Facilitators and Barriers to Glaucoma Medication Adherence

Précis: Patient education and dosing self-efficacy are important factors related to ongoing adherence to glaucoma topical treatment, and patients view their disease management as a shared effort with their provider. Purpose: Glaucoma affects nearly 3 million Americans, and medication adherence has been reported to be as low as 20% in this patient population; however, key limitations to our understanding of this behavior in adults with glaucoma exist. Patients and Methods: This research used an electronic survey including validated concepts related to topical medication use and an in-person interview to investigate the influencers of and solutions for challenges to medication adherence in adults with glaucoma. Patient eligibility was determined upon arrival to a regularly-scheduled visit to the Vanderbilt Eye Institute, during which they were asked for consent to complete the survey. Responses were captured by tablet and assessed using descriptive and inferential statistics. The primary focus was instrument correlations with the Adherence to Refills and Medications Scale score and were run between Adherence to Refills and Medications Scale, and the totaled score for each individual questionnaire as well as individual items. Recorded interviews were thematically assessed by multiple study team members. Results: Survey results of adults with glaucoma suggested that self-efficacy, forgetfulness, fear of side effects, and dosing ability were all related to self-reported medication adherence. Despite most having glaucoma for several years, discrepancies in disease knowledge were observed. Patient interviews uncovered 3 overall themes related to glaucoma treatment: (1) glaucoma management as a shared responsibility; (2) the importance of patient education; and (3) specific adherence facilitators and barriers. Conclusion: Glaucoma medication adherence interventions may benefit from focusing on developing patient medication-taking self-efficacy, disease-related education, and engagement with their provider.

A lthough medical treatments are effective, nonadherence to glaucoma medications is suboptimal, and those adhering to their glaucoma medication <80% of the time have worsening visual field defect severity. 1 A host of reasons and risk factors for patients' nonadherence to therapy have been accumulated across a number of studies. Evidence suggests that consistent barriers to glaucoma medication adherence include cost, 2-4 disease and/or medication knowledge, 5,6 treatment complexity, 7-9 patient engagement, 6 forgetfulness, and issues with drop technique. 10 However, key limitations to our understanding of this behavior in adults with glaucoma exist, including the extent to which these factors interplay and whether their influence changes over time.
In interpreting how multiple factors may interact, a number of behavioral theories, including the Health Belief Model, Social Cognitive Theory, Self-Determination Theory, and others, have been applied to interpreting glaucoma medication use behaviors. Constructs from these frameworks have been instrumental to improving our understanding of this patient population, suggesting that glaucoma medication adherence may be affected by self-efficacy, motivation, social norms, autonomy support, and perceived necessity of treatment. 1,6,[10][11][12][13][14][15][16][17][18][19][20][21][22][23][24] Building on these findings, multiple frameworks specific to glaucoma medication management have emerged, focusing on the role of patient knowledge in overcoming adherence barriers as well as the need for a holistic approach to glaucoma patient care that addresses the impact of the diagnosis, need for adherence, and managing of stress over the course of time. 25,26 While these frameworks serve as helpful guides to both interpreting and addressing medication use behaviors in glaucoma patients, the range of theoretical constructs observed in this patient population suggest that a more integrated model of behavior may be more appropriate and deserves testing.
Acknowledging a need to better understand glaucoma patient behavior and in pursuit of building additional frameworks in this population, this study sought to gain added insight from adults with this condition in terms of both challenges to initiating therapy and solutions to achieving adherence. In focusing on both barriers and solutions, findings can guide patient care throughout the continuum of therapy by assisting providers in preparing for and then navigating patients toward effective medication use strategies. Specifically, such insight will be used to develop a tailored behavioral intervention among adults with glaucoma to improve the odds of establishing and maintaining appropriate medication adherence in this population from the onset of treatment initiation.

PATIENTS AND METHODS
This study employed mixed methods to uncover reasons for and solutions to medication nonadherence among adults with glaucoma. This involved in-clinic recruiting of willing respondents to complete a tablet-based survey during a scheduled clinic visit. Guided by the survey results, a semistructured interview guide was then developed by the study team to corroborate survey findings and uncover solutions patients found to managing their glaucoma medication regimen early in their treatment. Individual patient interviews were conducted in-clinic during scheduled visits with recruitment conducted over the phone before the encounter. Survey respondents and interviewees were provided a gift card for participating. This study was approved by institutional review boards at the University of Tennessee Health Science Center and the Vanderbilt University Medical Center.

