key: cord-0264983-gutho883
authors: Yong, A. C.; Chuluunkhuu, C.; Chan, V. F.; Stephan, T.; Congdon, N.; O'Neill, C.
title: Cost-Benefit Analysis of a Children's Spectacle Reimbursement Scheme: Evidence for Including Children's Spectacles in Mongolia's Social Health Insurance
date: 2021-08-12
journal: nan
DOI: 10.1101/2021.08.12.21261541
sha: f8f28b9a90457b6cf74039c4967eb5cad55f34c5
doc_id: 264983
cord_uid: gutho883

Background and aim: Globally, 12.8 million children have vision impairment due to uncorrected refractive error (URE). In Mongolia, one in five children needs but do not have access to spectacles. This cost-benefit analysis aims to estimate the net benefits of a children's spectacles reimbursement scheme in Mongolia. Methods: A willingness-to-pay (WTP) survey using the contingent valuation method was administered to rural and urban Mongolia respondents. The survey assessed WTP in additional annual taxes for any child with refractive error to be provided government-subsidised spectacles. Net benefits were then calculated based on mean WTP (i.e. benefit) and cost of spectacles. Results: The survey recruited 50 respondents (mean age 40.2 +/- 9.86 years; 78.0% women; 100% response rate) from rural and urban Mongolia. Mean WTP was US$24.00 +/- 5.15 (95% CI US$22.55 to 25.35). The average cost of a pair of spectacles in Mongolia is US$15.00. Subtracting the average cost of spectacles from mean WTP yielded a mean positive net benefit of US$9.00. Conclusion: A spectacle reimbursement scheme is a potentially cost-effective intervention to address childhood vision impairment due to URE in Mongolia. These preliminary findings support the inclusion of children's spectacles into existing Social Health Insurance. A much larger random sample could be employed in future research to increase the precision and generalisability of findings.

The survey recruited 50 respondents (mean age 40.2 ± 9.86 years; 78.0% women; 100% response rate) from rural and urban Mongolia. Mean WTP was US$24.00 ± 5.15 (95% CI US$22.55 to 25.35) . The average cost of a pair of spectacles in Mongolia is US$15.00.

Subtracting the average cost of spectacles from mean WTP yielded a mean positive net benefit of US$9.00.

A spectacle reimbursement scheme is a potentially cost-effective intervention to address childhood vision impairment due to URE in Mongolia. These preliminary findings support the inclusion of children's spectacles into existing Social Health Insurance. A much larger random sample could be employed in future research to increase the precision and generalisability of findings.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

Uncorrected refractive error (URE) accounts for 61% of the total global burden of vision impairment (VI), [1] and affects 12.8 million children worldwide. [2] Despite the existence of evidence-based and cost-effective strategies, limited affordability and access to high-quality refractive services remain the major barriers to better spectacles coverage in low-and middleincome countries. [3] [4] [5] Delivery of spectacles is shown to improve cognitive development, educational achievement, work productivity, and psychosocial well-being. [6] [7] [8] [9] [10] The annual global economic loss caused by VI due to URE is estimated to be US$202 billion. [11] Mongolia, the least densely populated nation on earth (2 people/km 2 ), is a landlocked country in North-Central Asia. [12] The Mongolian government provides citizens universal access to health care services covered under Social Health Insurance (SHI), [13] and children receive free at point of use health and dental care. [14, 15] Ready-made spectacles are offered with lenses of the same spherical prescription in both eyes, while custom spectacles can be offered with combination of prescriptions to correct any magnitude of refractive errors (RE). Studies suggest that low cost, ready-made spectacles are effective at correcting RE, without compromising spectacle wear while reducing costs and solving logistical challenges of school-based refractive service programmes. [18, 19] The eligibility of children for ready-made spectacles ranges from 70-83% in China, India, and Cambodia. [18, 20, 21] According to an unpublished paper, 51.4% of Mongolian children are eligible for ready-made spectacles (OneSight, 2021).

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint To inform policymakers of the potential benefits of such a scheme, a cost-benefit analysis (CBA) is preferable over other health economic analyses because CBA reports outcomes in monetary terms, which are easily presented to decision-makers. [22] Langabeer et al.'s CBA on telemedicine demonstrates potential annual savings of US$928,000 in Houston, United States, when compared to traditional emergency medical services. [23] A willingness-to-pay (WTP) survey is one way in which preferences can be elicited for use in CBA that can hypothetically estimate an intervention's benefits.

