Patient barriers and facilitators for making environmental and behavioral modifications for dry eye in the United States

Abstract

Background:

Managing dry eye disease (DED) is expensive. Often, prescribed treatments improve clinical signs but not patient-reported symptoms. In large surveys, clinicians and patients ranked environmental and behavioral modifications among the most important DED-related research priorities. Our purpose was to investigate the barriers and facilitators of use of these modifications by patients with dry eye disease in the U.S. and how their use may be impacted by socioeconomic status (SES).

Methods:

Using Qualtrics, we conducted an anonymous online survey of adults with DED living in the U.S. in August-September 2022. Patients were identified through the Dry Eye Foundation, Sjogren’s Foundation, and a DED clinic in Colorado. We used an established index for classifying respondent SES based on education, household income, and employment. Outcomes included use of environmental and behavioral modifications and barriers and facilitators of their use.

Results:

We included 754 respondents (SES: 382 low, 275 high, 97 unclear). Most were aged 18–49 years (67%), female (68%), White (76%), and reported dealing with DED for ≤5 years (67%). The most frequent modifications were taking breaks to rest eyes (68%), increasing water intake (68%), and using hot/cold compresses (52%). For these three, the biggest facilitators were: belief that the modification works (27%–37%), being recommended it (24%–26%), and ease of use/performance (21%–32%). Across modifications, the biggest barriers were difficulty of use (55%), lack of family/employer/social/community support (33%), and lack of awareness (32%). The data do not suggest discernible patterns of differences in barriers or facilitators by SES.

Conclusions:

Greater emphasis should be placed on explaining to patients how environmental and behavioral modifications might mitigate DED. Employers and members of patients’ support systems should be guided regarding how best to support patients in managing DED symptoms.

Social determinants of health are “the conditions in the environments where people are born, live, work, play, worship, and age that affect a wide range of health, functioning, and quality of life outcomes and risks.”¹ The five key domains of social determinants of health include economic stability, education, healthcare, neighborhood and built environment, and social and community context (e.g., racism).¹ There is also increasing recognition that social determinants of health are highly important to vision health.² For example, national data suggest that less education and lower income are associated with a lower likelihood of having an eye care visit.³ Similarly, less education, trouble finding a doctor, and trouble paying medical bills have all been shown to be independently associated with visual difficulty.⁴

Dry eye disease is prevalent in the U.S. Our systematic review and meta-analysis of recent (2010–2021) studies found the overall prevalence in the U.S. to be 8.1% (95% confidence interval [CI] 4.9%–13.1%).⁵ Dry eye disease-related symptoms, such as dryness and visual difficulty, are among the leading concerns that make patients seek eye care. Managing dry eye disease is expensive, both for patients and society. An analysis published in 2020 found that annual treatment-related costs approximate $1,872 per patient.⁶ When lost wages and other consequences of dry eye disease are also considered, societal costs exceed $50 billion per year.⁷ Moreover, prescription treatments and in-office procedures for dry eye disease are often not covered or are under-reimbursed by insurance, which may present significant barriers to patients receiving them. Economic stability, a key domain of social determinants of health, can therefore have profound impact on patients with dry eye disease.

Despite the money spent on managing dry eye disease, there is a well-documented disconnect between the ocular surface signs of dry eye disease and patient-reported symptoms.^8–10 Although treatments may yield improvements in signs, patients may not experience improvement in symptoms, or vice versa.^8–10 Inadequate symptom relief may explain why patients often need additional interventions, such as over-the-counter artificial tears that are intended to directly lubricate the ocular surface and may be effective in alleviating symptoms of dry eye disease for short periods of time.^11,12 An estimated $2.64 billion was spent on over-the-counter artificial tears in the U.S. in 2019.¹³

In addition to eye treatments, there are modifications that patients can make to their indoor environment because various aspects, such as humidity, airflow, particulate matter, and bioaerosols, may also be related to dry eye disease.^14–16 Examples of environmental and behavioral modifications include using indoor air humidifiers, adjusting vents near workspaces, impeding upward airflow when wearing masks (especially relevant for the COVID-19 pandemic), using warm or cold compresses, and using moisture chamber spectacles. Although environmental and behavioral modifications may plausibly be useful, their effectiveness has not been widely studied for reducing dry eye disease-related symptoms. Indeed, in our previous large surveys, clinicians and patients ranked the effectiveness of environmental and behavioral modifications among the most important dry eye disease-related research priorities.^17,18

In addition to questions about their effectiveness, the feasibility and accessibility of environmental and behavioral modifications for managing dry eye disease-related symptoms are not established. Some modifications (e.g., humidifiers, taking breaks) can be expensive or impractical for certain populations, possibly presenting greater challenges. Hence, to supplement medical treatments for dry eye disease, the key to developing personalized strategies for dry eye disease symptom relief is understanding the barriers and facilitators of the use of these modifications by patients. This is critical information for the planning and implementation of equitable dry eye disease-related healthcare.