Quantitative Assessment
Surveying was completed in the clinic before being seen by a Vanderbilt Eye Institute physician. Potentially eligible adult patients were approached by a study team member to inquire about study interest, provide details, and confirm eligibility. Those eligible to participate must have been diagnosed with and taking at least 1 medication for glaucoma and be at least 18 years of age. Informed consent was then given in writing, and those willing to participate completed the survey on a tablet with responses recorded in REDCap.
Guided by previous research examining glaucoma medication adherence and using established instruments, a comprehensive survey was created by the study team. In addition to collecting patient demographic, medical history, and medication information, the instrument included the following established surveys: (1) Glaucoma Treatment and Compliance Assessment Tool 27 ; (2) Beliefs about Medicines Questionnaire 28 ; (3) Adherence to Refills and Medications Scale (ARMS) 29,30 ; (4) Glaucoma Medication Adherence Self-Efficacy Scale 31 ; and (5) the Short Test of Functional Health Literacy in Adults. 32 The intent of including each of these instruments on the combined survey was to identify individual constructs within and across instruments that correlate with self-reported medication adherence indicated by scores from the ARMS (range is 4 to 48 with a lower score indicating better adherence). Specifically, the constructs or factors included barriers, benefits, cues to action, patient-provider relationships, perceived severity, perceived susceptibility, and self-efficacy. Scoring of each individual instrument followed directions provided by the original survey designers. The final instrument included 114 questions.
Data from the survey were descriptively presented (means and SDs for continuous variables, counts and percentages for categorical variables) and compared across patient demographics by inferential statistics. The Pearson correlations were calculated between the ARMS and the Glaucoma Treatment and Compliance Assessment Tool, Beliefs about Medicines Questionnaire, the Short Test of Functional Health Literacy in Adults, and Glaucoma Medication Adherence Self-Efficacy Scale. To conduct subgroup analyses, select categorical variables had their responses dichotomized and were assessed using χ 2 tests. All quantitative analyses were conducted using SPSS 26 (IBM Corp., Armonk, NY) with a P-value <0.05 set a priori as the threshold for statistical significance.

Qualitative Assessment
Using the survey results, a semistructured interview guide was developed by the study team. The guide had several foci: (1) identifying barriers to glaucoma medication use at the initiation of therapy; (2) inquiring how patients might handle medication use issues reported by their peers (ie, survey findings) in both the first 6 months and since that timeframe; and (3) discussing changes in glaucoma-related knowledge since the initial diagnosis. A brief demographic survey accompanied the interview.
Potential subjects for the face-to-face interviews were recruited during scheduled visits to the Vanderbilt Eye Institute from December 2019 through February 2020 and continued until data saturation was achieved. Potentially eligible patients were identified by chart review from among pools of patients with scheduled visits. Before an upcoming visit, a study team member contacted patients meeting initial criteria (adults taking at least 1 glaucoma medication) and introduced them to the study. If interested, each patient was then approached by another study team member at their visit before encountering their physician. All interviews were conducted in private rooms in the clinic before the patient's scheduled provider encounter. At the beginning of each interview, informed consent was provided and documented, including obtaining permission to record the session. To maintain confidentiality and anonymity of study subjects, neither the location nor time of the interview was recorded. If names were mentioned unintentionally during the interview, they were redacted in the transcripts. Each interview ranged from 15 to 30 minutes, and participants were offered compensation for their participation in the study. The interview questions were developed by the team based on an extensive review of the literature, and respective team member expertise (eg, ophthalmology, health behavior, pharmacy) with the intention of mirroring content asked in the survey with the added focus of patients identifying solutions to their medication use problems early in their glaucoma treatment.
Following the interviews, the audio recordings were transcribed verbatim by a commercial company. Thematic analysis of the data followed the Braun and Clarke framework approach. 33 Specifically, the research team applied the 6 steps recommended to capture associations between categories while also extracting and conceptualize the themes. 33 One researcher (R.M.) independently read each transcript and inductively coded the content. The codes were clustered into categories based on their similarities. 33 The research team then met multiple times to discuss the identified themes and to ensure the codes and categories developed would capture all the data by discussing the similarities and differences for each emergent theme. Triangulation, using multiple analysts, provided a quality check on selective perception and blind interpretive bias that could occur through a single person doing all the analysis or through the employment of a single method. 34 Dedoose (Hermosa Beach, CA) was used for all qualitative assessment.

