Sociodemographic Factors and Determinants of Mental Health in the African American Population A Cross-Sectional Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

• INTRODUCTION

The introduction is well contextualized, presenting relevant historical facts and a consistent theoretical framework, which adequately situates the reader regarding the research problem.

It is suggested to revise the line spacing in lines 49 and 50.

The text presents a clear justification for conducting the study, particularly by highlighting the importance of local research for a better understanding of empirically observed problems, contributing to filling gaps in the existing literature.

 

• OBJECTIVES

The objectives are clearly formulated and coherent with the study proposal.

 

• HYPOTHESES

The hypotheses are explicitly presented, well defined, and aligned with both the objectives and the guiding research question, which strengthens the methodological consistency of the study.

 

• MATERIALS AND METHODS

The study reports the recruitment of 557 individuals, aiming to ensure representativeness by age and sex according to the local demographic profile. However, more detailed information on the sample size calculation is lacking, particularly regarding the parameters used (sampling error, confidence level, estimated prevalence), considering a total population of 5,118 inhabitants. It is recommended to clarify how this number was determined.

The data collection instruments are adequately described, well detailed, and grounded in validated literature, which confers methodological robustness to the study.

 

• RESULTS

The section presents a broad and detailed description of the sociodemographic profile, risk factors, and prevalence of mental disorders in the studied population.

The tables are well organized and allow clear visualization of frequencies, prevalences, and associations found.

In some sections, there is excessive interpretation of the findings. It is advisable to move such interpretations to the Discussion section.

Standardization of p-value presentation is recommended.

Some tables may generate interpretative confusion regarding percentages and multiple comparisons.

The inclusion of effect measures such as odds ratios or confidence intervals is recommended.

The high overall prevalence of mental disorders (60.7%) should be highlighted early in the section.

 

• DISCUSSION

The discussion is theoretically well grounded and articulates empirical findings with the literature.

Historical and social contextualization strengthens the analysis.

The relationship between stress, abuse/maltreatment, and mental disorders is consistently discussed.

Greater synthesis and a subsection on study limitations are recommended.

Implications for public policies and primary mental health care should be explicitly addressed.

 

• CONCLUSION

The conclusions are aligned with the study objectives and synthesize the main findings.

- Recommendations include strengthening primary health care actions, intersectoral strategies, and future longitudinal studies.

Author Response

Comments 1: Suggested reviewing line spacing on lines 49 and 50.

Response: Line spacing on lines 49 and 50 has been reviewed.

Comments 2: Lacks more detailed information on sample size calculation, particularly regarding parameters used (sampling error, confidence level, estimated prevalence), considering a total population of 5,118 inhabitants. Recommend clarifying how this number was determined.

Response: The reviewer's suggestion has been adopted by providing greater detail on sample size calculation and parameters used. Highlighted: Page 3, lines 127 to 136.

Comments 3: Some sections contain excessive interpretation of findings; advise moving such interpretations to the Discussion section.

Response: Thank you for this observation. Excessive interpretations have been reduced, compared with Discussion information, eliminating redundancy in results or adding relevant interpretations there. These changes are highlighted at the end of each table.

Comments 4: Recommend standardizing presentation of p-values. Some tables may cause interpretive confusion regarding percentages and multiple comparisons.

Response: Tables have been reviewed and p-value presentation standardized. Thank you for your observation.

Comments 5: Recommend including effect measures such as odds ratios or confidence intervals.

Response: The reviewer's suggestion has been adopted, adding the requested statistical data, which can be seen in Tables 4 and 6.

Comments 6: The high overall prevalence of mental disorders (60.7%) should be highlighted at the beginning of the section.

Response: This result has been highlighted at the beginning of the Results section, as suggested by the Reviewer.

Comments 7: In the Discussion: Public policy implications and primary mental health care must be explicitly addressed.

Response: Implications suggested by Reviewer 3 have been added at the end of the Discussion (Page 12, lines 382-387), along with limitations (lines 389-393).