Objectives

We conducted an anonymous, online survey of patients with dry eye disease living in the U.S. to investigate (1) the barriers and facilitators of the use of environmental and behavioral modifications for dry eye disease-related symptom relief and (2) how these barriers and facilitators may be impacted by socioeconomic status.

METHODS

The Colorado Multiple Institutional Review Board (COMIRB #21–2541) approved this cross-sectional survey. We offered each eligible respondent who completed the survey a $10 online gift card as compensation for participation.

Development of Survey

We developed questions based on our previous surveys among patients with dry eye disease.^18–21 To promote comprehension of the survey in lay language, a longtime patient advocate with dry eye disease (R.P.) and an expert in survey methodology revised the questions (Appendix 1, available at [LWW insert link]). All investigators pilot-tested and revised the survey (two rounds) to ensure comprehensiveness and clarity of all items in the survey.

The final instrument comprised 22 questions organized into five sections: (1) assessment of respondent eligibility, (2) duration of dry eye disease, (3) barriers and facilitators of the use of environmental and lifestyle modifications for dry eye disease-related symptom relief, (4) impact of dry eye disease, and (5) demographic information. To identify modifications for incorporation into the survey instrument, we used our collective dry eye-related research expertise and familiarity with the recent Tear Film and Ocular Surface Society (TFOS) report on the impact of environmental conditions on the ocular surface (all authors),²² familiarity with existing Cochrane systematic reviews (all authors), lived experience (RP and TL), and clinical expertise (CI, SGH, and DGG). Of note, we did not identify modifications based on their effectiveness but on what we perceived to be ones that are at least somewhat frequently used.

We estimated that survey completion would take no more than 15 minutes. Respondents had the option to save their responses for completion later.

Target Population

We targeted adults aged ≥18 years living in the U.S. with at least one of the following three self-reported characteristics:

1. Any dry eye disease-related symptom, such as dryness, irritation, burning, stinging, grittiness, and foreign body sensations, in the past 4 weeks,

2. Any current or prior diagnosis of dry eye disease by a healthcare professional, and

3. Current use of any treatment or environmental/behavioral modification for dry eye disease.

Source Population

We identified participants through two main sources:

1. To reach under-served populations, we contacted, via telephone, patients seen at a dry eye clinic at Denver Health and Hospital Authority. This is a level-one trauma and safety-net hospital that serves refugee, indigent, and homeless populations, among others. Approximately 80% of patients at Denver Health and Hospital Authority are covered by Medicaid, and another 10% are uninsured.

2. We contacted patients through the Dry Eye Foundation (Poulsbo, Washington) and the Sjögren’s Foundation (Reston, Virginia) online by (a) an invitation email to subscribers of the Dry Eye Foundation mailing list, and (b) invitations through social media platforms (Facebook, Instagram, and Twitter) of both Foundations as well as the authors’ personal social media platforms.

Survey Dissemination

We disseminated the survey through Qualtrics (Provo, Utah). To track the number of responses obtained from each of the two sources, we sent each source separate links to identical copies of the survey. All invitations were accompanied by a brief description and a link to the online survey, which began on August 25, 2022, and ended on November 1, 2022. We used the Fraud Detection functionality in Qualtrics to filter out responses that may have been completed by computer programs or “bots.”

Socioeconomic Status Classification

To classify respondent socioeconomic status, we used an established index used by the National Crime Victimization Survey (Appendix 2, available at [LWW insert link]).²³ This socioeconomic status index is calculated as the sum of three component scores: education (range 0–3), household income (0–3), and employment (0–1), with a total score ranging from 0 to 7. The socioeconomic status definition used in the current paper thus addresses two of the five broad domains of social determinants of health: education and economic stability.

We classified respondents with an socioeconomic status index total score of ≤4 as low socioeconomic status and ≥5 as high socioeconomic status. When respondents preferred to not provide information regarding one or more of the three components, we classified their socioeconomic status as unclear.

Statistical Analysis

We primarily used descriptive statistics to summarize group characteristics (e.g., demographics, dry eye disease characteristics) and to compare outcomes (e.g., use of modifications, barriers, facilitators) among socioeconomic status groups. We used STATA Version 17 for all analyses.

