Strategies to Improve Prevention and Management in Diabetic Retinopathy: Qualitative Insights from a Mixed-Methods Study

We designed two online patient education activities that could be viewed live and on demand (Table 1). The education goal was to empower patients to be their own advocates for optimal ophthalmological care. The study goal was to assess how the education initiative helped to change patient knowledge, competence, and practice.

Because the environments in which patients navigate health and health care are complex, we adopted a mixed-methods approach to data collection and analysis, using both quantitative data, which allows for statistical interpretation and generalization, and qualitative data, which supports depth of understanding and the acknowledgment of how context shapes behavior change. We collected and analyzed data across three phases of program delivery: 1) baseline data via an online pre-test survey concerning patients’ knowledge and beliefs, 2) survey data after patient exposure to education content to measure the immediate impact of education, and 3) follow-up data 5–16 weeks after activity completion via both an online survey and in-depth qualitative interviews with participants. Here, we report solely on the qualitative component of the study.

We used a combination of social media channels (YouTube, Facebook, and Twitter) and traditional email-based announcements to invite community-dwelling people with dia-betes, with or without DR, to participate in an online patient-focused education activity. Recruitment efforts were partially channeled through partnerships, including the online communities of DLife, The Diabetes Hands Foundation, and the Joslin Diabetes Center of Harvard Medical School in Boston, Mass. Recruitment included a combination of tweets on Twitter, posts on Facebook, and YouTube promotions for users with expressed interests in diabetes; email messages to patient databases of partner organizations; and announcements on the websites of the partners listed above. Total reach was difficult to estimate because of the social media elements; however, it is conservatively estimated that >300,000 individuals with expressed interests in diabetes (e.g., patients, patient advocates, and caregivers) were exposed to recruitment messages.

Program 1, titled Diabetic Retinopathy Part 1: What Is It, and Are You at Risk?, targeted people with diabetes and no DR diagnosis. Program 2, titled Diabetic Retinopathy Part 2: Preserving Your Eyesight, targeted people with diabetes and a diagnosis of DR. Participants completed knowledge test questions before and immediately after activity completion to gauge knowledge acquisition.

Approximately 1 month after completing either program, participants were invited via email to complete a screening survey to determine their eligibility to be interviewed. At the time we began screening program participants as potential interviewees, there were 160 participants for program 1 (August 2015) and 183 for Program 2 (November 2015). Participants were eligible to participate in interviews if they had a confirmed diagnosis of type 1 or type 2 diabetes, and, for Program 2, had been diagnosed with DR.

People who met these eligibility criteria were subsequently invited via email to participate in telephone interviews. Of 64 Program 1 participants who responded to interview invitations, 20 were eligible, 9 did not respond to scheduling information, and 11 were interviewed. Of 64 Program 2 participants who responded to interview invitations, 18 were eligible, 8 did not respond to scheduling information, and 10 were interviewed.

The Western Institutional Review Board provided ethical approval for this study, and we obtained informed consent from all participants before conducting interviews. Participants received a small honorarium after completing their interviews. We conducted telephone interviews that lasted ∼30–45 minutes per participant, 5–16 weeks after the educational activity in which interviewees took part. All interviews were semi-structured in format.

Interviews were audio-recorded and transcribed verbatim. We removed identifying information and typographical errors from all transcripts, which we imported into NVivo 11 Mac software (QSR International, Australia). We used grounded theory and a process of constant comparison to analyze participant responses both across and within interviews (13–15). In grounded theory, the data, rather than a pre-set hypothesis, guide the direction of analysis, which involves generation of themes and hypotheses as an ongoing, iterative process. Analysis included three steps: 1) data immersion and familiarization, 2) data coding according to broad areas of interest, and 3) theme identification.

