Association of Vision Loss With Hospital Use and Costs Among Older Adults | Geriatrics | JAMA Ophthalmology | JAMA Network
[Skip to Navigation]
Table 1.  Demographic Characteristics of Patients With No Vision Loss, Partial Vision Loss, and Severe Vision Loss
Demographic Characteristics of Patients With No Vision Loss, Partial Vision Loss, and Severe Vision Loss
Table 2.  Health Care Use for Patients With No Vision Loss, Partial Vision Loss, and Severe Vision Loss
Health Care Use for Patients With No Vision Loss, Partial Vision Loss, and Severe Vision Loss
Table 3.  Odds of Experiencing Different Outcomes for Persons With Vision Loss vs Persons With No Vision Loss
Odds of Experiencing Different Outcomes for Persons With Vision Loss vs Persons With No Vision Loss
1.
Chan  T, Friedman  DS, Bradley  C, Massof  R.  Estimates of incidence and prevalence of visual impairment, low vision, and blindness in the United States.  JAMA Ophthalmol. 2018;136(1):12-19. doi:10.1001/jamaophthalmol.2017.4655PubMedGoogle ScholarCrossref
2.
Congdon  N, O’Colmain  B, Klaver  CC,  et al; Eye Diseases Prevalence Research Group.  Causes and prevalence of visual impairment among adults in the United States.  Arch Ophthalmol. 2004;122(4):477-485. doi:10.1001/archopht.122.4.477PubMedGoogle ScholarCrossref
3.
Varma  R, Vajaranant  TS, Burkemper  B,  et al.  Visual impairment and blindness in adults in the United States: demographic and geographic variations from 2015 to 2050.  JAMA Ophthalmol. 2016;134(7):802-809. doi:10.1001/jamaophthalmol.2016.1284PubMedGoogle ScholarCrossref
4.
Gill  M, Chhabra  H, Shah  M, Zhu  C, Lando  H, Caldarella  F.  Association between provider specialty and healthcare costs and glycemic control for patients with diabetes.  J Med Econ. 2018;21(7):704-708. doi:10.1080/13696998.2018.1467324PubMedGoogle ScholarCrossref
5.
Elam  AR, Andrews  C, Musch  DC, Lee  PP, Stein  JD.  Large disparities in receipt of glaucoma care between enrollees in Medicaid and those with commercial health insurance.  Ophthalmology. 2017;124(10):1442-1448. doi:10.1016/j.ophtha.2017.05.003PubMedGoogle ScholarCrossref
6.
Stein  JD, Andrews  C, Musch  DC, Green  C, Lee  PP.  Sight-threatening ocular diseases remain underdiagnosed among children of less affluent families.  Health Aff (Millwood). 2016;35(8):1359-1366. doi:10.1377/hlthaff.2015.1007PubMedGoogle ScholarCrossref
7.
Stein  JD, Grossman  DS, Mundy  KM, Sugar  A, Sloan  FA.  Severe adverse events after cataract surgery among medicare beneficiaries.  Ophthalmology. 2011;118(9):1716-1723. doi:10.1016/j.ophtha.2011.02.024PubMedGoogle ScholarCrossref
8.
Centers for Medicare and Medicaid Services. CMS top ten Medicare diagnostic related groups by discharges FY2013. http://www.mmplusinc.com/news-articles/item/cms-releases-annual-medicare-payment-data. Accessed August 4, 2018.
9.
Charlson  ME, Pompei  P, Ales  KL, MacKenzie  CR.  A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.  J Chronic Dis. 1987;40(5):373-383. doi:10.1016/0021-9681(87)90171-8PubMedGoogle ScholarCrossref
10.
US Census Bureau. The nation’s older population is still growing, Census Bureau reports. https://www.census.gov/newsroom/press-releases/2017/cb17-100.html. Published June 22, 2017. Accessed August 10, 2018.
11.
Centers for Disease Control and Prevention. Number of overnight hospital stays during the past 12 months, by selected characteristics: United States, 2016. Summary Health Statistics: National Health Interview Survey, 2016. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/NHIS/SHS/2016_SHS_Table_P-10.pdf. Accessed August 19, 2018.
12.
Javitt  JC, Zhou  Z, Willke  RJ.  Association between vision loss and higher medical care costs in Medicare beneficiaries costs are greater for those with progressive vision loss.  Ophthalmology. 2007;114(2):238-245. doi:10.1016/j.ophtha.2006.07.054PubMedGoogle ScholarCrossref
13.
Frick  KD, Walt  JG, Chiang  TH,  et al.  Direct costs of blindness experienced by patients enrolled in managed care.  Ophthalmology. 2008;115(1):11-17. doi:10.1016/j.ophtha.2007.02.007PubMedGoogle ScholarCrossref
14.
Morse  AR, Yatzkan  E, Berberich  B, Arons  RR.  Acute care hospital utilization by patients with visual impairment.  Arch Ophthalmol. 1999;117(7):943-949. doi:10.1001/archopht.117.7.943PubMedGoogle ScholarCrossref
15.
Lord  SR, Dayhew  J.  Visual risk factors for falls in older people.  J Am Geriatr Soc. 2001;49(5):508-515. doi:10.1046/j.1532-5415.2001.49107.xPubMedGoogle ScholarCrossref
16.
Duffy  M. Accessibility barriers in medical and health care for people with vision loss: real issues, real problems. VisionAware. https://www.visionaware.org/blog/visionaware-blog/accessibility-barriers-in-medical-and-healthcare-for-people-with-vision-loss-real-issues-real-problems/. Published August 14, 2013. Accessed August 18, 2018.
17.
Zhi-Han  L, Hui-Yin  Y, Makmor-Bakry  M.  Medication-handling challenges among visually impaired population.  Arch Pharm Pract (Mumbai). 2017;8(1):8-14. doi:10.4103/2045-080X.199613Google ScholarCrossref
18.
Stuck  AE, Walthert  JM, Nikolaus  T, Büla  CJ, Hohmann  C, Beck  JC.  Risk factors for functional status decline in community-living elderly people: a systematic literature review.  Soc Sci Med. 1999;48(4):445-469. doi:10.1016/S0277-9536(98)00370-0PubMedGoogle ScholarCrossref
19.
Scotland  A. Catering for patients. http://www.audit-scotland.gov.uk/docs/health/2003/nr_031120_catering_patients.pdf. Published November 2003.
20.
Vision Australia. Caring for patients. https://www.visionaustralia.org/information/family-friends-carers/caring-for-patients. Accessed August 18, 2018.
21.
Marshall  S, Joffee  E. ADA checklist: health care facilities and service providers: ensuring access to services and facilities by patients who are blind, deaf-blind, or visually impaired. The Americans With Disabilities Act Communications Accommodations Project. http://www.afb.org/info/programs-and-services/public-policy-center/civil-rights/advocacy-resources/ada-checklist-health-care-facilities-and-service-providers/12345. Accessed August 25, 2018.
Original Investigation
April 4, 2019

