Assessment of Low Vision Referrals before and after Establishment of a Low Vision Program at an Academic Medical Center

Abstract

Significance:

We assessed the number of referrals for low vision (LV) services to determine if establishing a LV program at a large academic medical center impacted referral rates. Visual acuity (VA), referral outcome, location and specialty were examined as factors that could impact referrals.

Purpose:

To identify gaps in the referral process to LV services.

Methods:

Electronic medical records of patients were reviewed to ascertain the referral rate among those who qualified for services, both before (2014–2016) and after (2017–2019) the establishment of a LV program. The medical records were further sub-divided into 2 categories based on VA in the better seeing eye: 20/70 to 20/200 and 20/200 to worse vision.

Results:

2014 patient records with VA qualifying for LV services were reviewed. 91.7% of patients had a VA of 20/70 to 20/200 inclusive in their better eye. 89.8% of patients with VA of 20/70–20/200 and 74.4% of patients with VA worse than 20/200 were never referred. Before establishing a LV program, only 2.2% of patients with VA of 20/70 to 20/200 were referred for services on their first visit, which improved to 8% after the program established (OR 3.88, CI 2.37–6.33, P<.001). Also, before the program’s establishment, 12.5% of patients with VA worse than 20/200 were referred on their first visit, which increased to 31.9% after the program’s establishment (OR 3.29, CI 1.50–7.19, P=.002). Patients with VA worse than 20/200 were more likely to be referred (before OR 6.34, CI 3.03–13.28, P<.001; after OR 5.38, CI 3.09–9.37, P<.001). Our data also showed 10.3% of patients in this study declined referral to LV services.

Conclusions:

Referral rates to LV services are low among patients who qualify. The establishment of a LV program at the medical center significantly increased referral rates. However, more improvement is necessary to connect patients to LV services.

Low vision and blindness are considered one of the leading causes of disability for adults in the United States. Cases are predicted to more than double by 2050 highlighting the critical need for low vision exams and rehabilitation services.¹ As of 2012, 4.2 million Americans aged 40 years and older suffered from uncorrectable vision impairment. This number is predicted to reach 8.96 million by 2050 due to the increasing epidemics in diabetes and other chronic diseases as well as our rapidly aging population.² With an extensive number of stakeholders involved, including but not limited to medical providers, low vision specialists, patients, behavioral health, and social work services, successful low vision rehabilitation is complex. To ensure an individual has successfully taken advantage of low vision resources, it is important to 1) offer timely referrals by eye care providers, 2) administer prompt care by low vision specialists, 3) provide follow up with social work services and 4) carry out appropriate referrals to behavioral health services. However, in our community, little programmatic flow among these entities has created critical gaps in the rehabilitation process.³

The lack of a definitive referral process and perception by ophthalmologists towards low vision services ^{4, 5} in our academic medical center, may have implications for a low vision patient’s independence, employment, health, social status as well as mental health. In severe cases, this may lead to isolation, substance abuse and increase in co-morbidities - thereby negatively impacting the patient’s quality of life.^6–8 Vision impairment has been linked to causes of depression and burden among caregivers especially through emotional contagion and disconnection of non-verbal communication.⁹ This linkage indicates that efficient low vision rehabilitation program not only positively influences independence in the visually impaired person, but it also helps minimize caregiver burden and depression.

The total disability adjusted life-year gross domestic product purchasing power parity per capital loss due to moderate to severe blindness in the United States reached approximately $51.8 billion in 2011.¹⁰ Developing countries have the most people with moderate-to-severe blindness, however, the highest productivity loss was noted in the Unites States among high income countries. This impacts the global economy as a whole.¹⁰ Therefore, early low vision rehabilitation intervention has the potential to decrease the financial burden of loss of a functional member of society and productivity of its caregivers. Previous work in other countries have identified a need to implement programmatic change in the field of low vision to more efficiently meet the demands of an aging population.⁷ Likewise, recent work has shown earlier referral to low vision services help patients better cope with vision loss over time.^11–13

