Healthy Eating and Active Lifestyles for Diabetes (HEAL-D): study protocol for the design and feasibility trial, with process evaluation, of a culturally tailored diabetes self-management programme for African-Caribbean communities

Introduction Black British communities are disproportionately burdened by type 2 diabetes (T2D) and its complications. Tackling these inequalities is a priority for healthcare providers and patients. Culturally tailored diabetes education provides long-term benefits superior to standard care, but to date, such programmes have only been developed in the USA. The current programme of research aims to develop the Healthy Eating and Active Lifestyles for Diabetes (HEAL-D) culturally tailored T2D self-management programme for black British communities and to evaluate its delivery, acceptability and the feasibility of conducting a future effectiveness trial of HEAL-D. Methods and analysis Informed by Medical Research Council Complex Interventions guidance, this research will rigorously develop and evaluate the implementation of the HEAL-D intervention to understand the feasibility of conducting a full-scale effectiveness trial. In phase 1, the intervention will be developed. The intervention curriculum will be based on existing evidence-based T2D guidelines for diet and lifestyle management; codesign methods will be used to foster community engagement, identify the intervention’s underpinning theory, identify the optimal structure, format and delivery methods, ascertain adaptations that are needed to ensure cultural sensitivity and understand issues of implementation. In phase 2, the intervention will be delivered and compared with usual care in a feasibility trial. Process evaluation methods will evaluate the delivery and acceptability of HEAL-D. The effect size of potential primary outcomes, such as HbA1c and body weight, will be estimated. The feasibility of conducting a future effectiveness trial will also be evaluated, particularly feasibility of randomisation, recruitment, retention and contamination. Ethics and dissemination This study is funded by a National Institute of Health Research Fellowship (CDF-2015-08-006) and approved by National Health Service Research Ethics Committee (17-LO-1954). Dissemination will be through national and international conferences, peer-reviewed publications and local and national clinical diabetes networks. 
Trial registration number NCT03531177; Pre-results.

tailored intervention implies to a certain degree that patients of African-Caribbean origin are regarded as a homogenous group which is distinct from the homogenous group of the average population. There is huge body of literature which criticizes concepts of culture based on nationality/ancestry and emphasizes the importance of understanding diversity within one "culture". Numerous other factors might be more important for health-related behaviours than "culture". This might be especially important if a migrant already lives for a long time or in the second/third generation in the other country (it seems that this is not at all considered in the study). I think the paper would be more balanced if the authors stated that they are aware of this debate and describe how the intervention will be tailored if it turns out during the development phase that cultural issues are not a major barrier.
 Some epidemiological data to describe Afro-Caribbean communities in England would be interesting. What's their proportion of the general population, how many are in first/second/third generation in England, education etc.

Purpose and aims:
 It is not really clear to me which is the intervention to be evaluated and which parts of the intervention are already set and where is room for tailoring. It took me a while to understand that it is already defined that the Heal-Dprogramme will consist of an educational session in a group setting with an already elaborated curriculum (page 15/16). The aim of the study is to culturally tailor the content of this curriculum? If this is correct I would make this clear already in the introduction section and describe the evidence-base of this pre-set curriculum. Will other interventions, e.g. to engage key stakeholdersbe part of the Heal-Dprogramme and thus of the evaluation or will they be engaged only during the development phase? Methods and analysis  The part about phase 1 is very long and difficult to follow. It could be shortened or structured with more headlines to make it better readable.  Line 131-133: There is an undefined reference in brackets (REF NICE 2014). I think the sentence is grammatically not correct.  Line 316: It would be interesting to know what is "usual care" for diabetes in England? Are there for example Disease Management Programmes?  Line 331: Who will lead the educational sessions? Members of the study team?
 Line 338: Who will conduct the assessment, the study team?  The planned start date of the intervention period should be stated  Sample size calculation: The authors did not perform a sample size calculation since the study is planned as a feasibility study. Yet it would be interesting to know which effect size is needed to detect group differences with 120 patients.