[24] It assesses how much a target population is willing to pay for an intervention. WTP has been used to estimate the potential value of a proposed spectacle delivery scheme in rural Cambodia. [25] Despite a growing number of economic evaluation studies on URE programmes, few studies are on children. This CBA is designed to assist eye health non-governmental organisation (NGO)

Orbis International in informing Mongolian policymakers of the potential benefits of a proposed child's spectacle reimbursement scheme.

This study was approved by the Faculty of Medicine, Health and Life Sciences Research Ethics Committee, Queen's University Belfast (reference number MLHS 20_73). Verbal consent was obtained from each respondent upon agreeing to participate in the survey.

A WTP survey of rural and urban Mongolia was used to estimate the benefits of providing spectacles to any children with RE. Those estimated benefits were used in the CBA.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint

The Mongolian SHI is largely funded by the state central budget through general taxation.

Should the proposed scheme be adopted, the SHI would cover the reimbursement cost of children's spectacles. [26] Therefore, our target samples were taxpayers in Mongolia. This study used a trained local Mongolian-speaking enumerator to recruit 50 taxpayers, who were parents of children participating in a school-based vision screening programme conducted by Orbis International and OneSight (both are eye health NGOs). According to the central limit theorem for sample size of more than 30, [27] the sampling distribution was assumed to be normal. With this, a total of 50 parents were recruited. Parents from the Orbis contact list were randomly selected where samples were clustered into rural and urban schools. Criterion sampling was employed to recruit (i) 12 parents of children who do not need spectacles living in rural settings and 13 from urban settings, (ii) 12 parents of children who were provided with spectacles living in rural settings and 13 from urban settings.

A triple-bounded-dichotomous-choice experiment (TBDC) was used to facilitate value elicitation by leading respondents logically through consideration of their WTP. [28] The market cost of a pair of spectacles in Mongolia was used to determine realistic starting bids. The average cost of a pair of ready-made spectacles of US$5.00 (MNT15,000) and the cost of custom spectacles of US$25.00 (MNT75,000) were used to inform bids. [29] Three starting bids were established: a low bid -US$12.50, medium bid -US$17.50, and high bid -US$22.50. The subsequent bids were dependent on the acceptance (Yes) or rejection (No) of the former bid. In the case of respondents offering no maximum limit, the maximum WTP amount of US$30.00 was taken. To reduce anchoring bias, the starting bid used to initiate the survey was randomly selected.

[30] (Fig 1) . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint 

Three costs for spectacles were used to calculate the scheme's net benefits: US$5.00 for readymade spectacles, US$25.00 for custom spectacles, and US$15.00 for spectacles with an equal probability of being either. An unpublished paper's estimate showed that one in two Mongolian children would benefit from a pair of inexpensive, ready-made spectacles. Therefore, the costs of ready-made spectacles and custom spectacles were used to construct the cost of a mixed offering.

Due to COVID-19 safety considerations, data were collected via telephone survey. To ensure data quality and consistency, the trained enumerator used a standard script when conducting the survey. Demographic details such as sex, age, and educational level were collected.

Subsequently, three closed-ended questions were asked: (a) "Are you willing to pay [an amount] in additional taxes per year for any child who needs spectacles to have access to it?", followed CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint a lower bid was offered at the second question -US$12.50. The second and third question applied the same approach to arrive at the final WTP. Considering that children required annual prescription changes, the survey questions were structured to ask taxpayers how much additional annual taxes they would be willing to offer to cover the reimbursement. The WTP estimation was assumed to be conservative because if the spectacles can last more than a year, the amount paid by the taxpayers would be exceeding the costs.

Statistical Package for the Social Sciences V25 (SPSS Inc., Chicago, IL) was used for data management and analysis. Data were cleaned and checked for consistency. Because we met the criterion of central limit theorem, parametric methods were adopted for its greater statistical power and ability to use 95% confidence intervals (CI). [31] The study's primary outcome was the net benefits of a children's spectacle reimbursement scheme. The benefit in our study was determined using respondents' annual WTP for children's spectacles. [32] We assumed that the elements of benefit might include the aspect of additional lifetime income that can be attributed to the higher trajectory in earnings when the child is corrected with spectacles as URE has shown to have reduced the future income or increase children's educational inequalities. [33] The cost in our study referred to the cost of a pair of spectacles, which we reasonably assumed it to include the discounted stream of future replacement costs and the cost of collecting the tax. Net benefit was calculated by subtracting the cost of a pair of spectacles from the mean taxpayer's WTP. A positive net benefit means the . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint benefit outweighs the cost, while a negative net benefit means that cost outweighs the benefit.

