Downloaded from bjo.bmj.com on 6 February 2008
Using assessment of willingness to pay to improve a Cambodian spectacle service J Ramke, A Palagyi, R du Toit and G Brian Br. J. Ophthalmol. 2008;92;170-174; originally published online 22 Jan 2008; doi:10.1136/bjo.2007.122192
Updated information and services can be found at: http://bjo.bmj.com/cgi/content/full/92/2/170
These include:
References
Rapid responses Email alerting service
This article cites 20 articles, 6 of which can be accessed free at: http://bjo.bmj.com/cgi/content/full/92/2/170#BIBL You can respond to this article at: http://bjo.bmj.com/cgi/eletter-submit/92/2/170 Receive free email alerts when new articles cite this article - sign up in the box at the top right corner of the article
Notes
To order reprints of this article go to: http://journals.bmj.com/cgi/reprintform To subscribe to British Journal of Ophthalmology go to: http://journals.bmj.com/subscriptions/
Downloaded from bjo.bmj.com on 6 February 2008
Global issues
Using assessment of willingness to pay to improve a Cambodian spectacle service J Ramke,1,2,3 A Palagyi,1,2,3 R du Toit,1,2,3 G Brian1,2,3 1
The Fred Hollows Foundation (New Zealand), Auckland, New Zealand; 2 The International Centre for Eyecare Education (ICEE), Sydney, Australia; 3 The Vision Cooperative Research Centre, Sydney, Australia Correspondence to: G Brian, 5 Hazelmere Parade, Sherwood 4075, Australia;
[email protected] Accepted 4 November 2007
ABSTRACT Aims: To assess willingness to pay for spectacles in provincial Cambodia, and use this to inform creation of a financially self-sustaining spectacle scheme within a blindness prevention programme. Methods: An interview-based questionnaire was used to elicit willingness to pay for spectacles of all people dispensed spectacles during an outreach refraction service visit to three village health centres in Cambodia. Results: Of 293 people participating in the study, 252 (86%) provided internally valid willingness-to-pay responses from which data were analysed. 76.6% (193) were willing to pay at least KHR1500 (US$0.38) for spectacles. On multivariate analysis, an increased likelihood of being unwilling to pay at least KHR1500 for spectacles in the future was significantly and independently associated with being >60 years old, attending Kor or Svay Teap health centres, not being an income earner in the household and having a household monthly income of less than KHR50 000. There was no association with being vision-impaired, this being the first eye examination, occupation, not having motorised transport or previous spectacle wear. If the potential willingness to pay had been converted to actual on the day, there would have been a 28.0% increase in revenue, and a greater than fivefold increase in profit, for the spectacle scheme. Conclusions: Willingness-to-pay data may be useful for price-setting and developing a subsidisation protocol for poorer consumers that will ensure financial accessibility for all and financial sustainability for the provision of spectacles.
The reduction of blindness and vision impairment due to uncorrected refractive error is often described as simple,1 with spectacles cited as a cost-effective treatment.1 2 However, documented examples of successful financially self-sustaining spectacle programmes within national strategies for blindness prevention are lacking. Although management and financing systems to support the equitable and affordable provision of medicines within lowresource countries have been described,3–6 these have not been reported for refraction services or the dispensing of affordable spectacles. Consequently, government and non-government eye-care services aiming to provide equitable and affordable spectacle correction at the community level have little published experience to draw upon when attempting to initiate strategies for improvement. A comprehensive refractive error correction programme that attempts financial sustainability through cost-recovery mechanisms needs to consider much, including the implications of payment for spectacles. To ensure equitable access and utilisation of good quality, functional, durable spectacles, 170
regardless of consumer financial circumstances, such a programme needs to instigate cross-subsidisation, maximising profit from those requiring and able to pay for spectacles to subsidise those unable to afford to pay the full or any price. To this end, willingness to pay for spectacles requires quantification for the population to be targeted by the programme. An individual’s willingness to pay for healthcare may be influenced by perceived need and value of treatment,7 and prior experience and knowledge of an intervention.8–11 Willingness to pay for treatment has been reported for cataract surgery and trachoma control.7 10 11 However, apart from a study in Timor-Leste,12 willingness to pay for spectacles seemingly remains uninvestigated. This paper reports willingness to pay for spectacles in provincial Cambodia. The Kingdom of Cambodia has a population of 13.7 million:13 most (81%) reside in rural areas, and 40% of this rural population live below the poverty line.14 The country’s most populated province, Kampong Cham, is 120 km north east of its capital, Phnom Penh, along the Mekong River. Although the province’s two eye units provide cataract surgery, there is no permanent refraction or spectacle service. Spectacles are available from market-vendors in the urban centres, or through an infrequently visiting outreach refraction service partnering with the provincial eye units. At present, this service is reliant on external donation of spectacle stock and expenses, with loss of donor support likely to place continuation in jeopardy. Financial autonomy would improve the likelihood of sustainability. An appraisal is made of the financial health of the Kampong Cham outreach spectacle-dispensing programme. Willingness-to-pay information is then used to suggest how this may be improved without compromising equitable utilisation by all.