Tables and figures:
 Figure 2 is a table and should be named as such  Figure 3 is (2) a feasibility trial and process evaluation of, a culturally-tailored diabetes self-management program targeting Afro-Carribean individuals with type 2 diabetes. This represents an important body of work that may result in substantial benefit for a community experiencing disparities of type 2 diabetes incidence and control.
Major Concerns: 1. My most pressing concern is that this manuscript primarily describes the planned methodology for the development process of an intervention, rather than the evaluation of an intervention. The authors have included tentative plans for intervention implementation and evaluation, but many aspects of the intervention content and evaluation are yet to be determined. Thus, I find the current title and abstract of the manuscript to be inconsistent with its contents, and both should be modified to describe the manuscript more appropriately. I am unsure whether it is common practice for BMJ Open to publish manuscripts describing the methodology of an intervention trial before the intervention has been fully developed. Further, while the intervention development process described appears to be rigorous and well-founded (and could be of utility to other researchers), I am not sure of the utility to readers of a methods paper describing a feasibility trial and process evaluation before core elements of the intervention (e.g., setting, length, content, cultural tailoring procedures) have been established. Basically, I am unsure whether the second half of this paper is ready for publication at the current time. It makes sense to me for authors to complete the intervention development process, and THEN publish a methods paper which describes the development process but can also describe in more detail the intervention itself and plans for evaluation. I will defer to the editors regarding whether this issue is problematic enough to reject the paper or recommend that the authors to publish the development methods separately. Because I strongly believe in the potential and importance of the work the authors are undertaking, I do hope there is a way to successfully publish a revised version of this manuscript in BMJ-Open.
2. Authors state, "The theoretical basis for HEAL-D will be identified through two processes; firstly an evidence synthesis of key themes through published literature relating to adapting health promotion interventions for ethnic minority groups, and secondly through new primary research." The former of these two seems to be something that the authors can and should complete now, to include in the publication of this article. Authors continue on page 8 to cite some important methodological themes and findings from prior research, but these are not theories.
3. Relatedly, while I absolutely understand the important of the codesign process for determining optimal intervention implementation methods, it is not clear to me how the authors expect the co-design procedures to expand the theoretical basis on which the intervention will be developed.
4. The process evaluation section needs to be carefully revised for clarity and detail. Throughout this section, authors are not specific about how their determinations of evaluated domains (e.g., high versus low acceptability or feasibility) will be made. As is, this section remains quite vague. Typically, in a methods paper you want to give enough information do that other researchers could replicate what you have done (or are planning to do).
5. Authors need to find a way integrate Tables 1-3 and improve their readability. All three tables provide related information about the process evaluation, but how they are separated is confusing and not intuitive. It would more clear and efficient and less repetitive to see the planned process evaluation (domains, data source, evaluation method) in one table. Further, please include concise but specific information about how each element will be evaluated. Currently authors list data sources but (as mentioned above) authors do not say how determinations of high or low acceptability, for example, will be determined. Finally, please label tables appropriately with numbers and title (or as specified by the journal), and clarify why there appears to be repeated content in the main manuscript document and the supplementary tables.
Minor points: 1. For your citation of the percentage of the NHS budget and amount dedicated to T2D (in the introduction), is there a more updated estimate than 2011? I imagine costs have increased since then.
2. Be careful with your verb tenses and be consistent throughout (e.g., our intervention "is designed" vs. "will be designed") 3. Do African-Carribean individuals exhibit worse diabetes control only at the time of diagnosis? Or in general/over time? If in general, you might state that instead.
4. It would be helpful in your introduction to include further background and demographics for readers regarding the African-Caribbean communityhow many African-Caribbean individuals are there currently/what percentage of the UK population, whether this a growing demographic, whether they experience other health disparities or major structural barriers, etc. Also please be specific about the younger age at which T2D is diagnosed… at what age is first diagnosis common?
5. Please spell out any acronyms before they are used (such as "MRC" Complex interventions framework)and it would be helpful to explain this framework as well, or show a visual representation in a figure.
6. It does not appear that all in-text references have been appropriately formatted (see p. 7 line 132).

Please explain what a Service User
Group is or is meant to be.
8. It would be helpful to show the COM-B Framework and Behavior Change Wheel in a figure, if possible.
9. Authors alternate between the term "enablers" and "facilitators"are these synonyms? Please make it clear if they are synonyms, or if not, what is the distinction.
10. This sentence was unclear to me: "Themes that do not map clearly onto the COM-B framework will also inform the programme theory e.g. contextual themes at the community and health system levels." Please explain further.
11. How will you handle disagreements/differences in opinion in the co-design workshops, if participants do not manage to achieve consensus? Will stakeholders have the time and resources to be able to attend 3 half-day workshops?
12. It appears that there is an incomplete sentence in line 311.
13. Did authors consider an attention control instead of usual care? Is it possible that intervention effects may in part be related to increased attention and group social support?
14. Are there any planned adaptations for low literacy or numeracy, if these are concerns in this population? 15. Have the proposed self-report measures been validated in Afro-Carribean populations? If not, have you considered including them in the co-design workshops, for review of cultural relevance and comprehensibility by community members?
16. This sentence was particularly unclear to me: "Sustainability will be considered by assessing the scope for the intervention to be embedded within current care pathways, and contextual factors that may influence decision-making around continuance." (p. 18). Please clarify and be specific about what you mean by scope and continuanceand how an assessment regarding sustainability will actually be determined.