Reviewer 2 Report

Comments and Suggestions for Authors

The topic is important and timely: estimating psychiatric morbidity and associated determinants in historically marginalized Afro-descendant communities in northern Ecuador has clear relevance for public health planning and equity-oriented interventions. The study also benefits from a reasonably large community sample (n = 557).

That said, substantial revisions are needed before the paper can be interpreted with confidence. My main suggestions are below.

What the manuscript does well

• Addresses the important public health equity topic of mental health burden and correlates in a historically marginalized Afro-descendant population in the Chota River basin.
• Uses recognized instruments (GMHAT/PC, WHO STEPS items, IPAQ) and reports a clear overall prevalence estimate and diagnostic distribution. 
• Provides ethical approvals and informed consent statements, and includes an AI-tool disclosure, which is transparent. 

Issues requiring revision (high priority)

1. Correct and standardize the population description throughout (critical).
• The title and parts of the Discussion repeatedly refer to an “African American population,” yet the methods and setting clearly describe Afro-descendant communities in northern Ecuador (Chota River basin). This is not a minor wording issue; it affects interpretability and indexing. Please revise title/abstract/discussion terminology to be fully consistent and accurate. 
2. Align the statistical methods with the results actually reported.
• The Methods state that logistic regression models were used to examine associations, but the Results largely present bivariate tables and Mann–Whitney U tests without providing regression outputs (e.g., adjusted odds ratios, confidence intervals, model covariates, model fit). Either (a) add the full regression results, or (b) revise the Methods to match what was done and justify the analytic approach. 
3. Repair and clarify tables (accuracy + interpretability).
   Several tables appear to have incorrect or unclear percentage calculations and labeling (e.g., Table 4 “% with mental disorder” does not appear to reflect the within-group proportion; inconsistent totals; typos like “Mal”; “NaN” entries). Please ensure each table clearly states:
• the denominator used for each percentage (within-row vs. overall sample),
• the statistical test used and why it is appropriate (including small-cell handling),
• consistent rounding and formatting,
• corrected variable labels and headings. 
4. Avoid over-interpretation from very small cell counts and cross-sectional data.
• Statements such as “100% association” for certain abuse categories are likely driven by extremely small n in those subgroups and should be framed cautiously (and ideally accompanied by exact counts, CIs, and/or a note about sparse data bias).
• Reframe causal language (“influenced by,” “leads to,” “genesis”) to association language unless supported by design. 
5. Strengthen methods reporting for sampling and measurement validity in-context.
   Please add:
• cluster sampling steps (how clusters were defined/selected; household selection; within-household selection),
• inclusion/exclusion criteria, response rate/nonresponse handling, missing data handling,
• definitions for “young/middle-aged/older adults,” stress categories, and how abuse variables were operationalized,
• instrument language and cultural/linguistic adaptation (especially for GMHAT/PC), interviewer training, and quality assurance. 

Moderate-priority improvements

• Introduction and literature positioning: tighten the logic from broad Latin America/racism literature to the specific Ecuadorian context; ensure claims are supported by references that clearly match the population and geography. 
• References: formatting is inconsistent and several entries appear incomplete or include database-export artifacts (e.g., “LID,” incomplete journal details). Please bring references into the journal’s required format and verify completeness. 
• Language editing: professional copyediting is recommended to address grammar, typos, and readability.

Comments on the Quality of English Language

The manuscript would benefit from substantive English-language editing to improve clarity, consistency, and readability. While the overall structure (Introduction/Methods/Results/Discussion) is recognizable, multiple language issues currently interfere with accurate interpretation of the study.

Key issues to address

• Inconsistent and sometimes incorrect terminology for the study population and setting. The text alternates between “African American” and Afro-descendant communities in Ecuador, which creates confusion and should be standardized throughout (title, abstract, main text, and discussion).

