Methodological quality of the review: Low confidence

Author: Lewallen S, Courtright P.

Region: China, India, Malawi, Nepal, South Africa and Saudi Arabia

Sector: Cataract surgical services, cataract surgical coverage rates

Sub-sector: Women

Type of cataract: Age-related cataract

Equity focus: Gender: Review determines cataract surgical coverage rates by sex in a range of developing countries.

Review type: Effectiveness review.

Quantitative synthesis method: Meta-analysis

Qualitative synthesis methods: Not applicable

Background

Women bear approximately two-thirds of the global burden of blindness, with cataract being a major cause in developing countries. There is a higher prevalence of cataract in women than in men and that looking at developed country figures, women will comprise of 60-70% of all cataract surgeries when coverage rates are equal.

Research objectives

To determine, from the existing literature, cataract surgical coverage rates by sex and the proportion of cataract blindness that could be eliminated if women and men had equal access to cataract surgical services.

Main findings

Eight population-based cataract surveys were identified for inclusion in the review, reporting the number of cataract blind and the number having cataract surgery, by sex. Surveys included were conducted in China, India, Malawi, Nepal, South Africa, and Saudi Arabia. Authors reported the following results:

Cataract surgical coverage rates were lower for females than males; the odds ratio for females in the meta-analysis was 0.67 (95% CI: 0.60-0.74); the cataract surgical coverage rate was 1.2 -1.7 times higher for males than for females. Despite women’s lower coverage rate, women ‘accounted for approximately 63% of all cataract cases in the study populations, and if they received surgery the prevalence of cataract blindness would be reduced by a median of 12.5% (range 4-21%)’.

Authors noted that reducing the gender gap in coverage rates could significantly decrease the prevalence of cataract blindness, and recommend that gender sensitive programs are required to improve cataract surgical coverage among women. They noted that researchers could use ‘sight restoration rate’ to measure the cataract coverage rate. Currently cataract surgical coverage rates ‘do not take into account the outcome of the surgery, and most of the studies we analysed reported only whether cataract surgery was done, not whether it successfully restored sight’.

Methodology

Inclusion criteria included methodologically sound population-based prevalence surveys that included (but not limited to) adults; had at least a sample size of 1000 and 80% survey response; published between 1980 and 1999; and written in English, French, Chinese, Spanish or Portuguese.

Authors searched databases including MEDLINE, EMBASE, HEALTHSTAR, and LILACS; and references cited in the retrieved surveys were also reviewed for additional surveys.  It was not clear if authors contacted key experts as part of the search strategy, and also if screening of articles and data extraction were conducted by two reviewers independently.

Authors conducted a meta-analysis including seven out of the eight included studies in this review as this study consisted of a ‘non-standard’ methodology using Peto odds ratio and a fixed effects model to pool data of included studies.

Applicability/external validity

Authors recognized that the data included in the review is limited in that the studies are from a few regions in Asia, with only two African countries included and no Latin American countries. The results were more applicable to low-middle income settings and the developed countries. It should also be noted that reasons for underuse of services by women will vary region to region.

Geographic focus

Please see above.

Quality Assessment

This systematic review was based on a search of relevant databases with no restriction of inclusion based on publication status. However, reviewers did not contact authors/experts of included studies for further information, and language bias was not avoided in the review. There was a data restriction in that only studies published between 1980 and 1999 were included in the review. However, this seemed appropriate since authors were looking for current evidence of cataract surgical coverage rates. It was also not clear from review, if screening of articles or data extraction was conducted by two reviewers independently.

Data synthesis was clearly reported and appropriately conducted. Nevertheless, little is mentioned in the review about the likelihood of bias within the included studies and methodological quality of each study is not fully reported. Therefore, there was low confidence in conclusions about the effects of this study.