Diabetes‐related complications: Which research topics matter to diverse patients and caregivers?

Abstract Background Diabetes is a chronic disease with increasing prevalence worldwide. Although research has improved its treatment and management, little is known about which research topics matter to people living with diabetes, particularly among under‐represented groups. Objectives To explore the importance of research topics among a diverse range of people living with any type of diabetes or caring for someone living with any type of diabetes. Methods We used a convergent mixed‐method design with quantitative and qualitative aspects. We surveyed a national sample of people living with diabetes and caregivers of people with diabetes, asking them to rate the importance of 10 predetermined important research topics. We also held three focus groups in two major cities to explore research concerns of people who are under‐represented in research. Results 469 adults (57% men, 42% women) in Canada completed the online survey, indicating that all 10 areas of research mattered to them, with the highest ratings accorded to preventing and treating kidney, eye and nerve complications. Fourteen individuals participated in three focus groups and similarly noted the importance of research on those three complications. Additionally, focus group participants also noted the importance of research around daily management. No new topics were identified. Conclusions This study confirmed the importance of research topics among a population of people living with or caring for someone with diabetes. Findings from this study were used to inform the vision for Diabetes Action Canada—a pan‐Canadian Strategy for Patient‐Oriented Research (SPOR) Network on diabetes and its complications.


| INTRODUCTION
Diabetes is a chronic disease with increasing prevalence worldwide. 1 In 2014, an estimated 422 million adults, representing 8.5% of the global population, were living with diabetes. 2 The economic burden of this disease and its complications account for a growing proportion of local and national budgets. 3,4 For individuals, diabetes has negative psychosocial consequences that diminish quality of life. 5 While research has improved the treatment and management of diabetes and increased longevity, 6 mismatches between the focus of research and what matters to patients may lead to research waste. 7,8 Involving patients in the early stages of research is the first step in reducing such waste, as it helps increase the relevance of topics studied regarding such chronic diseases as diabetes. 9 People living with chronic diseases may develop a high degree of expertise which can provide new insight into how to improve their conditions and self-care. 10,11 The patient perspective may complement that of the clinician and researcher by providing a more holistic interpretation of health and the experience of a health condition. 11 When seeking to involve patients as partners in research, it is critical to avoid reproducing or even exacerbating health inequities. Major disparities persist in the diagnosis, treatment, disease management and health outcomes of groups such as ethnic minorities, immigrants, people living in poverty, people whose mental health require regular follow-up with a psychiatrist and seniors, all of whom are more vulnerable to diabetes-related complications. [12][13][14][15][16] In spite of these continuing disparities, minority groups continue to be under-represented in research, and engagement in defining research questions is no exception.
Additionally, there is little guidance on how to facilitate the full participation of members of these groups in setting research priorities. 17,18 For example, although previous research in the United Kingdom has identified research priorities among people living with type 1 diabetes, those involved were predominantly white and female. 19 This study aimed to involve a national sample of people living with diabetes and caregivers of people living with diabetes in rating the importance of research topics around diabetes-related complications.
We further sought to capture the perspectives of people who are under-represented in research. Our primary research question was as follows: What topics are most important to people living with or caring for someone living with diabetes regarding disease-related complications as a means to help better orient future research priorities?

| Study design
We used a multipronged mixed-methods (QUAN + QUAL) approach with a convergent design 20 to capture what was important to people living with diabetes and caregivers regarding research on diabetes-related complications. 21,22 According to the convergent mixed-methods design, quantitative and qualitative methods are complementary during data collection, data analysis or both. In our case, we combined 21 the quantitative and qualitative data after we completed both sets of data collection. The study consisted of two components accordingly: (i) quantitative: an anonymous online survey to poll a national sample of people living with diabetes or caregivers of people living with diabetes on the importance of 10 predetermined research topics; and (ii) qualitative: holding focus groups with people living with diabetes who are members of under-represented groups, in order to explore the views and experiences of those predicted to be under-represented in the online survey. 23

| Research ethics
This study was approved by the Research Ethics Board of the Centre Hospitalier Universitaire de Québec (Quebec City, approval #: 2016-2578). In agreeing to follow the link and take the online survey, participants provided implied consent. No survey questions were mandatory, meaning that respondents could skip questions if they wished. No attention filter was included. Prior to each focus group, we described the project and allowed participants to ask questions.
Verbal consent of focus group participants was recorded.