• Grammar and word-choice problems (e.g., incorrect articles/prepositions, awkward phrasing, and unclear sentence construction) that occasionally obscure meaning, particularly in the Introduction and Discussion.

• Typos and editing artifacts (e.g., “Mal” instead of “Male,” “p-valor,” and occasional garbled or placeholder text), suggesting the manuscript needs a careful proofread prior to resubmission.

• Table labeling and statistical notation in English are inconsistent (e.g., mixed language labels, unclear denominators, “NaN”), which should be corrected to ensure readers can interpret results without ambiguity.

Recommendation

I recommend a professional copyedit or thorough language revision by a fluent academic English speaker, with special attention to:

1. consistent terminology (population/location),

2. sentence-level clarity in the Introduction/Discussion, and

3. table/figure labels and statistical reporting conventions.

Author Response

Comments 1: Correct and standardize the population description throughout the text (critical).

Response: This correction has been made. Thank you for improving our work.

Comments 2: Align statistical methods with reported results. Methods indicate logistic regression models were used to examine associations but results mainly present bivariate tables and Mann-Whitney U tests without providing regression outputs (e.g., adjusted odds ratios, confidence intervals, model covariates, model fit). Either (a) add complete regression results, or (b) revise Methods to match what was done and justify the analytical approach.

Response: Thank you for this observation, which helped identify an important methodological inconsistency. Regression tables were removed to simplify and standardize descriptive analysis, omitting linear regressions (Page 4, line 178). Changes can also be seen in Table 6, where Mann-Whitney U test results were replaced with odds ratio tests.

Comments 3: Fix and clarify tables (precision + interpretability). Several tables appear to have incorrect or unclear percentage calculations, and labeling (e.g., Table 4 "% with mental disorder" doesn't seem to reflect proportion within group; inconsistent totals; errors like "Mal"; "NaN" entries). Ensure each table clearly states denominator used for each percentage (within row vs. total sample), statistical test used and why appropriate (including small cell handling), consistent rounding and formatting, corrected variable labels and headers.

Response: Thank you for the suggestion. The recommended changes have been implemented.

Comments 4: Avoid overinterpretation of very small cell counts and cross-sectional data. Statements like "100% association" for certain abuse categories are likely driven by extremely small n in those subgroups and should be framed cautiously (ideally accompanied by exact counts, CI, and/or note on sparse data bias). Reframe causal language ("influenced by," "leads to," "genesis") to association language unless supported by design.

Response: Overinterpretation has been changed at the foot of each results table. Thank you for suggesting this.

Comments 5: Strengthen methods reporting for sampling and measurement validity in context. Add: cluster sampling steps (how defined/selected; household selection; selection within household), inclusion/exclusion criteria, response rate/non-response handling, missing data handling, definitions for "young adults/middle-aged/elderly," stress categories, and how abuse variables were operationalized, instrument language and cultural/linguistic adaptation (especially for GMHAT/PC), interviewer training and quality control.

Response: Following recommendations from Reviewers 1 and 2, the materials and methods section has been modified, expanding inclusion/exclusion criteria. Age group definitions follow conventional Latin American classifications inspired by WHO recommendations. As GMHAT/PC is a semi-structured interview instrument, interviewers were trained to ensure respondents understood questions until obtaining clear criteria, noting responses based on interviewer perception.

Comments 6: Introduction and literature positioning: adjust logic from broad Latin America/racism literature to specific Ecuadorian context; ensure claims supported by references clearly matching population and geography.

Response: We respectfully disagree, as the topic is under investigated with scarce data, limiting available references—especially for the specific geography addressed.

Comments 7: References: format inconsistent, several entries incomplete or include database export artifacts (e.g., "LID," incomplete journal details). Bring to journal-required format and verify completeness.

Response: Thank you for the observations. These inconsistencies have been reviewed, and we have decided to submit a copy of the document for linguistic and formatting reviews to achieve publication standards.

Comments 8: Inconsistent and sometimes incorrect terminology for study population and context. Text alternates between "African American" and Ecuadorian Afro-descendant communities, creating confusion—standardize throughout (title, abstract, main text, discussion).

Response: This suggestion has been adopted, standardizing terminology homogeneously throughout the article text.

Comments 9: Manuscript would benefit from substantive English language editing for clarity, consistency, and readability. While general structure (Introduction/Methods/Results/Discussion) is recognizable, multiple language issues currently interfere with precise study interpretation. Recommend professional copy or thorough linguistic review by fluent academic English speaker, focusing on consistent terminology (population/location), sentence-level clarity in Introduction/Discussion, table/figure labeling conventions and statistical reporting. Grammar and word choice problems (e.g., incorrect articles/prepositions, awkward phrasing, unclear sentence constructions) occasionally obscure meaning, particularly in Introduction and Discussion. Typos and editing artifacts (e.g., "Mal" instead of "Male," "p-valor," scribbled text or occasional placeholders), suggesting manuscript needs careful review before resubmission. Inconsistent English table labeling and statistical notation (e.g., mixed-language labels, unclear denominators, "NaN") must be corrected for unambiguous reader interpretation of results.

Response: We welcome this suggestion. As English is not our native language, these errors occurred. Therefore, we will request the journal to perform a review to ensure readers encounter no confusion.

Reviewer 3 Report

Comments and Suggestions for Authors The presented article concerns an important issue, but at the same time it needs major changes that help to highlight its most important merits.   The title of the article does not correspond with its content since the analyses were conducted in a specific respondents' group due to my understanding not representative of the population. The introductory part of the text does not present a clear description of the aim of the analyses conducted in the study described in the paper, the introduction of the first figure corresponding to the hypotheses is not clear.  As stated above, since the majority of respondents were females it is not clear whether the group was representative for the population it was taken from. The same concern applies for other sociodemographic characteristics of the group, and needs clear and thorough description in the article. . The Authors did not provide readers with clear information how certain risk factors described in the study were identified. It needs to be addressed in the paper.  Additionally, some statements given in the text are not supported by the data, e.g. the information about correspondence between depressive and anxiety symptoms. It is not true that such correspondence is observed only in the communities with limited access to the treatment.  I recommend re-verification of the analyses and results obtained in the study since according to the data given in the table 6 Authors verified the associations between the variables whereas the description implies they analyzed the mean differences between the groups. Regression analyses would be useful for the presented data as well.   Also, the Authors should be more careful with the discussion of their results. Since there is no data considering other populations Authors can only hypothesize instead of giving unequivocal explanations of their results. What is more, the discussion part of the text needs a section devoted to the practical implications of the results obtained in the study.  The conclusions introduced in the abstract of the paper do not correspond to the main body of the text. As a result, I see great potential in your paper. The article concerns a very important topic, and definitely should be reconsidered for publication after major changes in the text and re-review.

Author Response

Comments 1: The article title does not correspond with its content, as analyses were conducted on a specific group of respondents that, to my understanding, is not representative of the population.

Response: We respectfully disagree with the interpretation that the sample is unrepresentative and that the title does not correspond. With suggestions from Reviewers 1 and 2 that were adopted, the sampling type should become clearer. We reiterate the sample size calculation: Frame population: 5,118 inhabitants from 5 Afro-descendant communities in the Río Chota river basin (all included in the sampling frame). Stage 1: Random selection of communities by lottery (primary clusters). Stage 2: Within each community, random selection of zones per official MSP division (subclusters). Sample size: n = 557 participants, calculated for expected prevalence = 30%, error = 5%, confidence level = 95% (minimum required: 323; actual sample more than sufficient). Page 3, lines 128 to 136. The title remains as it accurately reflects both the population and study objectives.

Comments 2: The introductory text section does not present a clear description of the objective of the analyses conducted in the described study.

