diversity, both non-metric multidimensional scaling (NMDS) using the\nBray-Curtis distance algorithm (Figure 1b) and Partial Least Squares Discriminant Analysis\n(PLS-DA) (Figure 1c)\" The p-values should be expressed on each comparison, even if there are\np=ns. The Beta-diversity comparisons in Panel B and Panel C are possibly significant.\nPage 10, Line 187 \"At the genus level, Candida, Cutaneotrichosporon, Apiotrichum, and\nSarocladium were the most abundant genera with the highest proportion in the PH group, while\nCandida, Fusarium, Cutaneotrichosporon, and Cladosporium in the CON group (Figure 1e). \"\nCan the authors reconcile this with candida found in COVID19 and ILD.\nPage 12, Line 222 \"As seen in Figure 2b, divergent fungal diversity was found by both #-\ndiversity (Shannon index) and #-diversity (NMDS) analyses among three subgroups, with\ngenera Holtermanniella, Trametes, Itersonilia, Psathyrella, and Geotrichum showing dramatic\nenrichment in IPAH but not in CHD-PAH or CTD-PAH.\" In Panels 2a, 2b, and 2c - needs p-\nvalues to be placed on the figure.\nPage 13, Line 259 \"...PH subgroups based on the mPAP levels. Simultaneously, Candida\nincreased dramatically in the PH-low and PH-medium groups (Figure 2d).\" As it's presented,\nthe role of candida is very interesting, have the authors considered the role of candida in other\ndiseases and how it may relate to pulmonary hypertension?\nPage 14, Line 265 \"ANT/CORT (Supplementary Figure S7b), and ANT (Supplementary Figure\nS7c), and CORT (Supplementary Figure S7d), and ANT+CORT groups (Supplementary Figure\nS7e), using Random Forest analysis.\" Can the authors also include the type of antibiotics that\nthe patients were exposed to? Could you include this as a table?\nPage 14 Line 280 \"he abundance of Ascomycota exhibited a gradual increase in non-ANT/\nCORT, ANT, CORT, and ANT+CORT subgroups, whereas the abundance of Basidiomycota\nshowed a gradual decline, and only ANT+CORT group and CON group had statistical difference\n(Figure 3d). \" Is this due to relative abundance diagnosis, can the authors pull out read counts\nto see if this pattern is still true?\nPage 15, Line 289 \". LEfSe multilevel species difference discriminant analysis also indicated\nsimilar fungal genera between the CON and non-ANT/CORT (Figure 4a), or ANT (Figure\n4b), or CORT (Figure 4c), or ANT+CORT (Figure 4d) groups. Moreover, the LEfSe multi-\nlevel species difference discriminant analysis also suggested similar and distinct genera between\nthe non-ANT/CORT and ANT (Figure 4e), or CORT (Figure 4f), or ANT+CORT (Figure 4g)\ngroups.\" Can the authors also use different orthogonal approaches such as DESEq2 or EdgeR\nto calculate other differential enrichment.\nPage 17, Line 337 \"White blood cells and cardiac index were excluded, and the correlation\nbetween the remaining 30 clinical factors and different mycobiome taxa (phyla and genera) was\nanalyzed\" Could the authors explain why this was done? It is not clear that this is actually helpful\nor why these were excluded.\nDiscussion:\nPage 20, Line 380 \"Blockage of the metabolite, trimethylamine N-oxide, which is derived from\nthe intestinal microbial flora and has been proven to be enriched in patients with PH...\" This\nseems as though it's a non-sequitur - is there anything that ties the trimethylamine N-oxide to\nthe mycobiome? I suggest this introduction is tied closer to the mycobiome.\nPage 21, Line 404 \"Notably, through multiple comparisons, we identified stable and\nconsistent mycobiome signatures in PH (and subgroups) compared to CON without being\ninfluenced by ANT, CORT, or both, characterized by specific enrichment of the fungal\ngenera Purpureocillium, Issatchenkia, and Cyberlindnera and a decrease in the fungal genera\nPeroneutypa, Simplicillium, and Metarhizium in PH (and subgroups) compared to CON.\" I am\nnot entirely sure that the authors performed multiple comparisons per se, I understand there are\nmultiple comparisons, but the authors should consider also applying multiple comparisons to\ntheir analysis. LEFSE does not put in adjustment for multiple comparisons, but would be great\nif they could do this in EdgeR.\nPage 21, Line 411 \"Moreover, a specific correlation pattern was observed between the airway\nmycobiome signature and clinical parameters of PH, strongly indicating a potential association\nbetween the altered airway mycobiome and disease progression of PH\" I would recommend that\nwhile this is the hypothesis that the authors are testing, the presentiation of this is difficult to\nunderstand. I suggest that the authors consider significant re-organization of their work - for\nexample - the redundancy of clinical groups and WHO groups does not help discern the message\nand hypothesis of the patients/particiapnts with PH. Especially, when clinicians tend to think\nabout pulmonary hypertension as a syndrome or symptom of a larger issue - thus - I suggest to\neither discuss among authors which Ph groups you would like to use, but I think the use of both\nclinical groups and WHO groups confounds the conclusions. I suggest picking one and moving\nthe other analysis to the supplementary.\nPage 21, Line 418 \"Antibiotics modulate host microbiome rapidly and remarkably. 33,34\nNg et al. reported that antibiotics significantly reshape the gut microbiota composition, and\npostantibiotic recovery depends on multiple factors, including the host diet, community context,\nand environmental reservoirs. 35 Ward et al. showed that antibiotics can alter the #-diversity and\n#-diversity of lung mycobiome. 36 Corticosteroids are immunosuppressive agents that affect\nhost microbiomes. In a meta-analysis of patients with COPD, asthma, and chronic rhinosinusitis,\nHartmann et al. demonstrated that corticosteroids significantly affected the airway microbiome\ncomposition. 37 Our study showed similar results: the #-diversity, #-diversity, and mycobiome\ncomposition were remarkably reduced in the ANT, CORT, or ANT+CORT groups versus the\nnon-ANT/CORT group. \" I think that the authors should consider explaining how antibiotics\nwhich do not affect the mycobiome but the microbiome, how the use of antibiotics may be\nassociated with mycobiome changes. There is a logical leap between the mycobiome and\nmicrobiome, the authors may consider bridging this gap during the discussion. Corticosteroids\nis much more understandable, but needs to be contextualized into the change associated with\npossible immune suppression.\nPage 23, Line 450 \"A recent study reported that P. lilacinum might induce fetal fungal\npneumonia in a green tree Python by excluding other potential pathogens.\" Not sure if this is\nactually helpful.\nMinor\nPage 18, Line 369 \"animal pathogens,s \" there is a typo here.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03982-7	Reviewed - bmc_medicine	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 1 Date: 17 Jan 2025 Reviewer's report: On review, the authors have addressed my major concerns and criticism. There are couple of small corrections to the discussion I would recommend: Page 21, Line 403 "don’t directly " please avoid conjunctions change to "do not ..." Page 22, Line 435 "Antibiotics modulate ..." I am not sure if "modulation" is the correct way we would use or describe the actions of antibiotics. I recommend a better word choice here. Page 24, Line 469 "Purpureocillium (specifically, P. lilacinum) causes several human diseases." This is an overstatement and needs to either include the human diseases or include citations. Page 25, Line 507 "Salivarius" should be not capitalized "salivarius." Page 25, Line 508 "strongly indicating a causal link between lung/airway bacteria and disease progression of PH." I think this is a slight overstatement and needs to be softened a bit. There is not necessarily a causal link as it may be inflammation from the xposure driving PH? The authors should consider slightly reducing the causal aspect of this sentence and say the association with microbiota exposure may correlate with PH progression.	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 1 Date: 17 Jan 2025 Reviewer's report: On review, the authors have addressed my major concerns and criticism. There are couple of small corrections to the discussion I would recommend: Page 21, Line 403 "don’t directly " please avoid conjunctions change to "do not ..." Page 22, Line 435 "Antibiotics modulate ..." I am not sure if "modulation" is the correct way we would use or describe the actions of antibiotics. I recommend a better word choice here. Page 24, Line 469 "Purpureocillium (specifically, P. lilacinum) causes several human diseases." This is an overstatement and needs to either include the human diseases or include citations. Page 25, Line 507 "Salivarius" should be not capitalized "salivarius." Page 25, Line 508 "strongly indicating a causal link between lung/airway bacteria and disease progression of PH." I think this is a slight overstatement and needs to be softened a bit. There is not necessarily a causal link as it may be inflammation from the xposure driving PH? The authors should consider slightly reducing the causal aspect of this sentence and say the association with microbiota exposure may correlate with PH progression.	BMC	12916_2025_3982_ReviewerReport_V1_R3.pdf	{ "criticism": 3, "example": 5, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 1, "presentation_and_reporting": 6, "results_and_discussion": 2, "suggestion_and_solution": 6, "total": 10 }	{ "criticism": 0.3, "example": 0.5, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0.1, "presentation_and_reporting": 0.6, "results_and_discussion": 0.2, "suggestion_and_solution": 0.6 }	2.3	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and\nSubgroups\nVersion: 1 Date: 17 Jan 2025\nReviewer's report:\nOn review, the authors have addressed my major concerns and criticism. There are couple of\nsmall corrections to the discussion I would recommend:\nPage 21, Line 403 \"don’t directly \" please avoid conjunctions change to \"do not ...\"\nPage 22, Line 435 \"Antibiotics modulate ...\" I am not sure if \"modulation\" is the correct way\nwe would use or describe the actions of antibiotics. I recommend a better word choice here.\nPage 24, Line 469 \"Purpureocillium (specifically, P. lilacinum) causes several human diseases.\"\nThis is an overstatement and needs to either include the human diseases or include citations.\nPage 25, Line 507 \"Salivarius\" should be not capitalized \"salivarius.\"\nPage 25, Line 508 \"strongly indicating a causal link between lung/airway bacteria and disease\nprogression of PH.\" I think this is a slight overstatement and needs to be softened a bit. There is\nnot necessarily a causal link as it may be inflammation from the xposure driving PH? The authors\nshould consider slightly reducing the causal aspect of this sentence and say the association with\nmicrobiota exposure may correlate with PH progression.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03982-7	Reviewed - bmc_medicine	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 1 Date: 29 Jan 2025 Reviewer's report: I am pleased to review the revised manuscript “Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension Subgroups.” As I have mentioned previously, investigating the mycobiome is novel in pulmonary hypertension. The authors have made some improvements to enhance the presentation of the data since the initial submission May I please get further clarification on the following: 1. During my initial review, I requested for analyses to be performed to determine whether the pharyngeal mycobiome can distinguish PH Group types (e.g. WHO Group 1 vs 3, Group 1 vs 4, Group 3 vs 4). Although the analyses were shown in the response letter, I believe that they should also be shown in the manuscript (in the supplement) for transparency to the readers. Both ROC curves and random forest analyses should be included because the potential advantage of evaluating the mycobiome is not just that there is a difference between control vs PH (which could be due to medication effects) but that the mycobiome may be distinct between these PH groups. 2. Were there any differences in diversity indices (Shannon, Bray-Curtis) when distinguishing by PVR instead of mPAP (the low, medium, and high group)? It appears from Figure 6, which looks at the correlation between phylum level (Ascomycota and Basidiomycota) and clinical factors of PH, PVR has strong correlations. Minor comments: 1. In figure 1G, no p-value was shown for Basidiomycota plot. 2. Figure legend for Figure 2 is incorrect. What is described for (B) in the legend is (C). 3. The figure 4 legend is incorrect here: “e-f: LEfSe analysis showing the most discriminant fungal genera between non-ANT/CORT and ANT (a), CORT (b) or ANT+CORT (c), respectively. Only taxa with LDA scores > 3 was presented.” 4. Table S2 should have row for PVR. 5. For Figure S4, S5, and S6, please include p values for the boxplots. Additionally, as the graphs are so similar, please change nomenclature of the graphs so it is clear whether it is with ANT/ CORT just by looking at the figure without referencing the figure legend.	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 1 Date: 29 Jan 2025 Reviewer's report: I am pleased to review the revised manuscript “Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension Subgroups.” As I have mentioned previously, investigating the mycobiome is novel in pulmonary hypertension. The authors have made some improvements to enhance the presentation of the data since the initial submission May I please get further clarification on the following: 1. During my initial review, I requested for analyses to be performed to determine whether the pharyngeal mycobiome can distinguish PH Group types (e.g. WHO Group 1 vs 3, Group 1 vs 4, Group 3 vs 4). Although the analyses were shown in the response letter, I believe that they should also be shown in the manuscript (in the supplement) for transparency to the readers. Both ROC curves and random forest analyses should be included because the potential advantage of evaluating the mycobiome is not just that there is a difference between control vs PH (which could be due to medication effects) but that the mycobiome may be distinct between these PH groups. 2. Were there any differences in diversity indices (Shannon, Bray-Curtis) when distinguishing by PVR instead of mPAP (the low, medium, and high group)? It appears from Figure 6, which looks at the correlation between phylum level (Ascomycota and Basidiomycota) and clinical factors of PH, PVR has strong correlations. Minor comments: 1. In figure 1G, no p-value was shown for Basidiomycota plot. 2. Figure legend for Figure 2 is incorrect. What is described for (B) in the legend is (C). 3. The figure 4 legend is incorrect here: “e-f: LEfSe analysis showing the most discriminant fungal genera between non-ANT/CORT and ANT (a), CORT (b) or ANT+CORT (c), respectively. Only taxa with LDA scores > 3 was presented.” 4. Table S2 should have row for PVR. 5. For Figure S4, S5, and S6, please include p values for the boxplots. Additionally, as the graphs are so similar, please change nomenclature of the graphs so it is clear whether it is with ANT/ CORT just by looking at the figure without referencing the figure legend.	BMC	12916_2025_3982_ReviewerReport_V1_R1.pdf	{ "criticism": 2, "example": 4, "importance_and_relevance": 0, "materials_and_methods": 8, "praise": 2, "presentation_and_reporting": 9, "results_and_discussion": 1, "suggestion_and_solution": 7, "total": 21 }	{ "criticism": 0.09523809523809523, "example": 0.19047619047619047, "importance_and_relevance": 0, "materials_and_methods": 0.38095238095238093, "praise": 0.09523809523809523, "presentation_and_reporting": 0.42857142857142855, "results_and_discussion": 0.047619047619047616, "suggestion_and_solution": 0.3333333333333333 }	1.571429	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and\nSubgroups\nVersion: 1 Date: 29 Jan 2025\nReviewer's report:\nI am pleased to review the revised manuscript “Distinct Airway Mycobiome Signature\nin Patients with Pulmonary Hypertension Subgroups.” As I have mentioned previously,\ninvestigating the mycobiome is novel in pulmonary hypertension. The authors have made some\nimprovements to enhance the presentation of the data since the initial submission\nMay I please get further clarification on the following:\n1. During my initial review, I requested for analyses to be performed to determine whether the\npharyngeal mycobiome can distinguish PH Group types (e.g. WHO Group 1 vs 3, Group 1 vs\n4, Group 3 vs 4). Although the analyses were shown in the response letter, I believe that they\nshould also be shown in the manuscript (in the supplement) for transparency to the readers. Both\nROC curves and random forest analyses should be included because the potential advantage of\nevaluating the mycobiome is not just that there is a difference between control vs PH (which\ncould be due to medication effects) but that the mycobiome may be distinct between these PH\ngroups.\n2. Were there any differences in diversity indices (Shannon, Bray-Curtis) when distinguishing\nby PVR instead of mPAP (the low, medium, and high group)? It appears from Figure 6, which\nlooks at the correlation between phylum level (Ascomycota and Basidiomycota) and clinical\nfactors of PH, PVR has strong correlations.\nMinor comments:\n1. In figure 1G, no p-value was shown for Basidiomycota plot.\n2. Figure legend for Figure 2 is incorrect. What is described for (B) in the legend is (C).