RESULTS

Respondents

In total, 976 individuals began the survey and 825 completed it (completion rate=85%). We excluded 71 of 825 respondents (8.6%) who did not meet eligibility criteria: 43 reported not living in the U.S.; 24 reported being younger than 18 years old; and four did not fulfill any of our inclusion criteria related to dry eye disease-related symptoms, diagnosis, or use of treatments or modifications. This study includes 754 adult respondents: 56 from the University of Colorado (source 1) and 698 from online (source 2).

Demographic Characteristics

We classified 382 respondents (51%) as low socioeconomic status, 275 (37%) as high socioeconomic status, and 97 (13%) as unclear socioeconomic status (Figure 1). Non-reported income data was the most frequent reason for classifying respondents as unclear socioeconomic status: 92/97 respondents (95%). There was a higher percentage of respondents classified as low socioeconomic status among respondents identified from the University of Colorado (source 1: 38/56; 68%) than from online (source 2: 344/698; 49%). Across all 754 included respondents, 65% had completed college or a graduate degree, 43% had total household income ≥$100,000, 68% were employed, and 17% were retired (Appendix 3, available at [LWW insert link]).

Figure 1.

Respondent demographic characteristics.

Approximately a third of all 754 respondents (33%) were ≥50 years old, and approximately two-thirds were female (68%). The low and high socioeconomic status groups had similar age and sex distributions, but the unclear socioeconomic status group had a higher percentage of individuals ≥50 years old (77%) and females (86%). Most respondents in the entire group were White (76%). The percentage of Hispanic or Latino respondents in the high socioeconomic status group was about half that in the low socioeconomic status or unclear socioeconomic status groups (5% vs. 9% or 10%, respectively).

Dry Eye Disease Severity, Duration, and Impact

The dry eye disease severity, duration, and impact were generally comparable across socioeconomic status groups (Appendix 4, available at [LWW insert link]). Most respondents reported that their eyes felt uncomfortable in the past 4 weeks: either some of the time (44%) or most of the time (41%). Almost all respondents had been diagnosed with dry eye disease by a healthcare professional (94%) and were currently using treatments or modifications for dry eye disease-related symptom relief (93%). More than half the respondents (55%) had been dealing with dry eye disease for ≥3 years. A considerably higher percentage of respondents in the unclear socioeconomic status group than in the low socioeconomic status or high socioeconomic status groups had been dealing with dry eye disease for ≥10 years (44% vs. 18% or 17%, respectively). This may be related to the higher percentage of older respondents in the unclear socioeconomic status group.

On a scale of 0–9, with higher scores indicating greater impact, the three groups reported comparable impact of dry eye disease-related symptoms on productivity during close-vision activities (overall median 5, interquartile range [IQR] 3–6) as well as productivity during other activities for which they use their eyes the most (overall median 5, IQR 3–6).

Patterns of Use of Modifications

Across modifications, sizable percentages of respondents were currently using them, ranging from 26% to 68% (Figure 2). The most frequently used modifications were taking breaks to rest eyes (68%), increasing water intake (68%), and using hot/cold compresses (52%).

Figure 2.

Patterns of use of environmental and behavioral modifications for dry eye disease.

Willingness to Consider Untried Modifications

Among all respondents, the percentage of respondents who had never tried and would not consider modifications ranged from 3% (increasing water intake) to 21% (avoiding/minimizing mask use when possible). However, among respondents those who had not tried modifications, there was a high degree of willingness to consider the modifications, ranging from 68% (reducing/blocking floor and/or ceiling vents) to 86% of respondents (keeping computer screen/reading device/book below eye level).

Discontinuation of Modifications

Among all respondents, the percentage of respondents who had tried but discontinued modifications ranged from 13% (taking breaks) to 23% (using hot/cold compresses). However, among respondents who had tried modifications, the modifications that were most frequently discontinued were avoiding mask use when possible and reducing/blocking floor and/or ceiling vents (40% each). The modifications with the lowest percentages of discontinuation were taking breaks to rest eyes (16%) and increasing water intake (17%).

There were no discernible patterns by socioeconomic status regarding willingness to consider untried modifications or discontinuation of modifications (Appendix 5, available at [LWW insert link]).

Barriers to Use of Modifications

When asked to name the two biggest barriers that they experienced (across modifications), respondents most frequently noted difficulty/inconvenience of modifications (55%), lack of family/employer/social/community support (33%), and lack of awareness regarding modifications (32%) (Appendix 4, available at [LWW insert link]). The reported biggest barriers were generally similar by socioeconomic status, except that the unknown socioeconomic status group less frequently than the low socioeconomic status or high socioeconomic status groups reported difficulty/inconvenience of modifications (31% vs. 56% or 61%, respectively) or lack of family/employer/social/ community support (8% vs. 32% or 41%, respectively) and more frequently reported no barriers (43% vs. 20% or 15%, respectively).

