Social Disability as Disaster: Case Studies of the COVID-19 Pandemic on People Living with Disabilities

Round 1

Reviewer 1 Report

I had enormous difficulty reviewing this paper! It is extremely well written, highly cogent, interestingly argumentative and very persuasive.  It deals with the reactions of many dozens of people who were multiply interviewed during and after after Covid 19 struck the world and their small, somewhat isolated, relatively affluent older communities in the US and Scotland. The authors focused on the impact of Covid 19 and the governmental response to Covid 19 on the social, interpersonal and self-reflective responses of the respondents. The 20-page paper contained no tables, no graphs or other kinds of numerical summary. The paper apparently reflects a consensual group impression of individual researchers who knew the communities well and had available to them an enormous amount of information about each participant. Based on their very nice introductory discussion that contrasted the "medical model" of disabilities that follow natural disasters with a "relational approach", it was clear that the authors' bias favored the relational approach and this paper is written from that perspective. Within that perspective, it is a job well done. Moreover, I believe that their recommendations are well worth listening too and seriously considering.

Author Response

Dear reviewer, 

Many thanks for taking the time to review our paper and for the positive and encouraging comments. It is very much appreciated!

Reviewer 2 Report

Using interviews conducted in two different study locations, this manuscript focuses on social disability as a disaster. The author(s) effectively discuss the two case studies and the study findings to highlight COVID-19 pandemic as a "social disability disaster". The manuscript is well written and easy to comprehend. The inclusion of literature and detail discussion of results makes the paper strong. However, the author(s) can improve the paper by focusing on the following points:

·     Under the methods section, provide some of the questions utilized for the interviews.

·     Currently, the results section is divided into two subheadings (5.1 and 5.2) based on the study area. I recommend the authors to further categorize the results for each case study based on the themes identified by the data analysis. This will help to organize the findings section better. 

· On page 6, the author(s) mention that they identified key themes from each case to develop the theory.  Rather than just presenting the findings in a paragraph format, presenting the key themes in a table format for the two cases can help readers to  identify the themes that resulted in the theory.

Author Response

Dear reviewer, 

Thank you very much for taking the time to review our paper and for your comments and helpful suggestions. 

Here are our responses to your comments and suggestions:

Comment 1: Using interviews conducted in two different study locations, this manuscript focuses on social disability as a disaster. The author(s) effectively discuss the two case studies and the study findings to highlight COVID-19 pandemic as a "social disability disaster". The manuscript is well written and easy to comprehend. The inclusion of literature and detail discussion of results makes the paper strong.

Our Response 1: Thank you for reviewing our paper and for your positive and encouraging comments.

Comment 2: However, the author(s) can improve the paper by focusing on the following points: - Under the methods section, provide some of the questions utilized for the interviews.

Authors’ Response 2: Thanks for the suggestions and guidance for how to further improve our paper. We have now provided further details in the methods section about how in both case studies, the information/data obtained about the pandemic and pandemic interventions from research participants from which our article and theory derives was generated in the course of working on two different research projects that were initially focused on other topics (ageing in place and responses to flooding). We have also explained that in addition to the interview information, the data contained in this study was also obtained in the course of discussions with participants as it emerged in the context of interactions between the researcher and participants during ethnographic field visits to the field site. We have added further these further details to our methodology section (on page 7) and also in response to a comment from another reviewer to make it explicitly clear that this is anthropological, ethnographic fieldwork where the formal interviews were only a part of the full data set - supplementing earlier discussions and later discussions, observations and participation in activities. We have also provided details of the four interview research questions focusing on participant experiences the pandemic at the bottom of page 7 and top of page 8 in the methodology section. These specific questions were: 1. Tell me about your experiences of the first and second periods of lockdown? 2. In what ways did/has the pandemic affect(ed) your regular activities and routines? 3. In what ways did/has the pandemic affect(ed) your relationships with family, friends, carers, and/or others living in the community? 4. How do you feel life has changed for you now compared to before the pandemic?

Comment 3: Currently, the results section is divided into two subheadings (5.1 and 5.2) based on the study area. I recommend the authors to further categorize the results for each case study based on the themes identified by the data analysis. This will help to organize the findings section better.

Authors’ Response 3: Thank you for this suggestion for ensuring that we draw out the emergent themes more clearly and increase their visibility to the reader. For this we have now categorized the results section by breaking each of the two case studies down into further sub-sections corresponding to five overlapping themes identified in each of the case studies: These sub-headings are disruption, coping with disruption, short term impact, longer term impact, and ‘silver linings.’ We have also added an additional subsection to the results section after the two case studies that focuses on similarities and also divergences in each of the two case study sites and how these points of convergence/divergence lead to our theory of social disability, including its drivers, how it is experienced, factors that influence its risk, as well as what influences how it may be offset or resisted.

