Article Text
Abstract
Aim To review the evidence for the presence of lower levels of psychosocial well-being in working-age adults with visual impairment and for interventions to improve such levels of psychosocial well-being.
Methods Systematic review of quantitative studies published in English from 2001 to July 2008 that measured depression/mental health, anxiety, quality of life, social functioning or social support.
Results Included were 29 studies that measured one or more outcomes (N=52). Working-age adults with visual impairment were significantly more likely to report lower levels of mental health (mean difference=14.51/100), social functioning (MD=11.55/100) and quality of life. Studies regarding the prevalence of depressive symptoms produced inconsistent results but had methodological limitations.
Conclusions Future research is required into the prevalence of loneliness, anxiety and depression in adults with visual impairment, and to evaluate the effectiveness of interventions for improving psychosocial well-being such as counselling, peer support and employment programmes.
- Middle aged
- ophthalmology
- psychology
- quality of life
- rehabilitation
- public health
Statistics from Altmetric.com
It is estimated that 80 000 adults in the UK aged 18–64 (16–64 in Scotland) have a visual acuity of <6/18, which represents 8.16% of those eligible for registration as blind or partially sighted.1 Older people are the clear majority of the population with visual impairment, and macular degeneration is the major cause of blindness in Europe and developed countries.2 However, working-age adults live with visual impairment longer; their ability to remain socially included, contribute to the workforce and perform care giving roles may be compromised; and they experience different visual impairments to older people with diabetic retinopathy as the major cause of blindness.3 4 Consequently, working-age adults may present specific needs that require investigation, especially as visual impairment may pose a greater risk to mental health in working-age adults than older people.5
Reviews on the psychosocial impact of visual impairment in older people6–17 demonstrate an elevated risk of depression and reduced mental health, quality of life and social functioning compared to sighted peers (Nyman, Gosney, Victor, submitted for publication). Social support and psychosocial interventions also show promise for buffering against the negative psychosocial impact of visual impairment (Nyman, Gosney, Victor, submitted for publication). However, there is no review of working-age adults, and this paper presents a review of the psychosocial impact of visual impairment in working-age adults and the effectiveness of rehabilitation or psychosocial interventions in alleviating negative psychosocial consequences.
Methods
We electronically searched PsycINFO and Medline for quantitative studies published in English from 2001 to July 2008, which provided data on the psychosocial impact of visual impairment. Inclusion criteria were recruitment of working-age adults (mean age or at least 66% aged 18–59 years), had visual impairment or were part of the supportive network (eg, spouse of someone with visual impairment), and provided statistical data on one or more of seven psychosocial outcome measures: depression/mental health, anxiety, quality of life, social functioning, social support, loneliness or the experience of complex visual hallucinations. Studies were excluded if only the abstract was published and if they recruited participants with a significant comorbidity (eg, hearing loss, brain injury, cancer). Although physical functioning such as the ability to independently carry out tasks of daily living is related to psychosocial well-being, studies that only reported physical functioning outcomes or fused them with psychosocial outcomes were excluded.
We used previous search terms selected from the thesaurus of keywords for the databases (see table 1)8 and retrieved 5393 abstracts (339 Medline, 5054 PsycINFO). Further papers were identified from reference lists and previous literature reviews. Review of the abstracts identified 42 papers that met the inclusion criteria, of which 13 were excluded because they either only recruited participants with correctable visual impairment18–29 or only provided data on scale development.30 For the included papers, study quality was analysed using a modified Downs and Black's score,31 where total scores could range from 0 to 18 for observational studies and from 0 to 25 for experimental studies, with better quality reflected in higher scores.
Results
Twenty-nine studies were reviewed, of which 11 measured one outcome variable and 18 measured either two (n=16) or three (n=2), resulting in 52 separate outcomes for analysis (see table 2). Three studies measured both depression and mental health,32–34 which were only included once in the following analyses (n=49). The context for the outcomes analysed were that 27 842 participants were recruited; mainly from the USA (n=32), with the remaining from Canada (n=3), the UK (n=2) and 12 other countries. Most of the outcomes were from observational (n=46) and cross-sectional (n=45) data. The mean quality score of the observational studies was 13.11 of 18 and ranged from 9 to 16/18. The remaining studies were trials (n=3), each rated with a quality score of 14/25, of which two were randomised controlled trials. Measures of effect size against a control group including ORs are reported, and where these values were not reported, mean differences (MDs) between groups on mean scores or percentage points (−/+%) are provided with an indication of the total points on the scale (eg, M=70.65/100 denotes the average mean score on a scale ranging from 0 to 100). Tables 3–8 are available online as a supplementary file at http://bjo.bmj.com.
Depression and mental health
Twenty studies investigated depression (n=9) and mental health (n=11) in working-age adults with visual impairment (see supplementary table 3).
Depression
Nine studies screened for symptoms of depression using a range of validated scales, but no study used a clinical diagnosis of depression. Four studies found either no significant or a minimal relationship between visual impairment and depressive symptoms.33–36 However, there were studies suggesting a relationship (MD on the Hamilton Depression Rating Scale=+7.78% and MD on the Montogomery–Asberg Depression Rating Scale=+13.13%),37 with depressive symptoms associated with worse diabetic retinopathy in African-Americans with type 1 diabetes (OR=3.41, R2=0.06)38 and poorer visual functioning (MD=12.8/100, r=0.33), but not visual status.32 39 In addition, a study investigated the influence of changes in “world views”; beliefs about susceptibility to harm from events beyond one's control.40 Compared to those with a negative change, those with a positive change in world views reported fewer depressive symptoms.41
Mental health
Ten of 11 studies measured mental health with a version of the National Eye Institute—Visual Function Questionnaire (NEI-VFQ), and those with visual impairment (M=70.65/100, range=48.85–90.00/100),32 34 42–45 reported modestly lower scores than controls (average MD=14.51/100, range=0.62–35.00/100).34 42 44–47 Those with poorer vision reported poorer mental health (average MD=15.03/100, range MD=9.88–19.6/100),34 44 47 and those who reported depressive symptoms reported poorer mental health (MD=9.5/100 and 23.1/100).43 48 The Short-Form survey was less sensitive than the NEI-VFQ in detecting lower mental health in individuals with visual impairment compared to a control group (36-item Short-Form MD=3/100 vs NEI-VFQ MD=35/100).45
Anxiety
Eight studies investigated anxiety (see supplementary table 4). Worry about blindness has been identified in half of a group of patients with glaucoma, but this decreased to 25% at 12 months.49 Reduced visual functioning but not vision status was associated with anxiety in another study.39 However, anxiety was often only elevated to a minimal degree or not at all in individuals with visual impairment.33 35 Indeed, in patients with glaucoma, there was no elevation in levels of anxiety when measured either by a clinical diagnosis or on a validated scale for symptoms.37
Quality of life
Three papers investigated quality of life (see supplementary table 5). Across a range of measures, lower scores of quality of life were found in those with visual impairment,50 such as those with bilateral visual impairment being four times more likely to report “not feeling full of life” (OR=4.63, +9.47%).51 Life satisfaction was higher in those with positive versus negative changes in views towards the world and themselves (self-esteem).41
Social functioning
Thirteen studies investigated social functioning (see supplementary table 6). Nine studies that used the NEI-VFQ found those with visual impairment reported an average mean score of 81.68/100 (51–96/100), which was modestly but significantly lower than controls (average MD=10.32/100, range=0.45–26.40/100, R2=0.29–0.30),32 34 42 43 45–47 and those reporting depressive symptoms had worse scores (MD=7.2/100 to 15.5/100).32 43 For the four remaining studies, a range of scales were used that found either non-significant or minimal differences in social functioning within groups or when compared to controls.36 48 50 52
Social support
Five studies investigated social support (see supplementary table 7). Studies found only minimal differences in reports of receipt of social support as a function of vision status and employment status in those with visual impairment.53–55 Depressive symptoms were significantly inversely predicted by perceived social support from family (β=−0.19) and friends (β=−0.23) (R2=0.12), and positively predicted by perceived overprotection (β=0.31) (R2= 0.08). Although anxiety was not significantly predicted by perceived social support from family and friends, it was positively predicted by perceived overprotection (β=0.21) (R2=0.04).56 Finally, those with retinitis pigmentosa reported significantly worse perceptions of understanding from healthcare staff (MD=15).52
Psychosocial outcomes of interventions
Three studies investigated psychosocial outcomes of rehabilitation and interventions (see supplementary table 8). Rehabilitation has potential for meeting emotional support needs, although this may not always be provided.57 Rehabilitation can also focus life goals on relationships, which are associated with greater satisfaction with life (r=0.21) and lower levels of depressive symptoms (r=−0.27).58 A study observed that those with visual impairment in extended-employment programmes reported greater satisfaction (MD=+47.33%) and empowerment/independence (MD=+49.33%) if they were registered with the programme for ≥8 years compared to those registered for fewer years.59 Further research is required to explain this observation and to test whether any short-term increases in quality of life can be observed.
