Abstract
Purpose:
The purpose of the study was to explore various forms of diabetes self-management education (DSME), including group and individual sessions, for persons with lived experiences of homelessness (PWLEH) in Canada.
Methods:
A qualitative descriptive study using open-ended interviews with health care and homeless sector service providers was utilized to serve those experiencing homelessness in 5 cities across Canada. NVivo qualitative data analysis software was used to facilitate thematic analysis, focusing on variations in DSME for PWLEH.
Results:
We conducted interviews with 96 unique health and social care providers. Four themes were identified through focused coding of interviews. First, the use of a harm reduction approach during diabetes education tailored to PWLEH considered patients’ access to food, medications, and supplies and other comorbidities, including mental health and substance use disorders. The second theme related to the unsuitability of the curriculum in mainstream diabetes education in a group setting for PWLEH. Third, the role of group education in community building is to create supportive relationships among members. The final theme was the importance of trust and confidentiality in DSME, which were most easily maintained during individual education, compared to group formats.
Conclusions:
Overall, PWLEH experience unique challenges in managing diabetes. DSME adapted to these individuals’ unique needs may be more successful and could be delivered in both individual and group settings.
There are estimated to be more than 400 000 people in the United States and 35 000 people in Canada experiencing homelessness every night.1,2 Management of chronic conditions is more challenging for people experiencing homelessness and housing instability; the multifaceted nature of diabetes management creates especially significant difficulties. This is manifest in diabetes-related complications and outcomes for this population. Individuals with diabetes who have experienced homelessness have a mortality rate of 1.45 times that of comparable individuals who have never been homeless. 3 Furthermore, we have also demonstrated that the risk of diabetes complications is significantly higher, notably for myocardial infarctions, strokes, skin and soft tissue infections, and amputations. 3 Persons with lived experiences of homelessness (PWLEH) and diabetes in Canada have been found to have more than 20 times the rate of mental health and substance use disorders compared to those with diabetes who have never been homeless, which makes self-management even more challenging. 4 These individuals also have to manage their diabetes while dealing with unstable housing, a lack of healthy food options, and limited access to medications and diabetes supplies.5-7
Diabetes self-management education (DSME) is recommended for all individuals with diabetes to enhance their knowledge, help them adopt healthy behaviors, and improve clinical outcomes. 8 DSME is essential for reducing all-cause mortality, cardiovascular risks, foot ulcers and infections, and amputations.9,10 DSME can be provided in individual or group settings, although the most significant improvements in glycemia have been seen in those who participate in group settings.9,10 Group DSME has also been associated with sustained knowledge and improved quality of life, suspected to be in part due to the relationships developed between group members.11,12
PWLEH face many barriers to attending structured group DSME stemming from their transient and tenuous housing situation, typically lower health literacy, and many challenges to attending prescheduled follow-up visits. 5 Despite these barriers, there have been successful group diabetes programs for PWLEH.5,13-16 An informal group diabetes education session at a shelter in Canada was associated with a significant improvement in A1C of 1.1%. 5 In addition to group diabetes education, group medical visits for PWLEH, which include direct medical care like blood pressure measurements, specific insulin adjustments, and education, 17 have been shown to enhance diabetes management through problem-solving, modeling, emotional support, and social comparison. 16 The provision of one-on-one DSME in this population can be similarly challenging, particularly the retention of patients who face social and structural disadvantages. A pilot study of individual DSME focused on the needs of PWLEH was only able to retain 5 out of 9 participants over, even with flexibility in session scheduling and offering incentives at every visit. 18
Overall, we need a better understanding of how DSME for PWLEH can be structured to meet the unique needs of this population. More data must be collected on how DSME programs are organized and the daily challenges of running such tailored education programs. Notably, we are unaware of any study that has sought perspectives on the advantages and disadvantages of individual versus group DSME for PWLEH. The aim of this study was to explore the types of DSME that are provided for PWLEH in a number of Canadian centers. We were specifically interested in which forms of education were perceived to be most successful for this population.
Methods
Study Design
A multisite qualitative descriptive study design embedded within the constructivist paradigm was conducted. This method was chosen to accurately and comprehensively recount participants’ perspectives. Qualitative description is a methodology that allows researchers to “offer a comprehensive summary of an event in the everyday terms of those events” by staying close to the data and the actual words and terms used by participants. 19 This was accomplished by using participants’ voices to generate codes and themes, including the use of “in vivo codes.” We believe that this was the ideal methodology to allow the viewpoints of our participants to lend a rich understanding of the state of diabetes education for PWLEH in a straightforward manner.