VERSION 1 -AUTHOR RESPONSE
REVIEWER 1. This represents an important body of work that may result in substantial benefit for a community experiencing disparities of type 2 diabetes incidence and control. RESPONSE: We thank the reviewer for their support for our work.
My most pressing concern is that this manuscript primarily describes the planned methodology for the development process of an intervention, rather than the evaluation of an intervention. The authors have included tentative plans for intervention implementation and evaluation, but many aspects of the intervention content and evaluation are yet to be determined. Thus, I find the current title and abstract of the manuscript to be inconsistent with its contents, and both should be modified to describe the manuscript more appropriately. RESPONSE We apologise for this. We have amended the manuscript title and abstract to include more reference to the development work. Due to our focus on communities that are normally neglected, there is a lot of emphasis on our development work, however, we have ensured that the section describing our feasibility trial includes all the measures that are recommended for feasibility trials (NIHR 2017).

2.
I am unsure whether it is common practice for BMJ Open to publish manuscripts describing the methodology of an intervention trial before the intervention has been fully developed. Further, while the intervention development process described appears to be rigorous and well-founded (and could be of utility to other researchers), I am not sure of the utility to readers of a methods paper describing a feasibility trial and process evaluation before core elements of the intervention (e.g., setting, length, content, cultural tailoring procedures) have been established. Basically, I am unsure whether the second half of this paper is ready for publication at the current time. It makes sense to me for authors to complete the intervention development process, and THEN publish a methods paper which describes the development process but can also describe in more detail the intervention itself and plans for evaluation. I will defer to the editors regarding whether this issue is problematic enough to reject the paper or recommend that the authors to publish the development methods separately. Because I strongly believe in the potential and importance of the work the authors are undertaking, I do hope there is a way to successfully publish a revised version of this manuscript in BMJ-Open. RESPONSE Thank you for your support for our work, we are delighted that you recognise the potential and importance of our study. We aim to publish this protocol to describe the entirety of our planned work; we plan to subsequently publish an intervention development paper that details the findings of our co-design work.

3.
Authors state, "The theoretical basis for HEAL-D will be identified through two processes; firstly an evidence synthesis of key themes through published literature relating to adapting health promotion interventions for ethnic minority groups, and secondly through new primary research." The former of these two seems to be something that the authors can and should complete now, to include in the publication of this article. Authors continue on page 8 to cite some important methodological themes and findings from prior research, but these are not theories. RESPONSE Thank you for highlighting this, we agree with your comment. We have amended this section accordingly: in lines 179-189 we describe methodologies from the literature, and in lines 198-201 we have described theories which have been identified in the literature and which will inform our intervention.

4.
Relatedly, while I absolutely understand the important of the co-design process for determining optimal intervention implementation methods, it is not clear to me how the authors expect the co-design procedures to expand the theoretical basis on which the intervention will be developed. RESPONSE We apologise if this is not clear. We have added information in lines 193-195 & 203-208 to try and clarify that our co-design work will help us to explore whether the themes from the literature are relevant to the UK context.

5.
The process evaluation section needs to be carefully revised for clarity and detail. Throughout this section, authors are not specific about how their determinations of evaluated domains (e.g., high versus low acceptability or feasibility) will be made. As is, this section remains quite vague. Typically, in a methods paper you want to give enough information do that other researchers could replicate what you have done (or are planning to do). RESPONSE Thank you for this comment. We have made extensive revision to the process evaluation section, pages 18-20, to improve the clarity and detail. We have provided more detail of our methods in a revised table (now Table 2).