Survey Findings
A total of 55 patients diagnosed with glaucoma and currently prescribed antiglaucoma ophthalmic drops were surveyed and included in the analysis ( Table 1). The mean Approximately half of patients disagreed that their eye drops are reasonably priced (49.1%). Almost a third of patients at least agreed that they sometimes fall asleep before dosing time (30.9%), the same proportion indicated the drops are not with them at dosing time (30.9%), and 38.2% of patients agreed that they use reminders as a means to remember to take their eye drops. Importantly, 16.4% of patients agreed to suffering from side effects, and over a third of patients agreed that the eye drops cause pain or discomfort (34.5%).
Despite a majority suggesting an excellent knowledge of glaucoma risk factors (76.4%), discrepancies in specific knowledge-based items existed. Specifically, over a third of patients were uncertain if glaucoma is always genetic (43.6%), can occur with normal eye pressure (49.1%), can be caused by diabetes (41.8%), can be caused by an eye injury (36.4%), and the reasons people get glaucoma are not well understood (36.4%). In contrast, patients did have a basic understanding of some aspects of the disease seeing as a large majority of patients agreed that all vision could be lost to glaucoma (80%), a person can have glaucoma and not know it (89.1%), vision lost from glaucoma is permanent (81.8%), and being older increases the chance of developing glaucoma (81.8%).
When the 7 validated scales were compared with the ARMS, the only questionnaire that significantly correlated was the Glaucoma Medication Self-Efficacy Questionnaire (r = −0.319; P = 0.021), which suggests a positive relationship between the ability to manage therapy and self-reported adherence to treatment.
Additional relationships were observed when the ARMS was compared with other specific items in the survey. Suffering side effects (r = 0.305; P = 0.025), having difficulty using the eye drops (r = 0.430; P = 0.001), and forgetting to administer at dosing time (r = 0.673; P < 0.001) correlated with a lower level of adherence (Table 2). In contrast, a higher level of adherence was reported with increasing age (r = 0.299; P = 0.028) and was associated with concerns about the long-term effect of the eye drops (r = −0.336; P = 0.013) ( Table 2).
Comparisons were also made across categories of selfreported adherence levels by dividing the sample into good and poor adherer groups based on if they were above or below the median ARMS score. Compared with good adherers, poor adherers were more likely to report falling asleep before dosing time (16% vs. 46%, respectively; P = 0.018) and not having drops with them at dosing time (16% and 50%; P = 0.009) (Fig. 1).

Adherence Themes
In total, 15 patient interviews were conducted. The sample had a median age of 67 years (range: 45 to 84 y), was mostly white (12/15), and was well-balanced by sex (8 females, 7 males). Most subjects were unmarried and had glaucoma in both eyes, and nearly all had at least a college degree and some form of health insurance.
Thematic assessment of the interviews yielded 3 overarching themes with specific foci therein (Table 3). First, subjects indicated that managing glaucoma properly is not a task for a single patient but instead an interactive team effort involving both those who are diagnosed and the care team. Many patients reported being unaware of glaucoma and were not seeking regular eye appointments before their diagnosis. With this lack of understanding, the original