Given that the parent study from which this substudy was derived sought to understand diabetes programming for this population, 20 the participants in this study were health and social care providers. In this work, no individuals with lived experience of diabetes and homelessness were interviewed, although our body of work includes these perspectives in other studies.6,21 We interviewed providers from programs that were tailored to address the needs of either people with diabetes or people who were experiencing homelessness (or both). These providers were chosen specifically because they were felt to be able to speak to certain aspects of addressing the unique needs of this population. Although not all diabetes care providers may have expertise in providing care to PWLEH, their expertise in diabetes care was thought to be valuable to this study, and many of them have cared for PWLEH even if that was not the sole focus of their practice. Interviews were conducted in 5 large Canadian urban centers (Vancouver, Calgary, Edmonton, Toronto, Ottawa). This study was approved by the Research Ethics Boards of the University of Calgary and Unity Health Toronto/St Michael’s Hospital.
Sampling and Recruitment
To obtain broad input, we recruited participants who possessed a breadth of experience with diabetes and/or homelessness. This included 4 major categories: (1) diabetes-care professionals who focus on inner-city/homeless populations, (2) other health care providers with a specific clinical focus on inner-city/homeless populations, (3) endocrinologists/diabetes-care providers (without a specific focus on homelessness), and (4) other stakeholders (frontline shelter workers, managers, social workers, etc).
Given that there is no comprehensive listing of diabetes programs for those experiencing homelessness, we used an internet search to identify relevant programs in each city using the terms “diabetes” + “homeless” + (city name). This was then broadened by searching: “health” + “homeless” + (city name). We found phone numbers and email addresses for the programs using our internet searching strategy and reached out to individuals to invite them to participate in this study. This strategy was augmented by utilizing our professional networks in each city to ask contacts in primary care, endocrinology, and the homeless-serving sector to connect us with individuals who would have relevant perspectives to share. Furthermore, snowball sampling 22 was utilized asking each study participant to connect us with others who work in the field who could provide a unique perspective on their program. Sampling was purposive, in that we recruited only those with relevant perspectives on the topic, but also comprehensive, in that we sought to be as inclusive as possible. Therefore, all individuals identified using publicly accessible email addresses and phone numbers were invited to participate in the study.
Data Collection
Data were collected using semistructured qualitative interviews, conducted by experienced research personnel (RBC and DJTC). Predefined interview guides were created for each category of participant (Table 1). The interview guide was iterative and refined after each successive interview. Interviews ranged from 20 to 90 minutes. Written informed consent was obtained from each participant. Interviews were digitally recorded, and recordings were subsequently transcribed verbatim by a trained professional.
Interview Guide
In addition to the interviews, RBC and DJTC also sought opportunities to observe diabetes education programs in operation. Although this was generally not possible for individual education sessions, given the sensitive nature of those one-on-one sessions, wherever group diabetes programs tailored to this population were discovered, we sought an opportunity to observe. Observations were done as nonparticipants (noneducators), and detailed field notes were taken in these settings.
Data Analysis
Data analysis was facilitated by NVivo 12 qualitative data analysis software (Doncaster, Australia). A process of inductive team-based thematic analysis was undertaken. All members of the coding team received training and instruction on qualitative analysis and the use of NVivo software. The principal investigators (DJTC and RBC) generated a preliminary coding template. Initially, 2 interviews were coded by all members of the coding team, who were encouraged to add codes inductively throughout the process as needed. Each coder’s work was reviewed in depth during a team meeting moderated by the principal investigator (DJTC), who is a clinical content expert and has extensive content expertise with homelessness-related research and qualitative data analysis. Discrepancies were resolved through discussion of multiple viewpoints. Following the initial meeting, another 3 transcripts were coded by the entire team until all team members were comfortable with coding, and coding was adequately reviewed by the team. The next 30 transcripts were coded in duplicate by 1 novice coder and 1 experienced coder with homelessness expertise. Coding discrepancies and emerging codes were reviewed in regular team meetings after each team had completed 5 transcripts. The coding template was deemed complete/saturated by the end of 35 transcripts with no new codes emerging. The remainder of the transcripts were coded by 2 novice coders or 1 experienced coder. Following open coding, the codes related to diabetes education underwent focused coding via thematic analysis (led by BM in consultation with DJTC). Through an inductive approach, individual codes were collapsed into themes of similar codes. The original codes were reexamined in the context of the theme to ensure the data supported the theme. If the code was not congruent with the theme, the code would be removed or the theme itself would be reconstructed.
