Applying the resources and supports in self-management framework to examine ophthalmologist-patient communication and glaucoma medication adherence

Abstract

Little is known about how ophthalmologist-patient communication over time is associated with glaucoma patient long-term adherence. The purpose of our study was to examine the association between provider use of components of the resources and supports in self-management model when communicating with patients and adherence to glaucoma medications measured electronically over an 8-month period. In this longitudinal prospective cohort study, the main variables studied were ophthalmologist communication-individualized assessment, collaborative goal setting and skills enhancement. Patients with glaucoma who were newly prescribed or on glaucoma medications were recruited from six ophthalmology clinics. Patients’ baseline and next follow-up visits were videotape-recorded. Patients were interviewed after their visits. Patients used medication event monitoring systems (MEMS) for 8 months after enrollment into the study, and adherence was measured electronically using MEMS for 240 days after their visits. Two hundred and seventy-nine patients participated. Patient race and regimen complexity were negatively associated with glaucoma medication adherence over an 8-month period. Provider communication behaviors, including providing education and positive reinforcement, can improve patient adherence to glaucoma medications over an 8-month period.

Introduction

Glaucoma is similar to high blood pressure in that it is a chronic asymptomatic disease. Yet glaucoma differs from other chronic diseases in that rather than taking oral medications, most patients need to instill eye drops long term with the ultimate goal of reducing intraocular pressure to preserve vision [1]. Prior work has found that non-adherence to using glaucoma medications is a significant problem and thus a concern for vision loss from glaucoma [2–7].

Adherence to medications is a critical self-management behavior for a chronic condition like glaucoma. One framework that has been applied to improving self-management behavior in individuals with chronic diseases is the resources and supports for self-management [8]. This framework uses key provider communication behaviors including individualized assessment, collaborative goal setting and skills enhancement that are important for providers to engage in when interacting with patients with chronic disease to improve self-management skills. We applied this framework to examine how key provider communication behaviors were associated with long-term adherence to glaucoma medication over an 8-month period.

Individualized assessment involves exploring the patient’s personal and cultural perspective on the chronic disease and how it impacts their life [8, 9]. It is important for ophthalmologists to conduct an individualized assessment with glaucoma patients so that they can better understand patients’ views of glaucoma and glaucoma medication treatment options since each topical class of medications differs from other classes [2–4, 6, 10].

Collaborative goal setting involves empowering patients and asking for their input when treatment decisions are being made [8]. Collaborative goal setting has been examined in several different disease states such as cardiac rehabilitation, asthma, osteoporosis, psychiatry and diabetes [11–16]. In a prior study involving 51 patients who were newly diagnosed and prescribed glaucoma medications, we found that collaborative goal setting occurred infrequently [17].

Skills enhancement involves physicians providing information required for glaucoma self-management and encouraging and reinforcing positive patient behaviors [18]. It is important for ophthalmologists to educate patients about glaucoma so that patients understand why they are taking glaucoma medication. Additionally, ophthalmologists need to educate patients about the medications they are taking so that patients understand how to use them appropriately. At the same time as providing education to patients, ophthalmologists also need to motivate patients to adhere to their medications and give them positive reinforcement on their medication-taking.

Therefore, the purpose of our study was to (i) describe the extent to which providers use components of the resources and supports in self-management model (individualized assessment, collaborative goal setting and skills enhancement) when interacting with patients across time and (ii) examine the association between provider use of components of the resources and support in self-management model when communicating with patients and patient adherence to glaucoma medications measured electronically over an 8-month period based on assessment of percent doses taken over 240-day period, percent correct number of doses taken each day, and percent prescribed doses taken on time.

Methods

Procedure

This study, approved by the Institutional Review Boards at the University of North Carolina, Duke University, Emory University and the University of Utah, was HIPAA compliant. English-speaking adult glaucoma patients were enrolled at six geographically distinct ophthalmology clinics located in four states. Two sites were private offices and four were affiliated with academic ophthalmology departments. At each site, clinic staff referred eligible patients to research assistants who were based at the clinics. Written patient and provider consent was obtained. Providers completed a short demographic questionnaire after providing consent.

Participants had their medical visit videotape-recorded when they enrolled into the study. Videotapes were kept if they fit one of two criteria: (i) the patient was diagnosed with glaucoma and glaucoma medications were prescribed for the first time or (ii) patients were already taking glaucoma medications. Patients who met either of these criteria were followed for the 8-month study period.