METHODS Approval for this study was granted by the Cambodian Ministry of Health National Prevention of Blindness Committee. During May 2004, the outreach refraction service worked in the health centres of three typical Kampong Cham province villages: Kor, Svay Teap and Ta Ong. All people acquiring spectacles at these clinics were invited to participate in this study. Each participant had undergone vision screening and refractive assessment, and been issued with spherical ready-made spectacles. This refraction and spectacle information was made available to the study team. An interview-based questionnaire, developed in consultation with the Cambodian Optometry Association to ensure local relevance, was delivered by a non-associated local enumerator. Br J Ophthalmol 2008;92:170–174. doi:10.1136/bjo.2007.122192
Downloaded from bjo.bmj.com on 6 February 2008
Global issues Communications occurred in Khmer. Participant socio-demography and history of spectacle wear were ascertained. The outreach service delivery team had either prepurchased the spectacles from the local market or manufactured them. As per customary practice, the starting price for spectacles was KHR5000, but, at the discretion of the refractionist, and without any guidelines, a lower amount was accepted from those deemed unable to afford this price. Willingness to pay for spectacles was assessed using a binary-with-follow-up (BWFU) technique. The questionnaire price increments used were designed with an anticipated increase in the price of spectacles from KHR5000 to KHR6000 in mind. These began at KHR6000 (approximately US$1.50) and moved to either KHR8000 or KHR4000, depending on a positive or negative response to KHR6000. Once this line of questioning was completed, participants were asked the maximum price they were willing to pay for spectacles. Internal response validity was determined by ensuring the stated maximum price willing to pay was consistent with the response obtained from specific price increment selection and actual price paid on the day. The maximum willingness-to-pay data were analysed to provide recommendations for improved financial sustainability of the spectacle service. As the lowest cost of acceptable quality spectacles at the market in Phnom Penh is KHR1500, a specific investigation of the willingness to pay this amount was carried out. This was done to inform recommendations for a subsidisation protocol with the aim of, at a minimum, ensuring cost recovery of the spectacle service. Univariate analyses were performed using Fisher’s exact test or chi square to provide a crude association of socio-economic characteristics with unwillingness to pay KHR1500. The strength of association of factors with binary outcomes was summarised using odds ratios (OR) and 95% confidence intervals (95% CI). Multiple logistic regression models were used to evaluate the independent contributions of demographic factors and vision status. Statistical significance was set at 5%. SPSS Version 12 was used for statistical analysis.
(1.4%) had both astigmatism and anisometropia, and two people (0.7%) had a spherical prescription greater than 6DS. In total, this accounted for 9.2% of participants who would likely benefit from custom-made spectacles. The remaining 90.8% of participants would likely attain adequate vision with spherical correction of equivalent power in each eye. Ready-made spectacles between ¡6.0DS would be sufficient to correct them. Ready-made spectacles of ¡3.0DS would correct 78.9% of the sample.
Future willingness to pay less than KHR1500 for spectacles Using maximum amount willing to pay data from the 252 internally valid responses, 193 (76.6%) participants were willing to pay at least KHR1500 for spectacles in the future. An increased likelihood of being unwilling to pay this was associated with female gender, being 60 years of age or above, attending Kor or Svay Teap health centres, being vision impaired (unaided vision less than 6/18 in the better eye), subsistence farming or unemployment, not having motorised transport, and having a household monthly income of less than KHR50 000 (table 2). There was no association with this being the first eye examination, being an income earner in the household or previous spectacle wear. However, multivariate analysis suggested that an increased likelihood of being unwilling to pay at least KHR1500 for spectacles in the future was significantly and independently associated with being 60 years of age or above, attending Kor or Svay Teap health centres, not being an income earner in the household and having a household monthly income of less than KHR50 000 (table 3). There was no association with being vision impaired, this being the first eye examination, occupation, not having motorised transport or previous spectacle wear. Gender was forced into the model because it was not known if age and gender distribution in this sample were similar to the Cambodian population. The goodness of fit (Hosmer and Lemeshow’s) of the developed model was non-significant (p = 0.99), suggesting that the model performs poorly at predicting willingness to pay at least 1500 KHR.
RESULTS Sample
Financial performance of the spectacle service: comparison of price paid for spectacles and future willingness to pay
All 293 people invited to participate in the survey agreed to do so. Ninety-nine (33.8%) were from Kor, 102 (34.8%) from Svay Teap and 92 (31.4%) from Ta Ong. Internally valid responses for the willingness-to-pay questions were provided by 252 (86.0%) participants. Only these were included in the willingness-to-pay analysis. Participants at Kor Health Centre were more likely than those from elsewhere to give a valid response (OR 2.8, 95% CI 1.2–6.4). Otherwise, there were no differences between the groups giving valid and invalid responses, which most commonly (88%) were stating a maximum amount greater than the highest elicited by the binary questions. Participant characteristics are presented in table 1.
