This secondary analysis examined how differing definitions of visual acuity alter estimates of eye‑care need derived from Rapid Assessment of Avoidable Blindness (RAAB) surveys. RAAB is widely used to estimate prevalence and causes of blindness and vision impairment among adults aged 50 years and older. The authors aimed to quantify differences when using presenting visual acuity (PVA) measured in one or both eyes versus pinhole acuity in the better‑seeing eye.
The analysis used data from 12 RAAB surveys conducted between 2018 and 2023 in six countries: India, Nepal, Pakistan, Kenya, Tanzania and Uganda. In total, 48,185 participants were examined, with an overall response rate of 94.3%. The current study is a secondary analysis of de‑identified individual participant records from those surveys. Summary survey results are publicly available through the RAAB repository (raab.world).
Three person‑level measures were compared:
These three approaches reflect common definitions used in population surveys and program reporting and have different implications for the size and character of identified eye‑care need.
Estimates were adjusted for the two‑stage cluster survey design used in RAAB and were age‑standardised to the WHO world standard population. Multivariable analyses examined associations between impairment and age, sex and survey region. Results are reported with adjusted odds ratios (AORs) and 95% confidence intervals (CIs) where available from the source.
Prevalence estimates varied substantially by the acuity definition used. Across the 12 surveys:
Thus, estimates based on presenting acuity—particularly when counting impairment in either eye—were markedly higher than estimates based on pinhole acuity in the better eye.
Independent associations reported in the analysis included:
Age‑standardisation accentuated regional differences: the crude prevalence ratio between regions was 1.50, which widened to 1.66 after standardisation to the WHO population.
The large gap between presenting and pinhole acuity indicates a considerable addressable burden, much of which is most plausibly attributable to uncorrected refractive error. Using pinhole acuity in the better eye yields substantially lower prevalence estimates than using presenting acuity, and restricting reporting to bilateral impairment in the better‑seeing eye omits many people with unilateral or monocular presenting deficits. The authors note that the higher prevalence observed in South Asian survey locations most likely reflects survey site selection and a greater burden of uncorrected refractive error, rather than an underlying regional epidemiological difference.
These findings highlight that choice of case definition materially affects the measured need for eye‑care services and the implied unmet need for refractive correction or other interventions.
Each of the 12 RAAB surveys included in the analysis received ethical approval from the relevant national or institutional committee before data collection. Participants provided written informed consent. The present study analysed de‑identified participant records under data‑sharing agreements with national programme partners. Summary results for each survey are available via the RAAB repository; de‑identified aggregate data may be available from the corresponding author on reasonable request and subject to approvals.
Limitations reported or implied by the authors include potential survey site selection effects and the observational nature of RAAB data. The source notes that the higher prevalence in South Asian locations likely reflects site selection and uncorrected refractive error rather than a true regional epidemiological difference.
Substantial vision impairment persists across the surveyed districts in both regions studied. Prevalence estimates are highly sensitive to whether presenting or pinhole acuity is used and whether unilateral impairment is counted. The disparity between presenting and pinhole measures underscores a large, potentially treatable burden — especially from uncorrected refractive error — and suggests that restricting estimates to bilateral impairment in the better‑seeing eye underestimates eye‑care need. The reported regional differences should be interpreted in light of site selection and refractive error prevalence rather than assumed to represent intrinsic regional epidemiology.