6.
Authors need to find a way integrate Tables 1-3 and improve their readability. All three tables provide related information about the process evaluation, but how they are separated is confusing and not intuitive. It would more clear and efficient and less repetitive to see the planned process evaluation (domains, data source, evaluation method) in one table. Further, please include concise but specific information about how each element will be evaluated. Currently authors list data sources but (as mentioned above) authors do not say how determinations of high or low acceptability, for example, will be determined. Finally, please label tables appropriately with numbers and title (or as specified by the journal), and clarify why there appears to be repeated content in the main manuscript document and the supplementary tables. RESPONSE We have revised our tables and integrated the tables into one process evaluation table (Table 2). We have provided more detail about our data sources and evaluation methods, and ensured there is no repetition or duplication between the main manuscript and the tables. We have provided more detail about how we will evaluate our quantitative data e.g. intervention acceptability, however our data will be largely qualitative and we will be interested mainly in understanding whether further intervention adaptations are needed rather than dichotomising these domains.
For your citation of the percentage of the NHS budget and amount dedicated to T2D (in the introduction), is there a more updated estimate than 2011? I imagine costs have increased since then. RESPONSE We have expanded this part of the introduction, line 63; unfortunately there is not a robust updated estimate of actual costs available but we have cited a forecast for the increasing costs.

2.
Be careful with your verb tenses and be consistent throughout (e.g., our intervention "is designed" vs. "will be designed") RESPONSE Apologies, these have been corrected throughout.

3.
Do African-Carribean individuals exhibit worse diabetes control only at the time of diagnosis? Or in general/over time? If in general, you might state that instead. RESPONSE We have expanded this part of the introduction to show that greater management is needed also, line 69.

4.
It would be helpful in your introduction to include further background and demographics for readers regarding the African-Caribbean communityhow many African-Caribbean individuals are there currently/what percentage of the UK population, whether this a growing demographic, whether they experience other health disparities or major structural barriers, etc. Also please be specific about the younger age at which T2D is diagnosed… at what age is first diagnosis common? RESPONSE Thank you for raising this, we have added in extra information to provide background to the UK context, lines 67-68, 73-87.

5.
Please spell out any acronyms before they are used (such as "MRC" Complex interventions framework)and it would be helpful to explain this framework as well, or show a visual representation in a figure. RESPONSE We have ensured all acronyms are now spelt out in full and added a figure of the MRC framework, figure 1.

6.
It does not appear that all in-text references have been appropriately formatted (see p. 7 line 132). RESPONSE Thank you for pointing this out, this has been corrected.

7.
Please explain what a Service User Group is or is meant to be. RESPONSE We have added detail on this, lines 260-265.

8.
It would be helpful to show the COM-B Framework and Behavior Change Wheel in a figure, if possible. RESPONSE We have added a figure to show the COM-B framework and Behaviour Change Wheel, figure 3.

9.
Authors alternate between the term "enablers" and "facilitators"are these synonyms? Please make it clear if they are synonyms, or if not, what is the distinction. RESPONSE We have corrected our terminology to ensure we use 'facilitators' throughout.

10.
This sentence was unclear to me: "Themes that do not map clearly onto the COM-B framework will also inform the programme theory e.g. contextual themes at the community and health system levels." Please explain further. RESPONSE We apologise if this was not clear. We have expanded our description, lines 276-280, to provide more detail.

11.
How will you handle disagreements/differences in opinion in the co-design workshops, if participants do not manage to achieve consensus? Will stakeholders have the time and resources to be able to attend 3 half-day workshops? RESPONSE We have added detail to this section, lines 287-291. 12.
It appears that there is an incomplete sentence in line 311. RESPONSE Apologies, this has been amended.

13.
Did authors consider an attention control instead of usual care? Is it possible that intervention effects may in part be related to increased attention and group social support? RESPONSE Thank you for raising this interesting suggestion. We gave much consideration to our control, particularly whether it should be usual care or a standard education programme. The literature suggests that group social support may be important for the communities we are working with so we expect this to be part of the theoretical basis of our intervention. This would make it difficult to have an attention control.

14.
Are there any planned adaptations for low literacy or numeracy, if these are concerns in this population? RESPONSE We have added some detail to cover this point, line 287.

15.
Have the proposed self-report measures been validated in Afro-Carribean populations? If not, have you considered including them in the co-design workshops, for review of cultural relevance and comprehensibility by community members? RESPONSE Thank you for raising this important point. There are a lack of validated questionnaires for the communities we are working with and one of the aims of our work will be to look at how well the questionnaires work with our participants. One of the roles of our service user group is to give us feedback on the methods/tools we are proposing to use. We have added this to the manuscript, lines 263-265.

