Journal of Pediatric Ophthalmology and Strabismus

Original Article Supplemental Data

Regionally Specific Economic Impact of Screening and Treating Retinopathy of Prematurity in Middle-Income Societies in the Philippines

Mina M. Naguib, MD; Rebecca R. Soares, MD, MPH; Rachelle Anzures, MD; Joanne Kamel, MD; Eeshwar K. Chandrasekar, MPH; Michael Rothschild, MD; Alcides Fernandes, MD; R. V. Paul Chan, MD; Timothy W. Olsen, MD

Abstract

Purpose:

To estimate the economic effects of implementing a universal screening and treatment program for retinopathy of prematurity (ROP) in the Philippines with the Economic Model for Retinopathy of Prematurity (EcROP).

Methods:

The EcROP is a cost-effectiveness, cost-benefit, and cost-utility analysis. Fifty parents of legally blind individuals (aged 3 to 28 years) from three schools for the blind in the Philippines were interviewed to estimate the societal burden of raising a blind child. A decision tree analytic model, with deterministic and probabilistic sensitivity analysis, was used to calculate the incremental cost-effectiveness ratio (primary outcome) and the incremental monetary benefit (secondary outcome) for implementing an optimal national ROP program, compared to estimates of the current policy. Findings were extrapolated to estimate the national economic benefit of an ideal screening and treatment program.

Results:

The incremental cost-effectiveness ratio for a national program over the current policy was strongly favorable to the ideal program for the Philippines and represents an opportunity for substantial societal cost savings. The per-child incremental, annual monetary benefit of a national program over the current policy was $2,627. Extrapolating to the population of children at risk in 1 year showed that the national annual net benefit estimate would be $64,320,692, which is favorable to the current policy.

Conclusions:

The EcROP demonstrates that implementing a national ROP screening and treatment program is cost-saving and cost-effective, and would substantially decrease childhood blindness in the Philippines.

[J Pediatr Ophthalmol Strabismus. 2019;56(6):388–396.]

Abstract

Purpose:

To estimate the economic effects of implementing a universal screening and treatment program for retinopathy of prematurity (ROP) in the Philippines with the Economic Model for Retinopathy of Prematurity (EcROP).

Methods:

The EcROP is a cost-effectiveness, cost-benefit, and cost-utility analysis. Fifty parents of legally blind individuals (aged 3 to 28 years) from three schools for the blind in the Philippines were interviewed to estimate the societal burden of raising a blind child. A decision tree analytic model, with deterministic and probabilistic sensitivity analysis, was used to calculate the incremental cost-effectiveness ratio (primary outcome) and the incremental monetary benefit (secondary outcome) for implementing an optimal national ROP program, compared to estimates of the current policy. Findings were extrapolated to estimate the national economic benefit of an ideal screening and treatment program.

Results:

The incremental cost-effectiveness ratio for a national program over the current policy was strongly favorable to the ideal program for the Philippines and represents an opportunity for substantial societal cost savings. The per-child incremental, annual monetary benefit of a national program over the current policy was $2,627. Extrapolating to the population of children at risk in 1 year showed that the national annual net benefit estimate would be $64,320,692, which is favorable to the current policy.

Conclusions:

The EcROP demonstrates that implementing a national ROP screening and treatment program is cost-saving and cost-effective, and would substantially decrease childhood blindness in the Philippines.

[J Pediatr Ophthalmol Strabismus. 2019;56(6):388–396.]

Introduction

Retinopathy of prematurity (ROP) is a blinding disorder that originates from aberrant retinal blood vessel growth and affects preterm and low birth weight infants. Aberrant retinal blood vessel proliferation leads to retinal detachment and permanent blindness.1 In industrialized nations with low infant mortality rates (< 9 in 1,000 live births), ROP is less common because of advanced neonatal intensive care unit (NICU) management. Although effective screening protocols are in place to guide appropriate treatment interventions, the penetrance of screening is not ideal.2 The incidence of ROP-related blindness can be decreased further in developed nations by improving screening, optimizing operational programs, and implementing effective treatments. In nations with high infant mortality rates (> 60 of 1,000 live births), ROP is not a substantial cause of blindness because NICUs are either not available or are limited in scope, so preterm and low birth weight infants typically do not survive. ROP primarily affects middle-income or developing nations.3 When infant mortality rates range from 9 to 60 per 1,000 live births and a NICU infrastructure is available with local medical expertise to treat preterm infants, there is an opportunity to also save their sight. However, many programs in developing nations lack the capacity to effectively manage the rapidly expanding ROP population. Gilbert et al.4 estimated that approximately 60% of childhood blindness in middle-income or developing countries is due to ROP.