Rigor/Trustworthiness
The rigor of this study was enhanced in multiple ways. Data triangulation was achieved by collecting data through interviews and observing the relevant programs in operation. In this way, multiple data sources were used to corroborate the themes in the analysis. In addition, various lenses were brought to the data through the analysis procedure because data were analyzed by a multidisciplinary team of trained coders, each of whom brought their own rich experience in diabetes care and knowledge of the homeless-serving sector to bear on the data. This yielded rich discussion among the team to fully understand the study participants’ perspectives.
Finally, member checking was undertaken so that all participants were invited to participate in a concept mapping project where the data shared in their interviews had been packaged into the various themes and concepts and where they had the chance to share opinions on these groupings. 6
Results
Of the 112 individuals approached, we interviewed 96 participants from 38 organizations, representing approximately 70 different programs/services. We had the opportunity to conduct observations at 3 different group diabetes programs tailored to the needs of this population. Approximately half of the interviewees’ practices focused on the general health of the homeless population, one quarter focused on diabetes care in the homeless population, and the remainder focused on diabetes care more broadly. Twenty-seven certified diabetes care and education specialists were interviewed, which included nurses, dietitians, and pharmacists. Nine endocrinologists, 26 primary care providers (family physicians or nurse practitioners) and 3 other specialist physicians who referred their patients for DSME (Table 2) were likewise interviewed.
Participants’ Demographics
Modalities of DSME included mainstream group education (targeting the general population living with diabetes), group education tailored specifically for PWLEH, and one-on-one educational sessions. The major themes around DSME were a harm reduction approach, assumptions in group diabetes education, community building, and the importance of trust and confidentiality.
Harm Reduction Approach
DSME designed for PWLEH was described as having a harm reduction approach in both group and individual formats. PWLEH were taught essential skills, like basic nutrition advice (applicable to the circumstances of the individual), blood glucose monitoring, and hypoglycemia management. The goal was not to attain target glycemic control or to strictly adhere to a specific dietary pattern but to keep participants safe and make small incremental improvements over time. Some topics pertained specifically to PWLEH, including healthy eating from a food bank or shelter, low-cost exercise options, and foot self-care. Another important role of the diabetes care and education specialist was to connect patients to community resources like shelters, meal services, and food banks.
During individual DSME, the diabetes care and education specialist was able to gauge the patient’s understanding of diabetes and understand their day-to-day life and the resources available to them. They would ensure the teaching was appropriate to the individual.
I talk to some people . . . they eat once a day. They sometimes don’t eat, they don’t know where their next meal is going to come from. And I mean am I going to sit there and talk to them about how to like carb count or like watch what they eat when they barely get a meal? No. What I’m going to do with them is I’m going to go over where in the area they can find like cheap meals or where they can find free meals. (Pharmacist 1)
PWLEH face more barriers to making healthy behavior changes, and many providers relayed that patients often felt guilty for not meeting the unrealistic goals set out by their health care providers. Diabetes care and education specialists described focusing on positive reinforcement, applauding small victories in their patients. They helped patients set small, relevant, and achievable goals.
So for instance, if a client has [a] hemoglobin A1C that’s double digits, say 10%, you know if they got down to almost like to the eight-something, we would just cheer them on and say that is absolutely fabulous. Then we would still keep working on it, but we would give them that encouragement. (Nurse 1)
This contrasts with mainstream group diabetes education, which was described as being paternalistic and usually did not acknowledge or address common barriers faced by PWLEH while managing diabetes, such as financial, mental health, or substance use challenges. Oftentimes, mainstream education was portrayed as being overly focused on dogmatic achievement of targets that were simply not realistic for this population.
Assumptions in Group Diabetes Education
When providing education to a group, curricula are designed to be generic and applicable to the largest audience. Mainstream group education was described as often relying on the provision of printed reading material and taught to people of a higher level of education than may be common among PWLEH. The dietary education is often focused on healthy eating as outlined by Canada’s food guide and other standardized self-management materials. 8 This education was focused on the “typical” Canadian culture in terms of diet and health beliefs and did not account for varying perspectives held by some individuals.