Overall Theme Representative Comments
Managing glaucoma is not solely the patient's responsibility "Well, I've learned that it's not, it doesn't have to be, uh, debilitating, the, um, disorder, um, health issue. But you've got to have the right doctor and you've got to do what the doctor says do. You can't neglect your care." (79-y-old female) "uh, ask your doctor questions in and be certain that you understand what your situation is and understand, uh, what you can do to, uh, to help yourself." (69-y-old male) "Didn't even know what the disease was. And then through an unfortunate incident, I'll tell you, I got a piece of something in my eye and I went to my medical doctor and he said, I can't see anything but I'm going to send you to this peak vision center. Well, when I went to them, they had dug something out of my eye and he checked the pressures and he was the one that freaked out and scared me, I told you it wasn't the best. Yes, and told me we set me up to come down here. So it was a freak thing that I even found out I had glaucoma and had, I knew what it was prior to, I would have obviously I'd, you know, pay more attention, going to eye doctors and stuff. So partly the importance of having an annual check with the doctor after, because the way they explained it to me is my glaucoma is at a point where it's right on the edge of being serious vision." (60-y-old male) "Uh, well, there weren't any real like challenges, but, um, I, I kind of missed, you know, I, I wasn't, um, taking them the drops as regularly as I probably should have because I didn't know what glaucoma was. And, um, and because there is like a total lack of symptoms, you know, I didn't take it very seriously initially." (48-y-old female) Patient education is key to long-term success "Well, I have no idea if I was getting the right amount, but I didn't have, you know, occasionally you miss your eye. Um, but like I said, my mom had it so I had watched her administer eye drops and the nurses and really she was in administer eye drops, so I had some exposure. That's good too. How to do it. Um, so there probably there, there should be a bit of training I think because these drops are really important. It's, I think there should be some training instead of just going, okay, here's a prescription for your drops. All of my put it in one eye twice a day. There should be some training because it's different than taking a pill." (60-y-old female) "My mom had glaucoma and she was taking two different drops and no doctor ever told her to wait five minutes in between the draws. And then a doctor ever told her to pinch the tear duct so that they don't go down. So she had her death was basically completely blind. And I think some of that is because she was not instructed on how to administer the eye drops correctly." (60-y-old female) "Um, just, you know, look up and understand, well, you know, what the disease is and you know, what the treatment, so you would tell yourself to look at all, I think, you know, initially, you know, I didn't understand what glaucoma was, but I didn't understand the goal of the, the treatment. Hmm … Uh, yeah, that's definitely something we should Def, uh, maybe get the doctor to also emphasize and then also tell patients they should look up more information on their own to get more knowledgeable." (48-y-old female) "… Because, uh, you know, when I was first diagnosed with the elevated pressure and told that I was at risk for glaucoma, um, I read about it and understood what the consequences could possibly be." (69-y-old male) "I think it's really … it's important to do because I've been taking drops for see 38 y. So if you don't do it, I mean you're thinking I did that. I've been doing this half amount of lifetime." (70-y-old female) Specific facilitators and barriers to adherence "… When I ran out I would think I'm going to get it reordered and then I w it would be like a day or two before I remembered to call." (57-y-old male) "They said we don't have any. They might go, that's not acceptable. I've got to have the sidetrack. And luckily they were able to get some, but apparently some pharmacies were not." (60-y-old female) "… Um, I, well I have, I'm now on Medicare, but I also have private insurance to backup Medicare. So between the two, for the most part, they do a good job of covering the cost." (70-y-old female) "… I'll tell you this, when I went to the preservative free drops, they were individually packaged and I was spending about $150 every 90 d. When Dr. Brown put me on the drops that I'm using today, it went from $150 to about 15. So that at first was a big challenge because it seemed like the prices kept going up and up and up. And then I don't know if he inadvertently changed my drop cause it's the same drives, just packaging different. But uh, um, or if he specifically changed it, I don't know. But when he did it worked and it was a whole lot cheaper and I don't want him to make any changes cause it was pretty expensive. They told me at the pharmacy that had I not had insurance that it would cost about $500 every 90 d. Yeah. There's worse than that too." (60-y-old male) "It was to the point that, um, now that I'm retired and can get the, I had to get the part D part D drug coverage before I was just paying out of pocket. But they kept going up and up and up on it. It made it prohibitive once I got that, uh, part D is quite manageable." (72-yold female) diagnosis comes as a shock to individuals as they try to come to terms with their new reality. Public awareness of the disease remains low, which then leads to a poor understanding of long-term consequences, misunderstandings with treatment, and an overall uneasiness with the disease itself. Consequently, one recommendation that emerged was an emphasis on creating an interactive team effort involving those who are diagnosed and their care team. In addition, another subject reported that the emphasis placed on treatment by their doctor helped them understand the need for proper adherence and that doctors' advice needs to be trusted and followed as close as possible. While patients become more comfortable with their diagnosis over time, this lack of knowledge, in the beginning, leads to poor adherence and a more rapid progression of the disease. Proper education in the beginning stages was beneficial; however, not all patients received adequate training on the course of treatment or technique, leading to further challenges for those diagnosed. Patients who researched the disease on their own, as opposed to relying solely on information from their doctor, showed a deeper understanding of the disease and how to better maintain it. A large number of patients reported a desire for more individualized information gained from outside resources is not catered to their individual health.
To overcome these challenges, patients reported strategies that have specifically worked for them to better adhere to their regimens and indicated preferences for certain medications which are easier to administer or reduce side effects. Keeping the medication on hand for individuals proved to be a multifactorial issue as pharmacies often do not have the medications in stock when they are needed, or high costs deter patients from picking up the medications when they run out. The cost of medications is fairly expensive for those without insurance coverage and often resulted in those individuals making tough decisions related to their care, however, those with adequate insurance coverage reported associated costs to be fairly manageable. Insurance companies often influence regimens as they might not cover the drops when patients ran out due to wasted drops or did not cover the patients preferred drops. This can force patients to use drops they are not comfortable with or that have increased side effects.

