Methodological quality of the review: High confidence
Author: Smeeth L, Iliffe S.
Region: United Kingdom (UK), Netherlands and United States (USA).
Sector: Visual impairment
Equity focus: Elderly (aged 65 years old and above)
Review type: Effectiveness review
Quantitative synthesis method: Meta-analysis
Qualitative synthesis methods: Not applicable
Studies over the past years have demonstrated high levels of visual impairment among older population. There seems to be a variety of adverse factors associated with visual impairment including reduced functional status, social interaction and quality of life. Much of the impairment could potentially be improved by treatment. The UK originally developed multicomponent assessment of older people, which has been introduced in many countries. While the aims of multicomponent screening of older people are broad, any benefit arising from the inclusion of a vision component in the assessment will necessarily be dependent on improving vision.
To assess the effects on vision of mass screening of older people for visual impairment
The authors included five randomized trials of multicomponent assessment from the UK, USA and Netherlands. Overall, trials were adequately conducted, the results of all trials in which individuals were randomized were very similar. There was no evidence of heterogeneity between trials.
Visual outcome data were available for 3,494 people in five trials of multicomponent assessment. Length of follow-up ranged from two to four years. All the trials used self-reported measures for visual impairment, both as screening tools and as outcome measures. In four of the trials people reporting visual problems were referred to either eye services or a physician. In one trial people reporting visual problems received information about resources in the community designed to assist those with poor vision. The proportions of participants in the intervention and control groups who reported visual problems at the time of outcome assessment were 0.26 and 0.23 respectively (risk ratio for visual impairment 1.03, 95% confidence interval (CI) 0.92 to 1.15). Visual outcome data were also available for 1,807 people aged 75 years and over in a cluster randomized trial in which physicians’ general practices were randomized to two different screening strategies; universal or targeted. Three to five years after screening, the risk ratio for visual acuity less than 6/18 in either eye comparing universal with targeted screening was 1.07 (95% CI 0.84 to 1.36, P = 0.58). The mean composite score of the National Eye Institute 25 item visual function questionnaire was 85.6 in the targeted screening group and 86.0 in the universal group, a difference of 0.4 (95% CI -1.7 to 2.5, P = 0.69).
Overall, the authors concluded that ‘there is no evidence that community-based screening of asymptomatic older people results in improvements in vision.’ They also noted that, given the importance of visual impairment among older people, further research into strategies to improve vision of older people is needed.
The authors included all randomized trials of visual screening alone or as part of the multicomponent screening in unselected people aged 65 years or over in a community setting. The outcome included was the level of visual impairment in the population at the end of the trial. Assessment of vision by any methods at least six months after the initial vision screening assessment was included.
The authors conducted a search of several databases including Central, Medline, Embase, PubMed and UK clinical Trials Gateway. No date or language restrictions were applied. The authors also scanned the reference lists of included articles and contacted named authors to obtain information about any other trials. Two reviewers independently assessed the titles and abstracts, and extracted data of included studies. Trials were assessed in terms of allocation concealment, attrition bias, intention-to-treat analysis and masking of outcome assessment, grading each parameter A (adequate), B (not clear) and C (inadequate).
The authors combined data from included studies to produce a summary risk ratio using the fixed-effect Mantel-Haenszel method. Additionally, authors also assessment the amount of between study heterogeneity using the I2 statistic and tested for heterogeneity between trials using a standard chi-squared test.
Review authors did not address the applicability/external validity of the results.
Although no restrictions were applied to specific income settings, data included from reviews were applicable to high-income settings only, as no data from low- and middle- income countries were identified to be included in the review.
There is a high confidence in the conclusions about the effects of this review. Search for literature was sufficiently thorough to ensure the inclusion of all relevant studies. The authors used appropriate methods to select studies and extract data of included studies, avoiding risk of bias. Sensible criteria were used to assess risk of bias of studies included. The authors appropriately combined data of included studies without reporting strong policy conclusions.