| Online survey
The online survey included questions on socio-demographics, the person's experience with diabetes-related complications and analog scales to rate the importance of 10 pertinent disease-related complication research topics. These topics were identified in the literature describing previous priority-setting exercises conducted with people living with type 1 diabetes, 19 and via email consultation with researchers, clinicians, representatives of patient organizations, caregivers and patient partners as part of a 6-month funding application planning process.
Demographic data gathered from participants included: age, gender, ethnicity, income and education levels, geographical location and country of birth (inside or outside of Canada). Prior to finalizing the survey, all survey questions were iteratively reviewed by a person living with type 1 diabetes, a person living with type 2 diabetes, and a parent of a child with type 1 diabetes. The survey also contained three validated scales [24][25][26][27][28] to measure fear or distress associated with living with diabetes and its complications (see Appendix S1). These scales were included because we believed that fear or distress might influence how individuals rate the importance of research topics relative to the levels of fear they experience regarding these complications.
If we were to observe large variations in ratings of importance, these data would allow us to explore potential reasons for the variation. The survey also included comment boxes where participants could provide additional information, including an open-ended question asking for their ideas on additional topics concerning diabetes and diabetesrelated complications that require more research.

| Survey participants
Over a 3-day period in September 2015, we recruited participants through Qualtrics online sampling services. 29 To be included in the study, participants had to be living in Canada, aged ≥18 years, living with type 1 or type 2 diabetes, or caring for a child or an adult with diabetes and able to complete the survey in English or French. To ensure demographic diversity and offset variations in response rates, we established desired quotas based on gender (50/50 men and women), type of diabetes and relationship with diabetes (people with diabetes themselves, parents of children with diabetes, caregivers for adults with diabetes). We could not put quotas in place regarding ethnicity due to sampling constraints. In keeping with standard amounts for surveys administered by panel services, participants who completed the survey received $1.00-$1.50 in compensation for their time answering our questions. We aimed for approximately 500 respondents.
This target was selected as an achievable sample size that would allow for a broad sample of respondents and aligned with previous, similar research that sought feedback from 583 people living with diabetes about research questions they would like to see addressed. 19

| Focus groups
Members of some groups may be less likely to complete online surveys, and thus, be under-represented in survey-based research. Therefore, we held 3 focus groups with patients and caregivers who were members of such groups. 30 To ensure variation in perspectives, we partnered with community organizations working with seniors, economically disadvantaged people, immigrants and people whose mental health requires follow-up with a psychiatrist. [31][32][33] Two experienced qualitative researchers (MJD and MDT) conducted the focus groups using an established protocol. During the focus groups, patients were invited to discuss their experience with diabetes and its related complications, their perspectives and their concerns about the long-term complications of diabetes. Participants also explained why, in their view, the concerns raised should be investigated by researchers.

| Focus group participants
We used a convenience sample of members of under-represented populations in the province of Quebec. We recruited focus group participants through three community-based organizations that provide services to seniors, immigrants and people whose mental health requires regular follow-up with a psychiatrist. To be eligible to participate in the focus groups, participants needed to be: living in Canada, aged ≥18 years, living with type 1 diabetes, type 2 diabetes or caring for a person with diabetes, and able to understand and express themselves in French. Participants who were unable to comfortably express themselves in French were excluded from the study.
To recruit participants, the organizations circulated information about the study to its clients or members. Interested participants contacted the research associate either by email or by phone. The research associate contacted all potential participants to explain the study, assess their eligibility, answer questions and discuss logistics. A reminder call and/or email was sent to all participants 2 days prior to the scheduled focus group to confirm the time and location.
We held the three focus groups at times convenient for participants. Furthermore, to increase accessibility, the focus groups were held in the offices of the partnering community organizations; a common practice when working with members of vulnerable populations. 34 We conducted two focus groups in Quebec City: (i) seniors; and (ii) people whose mental health requires regular follow-up with a psychiatrist. We conducted the third in Montreal with a group of immigrants. Each focus group was audio-recorded and lasted between 70 and 90 minutes. Participants received $50 in appreciation for their time and 10$ for transportation. 19

