Oral Diseases and Adverse Pregnancy Outcomes in Sub-Saharan Africa: A Scoping Review

Round 1

Reviewer 1 Report

The authors have reported a scoping review on the links between oral health status and adverse pregnancy outcomes in a sub-Saharan population. The study is important in its scope, but has the following concerns:

1.       Kappa value for study screening needs to be provided.

2.       The inclusion criteria does not adequately describe or capture the study aims. For example, if only the oral health status of pregnant women are included, then for example a cross sectional study assessing oral health outcomes of pregnant women may qualify that does not address the study aims at all.

3.       Figure 1 is cut off.

4.       The authors make a statement in section 3.9 advocating for screening with BANA. This technique has been associated with unreliable outcomes for diagnosis although it is excellent for detecting pathogens and has been out of clinical use for sometime. Hence, this recommendation does not follow standard of care procedures.

5.       Second 3.9 Line 313 – please define “Low periodontitis”

6.       The gaps in knowledge section (3.9) expands greatly on the scope of this study and as such the search strategy couldn’t have identified such studies since the scope was narrow. I would suggest narrowing its scope to match the scope of the review.

7.       Although table 2 is great from a practical point of view to associate conditions with exposures, the actual odds ratio, relative risk or other parameters have not been discussed anywhere in the manuscript. That information is extremely critical in a study evaluating the risks of a certain exposure to a condition as for example a statistically significant odds ratio of 1.8 is significantly different from a odds ratio of 3.

8.       More comprehensive references are needed for lines 387-398

Moderate editing of English language is required.

Author Response

The authors have reported a scoping review on the links between oral health status and adverse pregnancy outcomes in a sub-Saharan population. The study is important in its scope, but has the following concerns:

1. Kappa value for study screening needs to be provided.

The kappa value for inter-observer reliability was 0.70, inserted in lines 135-137

2. The inclusion criteria does not adequately describe or capture the study aims. For example, if only the oral health status of pregnant women are included, then for example a cross sectional study assessing oral health outcomes of pregnant women may qualify that does not address the study aims at all.

Many thanks for the observation on the inclusion criteria as written and its implications for the reproducibility of the study. It has been corrected “as studies that assessed oral disease status in pregnant women with reports of adverse pregnancy outcomes”. The PEO framework and the search strategy showed that studies were included based on linking the defined population, the exposure and the outcomes, lines 100-125.

3. Figure 1 is cut off.

Figure 1 has been revised and is now presented in full.

4. The authors make a statement in section 3.9 advocating for screening with BANA. This technique has been associated with unreliable outcomes for diagnosis although it is excellent for detecting pathogens and has been out of clinical use for sometime. Hence, this recommendation does not follow standard of care procedures.

We posit based on available literature that BANA may be useful for periodontal disease screening, subject to further studies in lines 449-460. Your caution on its unreliability is noted even as our literature search up till 2015 remained favorable to the use of BANA. While we could not find literature that recommended it was no longer standard of care for periodontal disease screening, based on your caution, the statement proposing its use has been removed and further studies on its usefulness advocated lines 449-460.

5. Second 3.9 Line 313 – please define “Low periodontitis”

The standard has been clarified to “low prevalence of periodontal disease” in line 321.

6. The gaps in knowledge section (3.9) expands greatly on the scope of this study and as such the search strategy couldn’t have identified such studies since the scope was narrow. I would suggest narrowing its scope to match the scope of the review.

The gaps in knowledge section (3.9) has been narrowed to the scope of this study, lines 312-325.

7. Although table 2 is great from a practical point of view to associate conditions with exposures, the actual odds ratio, relative risk or other parameters have not been discussed anywhere in the manuscript. That information is extremely critical in a study evaluating the risks of a certain exposure to a condition as for example a statistically significant odds ratio of 1.8 is significantly different from a odds ratio of 3.

Table 2 has been modified by deleting some columns. Your observation on the need to statistically show the level of significant association between dental conditions and adverse pregnancy outcomes is appreciated with thanks. We acknowledge that parameters such as odds ratio, relative risk and others strengthen the argument about a relationship. However, this scoping review was aimed at exploring the available literature from sub-Saharan Africa on the relationship as stated in lines 82-87, hence not pooling the outcome measures, statistical analysis to strengthen association is a limitation of the study, this was reflected in lines 482-484.

8. More comprehensive references are needed for lines 387-398

Lines 387-398 have been rewritten with appropriate references cited [28,40-43] lines 420-445.