Comment 4: On page 6, the author(s) mention that they identified key themes from each case to develop the theory.  Rather than just presenting the findings in a paragraph format, presenting the key themes in a table format for the two cases can help readers to identify the themes that resulted in the theory.

Authors’ Response: Thanks for your suggestion about how to improve the presentation of the key themes. However, we feel that inclusion of a table based on the interview findings alone would risk turning the undermining the totality of the findings from a qualitative, ethnographic project into a format that seems more appropriate for a more quantifiable project, or a project entirely based on formal interview findings. In our project we did not calculate results on a numerical basis to develop the theory with the theory, but rather, as we now explain in the methods section, looked for commonality of themes and points of convergence within these thematic categories from the data collected for each case study site. However, in line with another reviewer’s suggestion about drawing readers’ attention to the themes and points of convergence and divergence between the two case studies which led to the development of the theory, we have moved the theory section so that it now appears before the methodology section and added a new sub-section in the findings that highlights the thematic similarities and difference between the two case studies and shows how the theory was derived. This helps draw readers attention to the commonalities that underpin the theory of social disability and improves clarity and visibility as to how the key themes are embedded in the case studies in a way that is more consistent with stylistic conventions for a multi-method qualitative, ethnographic social anthropological project.

Once again we would like to thank you for your helpful and encouraging comments and suggestions for improving our work. 

Reviewer 3 Report

Comments for author File: Comments.pdf

Author Response

Dear reviewer, 

Thank you for taking the time to review the first version of our paper. We really appreciate your extensive, insightful comments and suggestions to help us improve the paper. We have now addressed each of your comments and revised the paper accordingly. 

Below is a breakdown of how we have addressed each comment and suggestion. 

Comment 1: This is an interesting and relevant paper which adds to the literature but is limited by methodology. It is well written (I spotted just a few minor typos) and is an easy read. There are some areas where the presentation and argument could be strengthened.

Authors’ Response 1: Thank you for reviewing our paper and for your encouraging comments. We have proofread the paper and hopefully now removed any remaining typos. Thank you for also helping us to strengthen our paper with your helpful suggestions. We have made addressed all your suggestions in our revised version of the paper.

Comment 2: Methodology – obviously the number of interviews is extremely small and therefore quite sweeping statements are drawn from very few interviews. Methodologically the research could be challenged as more akin to a pilot. Why weren’t online questionnaires used with a larger sample of people to test out some of the findings from the interviews? Why not comparator groups/locations – e.g. people not already experiencing social disability? Basically, the evidence base is very small upon which to draw often quite broad conclusions (albeit interesting). This needs more explicit recognition. In the Scotland case – the interviews were across people aged 29-93, but it is not clear how the ages translated to the quotes/examples used. Further, given the participants were all known to the researcher from previous work, I would expect to see some comment on bias in selection and whether familiarity may have influenced the engagement with the researcher.

Authors’ Response 2: Although the number of interviews is small the interviews were used to supplement ethnographic anthropological research (U.S Case study) and were undertaken in addition to online and in-person ethnographic research (Scottish case study). The findings are therefore based on the findings from more than the interviews alone. We have now expanded our methods section we know also clarify this by explaining that as this is an ethnographic, anthropological study we were not explicitly seeking or testing the generalisability of the findings but were interested in drawing out similarities from the two case studies from participants’ lived experience to develop a theoretical contribution to help explain the impacts of Covid on people’s live, using a ground-up approach – clarifying that we developed the theory from the observed and/or reported lived experiences of research participants.  This is at the bottom of page 6 and on page 7. He we also reference relevant methodological literature to support our research approach. However, we have also said in the further study section on page 25 in the conclusion that the theory we developed could ultimately be tested across comparative groups and locations to see if it is more generalisable and to further develop and refine the theory accordingly, even though studies such as these do not aim for generalizability as such.

We also explain how our study arose from two case studies that initially sought to explore other research topics but that owing to the impact of the pandemic upon people’s lives during this time the case studies were carried out, led to the emergence of the data contained within the findings and in the Scottish case study led to significant shift in the focus of the project. Therefore, participation was limited to those who were already engaged research in each of the two different projects. In the first case study we also explain how the original aim was for actually for a follow up study with participants who had been involved in an earlier study before the pandemic. We also explain that the pandemic placed limitations on ability to network and to meet with participants to further expand the study and focus on comparator groups. We explain this on pages 7 and 8. We also now clarify by making it explicitly clear in the findings for the first case study which had a broad age range of participants the points where divergences were evident between those aged over 60 (older adults) and under 60, and also made references to where divergences were also explicitly evident for those aged under 40.