16.
This sentence was particularly unclear to me: "Sustainability will be considered by assessing the scope for the intervention to be embedded within current care pathways, and contextual factors that may influence decision-making around continuance." (p. 18). Please clarify and be specific about what you mean by scope and continuanceand how an assessment regarding sustainability will actually be determined. RESPONSE This section has been rewritten in response to earlier comments.
REVIEWER 2 This is a very comprehensive study protocol with a very complex approach of intervention development. RESPONSE We thank the reviewer for their support for our work.
Some passages need clarifications: Abstract: • Information about the planned RCT is lacking. RESPONSE We have indicated that the feasibility study will inform the design of a future RCT but at this stage we cannot provide any further detail.
• See also my comment on purpose and aim. It might be necessary to make clear that the core curriculum of the HEAL-D-programme already exists and that the content will be tailored (if I understood correctly). RESPONSE Thank you for raising this. We have amended the abstract to include these details.
• Strength and limitations: Limitations of the study are not at all mentioned in the whole paper. RESPONSE Thank you for this recommendation. We have added a discussion to the end of the paper to detail our perceived strengths and limitations. Introduction: • The authors describe that the prevalence of T2D is 3 times higher in Black-British communities with poorer outcomes. It would be interesting to know if anything is known whether this is mainly due to worse health behaviour and worse access to health care (the target of the study) or which role the genetic risk plays. RESPONSE We have added this detail to the introduction, lines 70-72.
• The aim of the authors is to develop a cultural-tailored intervention. It would be worth to reflect on the definition and role of culture in a few sentences. Developing a culturally tailored intervention implies to a certain degree that patients of African-Caribbean origin are regarded as a homogenous group which is distinct from the homogenous group of the average population. There is huge body of literature which criticizes concepts of culture based on nationality/ancestry and emphasizes the importance of understanding diversity within one "culture". Numerous other factors might be more important for health-related behaviours than "culture". This might be especially important if a migrant already lives for a long time or in the second/third generation in the other country (it seems that this is not at all considered in the study). I think the paper would be more balanced if the authors stated that they are aware of this debate and describe how the intervention will be tailored if it turns out during the development phase that cultural issues are not a major barrier. RESPONSE We thank the reviewer for this important point, which we agree with. We have tried to draw on this in both the introduction, lines 109-122, and the discussion, lines 496-502.
• Some epidemiological data to describe Afro-Caribbean communities in England would be interesting. What's their proportion of the general population, how many are in first/second/third generation in England, education etc. RESPONSE We have added more detail to the introduction, lines 73-87.
Purpose and aims: • It is not really clear to me which is the intervention to be evaluated and which parts of the intervention are already set and where is room for tailoring. It took me a while to understand that it is already defined that the Heal-D-programme will consist of an educational session in a group setting with an already elaborated curriculum (page 15/16). The aim of the study is to culturally tailor the content of this curriculum? If this is correct I would make this clear already in the introduction section and describe the evidence-base of this pre-set curriculum. Will other interventions, e.g. to engage key stakeholdersbe part of the Heal-D-programme and thus of the evaluation or will they be engaged only during the development phase? RESPONSE We apologise for our lack of clarity. We have amended the manuscript to make this point clearer, lines 129-132, 141-142, 153-170. Methods and analysis • The part about phase 1 is very long and difficult to follow. It could be shortened or structured with more headlines to make it better readable. RESPONSE This part has been revised and some information tabulated to make it more succinct and easier to follow.
• Line 131-133: There is an undefined reference in brackets (REF NICE 2014). I think the sentence is grammatically not correct. RESPONSE Apologies, this has been corrected now.
• Line 316: It would be interesting to know what is "usual care" for diabetes in England? Are there for example Disease Management Programmes? RESPONSE We have added these detail to the introduction, lines 92-97.
• Line 331: Who will lead the educational sessions? Members of the study team? • Line 338: Who will conduct the assessment, the study team? RESPONSE We have added these details, lines 365-367, 385.
• The planned start date of the intervention period should be stated RESPONSE This has been added to the manuscript, line 150.
• Sample size calculation: The authors did not perform a sample size calculation since the study is planned as a feasibility study. Yet it would be interesting to know which effect size is needed to detect group differences with 120 patients. RESPONSE Thank you for this point. Unfortunately we have little data upon which to base a sample size calculation as very few studies to date have engaged UK African-Caribbean patients so we don't know if they are comparable to the general population in terms of HbA1c, body weight etc.

Tables and figures:
• Figure 2 is a table and should be named as such • Figure 3 is rather a box or a table RESPONSE These have been revised and corrected.

REVIEWER
Cornelia Straßner University Hospital Heidelberg Department of General Practice and Health Services Research REVIEW RETURNED 12-Nov-2018