We aim to lay the groundwork for implementation of a successful and sustainable low vision rehabilitation program in the studied academic medical center by identifying gaps in the referral process. Concurrent with the work of Kumar et. al., 2016,¹⁴ we believe patients with “better” acuity, i.e. 20/70–20/200 in the better-seeing eye, are referred at a lower rate than those with “worse” visual acuity, i.e. worse than 20/200 in the better-seeing eye.¹⁵ We also believe most referrals are made by retina specialist vs. other ophthalmology specialties in our department.^{14, 16} We are interested to know if referrals were more frequent at the location where the low vision clinic is situated compared to other eye clinic locations where there is no physical presence of services. This study explores whether the creation of a low vision rehabilitation program at our academic medical center at the end of 2016 had a significant impact on closing gaps in the referral process.

METHODS

A retrospective electronic medical record review was performed to identify patients with visual acuity of 20/70 or worse in the better seeing eye seen at the University of Wisconsin –Madison, Department of Ophthalmology and Vision Sciences, Eye Clinics between January 2014 and December 2019. This period includes when a Low Vision Clinic was established at the end of 2016 that was significantly advertised as a resource through various means of communication – announcements to eye care providers in our department, large print business cards and brochures to patients as well as promotions in community in the form of patient education and provider continuing education events. Nothing was indicated in the promotional materials or educational content to prompt providers to refer based on any visual acuity or visual field criteria. The study was approved by the institutional review board at the University of Wisconsin-Madison, adhered to the tenets of the Declaration of Helsinki and complied with the Health Insurance Portability and Accountability Act. Study data were collected and managed using Research Electronic Data Capture (REDCap) tools hosted at the University of Wisconsin-Madison, School of Medicine and Public Health.¹⁷ REDCap is a secure, web-based application designed to support data capture for research studies, providing: 1) an intuitive interface for validated data entry; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for importing data from external sources.

A RedCap search was performed on all encounters for adult patients seen at any of the seven ophthalmology clinics during the years 2014 through 2019 which had a visual acuity of 20/70 or worse in the better-seeing eye. Encounters were searched for common ophthalmological diagnoses which could lead to poor vision such as “macular degeneration”, “glaucoma” or “diabetic retinopathy” among others (Appendix Table A1). Encounters were also searched by visit types (Appendix Table A2) and for terms like “low vision” and the name of our institution’s low vision specialist (Appendix Table A3). Data were collected on date of encounter, location of encounter, specialty/sub-specialty of the encounter provider, visual acuity in each eye, whether they were referred for low vision services and outcome of the referral if made. Referral outcomes were grouped into four categories: (1) seen – patients pursued low vision services after a referral was placed, (2) declined – patients were offered services but they declined a referral, (3) lost – patients who were lost to follow up in eye care and we could not track referral outcome, and (4) unknown – patients were referred for services but the referral outcome is unknown or never discussed in future visits with eye care provider. Data were also collected on how many visits it took till a referral was made to low vision services. Patients with reversible diagnoses whose vision improved to better than 20/70 in the better-seeing eye during the period of the study were excluded. Statistical analysis was performed via logistical regression models with R statistical software version 4.0.5 (2021-03-31).¹⁸

RESULTS

A total of 2014 patient encounters met the inclusion criteria. 54.4% of encounters took place in 2014 through 2016, before the creation of our institution’s low vision rehabilitation program. 91.7% of patients had a visual acuity of 20/70–20/200 in their better-seeing eye at their first encounter during this period. Only 2.2% of patients with a visual acuity of 20/70–20/200 and only 12.5% of patients with a visual acuity worse than 20/200 were referred for low vision services on the first visit (Figure 1). After the program was established, the referral rates increased to 8% and 31.9% respectively on the first visit.

Figure 1.

Comparison of percentage of patients referred for each visual acuity category in the better seeing eye (20/70–20/200 vs. 20/200 and worse) before and after the establishment of a low vision rehabilitation program separated into three groups – first visit, after first visit and never referred.