DISCUSSION
Many chronic disease states are controlled with orally administered medications, but the management of glaucoma requires patients to administer 1 or more eye drops per day, posing unique and complex challenges to adherence. We found that patients' self-confidence in their ability to administer and adhere to their medications was associated with significantly higher self-reported adherence levels, as determined by validated scales. These findings highlight the significant role patients' individualized beliefs and attitudes may have on longitudinal medication use, suggesting the need for providers to engage their patients in active discussion about their views on topical therapy.
Echoing similar, earlier results, several barriers to adherence were observed in both the surveying and interviewing: suffering from side effects, difficulty remembering, not having the drops at dosing time, or having difficulty with drop administration. 2 In addition, the cost of glaucoma medications and delays in acquiring refills were barriers to reliably having medication on hand at dosing time, which negatively impacts adherence and is suggestive of the need for providers to investigate the availability of lower-cost generics if they are to more adequately address cost-driven nonadherence. However, in contrast to earlier observations, the current study found that older patients and those that reported having a better understanding of the disease after conducting their own research reported better adherence to their regimen.
The majority of the survey cohort reported having an excellent understanding of the contributing factors to and risks of uncontrolled glaucoma. Notably, the overwhelming majority of these patients had used eye drops for 1 or more years (92.7%), including 71% who had been using them for 3 years or more. However, interviews suggested that many patients lacked proper instruction on instillation technique in the early stages of their disease management, the results of which can include a lower overall level of adherence in the first year and may impact future adherence. Specifically, Newman-Casey et al 4 found that adherence patterns observed in patients during the first year of eye drop treatment were similar to patterns over the next 3 years. Therefore, behaviors that are established early have a lasting impact on disease management, whether positive or negative, which reinforces the need for early intervention on patients diagnosed and beginning topical treatment for glaucoma.
Further, we found that patients had a desire for more individualized information from their health care team to better understand their disease course, especially in the early stages. The learning curve at the time of diagnosis poses a challenge for patients who are not properly educated by their prescribers, nurses, or pharmacists at the time of prescribing or dispensing. This is supported by other studies that found patients who had fewer concerns about the long-term risks of glaucoma or a lower belief that their medications were necessary were associated with nonadherence. 6,12 Such findings highlight the need to provide more comprehensive information about the disease state and proper instillation technique at treatment initiation to establish and foster behaviors more likely to lead to higher and sustained levels of adherence.
Lacking prescribed medications at dosing time was also a leading barrier and due to several reasons. First, some patients, who presumably were prescribed multiple doses throughout the day, reported that they sometimes did not have their drops at dosing time, which suggests the need for ongoing reminders or assistive devices to aid in "nudging" patients with intentions to take the medication as prescribed. In addition, providers could encourage the dosing of medication at bedtime to pair use with a planned, daily activity. Medication cost is a common barrier to adherence, and half of our patients surveyed felt that their medications were unreasonably priced, and this was problematic even in patients with significant insurance coverage. Importantly, patients reported that higher quality coverage made it easier for them to afford their eye drops, which is an important factor for payers to consider as new therapies are expected to come to market. Relatedly, some patients have preferences for certain medications based on their ability to use a specific bottle or out of a desire for a more favorable side effect profile. Being able to use these medications over others can improve adherence; however, payers may prefer different products that prevent patients and doctors from having certain options. When the patient can afford the medication, a lack of pharmacy stock and forgetting to request refills were 2 factors that sometimes delay timely refills and also prevent adherence. Fortunately, community pharmacies remain one of the most accessible health resources, irrespective of location, providing mechanisms to ensure patients have ready access to their prescribed therapies. However, interruptions to such access (ie, requiring the use of in-network or specialty pharmacies) remain prominent threats in today's managed care environment.
These findings should be considered in light of several limitations in this study. First, although the adherence scales are validated, they were all self-reported in this analysis, so it is possible the actual level of adherence is different than what was reported. Second, the sample size is relatively low, and it was conducted at a single site of an academic medical center; therefore, findings may have limited generalizability beyond this clinic. Third, we did not track the characteristics of those who declined to participate, and this could introduce bias. In addition, patients volunteered and were incentivized to participate; therefore, findings may not reflect the beliefs and behaviors of those who did not respond.
Despite these limitations, this study provided new information and supported previous findings related to barriers and facilitators of adherence among patients taking glaucoma medications. These findings will be utilized to construct a planned mobile health intervention using a library of tailorable text messages to provide meaningful support to adults with glaucoma. Ultimately, the goal of such research is to provide a resource to assist patients in developing self-confidence in the management of glaucoma. Such support is hypothesized to increase or maintain high levels of adherence, which may, over time, improve disease control and preserve vision.