Patients were interviewed after their medical visits. The patient’s glaucoma eye drops were placed into large prescription vials with medication event monitoring system (MEMS) caps on top which electronically assessed patient adherence [3]. The patient’s next glaucoma visit was also video-tape recorded. Patients used the MEMS caps for 8 months after their baseline video-taped visit. The patient’s next follow-up visit, which typically occurred four to six weeks later, was also video-taped recorded.

Measurement

Personal characteristics

Self-reported patient race was measured as a categorical variable (White, African American, Asian, Native American and Hispanic) and then recoded into African American and non-African American. The majority of the non-African American patient sample was White (91%). The number of glaucoma medications a participant was taking was recorded. Whether the patient was prescribed glaucoma medication for the first time during the medical visit or was already on glaucoma medication before the visit was also recorded.

Each participant received the rapid estimate of adult literacy in medicine (REALM). This is a validated, rapid screening instrument designed to identify patients who have difficulty reading common medical and lay terms that are routinely used in patient education materials [19]. We chose the REALM because it has high face validity and high criterion validity, it has been well received by patients, and it only takes 2–3 min to administer and score [19]. Patient scores on the REALM correspond to reading levels (score of 0–60 = eighth grade and below, 61–66 = ninth grade and above).

Physicians reported their age, gender and race (White, African American, Asian, Native American and Hispanic).

All participants were administered a 10-item general glaucoma medication adherence self-efficacy questionnaire and a 6-item eye drop technique self-efficacy scale [20, 21]. Patients were given three possible response choices for the self-efficacy items: not at all confident (coded as 1), somewhat confident (coded as 2) and very confident (coded as 3). Scores on the glaucoma medication adherence scale could range from 3 (lower self-efficacy) to 30 (higher self-efficacy). The scale had a Cronbach’s alpha of 0.94. Scores on the eye drop technique self-efficacy scale could range from 3 (lower self-efficacy) to 18 (higher self-efficacy). The scale had a Cronbach’s alpha of 0.88. Both of the self-efficacy scales have strong psychometric properties [20, 21]. Outcome expectations were measured using four items designed to assess how much patients believed attending their ophthalmic clinical visits and taking medications would help their glaucoma and vision [20]. Responses ranged from 1 (not at all) to 9 (extremely) for each item. Scores could range from 4 to 36 on the outcome expectations scale. The scale had a Cronbach’s alpha of 0.94 in the current sample.

Communication measures

All of the medical visit video-tapes were transcribed into text verbatim with identifiers removed. A detailed coding tool was developed over a 1-year period. The transcripts were reviewed by a research assistant who met twice a month with the investigators to develop and refine the coding rules. The transcripts of the medical visits from each time point were coded. Table I presents the definition of each coding category.

Table I.