In the United States, screening guidelines for ROP include infants born at 30 weeks' gestational age or less and infants weighing 1,500 g or less.5 A recent study suggests that modified screening criteria may be indicated for varying populations.6 For example, in the Philippines, the current criteria of 32 weeks' gestational age or less and a birth weight of 1,500 g or less may miss approximately 16% of neonates who require ROP treatment.6 Genetic differences, varying levels of oxygen management, NICU nursing staff-to-infant ratios, and other environmental factors underline the need for country-specific screening criteria. Referrals from pediatricians who may primarily treat the neonates are not well characterized and may represent an opportunity for future educational efforts.

In the current study, we applied the previously described Economic Model for ROP (EcROP)2 to compare the cost-effectiveness, cost-utility, and cost-benefit of implementing an ideal ROP screening and treatment program in the Philippines.

Patients and Methods

We applied the methods as described in our prior EcROP study.2 The current economic analysis compared the effects of implementing an ideal national ROP screening and treatment program with 100% penetrance to current estimates from the Philippines. Sensitivity analyses were incorporated to adjust our calculated projections for variation and to demonstrate the likelihood of our financial estimates.

After obtaining appropriate institutional review board consent from both Emory University and the local research committee of Ospital ng Makati Hospital, country-specific economic data were compiled. Written informed parental consent was obtained in the preferred language (English or Tagalog) from each study participant or parent before conducting an interview; no stipend was provided. Parents of all individuals (aged 3 to 28 years) with blindness (≤ 20/200 in best eye), regardless of cause, were included because the burden of disease from blindness should confer similar personal and societal costs. The detailed economic data were acquired using known published and public data and face-to-face interviews, using a standardized form from our translated survey. Interview data were used to determine the society-specific direct and indirect costs to raise a blind child. The structure of and questions on the survey in the Philippines were adapted from previous EcROP surveys used in the United States, Mexico, and Peru.2,7

Decision Analytic Model

The cost-utility analysis and cost-benefit analysis were generated using decision analytic software (TreeAge Pro software program; TreeAge Software, Inc., Williamstown, MA). This analytic model has been described previously.2

The decision tree with only the ideal branch as cited by Rothschild et al.2 was used to represent the ideal branch for comparison in the current study. The decision tree for current practice uses an identical decision analysis referred to as an ideal national program, but it differs in costs, utilities, and probabilities. Table 1 summarizes the specific costs, utilities, probabilities, and assumptions from locally derived data extrapolated to estimate the national impact and incorporates published population figures specific to the Philippines.

ROP Estimates of Probability Parameters and Effectiveness Inputs

Table 1:

ROP Estimates of Probability Parameters and Effectiveness Inputs

Probability and Assumption Inputs

Both decision tree branches (current and ideal) assume screening and treatment of infants weighing 1,500 g or less. We assumed a 100% screening penetrance in the ideal national branch, compared with the 50% screening penetrance for the current-policy branch. The 50% value is based on data from the local NICU affiliated with regional Philippine ophthalmology collaborators. The ideal screening scenario (100%) assumes that all infants who met the criteria received the intervention, no deaths occurred after ROP screening, and ROP screening sensitivity and specificity were 100%.

The decision tree combined probability outcomes for untreated infants derived from published natural history data from the Cryotherapy for ROP study8,9 and the Early Treatment for Retinopathy of Prematurity study.10 Other probability analyses were derived from published or regional data (Table 1).11

Effectiveness Measures

A quality-adjusted life year (QALY) is a widely used measure of utility (1 = perfect health; 0 = death). We determined the cost per QALY saved from ROP screening and treatment on the basis of average published ranges.12 We discounted these QALYs by 3% annually during the expected lifetime to adjust for the future value of the QALY.13

Costs

Both direct and indirect costs are included in the EcROP model, discounted at 3% annually and reported in 2016 U.S. dollars. Caregiver and labor data were adjusted for known, in-country gender differences. Survey data variability was determined using a 95% confidence interval.

Direct costs of ROP screening were estimated using in-country (Ospital ng Makati) data that estimated depreciated equipment plus labor and supplies incurred during screening (Table A, available in the online version of this article). Physician and nursing labor costs were based on monthly salaries combined with estimated hours dedicated to ROP screening and treatment. We used an 8-year depreciation for reusable equipment and estimated that 3 neonates received treatment for every 10 neonates screened.

