The economics of vision impairment and its leading causes: a systematic review

Author: Marques AP, Ramke J, Cairns J, et al.

Geographical coverage: Worldwide   

Sector: Impact/economic evaluation

Sub-sector: Cost

Equity focus: None

Study population: Adults aged 15+ years

Review type: Other review

Quantitative synthesis method: Narrative review

Qualitative synthesis method: Not applicable

Background: Vision impairment can have wide-ranging economic impacts on individuals, households and health systems. A systematic review published in 2013 identified 22 studies that reported costs associated with VI from the main causes of VI in high income countries.

Objectives: To describe and summarise the costs associated with VI and its major causes – additionally drawing upon data from low- and middle-income (LMICs) and taking account of new studies and developments in eye treatment that have occurred since the 2013 review.

Main findings:

A total of 138 studies met criteria and were included in this analysis.

The review included 138 studies providing 147 estimates across seven Global Burden of Disease (GBD) super-regions and 155 estimates across 21 GBD regions. About 10% of estimates presented global data, with two-thirds from the high-income super-region. Western Europe (28%) and North America (23%) were the most common regional estimates. Eight regions had no estimates. The studies primarily focused on cost of illness (61%) or cost analysis (20%), with 17 cost-effectiveness studies (12%). Most studies (90%) used a prevalence-based approach to estimate costs. The cost estimation perspective was not stated in 38% of studies, with 35% using a societal perspective, and roughly equal proportions using a healthcare system (18%) or payer (17%) perspective. The average quality assessment score across studies was 78% (median 80%, range 35-100%). The aspects where studies most frequently met quality standards were in providing a complete or partially adequate description of the methods and appropriately presenting and discussing the results. The areas where studies fell short were in reporting sensitivity analysis and breaking down costs.

Authors found substantial variation across studies in average treatment costs per patient for most conditions, including refractive error correction (range $12−$201 ppp), cataract surgery (range $54−$3654 ppp), glaucoma (range $351−$1354 ppp) and AMD (range $2209−$7524 ppp).

To improve future cost estimates, authors recommend relevant stakeholders develop a “reference case” for eye health – a reference document of costing methods based on well-defined principles that can support better decision-making through standards for planning, conducting and reporting and enable more robust and consistent decisions over time.

Regular cost reports that ensure comparability and facilitate trend analysis over time, between settings and between other conditions, will be worthwhile once methods and reporting have been standardised. ‘Reference cases’ can be used for future studies in Africa, for example, to encourage a standard approach.

Methodology:

The study incorporated interventions that reported services for cataract and refractive error, irrespective of Visual Impairment (VI), as these are usually one-time interventions. Studies that reported costs for screening or treatment services for other VI causes but did not report costs for individuals with VI were excluded. The included studies reported outcomes among individuals with VI such as direct and indirect costs, productivity losses, informal care, intangible costs, transfer payments, and deadweight losses. Studies that only reported incremental costs, net costs, incremental benefits, or net benefits without also reporting actual costs were excluded. Regarding the study design, partial economic evaluation studies like cost of illness studies, burden of illness/diseases, and full economic evaluation studies like cost-effectiveness and cost-benefit studies were included. Model-based economic evaluation studies that did not report any costs, primary data, or were based on reviews of existing studies were excluded. No language or geographical restriction applied. To ensure contemporary estimates were identified, a search was restricted to papers published between 1 January 2000 and the search dates.

literature search was conducted in Medline and the Centre for Reviews and Dissemination database, which includes the NHS Economic Evaluation Database, the Database of Abstracts of Reviews of Effects, and the Health Technology Assessment database. The initial search was performed in November and December 2019, and repeated in January 2022 to identify newly published studies. References of all included studies were reviewed for additional relevant studies. The list of included studies was given to field experts, including health economists and eye care researchers, to identify further relevant studies and reports in the grey literature.

Full-texts were assessed by two investigators independently to establish eligibility for inclusion into the study. One author extracted all relevant data which was then verified by a second author.

Due to the absence of formal international guidelines for assessing the quality of economic studies, the authors used the British Medical Journal Checklist for economic submissions, adapted for cost of illness studies. Two investigators independently appraised all included studies. The assessment criteria included the definition of the disease, description of the epidemiological approach, cost disaggregation, data source description and assessment, method explanation, study perspective indication, resource utilization description, unit cost valuation explanation, results presentation and discussion, and sensitivity analysis performance. Each criterion was rated as yes (1 point), partial (0.5 points), no (zero points), or not applicable (zero points, and the item was removed from the denominator). A global score was calculated for each study as the total points allocated as a proportion of the total applicable points. Each checklist item was assigned equal weight. No study was excluded based on its quality score.

The included studies were divided into two groups: those reporting costs for people with general Visual Impairment (VI) and those reporting costs for people with one of the seven specified causes of vision loss. The studies were characterized based on various factors such as country of study, study design, perspective of analysis, epidemiological approach, type of reported costs, level of reporting, methods of resource quantification, and methods to handle uncertainty. Costs reported for any year prior to 2018 were inflated to 2018 values using a country-specific gross domestic product deflator and then converted to USD purchasing power parities to equalize the purchasing power of different currencies. All cost categories were compared against a standard framework and recategorized where indicated to increase comparability between studies. Costs were synthesized in a structured narrative way using seven summary tables and reported in four categories: direct costs, productivity costs, informal care costs, and intangible costs. The unit of observation was reported for average costs and included costs per episode of care, costs per patient, costs per surgery, and costs per spectacles. Direct costs were classified as direct medical or direct non-medical costs, and costs of productivity losses as morbidity-related or mortality-related productivity losses. Intangible costs were reported in non-monetary measures due to the controversy over objective monetary valuation of intangible costs.

Applicability/external validity: Authors indicate a high level of heterogeneity between studies in terms of methodology, due to the inclusive approach taken to review, which might reduce the comparability of their cost estimates.

Geographic focus: Authors intentionally aimed to include studies from LMICs in their review, which were absent from previous iterations. However, a bulk of data relates to high income countries, and cost estimates are not possible for some specific sub-regions which primarily contain LMICs.

Summary of quality assessment:

The approaches to the identification, inclusion and critical appraisal of studies were highly robust, with the key tasks being undertaken by at least two authors and a search that was inclusive in terms of language and publication status. While the approaches to the analysis of the data were generally robust, there was no specific information provided in relation to the results of individual studies. The authors also did not present separate analysis based on the different levels of quality assigned to individual studies. As the authors note, the inclusive nature of their review led to the included studies being highly heterogeneous in terms of their methodologies, which may reduce the comparability between their results. For these reasons, a medium confidence in the findings of this review was attributed.