It’s important to understand my clients generally have, on average, about a grade 6 education, so they are not very literate in the first place. Many can’t read. Some come from cultural backgrounds that are very, very different, you know. (Specialist Physician 1)
On the other hand, group education that was specifically designed for PWLEH was simplified, based on the assumption that their complex social situation, cognitive baseline, substance use disorder, or mental health limited their ability to retain provided knowledge. Although it may be true that many participants needed simplified education, some were highly educated with an interest in more complex topics like carbohydrate counting, making curriculum planning for group education challenging. One community health center stopped offering group DSME based on provider feedback that “often people were bringing a lot of personal questions about managing their own stuff to [the classes]” (Family Physician 1). The providers felt that some PWLEH struggled to apply the more generic didactic education to their own circumstances and required more tailored guidance, which was not feasible in group settings. Educational needs also vary between patients with type 1 and 2 diabetes, length of time living with diabetes, concomitant substance use disorders, and available social supports and personal finances, which group education is not ideally positioned to address.
Community Building
Group education has many benefits, like improved efficiency, peer-to-peer knowledge transfer, and reduced social isolation among patients. However, these groups are only successful if the individuals have a sense of belonging within the group. Most participants regarded group education designed for PWLEH as a safe and supportive community.
It’s the social support that people are getting. They are getting information from each other. They are getting support from each other. They are all living with the same illness and so that is the value in doing these groups. (Nurse Practitioner 1)
Group education tailored to PWLEH was often successful in contributing to the development of a sense of community among its members through more than just didactic diabetes education. Some had cooking classes, fitness classes, and grocery store tours, which were unique educational opportunities that fostered relationships among the group.
I’ve done actual cooking sessions where we are doing our dietary session, but we are actually learning how to make a sandwich from tuna and a whole wheat pita wrap, right. (Dietitian 1)
This contrasts with mainstream groups where providers expressed that they perceived PWLEH often felt out of place and self-conscious, possibly having slept on the streets the night prior: “It’s really hard for our guys to go into a place where people are dressed nicely and had a shower in the morning and all of those things” (Nurse 2). For PWLEH, this setting was not conducive to the sense of community required for group education to be successful.
Importance of Trust and Confidentiality
We were told that many PWLEH had a history of trauma and thus struggled to build trusting relationships with providers, which can be more challenging in a group setting than one-on-one. When discussing a prior unsuccessful group for PWLEH, a participant blamed its downfall on the lack of consistent facilitators to develop these relationships: “[The patients] weren’t comfortable and it wasn’t the same educators each time . . . I think that was the biggest part of the problem” (Nurse 3). There was also the concern of confidentiality within a group because others were privy to attendees’ personal health information by virtue of attending a disease-specific group. “It’s a small community and people talk and like to share other people’s personal information, it’s a form of power” (Nurse 4).
Individual education was felt to rely on a foundation of trust between the diabetes care and education specialist and the patient. Diabetes care and education specialists described how they emphasized confidentiality, tried to be judgement-free, and focused on getting to know their patients personally to help them achieve their goals.
I think it’s important to engage the person as much as possible and to build trust before offering too many suggestions. Of course, if they ask for suggestions, you give it immediately, but people may or may not feel comfortable doing that, so you have to give them the time that they need. (Nurse 5)
Further rapport was built between community members and diabetes care and education specialists when the latter visited rooming houses, shelters, and the local food bank. This was done to increase patients’ exposure to diabetes care and education specialists, hoping that they would feel more comfortable attending appointments with them.
[The outreach worker] does these rounds about the city. He walks around for 4 hours in a certain part of the city and he knocks on doors of . . . rooming houses . . . so [he] contacted me and this is what I’m doing now . . . so I go with my little tiny handouts and it’s very little words just to say who I am and the number to call. (Dietitian 2)
Discussion
DSME is essential to gain the skills necessary to improve diabetes outcomes. 8 PWLEH have many barriers to managing their diabetes, one of which is access to appropriate DSME. 5 In this study, we found providers did not suggest PWLEH attend mainstream group education due to concerns that the curriculum would be inapplicable or the patient may feel uncomfortable. Both group and individual education focused on PWLEH provided unique benefits for the patients served and are viable options for DSME.
