Increased Experiences of Multiple Forms of Discrimination in Healthcare Settings During the COVID-19 Pandemic Among African, Caribbean, and Black (ACB) People Across Canada: A Cross-Sectional Survey

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study is timely, policy-relevant, and makes a strong empirical contribution. With clearer methodological reporting, stronger theoretical integration, and language editing, the manuscript would be suitable for publication in Healthcare.

Here are some suggestions for improvement:

• Provide more detail on how discrimination items were developed.
• Justify your classification approach. 
• Incorporate intersectionality theory into the introduction and discussion.
• Discuss the policy and structural implications in more depth. 
• Explain why other sociodemographic factors were not significant.
• Describe handling of missing data and clarify sampling procedures.
• Recheck grammar for inconsistencies and reduce redundancy in the discussion and conclusion.
• Some web-based sources could be replaced with peer-reviewed literature.

Comments on the Quality of English Language

Edit for: 

• Grammar and verb tense consistency.

• Typographical errors (e.g., “experiend,” “orientate discrimination,” punctuation spacing issues).

• Redundant phrasing in Discussion and Conclusion.

• Minor inconsistencies in terminology.

Author Response

Dear reviewer,

Thank you so much for taking the time to review our manuscript. Please find the detailed point-by-point responses to comments below and the corresponding revisions/corrections in track changes in the re-submitted files.

Quality of English Language

(x) The English could be improved to more clearly express the research.

Response: Thank you; the manuscript has been reviewed and grammatical errors corrected.
( ) The English is fine and does not require any improvement.

 

	Yes	Can be improved	Must be improved	Not applicable
Does the introduction provide sufficient background and include all relevant references? Response: The introduction has been revised and all relevant references cited. Thank you.	( )	(x)	( )	( )
Is the research design appropriate?	(x)	( )	( )	( )
Are the methods adequately described? Response: We have revised the methods section on the manuscript. Thank you.	( )	(x)	( )	( )
Are the results clearly presented?	(x)	( )	( )	( )
Are the conclusions supported by the results?	(x)	( )	( )	( )
Are all figures and tables clear and well-presented?	(x)	( )	( )	( )

Comments and Suggestions for Authors

The study is timely, policy-relevant, and makes a strong empirical contribution. With clearer methodological reporting, stronger theoretical integration, and language editing, the manuscript would be suitable for publication in Healthcare.

Response: Thank you so much for appraising our manuscript and providing helpful comments which have helped us improved on the quality of the manuscript.

Here are some suggestions for improvement:

• Provide more detail on how discrimination items were developed.

Response: Thank you for this request. Our survey questions including the discrimination items were developed from questionnaires used in prior surveys by the National Tracks Surveillance System and other surveys that had measured the impact of COVID-19. We also gathered community input to ensure the questions and discrimination items were relevant to the population and aligned with the realities and needs of the community.

• Justify your classification approach. 

Response: Thanks for this request; our classification approach was based on items we found on previous standard surveys and community contributions. We followed this approach in order to ensure that we captured data that reflects the realities in our study population and community.

• Incorporate intersectionality theory into the introduction and discussion.

Response: Thank you; intersectionality theory as been incorporated into the introduction and discussion sections as requested.

 

• Discuss the policy and structural implications in more depth.

Response: We have discussed the policy and structural implications of discrimination in the discussion section of the manuscript; thank you.

 

• Explain why other sociodemographic factors were not significant.

Response: Thank you for this request. We think this could be influenced by limited variance in the sociodemographic characteristics of our study population – although this was a national survey open to all, it was online and the characteristics of the people who responded to the online survey might have been somehow homogeneous, causing the sociodemographic factors to be non-significant.

• Describe handling of missing data and clarify sampling procedures.

Response: To recruit participants for this study, we used networks, community leaders and stakeholders to distribute our online survey link to potential participants – ACB people were contacted through stakeholders in organizations working with ACB people and the stakeholders disseminated the survey information through their networks and organizational personnel. The survey was designed such that participants who responded that they did not experience any discrimination during the pandemic could skip questions requesting information about the specific type of discrimination experienced. Besides this, some participants did not complete all questions which led to missing data. However, the missing data in the study overall was very minimal (<10%) and as such was not included in our analysis.

 

• Recheck grammar for inconsistencies and reduce redundancy in the discussion and conclusion.

Response: Thank you for this comment; we have reviewed the manuscript and corrected any errors identified.

• Some web-based sources could be replaced with peer-reviewed literature.

Response: Thank you for this suggestion. We have replaced some web-based sources on the manuscript. However, we did not find peer-reviewed literature to replace some sources and the specific web-based sources used, are necessary as they are the direct sources from which we got our information.

 

Comments on the Quality of English Language

Edit for: 

• Grammar and verb tense consistency.

Response: We have reviewed the manuscript and corrected any grammatical errors. Thank you.

 

• Typographical errors (e.g., “experiend,” “orientate discrimination,” punctuation spacing issues).

Response: Thank you for identifying these errors; we have corrected them on the manuscript.

• Redundant phrasing in Discussion and Conclusion.

Response: Thank you for drawing our attention to this. We have reviewed the whole manuscript and made revisions to statements that were not clear on the manuscript.

 

• Minor inconsistencies in terminology.

Response: Thank you for this comment, we have reviewed the manuscript and revised some of our terminologies.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses an important public health issue: Discrimination in healthcare experienced by African, Caribbean, and Black people in Canada during the COVID-19 pandemic. The topic is timely, relevant, and significant for health equity, trust in health systems, and access to care. The manuscript’s strengths include its national scope, its focus on multiple forms of discrimination rather than race alone, and its attempt to examine perceived changes during the pandemic. However, the current version has several substantial weaknesses that limit confidence in the findings. The most important concerns are as follows:

• The outcome is not defined consistently across the manuscript.
• The analytic denominators and missing data are not reported clearly.
• There appears to be a serious ethics-timeline inconsistency.
• The sampling strategy is described more strongly than the methods justify.
• The paper makes stronger claims about change over time than the cross-sectional design supports.

Because of these considerations, I recommend the authors complete a major revision and re-submit the manuscript.