Some qualifying patients were indeed eventually referred for low vision services, however not on their initial visit (Table 1). For patients with visual acuity 20/70–20/200 before the low vision program establishment, on average 3 visits were required to receive the referral, whereas after the program’s establishment 1.3 visits were necessary. For patients with visual acuity worse than 20/200, 2 visits on average were needed before the program’s establishment versus 0.12 visits after the program’s establishment, meaning patients with visual acuity worse than 20/200 were nearly always referred on their initial visit after program establishment.

Table 1.

Mean number of visits required before a referral was made.

Referred ever	Vision category	Mean before	Mean after	Mean total
No	20/70 – 200	1.16 (2.063)	1.18 (1.88)	2.34 (2.971)
No	< 20/200	1.45 (2.441)	0.9 (1.703)	2.34 (3.173)
Yes	20/70 – 200	3.19 (3.851)	1.29 (2.246)	4.48 (5.183)
Yes	< 20/200	2 (1.195)	0.12 (0.354)	2.12 (1.356)

Of patients who were referred for low vision services, 14% progressed from visual acuity 20/70–20/200 to worse than 20/200. 89.6% patients with visual acuity 20/70 to 20/200 were never referred for low vision services prior to the establishment of a low vision clinic at our institute (Figure 1). This number did not change after the establishment of the low vision clinic and remained at 90%. In the 20/200 or worse visual acuity category, 79.2% of patients were never referred to low vision services which decreased to 68.1% after the establishment of the low vision clinic.

Patients with visual acuity 20/70–20/200 had four times higher odds to be referred for low vision services after the establishment of a low vision program at our institution (Odds Ratio (OR) = 3.88, Confidence Interval (CI) = 2.37–6.33, P<.05). Likewise, patients with visual acuity worse than 20/200 had 3 times higher odds to be referred after program establishment (OR 3.29, CI 1.50–7.19, P<.05). Patients with visual acuity worse than 20/200 were more likely to be referred for low vision services both before and after program establishment (OR 6.34, CI 3.03–13.28, P<.001 and OR 5.38, CI 3.09–9.37, P<.001 respectively).

After referral to low vision services, patients either pursued low vision services (Seen), declined low vision services (Decline), or were lost to follow-up (Lost) (Figure 2). There was also a proportion of patients who were referred to low vision services, but it was unknown what their outcome was (Unknown) (Figure 2). About half of patients with visual acuity 20/70 to 20/200 did follow through and receive low vision services, both before and after the establishment of our institution’s program (51.9% and 48.2% respectively). The proportion of patients with visual acuity worse than 20/200 who followed through with low vision services increased from 40% to about 61% after program establishment. The proportion of patients lost to follow-up was similar within visual acuity categories across both time points. In all the vision categories examined, 10.3% of patients declined referral to low vision services at any point during this study. A declined referral was noted for patients whose charts had provider notes specifying a decline by the patient. After the establishment of the low vision service, fewer patients in the worse visual acuity category (4.3%) declined referral than those with visual acuities 20/70–20/200 (15.3%).

Figure 2.

Percentage of subjects that were referred at any visit and that declined services (Decline), lost to follow up (Lost), seen (Seen) and unknown referral status (Unknown).

Data was also analyzed to examine if there were any significant differences between referral rates by sub-specialty or clinic location. However, these variables did not show statistical significance in a multiple regression model most likely due to a small referral sample size (Figure 1) divided between 6 specialties (Comprehensive, Glaucoma, Neuro-ophthalmology, Optometry, Pediatric and Retina).