Comments on the Quality of English Language

Moderate editing of English language is required.

Moderate editing of English language was carried out. The Grammarly software scored the edited paper 91.

Reviewer 2 Report

I reviewed the manuscript "Oral Health Status and Adverse Pregnancy Outcomes in Sub-Saharan Africa: A Scoping Review", whose stated purpose by the Authors was to explore the oral health status and adverse pregnancy outcomes.

The concept of oral health has been well defined in international scientific literature and does not coincide with the presence of caries, periodontitis or gingivitis. The authors evaluated the relationship between oral disease status and Adverse Pregnagncy Outcomes.

The outcomes they set out to assess were never defined before they began their research. In fact, the outcomes evaluated are those that evaluated the included studies.

Starting with the purpose, the concepts expressed are confusing and improper terms are often used. The reader has to get to the end of the results to understand which outcomes have been chosen, and discover that they are only the outcomes of the included studies and not primarily defined.

I would suggest that the Authors intensively rework all sections of the manuscript and, above all, define the research methodology beforehand and use the right terms in the right way.

Some comments for the various sections follow:

AFFILIATIONS:

- Affiliations 3, 6 and 9 should coincide. So all are denoted by the number 3 and the others are renumbered.

KEYWORDS: use MeSH terms.

ABSTRACT: The sentence on line 45 confirms the presence of an association, while the next sentence says that the results are inconclusive. There is a contradiction that confuses the reader, is the association present or not? Rephrase the concept you wanted to express.

M&M

- The definition of Outcomes cannot be 'pregnancy outcomes', one must primarily define Outcomes in detail.

Furthermore, I repeat my doubt. There is no mention of oral health in the article and one must, again, define what one is talking about in detail first. In fact, in the search string, the words gingivitis, periodontitis, caries were entered. If there are these conditions, there is no mention of oral health.

- The questions the authors asked themselves before starting the review predate the definition of PEO. The order of the M&Ms is incorrect. The definition of inclusion and exclusion criteria must also precede the selection of studies (sections 2.4 and 2.5).

- Who conducted the literature search? (section 2.2)

-  Leave only the search strings in the appendix or move them to the text in section 2.2. The rest of the information is already in the text and is therefore redundant in the appendix.

- Line 127: who was the third researcher?

- Line 131: who were the two authors?

RESULTS: 

- The flow chart does not read completely. Furthermore, there are errors in the compilation: the 626 excluded articles should have been entered in the second box below the one used. The box used is for not eligible records (not for excluded, as was used). Furthermore, the number of excluded articles for each reason must be indicated, not just the total number of excluded articles.

- Line 149: how many records were not eligible?

- Line 152: multiple citations in square brackets do not require a space between each number. Furthermore, subsequent references, e.g. 16,17,18,19, are indicated as 16-19. Read the journal guidelines for references and correct them in all manuscript sections.

- Why only report the type of studies included for 10 of the 14 articles in the flow chart?

- Line 191 'The' lowercase letter.

- Line 209: Figure 1 is flow chart. Recheck the numbering of the pictures and tables and correct the numbers in the text.

- Line 220: DMFT is an acronym never before defined.

- I reiterate the notion that all outcomes to be evaluated should have been defined before the research. Moreover, they should at least be mentioned, or better defined, in the introduction in order to prepare the reader (section 3.4).

- Line 263: what kind of intervention is it?

- Table 1 is very confusing and difficult to read. It should be reformatted, perhaps by merging the first columns that have short information, to make it easier visually.

- What does BANA consist of? The reader should possibly be prepared for it in the introduction, where all the concepts that will be covered should be explained (again, such as oral health/pathology and pregnancy outcomes in general).

- There is no SEVERITY of periodontal disease classified as absent, mild, moderate and severe! 

The 2017 World Workshop clearly defined the concept of severity of periodontal disease in its consensus report. The authors use this term improperly in several places in the manuscript. If any old article included defined the concept of severity in this way, it should be excluded or only the objective parameters to which it referred should be inserted. For this reason, it is important to have a clear conception and define in the text all the outcomes, exposures, and population to be investigated. Precisely, not generalized. 

- Table 1: no description was inserted. 

AUTHORS CONTRIBUTION: The role of only 5 out of 10 authors is specified. Did the other authors only contribute by reading the manuscript?

REFERENCES (SECTION): All references must be reformatted: check the journal guidelines. Also, remove the links and insert only the DOI number, not the link.