Definitions of the coding variables

Coding category	Definition
Glaucoma
Physical changes that can occur with glaucoma and/or how to manage these changes	This includes education about the physical changes associated with glaucoma and how to manage physical changes that can occur with glaucoma.
Emotional changes that can occur with glaucoma and/or how to manage these changes	This includes education about the emotional difficulty of having glaucoma, loss of vision, depression, or anxiety and how to manage emotional changes that can occur with glaucoma.
Diagnosis	This includes education about what glaucoma is.
Family history	This includes education about family history being a risk factor for glaucoma.
Goals of treatment	This includes education about what the goals of treatment are to prevent vision loss.
How to problem solve	This includes provider giving advice on how the patient can problem solve if he or she runs into trouble with managing their glaucoma.
Intraocular pressure	This includes education about what intraocular pressure (IOP) is.
Likelihood of long-term therapy	This includes education about glaucoma being a chronic condition that requires long-term therapy.
Management plan (ways to manage glaucoma without medications)	This includes education about other ways to manage glaucoma other than medications (e.g. surgery).
Prognosis	This includes education about the patient’s prognosis, how the patient is doing, and the potential for vision loss.
Glaucoma medications
Adherence and adherence strategies	This includes education about adherence, missed doses, and/or extra doses taken.
Amount/dose (number of drops)	This includes education about the number of drops to take each time.
Cost/insurance	This includes education about the cost of medications, prescription drug insurance, and/or drug assistance programs.
Eyelid closure and nasolacrimal occlusion when applying topical medications	This includes education about how to improve absorption of the medication or reduce local side effects.
Fear/concerns/barriers	This includes education about any fears or concerns (real or perceived) about the medications the patient is currently taking or will be taking in the future.
Frequency of use	This includes education about how often a medication should be used on a daily basis. Also, it includes when it should be used before, during, or after certain situations.
Generic/brand	This includes education about the equivalence of medications (brand and generic) or whether the medicine that is being prescribed is available as a generic or brand name.
How well medication is working	This includes education about how well the medication works, and how effective it is or is not.
How to administer	This includes education about how to administer eye drops. Also, it includes education about the amount of time to wait between administering two or more eye drop medications.
Side effects	This includes education about side effects of medication (real or perceived), as well as doctor offering explanations or reassurances about possible side effects brought up by patient.
Importance of use	This includes education about why the glaucoma medication is important to use.
Last time used drops	This includes information about the last time the patient used eye drops and how it relates to current IOP reading.
Length of use	This includes education about how long the patient will need to use the medicine.
Name of medication	This includes discussion about the name of the medication that the patient is prescribed.
Non-glaucoma medications	This includes any education of non-glaucoma medications, like steroid eye drops, Latisse, high blood pressure medicines, or cold medications.
Purpose	This includes education about the purpose of the eye drops (e.g. lowers IOP, etc.).
Supply	This includes education about how much medication the provider will prescribe, how long each bottle should last, how to keep track of how many doses are remaining in the bottle, and when they may need refills.
Which eye	Includes information about which eye to administer the drops.
Individualized assessment
Asks the patient about view of glaucoma or its treatment	This occurs when the provider asks the patient about his or her views on glaucoma or its treatment.
Asks the patient about how glaucoma will impact his or her life	This occurs when the provider asks the patient about how glaucoma will impact his or her life.
Asks about confidence in using medication regularly	This occurs when the provider asks about confidence in using glaucoma medication regularly.
Asks about intention to adhere in the future	This occurs when the provider asks the patient about his/her intention to adhere to glaucoma medications in the future.
Collaborative goal setting
Patient is given choices about treatment	This occurs when the provider presents one or more treatment options to the patient.
Provider asks for preferences or ideas on treatment	This occurs when the provider asks for preferences or ideas on treatment.
Provider asks patient to talk about treatment goals	This occurs when the provider asks the patient to talk about treatment goals.
Patient helps set treatment goals	This occurs when the patient helps set treatment goals
Encouragement/reinforcement
Encourages patient to take their glaucoma medications	This occurs when the provider encourages the patient to take their glaucoma medications.
Encourages patient to regularly come to their appointment	This occurs when the provider encourages the patient to regularly come to their appointments.
Gives positive reinforcement about glaucoma medication-taking	This occurs when the provider gives the patient positive reinforcement about the patient’s glaucoma medication-taking.

Using the coding tool for transcribed medical visits, coders recorded whether the provider educated the patient in the following areas about glaucoma: (i) physical changes with glaucoma and/or how to manage these changes, (ii) emotional changes with glaucoma and/or how to manage these changes, (iii) diagnosis, (iv) family history, (v) goals of treatment, (vi) how to problem solve, (vii) intraocular pressure, (viii) likelihood of long-term therapy, (ix) ways to manage glaucoma other than with medications and (x) prognosis. We summed the total number of areas educated about at each visit. We then added these two variables together to create ‘total amount of glaucoma education provided across both visits’.

Coders also recorded whether the provider educated the patient in the following areas about glaucoma medications: (i) adherence and adherence strategies, (ii) amount/dose, (iii) cost/insurance, (iv) eyelid closure and nasolacrimal occlusion when applying topical medications, (v) fear/concerns/barriers, (vi) frequency of use, (vii) generic/brand, (viii) how well medication is working, (ix) how to administer, (x) side effects, (xi) importance of use, (xii) last time used drops, (xiii) length of use, (xiv) name of medication, (xv) non-glaucoma medications, (xvi) purpose, (xvii) supply and (xviii) which eye to install the drops. We summed the total number of areas educated about at each visit. We then added these two variables together to create ‘total amount of glaucoma medication education provided across both visits’.

Coders recorded whether the provider conducted the following aspects of an individualized assessment: (i) asks the patient about their views of glaucoma and/or its treatment, (ii) asks the patient about how glaucoma will impact their life, (iii) asks about confidence in using glaucoma medication regularly and (iv) asks about intention to adhere to glaucoma medications in the future. For the multivariate analysis, we summed the total number of aspects of individualized assessment the provider performed. However, individualized assessment occurred so infrequently that we created a dichotomous variable ‘provider performed an individualized assessment during one or both visits’.