Major Comments:

1. The primary outcome is defined inconsistently and needs to be clarified. The paper does not use a single, consistent definition of the main outcome. In the methods section, the paper explains that participants were asked whether discrimination based on each attribute had increased, decreased, stayed the same, or had not been experienced during the pandemic relative to before the pandemic. The manuscript then states that participants were considered to have experienced discrimination if they selected increase, decrease, or no change. However, the regression tables are described as modeling “increased experience of discrimination,” while some parts of the Results and Discussion describe the outcome more broadly as “overall experience of discrimination.” These are not the same outcomes. This matters because the interpretation of the odds ratios depends entirely on whether the model predicts any discrimination, increased discrimination, or some combined summary of both. The authors should explicitly define the dependent variable used in each model, how it was coded, what the reference category was, and whether the models refer to any discrimination or increased discrimination only. Until that is made consistent across the Methods, Results, table titles, and Discussion, the core findings are difficult to interpret.
2. The manuscript does not report the analytic sample clearly enough. The paper states that 1,556 participants were included in the study, but the totals shown in the discrimination tables are far smaller for each type of discrimination. For example, the total counts in Table 2 imply that respondents were excluded from those analyses, but the paper does not explain whether that is due to item nonresponse, skip patterns, or some other reason. The final regression tables also do not clearly state the sample size included in the models. This is a major reporting problem. Readers need to know how many people consented, how many completed the survey, how many answered the discrimination items, how many were included in each analysis, and how missing data were handled. At least one clear paragraph describing these numbers would improve transparency.
3. The ethics dates raise a serious concern that must be clarified. The manuscript reports that data were collected between May 25, 2021 and July 12, 2021. However, the ethics section lists one approval date as May 5, 2022, which is after data collection ended. As written, this creates the appearance that at least one ethics approval post-dated the research activity. The manuscript needs to explain this clearly, since, in its current form, this is a serious issue.
4. The sampling strategy is described too strongly. The Methods section describes the sampling approach as “purposive, stratified, and proportional allocation sampling,” but the actual procedure described appears to be distribution of an online survey link through networks, community leaders, and stakeholders. That sounds much more like a non-probability, purposive, network-based recruitment strategy than a formally stratified proportional sample. If the sample truly was stratified with proportional allocation, the authors should specify the strata, sampling frame, intended allocation targets, and how proportionality was achieved. If not, the wording should be revised so it does not imply a stronger sampling design than was actually used.
5. The paper overstates novelty. The introduction states that there is little or no information in Canada about the levels and forms of discrimination experienced by African, Caribbean, and Black people when accessing healthcare during the pandemic. That claim is too broad. The manuscript may still be novel in its specific combination of features, especially its national scope and multi-domain discrimination framework, but it should not imply that there is essentially no prior Canadian evidence. A more accurate framing would be that less is known about national patterns, perceived pandemic-related change, and multiple forms of discrimination measured together in this population. This framing is narrower and, I think, more defensible.
6. The research design does not support strong claims about temporal change. This is a cross-sectional survey conducted in 2021, and participants were asked to compare their experiences during the pandemic to their experiences before the pandemic. That design measures retrospective self-perceived change, not actual measured change over time. The language throughout the paper should reflect this limitation. The authors should consider avoiding phrasing that suggests they observed discrimination increasing longitudinally. Instead, they should say respondents reported that discrimination had increased.

Other Comments:

• Abstract - There appears to be a numerical inconsistency between the abstract and the results. The abstract states that 75% of participants experienced at least one form of discrimination prior to the pandemic, but the results later state that 37.9% never experienced any form of discrimination prior to the pandemic. Those numbers do not align. That is, if 37.9% never experienced discrimination, then 62.1% did, not 75%.
• Introduction - Some of the citations in the introduction appear weakly matched to the claims they support. The use of a general WHO web source is not ideal for a specific scientific background claim, and one cited source appears to be a web-based institutional article rather than a peer-reviewed study. A stronger literature base would greatly improve the introduction.
• Introduction - There is a mismatch between some cited examples of gender discrimination and the paper’s actual outcome. Some of the cited material appears to concern workplace discrimination among healthcare workers rather than discrimination experienced by patients accessing healthcare. If so, those sources are not well aligned with the study's main research question.
• Methods - The connection between the parent survey and this specific discrimination analysis is not explained well enough. If the broader survey focused on STBBI-related services, the authors should clarify whether the present analysis is intended to generalize to all healthcare settings or to the subset of people reached through that topic area. 
• Methods - The paper should clarify how the survey was disseminated geographically and whether any provinces or regions were underrepresented.
• Methods - The paper does not report enough recruitment detail to assess selection bias. The fact that more than 5,000 organizations and individuals received the survey link is not, by itself, an interpretable denominator. We still do not know how many people actually saw the link, how many opened the survey, how many were eligible, and how many completed the relevant questions.
• Methods - The authors should clarify the implications of an item structure that combines occurrence and change in a single response format. And they should discuss it more directly as a limitation.
• Methods - The manuscript does not explain how missing data were handled in the models. Since the descriptive tables already suggest substantial item-level variation in sample size, this is a serious omission.
• Methods - The grouping of some demographic variables also deserves explanation. For instance, the categories appear to collapse distinct gender identities together, which may have been necessary statistically, but needs to be justified explicitly. 
• Results - The age categories used in Table 1 do not match those used later in the regression tables. That inconsistency makes the paper harder to read and interpret.
• Results - The denominators are unclear, since the number of responses contributing to each discrimination category is much smaller than the full sample. The paper does not explain why, however. As a result, the percentages are difficult to interpret confidently.
• Results - There is inconsistent temporal language in the Results section, with the paper referring to the pandemic period as “following the pandemic,” “during the pandemic,” and “since the start of the pandemic.” These phrases are not identical, so they should be standardized.
• Results - The results sometimes describe the model as predicting increased discrimination and sometimes as predicting overall discrimination. The authors should clarify the model outcome, report the final analytic N, and acknowledge sparse-data limitations. 
• Discussion - The discussion section does not sufficiently engage with the study's methodological limitations. It should be clearer that the study measures retrospective perceptions of change, rather than observed change over time. It should also discuss the implications of non-probability sampling and missing information at the item level more directly.
• Discussion - Not enough is being done with the age finding. There are several potential explanations for lower reported increased discrimination among older respondents, including cohort differences, expectations, healthcare-use patterns, and survey-selection effects, among other things.

This manuscript addresses an important topic and has the potential to make a useful contribution, but it is not yet ready for publication. The authors should first resolve the outcome/definition problem. They should also reconcile all descriptive numbers, clarify the analytic sample and missing information, and explain the ethics/IRB dates. These revisions will make the paper easier to evaluate on its own merits. 

Author Response

Dear reviewer,

Thank you so much for taking the time to review our manuscript. Please find the detailed point-by-point responses to comments below and the corresponding revisions/corrections in track changes in the re-submitted files.

 

Review Report Form

Open Review

(x) I would not like to sign my review report
( ) I would like to sign my review report

Quality of English Language

( ) The English could be improved to more clearly express the research.
(x) The English is fine and does not require any improvement.

Response: Thank you.