The text needs to be extensively revised. Increased focus on methodology and terminological accuracy.

Many sentences do not follow the normal English construction. Many terms are typical of spoken English. There are gaps in scientific terminology.

Author Response

I reviewed the manuscript "Oral Health Status and Adverse Pregnancy Outcomes in Sub-Saharan Africa: A Scoping Review", whose stated purpose by the Authors was to explore the oral health status and adverse pregnancy outcomes.

The concept of oral health has been well defined in international scientific literature and does not coincide with the presence of caries, periodontitis or gingivitis. The authors evaluated the relationship between oral disease status and Adverse Pregnagncy Outcomes.

Many thanks for the conceptual clarification of the term oral health status. The title has been appropriately changed to “Oral disease status and adverse pregnancy outcomes: a scoping review”

The outcomes they set out to assess were never defined before they began their research. In fact, the outcomes evaluated are those that evaluated the included studies.

Starting with the purpose, the concepts expressed are confusing and improper terms are often used. The reader has to get to the end of the results to understand which outcomes have been chosen, and discover that they are only the outcomes of the included studies and not primarily defined.

Outcomes to be utilized for determining eligible literature are now pre-specified in Material and Methods section 2.2. Some general outcomes were described in the Introduction but our scoping review was to identify the characteristics of available studies and identify the outcomes described in these studies from sub-Saharan Africa from lines 82-87, hence they could only be well described after the Results are available. We did not choose which outcomes to describe, rather in line with the aim and objectives, we reported the outcomes as described in the included literature based on the definitions of the various authors.

I would suggest that the Authors intensively rework all sections of the manuscript and, above all, define the research methodology beforehand and use the right terms in the right way.

Your suggestion is warmly appreciated. All sections of the manuscript starting with the title have been intensively reworked. The research methodology has been refined using the right terms the right way, lines 90-155.

Some comments for the various sections follow:

AFFILIATIONS:

- Affiliations 3, 6 and 9 should coincide. So all are denoted by the number 3 and the others are renumbered.

Many thanks for this observation. Affiliations 3,6,9 have been corrected so that affiliations 1,4, and 5 are shared by the affected authors.

KEYWORDS: use MeSH terms.

MeSH terms such as oral, periodontitis, pregnancy, outcomes, caries, dental are noew used in the Keywords.

ABSTRACT: The sentence on line 45 confirms the presence of an association, while the next sentence says that the results are inconclusive. There is a contradiction that confuses the reader, is the association present or not? Rephrase the concept you wanted to express.

The confusion caused by the affirmation of an association that was contradicted in the next sentence is regretted. The association found in the included studies is expressed clearly in lines 36-38 of the Abstract.

M&M

- The definition of Outcomes cannot be 'pregnancy outcomes', one must primarily define Outcomes in detail.

The outcomes have been defined to adverse pregnancy outcomes associated with maternal oral disease conditions lines 111-112.

Furthermore, I repeat my doubt. There is no mention of oral health in the article and one must, again, define what one is talking about in detail first. In fact, in the search string, the words gingivitis, periodontitis, caries were entered. If there are these conditions, there is no mention of oral health.

Many thanks for this observation. The concept of oral health has been substituted correctly with oral diseases which explains why gingivitis, periodontitis and dental caries are evaluated.

- The questions the authors asked themselves before starting the review predate the definition of PEO. The order of the M&Ms is incorrect. The definition of inclusion and exclusion criteria must also precede the selection of studies (sections 2.4 and 2.5).

Your observation is appreciated. The definition of PEO now comes in section 2.2 after the research questions. The order of M and Ms has been corrected and the definition of inclusion and exclusion criteria is now in section 2.3 and now precedes the selection of studies.

- Who conducted the literature search? (section 2.2)

The literature search in section 2.2 is now part of search strategy in section 2.3. It was conducted by two of the researchers identified using the acronyms ETA and ORA, lines 120-121.

-  Leave only the search strings in the appendix or move them to the text in section 2.2. The rest of the information is already in the text and is therefore redundant in the appendix.

The search strings mentioned in the text from the appendix are retained in lines 122-123 and the appendix deleted.

- Line 127: who was the third researcher?

The third researcher is now identified as one of the authors with the acronym MOF in line 134.

- Line 131: who were the two authors?

Section 2.6, Line 131 is now section 2.5, line 141 with the two authors identified using the acronyms ETA and ORA.