Direct Cost of ROP Screening and Follow-upa

Table A:

Direct Cost of ROP Screening and Follow-up

The direct cost of ROP treatment was determined from panretinal photocoagulation, the standard of care for ROP (Table B, available in the online version of this article). Approximately 30 infants are screened per year in each NICU with 8 years of equipment use.

Direct Cost of ROP Laser Treatmenta

Table B:

Direct Cost of ROP Laser Treatment

Direct costs for follow-up care were similarly estimated (Table A). Annual visits continued for 5 years and were also discounted at 3%.

Indirect costs were determined directly from family survey responses administered to parents and/or guardians of 50 children from three socioeconomically diverse schools for the blind: (1) a government school located in metropolitan Manilla; (2) a private school for the blind in a distinctly separate metropolitan area of Manilla; and (3) a rural, semi-private school (financed from both private donations and limited government funds) in Davao City on the southern Philippine island of Mindanao. Study authors (MMN, JK, or RA) administered the surveys in private family interviews. Survey questions gathered demographic information, parental income, gender-specific caregiver time allocation, and other unique or specific costs of raising a blind child as compared to a sighted child.

Overall societal cost calculations were based on the sum of productivity losses by the caregiver as calculated by multiplying a gender-adjusted per-capita contribution to gross domestic product per hour by the number of hours lost per year. In 2016, the contribution to gross domestic product in the Philippines was estimated to be $4.87 per hour.14 This was multiplied by either 0.608 (male) or 0.392 (female) to account for gender differences in labor participation rate15 and multiplied by the number of hours lost by the respective caregiver. Blind individuals in the Philippines are rarely independent of their parents after reaching adulthood (none in our survey), so we discounted (3%) household productivity losses annually for 37 years or the average number of work years was calculated16 until retirement (age 23 to 60 years).17 The annual productivity loss for blind individuals compared to individuals with sight was calculated by multiplying gross domestic product per capita in the Philippines ($2,943) by labor force participation rate (63%). The discounted (3%) annual productivity loss for the blind individual was calculated for 42 years, from adulthood to typical retirement age (age 18 to 60 years).

Analysis

The primary outcome of the current study, incremental cost-effectiveness ratio (ICER), was determined by using a cost-utility analysis. The ICER represents the change in cost to incremental benefit of a specific therapeutic intervention. For cost-utility analysis, benefit was defined as QALYs gained as a result of an intervention (screening and treatment of ROP at 100% penetrance). The cost-benefit analysis determined the net benefit for the intervention using willingness to pay (WTP). Incremental net monetary benefit, the secondary outcome, was calculated using the equation: net QALYs gained × WTP/QALY – additional cost.

Uncertainty and Variability

We performed both probabilistic and deterministic sensitivity analyses, as previously reported.2 For the probabilistic analysis, we used the decision tree as cited by Rothschild et al.2 with Monte Carlo simulations for the base case and again with subsequent best- and worst-case scenarios by changing inputs to values that increased and decreased the favorability of the national program, respectively.

A deterministic analysis was done to account for a possible variation in the ICER or incremental net monetary benefit by using a one-way sensitivity analysis of direct cost variations from the base scenario. Deterministic costs were varied by ±30%.

Results

Table 1 summarizes the inputs used for the decision tree model. We estimated that the total direct cost of raising a blind child from birth to age 18 years was $20,215. The total screening cost was calculated as $296 per neonate. The treatment cost was $637 per neonate, and the cost of follow-up visits was $349 per neonate, totalling $1,282 during the life of each child.

The estimated annual productivity losses by mothers and fathers, respectively, were $2,350 and $124, annualized to $2,474 per household. Total caregiver lifetime productivity loss was $54,850. The annual productivity loss for a blind individual was $1,863, and lifetime productivity losses for each blind individual were calculated as $25,551. The utility factors for blindness and sight were determined to be 0.61 and 0.89, respectively.

Table 2 summarizes the components for the ICER calculation. The ICER of an optimal national program over the current policy results in a negative value. A negative ICER indicates that universal screening would be superior to the current screening, cost-saving, and cost-effective. The incremental net monetary benefit per child of a national program over the current policy was $2,627. The incremental net monetary benefit, scaled to the population of children at risk (birth weight of < 1,500 g) born in 1 year (24,483 infants)18,19 was $64,320,692 (Table 2).