 

	Yes	Can be improved	Must be improved	Not applicable
Does the introduction provide sufficient background and include all relevant references?	( )	( )	(x)	( )
Is the research design appropriate? Response: The methods section has been revised to ensure the design is appropriate. Thank you.	( )	(x)	( )	( )
Are the methods adequately described?	( )	( )	(x)	( )
Are the results clearly presented?	( )	( )	(x)	( )
Are the conclusions supported by the results? Response: We have reviewed our conclusions to ensure they are supported by our results.	( )	(x)	( )	( )
Are all figures and tables clear and well-presented? Response: Thank you for this comment. All our figures are well labeled and tables provided with footnotes to improve clarity.	( )	(x)	( )	( )

Comments and Suggestions for Authors

This manuscript addresses an important public health issue: Discrimination in healthcare experienced by African, Caribbean, and Black people in Canada during the COVID-19 pandemic. The topic is timely, relevant, and significant for health equity, trust in health systems, and access to care. The manuscript’s strengths include its national scope, its focus on multiple forms of discrimination rather than race alone, and its attempt to examine perceived changes during the pandemic. However, the current version has several substantial weaknesses that limit confidence in the findings. The most important concerns are as follows:

• The outcome is not defined consistently across the manuscript.

Response: Thank you for appraising our manuscript and for this keen observation. The outcome is perceived increased experience of discrimination during the pandemic – one of the questions participants were asked was whether the frequency of their experiences of discrimination increased, decreased, or remained about the same when they accessed healthcare services during the pandemic. We changed this into a variable with two categories (increased experience of discrimination versus decreased experience/no change in experienced frequency) – this is the binary outcome variable described under statistical analysis section of the manuscript. The inconsistencies observed in the definition of this outcome in the manuscript were errors which have now been corrected. In fact, we used the definition “participants were considered to have experience discrimination if they selected increase, decrease or no change in any discrimination attribute” to show the overall proportion of participants that experienced each type of discrimination. In our model however, the outcome is simply perceived increased experience of discrimination coded as those who experienced increased discrimination versus those who experienced decreased/no change in frequency of experienced discrimination.

 

 

• The analytic denominators and missing data are not reported clearly.

Response: Thank you for this observation. Our survey was designed such that participants who responded that they did not experience any discrimination during the pandemic could skip questions requesting information about the specific type of discrimination experienced. Besides this, some participants did not complete all questions which led to missing data. So, the analytic denominators slightly vary with respect to the specific discrimination types due to missing data. Nevertheless, the proportion of people who reported experiencing any discrimination is calculated from the full sample excluding missing data and the proportion of people who experience specific types of discrimination is calculated from the sample that reported having experienced any discrimination (missing data also excluded). However, the missing data in the study overall was very minimal (<10%) and as such was not included in our analysis.

 

 

• There appears to be a serious ethics-timeline inconsistency.

Response: Thank you for this comment. This study involved a quantitative and a qualitative component. The quantitative component that had an online survey was first approved by the Public Health Agency of Canada (PHAC) Research Ethics Board (REB) in 2020 and this allowed the study to start. The qualitative component that involved focus group discussions was introduced later under the leadership of a University of Ottawa (uOttawa) professor – this component also needed to be approved prior to implementation by the uOttawa REB; that is why there are two separate ethics approvals from different institutions at different times. However, the first ethics approval was sufficient to start the quantitative component of the study.

• The sampling strategy is described more strongly than the methods justify.

Response: Thank you for this keen observation. Methodological details of the study have been reported elsewhere and the reference to where the details can be found have been provided on the manuscript (end of the first paragraph under methods). It was the general agreement of all authors to focus on the sampling strategy than elaborate on the methods which is already available online. However, some revisions have been made on the sampling strategy for more clarity.

• The paper makes stronger claims about change over time than the cross-sectional design supports.

Response: Thanks for this observation; the mention of change over time refers to comparison of participants’ perspectives on their experiences prior to the pandemic and during the pandemic. These perspectives were captured at a particular time in conformity with our cross-sectional design.

Because of these considerations, I recommend the authors complete a major revision and re-submit the manuscript.

Response: Thank you, we have revised the manuscript accordingly.

 

Major Comments:

1. The primary outcome is defined inconsistently and needs to be clarified. The paper does not use a single, consistent definition of the main outcome. In the methods section, the paper explains that participants were asked whether discrimination based on each attribute had increased, decreased, stayed the same, or had not been experienced during the pandemic relative to before the pandemic. The manuscript then states that participants were considered to have experienced discrimination if they selected increase, decrease, or no change. However, the regression tables are described as modeling “increased experience of discrimination,” while some parts of the Results and Discussion describe the outcome more broadly as “overall experience of discrimination.” These are not the same outcomes. This matters because the interpretation of the odds ratios depends entirely on whether the model predicts any discrimination, increased discrimination, or some combined summary of both. The authors should explicitly define the dependent variable used in each model, how it was coded, what the reference category was, and whether the models refer to any discrimination or increased discrimination only. Until that is made consistent across the Methods, Results, table titles, and Discussion, the core findings are difficult to interpret.

Response: We greatly appreciate this keen observation; the primary outcome is perceived increased experience of discrimination during the pandemic – one of the questions participants were asked was whether the frequency of their experiences of discrimination increased, decreased, or remained about the same when they accessed healthcare services during the pandemic. We changed this into a variable with two categories (increased experience of discrimination versus decreased experience/no change in experienced frequency) – this is the binary outcome variable described under statistical analysis section of the manuscript. The inconsistencies observed in the definition of this outcome in the manuscript were errors which have now been corrected. In fact, we used the definition “participants were considered to have experience discrimination if they selected increase, decrease or no change in any discrimination attribute” to show the overall proportion of participants that experienced each type of discrimination. In our model however, the outcome is simply perceived increased experience of discrimination coded as those who experienced increased discrimination versus those who experienced decreased/no change in frequency of experienced discrimination.

 

1. The manuscript does not report the analytic sample clearly enough. The paper states that 1,556 participants were included in the study, but the totals shown in the discrimination tables are far smaller for each type of discrimination. For example, the total counts in Table 2 imply that respondents were excluded from those analyses, but the paper does not explain whether that is due to item nonresponse, skip patterns, or some other reason. The final regression tables also do not clearly state the sample size included in the models. This is a major reporting problem. Readers need to know how many people consented, how many completed the survey, how many answered the discrimination items, how many were included in each analysis, and how missing data were handled. At least one clear paragraph describing these numbers would improve transparency.

Response: Thank you for this observation. The total number of participants in the study were 1,556; of these, only those who experienced any discrimination further responded to the type experienced. Also, among all those who experienced discrimination, not all provided responses to the type of discrimination they experienced. So, the total for each type of discrimination varies due to skip patterns and nonresponse. Since the nonresponse rate for each type of discrimination was less than 10%, those were considered as missing and not included in the analysis. The tables show only the number of participants that actually responded to each type of discrimination. We have added some sentences to the first paragraph of the forms of discrimination under the results section on the manuscript to clarify this.

 

1. The ethics dates raise a serious concern that must be clarified. The manuscript reports that data were collected between May 25, 2021 and July 12, 2021. However, the ethics section lists one approval date as May 5, 2022, which is after data collection ended. As written, this creates the appearance that at least one ethics approval post-dated the research activity. The manuscript needs to explain this clearly, since, in its current form, this is a serious issue.