RESULTS: 

- The flow chart does not read completely. Furthermore, there are errors in the compilation: the 626 excluded articles should have been entered in the second box below the one used. The box used is for not eligible records (not for excluded, as was used). Furthermore, the number of excluded articles for each reason must be indicated, not just the total number of excluded articles.

Our apology for incomplete flow chart. This has been corrected and the table clearly stated. Errors in the compilation are also acknowledged and have now been corrected. The 626 excluded articles are made up of 343 screened out after review of titles and abstracts (line 183) and 283 excluded as full articles for reasons stated in lines 185-191.

 - Line 149: how many records were not eligible?

After the screening using titles and abstracts, 297 titles were found eligible for retrieval as full articles, (lines 185-188).

- Line 152: multiple citations in square brackets do not require a space between each number. Furthermore, subsequent references, e.g. 16,17,18,19, are indicated as 16-19. Read the journal guidelines for references and correct them in all manuscript sections.

The correction of multiple citations is acknowledged and implemented according to the journal guidelines.

- Why only report the type of studies included for 10 of the 14 articles in the flow chart?

Lines 164-167 show that six articles were cross-sectional studies, four were case-control studies, one was cohort study and three were clinical trials to make up 14 articles from the flow chart.

- Line 191 'The' lowercase letter.

The lowercase letter for the word ‘The’ has been changed to ‘the’ in line 196.

- Line 209: Figure 1 is flow chart. Recheck the numbering of the pictures and tables and correct the numbers in the text.

The next figure after figure 1 has been correctly titled figure 2 in lines 218-220.

- Line 220: DMFT is an acronym never before defined.

The acronym DMFT has now been defined as Decayed, Missing and Filled Teeth now line 225.

- I reiterate the notion that all outcomes to be evaluated should have been defined before the research. Moreover, they should at least be mentioned, or better defined, in the introduction in order to prepare the reader (section 3.4).

Some of the outcomes like the oral conditions of periodontitis, gingivitis, tooth mobility ere mentioned in the Introduction, lines 33-55, while those pertaining to adverse pregnancy outcomes like premature delivery, fluctuating birth weights and preeclampsia were also mentioned in lines 58-59 to prepare the reader for further definition in the Materials and Methods in lines 112-118 and the Results.

- Line 263: what kind of intervention is it?

Section 3.7 gives further details of the intervention as made up ot two groups. The test group had scaling and polishing done as treatment for periodontal disease while the control group had the procedure done after delivery or had oral hygiene instruction before delivery in lines 266-269.

- Table 1 is very confusing and difficult to read. It should be reformatted, perhaps by merging the first columns that have short information, to make it easier visually.

Your comments on the confusion caused by the difficulty in reading table 1 was well taken. Some columns containing short information have been merged. The information in the table especially under study outcomes have also been made more succinct to eliminate confusion and increase its visual appeal.

- What does BANA consist of? The reader should possibly be prepared for it in the introduction, where all the concepts that will be covered should be explained (again, such as oral health/pathology and pregnancy outcomes in general).

N-benzoyl-DL-arginine-2-napththylamide (BANA) is a chemical point of care test proposed for the identification of periodontopathogens by some workers cited in this review. it was not prepared for in the Introduction as this scoping review was exploratory in nature. A brief description of its use is in lines 275-280.

- There is no SEVERITY of periodontal disease classified as absent, mild, moderate and severe! 

The 2017 World Workshop clearly defined the concept of severity of periodontal disease in its consensus report. The authors use this term improperly in several places in the manuscript. If any old article included defined the concept of severity in this way, it should be excluded or only the objective parameters to which it referred should be inserted. For this reason, it is important to have a clear conception and define in the text all the outcomes, exposures, and population to be investigated. Precisely, not generalized. 

Your observations are acknowledged with thanks. However, one of our objectives was to present how articles from sub-Saharan Africa reported oral diseases associated with adverse pregnancy outcomes. authors in the included studies classified oral diseases in different ways, one of which was of periodontal severity as absent, mild, moderate and severe. Excluding to report an objective modality of reporting oral diseases because the paper is old or for any other reason is a form of selection bias and would be contrary to the aim and objectives of this scoping review. Having presented our findings, we observed the different parameters used for examining and reporting oral diseases in sub-Saharan Africa and the need for standardization, lines 389-393.

