Association of Perceived Discrimination With Emotional Well-being in Older Adults With Visual Impairment

Key Points

Questions  How prevalent is perceived discrimination among older people with visual impairment and to what extent is it associated with emotional well-being?

Findings  In this study of 7677 older adults (≥50 years of age), 476 (52.1%) of those with poor eyesight reported discrimination vs 2963 (43.8%) of those with good eyesight, a significant difference. People with poor eyesight who reported discrimination were more likely to be depressed, lonely, and dissatisfied with life, and their quality of life was lower than those who did not report discrimination.

Meaning  The findings suggest that older adults with impaired vision are at increased risk of discrimination, which, if experienced, may be associated with poor well-being.

Importance  A significant proportion of individuals with visual impairment report experiences of discrimination. However, evidence comparing perceived discrimination among people with visual impairment with the general population is lacking. In addition, poorer mental health and well-being have been detected in this population, but the association between discrimination and well-being in those with visual impairment is unknown.

Objective  To investigate perceived discrimination among people with visual impairment and its association with well-being in a population-based sample of older adults.

Design, Setting, and Participants  This study collected data from 7677 participants 50 years or older from the English Longitudinal Study of Ageing, a representative sample of older men and women in England. Experiences of perceived discrimination were reported from July 2010 to June 2011. Depressive symptoms, life satisfaction, quality of life, and loneliness were assessed from July 2010 to June 2011 and May 2016 to June 2017. Data analysis was performed from September 27, 2018, to October 10, 2018.

Exposures  Self-rated eyesight, categorized as poor (ratings of fair, poor, or blind) or good (good, very good, or excellent), which was not previously validated for this population.

Main Outcomes and Measures  We used logistic regression to analyze differences in perceived discrimination between participants reporting poor vs good eyesight and cross-sectional and prospective associations between perceived discrimination and well-being in those with poor eyesight.

Results  A total of 7677 participants (mean [SD] age, 66.71 [9.17] years; 4023 [52.4%] female) were included in the study. Participants with poor eyesight had increased odds of reporting perceived discrimination compared with those with good eyesight (odds ratio [OR], 1.41; 95% CI, 1.23-1.63; P < .001). Cross-sectionally, participants who reported poor eyesight and discrimination had increased odds of depressive symptoms (OR, 2.14; 95% CI, 1.57-2.92; P < .001) and loneliness (OR, 2.17; 95% CI, 1.61-2.92; P < .001) and lower quality of life (B = −4.06; 95% CI, −5.29 to −2.84; P < .001) and life satisfaction (B = −2.37; 95% CI, −3.28 to −1.46; P < .001) compared with poor eyesight and no reported discrimination. Prospectively, perceived discrimination was associated with increased risk of depressive symptoms among participants reporting poor eyesight at 6-year follow-up (OR, 1.72; 95% CI, 1.08-2.76; P = .02).

Conclusions and Relevance  These findings suggest that older adults with impaired vision are at increased risk of perceived discrimination. Those who reported experiencing discrimination had higher levels of depressive symptoms and loneliness and lower quality of life and life satisfaction. Action to address discrimination may help mitigate the increased risk of poor well-being in this population.

Visual impairment (defined as presenting visual acuity worse than 6/12¹) is estimated to affect approximately 2 million people in the United Kingdom.² Approximately 10% of those 65 years or older report some sight loss.³ The prevalence of visual impairment is expected to increase because of population aging and increased diabetes.^2-5 This increase poses concern about the financial consequences of sight loss, estimated at £28.1 billion (US $36.7 billion) annually, and the disease burden involved.² People with visual impairment commonly experience difficulties with activities of daily living,³ and most require practical support.⁶

These difficulties may be compounded by discrimination. Charity- and government-commissioned research suggests that discrimination is experienced by a significant proportion of those with visual impairment in the United Kingdom.^6-9 In a survey of more than 1200 individuals with registered visual impairment, 35% reported experiencing negative attitudes from the public and 47% perceived they had been treated unfairly because of their sight loss.⁶ Observational data revealed that those with sight loss were twice as likely as those with another impairment to have experienced discrimination.^7,9

An increasing body of literature suggests that discrimination is a determinant of mental health and well-being.^10-18 For example, a meta-analysis¹¹ of 110 studies found that discrimination was associated with depression, psychiatric distress, and reduced mental well-being. However, the existing evidence is dominated by studies on racism or experiences of discrimination in general.¹⁶

There is evidence that individuals with visual impairment have poorer mental health than their impairment-free counterparts.^6,15 In health care settings, an association between visual impairment and depression has been consistently reported.^19-21 Greater prevalence of depressive symptoms in those with visual impairment has also been reported.^7,9,22,23 In a cross-sectional study of 13 900 older adults in Britain, those with visual impairment were 3 times more likely to experience depression than those with good vision,²³ with similar rates reported in later population-based studies.^7,9,22 Most of the evidence linking depression and visual impairment has been cross-sectional.^19-23 However, recent evidence suggests that vision loss may also be associated with new-onset depression.²⁴

Visual impairment has also been linked with poorer outcomes in other domains of well-being. Heightened levels of loneliness in those with visual impairment compared with those with good sight has been reported in Dutch and US samples.^25-27 However, no significant increase in levels of loneliness was reported in a German study.²⁸ In the United Kingdom, no study has assessed loneliness specifically, but several surveys indicate that social exclusion and reduced social contact may be experienced after sight loss.^6,7,9,29

Associations between visual impairment and lower life satisfaction have been reported,^7,30-32 with UK data suggesting that those with sight loss were 3 times more likely to be dissatisfied than those with no impairment.⁷ A link between visual impairment and reduced quality of life (QOL) has been documented by a review of qualitative studies.³³ Quantitative studies^29,32,34,35 have also found evidence of an association cross-sectionally, and longitudinal evidence suggests that improvements in vision may lead to a corresponding increase in QOL.³⁶

Taken together, several reports^6-9 suggest that many individuals with visual impairment experience discrimination and poorer well-being. However, evidence comparing the prevalence of discrimination among people with visual impairment with that in the general population is lacking. The association between discrimination and well-being in those with visual impairment is unknown. Our study, therefore, investigated this issue in a community sample of older English adults.

Data were obtained from the English Longitudinal Study of Ageing (ELSA), a study of adults 50 years or older.³⁷ Participants are followed up every 2 years. The present analyses were not planned before data collection. The present study uses data from wave 5 (collected July 2010 to June 2011; the only wave that assessed discrimination) and wave 8 (May 2016 to June 2017). Of the 9090 participants interviewed in wave 5, a total of 8107 (89.2%) answered the perceived discrimination questions. We excluded 430 participants (5.3%) with missing data on eyesight or covariates, leaving a final sample of 7677 participants. Data analysis was performed from September 27, 2018, to October 10, 2018. Ethical approval was obtained from the National Research Ethics Service. All participants gave full written informed consent, and all data were deidentified.

The measure of visual impairment was a single-item rating: “Is your eyesight (using glasses or corrective lenses, if you use them) excellent, very good, good, fair, or poor?” Spontaneous responses of those who were legally or registered blind were recorded. We dichotomized responses, defining visual impairment as blindness or fair or poor self-rated eyesight. Participants were also asked, “How good is your eyesight for seeing things up close, like reading ordinary newspaper print (excellent, very good, good, fair, or poor)?” Response options were categorized as above. Hereafter, we refer to fair, poor, or blind eyesight as poor eyesight and good, very good, or excellent eyesight as good eyesight. We coded participants who reported being blind as having poor self-rated eyesight up close. Two participants had missing data for eyesight up close, so we imputed these values with their response to the rating of overall eyesight. To our knowledge, these items have not been validated, although they have been used in other publications of ELSA data.^38,39

Participants also reported whether they had ever been diagnosed with glaucoma, diabetic eye disease, macular degeneration, or cataracts. We considered those who reported any of these as having a history of eye disease.

Items on discrimination were based on measures used in other studies.^40-42 Participants were asked about the frequency of 5 discriminatory experiences: you are treated with less respect or courtesy²; you receive poorer service than other people in restaurants and stores³; people act as if they think you are not clever⁴; you are threatened or harassed⁵; or you receive poorer service or treatment than other people from doctors or hospitals (almost every day, at least once a week, a few times a month, a few times a year, less than once a year, or never). Because data were skewed, with most participants reporting never experiencing discrimination, we dichotomized responses to indicate whether participants had experienced discrimination in the past year (a few times or more a year vs less than once a year or never), with the exception of the fifth item, which was dichotomized to indicate whether respondents had ever experienced discrimination from doctors or hospitals (never vs all other options).⁴³

Our primary outcome for analyses of well-being was depressive symptoms. Secondary outcomes were QOL, life satisfaction, and loneliness.

Depressive symptoms were assessed with an 8-item version of the Center for Epidemiologic Studies Depression Scale,⁴⁴ widely validated for use in older adults.⁴⁵ This scale asks about feelings during the last week (eg, “Over the last week have you felt sad?”), with binary response options (1 indicating yes and 0 indicating no). Positively framed items were reverse scored. Data were dichotomized using an established cutoff, with scores of 4 or higher indicating significant symptoms.⁴⁵

Quality of life was assessed with the CASP-19 (control, autonomy, self-realization, and pleasure),⁴⁶ a scale designed to measure QOL in older people. Respondents are asked how often each statement applies to them (0 indicating often to 3 indicating never). Positively worded items were reverse scored. A higher score indicates higher QOL (range, 0–57).

Life satisfaction was assessed with the Satisfaction With Life Scale,⁴⁷ which asks respondents to rate the extent to which they agree with 5 statements (eg, “In most ways my life is close to my ideal”) on a scale from 0 (strongly disagree) to 6 (strongly agree). Responses were summed to produce a total score (range, 0-30), with higher scores indicating greater life satisfaction.

Loneliness was measured using the 3-item Revised UCLA Loneliness Scale.⁴⁸ An example item was, “How often do you feel you lack companionship?” with response options ranging from hardly ever or never (scored 1) to often (scored 3). Total scores ranged from 3 to 9, with higher scores indicating greater loneliness. They were dichotomized at 6 or higher vs lower than 6 to indicate high vs low loneliness.⁴⁹

Information on age, sex, race/ethnicity (white vs nonwhite), marital status (married vs single, separated, divorced, or widowed) and household nonpension wealth (an indicator of socioeconomic status in this population⁵⁰) was recorded. Objectively measured body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared) (collected in wave 6 because BMI was not assessed in wave 5) was also included in a sensitivity analysis. Trained research nurses measured weight to the nearest 0.1 kg using portable electronic scales and height to the nearest millimeter using a portable stadiometer. Nurses recorded any factors that might have compromised measurement reliability (eg, participant was stooped or unwilling to remove shoes), and these cases were excluded.

All analyses were conducted using SPSS software, version 24 (SPSS Inc). Data were weighted to correct for sampling probabilities and for differential nonresponse and to calibrate back to the 2011 English Census population distributions for age and sex. For cross-sectional analyses, the weights accounted for the differential probability of being included in wave 5 and for nonresponse to the self-completion questionnaire that measured discrimination. For prospective analyses, we applied a longitudinal weight that accounted for nonresponse at wave 8 based on the sample who participated in wave 4.

Our primary outcome was perceived discrimination in any of 5 discriminatory situations. Secondary outcomes were perceptions of discrimination in each of the 5 situations. Logistic regression was used to examine associations between poor eyesight and perceived discrimination, controlling for covariates. In addition, these models were performed again to analyze the odds of reporting perceived discrimination in participants with poor eyesight who reported a history of eye disease and participants with poor eyesight with no history of eye disease compared with the group with good eyesight, adjusting for covariates.

We examined cross-sectional and prospective differences during 6-year follow-up among those who reported poor eyesight in depressive symptoms, QOL, life satisfaction, and loneliness between those who reported perceived discrimination and those who did not. Categorical outcomes were analyzed using logistic regression, and continuous outcomes were analyzed using linear regression. All models controlled for age, sex, race/ethnicity, marital status, and wealth, and prospective analyses also adjusted for baseline status or score on the outcome of interest.

We performed a sensitivity analysis in which models that compared the groups reporting poor vs good eyesight were repeated with additional adjustment for BMI because weight status is known to be associated with conditions associated with visual impairment (eg, diabetes⁵¹) and perceived discrimination.⁵² Body mass index was not included as a covariate in the primary analyses because data were not available in the same wave as discrimination was and we wanted to maximize the sample size. A 1-sided P < .05 was considered to be statistically significant.

