Task-shifting in eye health: Assessing the accuracy and feasibility of community worker-led vision screening. MEDiS report # 2

Dear Editor, Vision is vital for daily functioning and quality of life.1 In India, nearly 8 million people are blind, and 62 million are visually impaired.2 Urban slums reflect the intersection of poverty, limited healthcare, and preventable vision loss. With a shortage of optometrists and dependence on ophthalmologists,3,4 alternative workforce models are urgently needed. We recruited and trained 14 women from Mumbai slums as Allied Ophthalmic Personnel5 (“Sakhis”) to conduct door-to-door vision screening using a standardized protocol. This study compared their performance with qualified optometrists in detecting reduced vision in community settings. Sakhis received 1 month of structured training, with emphasis on accurate measurement of distance and near visual acuity (VA). Distance VA was tested using the validated Peek Acuity smartphone app (Peek Vision Ltd., London, UK)6 and near VA with the Shakuntla near vision chart (India). From a list of participants screened by Sakhis, a random subset was re-examined by a certified optometrist in the community itself, whose findings served as the reference standard. Participants were classified as ‘Fail’ if either eye had distance VA <6/12 or near VA <N6 (or <6/6 with correction) and as ‘Pass’ if both eyes met or exceeded these thresholds. The same criteria were applied to Sakhi results. Sakhi outcomes were compared against optometrist evaluations. Diagnostic metrics, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), were calculated using Pass/Fail classification. After converting the VA to logMAR values, a comparative analysis using t-tests was performed for right- and left-eye distance and near VA measurements. Agreement between one expert rater and each trainee rater was evaluated using the intraclass correlation coefficient. A total of 444 participant records validated by optometrists were included in the analysis. Sakhis demonstrated high sensitivity (89%) and specificity (90%) in detecting reduced vision, indicating reliable identification of individuals who require referral. The PPV of 88% suggests that nearly 9 of 10 referrals were accurate, while the NPV of 92% indicates strong confidence that no referrable individuals were missed Table 1.Table 1: Vision screening performance of Sakhi compared to optometrist evaluationPaired t-tests were performed to compare distance and near VA measurements (in logMAR) between Sakhis and optometrists Table 2. A statistically significant difference was found only for near VA in the left eye (P = 0.012), possibly due to testing bias (the right eye was tested first), participant fatigue, or optometrists’ greater precision in detecting subtle refractive errors. As this was limited to one parameter, a larger sample may clarify whether it reflects a true difference or is due to chance.Table 2: Comparison of VA measurements between Sakhis and optometrists and intraclass correlation coefficient (ICC) for distance and near visionInter-rater reliability for vision assessment was high, with ICC consistently above 0.75, indicating strong consistency between the two raters. Sakhis showed excellent agreement with optometrists, with high sensitivity and specificity and minimal interobserver variability. The strong predictive values highlight the accuracy and reliability of their screening. Minor discrepancies in near-vision testing of the second eye can be reduced through standardized training and periodic reinforcement. Given a good concordance between optometrists and allied ophthalmic personnel (Sakhis) trained for a specific task, it is reasonable to deploy this group of the workforce for community-level eye screening. Financial support and sponsorship: Shantilal Shanghvi Foundation (SSF). Conflicts of interest: There are no conflicts of interest.