ROP Base-Case Analysis of Cost-Effectiveness and Net Health Benefits

Table 2:

ROP Base-Case Analysis of Cost-Effectiveness and Net Health Benefits

Deterministic Sensitivity Analysis

Deterministic sensitivity is reported in Table 3. As compared to other regions, and unique to the Philippines, we were unable to determine a threshold value. Therefore, we found no value at which the current policy was preferred to a national screening program for any single variable modified within its given range.

One-Way, Deterministic Sensitivity Analysis of Direct Costs in ROP Managementa

Table 3:

One-Way, Deterministic Sensitivity Analysis of Direct Costs in ROP Management

Probabilistic Sensitivity Analysis

Using probabilistic sensitivity analysis, 10 variables cumulatively accounted for more than 99.999% of uncertainty: (1) QALY of a sighted individual; (2) prevalence of ROP; (3) QALY of a blind individual; (4) caregiver productivity loss; (5) probability that untreated individuals would have a favorable structural outcome; (6) probability that individuals who screened positive for ROP would receive treatment; (7) direct cost of raising a blind individual; (8) probability that a treated individual would have a favorable structural outcome; (9) blind individual productivity loss; and (10) discount rate (Table 1).

Base-Case Scenario

The comparison of incremental cost-effectiveness to QALYs (Figure 1A) indicates that a national program would be cost-saving or cost-effective 86% of the time, which is below the standard $50,000/QALY threshold. Most of the time (64%), an iteration of EcROP will fall in the bottom right quadrant, indicating an increase in QALYs while also resulting in cost savings. Less than one-fourth of the time (22%), an iteration will fall in the top right quadrant. Only 9% of iterations fell in the top right quadrant above the dashed line, representing an ICER greater than the WTP threshold

Cost-effectiveness for the national retinopathy of prematurity program versus current practice. (A) Incremental cost-effectiveness. Each point represents a probabilistic value. Cost-saving possibilities are in the bottom right quadrant; cost-effective possibilities are in the bottom right and top right quadrants. Dashed line is the $50,000/quality-adjusted life year (QALY) threshold. The ellipse represents 95% of iterations. (B) Acceptability curve.

Figure 1.

Cost-effectiveness for the national retinopathy of prematurity program versus current practice. (A) Incremental cost-effectiveness. Each point represents a probabilistic value. Cost-saving possibilities are in the bottom right quadrant; cost-effective possibilities are in the bottom right and top right quadrants. Dashed line is the $50,000/quality-adjusted life year (QALY) threshold. The ellipse represents 95% of iterations. (B) Acceptability curve.

Figure 1B represents an acceptability curve ranging from $0/QALY to $100,000/QALY. The probability that the intervention will have a positive economic benefit (ie, either cost-effective or cost-saving) is 64% to 90%, from $0 to $100,000 WTP, respectively.

Best- and Worst-Case Scenario

The best-case scenario resulted in an 89% probability that the intervention would have a positive economic benefit on the basis of a WTP of $50,000/QALY and a 70% probability that the intervention would result in cost savings. The worst-case scenario resulted in a 56% probability that the intervention would be cost-effective.

Cost-Benefit Outcomes

Table 4 shows the incremental net monetary benefits of the intervention with simultaneous adjustment for the WTP. The incremental net benefit per child and per year varies because the WTP/QALY varies from the typical $50,000/QALY. The incremental net monetary benefit ranged from $400 to $4,981 per child at a WTP of $0/QALY to $100,000/QALY, respectively.

Incremental Net Monetary Benefit of a National ROP Program Over Current Practice, per Child,Varied by WTPa

Table 4:

Incremental Net Monetary Benefit of a National ROP Program Over Current Practice, per Child,Varied by WTP

Discussion

With results showing a cost savings per child of $2,627 and a yearly net monetary benefit to society of $64,320,692, the current study reinforces the need for a universal ROP screening and treatment program in the Philippines. With a robust sensitivity analysis to account for uncertainty in our model, our findings for implementing a national program indicate substantial cost savings. If no value ($0 WTP) were applied to one QALY for one child, society would save $400 per child per year.