In prior studies, the DSME curriculum for PWLEH has included general information about diabetes, blood glucose monitoring, complications of diabetes, and dietary advice. These topics were taught with an understanding of the environment and resources available to PWLEH.15,16 In this study, diabetes care and education specialists working with PWLEH often took a harm reduction approach and addressed their patient’s unique needs, including managing diabetes in the context of food insecurity, substance use disorders, and lack of medication coverage. Health care providers expressed hesitancy to refer their patients to mainstream education because it was not suited to the needs of PWLEH.
Study findings found health care providers were split on whether group education for PWLEH was overly generalized and simplified for the population. Group DSME was susceptible to delays and tangents due to patient questions regarding highly personalized issues, which were often less applicable to the other attendees. However, multiple studies have described successful group education and group medical visits specifically designed for PWLEH. 14 Group education for PWLEH has been shown to improve knowledge, patient empowerment, and self-care behavior. 15 Davachi and Ferrari 5 showed a 1.1% A1C reduction between 3 and 12 months with informal group education at a shelter and other shelter changes like providing a snack at bedtime. These studies, along with the perspectives of the providers we interviewed, argue that group education can be successful for PWLEH if the programs are specialized to fit the population’s needs.
Group diabetes visits are education sessions that also include medical care, like point-of-care glycemic monitoring and medication adjustments. 17 Group medical visits have the benefit of peer relationships and support with the addition of individualized medical care. Thompson et al 16 found group diabetes visits for PWLEH demonstrated the benefits of relationship building with enhanced problem-solving, modeling, emotional support, and accountability. These visits rely on some degree of homogeneity or shared lived experience within the group to help develop a sense of community, thus building a more supportive environment. 16 This is consistent with what we found with health care providers seeing the benefit of peer identification within groups focused on PWLEH but not with mainstream education. Providers were hesitant to send PWLEH to mainstream groups because they felt their patients may feel ostracized.
Relationship development in group DSME has been cited as one of the reasons group education has a more dramatic reduction in A1C and enduring knowledge compared to individual education. 12 Having a consistent diabetes care and education specialist is important for building trusting relationships among a group and was a reason why one of the DSME groups failed. Building trust and confidentiality is of utmost importance and was often easier to do with individual DSME because the diabetes care and education specialist usually remained the same over time. Thompson et al 16 argued that diabetes care and education specialists who lead groups for PWLEH should have expertise in group facilitation to ensure they have the ability to navigate barriers to diabetes management, provide leadership, foster participation, and model warmth and empathy. Davis et al 15 used peers to provide DSME, which ensured it was applicable and resulted in positive relationships within the group. These examples demonstrate how an experienced and competent group diabetes care and education specialist can change the group dynamic. It is important that they are consistent, trained in group facilitation, and have empathy for their patients’ experiences.
Strengths of the study included diverse perspectives found by interviewing health care and homeless service providers in 5 different cities across Canada. It allowed us to explore various education programs for both PWLEH and the mainstream population. Using the open-ended interview style allowed for in-depth and complete responses by all participants.
Limitations of the study included only capturing the perspectives of the health care and homeless service providers and not PWLEH. However, we continue to engage with PWLEH in parallel research endeavors to ensure their voices are heard and amplified.5,6,21 Given that this was a qualitative study, the results are based on subjective experiences, and outcome measures were not assessed. We cannot definitively conclude that one form of education resulted in improved health outcomes for PWLEH. Finally, this research was conducted only in Canada, where there is universal public health care insurance coverage, so our findings may not be fully applicable to the unique challenges related to DSME seen in other settings where private health care coverage is the norm.
DSME is essential for all patients with diabetes to improve self-management skills and reduce the risks of complications. 8 PWLEH face unique challenges in managing diabetes that must be addressed in their education. Both individual and group education focused on PWLEH and that took a harm reduction approach were reported to be successful in improving patients’ knowledge and diabetes management. Group education focused on PWLEH had the additional benefit of building relationships and peer support compared to individual education; however, it lacked privacy between members and was more generic and simplified. Overall, PWLEH would be better served if diabetes care and education specialists provided tailored DSME in a harm reduction framework instead of expecting these patients to access diabetes education through groups not designed for their unique needs.