Response: Thank you for this comment. This study involved a quantitative and a qualitative component. The quantitative component that had an online survey was first approved by the Public Health Agency of Canada (PHAC) Research Ethics Board (REB) in 2020 and this allowed the study to start. The qualitative component that involved focus group discussions was introduced later under the leadership of a University of Ottawa (uOttawa) professor – this component also needed to be approved prior to implementation by the uOttawa REB; that is why there are two separate ethics approvals from different institutions at different times. However, the first ethics approval was sufficient to start the quantitative component of the study.

 

 

1. The sampling strategy is described too strongly. The Methods section describes the sampling approach as “purposive, stratified, and proportional allocation sampling,” but the actual procedure described appears to be distribution of an online survey link through networks, community leaders, and stakeholders. That sounds much more like a non-probability, purposive, network-based recruitment strategy than a formally stratified proportional sample. If the sample truly was stratified with proportional allocation, the authors should specify the strata, sampling frame, intended allocation targets, and how proportionality was achieved. If not, the wording should be revised so it does not imply a stronger sampling design than was actually used.

Response: Thank you for this observation. The sampling strategy has been revised on the manuscript as advised.

 

1. The paper overstates novelty. The introduction states that there is little or no information in Canada about the levels and forms of discrimination experienced by African, Caribbean, and Black people when accessing healthcare during the pandemic. That claim is too broad. The manuscript may still be novel in its specific combination of features, especially its national scope and multi-domain discrimination framework, but it should not imply that there is essentially no prior Canadian evidence. A more accurate framing would be that less is known about national patterns, perceived pandemic-related change, and multiple forms of discrimination measured together in this population. This framing is narrower and, I think, more defensible.

Response: Thank you so much for the suggested framing; the sentence has been revised accordingly on the manuscript.

 

1. The research design does not support strong claims about temporal change. This is a cross-sectional survey conducted in 2021, and participants were asked to compare their experiences during the pandemic to their experiences before the pandemic. That design measures retrospective self-perceived change, not actual measured change over time. The language throughout the paper should reflect this limitation. The authors should consider avoiding phrasing that suggests they observed discrimination increasing longitudinally. Instead, they should say respondents reported that discrimination had increased.

Response: Thanks again for this keen observation; we have modified the language in presenting the results and discussion accordingly.

 

Other Comments:

• Abstract - There appears to be a numerical inconsistency between the abstract and the results. The abstract states that 75% of participants experienced at least one form of discrimination prior to the pandemic, but the results later state that 37.9% never experienced any form of discrimination prior to the pandemic. Those numbers do not align. That is, if 37.9% never experienced discrimination, then 62.1% did, not 75%.

Response: This was a typographical error; the correct value is 62.1%. Correction has been made on the abstract.

 

• Introduction - Some of the citations in the introduction appear weakly matched to the claims they support. The use of a general WHO web source is not ideal for a specific scientific background claim, and one cited source appears to be a web-based institutional article rather than a peer-reviewed study. A stronger literature base would greatly improve the introduction.

Response: We greatly appreciate this keen observation; the WHO web source has been replaced with a peer-reviewed citation. Thank you.

 

• Introduction - There is a mismatch between some cited examples of gender discrimination and the paper’s actual outcome. Some of the cited material appears to concern workplace discrimination among healthcare workers rather than discrimination experienced by patients accessing healthcare. If so, those sources are not well aligned with the study's main research question.

Response: Again, we greatly appreciate this keen observation. The statement that concern workplace discrimination and citation has been removed in the second paragraph of the introduction. Thank you.

 

• Methods - The connection between the parent survey and this specific discrimination analysis is not explained well enough. If the broader survey focused on STBBI-related services, the authors should clarify whether the present analysis is intended to generalize to all healthcare settings or to the subset of people reached through that topic area. 

Response: Thank you for this request; the analysis presented in our study is intended to generalize only to the subset of the people reached and not to all healthcare settings. We have also clearly stated the limitations of the design we used and the representativeness issues of our sample under the discussion section of the manuscript.

 

• Methods - The paper should clarify how the survey was disseminated geographically and whether any provinces or regions were underrepresented.

Response: The Public Health Agency of Canada contacted stakeholders throughout the country and forwarded the link of the survey to them. These stakeholders helped in disseminating the survey in their respective geographical areas. From our results on Table 1, we can see that the survey was disseminated in all provinces even though it is hard to say whether the responses from each province was as a result of underrepresentation or overrepresentation.

 

• Methods - The paper does not report enough recruitment detail to assess selection bias. The fact that more than 5,000 organizations and individuals received the survey link is not, by itself, an interpretable denominator. We still do not know how many people actually saw the link, how many opened the survey, how many were eligible, and how many completed the relevant questions.

Response: Thank you for this critical observation. We have clearly stated the limitations of our study design and methods which do not exclude selection bias in our discussion section. We were not able to capture information about the number of people who saw the survey link and those who were eligible amongst them; so, we could not report these details. Our focus was to analyze the data for those who responded to the survey and submitted their responses at the end of the survey.

 

• Methods - The authors should clarify the implications of an item structure that combines occurrence and change in a single response format. And they should discuss it more directly as a limitation.

Response: Thank you for this request. We have revised the manuscript and elaborated on the study limitations under the discussion section of the manuscript.

 

• Methods - The manuscript does not explain how missing data were handled in the models. Since the descriptive tables already suggest substantial item-level variation in sample size, this is a serious omission.

Response: Thank you for this keen observation. During the analysis, we computed the proportion of missing information from the data. We realised that the proportion of the missing information was less than 10% for most of the variables that had missing information. Since the proportion of the missing data was insignificant, this was not included in the analysis. We have revised the data analysis section to include this information on the manuscript.

 

• Methods - The grouping of some demographic variables also deserves explanation. For instance, the categories appear to collapse distinct gender identities together, which may have been necessary statistically, but needs to be justified explicitly. 

Response: Thank you for this observation. When conducting the regression analysis, we realised that some variable categories had no values when matched with the outcome variable. We had to regroup the variable categories in a way that would permit the analysis to be conducted.

 

• Results - The age categories used in Table 1 do not match those used later in the regression tables. That inconsistency makes the paper harder to read and interpret.

Response: Thank you for this keen observation. When doing the regression analysis, we realized that some of the age categories presented in Table 1 had no values when matched with the outcome variable; so, we had to recategorize the age variable such that each category had values that permitted a regression analysis to be conducted.

 

• Results - The denominators are unclear, since the number of responses contributing to each discrimination category is much smaller than the full sample. The paper does not explain why, however. As a result, the percentages are difficult to interpret confidently.

Response: Thank you for this keen observation. The percentages of each discrimination category are not calculated from the full sample but only from the sample that experienced any category of discrimination during the pandemic.

 

• Results - There is inconsistent temporal language in the Results section, with the paper referring to the pandemic period as “following the pandemic,” “during the pandemic,” and “since the start of the pandemic.” These phrases are not identical, so they should be standardized.

Response: Thank you for this observation; “during the pandemic” has been adopted and standardized throughout the results.

