Methodological quality of the review: Low confidence

Author: Burgess PI, MacCormick IJC, Harding SP, Bastawrous A, Beare NAV, Garner P.

Region: Egypt, Mauritius, Nigeria, South Africa, Sudan, Cameroon, Seychelles, Ethiopia, Tanzania, Malawi and Botswana

Sector: Diabetic retinopathy, maculopathy.

Sub-sector: None

Equity focus: None specified

Review type: Effectiveness review

Quantitative synthesis method: Narrative analysis

Qualitative synthesis methods: Not applicable

Background

The epidemic rise of diabetes in Africa poses significant public health and socio-economic challenges for the continent and potentially signals a rising prevalence of diseases such as diabetic retinopathy and diabetic maculopathy.

Research objectives

To summarize findings from studies reporting the prevalence and incidence of diabetic retinopathy and diabetic maculopathy in African countries in light of the rising prevalence of diabetes mellitus.

Main findings

The review included 62 studies from 21 countries, all reporting the prevalence or incidence or progression of diabetic retinopathy or diabetic maculopathy in Africa. Study designs included three population-based surveys, two cohort studies, five case-control studies; 32 clinic-based, nine eye clinic-based and eleven other hospital-based surveys.

In population based studies the prevalence range in patients with diabetes for diabetic retinopathy was 30.2% to 31.6%, proliferative diabetic retinopathy 0.9% to 1.3% and any maculopathy 1.2% to 4.5%. In clinic-based surveys, the reported prevalence range for diabetic retinopathy was 7.0% to 62.4%, proliferative diabetic retinopathy 0% to 6.9% and any maculopathy 1.2% to 31.1%. Authors noted that there was no obvious association between prevalence and income level of the country was detected.

The review identified rates of prevalence of diabetic retinopathy in many areas of Africa comparable with high income countries. The prevalence of proliferative diabetic retinopathy and maculopathy was high in recent studies, particularly those from southern and eastern Africa.

Authors concluded that large, community-based cross-sectional and cohort studies were needed to investigate rates and determinants of prevalence of diabetic retinopathy, incidence and progression in Africa. Consensus was needed on the most appropriate methods of identification and classification of retinopathy for research and clinical practice.

 Methodology

Authors searched databases including MEDLINE, EMBASE (OVID), EMBASE Classic and Open-SIGLE (grey literature database) from 1948 to February 2011. No language, publication status or time restrictions were applied to the searches.

Studies were included if reported the prevalence or incidence or progression of diabetic retinopathy or diabetic maculopathy in Africa and included patients with diabetes resident in African countries. Study designs included population-based surveys, cohort studies, case-control studies; clinic-based or hospital-based surveys. Although no time restriction was applied at search stage, studies published before 1990 were excluded.

The main outcome measures included the prevalence of diabetic retinopathy, proliferative diabetic retinopathy and diabetic maculopathy and the incidence of diabetic retinopathy, proliferative diabetic retinopathy and diabetic maculopathy.

Applicability/external validity

It was noted in the review that few high-quality population-based studies were identified for inclusion in the review and the majority of studies included were hospital clinic based surveys of hospital attendees. Authors noted that with this type of study, selection bias was an issue and the findings should be generalized to other settings with caution.

Geographic focus

The review focusses specifically on low- to middle-income countries including: Egypt, Mauritius, Nigeria, South Africa, Sudan, Cameroon, Seychelles, Ethiopia, Tanzania, Malawi and Botswana. Therefore, findings are applicable to low- and middle-income settings.

Quality assessment             

Authors conducted a partially comprehensive search of the literature, although covering relevant databases and reference lists in included studies, language bias was not avoided and authors/experts were not contacted as part of the search strategy.

The studies included were very heterogeneous in terms of patient selection and the classification of retinopathy, therefore a narrative summary was appropriate. Although authors do not report the criteria used to assess the quality of included studies, it was noted that few high-quality population-based studies were identified for inclusion in the review and the majority of studies included were hospital clinic-based surveys of hospital attendees. Authors noted that with this type of study, selection bias was an issue and the findings should be generalized to other settings with caution. As such, a low confidence was attributed in conclusions about the effects of this study.