We coded transcript content into descriptive categories, then identified potential themes of relevance until we achieved thematic saturation (Table 2). Indicators of themes included words, phrases, or segments of text that were used in a similar manner by respondents across or within interviews and that pointed to an emerging idea or concept. We explored emerging themes in node matrices within and across transcripts and cross-checked all emerging themes or critical issues with participant attributes (i.e., sex, age, ethnicity, duration of diabetes, duration of DR, and time from program participation).

We conducted 21 in-depth telephone interviews with patients who participated in one of the two education programs (Table 3). Program 1 included 11 patients with diabetes (10 with type 2 diabetes and 1 with type 1 diabetes) who were 45–65 years of age (mean age 59.6 years) who had not been diagnosed with DR. Participants’ time from diagnosis of diabetes ranged from 3 to 38 years (mean 13.7 years). Program 2 included 10 patients with type 1 diabetes who were 40–68 years of age (mean age 56.9 years) who had an established diagnosis of DR or, in one case, pregnancy-related microaneurysms. These participants had been living with diabetes for a mean of 26.8 years and had a time from diagnosis ranging from 6 months to 15 years (mean 5.3 years).

We explored participants’ awareness of how diabetes can affect eyesight by asking them to describe vision problems typically associated with diabetes and what they knew about DR. Most participants were aware that diabetes can affect eyesight and repeatedly voiced concerns about the potential for blindness as a result of diabetes. Many Program 1 participants had already experienced some vision problems since being diagnosed with diabetes or knew someone close to them who had vision deficits; however, before program participation, approximately half of Program 1 participants were uncertain whether eye conditions such as cataracts, glaucoma, dry eye, and DR were linked to diabetes or were age-related.

Most participants had heard about DR before participating in the activity. Nonetheless, they typically described not knowing much about DR, and few participants reported that their HCPs had spoken to them specifically about DR—a theme that recurred throughout interviews with participants in both programs.

It proved challenging for participants to describe DR and identify its symptoms after viewing the activity. Overall, participants used imprecise language to describe retinopathy. Nonetheless, participants worked hard in interviews to recall their new knowledge about DR, and they did appreciate some of the important points about the condition, if not the precise technical language, as evident in the following quotes:

Program 1 participants were also better able to identify risk factors for DR after program participation. The predominant risk factors that participants identified were poorly controlled blood glucose or poorly managed diabetes, followed by insufficient diet and exercise, as illustrated in the following comments:

With prompting, a small number of participants also noted weight and blood pressure, but no one mentioned kidney disease, high cholesterol, anemia, stomach problems (gastroparesis), or pregnancy.

Program 2 participants all had an established diagnosis of DR or microaneurysms and regularly saw an ophthalmologist for general eye care and/or a retina specialist for retinopathy care. They were generally unaware of DR as a diabetes-related eye condition before their own diagnosis and were typically surprised at their own diagnosis, which all but three participants received during a routine eye exam. Despite emphasizing their surprise, some participants also recalled experiences that are redolent of retinopathy symptoms. For example, one participant noted:

Similarly, another participant reflected that although before his retinopathy diagnosis he had experienced “no major vision problems,” the ophthalmologist had noted in previous visits that there was slight bleeding in his eyes and had commented to him, “As long as you keep your blood sugar in control, there should be no problem with it.” Although his eyesight was sometimes unclear when his blood glucose was high, this participant said:

These comments suggest a tendency for people with diabetes—even those who are vigilant about routine eye examinations—to ignore signs or symptoms of poor eye health on the grounds that such symptoms are a normal part of diabetes. Furthermore, they suggest that before participation in the education program, not all participants fully grasped that retinopathy is initially a disease without symptoms.

After participation, when we asked Program 2 participants how they would describe retinopathy to a friend or family member, they were likely to highlight the inevitability of retinopathy for people with diabetes and the importance of blood glucose control, as shown in the following comments:

After program participation, interviewees were generally adamant about the need to be aware of risk factors for retinopathy and to educate others about this risk.

These comments suggest that participants absorbed two key messages of Program 2, which were 1) that everyone with diabetes is at risk for retinopathy and 2) that good blood glucose control is essential.