\n3. The figure 4 legend is incorrect here: “e-f: LEfSe analysis showing the most discriminant\nfungal genera between non-ANT/CORT and ANT (a), CORT (b) or ANT+CORT (c),\nrespectively. Only taxa with LDA scores > 3 was presented.”\n4. Table S2 should have row for PVR.\n5. For Figure S4, S5, and S6, please include p values for the boxplots. Additionally, as the graphs\nare so similar, please change nomenclature of the graphs so it is clear whether it is with ANT/\nCORT just by looking at the figure without referencing the figure legend.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03982-7	Reviewed - bmc_medicine	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 2 Date: 07 Feb 2025 Reviewer's report: On re-review, the authors have addressed my major concerns and criticism. The issues with association/causation has been addressed . I appreciate the authors attention to detail and work.	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 2 Date: 07 Feb 2025 Reviewer's report: On re-review, the authors have addressed my major concerns and criticism. The issues with association/causation has been addressed . I appreciate the authors attention to detail and work.	BMC	12916_2025_3982_ReviewerReport_V2_R3.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 2, "presentation_and_reporting": 1, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 3 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0.6666666666666666, "presentation_and_reporting": 0.3333333333333333, "results_and_discussion": 0, "suggestion_and_solution": 0 }	1	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and\nSubgroups\nVersion: 2 Date: 07 Feb 2025\nReviewer's report:\nOn re-review, the authors have addressed my major concerns and criticism. The issues with\nassociation/causation has been addressed . I appreciate the authors attention to detail and work.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03982-7	Reviewed - bmc_medicine	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 2 Date: 10 Feb 2025 Reviewer's report: The authors have addressed my comments.	Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and Subgroups Version: 2 Date: 10 Feb 2025 Reviewer's report: The authors have addressed my comments.	BMC	12916_2025_3982_ReviewerReport_V2_R1.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 1 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0 }	0	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Distinct Airway Mycobiome Signature in Patients with Pulmonary Hypertension and\nSubgroups\nVersion: 2 Date: 10 Feb 2025\nReviewer's report:\nThe authors have addressed my comments.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03976-5	Reviewed - bmc_medicine	Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles among postmenopausal women in China Version: 0 Date: 08 Sep 2024 Reviewer's report: In their manuscript, Sun and colleagues estimate the fraction of stroke (overall, ischemic, hemorrhagic, and subarachnoid hemorrhagic attributable to six reproductive factors (age at menarche, age at menopause, high parity, history of stillbirth, history of miscarriage, history of termination, history of lactation). Data on reproductive factors and covariates were collected from postmenopausal women in China via baseline survey and stroke was ascertained from hospital, disease registry, and insurance claims. Additionally, they conducted analyses stratified by socioeconomic status. They report generally modest population attributable fractions for reproductive factors in the overall population (the highest proportion being 8.1% of all strokes attributed to high parity, ~14% attributable to all factors combined). However, PAFs for investigated factors tended to be much higher among low-SES women compared to high-SES women. These results are generally in line with prior studies that have found higher risk of cardiovascular and cerebrovascular disease among women with early age at menarche, early age at menopause, high parity, and history of adverse pregnancy outcomes. The study benefits from a large sample, and the calculation of PAFs is an interesting one. It is typically interpreted as the fraction of cases of the outcome in a population that could be eliminated if a risk factor could be eliminated. For the risk factors in question in this study, it is unclear the extent to which they may be causal versus markers for other causal factors for stroke, such as cardiometabolic disease. Also, some of these factors are not readily modifiable, and, for those that are, such decisions may not be likely to be made with future stroke risk in mind. However, I feel that it can still be used to identify women at elevated risk of stroke based on reproductive history, as the authors discuss. The results could also, e.g., be used to help highlight risks and challenges faced by low-SES women in their childbearing years. Specific comments are listed below: • Lines 65-67: To what extent do the authors believe these factors are markers for other risk factors versus causal factors in stroke? • Lines 117-19: Dichotomizing early age at menarche at #15 versus >15 seems to be a little high to me. In US studies, early menarche may be defined as <10 or <12 years. Might the association be diluted by using this cutoff? Further, late age at menarche may also be associated with chronic disease. Was this explored at all? Factors such as history of stillbirth, miscarriage, termination, or lactation would more appropriately be restricted to parous women to avoid mixing of the potential impact of infertility and its associated causes and the reproductive risk factor in question. Was this done? • Lines 185-189: Were the population prevalence estimates derived from the survey data or other sources? If from the data, was the uncertainty in these estimates included in the calculation of the confidence intervals for the PAFs? Admittedly, this may not add much given the sample size. • Line 193: it is unclear whether the estimates of the combined PAFs assumed that each of the reproductive factors had an independent association with stroke • Overall Results: differences in PAFs among low and high SES women was striking. However, some estimate of the baseline stroke risk amongst women in these two groups would really help to further emphasize this contrast • Overall Discussion: I felt that a bit more discussion of whether or not the factors investigated were causal or markers for other risk factors, and the potential impact on the interpretation of results would strengthen this section. In addition, how might potential changes in employment over time impact the generalizability of the results?	Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles among postmenopausal women in China Version: 0 Date: 08 Sep 2024 Reviewer's report: In their manuscript, Sun and colleagues estimate the fraction of stroke (overall, ischemic, hemorrhagic, and subarachnoid hemorrhagic attributable to six reproductive factors (age at menarche, age at menopause, high parity, history of stillbirth, history of miscarriage, history of termination, history of lactation). Data on reproductive factors and covariates were collected from postmenopausal women in China via baseline survey and stroke was ascertained from hospital, disease registry, and insurance claims. Additionally, they conducted analyses stratified by socioeconomic status. They report generally modest population attributable fractions for reproductive factors in the overall population (the highest proportion being 8.1% of all strokes attributed to high parity, ~14% attributable to all factors combined). However, PAFs for investigated factors tended to be much higher among low-SES women compared to high-SES women. These results are generally in line with prior studies that have found higher risk of cardiovascular and cerebrovascular disease among women with early age at menarche, early age at menopause, high parity, and history of adverse pregnancy outcomes. The study benefits from a large sample, and the calculation of PAFs is an interesting one. It is typically interpreted as the fraction of cases of the outcome in a population that could be eliminated if a risk factor could be eliminated. For the risk factors in question in this study, it is unclear the extent to which they may be causal versus markers for other causal factors for stroke, such as cardiometabolic disease. Also, some of these factors are not readily modifiable, and, for those that are, such decisions may not be likely to be made with future stroke risk in mind. However, I feel that it can still be used to identify women at elevated risk of stroke based on reproductive history, as the authors discuss. The results could also, e.g., be used to help highlight risks and challenges faced by low-SES women in their childbearing years. Specific comments are listed below: • Lines 65-67: To what extent do the authors believe these factors are markers for other risk factors versus causal factors in stroke? • Lines 117-19: Dichotomizing early age at menarche at #15 versus >15 seems to be a little high to me. In US studies, early menarche may be defined as <10 or <12 years. Might the association be diluted by using this cutoff? Further, late age at menarche may also be associated with chronic disease. Was this explored at all? Factors such as history of stillbirth, miscarriage, termination, or lactation would more appropriately be restricted to parous women to avoid mixing of the potential impact of infertility and its associated causes and the reproductive risk factor in question. Was this done? • Lines 185-189: Were the population prevalence estimates derived from the survey data or other sources? If from the data, was the uncertainty in these estimates included in the calculation of the confidence intervals for the PAFs? Admittedly, this may not add much given the sample size. • Line 193: it is unclear whether the estimates of the combined PAFs assumed that each of the reproductive factors had an independent association with stroke • Overall Results: differences in PAFs among low and high SES women was striking. However, some estimate of the baseline stroke risk amongst women in these two groups would really help to further emphasize this contrast • Overall Discussion: I felt that a bit more discussion of whether or not the factors investigated were causal or markers for other risk factors, and the potential impact on the interpretation of results would strengthen this section. In addition, how might potential changes in employment over time impact the generalizability of the results?	BMC	12916_2025_3976_ReviewerReport_V0_R2.pdf	{ "criticism": 3, "example": 4, "importance_and_relevance": 1, "materials_and_methods": 14, "praise": 1, "presentation_and_reporting": 0, "results_and_discussion": 8, "suggestion_and_solution": 4, "total": 26 }	{ "criticism": 0.11538461538461539, "example": 0.15384615384615385, "importance_and_relevance": 0.038461538461538464, "materials_and_methods": 0.5384615384615384, "praise": 0.038461538461538464, "presentation_and_reporting": 0, "results_and_discussion": 0.3076923076923077, "suggestion_and_solution": 0.15384615384615385 }	1.346154	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles\namong postmenopausal women in China\nVersion: 0 Date: 08 Sep 2024\nReviewer's report:\nIn their manuscript, Sun and colleagues estimate the fraction of stroke (overall, ischemic,\nhemorrhagic, and subarachnoid hemorrhagic attributable to six reproductive factors (age at\nmenarche, age at menopause, high parity, history of stillbirth, history of miscarriage, history\nof termination, history of lactation). Data on reproductive factors and covariates were collected\nfrom postmenopausal women in China via baseline survey and stroke was ascertained from\nhospital, disease registry, and insurance claims. Additionally, they conducted analyses stratified\nby socioeconomic status. They report generally modest population attributable fractions for\nreproductive factors in the overall population (the highest proportion being 8.1% of all strokes\nattributed to high parity, ~14% attributable to all factors combined). However, PAFs for\ninvestigated factors tended to be much higher among low-SES women compared to high-SES\nwomen.\nThese results are generally in line with prior studies that have found higher risk of cardiovascular\nand cerebrovascular disease among women with early age at menarche, early age at menopause,\nhigh parity, and history of adverse pregnancy outcomes. The study benefits from a large sample,\nand the calculation of PAFs is an interesting one. It is typically interpreted as the fraction of\ncases of the outcome in a population that could be eliminated if a risk factor could be eliminated.\nFor the risk factors in question in this study, it is unclear the extent to which they may be causal\nversus markers for other causal factors for stroke, such as cardiometabolic disease. Also, some\nof these factors are not readily modifiable, and, for those that are, such decisions may not be\nlikely to be made with future stroke risk in mind. However, I feel that it can still be used to\nidentify women at elevated risk of stroke based on reproductive history, as the authors discuss.\nThe results could also, e.g., be used to help highlight risks and challenges faced by low-SES\nwomen in their childbearing years.\nSpecific comments are listed below:\n• Lines 65-67: To what extent do the authors believe these factors are markers for other risk\nfactors versus causal factors in stroke?\n• Lines 117-19: Dichotomizing early age at menarche at #15 versus >15 seems to be a little\nhigh to me. In US studies, early menarche may be defined as <10 or <12 years. Might the\nassociation be diluted by using this cutoff? Further, late age at menarche may also be associated\nwith chronic disease. Was this explored at all? Factors such as history of stillbirth, miscarriage,\ntermination, or lactation would more appropriately be restricted to parous women to avoid\nmixing of the potential impact of infertility and its associated causes and the reproductive risk\nfactor in question. Was this done?\n• Lines 185-189: Were the population prevalence estimates derived from the survey data or other\nsources? If from the data, was the uncertainty in these estimates included in the calculation of the\nconfidence intervals for the PAFs? Admittedly, this may not add much given the sample size.\n• Line 193: it is unclear whether the estimates of the combined PAFs assumed that each of the\nreproductive factors had an independent association with stroke\n• Overall Results: differences in PAFs among low and high SES women was striking. However,\nsome estimate of the baseline stroke risk amongst women in these two groups would really help\nto further emphasize this contrast\n• Overall Discussion: I felt that a bit more discussion of whether or not the factors investigated\nwere causal or markers for other risk factors, and the potential impact on the interpretation of\nresults would strengthen this section. In addition, how might potential changes in employment\nover time impact the generalizability of the results?", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03976-5	Reviewed - bmc_medicine	Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles among postmenopausal women in China Version: 0 Date: 16 Sep 2024 Reviewer's report: Thank you for reviewing this manuscript! The authors examined population-attributable fractions of stroke and its subtypes associated with reproductive factors among Chinese postmenopausal women. There are the following comments: 1. Figure 1: Regarding gynecological operations, it’s suggested that a specific definition be given. It is inappropriate to exclude participants who underwent surgical menopause or surgical delivery. 2. Figure 1: Excluding missing data on covariates, it is better to use the multiple imputations and then do the sensitivity analysis. 3. Figure 1: Follow-up: It’s suggested that more details be added, such as the initial time and number of visits. 4. Table 1: Why is BMI described as both a continuous and categorical variable? It is fine to select continuous BMI. The same comments are for the age of menarche and the age of menopause. 5. Age of menarche: <15 is defined as early menarche. Please provide the references. I didn’t agree with it at all. The early menarche generally refers to #10 years, and the late menarche refers to #15 years. 6. High parity: #3# Please provide the references for the definition. 7. Why not include age at childbirth? It’s a critical factor. 8. It should be multilinear for BMI and Waist circumference. Please choose one of the two variables. 9. For economic status, if the raw data is continuous. It’s unnecessary to categorize them intentionally. It would be better to include continuous income. 10. Table 1 is too lengthy. The rows of “0” and “No” should be removed. 11. Table 3: The association is insignificant with 0.1(-1.6-1.9). However, the prevalence (%) is 37.2%. It doesn’t make sense.	Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles among postmenopausal women in China Version: 0 Date: 16 Sep 2024 Reviewer's report: Thank you for reviewing this manuscript! The authors examined population-attributable fractions of stroke and its subtypes associated with reproductive factors among Chinese postmenopausal women. There are the following comments: 1. Figure 1: Regarding gynecological operations, it’s suggested that a specific definition be given. It is inappropriate to exclude participants who underwent surgical menopause or surgical delivery. 