| Data analysis
Our interest in conducting both qualitative and quantitative portions was to ensure inclusion of diverse perspectives. In other words, while research often uses quantitative and qualitative methods to collect different types of data from the same population to inform a research question, we used different methods to collect data from groups both more and less likely to participate in different types of research, in an attempt to capture more representative results. Therefore, we carried out quantitative and qualitative analyses separately before bringing both parts together. Our first step was to conduct descriptive statistics using SPSS version 22 (Armonk, NY, USA: IBM Corp.) to measure central tendency and examine the range of variation in responses to our questions about the importance of 10 important diabetes research areas. We recorded focus group discussions and transcribed them verbatim. We performed a six-stage thematic analysis 35,36 using NVivo qualitative analysis software (QSR International Pty Ltd. Version 10, 2012). We started by generating initial codes and themes, and inductively refining these themes based on the data. MTD analysed focus group data under the guidance of MJD. The codes were labelled with short phrases using the words of participants. Then, MTD sorted codes into potential themes and collated all relevant coded data extracts within the identified themes and subthemes. During this analysis, the codes, themes and subthemes were revised and refined. We used field notes 37 to validate and complete the information gathered during the focus groups. After separate analyses were completed, we combined the findings from each study to analyse how complementary or contradictory they were. We additionally examined how focus group findings could improve our interpretation of the statistical analysis.

| Online survey
Of the 500 participants surveyed, 31 were excluded from our analyses because they either completed the survey in a time deemed too fast to provide thoughtful answers (ie, 10 minutes or less) or because their responses were inconsistent with the questions. The remaining 469 participants were 57% men, had a mean age of 44 (SD = 15), came from across the 10 provinces and 3 territories of Canada, and represented a broad range of educational backgrounds and income levels. In line with our concerns and predictions about representation, participants predominantly identified as White or Caucasian (93%). Participants completed the survey in English (78%) or French (22%) and were either living with diabetes (96%) and/or caring for a child (<1%) or adult with diabetes (3%). Ten percent (10%) of participants were dealing with type 1 diabetes; 89% with type 2 diabetes; and 1% with another or unknown type.
A vast majority of participants with type 1 or type 2 diabetes (45% and 60%, respectively) reported other health concerns, some of which may be diabetes-related complications (see Table S1).
These concerns were, for types 1 and 2 respectively, eye complications (34% and 15% of participants), heart complications (13% and 24% of participants), kidney complications (22% and 8% of participants), mental health complications (34% and 27% of participants) and nerve complications (40% and 30% of participants). Many participants reported not having been screened for these complications in the previous year. Of those with type 1 and type 2 diabetes, respectively, 63% and 78% reported not receiving screening for eye complications within the past year; 71% and 68% reported not receiving screening for heart complications; 53% and 70% reported not receiving screening for kidney complications; 68% and 86% reported not receiving screening for mental health complications; and 61% and 70% reported not receiving screening for nerve complications.

| Focus groups
Of the 23 people who initially expressed an interest in participating in the study, 5 were ineligible because they neither had diabetes nor cared for a person with diabetes; 2 withdrew because they were unavailable on the day of the focus group and 2 withdrew without explanation. Of the 14 remaining individuals who participated in the 3 focus groups, 7 (50%) were female and 3 total (21%) were living with type 1 diabetes. The characteristics of participants are shown in Table 2.

| Online survey
We report here the medians rather than the means because the distri-  and treating kidney, eye, heart and nerve problems. Research topics for which participants had the widest interquartile range in scores were as follows: preventing and treating mental health problems, developing and testing smart insulin, patient and caregiver education, and artificial pancreas research (type 1) (see Table 3).  Table S1).
Comments provided by participants in the open box sections of the survey aligned with the quantitative findings and illustrate the emotional distress linked to diabetes and diabetes management, the fear associated with episodes of hypoglycaemia and its consequences, and with the long-term complications of the disease (see Table 3).