Our findings are similar to those of our original economic analysis in Peru,7 and further highlight the value of EcROP. Our methodology also predicted a strongly favorable economic impact for implementing a similar national screening and treatment effort in the United States and Mexico.2 In the current study, the EcROP supports the economic and societal benefits of implementing an effective ROP screening and treatment program in another middle-income country and the first country in Asia. We believe that the EcROP is generalizable across regions and can predict the economic and societal impact of universal ROP screening and treatment in any region of the globe. In-country survey data are critical to accurately assess and understand the local economic and social systems in a region, such as the Philippines, although we found key differences from Peru, Mexico, and the United States. Nevertheless, substantial barriers limit adequate screening in many areas. We believe that by working through the ministers of health, advocacy and political systems, and the ophthalmology community, blindness from ROP could be markedly reduced, along with a meaningful secondary economic gain for countries willing to address the findings from our EcROP analysis.

One major difference between the Philippines and previously studied countries is that current governmental health policy covers the cost of ROP screening and treatment in government-run hospitals. The Philippines has the advantage of a public health insurance program (PhilHealth) that operates similarly to Medicaid in the United States. Phil-Health covers the majority of payments for ROP management, whereas families contribute a small co-pay.

The current universal program in the Philippines has not yet adequately addressed or decreased the burden of ROP-related blindness, and there are several possible reasons. First, the population of the Philippines exceeds 90 million,20 with an estimated 24,483 infants born each year weighing 1,500 g or less.18,19 Yet in 2015, approximately 175 ophthalmologists in the Philippines were trained to screen for ROP and only 48 were able to effectively treat patients with ROP.21 Next, the majority of qualified professionals are located in urban areas that are too far away for many rural families. Although the NICU locations are well distributed through-out the Philippines, many areas are underserved by ophthalmic professionals. Solutions include redistributing health care expertise, providing education on screening and management to geographically diverse regions, and instituting centralized screening and transfer of care for at-risk neonates. As used in India,22,23 photoscreening may be helpful for remote ROP screening. In many developing nations, and especially underdeveloped nations, there is a lower socioeconomic segment of the population where health care access is so limited that premature infants do not survive. In such populations, ROP does not exist and the emphasis is on developing health care infrastructure.

Another opportunity unique to the Philippines is to improve ROP screening through educational efforts to optimize referrals from pediatricians to ophthalmologists. A recent study24 that surveyed 409 Filipino pediatricians from 40 randomly selected government and private hospitals found that referrals may not follow the evidence-based screening criteria. The four most common barriers to proper screening were: (1) the inability of families to return for follow-up (41%); (2) the expense of the screening (38%); (3) the limited access to ophthalmologists (16%); and (4) parental safety concerns for treatment (13%). Only 33% of the pediatricians reported no barriers.

Our study has several limitations. The EcROP study methods presented are limited by a small sample size with extrapolated data to represent the country. To add perspective to the validity of our overall data and conclusions, probability analysis indicates that our data have a high likelihood to be cost-saving. Also, as in many developing nations, populations in an urban area do not always reflect the country's broad health care variations. We intentionally investigated both urban and rural areas. Also, we included assumptions about ROP prevalence in infants weighing less than 1,500 g and the unknown penetrance of current screening patterns. Our data were based on best estimates by those actively treating ROP in-country. Additionally, cost data were collected from a small number of hospitals, and labor costs may vary. Uncertainties were addressed by implementing a robust probabilistic sensitivity analysis that strongly supported a high likelihood for overall cost savings and cost effectiveness.

One of the costs not specifically addressed in the EcROP model is the cost of training personnel to screen for and treat ROP. Although training costs represent a significant cost of scale in a universal screening program, the cost per child remains negligible. For example, the average cost to society of training one retina specialist in the Philippines is approximately $40,000 (1,040,000 pesos for 4 years of medical school and doubling it to overestimate the cost to society of training an ophthalmology resident and retina fellow for 6 years).25 If each retina specialist then sees 10 children a week for 30 years, the cost per child of training an additional retina specialist is approximately $3. If this amount were doubled to include the cost per child of training screening personnel, the additional cost of screening and treating a child would be $6. This added cost of training, although not explicitly addressed in our model, is well contained within our sensitivity analyses for cost, with the 30% higher, worst-case scenario costs for screening and treating children being highly cost-effective.

The current study clearly demonstrates that a national, universal screening and treatment program for ROP in the Philippines would be not only cost-effective but also cost-saving, with a high likelihood of incremental societal and net monetary benefit. The results of this study affirm that ROP in the Philippines must be addressed and a failure to do so will result in an expanding population of blind individuals, with the associated societal and financial burdens.3 The use of a multifaceted approach centered on the education of providers, operational optimization, and economic support to address the key local and regional barriers to ROP screening and treatment should promote long-term gains for the Philippines. Our EcROP data should be used by the Ministry of Health in the Philippines to advocate for a national program that would optimize the societal benefit and apply limited health care resources in an economically responsible manner. Our data indicate a high degree of analytic certainty that such efforts will yield robust rewards. In fact, screening and treatment for ROP may be one of the best uses of health care financial resources in all of medicine.