Note was taken of the 2017 consensus classification of periodontal and peri-implant disease in lines 389-393, but these are more applicable at individual patient level while the WHO Manuals for Basic Oral Surveys also cited are for population-based oral surveys.

- Table 1: no description was inserted. 

The description of table was “Characteristics of the included studies”, it is written in line 283.

AUTHORS CONTRIBUTION: The role of only 5 out of 10 authors is specified. Did the other authors only contribute by reading the manuscript?

All authors conceptualized the study, ETA, ORA, IEA, OAS, ME-T and MOF participated and finalized the study protocol while ETA, ORA and MOF undertook the literature search and selection of studies. Data analysis was carried out by ETA and MOF while ETA, IEA, and MOF wrote the results and analyzed the data. All authors made intellectual contributions, read and approved the published version of the manuscript.

REFERENCES (SECTION): All references must be reformatted: check the journal guidelines. Also, remove the links and insert only the DOI number, not the link.

References have been reformatted to conform with the guidelines of this journal. Links are retained for references that have no doi as they are websites.

The text needs to be extensively revised. Increased focus on methodology and terminological accuracy.

The text has been extensively revised with focus on the methodology and terminology.

Comments on the Quality of English Language

Many sentences do not follow the normal English construction. Many terms are typical of spoken English. There are gaps in scientific terminology.

The sentences have been reconstructed to suit written rather than spoken English language. The Grammarly software was also used to edit the paper.

Round 2

Reviewer 1 Report

I would like to thank the authors for the edits. Please find attached below additional comments based on the edits. 

1. Please clarify this line: The two researchers (ETA and ORA) had earlier been 164 tested using similar literature, with an inter-observer reliability of 0.70 indicating substantial agreement. Is the 0.7 the actual value for the screening or some other pilot testing that was carried out?

2. The search strategy does not appear to be well defined. For example, if the word Sub-Saharan population was used, other iterations of the same word would not be picked up by the search. Additionally, only periodontitis and caries appears to be included. Why not other oral conditions? A generic term oral or oral diseases does not appear to be all inclusive. In general, it appears that the search strategy is likely not comprehensive and sensitive enough.

3. In response to this comment from the authors:

Table 2 has been modified by deleting some columns. Your observation on the need to statistically show the level of significant association between dental conditions and adverse pregnancy outcomes is appreciated with thanks. We acknowledge that parameters such as odds ratio, relative risk and others strengthen the argument about a relationship. However, this scoping review was aimed at exploring the available literature from sub-Saharan Africa on the relationship as stated in lines 82-87, hence not pooling the outcome measures, statistical analysis to strengthen association is a limitation of the study, this was reflected in lines 482-484.

If the authors do not wish to add the strengths (odds ratio etc) of association, then it is imperative that the sections 4.3, conclusion and abstracts be re-written. If the true aim was to perform a scoping review without analyzing or commenting on the strengths of such association, making conclusive statements about associations appears to be outside the scope of the paper as has been made in the sections 4.3 and conclusion. In such a case the authors will need to re-write these sections to further narrow the scope of their paper due to the lack of available statistics (even if only repeated from representative papers).

For future iterations of edits, please include a clean version of the manuscript along with a tracked changes version.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

I congratulate the Authors for having signicantly improved the manuscript in all its sections and for having accepted the suggestions with scientific maturity, as well as clarifying the doubts.

Author Response

Round 3

Reviewer 1 Report

I would like to thank the authors for the edits and response to comments. 

In response to the comment 1: The authors need to provide the kappa or inter-rater reliability for all the screening each stage individually and not just the pilot studies.

In response to comment 2: the full search string is usually required to be present for all databases individually. As it is presented now, it is unclear for what database and if what boolean operators were included to combine the terms. More details are required to know if the search was adequately conducted and for repeatability of the search string.

Additionally, the authors state a search was carried out in Medline, PubMed, Cochrane Library, Google Scholar, and African Journals Online but only Pubmed and Google scholar data is presented in Figure 1. Were the other databases not searched?

The authors also state that both Medline and Pubmed were searched independently - please comment how this was achieved.

In response to comment 3 and the amended discussion: Several points in the new discussion are lacking references;

Some statements such as "First, we observed that there were very few published studies in English on the link between oral disease conditions and adverse pregnancy events" are not substantiated by literature - significant number of studies, reviews and systematic reviews exist on this topic and the study also managed to identify 15 papers in the Sub-Saharan region. Not sure what the authors imply by the statement.

English language appears adequate.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf