Affiliations 

  • 1 Universiti Sultan Zainal Abidin, Faculty of Medicine, Kuala Terengganu, Terengganu, Malaysia
  • 2 Universiti Sultan Zainal Abidin, Faculty of Medicine, Kuala Terengganu, Terengganu, Malaysia. abdulmutalib@unisza.edu.my
Med J Malaysia, 2020 11;75(6):685-690.
PMID: 33219178

Abstract

OBJECTIVES: To design and develop a simple vision test algorithm for mobile application and perform a pilot study to determine its validity and reliability as a tool for vision test in the community.

METHODS: A simple visual acuity test algorithm in the form of a single letter E display was designed as the optotype for development of a mobile application. The standardised optotype is presented at random to test visual acuity for corresponding level of 3/60, 6/60, 6/18, and 6/12. The final result is auto-generated based on the classification of the WHO for visual impairment and blindness. The Snellen chart was used as the gold standard to determine its validity while five different users were involved to determine its inter-rater reliability. A pilot study was performed between April till November 2019, in the Universiti Sultan Zainal Abidin Medical Centre (UMC) at Kuala Nerus and Mooris Optometrist Centre at Marang, Terengganu. A total of 279 participants aged four years old and above were involved in this study.

RESULTS: The highest sensitivity was found at the vision level cut-off point of 6/12 with the percentage of 92.7% and 86.8% for the right and left eye, respectively. The specificity was more than 89% for all vision levels in both eyes. The Krippendorff's alpha value for the inter-rater reliability was 0.87 and 0.83.

CONCLUSION: The relatively high level of validity and reliability obtained indicate the feasibility of using the designed optotype to develop a valid and reliable mobile app for vision test. The app can be used to screen vision by nonmedical persons, at anytime and anywhere to help improve public awareness and capability to correctly determine their visual status.

* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.