References

  1. Hartnett ME, Penn JS. Mechanisms and management of retinopathy of prematurity. N Engl J Med. 2012;367(26):2515–2526. doi:10.1056/NEJMra1208129 [CrossRef]23268666
  2. Rothschild MI, Russ R, Brennan KA, et al. The Economic Model of Retinopathy of Prematurity (EcROP) screening and treatment: Mexico and the United States. Am J Ophthalmol. 2016;168:110–121. doi:10.1016/j.ajo.2016.04.014 [CrossRef]27130372
  3. Gilbert C. Retinopathy of prematurity: a global perspective of the epidemics, population of babies at risk and implications for control. Early Hum Dev. 2008;84(2):77–82. doi:10.1016/j.earlhumdev.2007.11.009 [CrossRef]18234457
  4. Gilbert C, Fielder A, Gordillo L, et al. International NO-ROP Group. Characteristics of infants with severe retinopathy of prematurity in countries with low, moderate, and high levels of development: implications for screening programs. Pediatrics. 2005;115(5):e518–e525. doi:10.1542/peds.2004-1180 [CrossRef]15805336
  5. Fierson WMAmerican Academy of Pediatrics Section on Ophthalmology; American Academy of Ophthalmology; American Association for Pediatric Ophthalmology and Strabismus; American Association of Certified Orthoptists. Screening examination of premature infants for retinopathy of prematurity. Pediatrics. 2018;142(6):e20183061. doi:10.1542/peds.2018-3061 [CrossRef]30478242
  6. Corpus K, Jimenez JM IV, Anzures R, et al. Retinopathy of Prematurity Working Group. Proposed new retinopathy of prematurity screening criteria: evidence for including older and heavier Filipino premature babies. Philipp J Ophthalmol. 2013;38:72–79.
  7. Dave HB, Gordillo L, Yang Z, Zhang MS, Hubbard GB III, Olsen TW. The societal burden of blindness secondary to retinopathy of prematurity in Lima, Peru. Am J Ophthalmol. 2012;154(4):750–755. doi:10.1016/j.ajo.2012.04.003 [CrossRef]22831839
  8. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter Trial of Cryotherapy for Retinopathy of Prematurity: ophthalmological outcomes at 10 years. Arch Ophthalmol. 2001;119(8):1110–1118. doi:10.1001/archopht.119.8.1110 [CrossRef]11483076
  9. Palmer EA, Hardy RJ, Dobson V, et al. Cryotherapy for Retinopathy of Prematurity Cooperative Group. 15-year outcomes following threshold retinopathy of prematurity: final results from the multi-center trial of cryotherapy for retinopathy of prematurity. Arch Ophthalmol. 2005;123(3):311–318. doi:10.1001/archopht.123.3.311 [CrossRef]15767472
  10. Good WV, Hardy RJ, Dobson V, et al. Early Treatment for Retinopathy of Prematurity Cooperative Group. The incidence and course of retinopathy of prematurity: findings from the early treatment for retinopathy of prematurity study. Pediatrics. 2005;116(1):15–23. doi:10.1542/peds.2004-1413 [CrossRef]15995025
  11. Chen J, Stahl A, Hellstrom A, Smith LE. Current update on retinopathy of prematurity: screening and treatment. Curr Opin Pediatr. 2011;23(2):173–178. doi:10.1097/MOP.0b013e3283423f35 [CrossRef]
  12. Wittenborn J, Rein D. Cost of Vision Problems: The Economic Burden of Vision Loss and Eye Disorders in the United States. Presented at NORC at the University of Chicago. ; June 11, 2013. ; Chicago, IL. . https://www.preventblindness.org/sites/default/files/national/documents/Economic%20Burden%20of%20Vision%20Final%20Report_130611_0.pdf. Accessed January 1, 2016.
  13. Siegel JE, Weinstein MC, Russell LB, Gold MRPanel on Cost-Effectiveness in Health and Medicine. Recommendations for reporting cost-effectiveness analyses. JAMA. 1996;276(16):1339–1341. doi:10.1001/jama.1996.03540160061034 [CrossRef]8861994
  14. Asian Productivity Organization. APO Productivity Databook2013. http://www.apo-tokyo.org/publications/wp-content/uploads/sites/5/APO_Productivity_Databook_20131.pdf. Published 2013.
  15. The World Bank. Labor force, female (% of total labor force). http://data.worldbank.org/indicator/SL.TLF.TOTL.FE.ZS. Published 2014. Accessed December 16, 2016.
  16. Central Intelligence Agency. Mother's mean age at first birth. The World Factbook. https://www.cia.gov/library/publications/resources/the-world-factbook/fields/352.html. Published 2013. Accessed December 16, 2016.
  17. Department of Labor and Employment (DOLE). Labor Code of the Philippines: Renumbered. DOLE Edition. https://www.dole.gov.ph/wp-content/uploads/2017/11/LaborCodeofthePhilip-pines20171.pdf. Published 2015.
  18. Stanford Children's Health. Very Low Birthweight. http://www.stanfordchildrens.org/en/topic/default?id=very-low-birthweight-90-P02424. Published 2011. Accessed December 16, 2016.
  19. Philippine Statistics Authority. Live Births in the Philippines: 2014. https://psa.gov.ph/content/live-births-philippines-2014. Published 2014. Accessed December 16, 2016.
  20. The World Bank. Population, total Philippines. http://data.worldbank.org/indicator/SP.POP.TOTL. Published 2015. Accessed December 16, 2016.
  21. Raihan A, Azad R. APAO 2015 ROP Round Table Discussion. http://2015.apaophth.org/congress/register/PDS_program_schedule.php?session_id=318&SUPPRESS_MENU_BAR=1. Published 2015.
  22. Vinekar A, Mangalesh S, Jayadev C, Gilbert C, Dogra M, Shetty B. Impact of expansion of telemedicine screening for retinopathy of prematurity in India. Indian J Ophthalmol. 2017;65(5):390–395. doi:10.4103/ijo.IJO_211_17 [CrossRef]28573995
  23. Vinekar A, Jayadev C, Kumar S, et al. Impact of improved neonatal care on the profile of retinopathy of prematurity in rural neonatal centers in India over a 4-year period. Eye Brain. 2016;8:45–53. doi:10.2147/EB.S98715 [CrossRef]
  24. Te NDA, Loy MJ, Chua CE, Jimenez JM, Anzures R. Awareness and practice patterns of pediatricians regarding retinopathy of prematurity: a multicenter study. Philipp J Ophthalmol. 2013;38:86–93.
  25. University of Santo Tomas. Tuition Fees. www.ust.edu.ph/tuition-fees/. Published 2016.