Response: Thank you for this valuable observation: We recognize that, although the introduction provided solid historical and epidemiological context, the specific study objective was not highlighted explicitly enough at the end of the section. Changes made:

The final introduction paragraph has been rewritten (page 2, lines 84-90, text highlighted in bold) to clearly and concisely present:

"Therefore, this study had as its main objective to estimate the prevalence of mental disorders and identify their social and economic determinants in adults from Afro-descendant communities in the Río Chota river basin, Ecuador. Specifically, associations were evaluated between mental disorders diagnosed via GMHAT/PC and factors such as perceived stress, abuse experiences, socioeconomic level, and medical comorbidities, using probabilistic cluster sampling (n = 557). The results will provide local evidence to guide culturally appropriate preventive interventions."

Comments 3: As mentioned previously, since most respondents were women, it is unclear whether the group was representative of the population from which it was taken. The same concern applies to other sociodemographic characteristics of the group, requiring clear and exhaustive description in the article.

Response: Respectfully, the two-stage probabilistic cluster sampling design employed ensures population representativeness regardless of actual sociodemographic distribution. The observed female oversampling (64% vs. estimated 52%) is explained by differential availability during home visits (women available during daytime hours), which we did not anticipate since prior informal discussions (without scientific evidence) suggested women in some communities engaged more in commerce while men stayed home.

Comments 4: Authors did not provide readers with clear information on how certain risk factors described in the study were identified. This needs to be addressed in the article. Additionally, some statements in the text are not supported by the data, e.g., information on correspondence between depressive and anxious symptoms. It is not true that such correspondence is observed only in communities with limited treatment access.

Response: We appreciate these precise observations that significantly improve the manuscript's reproducibility and rigor. Changes made: (page 4, lines 142-148, highlighted text): Risk factors such as perceived stress and abuse/maltreatment experiences were obtained from the same GMHAT-PC. Questionnaire questions allowed extraction of specific data like abuse/maltreatment experiences as dichotomous variables. Other sociodemographic variables like economic income were obtained through additional questions in the semi-structured interview administered by trained surveyors.

Reviewer 3 Comment 5: I recommend re-verifying the analyses and results obtained in the study, as according to data in Table 6, Authors verified associations between variables, while the description implies, they analyzed mean differences between groups. Regression analysis would also be useful for the data presented.

Response: Thank you for this observation that helped identify an important methodological inconsistency. We confirm Table 6 presented Mann-Whitney U test results not described in the Methods section. Action taken: Following the suggestion, Table 6 has been reformulated, replacing statistical analysis with odds ratio tests.

Comments 6: Authors should be more careful in discussing their results. Since there are no data on other populations, Authors can only hypothesize rather than provide unequivocal explanations of their results.

Response: Thank you for the suggestion. The Discussion has been carefully reviewed, changing statements to present hypotheses. Pages 11 and 12.

Comments 7: Additionally, the Discussion section needs a dedicated part on the practical implications of the results obtained in the study.

Response: Implications suggested by the Reviewer have been added at the end of the Discussion. Page 12, lines 382-387.

 

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript explores the mental health of Afro-descendant communities in the Chota River basin in Ecuador, a population that is rarely the focus of empirical mental health research. The topic is clearly relevant and the authors manage to bring attention to a social reality that is often discussed in general terms but seldom documented with primary data. The sample size is adequate, the fieldwork appears carefully conducted, and the use of standardized instruments such as the GMHAT/PC, WHO STEPS, and IPAQ supports the overall credibility of the findings. The results point to a very high burden of mental disorders, especially depression and anxiety, and underline how closely psychological suffering is tied to stress, experiences of abuse, and persistent economic hardship in these communities.

At the same time, some aspects of the manuscript would benefit from further clarification. The overall prevalence of mental disorders reported is unusually high when compared with most international data, and this raises questions that should be addressed more openly in the discussion. In particular, it would be helpful to reflect more explicitly on the screening nature of the GMHAT/PC and on the possibility that the results capture psychological distress and probable disorders rather than confirmed clinical diagnoses. There is also some inconsistency in the way the study population is labeled, as the term “African American” is used alongside “Afro-descendant,” which may be misleading in a Latin American context and should be corrected for conceptual accuracy.