Coders recorded whether the provider participated in the following aspects of collaborative goal setting: (i) patient is given choices about treatment, (ii) provider asks for preferences or ideas on treatment, (iii) provider asks patient to talk about treatment goals and (iv) patient helps set treatment goals. For the multivariate analysis, we summed the total numbers of aspects of collaborative goal setting the provider performed. However, collaborative goal setting occurred so infrequently that we created a dichotomous variable ‘provider engaged in collaborative goal setting in one or both visits’.

Additionally, coders recorded if the ophthalmologist encouraged the patient to take their glaucoma medications and encouraged the patient to regularly come to their appointments. Coders also recorded whether the provider gave the patient positive reinforcement about the patient’s glaucoma medication-taking.

Two clinics had ophthalmology fellows see some of the enrolled patients while two other clinics also had ophthalmic technicians see some of the enrolled patients. Informed consent was obtained from these providers as well. If any one of these providers, including the physician, educated the patient, it was counted as education in the categories discussed earlier.

Three research assistants coded 25 of the same transcripts throughout the study period to assess inter-rater reliability which was calculated using inter-rater correlations. Inter-rater reliability for the variables ranged from 0.70 to 1.0. If there was not enough variability to calculate reliability, we calculated percent agreement between the coders; percent agreement was 100% for these variables.

Adherence measures

Medication adherence over an 8-month period after the video-taped visit was evaluated via electronic data from the MEMS caps system (MeadWestvaco AARDEX) [3]. Whether the patient took 80% or more of their prescribed doses was measured from the MEMS caps using the following formula: adherence = (number of doses used during the past 8 months divided by the number of prescribed doses) multiplied by 100. We dichotomized the variable since it was skewed toward patients being highly adherent. Participants were considered adherent if they used 80% or more of the prescribed doses (79.5% or above was rounded to 80% or more) and they were classified as non-adherent if they used <80% of the prescribed doses as suggested in prior research [23].

We also used the electronic data from the MEMS caps to examine the percent of doses taken on time during the 8-month period after the video-taped visits [3]. If patients were on once a day dosing, taking it on time was taking it every 24 h plus or minus 6 h. If it was twice a day dosing, taking it on time was every 12 h plus or minus 4 h. We also examined using the MEMS data the percent of correct number of prescribed doses taken each day during the 8-month period after the video-taped visit [3]. We dichotomized these adherence measures since they were skewed toward patients being highly adherent. Participants were considered adherent if they used 80% or more of the prescribed doses (79.5% or above was rounded to 80% or more) on time and if they took the correct number of doses each day 80% or more of the time.

If the participant was on more than one glaucoma medication, an adherence measure was created for each medication and then an overall adherence variable was created by adding together the participant’s adherence for each glaucoma medication and dividing it by the number of glaucoma medications the participant was using. This was done for each of the three adherence measures.

Analysis

We set the a priori level of statistical significance at P < 0.05. First, we ran descriptive statistics. Second, we examined the bivariate relationships between variables using Pearson correlation coefficients, chi-square statistics and t-tests as appropriate. We conducted generalized estimating equations (GEEs) to examine how patient age, gender, race, glaucoma medication adherence self-efficacy, glaucoma outcome expectations, whether the patient was newly prescribed glaucoma medication on the day of the baseline video-taped visit, number of glaucoma medications the patient was on, provider engaged in collaborative goal setting at one or both visits, provider conducted an individualized assessment at one or both visits, provider encouraged patient to take medications at one or more visits, provider gave positive reinforcement about glaucoma medication taking at one or both visits, total amount of glaucoma medication education provided at both visits, total amount of glaucoma education provided at both visits, physician age and physician gender were associated with: (i) whether the patient was 80% or more adherent to their glaucoma medications, (ii) whether the patient took 80% or more of their doses on time and (iii) whether patients took the correct number of doses each day 80% or more of the time according to the MEMS caps during the 240-day period after the video-taped visit. Physician race could not be included in the multivariable analysis because there was only one non-White physician.

Results

Fifteen physicians who cared for glaucoma patients agreed to participate in the study; one physician refused to participate for a participation rate of 94%. Fourteen physicians were White and one was African American. Ten physicians were male (66.7%). Physician age ranged from 26 to 66 years (mean 40.8 years, SD 11.7 years).

Eighty-six percent of eligible patients participated in the study (N = 279). We have useable video-tapes of the baseline visits for 275 of the 279 patients. Six patients were lost between their baseline visits and their next follow-up visit. We had useable video-tapes of the follow-up visit for 264 of the 273 patients who returned for follow-up visits.