Discussion
We reviewed quantitative studies published in English from 2001 to 2008 that investigated the psychosocial impact of visual impairment in working-age adults, and the effectiveness of rehabilitation or psychosocial interventions to alleviate this impact. We included 29 studies that measured one of six outcomes (N=52). Working-age adults with visual impairment are likely to report a modest reduction in mental well-being and social functioning, and a reduction in quality of life. Visual impairment was not consistently linked with an increase in depressive symptoms and did not predict a reduction in perceived social support, but low levels of perceived social support predicted depressive symptoms and high levels of perceived overprotection predicted symptoms of depression and anxiety.
Our findings are broadly similar to those of older people with visual impairment, but for this group the link between visual impairment and risk of depression is more robust (Nyman, Gosney, Victor, submitted for publication). Surprisingly, depressive symptoms were not linked with visual impairment in four of nine studies with working-age adults. However, this may be an artefact reflecting issues in study design. One study used an insensitive measure (a single item from the Short-Form survey)34 and three studies used samples that may have been particularly less likely to report depressive symptoms given that they could either read written text or Braille,35 were newly diagnosed with glaucoma and therefore may not have had time to become depressed,36 or were recruited into the control arm of an randomised controlled trial that may have inspired hope in treatment to restore vision.33 Similar to research with older people, (Nyman, Gosney, Victor, submitted for publication) few studies have evaluated interventions to improve psychosocial well-being in working-age adults with visual impairment, including the potential for employment programmes to improve quality of life.
Challenges
The studies reviewed were limited in that most were cross-sectional observations and so cannot inform of causal relationships or about the longitudinal natural history of psychosocial well-being. Most studies did not recruit matched control groups for comparison but used previously published normative data that may be confounded by differences in demographic characteristics and the environment and method in which the samples were recruited. In addition, as we included studies that recruited adults aged 60 and above as a small proportion of their sample, the prevalence of low psychosocial well-being in working-age adults may be inflated by potentially lower well-being in older people. There was heterogeneity in the measures used, limiting comparisons between studies, and validated scales were not always used and so the use of insensitive measures may explain why some studies did not find significant relationships.
Future research
There was heterogeneity in how visual impairment was defined, which complicates our synthesis of previous findings. Uniformity in defining blindness (<3/60 in the better eye with correction) and low vision (<6/18) in accordance with the World Health Organization60 would facilitate the interpretation of different studies and be conducive for future meta-analyses in this area. Further research is required to clarify if depression and anxiety are more prevalent in working-age adults with visual impairment, with the use of both clinical diagnoses and validated screening tools for symptoms. For this, reduced visual functioning rather than visual impairment per se may well be more predictive of increased levels of depression and anxiety, as suggested by a study on anxiety in this review39 and research with older people (Nyman, Gosney, Victor, submitted for publication). Future research on mental health and social functioning are likely to find vision-specific measures such as the NEI-VFQ much more sensitive than generic measures of quality of life such as the Short-Form survey, as indicated by a study on mental health in this review45 and research with older people (Nyman, Gosney, Victor, submitted for publication).
Research could also explore whether working-age adults report higher levels of perceived social support than older people, using measures of perceived adequacy of social support from family and friends as measured with older people.61 The prevalence of loneliness in working-age adults with visual impairment also remains unexplored. Much more work is required on the development and evaluation of interventions to improve psychosocial well-being in working-age adults with visual impairment. In addition, the inter-relationships of variables require investigation along with causal pathways including the role of personality.
Conclusion
Working-age adults with visual impairment report lower levels of psychosocial well-being than their sighted peers on measures of mental health, social functioning and quality of life. There was no evidence for the social support network to be reduced and mixed results for the prevalence of depression. Future research is required to document the prevalence of loneliness, clarify the prevalence of anxiety and depression, and test the effectiveness of interventions to improve psychosocial well-being directly through counselling/peer support or indirectly through employment programmes using validated and appropriate tools.
Acknowledgments
We thank Dr Karen Burnell for her assistance with preliminary work that was submitted to Thomas Pocklington Trust upon which this article is based and who declined an invitation to contribute to this paper. We thank the advisory group for their comments on a summary presentation of the initial findings of this review: Alison Hood (Guide Dogs), Iain Hopkin (Thomas Pocklington Trust R&D Committee), Martin Leyland (Royal Berkshire NHS Foundation Trust, West Berkshire Community Hospital, and Oxford Eye Hospital), Angela McCullagh (Thomas Pocklington Trust), Mary Norowzian (RNIB), and Femi Nzegwu (Guide Dogs).
References
Supplementary materials
Web Only Data bjo.2009.164814
Files in this Data Supplement:
Footnotes
Funding Thomas Pocklington Trust, 5 Castle Row, Horticultural Place, Chiswick, London W4 4JQ.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- At a glance