The prevalence of visual impairment in our study was 5% (95% CI = 3.3%, 7.6%); mostly due to refractive errors. This figure is comparable with other populations. In the Baltimore Pediatric Eye Study, the percentage of preschool-aged children requiring spectacle correction was about 1.2% among White children and 1.8% among African-American children . Jamali et al. reported that about 6.3% of the children entering school in Iran were at risk of amblyopia; mostly due to refractive errors . In Hong Kong, about 4.4% of preschool children had either reduced visual acuity or strabismus . Chia et al. reported a prevalence of 1.19% among Singaporean children aged 30 to 72 months . While in Nepal, Karki reported that 5.97% of children aged 4 to 5 years have amblyopia .
The type of refractive error varies among different populations. Astigmatism was the commonest type of refractive error in our population. The axis of astigmatism in our study population was mostly with-the-rule. Significant refractive errors were uncommon. Hyperopia was found to be the most prevalent refractive error among preschool children in the Baltimore Pediatric Eye Study and in Iran [10, 14]. In Singapore, where the prevalence of myopia is one of the highest in the world, preschool children had a high prevalence of myopia (11%-15%) [15, 16]. The type of refractive errors in a population may change with time. Fan et al. found that in Hong Kong, the commonest type of refractive error shifted from astigmatism to myopia over a decade (1996–2007) .
Strabismus was found in one of the 20 children who did not pass the screening tests. The prevalent type of concomitant strabismus varies in different study population. Esotropia is more common in White population while exotropia in Asians . In the Baltimore Pediatric Eye Study, the prevalence of strabismus was 0.3%; most due to esotropia . Chia et al. reported a 0.80% prevalence of strabismus among preschool Singaporean Chinese. The exotropia-esotropia was about 7:1 .
The rate of testability increases with age . As our sample consisted of older preschool children (aged 4 to 6 years), all 400 children in our study were testable for distant visual acuity and stereopsis. This is comparable with the findings of other studies [18, 19]. This finding will provide a basis for future research to the evaluation of the effectiveness of these (Sheridan Gardiner Test Complete (Keeler, UK) and Langs stereotest) screening tests to detect amblyopia in our population.
The main limitation is due to the inherent weakness of a cross-sectional study. Data regarding the developmental milestone of the children were not collected. The presence of developmental delay may indicate the presence of associated visual impairment .
The non-respondent rate was about 11% and this may result in selection bias. The bias was influenced by caregiver characteristics where children who are already wearing spectacles may or may not be more prone to participate in the study. It is difficult to determine the direction of this bias. However, a higher respondent rate does not necessarily prevent selection bias .
The children were only tested for visual acuity and stereopsis. Comprehensive eye examinations were only offered to those who failed the screening tests. In other prevalence studies, all children were examined [3, 10, 12, 13, 16]. It is therefore difficult to compare our findings because of the statistics. Children with mild strabismus may not be detected as Hirschberg test and cover test was not performed on the field. Therefore strabismus is under-reported in this study. Retesting was not performed in this cross-sectional study. Although pre-test was given prior to actual screening, the possibility of poor cooperation cannot be ruled out.
The study was conducted among the urban population in Kuching district. It is reasonable to assume that the findings should apply to major towns with similar socioeconomic class such as Miri, Sibu or Bintulu. Generalization of the findings to the rural population should be viewed with caution. The prevalence of decreased presenting visual acuity may be higher in rural communities where spectacles are less available.