Table II presents the patient demographics. Forty-one percent of the sample was male and 35.5% were African American. Eighteen percent of patients were prescribed glaucoma medications for the first time. Eighty-nine percent of the enrolled patients had MEMS caps data for the 240-day period after their video-taped visit (N = 248). Those who did not have data either did not return their MEMS caps (N = 24) or if they returned their MEMS caps, there was not useable data (N = 7) (e.g. data was not downloadable due to hardware problems in the caps).

Table II.

Participant characteristics (N = 279)

	Percent (N)
Gender
Male	40.9 (144)
Female	59.1 (165)
Race
African American	35.5 (99)
Non-African American	64.2 (179)
Newly prescribed glaucoma medications at visit or was on glaucoma medication before visit
Newly prescribed at visit	18.3 (51)
Was on glaucoma medications before visit	81.7 (228)
REALM
Eighth grade or lower	14.0 (39)
Ninth grade or higher	84.2 (235)
Age in years (mean ± SD)	65.8 ± 12.8

Table III presents whether physicians educated about different aspects of glaucoma at one or both visits. The areas educated about most often during at least one of the visits were (i) intraocular pressure (42%), (ii) the physical changes that occur with glaucoma (35%) and (iii) the treatment goals of glaucoma (29%). Very few areas were discussed at both video-taped visits. Intraocular pressure was discussed the most, with it being discussed at both visits for 29% of the patients.

Table III.

Areas of education provided about glaucoma and glaucoma medications at the initial and follow-up visits (N = 279)

Areas provider educated about	Neither visit % (N)	One visit % (N)	Both visits % (N)
Glaucoma
Physical changes that can occur with glaucoma and/or how to manage these changes	58.4 (163)	35.1 (98)	6.5 (18)
Emotional changes that can occur with glaucoma and/or how to manage these changes	99.6 (278)	0.4 (1)	0 (0)
Family history	90.7 (253)	9.3 (26)	0 (0)
Goals of treatment	69.2 (193)	29.0 (81)	1.8 (5)
How to problem solve	96.1 (268)	3.9 (11)	0 (0)
Intraocular pressure	28.3 (79)	42.3 (118)	29.4 (82)
Likelihood of long-term therapy	91.4 (255)	8.2 (23)	0.4 (1)
Management plan (ways to manage glaucoma without medications)	81.7 (228)	15.8 (44)	2.5 (7)
Glaucoma medications
Adherence and adherence strategies	72.0 (201)	25.8 (72)	2.2 (6)
Amount/dose (number of drops)	93.5 (261)	6.5 (18)	0 (0)
Fears/concerns/barriers	95.0 (265)	4.7 (13)	0.4 (1)
Frequency of use	75.6 (211)	22.9 (64)	1.4 (4)
Generic/brand	88.5 (247)	10.8 (30)	0.7 (2)
How to administer	83.5 (233)	16.1 (45)	0.4 (1)
Side effects	65.9 (184)	30.1 (84)	3.9 (11)
Last time used drops	99.3 (277)	0.4 (1)	0.4 (1)
Length of use	100 (279)	0 (0)	0 (0)
Name of medication	91.4 (255)	8.6 (24)	0 (0)
Purpose	82.4 (230)	17.2 (48)	0.4 (1)
Which eye	89.2 (249)	10.0 (28)	0.7 (2)

Table III also presents whether physicians educated about different aspects of glaucoma medication at one or both visits. The areas educated about most often during at least one of the visits included (i) side effects (30%), (ii) adherence and adherence strategies (26%), (iii) frequency of use (23%), (iv) purpose (17%) and (v) how to administer (16%). Physicians almost never educated on any of the areas about glaucoma medications at both visits.

Table IV presents whether physicians conducted an individualized assessment and whether they used collaborative goal setting at one or both visits. The only areas where physicians conducted an individualized assessment at one visit were asks about confidence in using glaucoma medication regularly (20%) and asks the patient their point of view about glaucoma and/or its treatment (8%). Physicians rarely conducted an individualized assessment in any area at both visits.

Table IV.