| Focus groups
The thematic analysis allowed us to identify a set of general concerns about diabetes-related complications as reported by members of under-represented groups.

| General concerns about diabetes-related complications
Participants in the focus groups provided further insight into the nature of their concerns about the impact of diabetes on their quality of life, life-expectancy ( In addition to these general and common concerns, four specific themes arose from the focus group discussions: With respect to the second point, discussions in both focus groups focused on the need for studies examining the cost of diabetes treatment (Table 4, citation 9). For example, some participants argued that they sometimes had to choose between paying their rent and buying insulin and complained that this should be a concern to researchers.

Theme 4: The need for better dissemination of the research results on diabetes
Focus group participants also expressed concerns about not having access to updated information on diabetes. They reported being aware of on-going research, but were never informed by community organizations about the research results (Table 4, citations 14 and 15).
Participants also pointed to a need for better information for their loved ones and relatives, to help them understand and provide better support in the management of the disease (Table 4, citations 16, 17, 18). This would represent a miracle in the area of blood sugar management. For an artificial insulin to mimic naturally produced insulin will result in better management of blood sugar levels and reduce dramatically the physical complications associated with them. 43 y old man living with type 1 diabetes.

| DISCUSSION
This would be a boon to older people or persons such as my mother who has macular degeneration and has problems testing and then seeing the results. 48 y old man living with type 2 diabetes.

Theme 1: General concerns about diabetes-related complications
Citation 1: The problem with … any disease that it may, uh … it will get worse if you have diabetes. That's my greatest fear. Ok. my mother is 99 y old, next year she will be 100. And last year, they took her to the hospital and because of her age, we thought: it's over. After a week with the medication, the doctor said, "She can go home.". And, we say: "What do you mean she can go home? ". Yes, because she is not sick, and there are no complications. We gave her the standard treatment for pneumonia and it's gone. But my fear is that anything we catch, those of us who have diabetes, it can complicate things. It is more serious compared to those who don't have diabetes.

F3FG3T2D
Citation 2: Those of us with diabetes are very vulnerable. We can have, have any number of diseases, our organs are not properly supplied with blood sugar … so if we wait too long to eat, uh … there is damage. We start, for example, to experience symptoms such as arthritis … … circulation problems and the worst is … that once the onset begins, once we have arthritis … is too late […]. We can't go back and say: "Treat it. " So the fear is […] we're on a tightrope. We never know when we might develop it. [Laughs] I can't wait for them to invent a machine. They have invented insulin pumps, stuff like that, but will they end up inventing a machine for us to take our blood sugar without always needing to prick our fingers?

F4FG2T1D
Citation 7: And uh … well it's perhaps not related, but I will mention it […] me my mother had diabetes. And uh … I took her to the hospital in a diabetic coma. So I have always been scared of developing a … diabetic coma. So sometimes I ate more, when I worked out, because I was afraid of falling. So I did not pay close attention to my blood sugar, to take it … so sometimes … I raised my blood sugar to ensure that it would not fall too low while I was exercising. women, Indigenous peoples, parents or guardians of children with diabetes, as well as caregivers) were under-represented in the online survey and were absent in the focus groups. Therefore, our sample lacks representation of some other under-represented populations in Canada. Additionally, language barriers may have limited our selection of participants and excluded individuals, particularly those from under-represented groups such as immigrants. Furthermore, because this online survey and focus group based study relied on participant self-reports, the data could be limited by the subjects' ability for introspection, their individual interpretations and social desirability bias. 47 Finally, because this was a preliminary study aimed at exploring the importance of different research topics to those living with diabetes and caregivers in Canada, we did not undertake prioritization activities that require trading-off one priority against another to produce a ranked list. Such activities are planned for future research.
One strength of this study is its use of qualitative and quantitative methods to help capture the experiences of under-represented groups and diverse participants from across Canada. This approach proved feasible as a method for efficiently exploring patients' and caregivers' preliminary views on research topics within a short period of time.

| CONCLUSIONS
This study confirmed the importance of research topics regarding diabetes-related complications within a population of people living with diabetes or caring for someone with diabetes, and further