Methodological quality of the review: Medium confidence
Author: Chen-Wei Pan, Mohamed Dirani, Ching-Yu Cheng, Tien-Yin Wong, and Seang-Mei Saw
Region: Japan, China, Malaysia, Bangladesh, Pakistan, India, Indonesia, Iran, Burmese, Mongolia, Bengal, Korea, Cambodia
Subsector: Epidemiology, prevalence of myopia
Equity focus: No
Study population: Population of any age
Type of programme: Community based
Review type: Other review
Quantitative synthesis method: Meta-analysis
Qualitative synthesis method: Not applicable
Background: Myopia is a complex eye condition that poses major public health and socioeconomic problems globally. Of the leading causes of vision impairment and blindness in the world, uncorrected myopia has been grouped with cataract, macular degeneration, infectious disease and vitamin A deficiency. Most cases of myopia can be treated with conservative treatments; however, more severe stages of myopia may be associated with a range of vision-threatening ocular complications, including agerelated cataract and open-angle glaucoma in older adults that may require surgical interventions. The prevalence of myopia varies with age, and the age-specific prevalence of myopia in Asia is not well-evaluated and summarised. A comprehensive overview of the age-specific prevalence of myopia in Asia is essential to guide future eye health care, intervention and clinical management in this area, and provide a perspective for other continents of the current and future challenges of myopia.
Objectives: The objective of this review was to estimate the age-specific prevalence of myopia in Asia.
Main findings: We identified 50 eligible population-based studies, including 215,672 subjects aged 0 to 96 years, reporting the prevalence of myopia from 16 Asian countries or regions. Myopia was found to be most prevalent (96.5%; 95% confidence interval (CI), 96.3 to 96.8) in Koreans aged 19 years. There was no significant linear age group effect on the prevalence of myopia in the whole Asian population, but there was a U-shaped relationship between both age and year of birth and the prevalence of myopia. The prevalence of myopia was also higher in those older than 70 years (36.3%; 95% CI, 27.6 to 45.0) compared with other age groups, which revealed nuclear cataract-myopia shifts in refraction.
Methodology: Inclusion criteria included: i) population-based surveys in Asia on the populations of any age range with response rates of at least 60% were included in the current review; ii) in addition, we included the studies on military conscripts in some Asian countries where military service is mandatory; iii) conscripts with visual defects or refractive errors are not exempted. This review followed the Meta-analysis of Observational Studies in Epidemiology guidelines for the conduct of systematic reviews and meta-analyses of observational studies. We conducted searches in PubMed (1950 onward), Embase (1960 onward), and Web of Science (1970 onward) electronic databases. Two authors performed the literature search independently. We compared the extracted studies and resolved inconsistencies by consensus. The electronic search was completed in September 2013. Duplicate reports were removed. Reference lists of identified reports were scanned to identify other relevant studies. A comprehensive list of variables was extracted from each eligible study. The quality of the studies included in this review was evaluated using the methodological criteria for prevalence studies developed by Leboeuf-Yde and Lauritsen and Walker. Authors performed the meta-analysis using Stata version 12.0. Heterogeneity was found to be high among the identified studies, and therefore, random-effects models were used. Statistical heterogeneity among studies was evaluated using I2 A meta-regression model was developed with age group and year of birth as the dependent variables to determine the linear age group effects on the prevalence of myopia. Regression models were also set up, with age 2 or year of birth2 (age or year of birth ‘‘squared’’) as an independent variable to test the significance of the possible U-shaped distribution of myopia in different age groups. Publication bias was assessed using the Egger test.
Applicability/external validity: The authors reported that the analysis included a number of limitations that should be considered when interpreting the results. First, we assumed that the contributing studies in the meta-analysis are representative of the general populations in Asia. In addition, it is difficult to compare the prevalence of myopia between Asians and western populations precisely because of the methodological issues attributed to differences in age, definition of myopia, and sampling strategy.
Geographic focus: This review was conducted in some LMIC in Asia and the Middle East; unfortunately, population-based data was not available in many countries.
Summary of quality assessment:
Overall, there is medium confidence in the conclusions about the effects of this study. Although authors used appropriate methods to analyse findings of included studies, methods used to search literature was not comprehensive enough to ensure that all relevant studies were identified. In addition, language bias was not avoided and the search period was not comprehensive enough that relevant literature is unlikely to be omitted.