Caring for Older Adults With Self-Reported Vision Impairment: Findings from the National Study of Caregiving


  • In this nationally representative study, caregivers of older adults who reported vision impairment spent 36% more hours providing care and reported that 61% more of their valued activities were affected per month.

  • They also had 46% greater odds of experiencing substantial emotional difficulty, per month, than caregivers of adults without vision impairment.

  • No differences between caregivers of older adults with and without vision impairment were noted in terms of experiencing substantial physical or financial difficulty.

  • Caring for older adults with vision impairment requires greater time commitments and has a greater impact on participation in valued activities and emotional well-being than does caring for adults without vision impairment.


The purpose of this study was to examine caregiving relationships for older adults with vision impairment (VI).


Cross-sectional study with a nationally representative sample.


Setting: the National Health and Aging Trends Study linked to the National Study of Caregiving, Year 2011. Study population: 1,776 family or unpaid caregivers to community-dwelling Medicare beneficiaries age ≥65 years old. Outcome measurement: in the preceding month, 1) the number of hours of care provided; 2) the valued activities affected by caregiving; and 3) the odds of experiencing substantial emotional, financial, and physical difficulty related to providing care. Exposure: VI was defined as a report of blindness or difficulty with distance or near vision.


Among 1,776 caregivers, 428 caregivers spent an average ± standard error (SE) of 111 ± 9.1 hours per month assisting older adults with VI, whereas 1,348 spent an average of 72 ± 3.3 hours assisting older adults without VI. In fully adjusted negative binomial regression analyses, caregivers of older adults with VI spent 36% more hours (incident rate ratio [IRR]: 1.36; 95% confidence interval [CI]: 1.15-1.60) providing care and reported having 61% more valued activities affected (IRR: 1.61; 95% CI: 1.23-2.10) than caregivers of older adults without VI. In fully adjusted logistic regression analyses, caregivers of older adults with VI had greater odds of emotional (odds ratio [OR]: 1.46; 95% CI: 1.04-2.03) but not financial (OR: 1.33; 95% CI: 0.87-2.03) or physical (OR: 1.13; 95% CI: 0.74-1.74) difficulty related to providing care than caregivers of older adults without VI.


These results suggest that caring for older adults with VI places different demands on time and emotional wellbeing than caring for older adults without VI, but no differences in financial or physical difficulties.

A s the aging population continues to grow in the United States, the prevalence of vision impairment (VI) is also projected to increase. In 2015, among US adults ≥40 years old, approximately 3 million people were visually impaired. By 2050, this number is projected to more than double to approximately 7 million people. Greater numbers of older adults living with impairment portend rising care needs for the aging population.

There is ample evidence to suggest that VI is associated with a decrease in functional status, specifically mobility and physical performance. , Family members play a key role in assisting visually impaired older adults with daily functioning and managing health. , However, despite evidence of the influence of caregiver support on older adults’ health outcomes, there is limited research examining in a comprehensive manner the impact of caregiving for persons with VI. A recent systematic review found that caregivers are at an increased risk of depression, especially when caring for individuals with VI who have greater difficulty with activities of daily living and require more extensive supervision. In addition, studies that have focused on specific eye diseases such as age-related macular degeneration have documented the increased cost and magnitude of care associated with worsening visual acuity. Although there have been some efforts to understand how VI affects older adults and their caregivers, existing research is largely based on small samples and does not go beyond examining the effect on caregivers’ psychosocial health.

A unique partnership exists between a caregiver and the person they assist that requires sustained efforts to preserve individual autonomy while recognizing the evolving needs of and challenges faced by both the older adult and their caregiver. Given evidence gaps, there is a need to investigate and understand caregiving for older adults with VI in order to effectively support caregivers and educate health care professionals in providing comprehensive care and cultivating complex partnerships. Therefore, this study extends the existing medical literature by analyzing the caregiving needs of older adults with VI and the impact of providing care on family caregivers, using nationally representative data from the United States. Findings may provide evidence to optimize the bilateral support system between the caregivers and older adults with VI.