Association of Vision Loss With Hospital Use and Costs Among Older Adults

Author Affiliations
  • 1Lighthouse Guild, New York, New York
  • 2Department of Ophthalmology, Columbia University, New York, New York
  • 3Department of Ophthalmology, New York University School of Medicine, New York
  • 4Department of Ophthalmology and Visual Sciences, University of Michigan Medical School, Ann Arbor
  • 5Center for Eye Policy and Innovation, University of Michigan, Ann Arbor
  • 6Department of Health Management and Policy, University of Michigan School of Public Health, Ann Arbor
JAMA Ophthalmol. 2019;137(6):634-640. doi:10.1001/jamaophthalmol.2019.0446
Key Points

Question  Do hospitalized patients with vision loss experience greater resource use and costs compared with hospitalized patients without vision loss?

Findings  In this study of 12 330 Medicare beneficiaries and 11 858 commercial health insurance enrollees with or without vision loss, severe vision loss was associated with longer mean length of stay, higher readmission rates, and higher costs during hospitalization and 90 days after discharge.

Meaning  These results suggest that, by addressing vision-related issues, opportunities exist to reduce lengths of stay, readmission rates, resource use, and costs while enhancing outcomes and patient satisfaction.

Abstract

Importance  Patients with vision loss who are hospitalized for common illnesses are often not identified as requiring special attention. This perception, however, may affect the outcomes, resource use, and costs for these individuals.

Objective  To assess whether the mean hospitalization lengths of stay, readmission rates, and costs of hospitalization differed between individuals with vision loss and those without when they are hospitalized for similar medical conditions.

Design, Setting, and Participants  This analysis of health care claims data used 2 sources: Medicare database and Clinformatics DataMart. Individuals with vision loss were matched 1:1 to those with no vision loss (NVL), on the basis of age, years from initial hospitalization, sex, race/ethnicity, urbanicity of residence, and overall health. Both groups had the same health insurance (Medicare or a commercial health plan), and all had been hospitalized for common illnesses. Vision loss was categorized as either partial vision loss (PVL) or severe vision loss (SVL). Data were analyzed from April 2015 through April 2018.

Main Outcomes and Measures  The outcomes were lengths of stay, readmission rates, and health care costs during hospitalization and 90 days after discharge. Multivariable logistic and linear regression models were built to identify factors associated with these outcomes among the NVL, PVL, and SVL groups.

Results  Among Medicare beneficiaries, 6165 individuals with NVL (with a mean [SD] age of 82.0 [8.3] years, and 3833 [62.2%] of whom were female) were matched to 6165 with vision loss. Of those with vision loss, 3401 (55.2%) had PVL and 2764 (44.8%) had SVL. In the Clinformatics DataMart database, 5929 individuals with NVL (with a mean [SD] age of 73.7 [15.1] years, and 3587 [60.5%] of whom were female) were matched to 5929 individuals with vision loss. Of the commercially insured enrollees with vision loss, 3515 (59.3%) had PVL and 2414 (40.7%) had SVL. Medicare enrollees with SVL, compared with those with NVL, had longer mean lengths of stay (6.48 vs 5.26 days), higher readmission rates (23.1% vs 18.7%), and higher hospitalization and 90-day postdischarge costs ($64 711 vs $61 060). Compared with those with NVL, Medicare beneficiaries with SVL had 4% longer length of stay (estimated ratio, 1.04; 95% CI, 1.01-1.07; P = .02), 22% higher odds of readmission (odds ratio, 1.22; 95% CI, 1.06-1.41; P = .007), and 12% higher costs (estimated cost ratio, 1.12; 95% CI, 1.06-1.18; P < .001). Similar findings were obtained for those with commercial health insurance. When these findings were extrapolated to hospitalizations of patients with vision loss nationwide, an estimated amount of more than $500 million in additional costs annually were spent caring for these patients.

Conclusions and Relevance  These findings suggest that opportunities for improving outcomes and reducing costs exist in addressing patients’ vision loss and concomitant functional difficulties during hospitalization and thereafter.

Introduction

Visual loss or blindness affects nearly 4 million adults in the United States.1 The number of persons with vision loss is expected to increase substantially as prevalence rates of macular degeneration, glaucoma, diabetic retinopathy, and other eye diseases are projected to rise.2,3 Typically, when patients with vision loss are hospitalized for common illnesses, they are not identified as requiring any special attention. Empirical evidence suggests that persons with vision loss may have difficulty following hospital routines and, once discharged, may struggle to read discharge orders and medication instructions, which may result in poor outcomes. If patients with vision loss use greater resources during and after hospitalization and incur greater costs, opportunities may exist to improve outcomes and reduce costs by employing targeted, assistive, patient-centric interventions during the hospitalization and immediately after discharge.

Using data from nationwide samples of more than 12 000 individuals with vision loss matched to others with no vision loss, all of whom were hospitalized for common illnesses, we assessed whether vision loss was associated with lengths of stay (LOS), readmissions, and costs during hospitalization and 90 days after discharge.

Methods
Data Sources

We used 2 health care claims databases: Medicare database and Clinformatics DataMart. Medicare enrollee data came from a randomly selected 20% nationally representative sample of Medicare beneficiaries with coverage from January 1, 2008, through December 31, 2014. The Clinformatics DataMart (OptumInsight) data set covered persons in a nationwide managed care network from January 1, 2001, through December 31, 2014. Both databases capture inpatient and outpatient care and include diagnoses coded according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Visits and diagnostic and therapeutic procedures were captured using Current Procedural Terminology, Fourth Edition, codes and diagnosis-related groups (DRGs). The Medicare database contains payment data for all services rendered. The Clinformatics DataMart provides closely approximated payments using standard prices.4 These data sources have been used previously to study patients with ocular diseases.5-7 The University of Michigan Institutional Review Board approved this research as a nonregulated study. No informed consent was required because all data were deidentified to the researchers. Data were analyzed from April 2015 through April 2018.

Sample Selection

We identified all enrollees with a record of 1 or more inpatient hospitalizations for which the primary diagnosis was 1 of the following: major joint replacement (DRG 470), pneumonia (DRG 193 and 194), digestive disorder (DRG 393), heart failure (DRG 291), urinary tract infection (DRG 690), sepsis (DRG 871), renal failure (DRG 683), and chronic obstructive pulmonary disease (DRG 190). These diagnoses are consistently among the top 10 reasons for US hospitalizations.8 For Medicare recipients, we also required an age of 65 years or older, continuous enrollment in traditional fee-for-service Medicare for 1 year or more before hospitalization, and plan enrollment for 90 days or more after the date of initial discharge. All enrollees had Medicare Parts A, B, and D coverage. We excluded Medicare Advantage Plan enrollees as our data source lacked all claims for these patients. For patients in the Clinformatics DataMart database, we used similar inclusion and exclusion criteria, except we included persons as young as 21 years of age.

Sample Categorization

Quiz Ref IDWe reviewed all claims submitted for 2 years before the initial hospitalization and characterized each eligible enrollee as having severe vision loss (SVL), partial vision loss (PVL), or no vision loss (NVL). To identify enrollees with vision loss we used ICD-9-CM billing codes 369.xx, reflecting the extent of vision loss and whether one or both eyes are affected. Enrollees were grouped as SVL if they had 1 or more records of ICD-9-CM codes 369.0 × to 369.4x. Those with 1 or more records of ICD-9-CM codes 369.6 × to 369.9 × and no SVL codes were categorized as PVL. Individuals without a record of any 369.xx code composed the NVL group. Because some eye care practitioners may not use ICD-9-CM codes 369.xx but instead use codes for the conditions responsible for the vision loss, we excluded enrollees with any common sight-threatening ocular disease from the NVL group. For inclusion in the NVL group, we required that individuals had visited an ophthalmologist or optometrist in the year before their hospitalization to give them an opportunity to receive a sight-threatening diagnosis.

Sample Matching

In the Medicare sample, 6179 persons had codes for vision loss and 248 653 for NVL. In the Clinformatics DataMart sample, 5963 persons had vision loss codes and 80 622 had NVL codes. In both databases, we matched 1:1 the individuals with any vision loss (PVL or SVL) to those with NVL on the basis of age (SD, 5 years), calendar years from the initial hospitalization (SD, 2 years), sex, race/ethnicity, urbanicity of residence, and overall health according to the Charlson Comorbidity Index in the year before the initial hospitalization.9 In the Medicare sample, we successfully matched 6165 (99.8%) of the 6179 persons with vision loss to 6165 individuals with NVL, for a total sample of 12 330 enrollees. In the Clinformatics DataMart sample, we matched 5929 (99.4%) of the 5963 persons with vision loss to 5929 individuals with NVL, for a total sample of 11 858 enrollees.

Statistical Analysis

Statistical analyses were performed using SAS, version 9.4 (SAS Institute, Inc). The continuous variable descriptive statistics were compared with 2-tailed t tests (when comparing vision loss with NVL) and analysis of variance (when comparing the 3 groups). χ2 Tests were used for categorical variables (whether comparing vision loss to NVL or the 3 groups). Significance levels were set at P < .05.

Enrollee characteristics were summarized using means and SDs for continuous variables and log-transformed for skewed data. Frequencies and percentages are reported for categorical variables. Outcomes of interest were LOS, hospital readmission within 1 month of discharge, and mean total of health care cost during hospitalization and 90 days after discharge. We built multivariable stratified logistic regression models to identify factors associated with hospital readmissions and multivariable linear regression models to identify factors associated with LOS of the initial hospitalization and costs. We ran separate models for the Medicare and Clinformatics DataMart data sets. For each model, the key indicator was whether the enrollee had SVL or PVL compared with NVL. Other model covariates included the primary diagnosis for the hospitalization, selected medical (eg, diabetes, dementia, and depression) and ocular (eg, glaucoma, macular degeneration, cataract, and diabetic retinopathy) comorbidities. In the Clinformatics DataMart analyses, we also adjusted for household income and educational level.

Excess Costs Calculation

We estimated excess costs to the US health care system for hospitalizations of enrollees with vision loss compared with hospitalizations of those with NVL for the same medical conditions. Using data for the total number of older adults hospitalized in the United States in 2016,10,11 the proportion of the population with visual impairment,1-3 and the additional costs associated with caring for patients with VL from our analyses, we calculated the excess annual costs associated with hospitalization of enrollees with vision loss and extrapolated the study findings to the entire US population aged 65 years or older.

Results

Among Medicare beneficiaries, 6165 individuals with NVL (with a mean [SD] age of 82.0 [8.3] years, and 3833 [62.2%] of whom were female) were matched to 6165 individuals with vision loss. Of the beneficiaries with vision loss, 3401 (55.2%) had PVL (with a mean [SD] age of 80.4 [7.9] years, and 2050 [60.3%] of whom were female) and 2764 (44.8%) had SVL (with a mean [SD] age of 83.9 [8.3] years, and 1783 [64.5%] of whom were female). In the Clinformatics DataMart database, 5929 individuals with NVL (with a mean [SD] age of 73.7 [15.1] years, and 3587 [60.5%] of whom were female) were matched to 5929 individuals with vision loss. Of enrollees with vision loss, 3515 (59.3%) had PVL (with a mean [SD] age of 71.8 [14.8] years, and 2085 [59.3%] of whom were female), and 2414 (40.7%) had SVL (with a mean [SD] age of 76.6 [15.3] years, and 1502 [62.2%] of whom were female) (Table 1).

Lengths of Stay

Quiz Ref IDFor the Medicare sample, the mean (SD) LOS of hospitalization was 5.26 (7.6) days for individuals with NVL, 5.01 (5.4) days for individuals with PVL, and 6.48 (15.00) days for individuals with SVL. Compared with those with NVL, beneficiaries with PVL had 4% shorter LOS (estimated LOS ratio, 0.96; 95% CI, 0.94-0.99; P = .01), whereas those with SVL had 4% longer LOS (estimated LOS ratio, 1.04; 95% CI, 1.01-1.07; P = .02). In the Clinformatics DataMart sample, the mean (SD) LOS of hospitalizations were 4.61 (5.9) days for individuals with NVL, 4.94 (6.5) days for individuals with PVL, and 5.97 (9.2) days for individuals with SVL. Compared with the NVL group, the PVL group had 5% longer LOS (estimated LOS ratio, 1.05; 95% CI, 1.01-1.08; P = .005), and the SVL group had 4% longer LOS (estimated LOS ratio, 1.04; 95% CI, 1.00-1.08; P = .03) (Table 2 and Table 3).

Hospital Readmissions

Quiz Ref IDIn the Medicare sample, 1154 individuals (18.7%) with NVL, 708 (20.8%) with PVL, and 639 (23.1%) with SVL were readmitted to the hospital within 1 month of discharge. Compared with those with NVL, beneficiaries with PVL had 15% higher odds of readmission (odds ratio [OR], 1.15; 95% CI, 1.02-1.31; P = .02), and those with SVL had 22% higher odds of readmission (OR, 1.22; 95% CI, 1.06-1.41; P = .007). In the Clinformatics DataMart sample, 871 individuals (14.7%) with NVL, 563 (16.0%) with PVL, and 510 (21.1%) with SVL were readmitted to the hospital within 1 month of discharge. Compared with those with NVL, enrollees with PVL had 30% higher odds of readmission (OR, 1.30; 95% CI, 1.08-1.55; P = .005), and those with SVL had 32% higher odds of readmission (OR, 1.32; 95% CI, 1.07-1.62; P = .01) (Table 2 and Table 3).

Costs

For the Medicare sample, the mean costs of health care services during hospitalization and 90 days after discharge were $61 060 for individuals with NVL, $61 261 for individuals with PVL, and $64 711 for individuals with SVL. Compared with costs for those with NVL, the costs for those with PVL were not statistically significant (estimated cost ratio, 1.02; 95% CI, 0.97-1.07; P = .40), whereas those with SVL had 12% higher costs (estimated cost ratio, 1.12; 95% CI, 1.06-1.18; P < .001). For the Clinformatics DataMart sample, the mean (SD) 90-day costs were $47 287 ($45 585) for the NVL group, $48 870 ($48 240) for the PVL group, and $51 133 ($51 237) for the SVL group. Compared with enrollees with NVL, those with PVL had 5% higher costs (estimated cost ratio, 1.05; 95% CI, 1.01-1.09; P = .01), whereas those with SVL had 8% higher costs (estimated cost ratio, 1.08; 95% CI, 1.03-1.13; P = .001) (Table 2 and Table 3).

Excess Costs to the US Health Care System

Quiz Ref IDTo estimate the excess 90-day cost associated with hospitalization of enrollees with vision loss, we used data from several sources. The US Census Bureau reported that 49 115 382 people in the United States were aged 65 years or older in 2016.10 According to the National Health Interview Survey, 15.2% of adults aged 65 years or older reported 1 or more hospitalizations in 2016.11 Applying the 15.2% hospitalization rate for this age group, we calculated that older adults had approximately 7 465 538 hospitalizations. Based on estimates from Chan et al,1 among all hospitalizations of older adults, 59 724 persons had SVL and 223 966 persons had PVL. Using the differences in 90-day costs between the Medicare enrollees in the NVL group and the PVL and SVL groups (Table 2), we estimated that older people with SVL had an excess of $229 699 504 in costs and those with PVL accrued an excess of $354 538 178, for a total of $584 237 682 additional costs associated with hospitalizations for patients with vision loss.

Discussion

Using 2 large health care claims databases, we identified individuals with and without vision loss who were hospitalized for common medical conditions. Medicare beneficiaries and those with commercial health insurance with vision loss had considerably greater health care resource use and costs during and immediately after hospitalization, compared with those without vision loss. Extrapolating these findings to older adults suggests that hospitalization of patients with vision loss is associated with excess estimated health care costs of more than $500 million annually.

These findings align with previous research demonstrating that managing patients with vision loss tended to be more costly compared with those with NVL. For example, using data from 1999 to 2003, Javitt and colleagues12 found that the mean cost for non–eye-related medical care services was substantially higher for persons with SVL or blindness than for others with NVL and that persons with vision loss were at increased risk for injury and depression and had greater use of skilled-nursing and long-term care facilities. Extrapolating their findings to the entire Medicare population, Javitt et al12 calculated that blindness and vision loss resulted in more than $2 billion in non-eye–related medical costs. Using 1996 to 2002 data from the Medical Expenditure Panel Survey, Frick and colleagues13 reported that, in the first year of enrollment, persons with blindness incurred $7356 in higher costs and $1329 in higher costs in subsequent years, compared with individuals with NVL. Excess medical care expenditures for persons with visual impairment or blindness were estimated to total more than $5 billion, much of which is owing to home care needs, costs that we did not consider in our analyses. Previously, Morse et al14 found that hospital LOS was greater by 2.4 days for patients with a vision loss diagnosis, compared with patients with NVL, admitted for nonocular disorders. The absence of detailed ICD-9-CM codes analysis precludes the direct comparison of the present study with that of Morse et al.14 However, changes in health care practices, including an increase in availability and use of home health care have helped to reduce LOS over the past several decades. That said, our findings are directionally consistent and continue to indicate that health care resource use and costs remain higher for older adults with vision loss or blindness.

For most outcomes, the PVL group’s health care use was at an intermediate level between the NVL and the SVL groups. Although this finding might be unexpected given that visual acuity may be maintained in one eye for individuals classified as PVL, monocular vision loss can result in subtle defects in visual processing. For example, disparities in binocular input can result in decreased stereopsis, a function important for tasks requiring hand-eye coordination, balance, and gait. Lord and Dayhew15 reported that individuals with good vision in one eye and moderate or poor vision in the other eye had increased fall rates equivalent to the fall rates of those with bilateral moderate or poor vision. Deficits in binocular function may be one cause of the increased health care use observed in the PVL group.

Hospitalization presents challenges for patients with vision loss. Consent forms, preadmission protocols and instructions, and postdischarge routines and regimens are often not accessible for these patients. Some patients with visual impairment require these documents to be in larger fonts, whereas others with more advanced vision loss may require braille or audio methods of communication. Few health care facilities are equipped to address these issues. One study found that only 23% of physician offices and hospitals had large-print materials available.16 Moreover, simple tasks such as indicating food choices, locating nursing call buttons, or identifying support staff can be difficult for hospitalized patients with vision loss. Ambulation is generally desirable during hospitalization to reduce risk of venous thrombosis and pressure ulcers. Yet for many patients with visual impairment, ambulation requires assistance from someone to address possible obstacles in hallways and patient rooms, which may increase the risk for injuries.

At discharge, coordinating care among multiple health care practitioners and across different care settings when few or none of the clinicians are aware of, or able to, adequately address vision loss–associated needs may mean that patients with blindness inevitably do not receive the care they need and continue to experience challenges. They may struggle with medication adherence and be prone to medication errors,17 especially if they are required to take multiple medications and cannot properly distinguish them without nonvisual cues. This problem is compounded if patients lack caregiver assistance. Furthermore, vision loss is a risk factor for functional decline, an aspect rarely considered in the health care context.18 These factors likely contributed to the higher proportion of patients with vision loss in the present study who required hospital readmission.

Although many hospitals may be ill equipped to address care needs of patients with vision loss and other disabilities, some hospitals and health care systems have invested resources to enhance the experience of patients with disabilities during hospitalization. For example, to facilitate selection of food choices for hospitalized patients with sensory impairments, the Lanarkshire Primary Care National Health Service Trust in Scotland requires menus to be available on audiotape.19 Likewise, Vision Australia and the Public Policy and Research Center of the American Foundation for the Blind provide recommendations on caring for patients with visual disabilities when they are hospitalized.20,21

As health systems, insurers, and policymakers look for ways to reduce LOS, readmissions, and their associated costs, opportunities for improving care quality may also emerge. For example, hospital staff should consider comorbidities such as vision loss, hearing loss, and cognitive impairment at the time of hospital admission, throughout the hospitalization, and during the discharge-planning process. During the admission process, a quick vision evaluation could be done to determine whether the patient can read printed text. If a deficit is found, a hospital bracelet indicating vision loss could be placed on the patient’s wrist and an indicator placed on the patient’s room door, similar to what is often done for patients at increased risk for falls. This indicator would alert physicians and health care staff to the patient’s visual impairment and potential need for additional assistance. Findings of vision loss should be documented in the patient’s electronic health record to ensure proper accommodation throughout the hospitalization and on discharge. Patient care instructions could be provided in large font sizes or in accessible formats. Patients should be referred to an eye care practitioner to follow up on newly identified vision loss during the hospitalization, as is commonly done for patients with physical or occupational therapy needs. These procedures could improve patient care and satisfaction while reducing costs downstream.

These results may encourage health policymakers to identify ways to provide incentives to hospitals and health care systems to work with researchers, patients and their caregivers, and staff at facilities to identify strategies to enhance the care of older adults with vision or other sensory loss. Although some costs may be incurred to make facilities and hospital personnel better equipped to care for these patients, the potential savings and improvement in quality of care may make this undertaking a good investment.

Limitations

This study has several limitations. First, using only claims data, we lacked access to pertinent clinical information such as best-corrected visual acuity and problems like extensive peripheral vision loss or metamorphopsia. Presumably, if enrollees in the NVL group had conditions known to be associated with vision loss, they would have received such diagnoses from their eye care professional. Moreover, misclassifying persons with vision loss as NVL would bias these findings to the null; that is, the statistical significance of the findings would be understated. Second, some eye care practitioners use the ICD-9-CM codes for conditions they are treating rather than the vision loss (369.xx) billing codes. We compiled a list of common sight-threatening diseases and screened the persons in the NVL group to exclude persons with these conditions. Despite this screening, some persons in the NVL group may have been misclassified, which would also bias these results to the null. Because ophthalmologic ICD-9-CM codes are frequently associated with claims for low vision rehabilitation services, the study sample may overrepresent these patients and, therefore, underestimate excess resource use and cost given that patients who have already received vision rehabilitation would have learned methods of accommodation and compensation. Third, these results may not be generalizable to persons without insurance or with types of health insurance other than Medicare or a commercial plan.

Conclusions

These findings suggest that identifying the presence of vision loss during hospitalization or the discharge-planning period and employing strategies to assist these patients may be associated with improved outcomes, fewer readmissions, shorter LOS, better patient satisfaction, and (if applied across the United States) a cost savings of more than $500 million annually.

Back to top
Article Information

Accepted for Publication: January 31, 2019.

Corresponding Author: Alan R. Morse, JD, PhD, Lighthouse Guild, 250 W 64th St, New York, NY 10023 (armorse@lighthouseguild.org).

Published Online: April 4, 2019. doi:10.1001/jamaophthalmol.2019.0446

Author Contributions: Dr Stein had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Morse, Stein.

Acquisition, analysis, or interpretation of data: Seiple, Talwar, Lee, Stein.

Drafting of the manuscript: Morse, Seiple, Stein.

Critical revision of the manuscript for important intellectual content: Morse, Seiple, Talwar, Lee.

Statistical analysis: Seiple, Talwar.

Supervision: Morse.

Conflict of Interest Disclosures: Dr Morse reported grants from Linder Fund outside of the submitted work. Dr Stein reported grants from the National Eye Institute, Lighthouse Guild, and Research to Prevent Blindness during the conduct of the study. No other disclosures were reported.

Funding/Support: This study was supported by Linder Fund (Dr Morse), W. K. Kellogg Foundation (Drs Lee and Stein), Research to Prevent Blindness (Dr Stein), Department of Veterans Affairs RR&D (Dr Seiple), and grant R01EY026641 from the National Institutes of Health (Dr Stein).

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentations: Portions of this paper were presented at the Association for Research in Vision and Ophthalmology meetings, May 4, 2016, Seattle, Washington, and May 10, 2017, Baltimore, Maryland.

References
1.
Chan  T, Friedman  DS, Bradley  C, Massof  R.  Estimates of incidence and prevalence of visual impairment, low vision, and blindness in the United States.  JAMA Ophthalmol. 2018;136(1):12-19. doi:10.1001/jamaophthalmol.2017.4655PubMedGoogle ScholarCrossref
2.
Congdon  N, O’Colmain  B, Klaver  CC,  et al; Eye Diseases Prevalence Research Group.  Causes and prevalence of visual impairment among adults in the United States.  Arch Ophthalmol. 2004;122(4):477-485. doi:10.1001/archopht.122.4.477PubMedGoogle ScholarCrossref
3.
Varma  R, Vajaranant  TS, Burkemper  B,  et al.  Visual impairment and blindness in adults in the United States: demographic and geographic variations from 2015 to 2050.  JAMA Ophthalmol. 2016;134(7):802-809. doi:10.1001/jamaophthalmol.2016.1284PubMedGoogle ScholarCrossref
4.
Gill  M, Chhabra  H, Shah  M, Zhu  C, Lando  H, Caldarella  F.  Association between provider specialty and healthcare costs and glycemic control for patients with diabetes.  J Med Econ. 2018;21(7):704-708. doi:10.1080/13696998.2018.1467324PubMedGoogle ScholarCrossref
5.
Elam  AR, Andrews  C, Musch  DC, Lee  PP, Stein  JD.  Large disparities in receipt of glaucoma care between enrollees in Medicaid and those with commercial health insurance.  Ophthalmology. 2017;124(10):1442-1448. doi:10.1016/j.ophtha.2017.05.003PubMedGoogle ScholarCrossref
6.
Stein  JD, Andrews  C, Musch  DC, Green  C, Lee  PP.  Sight-threatening ocular diseases remain underdiagnosed among children of less affluent families.  Health Aff (Millwood). 2016;35(8):1359-1366. doi:10.1377/hlthaff.2015.1007PubMedGoogle ScholarCrossref
7.
Stein  JD, Grossman  DS, Mundy  KM, Sugar  A, Sloan  FA.  Severe adverse events after cataract surgery among medicare beneficiaries.  Ophthalmology. 2011;118(9):1716-1723. doi:10.1016/j.ophtha.2011.02.024PubMedGoogle ScholarCrossref
8.
Centers for Medicare and Medicaid Services. CMS top ten Medicare diagnostic related groups by discharges FY2013. http://www.mmplusinc.com/news-articles/item/cms-releases-annual-medicare-payment-data. Accessed August 4, 2018.
9.
Charlson  ME, Pompei  P, Ales  KL, MacKenzie  CR.  A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.  J Chronic Dis. 1987;40(5):373-383. doi:10.1016/0021-9681(87)90171-8PubMedGoogle ScholarCrossref
10.
US Census Bureau. The nation’s older population is still growing, Census Bureau reports. https://www.census.gov/newsroom/press-releases/2017/cb17-100.html. Published June 22, 2017. Accessed August 10, 2018.
11.
Centers for Disease Control and Prevention. Number of overnight hospital stays during the past 12 months, by selected characteristics: United States, 2016. Summary Health Statistics: National Health Interview Survey, 2016. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/NHIS/SHS/2016_SHS_Table_P-10.pdf. Accessed August 19, 2018.
12.
Javitt  JC, Zhou  Z, Willke  RJ.  Association between vision loss and higher medical care costs in Medicare beneficiaries costs are greater for those with progressive vision loss.  Ophthalmology. 2007;114(2):238-245. doi:10.1016/j.ophtha.2006.07.054PubMedGoogle ScholarCrossref
13.
Frick  KD, Walt  JG, Chiang  TH,  et al.  Direct costs of blindness experienced by patients enrolled in managed care.  Ophthalmology. 2008;115(1):11-17. doi:10.1016/j.ophtha.2007.02.007PubMedGoogle ScholarCrossref
14.
Morse  AR, Yatzkan  E, Berberich  B, Arons  RR.  Acute care hospital utilization by patients with visual impairment.  Arch Ophthalmol. 1999;117(7):943-949. doi:10.1001/archopht.117.7.943PubMedGoogle ScholarCrossref
15.
Lord  SR, Dayhew  J.  Visual risk factors for falls in older people.  J Am Geriatr Soc. 2001;49(5):508-515. doi:10.1046/j.1532-5415.2001.49107.xPubMedGoogle ScholarCrossref
16.
Duffy  M. Accessibility barriers in medical and health care for people with vision loss: real issues, real problems. VisionAware. https://www.visionaware.org/blog/visionaware-blog/accessibility-barriers-in-medical-and-healthcare-for-people-with-vision-loss-real-issues-real-problems/. Published August 14, 2013. Accessed August 18, 2018.
17.
Zhi-Han  L, Hui-Yin  Y, Makmor-Bakry  M.  Medication-handling challenges among visually impaired population.  Arch Pharm Pract (Mumbai). 2017;8(1):8-14. doi:10.4103/2045-080X.199613Google ScholarCrossref
18.
Stuck  AE, Walthert  JM, Nikolaus  T, Büla  CJ, Hohmann  C, Beck  JC.  Risk factors for functional status decline in community-living elderly people: a systematic literature review.  Soc Sci Med. 1999;48(4):445-469. doi:10.1016/S0277-9536(98)00370-0PubMedGoogle ScholarCrossref
19.
Scotland  A. Catering for patients. http://www.audit-scotland.gov.uk/docs/health/2003/nr_031120_catering_patients.pdf. Published November 2003.
20.
Vision Australia. Caring for patients. https://www.visionaustralia.org/information/family-friends-carers/caring-for-patients. Accessed August 18, 2018.
21.
Marshall  S, Joffee  E. ADA checklist: health care facilities and service providers: ensuring access to services and facilities by patients who are blind, deaf-blind, or visually impaired. The Americans With Disabilities Act Communications Accommodations Project. http://www.afb.org/info/programs-and-services/public-policy-center/civil-rights/advocacy-resources/ada-checklist-health-care-facilities-and-service-providers/12345. Accessed August 25, 2018.
×