Descriptive analysis was performed to obtain the mean with standard deviation (SD) along with 95% CI. The differences in demographic characteristics among respondents in rural and urban Mongolia were tested using Chi-square test for categorical variables, and t-test for continuous variables, with a significance level of 5%.

The association of mean WTP with predictor variables, including geographic setting, age, sex, educational level, and children's RE status (those with a child or children prescribed spectacles due to RE versus those with children not needing spectacles), were assessed using t-test and ANOVA.

All persons contacted (mean age 40.2 ± 9.86 years; 78.0% female) agreed to participate in the survey (n=50, response rate=100%). There was no statistical difference between respondents' mean age (p=0.411), sex (p=0.172), and educational levels (p=1.00) in rural and urban settings.

Among all respondents, more than two-thirds (84.0%) were below age 50 years, and over 90.0% completed either secondary or tertiary level education. (Table 1) . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint 

The mean amount respondents were willing to pay in additional annual taxes for any child with RE to get a pair of free spectacles was US$24.00 ± 5.15 (95% CI US$22.55 to 25.35) . Table 2 shows the calculation of net benefits. The calculations for ready-made spectacle revealed a positive net benefit of US$19.00. For custom spectacles, calculations found a negative net . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint benefit of US$1.00. An analysis of the cost of mixed provision of spectacles found a positive net benefit of US$9.00, with benefits 1.6 times outweighing the cost. (Table 3) . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint

We found a positive net benefit of US$9.00 in this CBA of a children's reimbursement scheme with equal probability of offering custom spectacles as opposed to ready-made spectacles. The mean WTP is independent of respondents' demographic characteristics, except children's RE status. Perhaps unexpectedly, parents of children without RE had significantly higher WTP in additional annual taxes for any child with RE to get spectacles than did parents of affected children though this may reflect strategic bias in the sense that as their children already had glasses they had little to gain directly from the scheme.

The mean WTP in urban (US$23.70) and rural (US$24.30) settings in Mongolia are both higher than those found in a recent study assessing parental WTP for children's spectacles in Cambodia (US$18.60 and US$13.90 in the capital and rural settings, respectively). [34] This variation can be explained by higher gross domestic product per capita in Mongolia than Cambodia (US$4,339 versus US$1,643). [35, 36] Furthermore, our study uses additional annual taxes as the payment vehicle, while in Cambodia, the payment was through out-of-pocket expenses. This suggests that the Cambodian respondents would be more conservative in offering their WTP.

Several studies demonstrate that exposure to "health shocks", such as the loss of vision associated with URE, can increase WTP. [29, 37] In Cambodia, parents of children with an eye disorder were significantly more willing to pay US$17.50 or more than parents who were unaware of their children's RE status. [34] Interestingly, in our study, respondents having children with RE had a lower WTP than respondents of children without. As noted this may be because these parents' children already received free spectacles, their responses are intended to strategically avoid any welfare loss associated with the programme. However, the sample size in our study is not powered to detect these differences and therefore, interpretation of these differences should be treated with caution.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint Were only ready-made spectacles offered our study suggests a positive net benefit (US$19.00) while were only custom spectacles offered our study suggests a slightly negative net benefit (US$1.00). This suggests the benefits outweigh the costs of providing inexpensive ready-made spectacles but not custom spectacles. Concerning this, one proposed strategy would be for parents to "top-up" the government-subsidised spectacles when the cost of spectacles exceeds the subsidised amount. For example, if the government subsidises US$10.00 for any type of spectacles, parents will have to pay for the additional costs. This is especially referring to custom spectacles where the cost is often higher than the ready-made spectacles.

WTP has been widely used in the eye care sector to aid in service delivery planning, such as scheme for the cross-subsidisation of cataract surgery or spectacles. [34, 38] We used WTP to estimate the potential benefits to inform a CBA, a novel approach in evaluating interventions related to children's URE. Due to limited resources and high demand for children's refractive services in Mongolia, policymakers must be informed of the value added by the intervention.

This CBA should serve as a reference for the Mongolian government in making an informed policy decision.