The specific modifications with the highest percentages of respondents who noted barriers were reducing/blocking floor and/or ceiling vents, avoiding/minimizing mask use when possible, and impeding upward airflow from masks (Appendix 6, available at [LWW insert link]). The most frequently reported barriers for these modifications were lack of awareness (23%–25%), lack of belief in their effectiveness (13%–16%), and difficulty to use/perform (11%–12%). The data do not suggest discernible socioeconomic status-specific differences for individual modifications.

Facilitators of Use of Modifications

For the modifications that were most frequently reported as being currently used (i.e., taking breaks, increasing water intake, and using hot/cold compresses), respondents generally reported believing that these modifications work for them (32%–37%), are easy to use/perform (31%–32%), have been recommended (25–26%), and are affordable (20–24%) (Appendix 6, available at [LWW insert link]). The data do not suggest discernible differences in facilitators by socioeconomic status.

DISCUSSION

In this large online survey of patients of varying socioeconomic status with dry eye disease in the U.S., we found that the environmental and behavioral modifications that patients most frequently use are taking breaks to rest eyes, increasing water intake, and using hot/cold compresses. Respondents reported a high degree of willingness to consider modifications that they had not tried. The most frequently noted biggest barriers pertained to difficulty/inconvenience of modifications and lack of family/employer/social/community support. Facilitators of the most frequently used modifications were belief that they work, ease of use/performance, recommendation by providers, and affordability. Surprisingly, neither the barriers nor the facilitators appeared to differ based on respondent socioeconomic status.

Implications for Clinical Practice

We are not aware of other studies that have evaluated barriers and facilitators of using environmental or behavioral modifications for managing dry eye either in the U.S. or elsewhere. Future research should explore these issues further. The current study finds that low socioeconomic status does not appear to impact patient perceptions of barriers and facilitators for using environmental and behavioral modifications for managing dry eye disease. However, two of the three most frequently used modifications (increasing water intake and using hot/cold compresses) are inexpensive modifications. Furthermore, although one may hypothesize that individuals with financial insecurity or in underpaid jobs may have less flexibility in taking breaks to rest their eyes, we did not observe socioeconomic status-related differences in the frequency with which respondents reported taking breaks to rest eyes or the extent to which they experienced barriers or facilitators of doing so.

It is also notable that sizable percentages of respondents (26% to 68%) were currently using the various modifications. To our knowledge, this is the first study that documents the widespread usage of these modifications. It is also reassuring that we did not observe differences in patterns of use of these modifications by socioeconomic status.

Another important implication of our findings is that, for various specific modifications, sizable percentages of respondents (15% to 39%) reported that they had never tried but would be willing to try them. This may represent opportunities for healthcare providers to educate individual patients regarding environmental and behavioral modifications that may be effective for their specific circumstance. An important caveat is that the evidence base supporting recommendations of these interventions needs to be built and strengthened. For example, we are aware of only two systematic reviews of specific environmental modifications for dry eye disease.^24,25 One review found that indoor air humidification at the workplace may have little to no effect on ocular dryness symptoms.²⁴ The other review found that the use of blue-blocking spectacles did not reduce symptoms of visual fatigue associated with computer vision syndrome.²⁵ Regarding moisture chambers, we are aware of only one small, randomized trial (30 patients), which evaluated the short-term (90-minute) use of moisture chambers.²⁶ Their use was associated with improvements in ocular discomfort and tear film stability.²⁶ However, even with this evidence, there are large gaps regarding the effectiveness of environmental and medical modifications for dry eye. McCann and colleagues recently published an overview of systematic reviews that confirmed the limited evidence.¹⁰

Until more research evidence is available regarding the effectiveness of environmental and behavioral modifications for managing dry eye disease, we suggest that clinicians make patients aware of available options but emphasize the limitations in the evidence underpinning them. It is worth noting that, in our experience, moisture chamber devices are generally used only when the clinician provides the patient with suggestions of specific vendors through which the devices can be obtained and well fit.