 

• Results - The results sometimes describe the model as predicting increased discrimination and sometimes as predicting overall discrimination. The authors should clarify the model outcome, report the final analytic N, and acknowledge sparse-data limitations. 

Response: Thank you for this comment. The model outcome is increased experience of discrimination. The number of participants included in our analysis is 1556 – some variables however have missing values and because the missing values were very few (<5%), they were dropped in the analysis.

 

• Discussion - The discussion section does not sufficiently engage with the study's methodological limitations. It should be clearer that the study measures retrospective perceptions of change, rather than observed change over time. It should also discuss the implications of non-probability sampling and missing information at the item level more directly.

Response: The limitation section of the discussion has now been revised to include the study’s methodological limitations. Thank you.

 

• Discussion - Not enough is being done with the age finding. There are several potential explanations for lower reported increased discrimination among older respondents, including cohort differences, expectations, healthcare-use patterns, and survey-selection effects, among other things.

Response: Additional discussion on age and experience of discrimination has been added on the manuscript; thank you.

 

This manuscript addresses an important topic and has the potential to make a useful contribution, but it is not yet ready for publication. The authors should first resolve the outcome/definition problem. They should also reconcile all descriptive numbers, clarify the analytic sample and missing information, and explain the ethics/IRB dates. These revisions will make the paper easier to evaluate on its own merits. 

Response: Thank you for this comment; we have now clearly defined our outcome variable and addressed other issues raised.

 

Reviewer 3 Report

Comments and Suggestions for Authors

I read with interest the study presented by Etowa and colleagues on the experiences of multiple forms of discrimination in healthcare settings among African, Caribbean, and Black (ACB) people during the COVID-19 pandemic in Canada. Two key findings emerge from this work. First, a large proportion of participants reported experiencing discrimination when accessing healthcare services during the pandemic, with more than two-thirds reporting at least one form of discrimination. Second, race-based discrimination appeared to be the most prevalent type, and the only sociodemographic factor that remained significantly associated with increased discrimination after multivariable adjustment was older age. 

The topic aligns well with the aims and scope of the journal Healthcare, which publishes research addressing health equity, healthcare access, and social determinants of health. Given the increasing interest in health disparities and systemic barriers to care, the manuscript is likely to be of interest to the journal’s readership.

Despite its merits, the manuscript is riddled with issues that need to be addressed before it can be accepted for publication.

 

1. The introduction provides a general overview of discrimination in healthcare but lacks a robust conceptual framework explaining how multiple forms of discrimination interact. The manuscript mentions several types of discrimination (race, gender, sexual orientation, disability, and economic status), yet the theoretical basis for examining multiple or intersecting forms of discrimination is not adequately developed. The authors should strengthen the introduction by incorporating theoretical frameworks such as intersectionality and structural discrimination in healthcare systems. This would clarify how multiple discrimination domains interact and justify the analytical approach adopted in the study.

2. Although the introduction states that there is limited information on discrimination experienced by ACB communities during the pandemic, the justification for the study could be articulated more clearly. The authors should explicitly define the knowledge gap addressed by the study and distinguish how their work advances current literature beyond previous research on healthcare discrimination and pandemic-related disparities.

3. Participants were recruited using purposive and network-based sampling through stakeholders and organizations. While understandable given the pandemic context, the implications of this sampling strategy for representativeness are not sufficiently discussed. The authors should clarify:

how sampling quotas or stratification were implemented, whether the sample reflects the demographic distribution of ACB populations in Canada, and the potential for selection bias and overrepresentation of specific networks.

This clarification is necessary to evaluate the external validity of the findings.

4. The manuscript does not sufficiently describe how discrimination variables were operationalized in the survey instrument. For example, the items used to measure discrimination and whether validated scales were employed are not clearly stated. The authors should provide:

the exact survey items used to measure discrimination, justification for the measurement approach, references for validated instruments (if used)

clarification on how “increase,” “decrease,” and “no change” responses were operationalized in the analysis.

Without this information, the reproducibility and validity of the findings are difficult to assess.

5. The statistical analysis describes the use of bivariate and multivariable logistic regression models. However, the rationale for selecting variables for the multivariable model (p ≤ 0.25 threshold) is only briefly mentioned. The authors should elaborate on: the model selection strategy

the assessment of multicollinearity

whether interaction effects were tested (particularly given the multi-discrimination framework), and goodness-of-fit diagnostics for the final model.

6. The discussion occasionally implies causal interpretations despite the cross-sectional design of the study. For instance, statements suggesting that discrimination “led to” certain healthcare outcomes should be interpreted cautiously. The authors should revise the discussion to emphasize associations rather than causal relationships, and explicitly acknowledge the limitations inherent in cross-sectional survey designs.

7. Although the tables present detailed data, some of them are overly dense and difficult to interpret quickly. For instance, the tables presenting discrimination categories and regression outputs could be simplified for readability. The authors should consider:

simplifying tables by grouping related variables, improving formatting of regression tables providing clearer titles and explanatory footnotes.

8. The manuscript acknowledges several limitations, including potential response bias and recall bias. However, additional limitations should be addressed, particularly regarding online survey methods and potential underrepresentation of individuals with limited internet access. The authors should expand the limitations section to include: digital access disparities, language limitations beyond English and French, potential social desirability bias in reporting discrimination experiences.

Author Response

Dear reviewer,

Thank you so much for taking the time to review our manuscript. Please find the detailed point-by-point responses to comments below and the corresponding revisions/corrections in track changes in the re-submitted files.

 

Review Report Form

Open Review

(x) I would not like to sign my review report
( ) I would like to sign my review report

Quality of English Language

( ) The English could be improved to more clearly express the research.
(x) The English is fine and does not require any improvement.

 

	Yes	Can be improved	Must be improved	Not applicable
Does the introduction provide sufficient background and include all relevant references? Response: The introduction section has now been revised and all relevant references cited.	( )	(x)	( )	( )
Is the research design appropriate?	(x)	( )	( )	( )
Are the methods adequately described?	( )	( )	(x)	( )
Are the results clearly presented? Response: Thank you for this comment. The results section has been revised to improve clarity.	( )	(x)	( )	( )
Are the conclusions supported by the results?	(x)	( )	( )	( )
Are all figures and tables clear and well-presented?	( )	( )	(x)	( )

Comments and Suggestions for Authors

I read with interest the study presented by Etowa and colleagues on the experiences of multiple forms of discrimination in healthcare settings among African, Caribbean, and Black (ACB) people during the COVID-19 pandemic in Canada. Two key findings emerge from this work. First, a large proportion of participants reported experiencing discrimination when accessing healthcare services during the pandemic, with more than two-thirds reporting at least one form of discrimination. Second, race-based discrimination appeared to be the most prevalent type, and the only sociodemographic factor that remained significantly associated with increased discrimination after multivariable adjustment was older age. 