Study Population

Cross-sectional analyses were performed of data from the National Health and Aging Trends Study (NHATS), a nationally representative survey of Medicare beneficiaries aged 65 years and older, linked to the National Study of Caregiving I (NSOC), a study of family and other unpaid caregivers to NHATS participants. The linked surveys together provide care recipient and caregiver perspectives on late-life care at a national level. These data are exempt from institutional review board approval as they are de-identified and publicly available. Informed written consent was obtained from NHATS participants, and oral consent was obtained from NSOC participants prior to data collection.

In the NHATS, participants completed an in-person survey by themselves or with the assistance of a proxy respondents who provided information about their health, environment, and ability to complete everyday tasks. Although the NHATS data were able to provide longitudinal estimates, as there have been multiple rounds of data collection, this study focused on Round 1 (2011) data to perform a cross-sectional analysis. The NHATS data were then linked to the NSOC data through a sample person identifier.

The NSOC surveyed the experience of up to 5 caregivers per NHATS participant, providing measurements such as care activities; duration of care; aspects of caregiving; support environment; participation; health of caregiver; employment and caregiving; and health insurance and income. Caregivers of NHATS participants who received assistance with self-care, mobility, or household activities (the last only if specifically for health of functioning reasons) were eligible for telephone interviews if they assisted the NHATS participants with mobility, self-care activities, household activities, transportation, or medical care.

Analytic Sample

Of the 2,423 NHATS participants eligible for NSOC, 4,935 of their caregivers met study eligibility criteria ( Figure 1 ). The NHATS participants did not provide contact information for 1,573 eligible family caregivers, and 1,355 of the remaining 3,362 eligible family caregivers could not be located or refused to respond, yielding 68.1% and 59.7% response rates at the first and second stages, respectively. , The study weights allowed adjustment for differential probabilities of selection at the NHATS sample person and caregiver levels, thereby providing nationally representative estimates. NSOC interviews were conducted with 2,007 caregivers of 1,369 older adults. Among them, 1,786 caregivers (1,684 spouses or offspring and 102 other relatives) who provided care to 1,199 community-dwelling older adults were included (those in residential care facilities who relied on the availability of supportive services were excluded). The study sample was further limited to 1,776 caregivers who had provided assistance with any activity in the previous month to 1,196 older adults from the NHATS.

Figure 1

Study participation selection. NHATS = National Health and Aging Trends Study; NSOC = National Study of Caregiving.

Caregiving Outcomes

Caregiving outcomes, all collected by self-report and pertaining to the previous month, were analyzed, as previously done. , Outcomes included 1) the number of hours of care provided; 2) the number of valued activities affected by caregiving; and 3) the substantial emotional, financial, and physical difficulty experienced by the caregiver due to providing care in the previous month. Valued activities refer to those reported as being very or somewhat important to the caregiver that were limited in the prior month because of caregiving and included: 1) visiting friends and family; 2) going out for enjoyment; 3) attending religious services; and 4) participating in club meetings or group activities. Restriction of participation in each valued activity was coded as a binary variable (yes/no) and then summed to obtain the composite score of 0-4 for that outcome. Caregivers rated emotional, financial, and physical difficulty on a scale 0-5 for each domain, and levels of difficulty rated as 3-5 were categorized as substantial difficulty in each domain (the reference group being caregivers who indicated 0-2 (ie, none, little, or some difficulty).


NHATS participants (or their proxies) reported if they were blind or had difficulty with distance or near vision while using contact lenses or glasses, if normally worn. Participants were classified as having VI if they responded “no” to any of the following questions: “When you use glasses or contacts, do you see well enough to: 1) recognize someone across the street?; 2) watch television across the room?; and 3) read newspaper print?” VI was defined as participant or proxy-reported blindness or difficulty with distance or near vision, as defined in previous analyses.


Sociodemographic and health data for the NHATS participants (age, race/ethnicity, sex, marital status, income, medical comorbidities) and for the NSOC caregivers (age, sex, education, self-reported health status, relationship to the older adult, and cohabiting status) were examined. Comorbidity was defined as the number of self-reported chronic diseases, categorized as 0-1, 2-3, or 4+ conditions and included high blood pressure, arthritis, osteoporosis, lung disease, stroke, heart disease, cancer, depression, probable dementia, and hip fracture. Diabetes was also ascertained by survey-reported physician diagnosis and was entered into models as a separate covariate.