From an analytical point of view, although logistic regression is mentioned in the methods section, the results are presented mainly as bivariate associations. Including clearer multivariate analyses, with effect sizes and confidence intervals, would strengthen the interpretation and help distinguish the independent contribution of factors such as gender, income, stress, and exposure to violence. This would also make the conclusions more solid and easier to compare with findings from other settings.

Overall, the manuscript has clear value and addresses an important gap in the literature on mental health inequalities. With revisions that improve conceptual clarity and strengthen the statistical presentation, the study could make a meaningful contribution to international discussions on social determinants of mental health and the specific vulnerabilities of Afro-descendant populations.

Author Response

Comments 1 
“The overall prevalence of mental disorders reported is unusually high when compared with most international data, and this raises questions that should be addressed more openly in the discussion.”

Response: Thank you very much for your thoughtful and encouraging comments. We are pleased that you recognize the importance of our research topic; we have taken your suggestions into account, and we were already aware of the need to address this issue more explicitly. Therefore, we have added several sentences in the Discussion to address the unusually high prevalence. Some studies suggest that this may be related to the instrument used in the study, as it can even detect subclinical cases (page 11, lines 319–323). If you consider that a more extensive explanation is necessary, we would be happy to expand this section further.

Comments 3
“There is also some inconsistency in the way the study population is labeled, as the term ‘African American’ is used alongside ‘Afro-descendant,’ which may be misleading in a Latin American context and should be corrected for conceptual accuracy.”

Response: We have carefully standardized the terminology for the study population throughout the entire document.

Comments 3
“From an analytical point of view, although logistic regression is mentioned in the methods section, the results are presented mainly as bivariate associations. Including clearer multivariate analyses, with effect sizes and confidence intervals, would strengthen the interpretation and help distinguish the independent contribution of factors such as gender, income, stress, and exposure to violence. This would also make the conclusions more solid and easier to compare with findings from other settings.”

Response: We thank you for this valuable observation, which will undoubtedly improve the quality of our work. In Table 4, we have expanded the data with bivariate analysis, and in Table 6, we have decided to replace the logistic regression analysis with odds ratios and confidence intervals.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors Dear Authors, Thank you for your work aimed at improving your article. Current version of the text is much clearer, and self-explanatory. All the data needed to understand your reasoning is presented in the text. Theoretical introduction and discussion are congruent with the empirical part of the paper. At the same time, I still recommend conducting additional regression analyses that can serve as a better foundation for interpretations given in the discussion. Practical implications presented in the article should be a bit more justified based on the data presented in the empirical part of the article.  Additionally, I still don't see the need to describe in the text use of the International Physical Activity Questionnaire since its results are not further discussed. I suggest either skip the description of this tool or to supplement it with additional information stating that the results concerning this questionnaire will be described elsewhere, but since it was one of the elements of the procedure it was also presented in the paper. I suggest introducing above mentioned changes to the paper which can serve as a valuable source of information.

Author Response

Comment 1 : I still recommend conducting additional regression analyses that can serve as a better foundation for interpretations given in the discussion. Practical implications presented in the article should be a bit more justified based on the data presented in the empirical part of the article.

Response: 

In response to your suggestion, we have added Table 7, titled 'Determinants of General Mental Disorders'. This table presents the multivariate analysis results using the presence of any mental disorder (GMHAT) as the dependent variable.

Comment 2 : Additionally, I still don't see the need to describe in the text use of the International Physical Activity Questionnaire since its results are not further discussed. I suggest either skip the description of this tool or to supplement it with additional information stating that the results concerning this questionnaire will be described elsewhere, but since it was one of the elements of the procedure it was also presented in the paper

Response: 

We agree with the reviewer’s suggestion to streamline the focus of the study. Consequently, we have excluded the information regarding the IPQ results. (page 4)