2. Figure 1: Excluding missing data on covariates, it is better to use the multiple imputations and then do the sensitivity analysis. 3. Figure 1: Follow-up: It’s suggested that more details be added, such as the initial time and number of visits. 4. Table 1: Why is BMI described as both a continuous and categorical variable? It is fine to select continuous BMI. The same comments are for the age of menarche and the age of menopause. 5. Age of menarche: <15 is defined as early menarche. Please provide the references. I didn’t agree with it at all. The early menarche generally refers to #10 years, and the late menarche refers to #15 years. 6. High parity: #3# Please provide the references for the definition. 7. Why not include age at childbirth? It’s a critical factor. 8. It should be multilinear for BMI and Waist circumference. Please choose one of the two variables. 9. For economic status, if the raw data is continuous. It’s unnecessary to categorize them intentionally. It would be better to include continuous income. 10. Table 1 is too lengthy. The rows of “0” and “No” should be removed. 11. Table 3: The association is insignificant with 0.1(-1.6-1.9). However, the prevalence (%) is 37.2%. It doesn’t make sense.	BMC	12916_2025_3976_ReviewerReport_V0_R1.pdf	{ "criticism": 5, "example": 4, "importance_and_relevance": 0, "materials_and_methods": 7, "praise": 0, "presentation_and_reporting": 10, "results_and_discussion": 0, "suggestion_and_solution": 12, "total": 37 }	{ "criticism": 0.13513513513513514, "example": 0.10810810810810811, "importance_and_relevance": 0, "materials_and_methods": 0.1891891891891892, "praise": 0, "presentation_and_reporting": 0.2702702702702703, "results_and_discussion": 0, "suggestion_and_solution": 0.32432432432432434 }	1.027027	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles\namong postmenopausal women in China\nVersion: 0 Date: 16 Sep 2024\nReviewer's report:\nThank you for reviewing this manuscript! The authors examined population-attributable\nfractions of stroke and its subtypes associated with reproductive factors among Chinese\npostmenopausal women.\nThere are the following comments:\n1. Figure 1: Regarding gynecological operations, it’s suggested that a specific definition be\ngiven. It is inappropriate to exclude participants who underwent surgical menopause or surgical\ndelivery.\n2. Figure 1: Excluding missing data on covariates, it is better to use the multiple imputations\nand then do the sensitivity analysis.\n3. Figure 1: Follow-up: It’s suggested that more details be added, such as the initial time and\nnumber of visits.\n4. Table 1: Why is BMI described as both a continuous and categorical variable? It is fine\nto select continuous BMI. The same comments are for the age of menarche and the age of\nmenopause.\n5. Age of menarche: <15 is defined as early menarche. Please provide the references. I didn’t\nagree with it at all. The early menarche generally refers to #10 years, and the late menarche\nrefers to #15 years.\n6. High parity: #3# Please provide the references for the definition.\n7. Why not include age at childbirth? It’s a critical factor.\n8. It should be multilinear for BMI and Waist circumference. Please choose one of the two\nvariables.\n9. For economic status, if the raw data is continuous. It’s unnecessary to categorize them\nintentionally. It would be better to include continuous income.\n10. Table 1 is too lengthy. The rows of “0” and “No” should be removed.\n11. Table 3: The association is insignificant with 0.1(-1.6-1.9). However, the prevalence (%) is\n37.2%. It doesn’t make sense.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03976-5	Reviewed - bmc_medicine	Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles among postmenopausal women in China Version: 1 Date: 18 Dec 2024 Reviewer's report: I appreciate the efforts of Sun et al. to address my comments on their manuscript calculating the of risk of stroke (overall and by subtype) attributable to reproductive factors among Chinese women of differing socioeconomic status. Those efforts have added much to what was already a very interesting manuscript. I have only a couple of additional comments 1. The numbers of the individual stroke subtypes in Figure 1 and the text are slightly higher than the total number of strokes. Did some cases fall under more than one subtype? If so, it would be helpful to note this in the footnote and in the text. Or is there some other explanation? Minor comments 2. Line 112: I suggest replacing “abnormal age” with “implausible age”. “Abnormal age” may be interpreted as “unusually early or late age at menopause”. 2. Line 372: I recommend replacing “while” with “however”. I believe this better conveys the authors’ intent.	Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles among postmenopausal women in China Version: 1 Date: 18 Dec 2024 Reviewer's report: I appreciate the efforts of Sun et al. to address my comments on their manuscript calculating the of risk of stroke (overall and by subtype) attributable to reproductive factors among Chinese women of differing socioeconomic status. Those efforts have added much to what was already a very interesting manuscript. I have only a couple of additional comments 1. The numbers of the individual stroke subtypes in Figure 1 and the text are slightly higher than the total number of strokes. Did some cases fall under more than one subtype? If so, it would be helpful to note this in the footnote and in the text. Or is there some other explanation? Minor comments 2. Line 112: I suggest replacing “abnormal age” with “implausible age”. “Abnormal age” may be interpreted as “unusually early or late age at menopause”. 2. Line 372: I recommend replacing “while” with “however”. I believe this better conveys the authors’ intent.	BMC	12916_2025_3976_ReviewerReport_V1_R2.pdf	{ "criticism": 0, "example": 3, "importance_and_relevance": 1, "materials_and_methods": 2, "praise": 2, "presentation_and_reporting": 3, "results_and_discussion": 1, "suggestion_and_solution": 4, "total": 14 }	{ "criticism": 0, "example": 0.21428571428571427, "importance_and_relevance": 0.07142857142857142, "materials_and_methods": 0.14285714285714285, "praise": 0.14285714285714285, "presentation_and_reporting": 0.21428571428571427, "results_and_discussion": 0.07142857142857142, "suggestion_and_solution": 0.2857142857142857 }	1.142857	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Socioeconomic variations in the proportions of stroke attributable to reproductive profiles\namong postmenopausal women in China\nVersion: 1 Date: 18 Dec 2024\nReviewer's report:\nI appreciate the efforts of Sun et al. to address my comments on their manuscript calculating\nthe of risk of stroke (overall and by subtype) attributable to reproductive factors among Chinese\nwomen of differing socioeconomic status. Those efforts have added much to what was already\na very interesting manuscript. I have only a couple of additional comments\n1. The numbers of the individual stroke subtypes in Figure 1 and the text are slightly higher than\nthe total number of strokes. Did some cases fall under more than one subtype? If so, it would be\nhelpful to note this in the footnote and in the text. Or is there some other explanation?\nMinor comments\n2. Line 112: I suggest replacing “abnormal age” with “implausible age”. “Abnormal age” may\nbe interpreted as “unusually early or late age at menopause”.\n2. Line 372: I recommend replacing “while” with “however”. I believe this better conveys the\nauthors’ intent.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03973-8	Reviewed - bmc_medicine	Title: Research prioritisation in preparedness for and response to outbreaks of high-consequence pathogens: a scoping review Version: 0 Date: 09 Jan 2025 Reviewer: Amesh A. Adalja Reviewer's report: I think this paper is interesting and straight forward providing good information on how this process occurs. One thing that I might suggest considering (optionally) is to add a line or two about how the pathogens are selected. In my experience, many extraneous considerations factor into decisions for which pathogens are considered “high consequence” and many are not truly pandemic prone	Title: Research prioritisation in preparedness for and response to outbreaks of high-consequence pathogens: a scoping review Version: 0 Date: 09 Jan 2025 Reviewer: Amesh A. Adalja Reviewer's report: I think this paper is interesting and straight forward providing good information on how this process occurs. One thing that I might suggest considering (optionally) is to add a line or two about how the pathogens are selected. In my experience, many extraneous considerations factor into decisions for which pathogens are considered “high consequence” and many are not truly pandemic prone	BMC	12916_2025_3973_ReviewerReport_V0_R2.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 1, "materials_and_methods": 1, "praise": 1, "presentation_and_reporting": 1, "results_and_discussion": 0, "suggestion_and_solution": 1, "total": 3 }	{ "criticism": 0.3333333333333333, "example": 0, "importance_and_relevance": 0.3333333333333333, "materials_and_methods": 0.3333333333333333, "praise": 0.3333333333333333, "presentation_and_reporting": 0.3333333333333333, "results_and_discussion": 0, "suggestion_and_solution": 0.3333333333333333 }	2	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Research prioritisation in preparedness for and response to outbreaks of high-consequence\npathogens: a scoping review\nVersion: 0 Date: 09 Jan 2025\nReviewer: Amesh A. Adalja\nReviewer's report:\nI think this paper is interesting and straight forward providing good information on how this\nprocess occurs. One thing that I might suggest considering (optionally) is to add a line or two\nabout how the pathogens are selected. In my experience, many extraneous considerations factor\ninto decisions for which pathogens are considered “high consequence” and many are not truly\npandemic prone", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03973-8	Reviewed - bmc_medicine	Title: Research prioritisation in preparedness for and response to outbreaks of high-consequence pathogens: a scoping review Version: 0 Date: 20 Jan 2025 Reviewer's report: Studies such as this are important and add pieces to the puzzle to ultimately help us prepare for the next inevitable public health threat. The paper is generally well-written, flows well, and is easy to read while being engaging. The article does a great job of sharing the most important results and aligned conclusions. The paper would be a valuable addition to BMC Medicine. I recommend accepting this manuscript.	Title: Research prioritisation in preparedness for and response to outbreaks of high-consequence pathogens: a scoping review Version: 0 Date: 20 Jan 2025 Reviewer's report: Studies such as this are important and add pieces to the puzzle to ultimately help us prepare for the next inevitable public health threat. The paper is generally well-written, flows well, and is easy to read while being engaging. The article does a great job of sharing the most important results and aligned conclusions. The paper would be a valuable addition to BMC Medicine. I recommend accepting this manuscript.	BMC	12916_2025_3973_ReviewerReport_V0_R1.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 3, "materials_and_methods": 0, "praise": 5, "presentation_and_reporting": 1, "results_and_discussion": 1, "suggestion_and_solution": 2, "total": 5 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0.6, "materials_and_methods": 0, "praise": 1, "presentation_and_reporting": 0.2, "results_and_discussion": 0.2, "suggestion_and_solution": 0.4 }	2.4	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Research prioritisation in preparedness for and response to outbreaks of high-consequence\npathogens: a scoping review\nVersion: 0 Date: 20 Jan 2025\nReviewer's report:\nStudies such as this are important and add pieces to the puzzle to ultimately help us prepare for\nthe next inevitable public health threat. The paper is generally well-written, flows well, and is\neasy to read while being engaging. The article does a great job of sharing the most important\nresults and aligned conclusions. The paper would be a valuable addition to BMC Medicine. I\nrecommend accepting this manuscript.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 19 Nov 2024 Reviewer's report: Duan et al examined the association between smoking-related gut microbiota alteration (represented by smoking-related microbiota index) and obesity, obesity-related disorders, and obesity-related outcomes. The manuscript is well-written, and data analysis seems ok. The major concern is the concept of “smoking-related microbiota index”. 1. What’s the necessity of using UKB, NHANES, CHNS, and CHARLS? It only shows the associations between smoking status (current vs never) and BMI, which has been admitted in your introduction. Could you explain the adding value of Figure 1 for this study? 2. In WELL-China cohort, 50 microbial genera displayed an association with current smoking while in the Lanxi cohort, only 19 microbial genera, and only 9 overlapped. The two populations were of similar age and geographical area. What caused the difference between these two cohorts? If the association between smoking and gut microbiota varied across populations, how to justify the validity and reliability of this smoking-related microbiota index? 3. The authors should use one cohort as a replicate cohort for the index. 4. Is it possible to quantify smoking exposure? According to authors, the index was similar between former smokers and current smokers. Does it suggest cumulative effects of smoking?	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 19 Nov 2024 Reviewer's report: Duan et al examined the association between smoking-related gut microbiota alteration (represented by smoking-related microbiota index) and obesity, obesity-related disorders, and obesity-related outcomes. The manuscript is well-written, and data analysis seems ok. The major concern is the concept of “smoking-related microbiota index”. 1. What’s the necessity of using UKB, NHANES, CHNS, and CHARLS? It only shows the associations between smoking status (current vs never) and BMI, which has been admitted in your introduction. Could you explain the adding value of Figure 1 for this study? 2. In WELL-China cohort, 50 microbial genera displayed an association with current smoking while in the Lanxi cohort, only 19 microbial genera, and only 9 overlapped. The two populations were of similar age and geographical area. What caused the difference between these two cohorts? If the association between smoking and gut microbiota varied across populations, how to justify the validity and reliability of this smoking-related microbiota index? 3. The authors should use one cohort as a replicate cohort for the index. 4. Is it possible to quantify smoking exposure? According to authors, the index was similar between former smokers and current smokers. Does it suggest cumulative effects of smoking?	BMC	12916_2025_3969_ReviewerReport_V0_R6.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 1, "materials_and_methods": 9, "praise": 1, "presentation_and_reporting": 2, "results_and_discussion": 0, "suggestion_and_solution": 2, "total": 18 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0.05555555555555555, "materials_and_methods": 0.5, "praise": 0.05555555555555555, "presentation_and_reporting": 0.1111111111111111, "results_and_discussion": 0, "suggestion_and_solution": 0.1111111111111111 }	0.833333	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 0 Date: 19 Nov 2024\nReviewer's report:\nDuan et al examined the association between smoking-related gut microbiota alteration\n(represented by smoking-related microbiota index) and obesity, obesity-related disorders, and\nobesity-related outcomes. The manuscript is well-written, and data analysis seems ok. The major\nconcern is the concept of “smoking-related microbiota index”.\n1. What’s the necessity of using UKB, NHANES, CHNS, and CHARLS? It only shows the\nassociations between smoking status (current vs never) and BMI, which has been admitted in\nyour introduction. Could you explain the adding value of Figure 1 for this study?\n2. In WELL-China cohort, 50 microbial genera displayed an association with current smoking\nwhile in the Lanxi cohort, only 19 microbial genera, and only 9 overlapped. The two populations\nwere of similar age and geographical area. What caused the difference between these two\ncohorts? If the association between smoking and gut microbiota varied across populations, how\nto justify the validity and reliability of this smoking-related microbiota index?\n3. The authors should use one cohort as a replicate cohort for the index.\n4. Is it possible to quantify smoking exposure? According to authors, the index was similar\nbetween former smokers and current smokers. Does it suggest cumulative effects of smoking?", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 24 Nov 2024 Reviewer's report: The study design is unclear. The authors initially describe the work as a systematic review. However, they perform association analyses by mixing cohort data with national survey data. Additionally, they include a meta-analysis section. If this is intended as a systematic review (SR) with a meta-analysis (MA), the protocol must be registered. The authors do not provide any details on the protocol registration. Systematic reviews and meta-analyses should be reported in accordance with the most recent update of the PRISMA statement (2020). The authors fail to include key sections from the PRISMA checklist, including, most notably, risk of bias and publication bias. The authors do not provide forest plots typically associated with meta-analyses. The authors do not present funnel plots that are standard for meta-analyses. Finally, the summary measure of association specific to the meta-analysis is not presented.	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 24 Nov 2024 Reviewer's report: The study design is unclear. The authors initially describe the work as a systematic review. However, they perform association analyses by mixing cohort data with national survey data. Additionally, they include a meta-analysis section. If this is intended as a systematic review (SR) with a meta-analysis (MA), the protocol must be registered. The authors do not provide any details on the protocol registration. Systematic reviews and meta-analyses should be reported in accordance with the most recent update of the PRISMA statement (2020). The authors fail to include key sections from the PRISMA checklist, including, most notably, risk of bias and publication bias. The authors do not provide forest plots typically associated with meta-analyses. The authors do not present funnel plots that are standard for meta-analyses. Finally, the summary measure of association specific to the meta-analysis is not presented.	BMC	12916_2025_3969_ReviewerReport_V0_R4.pdf	{ "criticism": 6, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 8, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 2, "total": 11 }	{ "criticism": 0.5454545454545454, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0.7272727272727273, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0.18181818181818182 }	1.454545	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 0 Date: 24 Nov 2024\nReviewer's report:\nThe study design is unclear. The authors initially describe the work as a systematic review.\nHowever, they perform association analyses by mixing cohort data with national survey data.\nAdditionally, they include a meta-analysis section.\nIf this is intended as a systematic review (SR) with a meta-analysis (MA), the protocol must be\nregistered. The authors do not provide any details on the protocol registration.\nSystematic reviews and meta-analyses should be reported in accordance with the most recent\nupdate of the PRISMA statement (2020).\nThe authors fail to include key sections from the PRISMA checklist, including, most notably,\nrisk of bias and publication bias.\nThe authors do not provide forest plots typically associated with meta-analyses.\nThe authors do not present funnel plots that are standard for meta-analyses.\nFinally, the summary measure of association specific to the meta-analysis is not presented.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 29 Nov 2024 Reviewer's report: Shankuan Zhu's team utilised population databases to deeply investigate the relationship between smoking and body weight, revealing a downward trend in BMI among smokers. In subsequent research, through microbiome data analyses of the WELL-China cohort and the Lanxi cohort, it was found that the Smoking-Associated Microbiome Index (SMI) was significantly positively correlated with BMI and other obesity indicators. Further analyses indicated that SMI was also closely associated with obesity-related diseases. Moreover, the study effectively addressed potential familial confounding factors through sibling cohort analyses. Based on these findings, the authors attempted to provide a plausible explanation for the smoking-obesity paradox. This study's topic is novel and its content highly compelling, offering a fresh perspective on the complex relationship between smoking and obesity while carrying significant public health implications and practical value. However, I still have some questions that require clarification from the authors. (1)Given that the authors have chosen both population-based studies and microbiome research to address this issue, it would be beneficial to reflect this dual research perspective clearly from the title. However, based on the current study content, the focus seems primarily on microbiome- related analyses. To more comprehensively explore the relationship between smoking and obesity, the inclusion of additional indicators beyond BMI could be considered. For instance, datasets like the UK Biobank or NHANES could provide a broader perspective and enrich the study with supplementary data. Most importantly, it is crucial to incorporate a more objective measure of smoking exposure, such as nicotine levels. Compared to self-reported questionnaires, nicotine exposure provides a more accurate reflection of smoking status, significantly enhancing the scientific robustness and credibility of the study's conclusions. I recommend structuring the study into two main sections: one focusing on population databases to systematically explore the association between smoking and obesity, and another delving into microbiome research to uncover potential biological mechanisms. If it is not feasible to address both aspects comprehensively, I suggest omitting the population database analysis and concentrating on the microbiome study to ensure a clearer and more focused research narrative. (2)The conclusions drawn by the author from the population database are not necessarily specific to the male population. Currently, the # values from the CHNS and CHARLS databases, which are also based on Chinese populations, are relatively large, whereas the # values from the WELL- China cohort and the Lanxi cohort are relatively smaller. How should the author explain this discrepancy? Are there feasible methods to address or adjust this issue? (3)Former smokers should not be overlooked. Existing research evidence indicates that individuals who quit smoking may experience phenomena such as weight gain; however, quitting smoking can also significantly reduce the risk of adverse health outcomes caused by smoking. This is a highly valuable and very interesting topic for research. (4)It is recommended to provide references for certain definitions described in the methodology section, such as the definition of obesity-related cancers and the reasons for choosing a fixed- effects model. Additionally, the author is advised to examine and explain relevant factors one by one to enhance the rigor and scientific validity of the study. (5)The greatest limitation of this study lies in the excessive complexity of the research content, while the description of each part is not detailed or rigorous enough, with some content missing. For example, in the cohort analysis of disease occurrence, the baseline characteristics of the study population (such as comorbidities) were not sufficiently elaborated. These issues are, in my opinion, the key areas that require revision. It is recommended that the author thoroughly refine the structure of the article to ensure clear logic and provide more detailed and scientifically robust descriptions of each section. (6)It is recommended to comprehensively polish the language of the article to ensure clear and fluent expression, as well as to standardize the use of capitalization. Additionally, the supplementary materials should be carefully reviewed and their order adjusted as necessary. (7)Considering the highly complex factors influencing gut microbiota and obesity, it is recommended to provide broader explanations from multiple perspectives regarding the contradictions between population studies and microbiome research. Furthermore, it is suggested to enhance the literature review in the introduction and discussion sections.	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 29 Nov 2024 Reviewer's report: Shankuan Zhu's team utilised population databases to deeply investigate the relationship between smoking and body weight, revealing a downward trend in BMI among smokers. In subsequent research, through microbiome data analyses of the WELL-China cohort and the Lanxi cohort, it was found that the Smoking-Associated Microbiome Index (SMI) was significantly positively correlated with BMI and other obesity indicators. Further analyses indicated that SMI was also closely associated with obesity-related diseases. Moreover, the study effectively addressed potential familial confounding factors through sibling cohort analyses. Based on these findings, the authors attempted to provide a plausible explanation for the smoking-obesity paradox. This study's topic is novel and its content highly compelling, offering a fresh perspective on the complex relationship between smoking and obesity while carrying significant public health implications and practical value. However, I still have some questions that require clarification from the authors. (1)Given that the authors have chosen both population-based studies and microbiome research to address this issue, it would be beneficial to reflect this dual research perspective clearly from the title. However, based on the current study content, the focus seems primarily on microbiome- related analyses. To more comprehensively explore the relationship between smoking and obesity, the inclusion of additional indicators beyond BMI could be considered. For instance, datasets like the UK Biobank or NHANES could provide a broader perspective and enrich the study with supplementary data. Most importantly, it is crucial to incorporate a more objective measure of smoking exposure, such as nicotine levels. Compared to self-reported questionnaires, nicotine exposure provides a more accurate reflection of smoking status, significantly enhancing the scientific robustness and credibility of the study's conclusions. I recommend structuring the study into two main sections: one focusing on population databases to systematically explore the association between smoking and obesity, and another delving into microbiome research to uncover potential biological mechanisms. If it is not feasible to address both aspects comprehensively, I suggest omitting the population database analysis and concentrating on the microbiome study to ensure a clearer and more focused research narrative. (2)The conclusions drawn by the author from the population database are not necessarily specific to the male population. Currently, the # values from the CHNS and CHARLS databases, which are also based on Chinese populations, are relatively large, whereas the # values from the WELL- China cohort and the Lanxi cohort are relatively smaller. How should the author explain this discrepancy? Are there feasible methods to address or adjust this issue? (3)Former smokers should not be overlooked. Existing research evidence indicates that individuals who quit smoking may experience phenomena such as weight gain; however, quitting smoking can also significantly reduce the risk of adverse health outcomes caused by smoking. This is a highly valuable and very interesting topic for research. (4)It is recommended to provide references for certain definitions described in the methodology section, such as the definition of obesity-related cancers and the reasons for choosing a fixed- effects model. Additionally, the author is advised to examine and explain relevant factors one by one to enhance the rigor and scientific validity of the study. (5)The greatest limitation of this study lies in the excessive complexity of the research content, while the description of each part is not detailed or rigorous enough, with some content missing. For example, in the cohort analysis of disease occurrence, the baseline characteristics of the study population (such as comorbidities) were not sufficiently elaborated. These issues are, in my opinion, the key areas that require revision. It is recommended that the author thoroughly refine the structure of the article to ensure clear logic and provide more detailed and scientifically robust descriptions of each section. (6)It is recommended to comprehensively polish the language of the article to ensure clear and fluent expression, as well as to standardize the use of capitalization. Additionally, the supplementary materials should be carefully reviewed and their order adjusted as necessary. (7)Considering the highly complex factors influencing gut microbiota and obesity, it is recommended to provide broader explanations from multiple perspectives regarding the contradictions between population studies and microbiome research. Furthermore, it is suggested to enhance the literature review in the introduction and discussion sections.	BMC	12916_2025_3969_ReviewerReport_V0_R2.pdf	{ "criticism": 4, "example": 0, "importance_and_relevance": 11, "materials_and_methods": 19, "praise": 3, "presentation_and_reporting": 10, "results_and_discussion": 5, "suggestion_and_solution": 16, "total": 32 }	{ "criticism": 0.125, "example": 0, "importance_and_relevance": 0.34375, "materials_and_methods": 0.59375, "praise": 0.09375, "presentation_and_reporting": 0.3125, "results_and_discussion": 0.15625, "suggestion_and_solution": 0.5 }	2.125	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 0 Date: 29 Nov 2024\nReviewer's report:\nShankuan Zhu's team utilised population databases to deeply investigate the relationship\nbetween smoking and body weight, revealing a downward trend in BMI among smokers.\nIn subsequent research, through microbiome data analyses of the WELL-China cohort and\nthe Lanxi cohort, it was found that the Smoking-Associated Microbiome Index (SMI) was\nsignificantly positively correlated with BMI and other obesity indicators. Further analyses\nindicated that SMI was also closely associated with obesity-related diseases. Moreover, the study\neffectively addressed potential familial confounding factors through sibling cohort analyses.\nBased on these findings, the authors attempted to provide a plausible explanation for the\nsmoking-obesity paradox. This study's topic is novel and its content highly compelling, offering\na fresh perspective on the complex relationship between smoking and obesity while carrying\nsignificant public health implications and practical value. However, I still have some questions\nthat require clarification from the authors.\n(1)Given that the authors have chosen both population-based studies and microbiome research\nto address this issue, it would be beneficial to reflect this dual research perspective clearly from\nthe title. However, based on the current study content, the focus seems primarily on microbiome-\nrelated analyses. To more comprehensively explore the relationship between smoking and\nobesity, the inclusion of additional indicators beyond BMI could be considered. For instance,\ndatasets like the UK Biobank or NHANES could provide a broader perspective and enrich the\nstudy with supplementary data. Most importantly, it is crucial to incorporate a more objective\nmeasure of smoking exposure, such as nicotine levels. Compared to self-reported questionnaires,\nnicotine exposure provides a more accurate reflection of smoking status, significantly enhancing\nthe scientific robustness and credibility of the study's conclusions. I recommend structuring\nthe study into two main sections: one focusing on population databases to systematically\nexplore the association between smoking and obesity, and another delving into microbiome\nresearch to uncover potential biological mechanisms. If it is not feasible to address both aspects\ncomprehensively, I suggest omitting the population database analysis and concentrating on the\nmicrobiome study to ensure a clearer and more focused research narrative.\n(2)The conclusions drawn by the author from the population database are not necessarily specific\nto the male population. Currently, the # values from the CHNS and CHARLS databases, which\nare also based on Chinese populations, are relatively large, whereas the # values from the WELL-\nChina cohort and the Lanxi cohort are relatively smaller. How should the author explain this\ndiscrepancy? Are there feasible methods to address or adjust this issue?\n(3)Former smokers should not be overlooked. Existing research evidence indicates that\nindividuals who quit smoking may experience phenomena such as weight gain; however,\nquitting smoking can also significantly reduce the risk of adverse health outcomes caused by\nsmoking. This is a highly valuable and very interesting topic for research.\n(4)It is recommended to provide references for certain definitions described in the methodology\nsection, such as the definition of obesity-related cancers and the reasons for choosing a fixed-\neffects model. Additionally, the author is advised to examine and explain relevant factors one\nby one to enhance the rigor and scientific validity of the study.\n(5)The greatest limitation of this study lies in the excessive complexity of the research content,\nwhile the description of each part is not detailed or rigorous enough, with some content missing.\nFor example, in the cohort analysis of disease occurrence, the baseline characteristics of the\nstudy population (such as comorbidities) were not sufficiently elaborated. These issues are, in\nmy opinion, the key areas that require revision. It is recommended that the author thoroughly\nrefine the structure of the article to ensure clear logic and provide more detailed and scientifically\nrobust descriptions of each section.\n(6)It is recommended to comprehensively polish the language of the article to ensure clear\nand fluent expression, as well as to standardize the use of capitalization. Additionally, the\nsupplementary materials should be carefully reviewed and their order adjusted as necessary.\n(7)Considering the highly complex factors influencing gut microbiota and obesity, it is\nrecommended to provide broader explanations from multiple perspectives regarding the\ncontradictions between population studies and microbiome research. Furthermore, it is\nsuggested to enhance the literature review in the introduction and discussion sections.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 04 Dec 2024 Reviewer's report: I enjoy reading this manuscript. I have the following comments for consideration. 1. Could the authors justify the exclusion of participants with BMI <18.5? 2. Why did the authors use a lenient threshold of 0.25 for the Q value (FDR-adjusted p-value)? 3. Why did the authors analyze SMI in tertiles rather than as a continuous variable? 4. Could the authors include overall cancer as an outcome to increase the power? 5. Figure 1 - What are the differences in BMI between never smokers and former smokers and between current smokers and former smokers? 6. Figure 3 - I recommend the use of log-scale x-axis for odds ratio. 7. Line 376 - What are the potential mechanisms of smoking decreasing obesity?	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 0 Date: 04 Dec 2024 Reviewer's report: I enjoy reading this manuscript. I have the following comments for consideration. 1. Could the authors justify the exclusion of participants with BMI <18.5? 2. Why did the authors use a lenient threshold of 0.25 for the Q value (FDR-adjusted p-value)? 3. Why did the authors analyze SMI in tertiles rather than as a continuous variable? 4. Could the authors include overall cancer as an outcome to increase the power? 5. Figure 1 - What are the differences in BMI between never smokers and former smokers and between current smokers and former smokers? 6. Figure 3 - I recommend the use of log-scale x-axis for odds ratio. 7. Line 376 - What are the potential mechanisms of smoking decreasing obesity?	BMC	12916_2025_3969_ReviewerReport_V0_R5.pdf	{ "criticism": 0, "example": 3, "importance_and_relevance": 0, "materials_and_methods": 6, "praise": 1, "presentation_and_reporting": 1, "results_and_discussion": 2, "suggestion_and_solution": 1, "total": 16 }	{ "criticism": 0, "example": 0.1875, "importance_and_relevance": 0, "materials_and_methods": 0.375, "praise": 0.0625, "presentation_and_reporting": 0.0625, "results_and_discussion": 0.125, "suggestion_and_solution": 0.0625 }	0.875	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 0 Date: 04 Dec 2024\nReviewer's report:\nI enjoy reading this manuscript. I have the following comments for consideration.\n1. Could the authors justify the exclusion of participants with BMI <18.5?\n2. Why did the authors use a lenient threshold of 0.25 for the Q value (FDR-adjusted p-value)?\n3. Why did the authors analyze SMI in tertiles rather than as a continuous variable?\n4. Could the authors include overall cancer as an outcome to increase the power?\n5. Figure 1 - What are the differences in BMI between never smokers and former smokers and\nbetween current smokers and former smokers?\n6. Figure 3 - I recommend the use of log-scale x-axis for odds ratio.\n7. Line 376 - What are the potential mechanisms of smoking decreasing obesity?", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 1 Date: 07 Jan 2025 Reviewer's report: The author has addressed all my concerns. However, there are still minor details that require revision. 1.The author needs to recheck the citations of figures and tables in the manuscript, as I noticed that Figure 1 seems to have not been uploaded. 2.Both BMI and smoking are associated with the recently proposed Life’s Essential 8 by the American Heart Association, and both are closely linked to vascular inflammation. Moreover, existing studies have supported the role of the gut microbiota as a key mediator of vascular wall inflammation. Additional discussion on this topic is warranted to expand on the public health significance of this study.(DOI: 10.3390/microorganisms12112341;DOI: 10.1016/ j.micinf.2024.105406)	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 1 Date: 07 Jan 2025 Reviewer's report: The author has addressed all my concerns. However, there are still minor details that require revision. 1.The author needs to recheck the citations of figures and tables in the manuscript, as I noticed that Figure 1 seems to have not been uploaded. 2.Both BMI and smoking are associated with the recently proposed Life’s Essential 8 by the American Heart Association, and both are closely linked to vascular inflammation. Moreover, existing studies have supported the role of the gut microbiota as a key mediator of vascular wall inflammation. Additional discussion on this topic is warranted to expand on the public health significance of this study.(DOI: 10.3390/microorganisms12112341;DOI: 10.1016/ j.micinf.2024.105406)	BMC	12916_2025_3969_ReviewerReport_V1_R2.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 2, "materials_and_methods": 1, "praise": 2, "presentation_and_reporting": 1, "results_and_discussion": 1, "suggestion_and_solution": 3, "total": 7 }	{ "criticism": 0.14285714285714285, "example": 0, "importance_and_relevance": 0.2857142857142857, "materials_and_methods": 0.14285714285714285, "praise": 0.2857142857142857, "presentation_and_reporting": 0.14285714285714285, "results_and_discussion": 0.14285714285714285, "suggestion_and_solution": 0.42857142857142855 }	1.571429	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 1 Date: 07 Jan 2025\nReviewer's report:\nThe author has addressed all my concerns. However, there are still minor details that require\nrevision.\n1.The author needs to recheck the citations of figures and tables in the manuscript, as I noticed\nthat Figure 1 seems to have not been uploaded.\n2.Both BMI and smoking are associated with the recently proposed Life’s Essential 8 by\nthe American Heart Association, and both are closely linked to vascular inflammation.\nMoreover, existing studies have supported the role of the gut microbiota as a key mediator of\nvascular wall inflammation. Additional discussion on this topic is warranted to expand on the\npublic health significance of this study.(DOI: 10.3390/microorganisms12112341;DOI: 10.1016/\nj.micinf.2024.105406)", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 1 Date: 11 Jan 2025 Reviewer's report: The authors' responses to my questions are reasonable.	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 1 Date: 11 Jan 2025 Reviewer's report: The authors' responses to my questions are reasonable.	BMC	12916_2025_3969_ReviewerReport_V1_R6.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 1 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0 }	0	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 1 Date: 11 Jan 2025\nReviewer's report:\nThe authors' responses to my questions are reasonable.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03969-4	Reviewed - bmc_medicine	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 1 Date: 22 Jan 2025 Reviewer's report: The authors have addressed all my concerns. I do not have further comments.	Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related diseases: results from two cohorts with sibling comparison analyses Version: 1 Date: 22 Jan 2025 Reviewer's report: The authors have addressed all my concerns. I do not have further comments.	BMC	12916_2025_3969_ReviewerReport_V1_R5.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 1, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 2 }	{ "criticism": 0.5, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0.5, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0 }	1	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Smoking-related gut microbiota alteration is associated with obesity and obesity-related\ndiseases: results from two cohorts with sibling comparison analyses\nVersion: 1 Date: 22 Jan 2025\nReviewer's report:\nThe authors have addressed all my concerns. I do not have further comments.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03978-3	Reviewed - bmc_medicine	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 0 Date: 15 Nov 2024 Reviewer's report: The current work is to raise awareness of multiple outcome event prediction, and present clinical examples where such prediction is essential to help inform individual decision-making. I have several comments: 1. This is not a original type article, it may be suitable as a technical note or statistical methods introduction. 2. in the example of CVD, this can be a composite outcome that include multiple events such as myocardial infarction, coronary disease and so on. 3. more explanations for the ISARIC 4C model in the main text is needed to facilitate the reading. 4. It is important to decide whether these outcome events are mutually exclusive; for example, bleeding and clotting can occur simutaneously in one patient. this can cause the situation different. The probability of events can sum to > 100%. 5. Machine learning/deep learning models can have multiple classification in the output; This is an important technical tool with the development of AI, they have special Taxonomy in machine learning context for the classification purpose. For example, in ensemble modeling ( https:// doi.org/10.1016/j.lers.2021.12.003), many models can handle regression and classification problem.	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 0 Date: 15 Nov 2024 Reviewer's report: The current work is to raise awareness of multiple outcome event prediction, and present clinical examples where such prediction is essential to help inform individual decision-making. I have several comments: 1. This is not a original type article, it may be suitable as a technical note or statistical methods introduction. 2. in the example of CVD, this can be a composite outcome that include multiple events such as myocardial infarction, coronary disease and so on. 3. more explanations for the ISARIC 4C model in the main text is needed to facilitate the reading. 4. It is important to decide whether these outcome events are mutually exclusive; for example, bleeding and clotting can occur simutaneously in one patient. this can cause the situation different. The probability of events can sum to > 100%. 5. Machine learning/deep learning models can have multiple classification in the output; This is an important technical tool with the development of AI, they have special Taxonomy in machine learning context for the classification purpose. For example, in ensemble modeling ( https:// doi.org/10.1016/j.lers.2021.12.003), many models can handle regression and classification problem.	BMC	12916_2025_3978_ReviewerReport_V0_R2.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 2, "materials_and_methods": 5, "praise": 1, "presentation_and_reporting": 2, "results_and_discussion": 4, "suggestion_and_solution": 3, "total": 12 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0.16666666666666666, "materials_and_methods": 0.4166666666666667, "praise": 0.08333333333333333, "presentation_and_reporting": 0.16666666666666666, "results_and_discussion": 0.3333333333333333, "suggestion_and_solution": 0.25 }	1.416667	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production\nof Clinical Prediction Models: defining a taxonomy of risk estimands\nVersion: 0 Date: 15 Nov 2024\nReviewer's report:\nThe current work is to raise awareness of multiple outcome event prediction, and present clinical\nexamples where such prediction is essential to help inform individual decision-making. I have\nseveral comments:\n1. This is not a original type article, it may be suitable as a technical note or statistical methods\nintroduction.\n2. in the example of CVD, this can be a composite outcome that include multiple events such\nas myocardial infarction, coronary disease and so on.\n3. more explanations for the ISARIC 4C model in the main text is needed to facilitate the\nreading.\n4. It is important to decide whether these outcome events are mutually exclusive; for example,\nbleeding and clotting can occur simutaneously in one patient. this can cause the situation\ndifferent. The probability of events can sum to > 100%.\n5. Machine learning/deep learning models can have multiple classification in the output; This is\nan important technical tool with the development of AI, they have special Taxonomy in machine\nlearning context for the classification purpose. For example, in ensemble modeling ( https://\ndoi.org/10.1016/j.lers.2021.12.003), many models can handle regression and classification\nproblem.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03978-3	Reviewed - bmc_medicine	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 0 Date: 09 Dec 2024 Reviewer: Ulrich Mansmann Reviewer's report: I do congratulate the authors for this comprehensive and needed paper. There are many medical contexts where healthcare decision-making needs to consider risks of multiple outcome events. Thus, CPMs need to estimate the probability of multiple outcome events co-occurring. Understanding when, and how, to incorporate multiple outcomes into CPM is challenging Relevant are intended estimand (the type of predicted risks that we wish the CPM to estimate). The submitted paper plans to give direction on how to manoeuvre through this difficult landscape. It raises awareness of multiple outcome event prediction, and presents clinical examples where such prediction is essential to help inform individual decision-making. It presents an important contribution to reflect on the ongoing algorithmizing of clinical decision making. The authors introduce four taxonomic levels: Level 1: Estimand is the Predicted Risk of a Single Outcome Level 2: Estimand is the Predicted Risk of Multiple Competing Outcome Events Level 3: Estimand is the Predicted Risk of a Composite of Multiple Outcomes Level 4: Estimand is the Predicted Risk of Different Combinations of Multiple Outcomes Here my comments: 1. General comment (1): The paper only looks on estimands for the risk estimation. It ignores that more complicated estimands are needed for the estimation of the patients’ advantage. The need to combine risk estimates with utilities or other values of patient preferences is ignores. 2. General comment (2): The paper ignores situations when the decision has to be performed regarding the choice on an optimal treatment out of a large set of options. For example a diabetic patient has to switch to a new therapy. 3. The introduction identifies two problems: 1.) Unappropriated use of the term multivariate analysis and 2.) Unprecise definitions of relevant estimands. It would be helpful to give a short example of an unprecise estimand (after line 20). 4. In 4.1. the prediction horizon is not clear. 5. Regarding box 1, it may helpful to describe risk estimands and risk estimates. The need to define timeframes and events (Outcome and prediction horizon) as well as the predictors used. The measurement(s) used for prediction that characterize the individual, that is, that individualize the predictions. The measurement(s) must be available at the moment(s) of intended use and measurement procedures must correspond to those in the setting of intended use. Especially the history of the patient or a clinical status at time of the prediction. 6. Chapter 2 uses the concept of utilities which is not mentioned in the introduction. There should be a short sentence preparing the use of utilities in chapter 2. 7. The paper does not touch the problem how to quantify the utilities. They may differ from patient to patient and may underly a multi-criteria problem (his own health, family aspects, professional aspects…). 8. Especially, each taxonomic level of the risk estimand needs a different concept to quantify utilities. We also use estimands for the utilities to get correct estimates of the utilities. 9. How helpful is the section between lines 99 and 109? The authors discuss very complex models which a non-trivial and in general unfamiliar background. Would it be possible to exemplify this in more detail? Maybe in an electronic methodological supplement? 10. Decision-making is more than risk estimation. Why is the estimand of interest in this paper not the final risk/utility combination. The title should take care about this by stating that the authors study risk estimands. 11. The examples given in lines 160 – 162 may also be seen as ordinal: Benign / borderline / metastatic. 12. It would be good to extend Box 1 from the estimand to the estimator and the estimate. 13. At the end of the paper, I was a bit confused. What is the purpose of the paper? Who is the addressed reader? Is it the clinician who wants to apply CPMs? Is it the clinician who evaluates the CPM? Is it the biostatistician starting the journey of CPM development? It would be good to give to a clinician an example how to apply the taxonomy and how to gain advantage out of it.	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 0 Date: 09 Dec 2024 Reviewer: Ulrich Mansmann Reviewer's report: I do congratulate the authors for this comprehensive and needed paper. There are many medical contexts where healthcare decision-making needs to consider risks of multiple outcome events. Thus, CPMs need to estimate the probability of multiple outcome events co-occurring. Understanding when, and how, to incorporate multiple outcomes into CPM is challenging Relevant are intended estimand (the type of predicted risks that we wish the CPM to estimate). The submitted paper plans to give direction on how to manoeuvre through this difficult landscape. It raises awareness of multiple outcome event prediction, and presents clinical examples where such prediction is essential to help inform individual decision-making. It presents an important contribution to reflect on the ongoing algorithmizing of clinical decision making. The authors introduce four taxonomic levels: Level 1: Estimand is the Predicted Risk of a Single Outcome Level 2: Estimand is the Predicted Risk of Multiple Competing Outcome Events Level 3: Estimand is the Predicted Risk of a Composite of Multiple Outcomes Level 4: Estimand is the Predicted Risk of Different Combinations of Multiple Outcomes Here my comments: 1. General comment (1): The paper only looks on estimands for the risk estimation. It ignores that more complicated estimands are needed for the estimation of the patients’ advantage. The need to combine risk estimates with utilities or other values of patient preferences is ignores. 2. General comment (2): The paper ignores situations when the decision has to be performed regarding the choice on an optimal treatment out of a large set of options. For example a diabetic patient has to switch to a new therapy. 3. The introduction identifies two problems: 1.) Unappropriated use of the term multivariate analysis and 2.) Unprecise definitions of relevant estimands. It would be helpful to give a short example of an unprecise estimand (after line 20). 4. In 4.1. the prediction horizon is not clear. 5. Regarding box 1, it may helpful to describe risk estimands and risk estimates. The need to define timeframes and events (Outcome and prediction horizon) as well as the predictors used. The measurement(s) used for prediction that characterize the individual, that is, that individualize the predictions. The measurement(s) must be available at the moment(s) of intended use and measurement procedures must correspond to those in the setting of intended use. Especially the history of the patient or a clinical status at time of the prediction. 6. Chapter 2 uses the concept of utilities which is not mentioned in the introduction. There should be a short sentence preparing the use of utilities in chapter 2. 7. The paper does not touch the problem how to quantify the utilities. They may differ from patient to patient and may underly a multi-criteria problem (his own health, family aspects, professional aspects…). 8. Especially, each taxonomic level of the risk estimand needs a different concept to quantify utilities. We also use estimands for the utilities to get correct estimates of the utilities. 9. How helpful is the section between lines 99 and 109? The authors discuss very complex models which a non-trivial and in general unfamiliar background. Would it be possible to exemplify this in more detail? Maybe in an electronic methodological supplement? 10. Decision-making is more than risk estimation. Why is the estimand of interest in this paper not the final risk/utility combination. The title should take care about this by stating that the authors study risk estimands. 11. The examples given in lines 160 – 162 may also be seen as ordinal: Benign / borderline / metastatic. 12. It would be good to extend Box 1 from the estimand to the estimator and the estimate. 13. At the end of the paper, I was a bit confused. What is the purpose of the paper? Who is the addressed reader? Is it the clinician who wants to apply CPMs? Is it the clinician who evaluates the CPM? Is it the biostatistician starting the journey of CPM development? It would be good to give to a clinician an example how to apply the taxonomy and how to gain advantage out of it.	BMC	12916_2025_3978_ReviewerReport_V0_R3.pdf	{ "criticism": 5, "example": 5, "importance_and_relevance": 2, "materials_and_methods": 26, "praise": 3, "presentation_and_reporting": 9, "results_and_discussion": 10, "suggestion_and_solution": 11, "total": 57 }	{ "criticism": 0.08771929824561403, "example": 0.08771929824561403, "importance_and_relevance": 0.03508771929824561, "materials_and_methods": 0.45614035087719296, "praise": 0.05263157894736842, "presentation_and_reporting": 0.15789473684210525, "results_and_discussion": 0.17543859649122806, "suggestion_and_solution": 0.19298245614035087 }	1.245614	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production\nof Clinical Prediction Models: defining a taxonomy of risk estimands\nVersion: 0 Date: 09 Dec 2024\nReviewer: Ulrich Mansmann\nReviewer's report:\nI do congratulate the authors for this comprehensive and needed paper.\nThere are many medical contexts where healthcare decision-making needs to consider risks of\nmultiple outcome events. Thus, CPMs need to estimate the probability of multiple outcome\nevents co-occurring. Understanding when, and how, to incorporate multiple outcomes into CPM\nis challenging Relevant are intended estimand (the type of predicted risks that we wish the\nCPM to estimate). The submitted paper plans to give direction on how to manoeuvre through\nthis difficult landscape. It raises awareness of multiple outcome event prediction, and presents\nclinical examples where such prediction is essential to help inform individual decision-making.\nIt presents an important contribution to reflect on the ongoing algorithmizing of clinical decision\nmaking.\nThe authors introduce four taxonomic levels:\nLevel 1: Estimand is the Predicted Risk of a Single Outcome\nLevel 2: Estimand is the Predicted Risk of Multiple Competing Outcome Events\nLevel 3: Estimand is the Predicted Risk of a Composite of Multiple Outcomes\nLevel 4: Estimand is the Predicted Risk of Different Combinations of Multiple Outcomes\nHere my comments:\n1. General comment (1): The paper only looks on estimands for the risk estimation. It ignores\nthat more complicated estimands are needed for the estimation of the patients’ advantage. The\nneed to combine risk estimates with utilities or other values of patient preferences is ignores.\n2. General comment (2): The paper ignores situations when the decision has to be performed\nregarding the choice on an optimal treatment out of a large set of options. For example a diabetic\npatient has to switch to a new therapy.\n3. The introduction identifies two problems: 1.) Unappropriated use of the term multivariate\nanalysis and 2.) Unprecise definitions of relevant estimands. It would be helpful to give a short\nexample of an unprecise estimand (after line 20).\n4. In 4.1. the prediction horizon is not clear.\n5. Regarding box 1, it may helpful to describe risk estimands and risk estimates. The need\nto define timeframes and events (Outcome and prediction horizon) as well as the predictors\nused. The measurement(s) used for prediction that characterize the individual, that is, that\nindividualize the predictions. The measurement(s) must be available at the moment(s) of\nintended use and measurement procedures must correspond to those in the setting of intended\nuse. Especially the history of the patient or a clinical status at time of the prediction.\n6. Chapter 2 uses the concept of utilities which is not mentioned in the introduction. There should\nbe a short sentence preparing the use of utilities in chapter 2.\n7. The paper does not touch the problem how to quantify the utilities. They may differ from\npatient to patient and may underly a multi-criteria problem (his own health, family aspects,\nprofessional aspects…).\n8. Especially, each taxonomic level of the risk estimand needs a different concept to quantify\nutilities. We also use estimands for the utilities to get correct estimates of the utilities.\n9. How helpful is the section between lines 99 and 109? The authors discuss very complex\nmodels which a non-trivial and in general unfamiliar background. Would it be possible to\nexemplify this in more detail? Maybe in an electronic methodological supplement?\n10. Decision-making is more than risk estimation. Why is the estimand of interest in this paper\nnot the final risk/utility combination. The title should take care about this by stating that the\nauthors study risk estimands.\n11. The examples given in lines 160 – 162 may also be seen as ordinal: Benign / borderline /\nmetastatic.\n12. It would be good to extend Box 1 from the estimand to the estimator and the estimate.\n13. At the end of the paper, I was a bit confused. What is the purpose of the paper? Who is the\naddressed reader? Is it the clinician who wants to apply CPMs? Is it the clinician who evaluates\nthe CPM? Is it the biostatistician starting the journey of CPM development? It would be good to\ngive to a clinician an example how to apply the taxonomy and how to gain advantage out of it.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03978-3	Reviewed - bmc_medicine	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 1 Date: 27 Jan 2025 Reviewer's report: my comments are well addressed.	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 1 Date: 27 Jan 2025 Reviewer's report: my comments are well addressed.	BMC	12916_2025_3978_ReviewerReport_V1_R2.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 1, "praise": 1, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 1 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 1, "praise": 1, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0 }	2	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production\nof Clinical Prediction Models: defining a taxonomy of risk estimands\nVersion: 1 Date: 27 Jan 2025\nReviewer's report:\nmy comments are well addressed.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03978-3	Reviewed - bmc_medicine	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 1 Date: 02 Feb 2025 Reviewer: Ulrich Mansmann Reviewer's report: Many thanks for this important and timely paper. The use of prediction models to improve clinical decision making is growing and getting more complex. The provided manuscript provides guidance to structrure this growing field of research activities. I miss a few aspects in the paper and it would be great if you could incorporate them: 1.) Please mention, that one problem of multi-outcome prediction is also the availability of appropriate data on the multiple outcomes under study of high quality. Clinical trial data may be helpfull. It is not clear how data based of electronisc patients recourds may help 2.) A central concept in your manuscript are utilities. From where do you get the utilities? In your paper they are simply given. No reference for realistic utilities can be found. Or is each patient defining her/his own utilities? What are tools to elucidate utilities? 3.) At the end you mention dynamic simulation models to derive for a patient a decision between different options. Multi-state models may be useful but there are may other simulation techniques for patient histories which could help to settle complex scenarios. There is a causal issue if you compare different scenarios for a specific patient. An interesting tool to derive such dynamic models from observational data are model around the g-formula. 4.) You should also mention the large body of literature on multi-criteria decision making.	Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production of Clinical Prediction Models: defining a taxonomy of risk estimands Version: 1 Date: 02 Feb 2025 Reviewer: Ulrich Mansmann Reviewer's report: Many thanks for this important and timely paper. The use of prediction models to improve clinical decision making is growing and getting more complex. The provided manuscript provides guidance to structrure this growing field of research activities. I miss a few aspects in the paper and it would be great if you could incorporate them: 1.) Please mention, that one problem of multi-outcome prediction is also the availability of appropriate data on the multiple outcomes under study of high quality. Clinical trial data may be helpfull. It is not clear how data based of electronisc patients recourds may help 2.) A central concept in your manuscript are utilities. From where do you get the utilities? In your paper they are simply given. No reference for realistic utilities can be found. Or is each patient defining her/his own utilities? What are tools to elucidate utilities? 3.) At the end you mention dynamic simulation models to derive for a patient a decision between different options. Multi-state models may be useful but there are may other simulation techniques for patient histories which could help to settle complex scenarios. There is a causal issue if you compare different scenarios for a specific patient. An interesting tool to derive such dynamic models from observational data are model around the g-formula. 4.) You should also mention the large body of literature on multi-criteria decision making.	BMC	12916_2025_3978_ReviewerReport_V1_R3.pdf	{ "criticism": 3, "example": 0, "importance_and_relevance": 4, "materials_and_methods": 13, "praise": 4, "presentation_and_reporting": 6, "results_and_discussion": 3, "suggestion_and_solution": 6, "total": 20 }	{ "criticism": 0.15, "example": 0, "importance_and_relevance": 0.2, "materials_and_methods": 0.65, "praise": 0.2, "presentation_and_reporting": 0.3, "results_and_discussion": 0.15, "suggestion_and_solution": 0.3 }	1.95	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: A Decision-Analytical Perspective on Incorporating Multiple Outcomes in the Production\nof Clinical Prediction Models: defining a taxonomy of risk estimands\nVersion: 1 Date: 02 Feb 2025\nReviewer: Ulrich Mansmann\nReviewer's report:\nMany thanks for this important and timely paper. The use of prediction models to improve\nclinical decision making is growing and getting more complex. The provided manuscript\nprovides guidance to structrure this growing field of research activities.\nI miss a few aspects in the paper and it would be great if you could incorporate them:\n1.) Please mention, that one problem of multi-outcome prediction is also the availability of\nappropriate data on the multiple outcomes under study of high quality. Clinical trial data may\nbe helpfull. It is not clear how data based of electronisc patients recourds may help\n2.) A central concept in your manuscript are utilities. From where do you get the utilities? In\nyour paper they are simply given. No reference for realistic utilities can be found. Or is each\npatient defining her/his own utilities? What are tools to elucidate utilities?\n3.) At the end you mention dynamic simulation models to derive for a patient a decision between\ndifferent options. Multi-state models may be useful but there are may other\nsimulation techniques for patient histories which could help to settle complex scenarios. There\nis a causal issue if you compare different scenarios for a specific patient. An interesting tool to\nderive such dynamic models from observational data are model around the g-formula.\n4.) You should also mention the large body of literature on multi-criteria decision making.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03952-z	Reviewed - bmc_medicine	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 0 Date: 08 Oct 2024 Reviewer's report: This is a large scale multinational study of psychiatric sequelae of COVID-19. The paper is clearly written, the methods are reasonable and thoroughly described. A main strength of the authors’ analytical approach is to examine time-dependence of the effect of COVID-19 on mental health, something that has not often been done. The authors acknowledge some important limitations, in my opinion lack of info on acute illness severity is a major one. My main comments regard the interpretation of the findings. Comment #1 Lines 309-312: “Biologically, the SARS-CoV-2 virus can invade the central nervous system, leading to neuroinflammation and potential neural damage, facilitated by the virus’s interaction with angiotensin-converting enzyme 2 (ACE2) receptors on neurons and glial cells, disrupting normal brain function and possibly resulting in psychiatric symptoms.35-37” In my opinion, the references provided by the authors do not give any empirical support for CNS invasion in humans. They are from 2020-2021 and only include conjectures and the theoretical possibility of this happening. In fact, recent reviews (eg. https:// doi.org/10.1016/j.mib.2024.102474) state that the mechanisms behind neurological and psychiatric manifestations connected to SARS-CoV-2 are not understood. Invasion of the CNS by SARS-CoV-2 has poor empirical support in humans. It is not resolved if invasion occurs, let alone the mechanism of entry (through PNS or hematogenically). Well conducted clinical studies that I am aware of (eg. https://pubmed.ncbi.nlm.nih.gov/35604688/, there may be several others) do not find credible evidence of viral CNS invasion as a direct cause of neurological manifestation in acute COVID-19. It then seems far-fetched that they should play a role in long- term sequelae. Systemic inflammation not directly related to viral invasion of CNS seems to me an important mechanism of neurological manifestations in acute COVID-19, potentially resulting in long-term sequelae. The level of systemic inflammation is obviously related to acute illness severity and can be seen as especially relevant in primary infections in unvaccinated individuals, which the current study mainly examines. If the authors wish to stand by the statement above, that invasion of the CNS may be a driver of neuroinflammation, the authors have to provide references that clearly support this mechanism. Otherwise I suggest that they revise this. Comment #2 The authors discuss (page 16, first paragraph) several explanations to the short-lived elevation of risk for mental disorders. The authors rightly acknowledge the possibility of detection bias (p. 18, line 377 and onwards). I think the authors may consider to even bring up detection bias as a reasonable explanation to their findings alongside the others, rather than a limitation. The authors have made reasonable efforts to correct for detection bias by excluding prevalent cases. Still, mental disorders are severely underdiagnosed or are associated with considerable diagnostic delay. Arguments for this interpretation: 1. It is difficult to understand how COVID-19 could double the incidence of chronic health conditions such as neurodevelopmental disorders (how can ADHD or autism be of new onset in adults, for example) and dementia (an in most cases slowly progressive disorder with relatively well defined pathophysiology). 2. Considerable heterogeneity between countries in relative risks are difficult to explain by any SARS-CoV-2 / COVID-19 related mechanism. 3. There is no established knowledge about such a mechanism (see previous comment). 4. The short-lived effect is at least compatible with detection bias, the longer effect on dementia in the UK can simply reflect longer waiting times for dementia work-up. Bringing up the possibility that authors simply find ongoing but undiagnosed disorders has important implications for further research. If findings are to some extent driven by detection bias, we should not put large effort into finding putative biological mechanisms linking COVID-19 with neurodevelopmental disorders or dementia in adults. Comment #3 Conclusions: “This phenomenon may be attributed to the redirection of healthcare resources, diverse pandemic management strategies, and differing rates of socioeconomic recovery rather than the direct impact of COVID-19 itself.” Comment #4 Is it not a limitation that the authors study sequelae after COVID-19 in 2020-2021, i.e. mostly primary infections in unvaccinated people? Such infections are now a small proportion of all cases of COVID-19. It is known that re-infection or breakthrough infection is associated with lower risk of adverse outcomes in general. Thus findings cannot be generalized to current situation as authors are not able to take illness severity, vaccination or previous infection into account.	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 0 Date: 08 Oct 2024 Reviewer's report: This is a large scale multinational study of psychiatric sequelae of COVID-19. The paper is clearly written, the methods are reasonable and thoroughly described. A main strength of the authors’ analytical approach is to examine time-dependence of the effect of COVID-19 on mental health, something that has not often been done. The authors acknowledge some important limitations, in my opinion lack of info on acute illness severity is a major one. My main comments regard the interpretation of the findings. Comment #1 Lines 309-312: “Biologically, the SARS-CoV-2 virus can invade the central nervous system, leading to neuroinflammation and potential neural damage, facilitated by the virus’s interaction with angiotensin-converting enzyme 2 (ACE2) receptors on neurons and glial cells, disrupting normal brain function and possibly resulting in psychiatric symptoms.35-37” In my opinion, the references provided by the authors do not give any empirical support for CNS invasion in humans. They are from 2020-2021 and only include conjectures and the theoretical possibility of this happening. In fact, recent reviews (eg. https:// doi.org/10.1016/j.mib.2024.102474) state that the mechanisms behind neurological and psychiatric manifestations connected to SARS-CoV-2 are not understood. Invasion of the CNS by SARS-CoV-2 has poor empirical support in humans. It is not resolved if invasion occurs, let alone the mechanism of entry (through PNS or hematogenically). Well conducted clinical studies that I am aware of (eg. https://pubmed.ncbi.nlm.nih.gov/35604688/, there may be several others) do not find credible evidence of viral CNS invasion as a direct cause of neurological manifestation in acute COVID-19. It then seems far-fetched that they should play a role in long- term sequelae. Systemic inflammation not directly related to viral invasion of CNS seems to me an important mechanism of neurological manifestations in acute COVID-19, potentially resulting in long-term sequelae. The level of systemic inflammation is obviously related to acute illness severity and can be seen as especially relevant in primary infections in unvaccinated individuals, which the current study mainly examines. If the authors wish to stand by the statement above, that invasion of the CNS may be a driver of neuroinflammation, the authors have to provide references that clearly support this mechanism. Otherwise I suggest that they revise this. Comment #2 The authors discuss (page 16, first paragraph) several explanations to the short-lived elevation of risk for mental disorders. The authors rightly acknowledge the possibility of detection bias (p. 18, line 377 and onwards). I think the authors may consider to even bring up detection bias as a reasonable explanation to their findings alongside the others, rather than a limitation. The authors have made reasonable efforts to correct for detection bias by excluding prevalent cases. Still, mental disorders are severely underdiagnosed or are associated with considerable diagnostic delay. Arguments for this interpretation: 1. It is difficult to understand how COVID-19 could double the incidence of chronic health conditions such as neurodevelopmental disorders (how can ADHD or autism be of new onset in adults, for example) and dementia (an in most cases slowly progressive disorder with relatively well defined pathophysiology). 2. Considerable heterogeneity between countries in relative risks are difficult to explain by any SARS-CoV-2 / COVID-19 related mechanism. 3. There is no established knowledge about such a mechanism (see previous comment). 4. The short-lived effect is at least compatible with detection bias, the longer effect on dementia in the UK can simply reflect longer waiting times for dementia work-up. Bringing up the possibility that authors simply find ongoing but undiagnosed disorders has important implications for further research. If findings are to some extent driven by detection bias, we should not put large effort into finding putative biological mechanisms linking COVID-19 with neurodevelopmental disorders or dementia in adults. Comment #3 Conclusions: “This phenomenon may be attributed to the redirection of healthcare resources, diverse pandemic management strategies, and differing rates of socioeconomic recovery rather than the direct impact of COVID-19 itself.” Comment #4 Is it not a limitation that the authors study sequelae after COVID-19 in 2020-2021, i.e. mostly primary infections in unvaccinated people? Such infections are now a small proportion of all cases of COVID-19. It is known that re-infection or breakthrough infection is associated with lower risk of adverse outcomes in general. Thus findings cannot be generalized to current situation as authors are not able to take illness severity, vaccination or previous infection into account.	BMC	12916_2025_3952_ReviewerReport_V0_R1.pdf	{ "criticism": 7, "example": 4, "importance_and_relevance": 4, "materials_and_methods": 9, "praise": 5, "presentation_and_reporting": 2, "results_and_discussion": 8, "suggestion_and_solution": 5, "total": 38 }	{ "criticism": 0.18421052631578946, "example": 0.10526315789473684, "importance_and_relevance": 0.10526315789473684, "materials_and_methods": 0.23684210526315788, "praise": 0.13157894736842105, "presentation_and_reporting": 0.05263157894736842, "results_and_discussion": 0.21052631578947367, "suggestion_and_solution": 0.13157894736842105 }	1.157895	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study\nVersion: 0 Date: 08 Oct 2024\nReviewer's report:\nThis is a large scale multinational study of psychiatric sequelae of COVID-19. The paper is\nclearly written, the methods are reasonable and thoroughly described. A main strength of the\nauthors’ analytical approach is to examine time-dependence of the effect of COVID-19 on\nmental health, something that has not often been done. The authors acknowledge some important\nlimitations, in my opinion lack of info on acute illness severity is a major one. My main\ncomments regard the interpretation of the findings.\nComment #1\nLines 309-312: “Biologically, the SARS-CoV-2 virus can invade the central nervous system,\nleading to neuroinflammation and potential neural damage, facilitated by the virus’s interaction\nwith angiotensin-converting enzyme 2 (ACE2) receptors on neurons and glial cells, disrupting\nnormal brain function and possibly resulting in psychiatric symptoms.35-37”\nIn my opinion, the references provided by the authors do not give any empirical support\nfor CNS invasion in humans. They are from 2020-2021 and only include conjectures\nand the theoretical possibility of this happening. In fact, recent reviews (eg. https://\ndoi.org/10.1016/j.mib.2024.102474) state that the mechanisms behind neurological and\npsychiatric manifestations connected to SARS-CoV-2 are not understood. Invasion of the CNS\nby SARS-CoV-2 has poor empirical support in humans. It is not resolved if invasion occurs, let\nalone the mechanism of entry (through PNS or hematogenically). Well conducted clinical studies\nthat I am aware of (eg. https://pubmed.ncbi.nlm.nih.gov/35604688/, there may be several\nothers) do not find credible evidence of viral CNS invasion as a direct cause of neurological\nmanifestation in acute COVID-19. It then seems far-fetched that they should play a role in long-\nterm sequelae. Systemic inflammation not directly related to viral invasion of CNS seems to\nme an important mechanism of neurological manifestations in acute COVID-19, potentially\nresulting in long-term sequelae. The level of systemic inflammation is obviously related to acute\nillness severity and can be seen as especially relevant in primary infections in unvaccinated\nindividuals, which the current study mainly examines.\nIf the authors wish to stand by the statement above, that invasion of the CNS may be a driver of\nneuroinflammation, the authors have to provide references that clearly support this mechanism.\nOtherwise I suggest that they revise this.\nComment #2\nThe authors discuss (page 16, first paragraph) several explanations to the short-lived elevation\nof risk for mental disorders. The authors rightly acknowledge the possibility of detection bias (p.\n18, line 377 and onwards). I think the authors may consider to even bring up detection bias as a\nreasonable explanation to their findings alongside the others, rather than a limitation. The authors\nhave made reasonable efforts to correct for detection bias by excluding prevalent cases. Still,\nmental disorders are severely underdiagnosed or are associated with considerable diagnostic\ndelay. Arguments for this interpretation: 1. It is difficult to understand how COVID-19 could\ndouble the incidence of chronic health conditions such as neurodevelopmental disorders (how\ncan ADHD or autism be of new onset in adults, for example) and dementia (an in most cases\nslowly progressive disorder with relatively well defined pathophysiology). 2. Considerable\nheterogeneity between countries in relative risks are difficult to explain by any SARS-CoV-2 /\nCOVID-19 related mechanism. 3. There is no established knowledge about such a mechanism\n(see previous comment). 4. The short-lived effect is at least compatible with detection bias,\nthe longer effect on dementia in the UK can simply reflect longer waiting times for dementia\nwork-up. Bringing up the possibility that authors simply find ongoing but undiagnosed disorders\nhas important implications for further research. If findings are to some extent driven by\ndetection bias, we should not put large effort into finding putative biological mechanisms linking\nCOVID-19 with neurodevelopmental disorders or dementia in adults.\nComment #3\nConclusions: “This phenomenon may be attributed to the redirection of healthcare resources,\ndiverse pandemic management strategies, and differing rates of socioeconomic recovery rather\nthan the direct impact of COVID-19 itself.”\nComment #4\nIs it not a limitation that the authors study sequelae after COVID-19 in 2020-2021, i.e. mostly\nprimary infections in unvaccinated people? Such infections are now a small proportion of all\ncases of COVID-19. It is known that re-infection or breakthrough infection is associated with\nlower risk of adverse outcomes in general. Thus findings cannot be generalized to current\nsituation as authors are not able to take illness severity, vaccination or previous infection into\naccount.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03952-z	Reviewed - bmc_medicine	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 0 Date: 21 Oct 2024 Reviewer's report: This study used data from five countries and investigated the short-, medium, and long-term risks of psychiatric and neuropsychiatric disorders following COVID-19 infection. The study is well-written, clear in introduction, results and discussions. I have some questions regarding the methods. 1. Line 189, “Only the first diagnosis of each outcome following the index date was used.” We know that psychiatric disorders are highly corelated with each other. Is it enough to only exclude those with targeted outcome? 2. Line 195, “. For the analysis of each outcome, individuals were excluded if they had the outcome of interest within 365 days prior to the start of the short- (the index date), medium-, and long-term observation to ensure the identification of incident cases.” Authors exclude individuals if they had the outcome of interest within one year prior to the start of the observations. But some psychiatric disorders can be exit for a longer period and are frequently relapsed. E.g., After having depression once, people may experience a relapse, either before recovery is complete or as a later, separate episode. How to address this issue? 3. Another question regarding comparators: How about if they had a diagnosis of COVID-19 during follow-up? When excluding individuals having the outcome of interest before observation, did the balances after propensity score matching remain?	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 0 Date: 21 Oct 2024 Reviewer's report: This study used data from five countries and investigated the short-, medium, and long-term risks of psychiatric and neuropsychiatric disorders following COVID-19 infection. The study is well-written, clear in introduction, results and discussions. I have some questions regarding the methods. 1. Line 189, “Only the first diagnosis of each outcome following the index date was used.” We know that psychiatric disorders are highly corelated with each other. Is it enough to only exclude those with targeted outcome? 2. Line 195, “. For the analysis of each outcome, individuals were excluded if they had the outcome of interest within 365 days prior to the start of the short- (the index date), medium-, and long-term observation to ensure the identification of incident cases.” Authors exclude individuals if they had the outcome of interest within one year prior to the start of the observations. But some psychiatric disorders can be exit for a longer period and are frequently relapsed. E.g., After having depression once, people may experience a relapse, either before recovery is complete or as a later, separate episode. How to address this issue? 3. Another question regarding comparators: How about if they had a diagnosis of COVID-19 during follow-up? When excluding individuals having the outcome of interest before observation, did the balances after propensity score matching remain?	BMC	12916_2025_3952_ReviewerReport_V0_R2.pdf	{ "criticism": 0, "example": 2, "importance_and_relevance": 0, "materials_and_methods": 6, "praise": 1, "presentation_and_reporting": 1, "results_and_discussion": 5, "suggestion_and_solution": 0, "total": 15 }	{ "criticism": 0, "example": 0.13333333333333333, "importance_and_relevance": 0, "materials_and_methods": 0.4, "praise": 0.06666666666666667, "presentation_and_reporting": 0.06666666666666667, "results_and_discussion": 0.3333333333333333, "suggestion_and_solution": 0 }	1	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study\nVersion: 0 Date: 21 Oct 2024\nReviewer's report:\nThis study used data from five countries and investigated the short-, medium, and long-term\nrisks of psychiatric and neuropsychiatric disorders following COVID-19 infection. The study\nis well-written, clear in introduction, results and discussions. I have some questions regarding\nthe methods.\n1. Line 189, “Only the first diagnosis of each outcome following the index date was used.” We\nknow that psychiatric disorders are highly corelated with each other. Is it enough to only exclude\nthose with targeted outcome?\n2. Line 195, “. For the analysis of each outcome, individuals were excluded if they had\nthe outcome of interest within 365 days prior to the start of the short- (the index date),\nmedium-, and long-term observation to ensure the identification of incident cases.” Authors\nexclude individuals if they had the outcome of interest within one year prior to the start of the\nobservations. But some psychiatric disorders can be exit for a longer period and are frequently\nrelapsed. E.g., After having depression once, people may experience a relapse, either before\nrecovery is complete or as a later, separate episode. How to address this issue?\n3. Another question regarding comparators: How about if they had a diagnosis of COVID-19\nduring follow-up? When excluding individuals having the outcome of interest before\nobservation, did the balances after propensity score matching remain?", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03952-z	Reviewed - bmc_medicine	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 0 Date: 31 Oct 2024 Reviewer's report: Authors should provide an abbreviated description of the databases they accessed across countries and are they comparable (e.g., do they represent diverse patients of socioeconomic backgrounds). - The authors noted that elevated risk for a psychiatric diagnosis the first 6 months following the infection. How do they know if the diagnosis is new versus continued? And those who had a diagnosis before and suddenly had exacerbation of symptoms? - The authors should consider completing separate analyses for those who had the infection during the peak of the pandemic (i.e., during social distancing and quarantine practices) as compared to towards the end or post-pandemic. I question if the initial elevated risk for psychiatric disorders are compounded by pandemic-related stressors such as reduced social supports, anxiety about pandemic-related disruptions to school/work/social life, etc. - What was the rational for the age cohorts (e.g. 18-24 years, 25-44 years, 45-64 years)? - Are sleep disorders defined as behavioral sleep disruptions (e.g. insomnia, hypersomnia, etc.) or sleep breathing disorders? - Neurodevelopmental disorders is a very broad group and is not very meaningful when authors collapse this all into one (Intellectual disability/IDD, autism and ADHD are all considered a developmental disorder but have different implications on functioning). Authors should strongly consider separating this into more common developmental disorders e.g., ADHD, IDD, Specific Learning Disability as it is much more likely some disorders are more evident with worsened symptoms (e.g., ADHD) whereas others neuropsychologist would not expect significant change following COVID (e.g. IDD). -Authors should include information on the cohort’s preexisting diagnoses (i.e., diagnoses pre- COVID infection) if they have the data, and if not, provide more in depth discussion about this.	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 0 Date: 31 Oct 2024 Reviewer's report: Authors should provide an abbreviated description of the databases they accessed across countries and are they comparable (e.g., do they represent diverse patients of socioeconomic backgrounds). - The authors noted that elevated risk for a psychiatric diagnosis the first 6 months following the infection. How do they know if the diagnosis is new versus continued? And those who had a diagnosis before and suddenly had exacerbation of symptoms? - The authors should consider completing separate analyses for those who had the infection during the peak of the pandemic (i.e., during social distancing and quarantine practices) as compared to towards the end or post-pandemic. I question if the initial elevated risk for psychiatric disorders are compounded by pandemic-related stressors such as reduced social supports, anxiety about pandemic-related disruptions to school/work/social life, etc. - What was the rational for the age cohorts (e.g. 18-24 years, 25-44 years, 45-64 years)? - Are sleep disorders defined as behavioral sleep disruptions (e.g. insomnia, hypersomnia, etc.) or sleep breathing disorders? - Neurodevelopmental disorders is a very broad group and is not very meaningful when authors collapse this all into one (Intellectual disability/IDD, autism and ADHD are all considered a developmental disorder but have different implications on functioning). Authors should strongly consider separating this into more common developmental disorders e.g., ADHD, IDD, Specific Learning Disability as it is much more likely some disorders are more evident with worsened symptoms (e.g., ADHD) whereas others neuropsychologist would not expect significant change following COVID (e.g. IDD). -Authors should include information on the cohort’s preexisting diagnoses (i.e., diagnoses pre- COVID infection) if they have the data, and if not, provide more in depth discussion about this.	BMC	12916_2025_3952_ReviewerReport_V0_R3.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 1, "materials_and_methods": 5, "praise": 0, "presentation_and_reporting": 2, "results_and_discussion": 1, "suggestion_and_solution": 4, "total": 15 }	{ "criticism": 0.06666666666666667, "example": 0, "importance_and_relevance": 0.06666666666666667, "materials_and_methods": 0.3333333333333333, "praise": 0, "presentation_and_reporting": 0.13333333333333333, "results_and_discussion": 0.06666666666666667, "suggestion_and_solution": 0.26666666666666666 }	0.933333	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study\nVersion: 0 Date: 31 Oct 2024\nReviewer's report:\nAuthors should provide an abbreviated description of the databases they accessed across\ncountries and are they comparable (e.g., do they represent diverse patients of socioeconomic\nbackgrounds).\n- The authors noted that elevated risk for a psychiatric diagnosis the first 6 months following\nthe infection. How do they know if the diagnosis is new versus continued? And those who had\na diagnosis before and suddenly had exacerbation of symptoms?\n- The authors should consider completing separate analyses for those who had the infection\nduring the peak of the pandemic (i.e., during social distancing and quarantine practices) as\ncompared to towards the end or post-pandemic. I question if the initial elevated risk for\npsychiatric disorders are compounded by pandemic-related stressors such as reduced social\nsupports, anxiety about pandemic-related disruptions to school/work/social life, etc.\n- What was the rational for the age cohorts (e.g. 18-24 years, 25-44 years, 45-64 years)?\n- Are sleep disorders defined as behavioral sleep disruptions (e.g. insomnia, hypersomnia, etc.)\nor sleep breathing disorders?\n- Neurodevelopmental disorders is a very broad group and is not very meaningful when authors\ncollapse this all into one (Intellectual disability/IDD, autism and ADHD are all considered a\ndevelopmental disorder but have different implications on functioning). Authors should strongly\nconsider separating this into more common developmental disorders e.g., ADHD, IDD, Specific\nLearning Disability as it is much more likely some disorders are more evident with worsened\nsymptoms (e.g., ADHD) whereas others neuropsychologist would not expect significant change\nfollowing COVID (e.g. IDD).\n-Authors should include information on the cohort’s preexisting diagnoses (i.e., diagnoses pre-\nCOVID infection) if they have the data, and if not, provide more in depth discussion about this.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03952-z	Reviewed - bmc_medicine	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 1 Date: 11 Dec 2024 Reviewer's report: I thank the authors for thoroughly addressing my comments. I am satisfied with their answers and revisions.	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 1 Date: 11 Dec 2024 Reviewer's report: I thank the authors for thoroughly addressing my comments. I am satisfied with their answers and revisions.	BMC	12916_2025_3952_ReviewerReport_V1_R1.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 2, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 2 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 1, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0 }	1	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study\nVersion: 1 Date: 11 Dec 2024\nReviewer's report:\nI thank the authors for thoroughly addressing my comments. I am satisfied with their answers\nand revisions.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03952-z	Reviewed - bmc_medicine	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 1 Date: 13 Dec 2024 Reviewer's report: I have no further comments.	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 1 Date: 13 Dec 2024 Reviewer's report: I have no further comments.	BMC	12916_2025_3952_ReviewerReport_V1_R2.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0, "total": 1 }	{ "criticism": 1, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0, "praise": 0, "presentation_and_reporting": 0, "results_and_discussion": 0, "suggestion_and_solution": 0 }	1	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study\nVersion: 1 Date: 13 Dec 2024\nReviewer's report:\nI have no further comments.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03952-z	Reviewed - bmc_medicine	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 1 Date: 08 Jan 2025 Reviewer's report: The manuscript “Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study” outlines a study that includes data from electronic medical records across countries to examine risk of psychiatric diagnoses including depression, anxiety disorder, and alcohol abuse/dependence, in addition to others such as sleep disorders, dementia and neurodevelopmental disorders within the first 6 months 6-12 months or 12-24 months post infection. Overall, I believe the authors addressed my concerns to their best ability by expanding their discussion/limitations sections or clarifying with appropriate references/supplemental information. I do not have further comments at this time.	Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study Version: 1 Date: 08 Jan 2025 Reviewer's report: The manuscript “Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A multinational cohort study” outlines a study that includes data from electronic medical records across countries to examine risk of psychiatric diagnoses including depression, anxiety disorder, and alcohol abuse/dependence, in addition to others such as sleep disorders, dementia and neurodevelopmental disorders within the first 6 months 6-12 months or 12-24 months post infection. Overall, I believe the authors addressed my concerns to their best ability by expanding their discussion/limitations sections or clarifying with appropriate references/supplemental information. I do not have further comments at this time.	BMC	12916_2025_3952_ReviewerReport_V1_R3.pdf	{ "criticism": 1, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 1, "praise": 1, "presentation_and_reporting": 1, "results_and_discussion": 1, "suggestion_and_solution": 1, "total": 3 }	{ "criticism": 0.3333333333333333, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0.3333333333333333, "praise": 0.3333333333333333, "presentation_and_reporting": 0.3333333333333333, "results_and_discussion": 0.3333333333333333, "suggestion_and_solution": 0.3333333333333333 }	2	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study\nVersion: 1 Date: 08 Jan 2025\nReviewer's report:\nThe manuscript “Psychiatric and neuropsychiatric sequelae of COVID-19 within two years: A\nmultinational cohort study” outlines a study that includes data from electronic medical records\nacross countries to examine risk of psychiatric diagnoses including depression, anxiety disorder,\nand alcohol abuse/dependence, in addition to others such as sleep disorders, dementia and\nneurodevelopmental disorders within the first 6 months 6-12 months or 12-24 months post\ninfection. Overall, I believe the authors addressed my concerns to their best ability by expanding\ntheir discussion/limitations sections or clarifying with appropriate references/supplemental\ninformation. I do not have further comments at this time.", "role": "assistant" } ]	2,025
Review for 10.1186/s12916-025-03958-7	Reviewed - bmc_medicine	Title: Effect of a Single-dose Denosumab on Mineral Homeostasis in Infertile Men: Insights from a Pilot Intervention Study and a Randomized Controlled Trial Version: 0 Date: 08 Oct 2024 Reviewer: Jan Št#pán Reviewer's report: Previously, the authors reported that a single injection of denosumab may be used to improve male fertility potential in a sub-population of infertile men. This study provides evidence that a single injection of denosumab in infertile men has no major persistent impact on bone or mineral homeostasis. The hypothesis to be tested is well defined. The methods, data and analysis are described in sufficient detail to be reproduced, ant they adequately test the hypothesis and support the conclusions. Appropriate figures have been provided. The information provided by the paper is clinically relevant. Minor comment Please, consider mentioning absence of other biomarkers in the limitations of the study. A more reliable marker of osteoblast activity, such as PINP was not provided. Total serum alkaline phosphatase was measured, but is it known whether there any effects of denosumab on the liver isoform? Were the extended effects of denosumab on PTH accompanied by changes in FGF23 and 1,25(OH)2D3? Serum betaCTX could provide evidence on suppression of osteoclastic bone resorption.	Title: Effect of a Single-dose Denosumab on Mineral Homeostasis in Infertile Men: Insights from a Pilot Intervention Study and a Randomized Controlled Trial Version: 0 Date: 08 Oct 2024 Reviewer: Jan Št#pán Reviewer's report: Previously, the authors reported that a single injection of denosumab may be used to improve male fertility potential in a sub-population of infertile men. This study provides evidence that a single injection of denosumab in infertile men has no major persistent impact on bone or mineral homeostasis. The hypothesis to be tested is well defined. The methods, data and analysis are described in sufficient detail to be reproduced, ant they adequately test the hypothesis and support the conclusions. Appropriate figures have been provided. The information provided by the paper is clinically relevant. Minor comment Please, consider mentioning absence of other biomarkers in the limitations of the study. A more reliable marker of osteoblast activity, such as PINP was not provided. Total serum alkaline phosphatase was measured, but is it known whether there any effects of denosumab on the liver isoform? Were the extended effects of denosumab on PTH accompanied by changes in FGF23 and 1,25(OH)2D3? Serum betaCTX could provide evidence on suppression of osteoclastic bone resorption.	BMC	12916_2025_3958_ReviewerReport_V0_R1.pdf	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 6, "praise": 4, "presentation_and_reporting": 4, "results_and_discussion": 1, "suggestion_and_solution": 2, "total": 11 }	{ "criticism": 0, "example": 0, "importance_and_relevance": 0, "materials_and_methods": 0.5454545454545454, "praise": 0.36363636363636365, "presentation_and_reporting": 0.36363636363636365, "results_and_discussion": 0.09090909090909091, "suggestion_and_solution": 0.18181818181818182 }	1.545455	[ { "content": "Please review this paper.", "role": "user" }, { "content": "Title: Effect of a Single-dose Denosumab on Mineral Homeostasis in Infertile Men: Insights\nfrom a Pilot Intervention Study and a Randomized Controlled Trial\nVersion: 0 Date: 08 Oct 2024\nReviewer: Jan Št#pán\nReviewer's report:\nPreviously, the authors reported that a single injection of denosumab may be used to improve\nmale fertility potential in a sub-population of infertile men. This study provides evidence that\na single injection of denosumab in infertile men has no major persistent impact on bone or\nmineral homeostasis. The hypothesis to be tested is well defined. The methods, data and analysis\nare described in sufficient detail to be reproduced, ant they adequately test the hypothesis and\nsupport the conclusions. Appropriate figures have been provided. The information provided by\nthe paper is clinically relevant.\nMinor comment\nPlease, consider mentioning absence of other biomarkers in the limitations of the study. A more\nreliable marker of osteoblast activity, such as PINP was not provided. Total serum alkaline\nphosphatase was measured, but is it known whether there any effects of denosumab on the liver\nisoform? Were the extended effects of denosumab on PTH accompanied by changes in FGF23\nand 1,25(OH)2D3? Serum betaCTX could provide evidence on suppression of osteoclastic bone\nresorption.", "role": "assistant" } ]	2,025

End of preview. Expand in Data Studio

BMC Medicine Peer Reviews

This dataset contains peer reviews from BMC, standardized to match the format of the pawin205/PeerRT dataset.

Dataset Structure

Each record contains the following attributes:

relative_rank: Default value (0).
win_prob: Default value (0.0).
title: Title of the paper.
abstract: Abstract of the paper.
full_text: Full text of the paper (or review text if unavailable).
review: The peer review text.
source: Source of the data ('BMC').
review_src: Original filename of the review.
criteria_count: Dictionary containing counts of specific criteria in the review (e.g., criticism, praise).
dimension_scores: Dictionary containing scores for various dimensions.
overall_score: Computed overall score (reward value).
thinking_trace: Placeholder (empty).
prompt: Placeholder (empty).
prompt_length: Placeholder (0).
conversations: List of dictionaries representing the conversation (User/Assistant format).
year: Year of publication (if available).

Usage

from datasets import load_dataset

dataset = load_dataset("JerMa88/BMC_Medicine_Peer_Reviews")
print(dataset['train'][0])

Downloads last month: 28

: (See dataset viewer)

Size of downloaded dataset files:

83.5 MB

Size of the auto-converted Parquet files:

83.5 MB

Number of rows:

19,492