Statistical Analysis

Sociodemographic and health characteristics were summarized across the NHATS participant groups with and without VI. Unweighted frequencies and weighted percentages for categorical variables and weighted means ± standard error (SE) for continuous variables are reported. Logistic regression models were used to assess the association between sociodemographic and health characteristics of older adults with VI and those of older adults without VI. Similarly, NSOC caregiver characteristics were summarized across the NHATS older adult groups (those caring for participants with and without VI). The nature and intensity of care provided by each group in terms of hours of care per month, types of activities for which help was provided, health system interactions, and health management tasks are described. Caregiving-related emotional, physical, and financial difficulty, participation effects in valued activities, and use of supportive services, stratified by the NHATS participant groups with and without VI are also described.

Negative binomial regression models were used to examine how VI was associated with intensity of caregiving (hours of caregiving in the previous month) and the impact on participation in valued activities (number of valued activities affected by caregiving). This method was chosen because outcomes, caregiving hours, and number of valued activities affected by caregiving are count data and had excessive zero counts. In other models, logistic regression was used to determine the odds of experiencing substantial emotional, financial, or physical difficulty related to providing care.

Sensitivity Analysis

Sensitivity analyses were conducted excluding caregivers of NHATS participants who had proxy-reported VI to examine if there was differential reporting of caregiving outcomes by proxy-report versus participant-report of VI (as the proxy could also be the caregiver). Sensitivity analysis was also conducted by type of VI (ie, subgroup analysis for distance and near VI to examine any differential impact on caregiving outcomes).

All multivariate models were adjusted for NHATS participant age, race/ethnicity, sex, marital status, income, comorbidities, diabetes, and NSOC caregiver age, caregiver sex, caregiver education, caregiver’s self-reported health, caregiver relationship to the older adult, and cohabiting status. These covariates were included due to clinical relevance and previously established associations with VI and caregiving needs. All regression models accounted for NSOC’s complex survey design (NSOC-specific survey weights were applied) and clustering by NHATS participants (to account for correlation between multiple caregivers for an NHATS older adult). All tests were 2-sided, and statistical significance was established as a P value less than .05. Analyses were conducted using SAS version 9.4 software (SAS Institute, Cary, North Carolina, USA) and STATA 15 software (StatCorp LLC, College Station, Texas, USA).


Caregiver Characteristics

Among 1,196 community-dwelling NHATS participants, 915 (using NHATS weights, estimated n = 3,305,196; 78.8%) did not have VI, and 281 (estimated n=891,254; 21.2%) had VI. Among 1,776 caregivers, 1,348 (using NSOC weights, estimated n = 11,643,724; 77%) assisted older adults without VI and 428 (estimated n = 3,485,249; 23%) assisted those with VI ( Tables 1 and 2 ). NHATS participants with and without VI had a similar number of caregivers (median: 1; range: 1-5 caregivers per older adult for both groups). Compared to NHATS participants (those receiving care) without VI, participants with VI were older, less likely to be married, and more likely to have a lower education and income ( Table 1 ). Compared to caregivers of older adults without VI, caregivers of older adults with VI were younger, less likely to be their spouse, more likely to be their offspring, and more likely to have been providing care for a duration longer than 4 years ( Table 2 ).


Characteristics of NHATS Participants Linked to NSOC.

Participant Characteristics Total (N = 1,196.0) Not Vision Impaired (n = 915.0; 78.8%) Vision Impaired (n = 281.0; 21.2%) P Value
Mean ± SE age, yrs 78.4 ± 0.23 78.0 ± 0.27 80.1 ± 0.53 .002
Age categories, %, yrs b



Females, % 65.0 65.9 61.9 .293
Race/ethnicity, %
Non-Hispanic white
Non-Hispanic black



Education, %
<High school
High school graduate
>High school



Marital Status, %
Married/living with a partner
Widowed/never married



Income, %



No. of comorbidities, % a



Diabetes, % 36.3 35.9 38.0 .625

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 10, 2021 | Posted by in OPHTHALMOLOGY | Comments Off on Caring for Older Adults With Self-Reported Vision Impairment: Findings from the National Study of Caregiving

Full access? Get Clinical Tree

Get Clinical Tree app for offline access