ROP Estimates of Probability Parameters and Effectiveness Inputs

CharacteristicBase CaseBest CaseWorst CaseDistribution Type
Parametera
  Prevalence of infants with ROP weighing < 1,500 g27%b99%1%Beta
  Screening penetrance under current practice50%c0%80%cNone
  Screening penetrance under ideal national program100%c100%81%None
  Probability of a favorable structural outcome in a treated child (±10%)e90%d99%81%Beta
  Probability of a favorable structural outcome in an untreated child (±10%)e50%f45%55Beta
  Probability that a child who screens positive will be treated (±10%)e10%g20%0%Beta
  Sensitivity of screening100cNone
  Specificity of screening100cNone
  Discount rate3h05Beta
Effectiveness inputs
  QALY of a blind individual0.61i0.390.8
  QALY of being blind discounted over lifetimej17.6911.3823.20Triangular
  QALY of a normal-seeing individual0.891210.69
  QALY of normal-seeing individual discounted over lifetimej25.8120.0129.00Triangular
Main costse,k
  Screening (±30%)$296$197.33$394.67None
  Treatment (±30%)$637.49$446.24$828.74None
  Follow-up (±30%)$349.06$244.35$435.78None
  Raising a blind child (0, upper limit 95% CI)$20,215.33$51,519.970Gamma
  Productivity loss, caregivers (0, upper limit 95% CI)$54,850.39$150,545.460Gamma
  Productivity loss, blind individual (±30%)$25,551.42$17,885.99$33,216.84Triangular

ROP Base-Case Analysis of Cost-Effectiveness and Net Health Benefits

ParameteraNational ProgramCurrent Practice
Incremental cost (cost national – cost current practice)−$8,508
Incremental effectiveness (QALY national – QALY current practice)0.0449
Incremental cost-effectiveness ratioCost-saving
Net monetary benefit national program$1,235,795
Net monetary benefit current practice$1,233,168
Incremental net monetary benefit of national program$2,627b
Incremental net monetary benefit, scaled to population at risk during 1 year$64,320,692c