Extent to which physicians engaged in individualized assessment, collaborative goal setting, encouragement/reinforcement or providing written/video materials at the initial and follow-up visits (N = 279)

	Neither visit % (N)	One visit % (N)	Both visits % (N)
Individualized assessment
Asks patient about view of glaucoma or its treatment	92.5 (258)	7.5 (21)	0 (0)
Asks the patient about how glaucoma will impact his or her life	98.9 (276)	1.1 (3)	0 (0)
Asks about confidence in using medication regularly	79.2 (221)	19.7 (55)	1.1 (3)
Asks about intention to adhere in the future	97.8 (273)	2.2 (6)	0 (0)
Collaborative goal setting
Patient is given choices about treatment	81.4 (227)	16.5 (46)	2.2 (6)
Provider asks for preferences or ideas on treatment	85.7 (239)	11.5 (32)	2.9 (8)
Provider asks patient to talk about treatment goals	100 (279)	0 (0)	0 (0)
Patient helps set treatment goals	99.6 (278)	0.4 (1)	0 (0)
Encouragement / reinforcement
Encourages patient to take their glaucoma medications	69.9 (195)	26.2 (73)	3.9 (11)
Encourages patient to regularly come to their appointment	97.1 (271)	2.9 (8)	0 (0)
Gives positive reinforcement about glaucoma medication-taking	59.1 (165)	33.7 (94)	7.2 (20)

The areas in which physicians most often engaged in collaborative goal setting at one visit were patient is given choice about treatment (17%) and physician asks for patient preferences or ideas on treatment (12%). Physicians rarely engaged in any aspects of collaborative goal setting during both visits. Providers rarely encouraged patients to take their glaucoma medications at both visits and they rarely gave positive reinforcement about medication-taking at more than one visit.

Table V presents the results of the generalized estimating equation predicting whether the patient was 80% or more adherent during the 240-day period after the baseline visit. African American patients were significantly less likely to be adherent than non-African American patients (odds ratio = 0.29, 95% confidence interval = 0.16, 0.52). Patients with higher glaucoma medication adherence self-efficacy were significantly more likely to be adherent than patients with lower self-efficacy (odds ratio = 1.10, 95% confidence interval = 1.02, 1.78). Patients seeing older physicians were significantly less likely to be adherent (odds ratio 0.95, 95% confidence interval 0.91, 095).

Table V.

Multivariable generalized estimating equation results predicting whether the patient is 80% or more adherent during the 240-day period after the visit according to MEMS caps (N = 239)

Independent variables	Patient is 80% or more adherent OR (95% CI)
Patient age	0.99 (0.96, 1.02)
Patient gender-female	1.21 (0.59, 2.46)
Patient race-African American	0.29 (0.16, 0.52)***
Glaucoma medication adherence
self-efficacy	1.10 (1.02, 1.78)*
Eye drop technique self-efficacy	0.98 (0.81, 1.19)
Glaucoma outcome expectations	1.15 (1.03, 1.29)*
Newly prescribed glaucoma medications	0.64 (0.25, 1.65)
Number of glaucoma medications	0.98 (0.51, 1.89)
Provider engages patient in collaborative
goal setting at one or both visits	0.54 (0.29, 0.98)*
Provider conducts an individualized assessment
at one or both visits	1.23 (0.68, 2.23)
Provider encourages patient to take medications
at one or both visits	1.11 (0.46, 2.71)
Provider gives positive reinforcement about
glaucoma medication taking	3.37 (1.69, 6.71)***
Total amount of glaucoma medication
education provided at both visits	0.89 (0.71, 1.11)
Total amount of glaucoma education
provided at both visits	1.09 (0.84, 1.42)
Physician age	0.95 (0.91, 0.99)**
Physician gender-female	0.50 (0.15, 1.63)

OR, Odds Ratio; 95%CI, 95%, Confidence Interval.

*P < 0.05, **P < 0.01, ***P < 0.001.

Interestingly, if providers engaged in collaborative goal setting with patients at either visit, patients were significantly less likely to be adherent (odds ratio = 0.54, 95% confidence interval 0.29, 0.98). If providers gave positive reinforcement about the patient’s medication taking at one or more visits, patients were significantly more likely to be adherent (odds ratio = 3.37, 95% confidence interval 1.69, 6.71).

Table VI presents the GEE results predicting whether patients took their doses on time 80% or more of the time during the 8-month period. African Americans were significantly less likely to take their doses on time than non-African Americans (odds ratio = 0.38, 95% confidence interval 0.25, 0.58). Patients on more glaucoma medications were less likely to take their doses on time (odds ratio = 0.62, 95% confidence interval = 0.39, 0.98).

Table VI.