Because patients may require help from their employers and support systems (e.g., family members) to implement their chosen modifications, these parties should be guided regarding how best to support patients in managing dry eye disease. For example, where feasible and requested by employees with dry eye disease, employers can support them in various ways, such as reducing or blocking floor and/or ceiling vents and allowing employees to take breaks to rest their eyes. Educating employers is particularly important because, especially in the context of office-based jobs, between a third and a half of a patient’s waking time is spent working indoors. The prevalence of dry eye disease symptoms in offices has been reported to range from 19% to 59% around the world, with regional and employment type-specific variations.²⁷

Implications for Research

Given that existing medical treatments for dry eye disease are not fully effective in improving symptoms, and that patients seek environmental and behavioral modifications to manage symptoms, there is a pressing need for rigorous studies (and subsequent systematic reviews) to identify effective modifications as well as approaches to make patients aware of them. Additionally, the lack of evidence of a difference between socioeconomic status groups in the barriers and facilitators suggests that new studies on effectiveness may be broadly applicable across socioeconomic strata within the U.S.

Limitations

There are certain limitations to the current study. First, we relied on respondent self-report for all information gathered, including key variables to define socioeconomic status and barriers and facilitators for use of the environmental and behavioral modifications. Respondents might have mischaracterized aspects of this information. Second, the socioeconomic status definition used in the current study focuses on two of the five key domains of social determinants of health (economic stability and education). Studies have shown that dry eye disease-related outcomes may be impacted by the three other domains as well: healthcare access and quality (e.g., we did not ask about whether the respondent had regular access to an eye care provider),^28,29 neighborhood and built environment (e.g., we did not ask about the respondent’s occupation, such as whether they worked indoors or outdoors),^{14,15,22,30–34} and social and community context (e.g., we did not ask about the respondent’s residential geographical region and weather-related context in the U.S.).^29,35,36 Third, we did not ask whether the respondent’s dry eye symptoms may have been related to individual factors, such as contact lens use and underlying health conditions. Fourth, the applicability of the study results to patients outside the U.S. is unclear because we restricted the survey to patients living in the U.S. However, likely related to the sources used to identify participants for this survey, the respondent sample was somewhat more diverse in terms of race and socioeconomic status than most past large surveys of patients with dry eye disease.^18–20,37 Fifth, minimizing mask use was most frequently reported by respondents as the modification that they had never tried and would not be willing to try. We included this item in the survey because the survey was disseminated in 2022, when COVID-19 was still a public health emergency in the U.S. This mask-related finding may be less relevant now given that, at the time of writing, fewer individuals in the general population are routinely wearing masks. We recently reported the findings of a study of healthcare workers with prolonged mask use and self-reported symptoms of dry eye disease.³⁸

Socioeconomic status does not appear to be related to barriers and facilitators of the use of environmental and behavioral modifications for managing dry eye disease-related symptoms. Greater emphasis should be placed on making patients aware and explaining to them how these modifications might (or might not) mitigate dry eye disease-related symptoms. Employers and members of patients’ support systems (e.g., family members) should be guided regarding how best to support patients in managing symptoms of dry eye disease.

Supplementary Material

Appendix 1

Appendix 1. Survey Instrument, available at [LWW insert link]. This appendix provides the full survey instrument that was used in the study. The survey was deployed in Qualtrics.

Appendix 2

Appendix 2. Socioeconomic status (SES) index used by the National Crime Victimization Survey (NCVS) – Index 3, available at [LWW insert link]: this appendix provides the index established by the NCVS that we used in the survey for classifying respondent SES.

Appendix 3

Appendix 3. Respondent education, household income, and employment/student status, available at [LWW insert link]. This appendix provides details about education, household income, and employment/student statuses for all respondents as well as separately by unknown socioeconomic status (SES), low SES, and high SES groups.

Appendix 4

Appendix 4. Self-reported dry eye disease severity, duration, and impact, available at [LWW insert link]. This appendix provides details about self-reported dry eye disease severity, duration, and impact for all respondents as well as separately by unknown socioeconomic status (SES), low SES, and high SES groups.

Appendix 5

Appendix 5. Patterns of use of environmental and behavioral modifications by socioeconomic status, available at [LWW insert link]. This appendix provides details about use of specific environmental and behavioral modifications for all respondents as well as separately by unknown socioeconomic status (SES), low SES, and high SES groups. For each modification, information is provided regarding the percentage of participants who have never tried and will not consider, never tried but would consider, tried but discontinued, and currently doing the modification.

Appendix 6

Appendix 6. Percentages of respondents who noted specific barriers and facilitators to use of environmental and behavioral modifications for dry eye disease, available at [LWW insert link]. This appendix provides details about the barriers and facilitators of use of specific environmental and behavioral modifications for all respondents as well as separately by unknown socioeconomic status (SES), low SES, and high SES groups. For each modification, information is provided regarding the percentage of participants who noted specific barriers and facilitators.