The topic aligns well with the aims and scope of the journal Healthcare, which publishes research addressing health equity, healthcare access, and social determinants of health. Given the increasing interest in health disparities and systemic barriers to care, the manuscript is likely to be of interest to the journal’s readership.

Despite its merits, the manuscript is riddled with issues that need to be addressed before it can be accepted for publication.

Response: We greatly appreciate your appraisal of our manuscript and the comments made to help improve the quality of the work. We have revised the manuscript accordingly, thank you.

 

 

1. The introduction provides a general overview of discrimination in healthcare but lacks a robust conceptual framework explaining how multiple forms of discrimination interact. The manuscript mentions several types of discrimination (race, gender, sexual orientation, disability, and economic status), yet the theoretical basis for examining multiple or intersecting forms of discrimination is not adequately developed. The authors should strengthen the introduction by incorporating theoretical frameworks such as intersectionality and structural discrimination in healthcare systems. This would clarify how multiple discrimination domains interact and justify the analytical approach adopted in the study.

Response: Thank you for this request; the intersectionality theory has been incorporated in the introduction section.

 

2. Although the introduction states that there is limited information on discrimination experienced by ACB communities during the pandemic, the justification for the study could be articulated more clearly. The authors should explicitly define the knowledge gap addressed by the study and distinguish how their work advances current literature beyond previous research on healthcare discrimination and pandemic-related disparities.

Response: Thanks again for this request; the last paragraph of the introduction has been revised and a statement clearly stating the knowledge gap included.

 

3. Participants were recruited using purposive and network-based sampling through stakeholders and organizations. While understandable given the pandemic context, the implications of this sampling strategy for representativeness are not sufficiently discussed. The authors should clarify:

how sampling quotas or stratification were implemented, whether the sample reflects the demographic distribution of ACB populations in Canada, and the potential for selection bias and overrepresentation of specific networks.

This clarification is necessary to evaluate the external validity of the findings.

Response: Thank you for this comment. The sampling strategy has been revised on the manuscript. Also, the limitations section has been revised to include the implications of our sampling for representativeness.

 

4. The manuscript does not sufficiently describe how discrimination variables were operationalized in the survey instrument. For example, the items used to measure discrimination and whether validated scales were employed are not clearly stated. The authors should provide:

the exact survey items used to measure discrimination, justification for the measurement approach, references for validated instruments (if used)

clarification on how “increase,” “decrease,” and “no change” responses were operationalized in the analysis.

Without this information, the reproducibility and validity of the findings are difficult to assess.

Response: Thank you for this request. Our survey questions including the discrimination items were developed from questionnaires used in prior surveys by the National Tracks Surveillance System and other surveys that had measured the impact of COVID-19. We also gathered community input to ensure the questions and discrimination items were relevant to the population and aligned with the realities and needs of the community. The survey instrument was pretested prior to use. In the analysis, “increase”, “decrease”, and “no change” responses were treated as categories of each discrimination type – each discrimination type was treated as a categorical variable with three categories (increase, decrease, and no change). However, for our model outcome, we created a binary outcome variable (increased experience of discrimination) that had categories as “increase” versus “decrease” and “no change”. We have submitted the survey questionnaire that was used.

 

5. The statistical analysis describes the use of bivariate and multivariable logistic regression models. However, the rationale for selecting variables for the multivariable model (p ≤ 0.25 threshold) is only briefly mentioned. The authors should elaborate on: the model selection strategy

the assessment of multicollinearity

whether interaction effects were tested (particularly given the multi-discrimination framework), and goodness-of-fit diagnostics for the final model.

Response: Variables for our multivariable model were selected following purposeful selection based on the theory put forth in the study by Bursac et al, 2008; this study explains that in selecting variables for logistic regression, the p-value ≤ 0.25 cutoff is set for a variable to be added to the model.  The study is cited under the statistical analysis section of the manuscript. We strictly adhered to the principles described in the study to select variables for our multivariable model.

The discussion occasionally implies causal interpretations despite the cross-sectional design of the study. For instance, statements suggesting that discrimination “led to” certain healthcare outcomes should be interpreted cautiously. The authors should revise the discussion to emphasize associations rather than causal relationships, and explicitly acknowledge the limitations inherent in cross-sectional survey designs.

Response: We greatly appreciate this comment. The discussion section has been revised accordingly.

6. Although the tables present detailed data, some of them are overly dense and difficult to interpret quickly. For instance, the tables presenting discrimination categories and regression outputs could be simplified for readability. The authors should consider:

simplifying tables by grouping related variables, improving formatting of regression tables providing clearer titles and explanatory footnotes.

Response: Thank you for this comment. In order not to distort the results, all authors have agreed to leave the tables as they are. However, all tables have footnotes explaining the abbreviations used on the table, reference variable categories and the cutoff p-values.

 

7. The manuscript acknowledges several limitations, including potential response bias and recall bias. However, additional limitations should be addressed, particularly regarding online survey methods and potential underrepresentation of individuals with limited internet access. The authors should expand the limitations section to include: digital access disparities, language limitations beyond English and French, potential social desirability bias in reporting discrimination experiences.

Response: Thank you so much for this comment. The limitations section has been expanded to include the suggested limitations.

 

Reviewer 4 Report

Comments and Suggestions for Authors

This is an interesting paper and for the most part it is written and presented well. The statistics and there presentation was conducted well and the results tables and figures are clear and easy to follow. As will be suggested below some explanation of some of these results might add to the paper. The limitations for the most part are appropriate and yet the strong possibility of discrimination already affecting the sample group might be more clearly noted. The conclusion is useful and does suggest some useful ideas for future research. However, there are some issues which in my view should be addressed.

On page 3 the methods clearly state that the participants are from patients seeking access to sexually transmitted and blood borne infection-related services. This appears to be a group that might sadly, already face discrimination. While it is acknowledged that this sample may limit the study, I think it should be more clearly noted that this sample probably already faces discrimination that may not be at the same level other ACB patients seeking other medical services might endure. 

On page 4 it states the questionnaire includes items regarding health providers knowledge and experience of working with ACB communities and their capacity to the needs of ACB patients before and during the pandemic. Yet, there appears to be no results presented on these questions. This could be interesting and I wonder why it is not included?

The results also indicate that patients over 50 and living in the northern territories are more likely to be discriminated against. Yet, no possible explanation for these results are given. Inclusion of suggested reasons for these results might add context especially to readers unfamiliar with Canada. 

There are also some small wording suggestions and errors and these are listed below for ease of change. 

Page 3, line 99 suggest placing a fullstops after USA. Therby splitting the sentence into two for readability. 

Page 4, line 174 can I suggested the word done be changed to 'conducted' as it seems more appropriate in light of the overall academic language of the paper. This is only a suggestion not a mandatory change. 

Page 7, line 221 I would suggest changing the word 'following' to 'during' to better reflect the time period participants were being surveyed about. 

Page 14, line 359 suggest changing 'orientate' to 'orientated' for readability.