Before education participation, interviewees in both programs identified routine blood glucose monitoring as their main diabetes self-management and DR prevention strategy, alongside regular eye exams and what one participant called the “diabetic triangle”—medications, diet, and exercise. After program participation, participants emphasized that the education had reinforced for them the importance of continuing to pursue the glucose monitoring and control regimens, and all expressed a heightened awareness of the importance of following their monitoring and control strategies more assiduously. As one interviewee noted:

Before viewing the activity, all participants described having “frequent,” “regular,” and “routine” comprehensive eye examinations, including retinal exams. Indeed, all participants were adamant about the importance of regular eye examinations as a key step to maintaining healthy eyesight and emphatic that others with diabetes should have their eyes routinely checked to make sure there are no changes in the retina.

However, although Program 1 participants were able to describe the process involved in eye examinations, after education participation, some participants remained diffident about its purpose, as illustrated in the following quotes.

Most Program 2 participants saw an ophthalmologist or a retinal specialist for routine eye exams. Although participants viewed a retinal eye exam as part of the steps they needed to take to monitor and control diabetes, they did not necessarily describe a retinal eye exam as a specific strategy to minimize DR, as illustrated by this quote from a participant with type 1 diabetes:

Participants in both programs viewed eye exams as simply something they knew they should do to help manage their diabetes, but not necessarily as part of DR prevention and management. These findings suggest a residual deficit in participants’ understanding of the specific purpose of retinal eye examinations for screening and monitoring DR, which potentially poses a barrier to self-efficacy in reducing risk for DR.

Most Program 2 participants had been, or were currently being, treated for DR, and three were under surveillance for what they described as “mild retinopathy” or pregnancy-related microaneurysms. Participants described a range of side effects that they attributed to treatment, including discomfort or pain (from both laser and injections); “pouchiness” under the eyes and dye allergy (from the laser); and infection, eye sensitivity, and redness (from injections) and were generally unequivocal about treatment effectiveness. Although few interviewees were aware of alternative treatment options before participation, the education helped participants better understand their therapeutic options and identity questions they wished to ask their HCPs.

Although participants in both programs felt that their HCPs would respond to any questions they had about DR, few participants identified their HCPs as sources of verbal or written information about DR. Participants typically noted that their providers “never discuss it at all,” are “not very communicative,” or “don’t tell me anything special.” When participants said their HCPs did refer to retinopathy, participants noted that the emphasis was typically on the importance of blood glucose control as a general tenet of diabetes management, rather than as a specific approach to preventing or minimizing the risk for DR. The main reason suggested for this communication deficit was lack of provider time; however, many participants also noted that unless they asked questions, HCPs were unlikely to volunteer unsolicited information unless there was “bad news” to report.

Similarly, before education participation, despite acknowledging the importance of asking questions and of being actively engaged in the management of their own eye health, active engagement was an exception rather than a rule for most participants. Although some participants described themselves as ready and able to ask questions of their primary diabetes care providers, saying, for example, “I'm not afraid to talk to doctors,” most assumed that providers knew what they were doing and therefore did not ask questions.

At the same time, there was a perceived communication disconnect with HCPs. Although participants felt that their primary diabetes care providers “had their hands full with the rest of the endocrine system” and assumed that specialists (i.e., either their endocrinologist or their eye specialist) would educate them about DR, they also noted that specialists infrequently volunteered to share information about DR with patients.

After viewing the activity, all participants were able to list questions they would like to ask their primary diabetes care and eye health providers based on themes addressed in the education. For Program 1 participants, these themes concerned self-efficacy strategies.

Almost half of Program 2 participants identified questions specifically related to treatment that they said they would like to ask their eye health provider at their next visit.

This qualitative study suggests that online education can empower patients with longstanding diabetes and established DR to more proactively engage with their primary diabetes care provider (primary care or endocrinologist) and eye health providers. Although many participants in both programs were somewhat aware of diabetes-related eye conditions and had heard of DR before the education, they knew little about the association of DR with diabetes. After the education activity, participants had absorbed key messages about the gradual nature of change and symptoms associated with later stages of DR and expressed a heightened awareness about DR as one of the many conditions that diabetes can precipitate.

