The development & evaluation of two vision screening tools for correctable visual loss in older people

Abstract

Objectives: In the UK, 20-50% of older people have undetected reduced vision and in most cases this results from correctable problems (refractive error and cataract). Many older people are not availing themselves of state-funded community optometric care. We assessed the efficacy of two vision screening instruments at detecting correctable visual problems and investigated the effect of optometric intervention on quality of life.

Methods: A computerised vision screener (CVS) was developed and refined after a preliminary study of 180 older people to include tests of: monocular presenting distance high contrast and low contrast visual acuities (VAs), binocular near acuities, and monocular visual fields. The modified CVS and a flip-chart vision screener (FVS) were evaluated on a second sample of 200 people aged 65+ (mean age 77 years). All participants in both studies were given an optometric eye examination, including high and low contrast VAs, refraction, binocular vision tests, tonometry, automated perimetry, and dilated fundoscopy including cataract grading and ARM grading. The target conditions were significant gain in monocular distance VA or binocular near VA with new refractive correction, significant cataract, or macular degeneration at risk of rapid progression. The Low Vision Quality of Life Questionnaire (LVQoL) was administered before and up to 3 months after testing.

Results: For the CVS, the best sensitivity (80.3%, 95% CI 72.4-86.4; specificity 66.7%, 95% CI 55.6-76.1) was obtained for a screener test combination of a fail on high contrast VA (>0.19 LogMAR) OR low contrast VA (>0.39 LogMAR) OR near VA (>N11.9). A screener test combination of high contrast VA OR near VA gave sensitivity of 79.5% (71.5-85.7) and specificity 67.9% (57-77.3). For the FVS, the best sensitivity was obtained for a test combination of a fail on high contrast VA OR low contrast VA OR near VA (sensitivity 82%, 95% CI 74.2-87.8; specificity 61.5%, 95% CI 50.4-71.6). A screener test combination of low contrast VA alone gave sensitivity of 75.4% (67.1-82.2) and specificity 76.9% (66.4-84.9). Significant improvements in LVQoL were found, with a significant correlation between gain in VA with new spectacles and improvement in LVQoL.

Conclusions: The vision screeners are effective tools for detecting those with reduced vision. Further work is required to determine their effectiveness as a tool for encouraging older people to engage in regular eyecare.