Methodological quality of the review: High confidence

Author: Ramke J, Petkovic J, Welch V, Blignault I, Gilbert C, Blanchet K, Christensen R, Zwi AB, Tugwell P.

Region: China

Sector: Cataract surgery

Sub-sector: Access to services

Type of cataract: Age-related cataract

Equity focus:  None specified

Review type: Other review

Quantitative synthesis method: Narrative synthesis

Qualitative synthesis method: Not applicable

Background:

Cataract is the leading cause of blindness in low- and middle-income countries (LMICs), and the prevalence is inequitably distributed between and within countries. Interventions have been undertaken to improve cataract surgical services, although the effectiveness of these interventions on promoting equity is not known.

Objectives:

To assess the effects on equity of interventions to improve access to cataract services for populations with cataract blindness (and visual impairment) in LMICs.

Main findings:

Two studies met the reviews’ eligibility criteria, both of which were cluster-RCTs conducted in rural China. Authors attributed unclear risk of bias to each trial. In both studies, villages were randomized to be either an intervention or control group. Adults identified with vision-impairing cataract, following village-based vision and eye health assessment, either received an intervention to increase uptake of cataract surgery (if their village was an intervention group), or to receive ’standard care’ (if their village was a control group).

One study (n = 434), randomly allocated 26 villages or townships to the intervention, which involved watching an informational video and receiving counselling about cataract and cataract surgery, while the control group were advised that they had decreased vision due to cataract and it could be treated, without being shown the video or receiving counselling. Authors report low-certainty evidence that providing information and counselling had no effect on uptake of referral to the hospital (odds ratio (OR) 1.03, 95% confidence interval (CI) 0.63 to 1.67, 1 RCT, 434 participants) and little or no effect on the uptake of surgery (OR 1.11, 95% CI 0.67 to 1.84, 1 RCT, 434 participants). Authors assessed the level of evidence to be of low certainty for both outcomes, due to indirectness of evidence and imprecision of results.

The other study (n = 355, 24 towns randomized) included three intervention arms: free surgery; free surgery plus reimbursement of transport costs; and free surgery plus free transport to and from the hospital. These were compared to the control group, which was reminded to use the “low-cost” (USD 38) surgical service. Authors report low-certainty evidence that surgical fee waiver with/without transport provision or reimbursement increased uptake of surgery (RR 1.94, 95% CI 1.14 to 3.31, 1 RCT, 355 participants). Authors assessed the level of evidence to be of low certainty due to indirectness of evidence and imprecision of results.

Neither of the studies reported authors’ primary outcome of change in prevalence of cataract blindness, or other outcomes such as cataract surgical coverage, surgical outcome, or adverse effects. Neither study disaggregated outcomes by social subgroups to enable review authors further assessment of equity effects. They sought data from both studies and obtained data from one; the information video and counselling intervention did not have a differential effect across the PROGRESS-Plus categories with available data (place of residence, gender, education level, socioeconomic status and social capital).

Methodology: 

Inclusion criteria for this review consisted of: 1) randomized and quasi-randomized controlled trials (RCTs, including controlled clinical trials (CCTs), and cluster-RCTs); controlled before-and-after studies (CBAs); and interrupted time series studies (ITSs) with a clearly defined point in time at which the intervention occurred and at least three data points before and after implementation of the intervention; 2) studies conducted in LMICs; 3) assessed interventions that may improve access to cataract services for those with visual impairments from cataract. Primary outcome measures include change in the prevalence of cataract blindness. Secondary outcome measures include prevalence of cataract visual impairment, service utilization, cataract surgical coverage (CSC), intraocular lens implantation rate, surgical outcome, unintended outcomes/adverse event of the intervention, and any measure of inequity.

Authors searched the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Trials Register) (2017, Issue 3), MEDLINE Ovid, Embase Ovid, LILACS, the ISRCTN registry, ClinicalTrials.gov, and the World Health Organization (WHO) International Clinical Trials Registry Platform. Authors did not use any date or language restrictions in the electronic searches for trials. Authors also searched the reference lists of included studies as part of the search strategy.

Two authors independently selected studies, extracted data and assessed them for risk of bias. Meta-analysis was not possible, so included studies were synthesized in table and text. Authors also assessed heterogeneity narratively.

Applicability/external validity:

Authors note that both included studies were conducted in rural china, and were unable to assess effects in different settings where healthcare systems are different. Both studies applied their interventions universally to rural populations, so there is no evidence on the effect of targeted interventions on access to services.

Geographic focus:

It was not possible for authors to discuss the applicability of the results to LMICs. Please see above.

Summary of quality assessment:

Overall there is high confidence in the conclusions about the effects of this study. Authors used appropriate methods to screen studies for inclusion, extract data and assess the quality of included studies, avoiding biases. Authors appropriately reported limitations of the studies included in the review, making clear which evidence is subject to high or low risk of bias.