Multivariable generalized estimating equations predicting the percent doses taken on time and the percent correct number of doses taken each day during the 240-day period after the visit according to MEMS caps (N = 239)

Independent variables	Percent doses Taken on time OR (95% CI)	Percent correct number of doses taken each day OR (95% CI)
Patient age	1.00 (0.98, 1.03)	1.00 (0.97, 1.03)
Patient gender-female	1.43 (0.65, 3.16)	1.28 (0.75, 2.20)
Patient race-African American	0.38 (0.25, 0.58)***	0.41 (0.27, 0.61)***
Glaucoma medication adherence self-efficacy	1.00 (0.92, 1.09)	0.99 (0.96, 1.03)
Eye drop technique self-efficacy	1.17 (0.94, 1.44)	1.13 (0.98, 1.31)
Glaucoma outcome expectations	1.08 (0.99, 1.18)	1.08 (0.96, 1.21)
Newly prescribed glaucoma medications	1.17 (0.47, 2.91)	0.78 (0.36, 1.67)
Number of glaucoma medications	0.62 (0.39, 0.98)*	1.07 (0.65, 1.76)
Provider engages patient in collaborative goal setting at one or both visits	0.82 (0.46, 1.45)	0.44 (0.29, 0.65)***
Provider conducts an individualized assessment at one or both visits	0.99 (0.58, 1.69)	0.75 (0.49, 1.14)
Provider encourages patient to take medications at one or both visits	1.45 (0.80, 2.65)	1.84 (1.24, 2.73)**
Provider gives positive reinforcement about glaucoma medication taking	2.80 (1.44, 5.43)**	2.86 (1.76, 4.65)***
Total amount of glaucoma medication education provided at both visits	0.81 (0.65, 1.00)	1.18 (0.97, 1.44)
Total amount of glaucoma education provided at both visits	1.35 (1.03, 1.78)*	0.96 (0.75, 1.25)
Physician age	0.97 (0.94, 1.01)	0.98 (0.96, 1.00)
Physician gender-female	0.65 (0.31, 1.36)	0.83 (0.47, 1.49)

OR, Odds Ratio; 95%CI, 95%, Confidence Interval.

*P < 0.05, **P < 0.01, ***P < 0.001.

If the provider gave positive reinforcement about the patient’s medication taking at one or more visits, the patient was significantly more likely to take their medications on time (odds ratio = 2.8, 95% confidence interval 1.44, 5.43). If providers gave more glaucoma education during the two visits, patients were significantly more likely to be adherent (odds ratio = 1.35, 95% confidence interval 1.03, 1.78).

Table VI also presents the GEE results predicting whether patients took the correct number of doses each day 80% or more of the time during the 8-month period. African Americans were significantly less likely to take the correct number of doses each day than non-African Americans (odds ratio = 0.41, 95% confidence interval 0.27, 0.61). If the provider engaged in collaborative goal setting with the patient at one or both visits, the patient was significantly less likely to take the correct number of doses each day (odds ratio = 0.44, 95% confidence interval 0.29, 0.65). If the provider encouraged the patient to take their medication at one or both visits then the patient was significantly more likely to take the correct number of doses each day throughout the 8-month period (odds ratio = 1.84, 95% confidence interval 1.24, 2.73). Also, if the provider gave positive reinforcement to the patient about their medication taking, the patient was significantly more likely to take the correct number of doses each day (odds ratio = 2.86, 95% confidence interval 1.76, 4.65).

Discussion

One of our most important findings is that the provider giving the patient positive reinforcement about their glaucoma medication taking was significantly associated with all three measures of adherence. This finding supports the resources and supports in self-management framework’s assertion that giving positive reinforcement to patients can improve their health behaviors, which in this case was patient medication taking [8]. In addition, if providers encouraged the patient to take their glaucoma medications, then the patient was significantly more likely to take the correct number of doses each day.

These findings emphasize the importance of providing positive reinforcement on patient medication taking. Simple statements such as ‘Alright, sounds like you’re doing a great job with the drop, so keep it up’ or ‘You’ve done a very nice job taking your drops. I’m impressed’ could reinforce their positive adherence behavior. The findings also indicate that encouraging patients to take their glaucoma medications could improve patient adherence. Provider could say something like ‘Keep up the good work’, ‘You’ve done great. I want you to keep it up’ or ‘So try your best to remember to take it. It only works if you take it.’. Future intervention studies should be designed to motivate providers to: (i) give patients positive reinforcement about their glaucoma medication taking behavior and (ii) encourage patients to take their glaucoma medications.