We also now explain in the methods section that although participants and the researcher in the first case study had known each other for a number of years, this ultimately proved to be beneficial within the context of the research rather than introduce a source of harmful bias. We explain that familiarity with the researcher and prior engagement in research meant rapport had al-ready been established and participants were already comfortable talking experiences of living with a disability and experiencing hardships with the researcher, including in an interview context. We also explain that this level of comfort and willingness to challenge the researcher was evidenced by how it was the participants themselves that initially raised the issue of the pandemic and, by doing so, ultimately steered the focus and development of the research towards its eventual outcome. We also note that the research aimed to capture the subjective experiences and views of participants, rather than objective, unbiased forms of data and that acknowledgement of people’s perspectives, values, and agency in the process of conducting research, as well as in the reporting of the findings, is also particular important within disability research for enhancing inclusion and helping rebalance uneven power dynamics between the researcher and research subjects. We do this in the second paragraph on page 8. We also explain that although participation was limited to those who were already engaged in research and had been engaged in an early study, engagement in this study was open to anyone living with any disability within the local community and were not selected by the researcher according to specific criteria (such as being of a specific age or living with a specific type of disability).

Comment 3: The US case has to stand alone from the Scotland case as the participants were in a very different environment – not only residential (with personal choice to be there) but also in terms of the official Covid response. Also, not such a large age range? While the results and analysis are presented separately, it would be useful to more explicitly pull-out commonalities and differences in response. No evidence of socio-economic differences is presented. Those with more personal resources being able to adapt more effectively? No evidence of the impact of family (either living with the individuals or in close proximity) is presented in terms of impact on social disability. While death rates in the US residential community are referred to the same would be useful in the Scotland community.

Authors’ response 3: As we explained above, in our study we focused on the issue of people living with a disability in two different residential environments, rather than upon age per se. We now explain in the methods section that the findings from the comparative analysis of our two case studies from which we developed our theory converged and was cantered upon the theme of disability. We do this in the final paragraph on page 8 where we acknowledge the differences in age ranges between the population samples. We have also added information about socio-economic circumstances and differences between participants in both case study sites (pages 9 and 16) and explain in our findings (on page 12) and discussion that ultimately, being financially better off in either case study, appeared to do very little to offset social disability or reduce or enhance the risks (bottom of page 21). We also include evidence of how having family in close proximity (but not in the same household) in the first case study site was actually associated with enhanced feelings of isolation during lockdown and that those with greater numbers of social connections before the pandemic also experienced a heightened sense of isolation owing to how restrictions required them to change their normally more social ways of living. We also explain that living with others in the same household did not help to offset social disability as it increased dependency on familial care providers and that household became more self-contained and distanced from others (on page 12). We also provided details of the death rates in Scotland information about case numbers from the actual official data as well as details of how this contrasted with what participants perceived was happening. We do this on page 14 in the second paragraph.

Comment 4: Care is needed in quite general statements without sufficient attention to how many people were commenting as suggested – e.g. page 11 – lines 564-571; page 12 – lines 586-590; page 13 – lines 667-669. I appreciate that this is due to be the limited empirical evidence collected but there is a need to be careful in terms of the presentation. Otherwise, there is a danger that pre-conceived views of responses are presented.

Authors’ response 4: Thank you for this comment and helpful warning. We have now hopefully offsetted this by providing further detail about numbers where appropriate in the reporting of the findings and also explained in the methods section that our findings are based on more than the interview data – also based on prolonged interaction and discussions with participants as part of a wider ethnographic study – and as such, our reporting of our findings (including what is evidenced from the interviews) is consistent with our observations from sustained personal interaction and discussion (including online). We do this by explaining that in ethnographic research, the researcher is the ‘interview instrument’ and uses multiple methods of recording and multiple data points through extended time in the field, supplemented with interviews (page 7, paragraph 2).

Comment 5: In the future research section, it would be relevant to consider how future research in relation to a pandemic might be pre-designed in order to be of high quality. UK evidence suggests that much Covid-related research was fragmented and therefore lost opportunities in terms of robust evidence gathering. Given the length of time of a pandemic, more explicit attention to how experiences and responses change (or not) with time with consequent impact on official responses and communication is important. Further, evidence in the UK suggests that ethnic minority communities who can have very different social interactions (e.g. through extended family living; faith groups etc etc) may need different communication modes and information (e.g. https://royalsocietypublishing.org/doi/epdf/10.1098/rsta.2023.0129). Here the evidence suggests that official responses and interactions need to be responsive to these engrained social and cultural characteristics (as the authors suggest on page 17 in the social disability context).