DISCUSSION

In this study, referral rates for low vision services were low overall. Our study revealed that 89.8% of patients with visual acuity 20/70 to 20/200 and 74.4% of patients with visual acuity of 20/200 or worse were never referred for low vision services. Perception towards low vision rehabilitative services may have played a key role to referral outcomes and providers play an important role in how this perception is driven. Despite referral rates being low, establishing a low vision rehabilitation service did improve referral rates for all vision categories at our institute. This is similar to findings by a recent study in India. This study looked at barriers and enablers to low vision services in a tertiary eye care hospital and concluded that creating awareness in the community and among eye care providers improved acceptance of low vision services and increased referral rates.¹⁹ Our referral rate remains lower than the final referral rate cited in this study. This is most likely because our only enabler was establishing a low vision rehabilitation service. An additional finding in our data confirmed our hypothesis that more patients reaching the level of legal blindness (20/200 or worse vision in the better seeing eye) are referred to low vision services than those who have low vision (20/70 or worse in the better seeing eye).²⁰

Another study examined the utilization of low vision services and found that patients were more likely to follow through with low vision services if they were seen at the same location where the low vision services are offered.²¹ Low vision services are currently offered only at our main eye clinic. However, we have seven satellite clinics, and we plan to focus on adding low vision services in the future to increase referral rates. Sharika et al. examined the perceived barriers to patients successfully undergoing low vision rehabilitation.¹⁹ The most prominent barriers among eye care providers were lack of understanding of appropriate referral criteria and lack of availability of low vision services. The most prominent barriers among patients were lack of understanding about the need for rehabilitation services. Conversely, positive factors influencing referral to low vision services were development of clear referral criteria, having a referral pathway to low vision services, creating awareness among providers and patients, stratifying different levels of low vision services, and implementing a low vision counseling service. The study group found that after implementing these positive factors, the referral rate improved from 25.6% to 51.2% and utilization of low vision services improved from 67.9% to 81.7%. These findings provide proof of concept that strategies to implement programmatic change can improve outcomes.

It is possible that our low utilization rate may be due to reasons of patients’ lack of knowledge about low vision services and refusal of low vision assistive devices: social stigma, fear of loss of employment, low perceived necessity, and denial of the magnitude of the illness.²² On top of this, visual impairment is an invisible disability. Patients feel society may not recognize their disability as they appear fine both physically and socially.¹⁵ Due to being an invisible disability, employers, caregivers, family and friends may not appreciate the extensiveness of the vision loss and why it affects the patient’s quality of life so profoundly. This has a significant effect on the patient’s behavioral health and their own perceptions of vision loss.⁴ Dependence on family and caregivers for transportation and activities of daily living especially in rural areas is a burden. Public transportation and paratransit access is readily available in urban areas. Caregiver burden is therefore magnified in a rural setting as they must take time off from employment and their daily lives to transport the patient to medical appointments. This leads to an increase in financial hardship and income stresses on the household.⁹

The far-reaching systemic effects of vision loss are becoming ever more apparent. Increasing numbers of studies are being published on the detrimental effects of unmitigated vision loss. Low vision’s relationship to dementia has been established by Paik et al. 2020, which found a positive relationship between low vision, dementia, Alzheimer’s disease, and vascular dementia.²³ Additionally, Sabel et al. reported that chronic stress can not only result from vision loss but can also contribute to worsening of vision loss through psychosomatic processes.²⁴ Eye care providers can play an important role in promoting positivity and optimism in their patients going through vision loss.

Much more work remains to be done to meet the low vision needs of the community. Further directions include de-stigmatizing and increasing awareness among both providers and patients, stratifying levels of low vision rehabilitation services to simplify access, and creating an interdisciplinary low vision counseling service to ensure success with rehabilitation. As referrals to low vision clinics increase, we should also address the current paucity of vision rehabilitation providers, to prevent patient frustration due to increased times for an appointment as more active referrals are made.

The World Health Organization (WHO) International Classification of Diseases (ICD) categorizes mild vision loss or near-normal vision as between 20/30 to 20/60. For diagnosis purposes, the ICD-10 code for category 1 low vision starts at 20/70.²⁵ We chose this acuity as our lowest criteria for visual acuity. Best corrected visual acuity of 20/200 or worse is categorized by WHO ICD-10 as severe visual impairment and in the United States, is recognized as the minimal criteria for legal blindness according to the Social Security Administration.²⁶ In 2017, the American Academy of Ophthalmology came out with a Preferred Practice Pattern for Vision Rehabilitation. It recommends referral of patients with any kind of functional vision problems to vision rehabilitation services as a continuum of eye care regardless of the best corrected visual acuity status.³ Providers who were familiar with and followed the Preferred Practice Pattern guidelines published by the American Academy of Ophthalmology were 2.5 times more likely to refer patients for low vision services than those not following the guidelines.²⁷ However, a referral to low vision services is not recommended by most eye care providers due to lack of communication by the patient in expressing difficulty with daily living activities, the specialists’ lack of time in addressing functional vision loss during clinical visits and the specialists’ perception regarding patient’s resources to undergo low vision rehabilitation.²⁸ Both the American Optometric Association and the American Academy of Ophthalmology practice guidelines for low vision rehabilitation indicate to refer for low vision services when a patient has a loss of functional vision regardless of visual acuity.²⁹ Referrals for patients with vision better than 20/70 were not included in the present study, but should be considered for patients with mild vision loss who have functional complaints due to vision.

A limitation of our study is that our patient cohort likely under-represents the population of patients with visual field loss who qualify for low vision services. We based our search criteria on central visual acuity, which may be preserved even in advanced glaucoma patients or patients with hemianopia who would qualify for low vision rehabilitation services based on visual field criteria. Our work was a review of electronic medical records and eye exam notes which do not normally capture the extent of visual field loss. It would be interesting in future studies to manually incorporate individual visual field measures into the data to quantify the degree that visual field loss contributes to referral trends. A confounding variable was that our retina specialists, who are responsible for most referrals to low vision services, are based predominately in the same building as the low vision rehabilitation services, making it difficult to parse the data based on referral source versus referral location. Our satellite clinics have fewer sub-specialty services adding to the bias. We have noted an uptick of referrals to low vision services at our academic medical center from community optometrists across the state since 2020 onwards due to increased awareness through continuing education events sponsored by our department. We did not include this revelation in our study as we did not formally track the referrals, and the timeline is outside of the limit of our study. Our findings do concur with what is reported by Malkin et al. on using continuing education events to address provider perception barriers on access to care.²⁹

FUTURE DIRECTION

We believe developing a multi-disciplinary healthcare team while simultaneously educating eye care providers and communities about low vision rehabilitation services will provide the most benefit to low vision patients. At other institutions,³⁰ patient peer advocates have been called upon to educate and counsel members of the community with low vision. We wish to apply a similar concept where patients receiving a diagnosis of irreversible vision loss are connected to a patient care advocate on site at their initial appointment. We hope to add the expertise of social workers, psychologists, geriatricians, and patient peer advocates to the expertise of our eye care providers to create a holistic approach to low vision care providing stronger and more sustainable care.

Our next aim is to more fully understand perspectives of the key groups involved in low vision care. We plan to survey patients, eye care providers, and low vision stakeholders to elucidate each group’s perception of the gaps and barriers to effective low vision care. Further work needs to be done in understanding the disposition of providers, patients, and low vision specialists regarding timely referrals, variation in referrals for services, and awareness of the role of each low vision specialty. We hope to break down patients’ misconceptions of low vision services and deconstruct providers’ assumption that low vision services are a last-resort treatment for those with poor vision. Ultimately, we want to break societal stigmas and promote a positive perception of low vision services in the community to increase access and timely referrals.

Supplementary Material

Dalzotto et al appendix tables A1-A3 final

Appendix 1: International Statistical Classification of Disease and Related Health Problems, Tenth Revision, Clinical Modification (ICD-10-CM) code ranges used during electronic medical record search for common diagnoses of eye diseases that can affect functional vision.

Appendix 2: Current Procedural Terminology (CPT) codes used to filter visit types during electronic medical records review.

Appendix 3: List of key phrases used to filter physician notes within the electronic medical records that qualify for referral to low vision services (based on visual acuity criteria 20/70 or worse) in order to determine if a referral was discussed or made and the subsequent outcome of the referral.