Despite the assumption that collaborative goal setting between providers and patients under the resources and supports in self-management framework should positively impact patient health behavior, we found that if providers engaged in collaborative goal setting with patients, then the patients were less likely to take the correct doses per day and they were less likely to be 80% or more adherent over the 8-month period [8]. Perhaps providers were more likely to engage in collaborative goal setting with patients who they perceived were less adherent in an attempt to increase adherence.

An interesting finding was that if providers educated more about glaucoma then the patients were significantly more likely to take their glaucoma medication doses on time. This supports the part of the resources and supports in self-management framework which emphasizes the importance of educating patients about their disease state so they can better understand their condition and be motivated to self-manage the condition on their own [8]. Our findings suggest that providers should make sure to educate patients about glaucoma because it can influence their medication adherence. Future work should design intervention studies to improve patient knowledge of their glaucoma.

We found that patients on more glaucoma medications were less likely to take 80% or more of their doses on time. Wide fluctuations in intraocular pressure, as would be expected from glaucoma medication doses taken off schedule, are associated with greater vision loss in patients with advanced glaucoma, despite low mean intraocular pressure [24]. Our finding that patients on simpler regimens were more adherent to their glaucoma medications is similar to what other researchers have found previously [3, 25, 26]. This suggests that ophthalmologists might want to simplify glaucoma patients’ medication regimens, if possible, so they have to take fewer medications each day.

Additionally, even after controlling for provider-patient communication, African American patients were significantly less adherent to their medications on all three measures over the 8-month period. Prior studies that have examined the relationship between African American race and adherence to glaucoma medications have had contradictory results [6, 21, 27–30]. Despite the contradictory prior results, our findings suggest that ophthalmologists should make sure to communicate with African American patients about what health beliefs or problems are preventing them from being adherent to their glaucoma medications. Future research is needed to better understand why patient race is significantly associated with medication adherence even after controlling for other personal and cultural characteristics and provider-patient communication.

Our study was the first longitudinal study to examine ophthalmologist-patient communication across two visits, prior research has only examined one visit [22, 27, 31, 32]. One of the most interesting findings from our work is that providers very rarely provide education about glaucoma or glaucoma medication areas during two consecutive visits, with the exception of intraocular pressure which was discussed at both visits for less than one-third of the patients. However, only discussing intraocular pressure with less than one-third of patients at both visits is concerning, because one of the main goals of glaucoma treatment is to lower intraocular pressure. Therefore, it is important for providers to discuss what a patient’s intraocular pressure is at each visit so that patients know what their pressure is and why lowering intraocular pressure is important.

This study has several limitations. Providers and patients both knew the visit was being recorded but they did not know the study hypotheses. Our study is limited in that we could not track the characteristics of non-participants so we cannot compare the characteristics of participants and non-participants. Thus, we cannot estimate selection bias and how that might impact the generalizability of our results. Additionally, our coders counted the patient being educated about glaucoma or glaucoma medications during visits regardless of whether a physician, technician or fellow provided it, which limits our ability to examine differences in education by provider type. A limitation is that we coded the data this way so we cannot separate out physician, technician and fellow provision of education. Another limitation of our study is the lack of variation in physician race and how provider race might affect physician-patient communication and adherence.

Despite the limitations, this longitudinal prospective cohort study provides important new findings which indicate that providers giving patients positive reinforcement about their medication taking, educating patients about glaucoma and encouraging patients to take their medications were all positively associated with patient adherence to glaucoma medications over an 8-month period.

Funding

This work was supported by the National Eye Institute at the National Institutes of Health (EY018400); and the National Center of Research Resources at the National Institutes of Health (1UL1TR001111). Dr Hartnett was also supported by the National Eye Institute at the National Institutes of Health (EY015130 and EY017011). National Institutes of Health had no role in the design or conduct of this research.

Conflict of interest statement

Drs Sleath, Carpenter, Blalock, Sayner Giangiacomo, and Tudor declare no conflicts of interest. Ms. Slota also declares no conflict of interest. Dr Robin has been a consultant for Biolight, Lupin Pharmaceuticals and Sucampo and he does paid lectures for Merck and Allergan. He has been a consultant and has stock options in Glaukos. Dr Robin is on the board of Aerie Pharmaceuticals. Dr Hartnett is a consultant for Axikin Pharmaceuticals. Dr Muir receives salary support from a VA HSR&D career development award.