Authors’ response 5: We completely agree with this as the pandemic took us all by surprise. We have added these excellent suggestions to the future research section. Specifically, at the bottom of page 24 we have said that one option  for future study would be to consider how new approaches and methods for conducting research with more technologically excluded populations, such as older people and people with disabilities, may be devised in anticipation of future pandemics or widespread disruptions, such as by enhancing opportunities for citizen science engagement for members of these communities to participate in the research process. We explain that this could help enhance research by helping to capture peoples’ experiences of events as they unfold rather than retrospectively, which could help to inform the development of interventions at an earlier stage to help reduce iatrogenic effects. We also explain this would also help to enhance inclusion of marginalized groups in research and reduce uneven power dynamics between the researchers and research subjects. We also say on page 25 in the final paragraph that another option could be to explore experiences of the pandemic within other U.S and UK communities with greater ethnic diversity to test and further refine our theory of social disability within the wider national contexts. We explain that this is especially important given that evidence in both the UK and U.S suggests that different ethnic minority groups may have different, culturally specific ways of interacting socially, which may result in different experiences of social disability and necessitate different forms of intervention to offset risk. We explain that this would also help ensure that future disaster interventions are ethnically inclusive in addition to ability inclusive.

Comment 6: Type of disaster – firstly, I think section 4 should come before 3 as it should lead to the methodology.

Authors’ response 6: Thank you for this suggestion. We have now changed the order of the two sections accordingly.

Comment 7: The important disaster element of Covid was its duration (over many, many months of direct impact) and also the global occurrence and reporting. The paper refers to floods and hurricanes as comparators but these are relatively location limited and also the actual impact (as opposed to the recovery periods) can be of relatively short duration (hours to days). The numbers of people directly impacted can be much smaller. With floods a very large literature shows that residents can adapt relatively quickly, as they are not learning anew and have often learnt how to cope. With floods there are frequently social systems set up in relevant communities to help people to respond. Floods of course in the west rarely kill so the risk is primarily loss or damage to property and physical interventions (barriers etc) are put in high risk areas. Whereas a pandemic is almost a once in a life-time experience that kills hundreds of thousands of people. Therefore, I am not entirely convinced that these different hazards and risks are directly comparable.

Authors’ response 7: Thank you for this comment. We completely agree about the differences between disasters caused by anthropogenic climate change (such as floods) on communities and the pandemic disaster. We have now also added more information in the introduction to clarify that we do acknowledge how different they are (Intro, paragraph 2, page 1). However, we also clarify that we are not actually trying to directly compare the two, but rather that what we are saying is that understanding the impacts of the pandemic and its interventions yields important lessons and considerations for dealing with increasing scale, frequency and severity of disasters, where local adaptational knowledge and methods as well as external resources become overwhelmed and exhausted. Thus, understanding of the pandemic helps us to think about how we can devise interventions that are responsive to how disasters are becoming more widespread and compressed. We do this in the final paragraph of page 1 and the first paragraph on page 2, referencing appropriate studies were possible – to strengthen our argument of the relevance of the pandemic for disaster risk reduction more broadly.

Comment 8: Covid responses/controls - Interestingly little reference is made to social distancing rules and their impact and none to mask wearing and – e.g. carers visiting individuals in the Scotland case would certainly have been taking a number of pre-cautions, some of which like mask wearing can in and of themselves create a sense of separation as well as reassurance. The only example given of responses is in relation to someone becoming terrified of such a visitor- but why?  

Authors’ response 8: We have now added some extra information about responses to mask wearing and social distancing in terms of their impacts in creating both a sense of security and separation in both case study findings (pages 11, top of page 13, and on page 18). We also clarify that in the example to which this comment refers (bottom of page 12). this was due to concerns about risk of contagion and that testing procedures were not being followed or were accurately reported and about whether people involved in giving care were now breaking rules after the initial first few months.

Once again, thank you very much for your helpful comments and suggestions. We hope that the revisions we have made to the paper based upon your comments are to your satisfaction. 

Round 2

Reviewer 3 Report

This is much improved and thank you to the authors for working through the comments diligently. I dont think that based on the evidence available it can at this stage be improved further. 

There are just a few small typos so worth a final edit.