Outcomes of recommendation for cataract surgery in control and intervention homes by the SEEING study.
Friedman DS, Muñoz B, Roche KB, Massof R, Broman A, West SK. Poor Uptake of Cataract Surgery in Nursing Home ResidentsThe Salisbury Eye Evaluation in Nursing Home Groups Study. Arch Ophthalmol. 2005;123(11):1581-1587. doi:10.1001/archopht.123.11.1581
To compare the uptake of cataract surgery in nursing homes in which assistance was provided in obtaining services with that in control homes.
The Salisbury Eye Evaluation in Nursing Home Groups (SEEING) project is a randomized clinical trial studying the effect of a comprehensive vision restoration-rehabilitation program, including the provision of cataract surgery services when needed. Twenty-eight nursing homes in the Eastern Shore area of Maryland and Delaware were matched in pairs by size and payment type. Nursing homes within each pair were randomized to usual care or targeted intervention. Persons with cataract causing visual acuity in the better eye to be worse than 20/40 were informed of the possible benefit of cataract surgery. For those in intervention homes, additional support was provided in obtaining cataract surgery.
Of residents with vision-impairing cataract in intervention homes, 31% underwent cataract surgery vs 2% in usual-care facilities. Residents with cataract compared with all residents without visual impairment, regardless of nursing home assignment, tended to be older (mean age, 86.7 vs 82.1 years; P<.001), were more likely to be black (age-adjusted P<.001), had lower Mini-Mental State Examination scores (mean, 11.7 vs 16.2; age-adjusted P<.001), and longer length of stay (mean, 42.3 vs 24.4 months; P<.001). Furthermore, cataract was associated with significant visual impairment and with functional limitations, with 20% of those recommended for surgery having a visual acuity of 20/100 or worse in the better-seeing eye.
Screening for and identifying cataracts as a cause of vision loss rarely results in uptake of cataract surgery services in nursing homes. The addition of a support system to facilitate the process of scheduling surgery and getting to and from the hospital dramatically increases uptake rates.
Cataract extraction is the most commonly performed operation in the Medicare population; more than 1.5 million procedures were performed in 1998. Surgery is highly successful in restoring good vision, and more than 95% of patients achieve 20/40 or better vision.1 More important, cataract extraction has been demonstrated to improve visual function in both first-eye and second-eye surgery.2- 7 In addition, cataract surgery can improve overall function as measured by several validated outcome measures.8- 10
Research in nursing home residents has documented high rates of vision-impairing cataract.11 In an ancillary study to the Baltimore Eye Survey, researchers found a 13-fold higher rate of blindness among black participants and a 16-fold higher rate of blindness among white participants compared with race-specific, noninstitutionalized populations. Overall, 37% of the nursing home residents had visual acuity worse than 20/40,with 27% of bilateral blindness attributable to cataract. Another survey, from Australia, reported that bilateral cataracts were responsible for bilateral blindness in 7 of 10 nursing home residents.12
The Salisbury Eye Evaluation in Nursing Home Groups (SEEING) project is a nursing home–based clinical trial to determine whether increased access to vision care services can benefit nursing home residents. Although numerous investigators have called for measures to improve sight in nursing home residents, to our knowledge none have attempted a study of the effect of a vision restoration-rehabilitation program on function and quality of life in this population. This is an article about the characteristics of nursing home residents with vision-impairing cataract and the factors associated with the uptake of cataract surgery services in these populations.
The SEEING study is a randomized clinical trial studying the effect of a comprehensive vision restoration-rehabilitation program, including the provision of cataract surgery services when needed. Nursing homes were randomly assigned to intensive visual rehabilitation (eyeglasses or cataract surgery), low-vision services (intervention), or usual care (control). The study was conducted in 28 nursing homes on the Eastern Shore area of Maryland and Delaware. Its design and protocols have been detailed elsewhere.13 In brief, the primary study aim was to determine whether individuals receiving assistance in scheduling cataract evaluation and attending cataract surgery were more likely to undergo cataract surgery than those receiving routine care. A secondary goal of this research was to assess the factors associated with untreated cataract and the effect of untreated cataract on function among nursing home residents.