Strengths of the current study include our having used a number of approaches recommended to increase the validity of the contingent valuation estimates: (i) using telephone interviews instead of surveys posted by mail; (ii) using WTP rather than willingness-to-accept; (iii) pretesting the survey before actual interviews; (iv) phrasing the WTP questions in a hypothetical scenario by indicating additional taxes that respondents would have to pay to subsidise free spectacles; and (v) collecting respondents' demographic characteristics. [39] Limitations of the study must also be acknowledged. The relatively small sample size may explain the lack of a significant association between WTP and most demographic factors.

Secondly, hypothetical WTP surveys using contingent valuation may be prone to overestimation of the actual WTP amount. [40] Further, it has been suggested that a visual aid should be used when possible in WTP studies, so that biases due to miscomprehension can be avoided when . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint participants are asked to make decisions about unfamiliar subjects. [24] We originally planned to present visual aids demonstrating to respondents the impact of spectacle wear. However, COVID-19 precautions made face-to-face interviews and the use of such aids impossible.

Finally, despite including respondents whose children participated in the vision screening programme in rural and urban schools (Orbis International's contact list), we did not include in our sample individuals not involved in the screening programme. This might cause selection bias and thus affecting the generalisability of the findings. We suggest employing a random sampling method and sample size power calculation for future studies.

Despite its limitations, our analysis is one of the few examining the cost-benefit of national programmes providing spectacles for children. Our preliminary findings suggest such a scheme is a good value for Mongolia.

Supporting information S1 File. Database and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020 : the Right to Sight: an analysis for the Global Burden of Disease Study. Lancet Glob Heal. 2020;9: E144-E160.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 12, 2021. ; https://doi.org/10.1101/2021.08.12.21261541 doi: medRxiv preprint

Causes of blindness

Global magnitude of visual impairment caused by uncorrected refractive errors in 2004

Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: a systematic review and meta-analysis

Barriers to use of refractive services in Mozambique

Barriers to uptake of prescribed refractive spectacles amongst Nigerian students

Ametropia, preschoolers' cognitive abilities, and effects of spectacle correction

The impact of providing vision screening and free eyeglasses on academic outcomes: evidence from a randomized trial in title I elementary schools in Florida

Disability-related-distress in primary school learners with vision impairment due to uncorrected refractive error in KwaZulu-Natal Province, South Africa -A qualitative study

Effect of providing near glasses on productivity among rural Indian tea workers with presbyopia (PROSPER): a randomised trial

Effect of providing free glasses on children's educational outcomes in China: Cluster randomized controlled trial

Potential lost productivity resulting from the global burden of uncorrected refractive error

World Population Review -Mongolia

Free and universal access to primary healthcare in Mongolia: The service availability and readiness assessment

CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted

World Health Organisation. Priority assistive products list. GATE Initiat. 2016. 17. Mongolia Government. Mongolia National Program for Non-Communicable Diseases -The Government Resolution No.34 "Adoption of the National Programme

Spectacle wear among children in a schoolbased program for ready-made vs custom-made spectacles in India a randomized clinical trial

A randomized, clinical trial evaluating ready-made and custom spectacles delivered via a school-based screening program in China

A Randomized Noninferiority Trial of Wearing Adjustable Glasses versus Standard and Ready-made Spectacles among Chinese Schoolchildren: Wearability and Evaluation of Adjustable Refraction III

Standard school eye health guidelines for low and middle-income countries

Bayoumi AM. The measurement of contingent valuation for health economics

Using assessment of willingness to pay to improve a Cambodian spectacle service

Law of Mongolia on Social Insurance. Mongolia

Central limit theorem: the cornerstone of modern statistics

Contingent valuation analysis of willingness to pay to reduce childhood obesity

Health shock and preference instability: assessing health-state dependency of willingness-to-pay for corrective eyeglasses

International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted

Investigating starting-point bias: A survey of willingness to pay for insecticide-treated nets

Non-Parametric Tests, and Large Studies-a Paradox of Statistical Practice

Willingness to pay as a measure of health benefits

A better vision for development: Eyeglasses and academic performance in rural primary schools in China

Parents' willingness to pay for children's spectacles in Cambodia

Mongolia Gross Domestic Product per capita

Cambodia Gross Domestic Product per capita

The utility of different health states as perceived by the general public

Correction of refractive error and presbyopia in Timor-Leste

Report of the NOAA panel on contingent valuation

Hypothetical versus real willingness to pay in the health care sector: results from a field experiment

International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

We would like to thank Orbis International and Orbis Mongolia for the supports given.