Aim To investigate the prevalence of, and demographic associations with, uncorrected refractive error (URE) in an older British population.

Methods Data from 4428 participants, aged 48–89 years, who attended an eye examination in the third health check of the European Prospective Investigation into Cancer-Norfolk study and had also undergone an ophthalmic examination were assessed. URE was defined as ≥1 line improvement of visual acuity with pinhole-correction in the better eye in participants with LogMar presenting visual acuity (PVA) <0.3 (PVA <6/12). Refractive error was measured using an autorefractor without cycloplegia. Myopia was defined as spherical equivalent ≤−0.5 dioptre, and hypermetropia ≥0.5 dioptre.

Results Adjusted to the 2010 midyear British population, the prevalence of URE in this Norfolk population was 1.9% (95% CI 0.6% to 3.1%). Lower self-rated distance vision was correlated with higher prevalence of URE (p_trend<0.001). In a multivariate logistic regression model adjusting for age, gender, retirement status, educational level and social class, independent significant associations with URE were increasing age (p_trend<0.001) and having hypermetropic or myopic refractive error. Wearing distance spectacles was inversely associated with URE (OR 0.34, 95% CI 0.21 to 0.55, p<0.001). There were 3063 people (69.2%) who wore spectacles/contact lenses for distance vision. Spectacle wear differed according to type of refractive error (p<0.001), and use rose with increasing severity of refractive error (p_trend<0.001).

Conclusion Although refractive error is common, the prevalence of URE was found to be low in this population reflecting a low prevalence of PVA<0.3.