 

 

Comments on the Quality of English Language

The English is very good and the only obvious errors were those listed above.

Author Response

Dear reviewer,

Thank you so much for taking the time to review our manuscript. Please find the detailed point-by-point responses to comments below and the corresponding revisions/corrections in track changes in the re-submitted files.

 

Review Report Form

Open Review

(x) I would not like to sign my review report
( ) I would like to sign my review report

Quality of English Language

(x) The English could be improved to more clearly express the research.

Response: Thanks for this comment. The paper has been revised and grammatical errors corrected.
( ) The English is fine and does not require any improvement.

 

	Yes	Can be improved	Must be improved	Not applicable
Does the introduction provide sufficient background and include all relevant references?	(x)	( )	( )	( )
Is the research design appropriate? Response: Thanks for this comment. We have revised the methods section to ensure the design is appropriate.	( )	(x)	( )	( )
Are the methods adequately described? Response: We have revised the methods section of the manuscript.	( )	(x)	( )	( )
Are the results clearly presented?	(x)	( )	( )	( )
Are the conclusions supported by the results? Response: Thank you; we have reviewed our conclusions and confirm that they are supported by results.	( )	(x)	( )	( )
Are all figures and tables clear and well-presented?	(x)	( )	( )	( )

Comments and Suggestions for Authors

This is an interesting paper and for the most part it is written and presented well. The statistics and there presentation was conducted well and the results tables and figures are clear and easy to follow. As will be suggested below some explanation of some of these results might add to the paper. The limitations for the most part are appropriate and yet the strong possibility of discrimination already affecting the sample group might be more clearly noted. The conclusion is useful and does suggest some useful ideas for future research. However, there are some issues which in my view should be addressed.

Response: Thank you so much for your appraisal of our manuscript and providing insightful comments and suggestions to improve the manuscript. Please see our point-by-point responses to your comments below.

On page 3 the methods clearly state that the participants are from patients seeking access to sexually transmitted and blood borne infection-related services. This appears to be a group that might sadly, already face discrimination. While it is acknowledged that this sample may limit the study, I think it should be more clearly noted that this sample probably already faces discrimination that may not be at the same level other ACB patients seeking other medical services might endure. 

Response: Thank you for this keen observation and suggestion. Suggestion has been accepted and added to the limitations section on the manuscript.

 

On page 4 it states the questionnaire includes items regarding health providers knowledge and experience of working with ACB communities and their capacity to the needs of ACB patients before and during the pandemic. Yet, there appears to be no results presented on these questions. This could be interesting and I wonder why it is not included?

Response: Thank you for this keen observation. Our focus on this paper was to present the experiences of ACB people in relation to discrimination during the COVID-19 pandemic; the analysis of the data for health providers will be conducted separately. Results on health providers’ knowledge and experiences will be presented in a different paper when the analysis has been completed.

The results also indicate that patients over 50 and living in the northern territories are more likely to be discriminated against. Yet, no possible explanation for these results are given. Inclusion of suggested reasons for these results might add context especially to readers unfamiliar with Canada. 

Response: We have added potential reasons for this in the discussion section of the manuscript. Thank you.

 

There are also some small wording suggestions and errors and these are listed below for ease of change. 

Page 3, line 99 suggest placing a fullstops after USA. Therby splitting the sentence into two for readability. 

Response: Thanks for this suggestion; change has been made on manuscript.

Page 4, line 174 can I suggested the word done be changed to 'conducted' as it seems more appropriate in light of the overall academic language of the paper. This is only a suggestion not a mandatory change. 

Response: Thank you for this suggestion; change has been made on manuscript as suggested.

Page 7, line 221 I would suggest changing the word 'following' to 'during' to better reflect the time period participants were being surveyed about. 

Response: Change has been made as suggested on manuscript.

Page 14, line 359 suggest changing 'orientate' to 'orientated' for readability.

Response: Thanks for this suggestion, correction has been made on manuscript.

Comments on the Quality of English Language

The English is very good and the only obvious errors were those listed above.

Response: Thank you.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The revised article is meaningfully improved but still has a few important problems. The authors did respond to several of my critiques, especially around novelty framing, cross-sectional/retrospective design language, non-probability sampling, STBBI context, and missing data acknowledgment. However, they did not fully resolve the two issues I would still consider the most serious: The ethics-timeline inconsistency and the continued lack of full analytic transparency about denominators/model sample construction. 

Below are my suggestions regarding what remains to be fixed in the manuscript before it is ready for publication: 

1. The ethics-timeline issue remains unresolved. The manuscript still says that recruitment and data collection occurred in 2021, but IRB approval was not obtained until 2022. Unless the authors explain that this was an amendment, a secondary-analysis approval, or some later institutional signoff on an already-approved study, this remains a serious issue.
2. The model interpretation is clearer, but the prose is still a bit loose in some places. For example, the Results still say "factors associated with participants' overall increased experience of discrimination" and "overall participants perceived increased experience of discrimination." This phrasing is awkward and a bit muddy. It suggests a composite outcome without ever fully defining how that summary variable was created across domains. Ultimately, the core binary outcome is more understandable now, but the writing still needs tightening. 
3. The age finding is still interpreted rather weakly. The authors now identify older age as the only significant adjusted correlate and attempt a brief interpretation, but the explanation is not very persuasive. They write that people over 50 in Canada’s Northern Territories face higher risk of discrimination, yet the model’s significant result is about older age overall, not specifically northern residence, and northern residence was not significant in the adjusted model. So the interpretation feels somewhat mismatched to the actual model results.
4. Some phrasing and grammar issues remain. There are still places where the prose is awkward or imprecise. For example,  "both individually and as a group did not accurately distinguish"; "data was collected" instead of "data were collected"; "other factors that could possibly contribute…needs to be explored” (in both abstract and conclusion). 
5. The limitations are stronger now, but the strengths section still says the statistical analysis allowed for an accurate description of the levels of discrimination and associated factors. That still feels too strong given the non-probability design, retrospective change measure, and acknowledged missing data issues. 

In sum, the manuscript is significantly improved, but still requires adjustments along the above lines if it is to be published in this journal. 

Author Response

Dear reviewer,

Thank you again for taking the time to review the revisions made on our manuscript. Please find the point-by-point responses to your comments below and the corresponding revisions/corrections in track changes in the re-submitted files.

 

Quality of English Language

( ) The English could be improved to more clearly express the research.
(x) The English is fine and does not require any improvement.

Response: Thank you.

