Elsevier

Ophthalmology

Volume 118, Issue 8, August 2011, Pages 1495-1500
Ophthalmology

Original article
Prevalence and Risk Factors for Visual Impairment in Preschool Children: The Sydney Paediatric Eye Disease Study

https://doi.org/10.1016/j.ophtha.2011.01.027Get rights and content

Purpose

To assess the prevalence and associations of visual impairment (VI) in preschool children.

Design

Cross-sectional, population-based study.

Participants

A total of 2461 children (73.8% participation rate), aged 6 to 72 months, were examined in the Sydney Paediatric Eye Disease Study during 2007–2009; of whom 1188, aged 30 to 72 months, with complete visual acuity (VA) data in both eyes, were included in this report.

Methods

Measurement of VA was attempted on all children using the Electronic Visual Acuity (EVA) system or a logarithm of the minimum angle of resolution (logMAR) chart. Visual impairment was defined as presenting VA <20/40 in children aged ≥48 months and <20/50 in those aged <48 months. Post-cycloplegic refraction was measured, and myopia was defined as spherical equivalent (SE) ≤−0.50 diopters (D), hyperopia was defined as SE ≥2.00 D, astigmatism was defined as cylinder ≥1.00 D, and anisometropia was defined as SE difference ≥1.00 D between 2 eyes. Ethnicity, birth parameters, and sociodemographic information were collected in questionnaires completed by parents.

Main Outcome Measures

Visual impairment prevalence and its associations with child demographic factors and birth parameters.

Results

Visual impairment was found in 6.4% of the worse eye and 2.7% of the better eye in our sample. Refractive errors (69.7%) and amblyopia (26.3%) were the principal causes of VI in the worse eye. Astigmatism (51.3%) and hyperopia (28.9%) were the main refractive errors causing VI. In regression analysis controlling for other factors, VI was independently associated with low birthweight of <2500 g (odds ratio 2.4, 95% confidence interval, 1.1–5.3), but not with age, gender, ethnicity, or measures of socioeconomic status (P > 0.05).

Conclusions

Visual impairment in at least 1 eye was found in 6.4% of Australian preschool children, with bilateral VI found in 2.7%. Uncorrected refractive errors and amblyopia were the principal ocular conditions associated with VI. Low birthweight was a significant risk factor independent of age, gender, and ethnicity.

Financial Disclosure(s)

The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Section snippets

Participants

The Sydney Paediatric Eye Disease Study (SPEDS) identified 3333 eligible children, aged between 6 and 72 months, from a door-to-door census using the 2006 Australian Bureau of Statistics data in 4 selected postcodes in metropolitan Sydney, Australia. The postcodes were selected using random cluster sampling, stratified by socioeconomic status, to represent each of the 3 major strata in Sydney (inner city, suburban, and outer suburban). The recruitment process involved leaving information

Visual Impairment

Visual impairment was defined as the best presenting VA <20/50 (Snellen equivalent 6/15, logMAR score 0.4) in children aged <48 months and <20/40 (Snellen equivalent 6/12, logMAR score 0.3) in children aged ≥48 months.12 The better eye was defined as the eye with better VA, and the worse eye was defined as the eye with the worse VA. Thus, all children with VI in their better eye (n = 32) had VI present in both eyes.

Refractive Errors

Myopia was defined as SE (sphere + ½ cylinder) ≤−0.50 diopters (D), hyperopia

Results

After excluding children with incomplete VA data, 1188 children were included in this report. The mean age was 52.0 months, with similar proportions of girls (48.7%) and boys (51.3%). Caucasian was the dominant ethnicity (47.1%), followed by East Asian (21.4%), South Asian (13.5%), other ethnicities/mixed ethnicity (9.8%), and Middle Eastern (8.2%). Approximately half of the tested children had at least 1 parent with a university degree (51.8%), 83.1% of children had at least 1 parent working

Discussion

By using essentially identical definitions, the MEPEDS12 reported a worse eye presenting VI prevalence of 5.9% in African-American and 7.1% in Hispanic preschool children, and a better eye VI prevalence of 2.8% in African-American and 2.5% in Hispanic children. The BPEDS reported similar VI prevalence in 1 or both eyes in 3.5% of white children and in 4.4% of black children.11 Our results are in keeping with the findings from these 2 studies. However, the Beijing Study of Visual Impairment in

References (25)

  • L. Xu et al.

    Causes of blindness and visual impairment in urban and rural areas in Beijing: the Beijing Eye Study

    Ophthalmology

    (2006)
  • J.J. Wang et al.

    Age-specific prevalence and causes of bilateral and unilateral visual impairment in older Australians: the Blue Mountains Eye Study

    Clin Experiment Ophthalmol

    (2000)
  • Cited by (66)

    • Choroidal nevi in children: prevalence, age of onset, and progression

      2021, Journal of AAPOS
      Citation Excerpt :

      Prevalence was calculated as a cross sectional value and plotted against the corresponding mean age of the study subjects. The pooled study sample of 5,533 unique children (7,059 examinations) was derived from a subset of various studies consisting of 1,085 (32.5%) young children from SPEDS (6 months to 6 years of age; mean, 4.3),20,21 1,756 from SMS6 (6 years of age; mean, 6.7),22 and 2307 from SMS12 (12 years of age; mean, 12.7)22 cohorts for whom good quality fundus photographs were available for assessment. The 6-year longitudinal follow-up study, SAVES, had follow-up on 1,526 (37.9%), with 516 (29.3%) from the younger cohort and 1,010 (43.7%) from the older cohort; 385 children (18 years of age; mean, 17.2) were also included from the older cohort of SAVES (Table 1).22

    View all citing articles on Scopus

    Manuscript no. 2010-994.

    Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

    The SPEDS was supported by the Australian National Health and Medical Research Council Grant 402425, and the Westmead Millennium Institute, University of Sydney.

    View full text