A total of 7677 participants (mean [SD] age, 66.71 [9.17] years; 4023 [52.4%] female) were included in the study. Of the 7677 participants in the sample, 913 (11.9%) reported poor overall eyesight and 658 (8.6%) reported poor eyesight up close. Sample characteristics are summarized in Table 1. On average, participants who reported poor eyesight were significantly older than those with good eyesight (70 vs 66 years), and a greater proportion were female, nonwhite, unmarried, and from the lowest wealth quintiles.

Associations between eyesight and discrimination are given in Table 2. Discrimination was more commonly reported by participants with poor eyesight (476 [52.1%]) than those with good eyesight (2963 [43.8%]). After adjustment for covariates, participants with poor eyesight had 1.41 times higher odds of reporting any discrimination than those with good eyesight (95% CI, 1.23-1.63, P < .001). The most common form of discrimination was being treated with less respect or courtesy (331 [36.3%] reported by participants with poor eyesight), and the least common was being threatened or harassed (90 [9.9%]). Poor eyesight was associated with increased odds of reporting discrimination in each of the 5 domains, with odds ratios (ORs) ranging from 1.24 (95% CI, 1.03-1.51, P = .03) for receiving poorer service in restaurants and stores to 1.41 (95% CI, 1.18-1.69; P < .001) for receiving poorer service or treatment in medical settings (Table 2).

As for overall ratings of eyesight, participants who reported poor eyesight for seeing things up close had significantly higher odds of perceived discrimination overall (OR, 1.45; 95% CI, 1.24-1.70, P < .001) and in all 5 domains (Table 2). Reports of being treated as if they were not clever were more common among the group with poor eyesight up close (174 [25.6%]) than in the group reporting poor overall eyesight (210 [23.0%]). Reports of being threatened or harassed were also more common among the group with poor eyesight up close (84 [12.4%]) than in the group reporting poor overall eyesight (90 [9.9%]).

Of the 913 participants who reported poor overall eyesight, 527 (57.7%) reported a history of eye disease. Compared with those with good eyesight, the odds of reporting discrimination in any domain were 1.66 times higher (95% CI, 1.30-2.12; P < .001) among participants with poor eyesight but no diagnosed eye disease and 1.39 times higher among those with poor eyesight who had a history of eye disease (95% CI, 1.11-1.74; P = .004).

Associations between discrimination and well-being in those reporting poor eyesight are summarized in Table 3. Cross-sectionally, participants who reported poor eyesight and discrimination had significantly increased odds of depressive symptoms (OR, 2.14; 95% CI, 1.57-2.92, P < .001) and loneliness (OR, 2.17; 95% CI, 1.61-2.92; P < .001) and lower QOL (B = −4.06; 95% CI, −5.29 to −2.84; P < .001) and life satisfaction (B = −2.37; 95% CI, −3.28 to −1.46; P < .001) compared with poor eyesight and no reported discrimination. Prospectively, discrimination was associated with increased risk of depressive symptoms among participants reporting poor eyesight at follow-up (OR, 1.72; 95% CI, 1.08-2.76; P = .02). Perceived discrimination was not significantly associated with other outcomes prospectively.

In a subsample of 5931 participants whose BMI was measured 2 years after baseline, additional adjustment for BMI revealed no notable differences in associations between eyesight and discrimination or discrimination and well-being among the group with poor eyesight.

In this population-based sample of older adults in England, participants with poor eyesight were 40% more likely to report perceived discrimination compared with those who rated their eyesight as good. Participants who reported poor eyesight and had experienced discrimination were more likely to report depressive symptoms both cross-sectionally and prospectively during 6-year follow-up than those with poor eyesight who did not report discrimination. They were also more likely to be lonely and reported poorer QOL and life satisfaction in cross-sectional analyses, although prospective associations were not significant.