Before the education, participants across both groups were aware of the need for strict glycemic control as a factor in reducing the long-term risk of vision loss and reported that they already engaged in blood glucose monitoring and control strategies. After education, participants emphasized poorly controlled blood glucose or diabetes as the predominant risk factor for developing or worsening DR and emphasized that the activity reinforced for them the importance of blood glucose monitoring more assiduously.

After activity participation, al-though some Program 2 participants remained unclear about alternative treatment options for DR, almost half of this group identified questions about therapeutic options as potential alternatives to their current treatment that they felt they could now ask their eye health providers.

Despite the importance of effec-tive communication and care coordination between HCPs and patients, interviewees described poor communication about disease prevention and progression as the norm. Few participants identified HCPs as known or established sources of verbal or written information about DR, and participants often commented that neither their ophthalmologist nor their diabetes provider had mentioned any tools or resources to enhance their vision.

Participants also generally ex-pressed the view that, although HCPs should be sharing information about patients with each other, few felt that providers did this in practice. Participants felt that their primary diabetes care providers assumed that their eye health providers were managing DR, and the eye health providers assumed that the primary diabetes care providers were managing the diabetes. This insight points to a distinct clinical division of labor that likely influences the assignation of responsibility for educating patients.

Finally, although participations acknowledged the importance of active engagement in the management of diabetes eye health before the education, active engagement was an exception rather than a rule. After education, participants were able to identify questions they would ask their primary diabetes care and eye health providers.

Participants who chose to view this educational intervention were not randomly selected. Rather, they had self-selected to seek further information about diabetes and DR. Accordingly, our sample may have been more likely to be proactive in inquiring about medication issues when interacting with their HCPs and also may have been atypical of the general population with diabetes. Additionally, there were key retinopathy and diabetes status differences between the two groups that may contribute to response differences. Nonetheless, our analysis of qualitative responses suggests that, even among such participants, there was some passivity with respect to discussing eye disease with HCPs, likely masked by comments that providers were “busy.” If the generally proactive participants in our sample still had a passive approach to addressing DR issues, then less proactive people with diabetes are likely to be even less likely to actively address this important aspect of eye health with their diabetes care providers. Also, although useful for many, this method of education is not suitable for all patients with diabetes. In fact, elderly patients may not be comfortable with online education, and patients without Internet access could not take advantage of such a program. However, our results do emphasize the importance of patient-provider communication as a basis from which to begin any preventive efforts.

This qualitative study demonstrates that online patient education is an effective tool in building patient knowledge and awareness about DR and in motivating action in DR self-care. Notably, such education cannot only help to reinforce self-management behaviors (e.g., glucose monitoring and control and regular eye exams), but also build confidence as a foundation for being more actively engaged in preventing and managing DR. Participants emphatically endorsed the importance of Internet-based, “free” education as part of their ongoing approach to diabetes self-management and felt this to be especially important since their HCPs rarely addressed retinopathy, which they viewed as a “hidden” dimension of diabetes. As one participant noted:

Participants also emphasized the need for continuous messaging to people with diabetes about the importance of glucose control and eye examinations as strategies to avoid retinopathy and also felt they had some personal responsibility for telling others about strategies to prevent vision loss. The use of Internet-based content and information technology continues to grow as a vehicle for enhancing education support in diabetes (16). Given this and the perceived communication deficits among HCPs, diabetes care providers and eye care providers may wish to consider encouraging their patients to participate in online patient-directed education about DR as a means of building patients’ self-efficacy, instilling motivation, and fostering patient engagement.

This study was funded by an unrestricted educational grant from Genentech.

No potential conflicts of interest relevant to this article were reported.

All authors researched data, contributed to the discussion, and wrote, reviewed, and edited the manuscript. A.H. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.