One-Way, Deterministic Sensitivity Analysis of Direct Costs in ROP Managementa

Direct Cost-Inputs Per ChildIncremental Net Monetary Benefit Per Child of National vs Current ProgrambIncremental Net Monetary Benefit Per Child of Nation During 1 Yearb
Base case$2,627.05$64,320,692
Screening
  Increased 30%$2,577.72$63,112,896.48
  Decreased 30%$2,676.39$65,528,732.76
Treatment
  Increased 30%$2,624.48$64,257,768.32
  Decreased 30%$2,629.64$64,384,105.76
Follow-up
  Increased 30%$2,625.65$64,286,414.6
  Decreased 30%$2,628.47$64,355,459.48

Incremental Net Monetary Benefit of a National ROP Program Over Current Practice, per Child,Varied by WTPa

WTP ($/QALY)Incremental Monetary Benefit Per ChildIncremental Net Monetary Benefit Per Year
0$400.45$9,804,617
1,000$858.46$21,018,534
10,000$2,627.05$64,320,692
50,000$4,980.65$121,946,234
100,000$400.45$9,804,617

Direct Cost of ROP Screening and Follow-upa

ItemUnitCost in U.S. Dollars
Labor
  Nursing
Cost per visit$3.69
No. of hours per visit1.5
Average hourly wage$2.46
  Physician
Cost per visit$15.93
No. of hours per patient screened1.5
Average hourly wage$10.62
  Total labor costsCost per visit$19.62
Reusable equipment
  Indirect ophthalmoscopeOne-time cost$5,179
  Eyelid speculumOne-time cost$18.66
  Scleral depressorOne-time cost$103.58
  Total reusable equipment costsCost per infant screenedb$22.09
Single-use equipment
  Phenylephrine and tropicamideCost per visit$12.47
  ProparacaineCost per visit$12.53
  Erythromycin eye ointmentCost per visit$9.52
  Total single-use equipment costCost per visit$34.51
Total cost of one screening or follow-up visitc$76.22
Total cost of screening per neonateb$304.88
Total cost of follow-up visits per neonated$381.10e

Direct Cost of ROP Laser Treatmenta

ItemUnitCost in U.S. Dollars
Labor
  Nursing
Cost per treatment$9.84
No. of hours per treatment4
Average hourly wage$2.46
  Physician
Cost per treatment$21.24
No. hours per patient treated2
Average hourly wage$10.62
  Total labor costsCost per treatment$31.08
Equipment
  Rented laserCost per treatment$155.38
  Eyelid speculumOne-time cost$18.66
  Scleral depressorOne-time cost$103.58
  28.00 D lensOne-time cost$414.34
  Total reusable equipment costsCost per infant screenedb$157.61
Single-use equipment
  Phenylephrine and tropicamideCost per visit$12.47
  ProparacaineCost per visit$12.53
  Basic saline solutionCost per visit$24.55
  Tobramycin and dexamethasone eye ointmentCost per visit$10.96
  Atropine eye dropsCost per visit$6.61
  Sodium hyaluronate eye dropsCost per visit$6.84
  Total single-use equipment costCost per visit$73.96
  Total direct cost of laser treatment$656.61c
Authors

From the Department of Vitreoretinal Diseases and Surgery, Emory Eye Center, Atlanta, Georgia (MMN, RRS, EKC, MR, AF, TWO); the Cullen Eye Institute, Baylor College of Medicine, Houston, Texas (MMN); Wills Eye Institute, Philadelphia, Pennsylvania (RRS); the Department of Ophthalmology, Ospital ng Makati, Makati, Manilla, Philippines (RA, JK); the Department of Ophthalmology, New York University School of Medicine, New York, New York (MR); the Department of Ophthalmology, University of Illinois, Chicago, Illinois (RVPC); and the Department of Ophthalmology, Mayo Clinic, Rochester, Minnesota (TWO).

Supported in part by the Mayo Clinic, Emory University School of Medicine, and an unrestricted departmental grant from Research to Prevent Blindness, Inc., New York, New York.

The authors have no financial or proprietary interest in the materials presented herein.

Correspondence: Timothy W. Olsen, MD, Department of Ophthalmology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905. E-mail: olsen.timothy@mayo.edu

Received: May 23, 2019
Accepted: August 30, 2019

10.3928/01913913-20190925-02

Sign up to receive

Journal E-contents