More than half (52.1%) of the sample who reported poor eyesight experienced discrimination in their day-to-day life. More than one-third believed they were treated with less respect than other people: 1 in 4 felt people acted as if they were not clever, 1 in 5 reported receiving poorer treatment than other people from doctors or hospitals, 1 in 6 believed they received poorer service than other people in restaurants and shops, and 1 in 10 reported being threatened or harassed. These findings are in line with a previous report⁶ in which 47% of people with visual impairment reported perceived discrimination, 35% reported negative attitudes from members of the public, and 14% reported unfair treatment from health care workers. We also found that prevalence of perceived discrimination in each domain was significantly higher among people with poor eyesight than in a control group with good eyesight despite relatively high prevalence of reports of discrimination among this control group (43.8%). Results also reveal that the association between poor eyesight and discrimination was not limited to participants with a history of eye disease, with associations at least as strong among participants with poor eyesight but no diagnosed eye disease.

Among participants with poor eyesight, discrimination was associated with significantly poorer well-being. In cross-sectional and prospective analyses, we found that individuals with poor eyesight who had experienced discrimination were twice as likely to have depressive symptoms above threshold. Previous studies^7,9,19-23 have reported an association between visual impairment and depression, with some evidence suggesting that vision loss is associated with the onset of depression.²⁴ Our results demonstrating a longitudinal association between discrimination and increased depressive symptoms among individuals with poor vision point to a potential role of perceived discrimination in the onset and maintenance of depression in this population group.

In addition to having a higher rate of depressive symptoms, people with poor vision who had experienced perceived discrimination were twice as likely to be lonely and reported poorer QOL and life satisfaction. This finding suggests that higher levels of discrimination may contribute to previously reported associations between visual impairment and increased loneliness, social exclusion and reduced social contact,^{6,7,9,25-27,29} lower QOL,^29,32 and life dissatisfaction.^7,30-32 Although these associations were significant in cross-sectional analyses, no prospective associations were observed with the onset of loneliness or decline in QOL or life satisfaction during 6-year follow-up. It seems plausible that the association between discrimination and these outcomes had been established by the time of the baseline assessment; thus, little change was evident over and above the cross-sectional results.

Strengths of this study include the large, representative sample, prospective design, and assessment of well-being using a range of measures. However, there were also limitations. Potential limitations are that poor eyesight was self-reported and that these data have not been compared directly with objective measures. However, similar self-reported ratings of eyesight have been validated.⁵³ Further research is needed to validate our findings using an objective measure of visual acuity. Perceived discrimination was determined by self-reports of past experiences and thus may be subject to recall bias. Recall bias may be a particular concern among participants at the older end of the age spectrum, who are also at increased risk of visual impairment,⁵⁴ which may have led to underestimation of the prevalence of discrimination in the group with poor eyesight. The discrimination questions asked about 5 situations, but there may be others relevant to visual impairment that were not assessed. Prospective analyses were restricted to participants with data at follow-up. Consistent with retention in other studies,⁵⁵ the analyzed sample was younger and wealthier than the total ELSA sample; thus, results may not be representative of the entire population.

The results of this study suggest that older adults with impaired vision are at increased risk of discrimination across a range of domains compared with those with good vision. In addition, those who experience discrimination reported higher levels of depressive symptoms and loneliness and lower QOL and life satisfaction. Action to address discrimination against people with visual impairment may help mitigate the increased risk of poor well-being in this population group. Health care practitioners may consider querying patients with visual impairment about their well-being to identify those who would benefit from additional support.

Accepted for Publication: February 26, 2019.

Corresponding Author: Sarah E. Jackson, PhD, Department of Behavioural Science and Health, University College London, 1-19 Torrington Pl, London WC1E 6BT, United Kingdom (s.e.jackson@ucl.ac.uk).

Published Online: May 30, 2019. doi:10.1001/jamaophthalmol.2019.1230

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

Concept and design: Jackson, Smith.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Jackson, Hackett, Pardhan, Smith.

Critical revision of the manuscript for important intellectual content: Hackett, Pardhan, Smith, Steptoe.

Statistical analysis: Jackson, Pardhan, Smith.

Obtained funding: Jackson, Steptoe.

Administrative, technical, or material support: Hackett, Pardhan, Smith, Steptoe.

Supervision: Pardhan, Steptoe.

Conflict of Interest Disclosures: None reported.

Funding/Support: This research was funded by grant ES/R005990/1 from the Economic and Social Research Council (Drs Jackson and Steptoe).

Role of the Funder/Sponsor: The funding source 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.

Additional Information: The raw data are available from the UK Data Service.