 

	Yes	Can be improved	Must be improved	Not applicable
Does the introduction provide sufficient background and include all relevant references? Response: Thank you for this comment. We have reviewed our introduction; all relevant references to the information provided have been cited. We did not find any other information that we could add to this section of the manuscript.	( )	( )	(x)	( )
Is the research design appropriate? Response: Thank you for this comment. We have reviewed our research design and all authors agree that it is appropriate for our study.	( )	(x)	( )	( )
Are the methods adequately described? Response: All requested revisions on the methods section have been made; thank you.	( )	( )	(x)	( )
Are the results clearly presented? Response: We have made revisions on the results as requested; thank you.	( )	( )	(x)	( )
Are the conclusions supported by the results? Response: Thank you for this comment. We have made some revisions to our conclusions, and the conclusions are supported by the findings of our study.	( )	(x)	( )	( )
Are all figures and tables clear and well-presented? Response: We have clearly labeled all our figures and presented tables with footnotes describing any abbreviations used on the tables. We have also placed tables and figures after the paragraphs or sections citing them for easy readability.	( )	(x)	( )	( )

Comments and Suggestions for Authors

The revised article is meaningfully improved but still has a few important problems. The authors did respond to several of my critiques, especially around novelty framing, cross-sectional/retrospective design language, non-probability sampling, STBBI context, and missing data acknowledgment. However, they did not fully resolve the two issues I would still consider the most serious: The ethics-timeline inconsistency and the continued lack of full analytic transparency about denominators/model sample construction. 

Below are my suggestions regarding what remains to be fixed in the manuscript before it is ready for publication: 

1. The ethics-timeline issue remains unresolved. The manuscript still says that recruitment and data collection occurred in 2021, but IRB approval was not obtained until 2022. Unless the authors explain that this was an amendment, a secondary-analysis approval, or some later institutional signoff on an already-approved study, this remains a serious issue.

Response: Thank you again for this keen observation. In fact, ethics approval for this study was obtained on 29/10/2020 form the Public Health Agency of Canada Research Ethics Board. This approval allowed for the collection of data in 2021. However, an amendment to the protocol to include a qualitative component in the study was later made and an approval for this was obtained from the Health Sciences and Science Research Ethics Board of the University of Ottawa on 05/05/2022. We have added a statement on the Ethics consideration section on the manuscript to clarify this.

1. The model interpretation is clearer, but the prose is still a bit loose in some places. For example, the Results still say "factors associated with participants' overall increased experience of discrimination" and "overall participants perceived increased experience of discrimination." This phrasing is awkward and a bit muddy. It suggests a composite outcome without ever fully defining how that summary variable was created across domains. Ultimately, the core binary outcome is more understandable now, but the writing still needs tightening.

Response: Thank you for this keen observation; we have revised this and made the appropriate corrections. 

1. The age finding is still interpreted rather weakly. The authors now identify older age as the only significant adjusted correlate and attempt a brief interpretation, but the explanation is not very persuasive. They write that people over 50 in Canada’s Northern Territories face higher risk of discrimination, yet the model’s significant result is about older age overall, not specifically northern residence, and northern residence was not significant in the adjusted model. So the interpretation feels somewhat mismatched to the actual model results.

Response: Thanks again for this observation; we have now added evidence from other studies supporting that older age is a correlate of discrimination.

1. Some phrasing and grammar issues remain. There are still places where the prose is awkward or imprecise. For example,  "both individually and as a group did not accurately distinguish"; "data was collected" instead of "data were collected"; "other factors that could possibly contribute…needs to be explored” (in both abstract and conclusion). 

Response: Thank you so much for identifying these errors, they have now been corrected on the manuscript.

1. The limitations are stronger now, but the strengths section still says the statistical analysis allowed for an accurate description of the levels of discrimination and associated factors. That still feels too strong given the non-probability design, retrospective change measure, and acknowledged missing data issues.

Response: Thank you so much for pointing this out; that sentence has been revised on the manuscript. 

In sum, the manuscript is significantly improved, but still requires adjustments along the above lines if it is to be published in this journal. 

Response: We have revised the manuscript following your comments; thank you so much.

 

Reviewer 3 Report

Comments and Suggestions for Authors

Well done to the authors for addressing most of the raised concerns. However, a few minor issues remain and require addressing:

1- The presentation of the results is opposite to the norm in scientific publishing. Authors should describe their results, and THEN refer to the tables/figures for further illustration, and not the other way around. In other words, authors should avoid starting their paragraphs with "Table x shows".

2- The subsections and headings under "Forms of discrimination experienced by participants" are unnecessary and should be removed.

3- Tables and figures should be placed immediately after their first reference in the text, rather than being compiled together. This would significantly improve the manuscript readership.

Author Response

Dear reviewer,

Thank you again for taking the time to review the revisions made on our manuscript. Please find the point-by-point responses to your comments below and the corresponding revisions/corrections in track changes in the re-submitted files.

 

Quality of English Language

( ) The English could be improved to more clearly express the research.
(x) The English is fine and does not require any improvement.

Response: Thank you.

 

	Yes	Can be improved	Must be improved	Not applicable
Does the introduction provide sufficient background and include all relevant references? Response: Thank you for this comment. We have reviewed our introduction; all relevant references to the information provided have been cited. We did not find any other information that we could add to this section of the manuscript.	( )	(x)	( )	( )
Is the research design appropriate?	(x)	( )	( )	( )
Are the methods adequately described? Response: All requested revisions on the methods section have been addressed on the manuscript; thank you.	( )	( )	(x)	( )
Are the results clearly presented? Response: Thanks; we have made some revisions on the results section and we think they are now clearly presented.	( )	(x)	( )	( )
Are the conclusions supported by the results?	(x)	( )	( )	( )
Are all figures and tables clear and well-presented? Response: We have clearly labeled all our figures and presented tables with footnotes describing any abbreviations used on the tables. We have also placed tables and figures after the paragraphs or sections citing them for easy readability.	( )	( )	(x)	( )

Comments and Suggestions for Authors

Well done to the authors for addressing most of the raised concerns. However, a few minor issues remain and require addressing:

• The presentation of the results is opposite to the norm in scientific publishing. Authors should describe their results, and THEN refer to the tables/figures for further illustration, and not the other way around. In other words, authors should avoid starting their paragraphs with "Table x shows".

Response: Thank you so much for pointing this out; we have made the corrections on the manuscript. All our tables and figures have been placed after each paragraph or section citing them.

• The subsections and headings under "Forms of discrimination experienced by participants" are unnecessary and should be removed.

Response: We greatly appreciate this suggestion; however, we are presenting the various forms of discrimination in this section. We think it is clearer to list each type of discrimination first, before providing the respective information reported by participants. For this reason, all authors agreed to maintain the listing of the forms of discrimination in this section of the results.

• Tables and figures should be placed immediately after their first reference in the text, rather than being compiled together. This would significantly improve the manuscript readership.

Response: Thank you so much for this suggestion. We have placed Table 2 after presenting the various forms of discrimination and Table 1 also follows the presentation of participants characteristics. Tables 3 and 4 however presents results of the factors associated with increased experiences of discrimination; the variables in Table 4 were selected from the results in Table 3. The authors agreed to place the two tables together to facilitate easy cross-checking of the variables on both tables by readers. With regards to our figures, they are all cited in one paragraph and we think it is better to place them together than break up the paragraph with the figures.