Twenty-eight nursing homes on the Eastern Shore area of Maryland and Delaware, representing all but 2 of the nursing homes on the Eastern Shore within a 2-hour drive of Salisbury, Md, at the time of study design, participated in the study. Nursing homes were matched in pairs by size (number of beds) and proportion of patients who self-pay. Stratification categories for matching on size were less than 80 beds, 81 to 110 beds, 111 to 140 beds, and more than 140 beds. Stratification categories for matching on proportion of persons who self-pay were less than 15%, 15% to 29%, 30% to 44%, and more than 44%. One member of each pair of matched homes was randomized to 1 of 2 groups. In the usual-care group, eligible residents underwent visual acuity screening, and those with visual impairment underwent a full ophthalmologic examination. The ophthalmologist reported the findings to the nursing home physician, the patient, and the patient’s guardian. Further provision of eye care services was at the discretion of the family. In the intervention group, eligible residents underwent visual acuity screening, and those with visual impairment underwent a full ophthalmologic examination, identical in all testing and examination procedures as in the usual-care group. However, if new eyeglasses, cataract surgery, or low-vision care was indicated, the project staff facilitated access to these services within the nursing home (see the “Examinations” subsection of the “Methods” section).
Thus the unit of randomization for this clinical trial was the nursing home, not the individual. This decision was based on the need for a standard approach to care for all residents of a single nursing home after the ophthalmologic examination. Moreover, some of the intervention consisted of staff training on low-vision aids, which was conducted at the facility level. The main outcome of this article is the uptake of cataract surgery services in intervention and control homes.
Subjects were eligible for the SEEING study if they were 65 years or older, were not in hospice care or at immediate risk for death, and were not short-stay residents. Those determined by nursing home staff and the SEEING study team members to be completely unresponsive to all stimuli, to be too ill to be tested, or unable to give informed consent and for whom no guardian was assigned were ineligible. The Johns Hopkins Institutional Review Board approved the study protocol. All studied residents or their legal guardians provided informed consent.
All visual acuity measurements were made both before and after refraction. Refraction was performed only in patients with visual acuity worse than 20/40 in the better-seeing eye. All individuals were tested first using Early Treatment Diabetic Retinopathy Study charts or Lea symbols to determine whether refraction was necessary.14,15 Grating acuity was also tested using Teller cards, as described previously.16 For this study, visual impairment was defined as visual acuity worse than 20/40 in the better-seeing eye.
Residents were given the Mini-Mental State Examination (MMSE) to determine the level of cognitive impairment.17 If a resident could not do parts of the MMSE because of physical reasons, the questions were excluded from the calculation of the score, which was calculated on the basis of the remaining questions and rescaled to the reference total of 30.
Age, gender, race, educational status, length of stay (LOS) in the nursing home, number and kinds of medications received, and chronic conditions, hospitalizations, and falls were ascertained from the medical records.
Data on physical function were collected from nursing staff caring for the resident. We used the minimum data set section G series of questions on level of physical function and need for assistance. Inter-rater reliability for this section has been reported at weighted κ = 0.69 to 0.92.18 The questions relate to basic activities of daily living (ADL) and mobility, and for each question a determination is made of the level of difficulty in performing the task and the level of supervision or assistance required by staff.
We used a validated scale of observed behaviors from a previous study of nursing home residents.19,20 Adaptation is measured on a scale from 0 to 22 and includes items on sociable behaviors directed toward the staff and other patients. The nursing staff provides data about the participant. The scale has been shown to be internally consistent with reasonable face validity.21
We used the minimum data set questions on activities, supplemented with other activities pertinent to goals for patients with low vision. In this section we also asked questions on interaction with others, such as family, friends, and staff.
Residents with visual impairment in all nursing homes underwent an ophthalmologic examination by the study ophthalmologist and were assigned to 1 of 3 groups. Residents with uncorrected refractive error less than 20/40 in the better eye in whom correction improved vision by at least 2 lines on the Early Treatment Diabetic Retinopathy Study logarithm of the minimum angle of resolution illumination chart so that at least 1 eye had visual acuity better than 20/40 were assigned to the uncorrected refractive error group.
In residents with visual acuity worse than 20/40 owing to cataract or posterior capsular opacification who also had visual symptoms, either reported by them or their family member or guardian, cataract surgery was recommended. The guidelines for cataract surgery specify that some visual disability should be part of the indication for cataract surgery.22 The ophthalmologist did not refer for cataract surgery residents with lower levels of vision loss who did not report any visual symptoms or, if cognitively impaired, for whom staff, family members, and guardians did not find evidence of visual impairment in the patient’s behaviors. Such patients were referred for low-vision rehabilitation. In intervention homes, the study provided transportation to and from the surgical facility for a preoperative visit, the surgery, and 1 postoperative visit. The project director worked with the nursing home staff, family, and patients to schedule visits and surgery. The costs of the surgery and the surgeon were paid by Medicare.
Those eligible for low-vision rehabilitation had visual acuity worse than 20/40 not correctable with proper refraction or surgery. If a participant was eligible for more than 1 intervention, such as proper refraction and low-vision aid training, both were provided.
Statistical testing of differences in baseline characteristics between participants referred for cataract surgery and participants without visual impairment were made using age, MMSE score, and LOS in the nursing home as continuous variables. Logistic regression models were used to compare the 2 groups, adjusting for age, cognitive status, and LOS when appropriate. To account for the correlation among residents in nursing homes, standard errors were corrected with the generalized estimated equation approach.
Residents in control nursing homes were similar to those in the intervention homes in comparisons of age, race, and gender, but the rate of visual impairment and low MMSE score was slightly higher in the control home population (Table 1). When comparing those with cataract surgery recommended with all residents without visual impairment, they tended to be older (mean age, 86.7 vs 82.1; P<.001) more often were black women (age adjusted P<.001), and were more likely to be women although this was not statistically significant (age adjusted P = .30;Table 2). Furthermore, they were resident in the nursing home for a longer period (mean LOS, 42.3 vs 24.4 months; age- and MMSE-adjusted P<.001).In addition, comparing individuals with visually significant cataracts with all residents without visual impairment, those with vision-impairing cataract had lower MMSE scores (mean, 11.7 vs 16.2; age-adjusted, P<.001). Using logistic regression, age, MMSE, and LOS were significant predictors of vision-impairing cataract (Table 3).
Cataracts reduced vision significantly, with 75% of those recommended to undergo cataract surgery having visual acuity worse than or equal to 20/60 in the better-seeing eye. Nineteen percent of those referred for cataract surgery had visual acuity worse than or equal to 20/100 in the better-seeing eye. Comparing those with cataract with age- and gender-matched control residents without visual impairment from the same nursing homes, those with cataract were more likely to require help with ADL (Table 4). Fifteen percent of those in whom cataract surgery was recommended were able to perform 4 to 5 ADL independently, compared with 28% of residents without visual impairment (P<.05). Mobility scores were also lower for those with visual impairment from cataract, with 45% of these residents having no independent mobility vs 32% in the visually unimpaired control residents. However, this finding was not statistically significant in the multivariate analysis.
A total of 1305 residents had testable vision using either grating cards or recognition testing (Early Treatment Diabetic Retinopathy Study or Lea symbols). Of the 496 individuals with visual impairment, 177 (44%) of the 410 in whom an ocular cause was found had cataract as the primary reason for decreased vision. Ninety-nine residents were identified in the control homes, and 78 residents were found in the intervention homes (1 of whom was reassigned to the refraction group [Figure]). Of the remaining 77 individuals, 3 had either died or were too ill to be followed up when we returned to schedule surgery. Of the 74 fully eligible individuals with bilateral visual impairment attributed to cataract for whom services and assistance were provided to assist in obtaining surgical services, the guardian or patient refused surgery in 37 patients (50%). In 11 additional residents (14%), the ophthalmologist refused to perform surgery. Of these 11 subjects, 3 had bilateral visual acuity worse than 20/200, and 3 had visual acuity between 20/40 and 20/60. Eight distinct ophthalmologists elected not to perform surgery in at least 1 resident, 7 of whom did operate on other residents as part of this study. A total of 26 residents (33%) were scheduled for surgery, 2 of whom had surgery postponed by the ophthalmologist. Six (33%) of the 26 residents with cataract had posterior capsular opacification requiring laser capsulotomy.
In the control homes, cataract surgery was recommended in 99 residents. Twelve (12%) of these residents obtained an ophthalmologic examination, and 2 individuals (2% of those referred) underwent cataract surgery on the basis of these examinations.
We found no difference in the likelihood of accepting cataract surgery when comparing those individuals consenting for themselves with those for whom a guardian gave consent. Of the 18 residents who were able to consent for themselves and who were offered cataract surgery, 61% (11 residents) ultimately underwent the procedure, which was similar to the 54% (30/56) for those who had a guardian. Age, race, gender, cognitive status, LOS, and visual acuity did not affect the likelihood of consenting for surgery (P>.2 for all factors).Older individuals were less likely to undergo cataract evaluation by a community ophthalmologist (Table 5).
This project highlights the many barriers to providing cataract surgery services to institutionalized, cognitively impaired elderly residents of nursing homes. Not only is testing vision in these residents more difficult than in community-dwelling individuals,16 but once residents are identified as having a cataract as the primary cause of vision loss, multiple obstacles prevent them from undergoing cataract surgery. We hypothesized these barriers to be transportation and the lack of a patient advocate to help in scheduling appointments. These were some of the barriers, as demonstrated by the improved uptake of surgery comparing the usual-care facilities (2 [2%]) with the intervention homes (24 [31%]). However, an important barrier that was unanticipated was the lack of willingness of family members, guardians, and the residents themselves to consent to surgery. Half of those found to have vision-impairing cataracts with no other clear explanation for vision loss refused cataract surgery. We found no difference in refusal rates when comparing self-consenting individuals with those dependent on guardians for consent.This is in contrast with the findings of Marx et al,23 who reported that guardians were less likely to give consent for residents than were residents themselves. The high rates of refusal (>50%) to be evaluated by a second ophthalmologist for a relatively benign, sight-restoring intervention points to the resignation that many residents and guardians feel about their poor vision. Many stated that they were unwilling to undergo any procedure, even if it improved their quality of life.
Depression is highly prevalent among older individuals, particularly those with cognitive impairment.24 This raises the difficult issue of what to do when a resident may not be behaving in his or her own best interest. Cataract surgery is not life-saving; rather, it is life-enhancing. How aggressive should caregivers be with reluctant, depressed, cognitively impaired individuals and their families who do not want to “bother” or “upset” the resident?
Another unanticipated barrier to care was the opinion of treating ophthalmologists. This project is community-based; therefore, ophthalmic surgeons not participating in the study evaluated the patients. Several physicians expressed concerns about the ability of residents to tolerate or benefit from surgery, even though all had agreed in advance to treat residents referred from nursing homes. More than 20% of residents referred to ophthalmologists for surgery were told not to bother, even though they had vision-impairing cataracts. The mandate to “do no harm” may drive this behavior, inasmuch as many of these individuals were severely cognitively impaired, but it may also represent a bias against treating very elderly or institutionalized individuals. Although caring for nursing home residents requires more time and effort, only 1 of 23 subjects who underwent cataract surgery required general anesthesia.
We found that individuals with visual impairment from cataract tend to be older and more often are black. Cataract is an age-related phenomenon, and it is not surprising that the oldest individuals tended to have more severe cataracts. Black persons have been documented in previous cross-sectional research to have a higher prevalence of vision loss from cataract25- 27 and recently were shown to be less likely to undergo cataract surgery than were white persons.28 This may be owing to multiple factors, including access to health care services, physician bias, and cultural beliefs regarding surgical interventions. In the present research, black and white residents refused surgery at similar rates, but our study did not have the power to detect differences in uptake of cataract surgery.
Other predictors of visual impairment from cataract include a lower MMSE score and a greater LOS in the nursing home. Having a lower MMSE may contribute to having untreated cataract in several ways: the cognitively impaired are less able to notify caregivers of their visual limitations, it is more difficult to test vision in these individuals, caregivers may be reluctant to subject the cognitively impaired to procedures, and the benefits of surgery may be less obvious for these individuals.
The greater likelihood of having untreated cataract in individuals residing in the nursing home for a longer time may be attributable to several factors. Eye care services are not routinely provided in nursing homes; therefore, residents in whom vision loss develops may be less likely to be evaluated. In addition, those who are resident the longest include some with worsening cognition, making them less able to express visual symptoms. Our finding of an association between LOS and untreated cataract suggests that increased levels of visual disability found in nursing home residents may be attributable to both a greater likelihood of nursing home admission among the visually impaired and a lower likelihood that visual impairment will be detected and treated after an individual is admitted to a nursing home.
It was surprising that age, race, gender, MMSE score, and LOS were not predictive of acceptance of cataract surgery. This may in part be because of a small sample size. Subjects who accepted cataract surgery were more likely (although not statistically significantly) to have higher MMSE scores and to be younger than those who refused.
Nursing home residents with bilateral vision-impairing cataract face significant obstacles to obtaining surgical services. Without a program to assist residents in identifying a surgeon, making it to the appointment, and getting to the hospital for surgery, only 2 (2%) of 99 identified by an ophthalmologist as having decreased vision due to cataract received surgery. In contrast, an intensive effort to support residents in the process led to one third (24 of 77 persons) being scheduled for surgery, all but 2 (30%) of whom had the surgery. If it is found that cataract surgery improves the quality of life of frail nursing home residents, then it will be essential to establish programs to remove barriers to access cataract surgery services.
Correspondence: David S. Friedman, MD, MPH, Wilmer Eye Institute, Room 120, 600 N Wolfe St, Baltimore, MD 21287 (David.Friedman@jhu.edu).
Submitted for Publication: June 11, 2004; final revision received January 7, 2005; accepted January 7, 2005.
Funding/Support: This study was supported by grant AG 15812 from the National Institute of Aging, National Institutes of Health, Bethesda, Md. Dr West is a Research to Prevent Blindness Senior Scientific Investigator. Dr Friedman is the recipient of the Research to Prevent Blindness Robert E. McCormick Scholarship and the American Geriatrics Society Dennis Jahnigen Scholars Award.