In total, the 299 spectacles dispensed to the 293 study participants were estimated to have cost the service KHR860 000 (table 4). The total and average cost of spectacles to the service were highest for those obtained in Kor and lowest in Ta Ong. However, the total and average income from each pair of spectacles were lowest in Kor and highest in Ta Ong. A loss was made in Kor, with profits in Svay Teap and Ta Ong. The total sales revenue was KHR910 800, for a profit of KHR50 800 (average KHR170/pair). If the spectacle nominal price to consumers had been raised to KHR6000 and those who stated they were willing to pay this amount actually did so, an additional KHR255 200 would have been taken on the day. This would have been a 28.0% increase in revenue, and a greater than fivefold increase in profit (table 4).
Refraction and spectacles Immediately prior to interview, 25.9% (76/293) and 72.0% (211/ 293) of participants had been dispensed ready-made spectacles for distance and presbyopia, respectively. A further 2.0% (6/293) had received two pairs of spectacles: one for each of distance and near vision. Of the 299 spectacles dispensed, 21.1% (63/299) were greater than ¡3.0DS and 1.3% (4/299) were greater than ¡6.0DS. Participant refraction data showed 16 (5.5%) had anisometropia .1DS, five (1.7%) had astigmatism .1DC in either eye, four Br J Ophthalmol 2008;92:170–174. doi:10.1136/bjo.2007.122192
DISCUSSION There is no single commonly accepted method of measuring willingness to pay. Environmental and cultural differences of sample populations prevent this.8 15–17 Further, only a few studies have measured the conversion of stated to actual willingness to pay.8 15 18 In Nigeria, compared with BWFU, the use of open-ended questions has been found to yield more valid estimates of actual 171
Downloaded from bjo.bmj.com on 6 February 2008
Global issues Table 1 Characteristics of participants Category
Total participants n = 293
Gender 112 Male 181 Female Age (years) 12 14–39 92 40–49 112 50–59 77 60+ Domicile 99 Kor 102 Svay Teap 92 Ta Ong Occupation 84 Paid employment 131 Subsistence farming 71 Unemployed 7 Other Household monthly income (KHR)* 40 ,10 000 70 10 000–49 999 50 000–99 999 28 116 100 000–499 999 17 500 000+ Income earner 127 Yes 166 No Proxy of household wealth{ 24 No vehicle 145 Bicycle 122 Moto/car Vision impairment (unaided vision worse than 6/18 in the better eye) 221 Not visually impaired 72 Visually impaired Previous eye examination 239 No 54 Yes Previous spectacle wear 154 No 139 Yes Satisfaction with previous spectacles 79 Satisfied 60 Not satisfied
n (%)
Participants with invalid willingness-to-pay responses n = 41
Participants with valid willingness-to-pay responses n = 252
13 28
(11.6) (15.5)
99 153
(88.4) (84.5)
2 9 22 8
(16.7) (9.8) (19.6) (10.4)
10 83 90 69
(83.3) (90.2) (80.4) (89.6)
7 16 18
(7.1) (15.7) (19.6)
92 86 74
(92.9) (84.3) (80.4)
12 21 8
(14.3) (16.0) (11.3) 0
72 110 63 7
(85.7) (84.0) (88.7) (100.0)
3 13 4 20 1
(7.5) (18.6) (14.3) (17.2) (5.9)
37 57 24 96 16
(92.5) (81.4) (85.7) (82.8) (94.1)
22 19
(17.3) (11.4)
105 147
(82.7) (88.6)
5 18 18
(20.8) (12.4) (14.8)
19 127 104
(79.2) (87.6) (85.2)
34 7
(15.4) (9.7)
187 65
(84.6) (90.3)
35 6
(14.6) (11.1)
204 48
(85.4) (88.9)
17 24
(11.0) (17.3)
137 115
(89.0) (82.7)
19/79 5/60
(24.1) (8.3)
60/79 55/60
(75.9) (91.7)
*22 unknown values; {2 missing.
willingness to pay.15 However, elsewhere, non-binary approaches have been shown to elicit invalid responses.19 20 Given that purchasing goods in Cambodia most commonly involves bargaining, as a truncation of the bidding game,15 BWFU was believed more appropriate for the Cambodian survey population. It also reduces the range within which respondents’ measured willingness-to-pay sits. This enables more rapid assessment and reduces the sample size required.15 19 21 In this Cambodian study, 86% of participants’ responses were internally valid, indicating that this method of questioning was well understood and therefore appropriate in this context. The potential introduction of bias through selection of the starting price increment with BWFU has been identified as a limitation of this willingness-to-pay technique.15 22 In this study, a related bias may have arisen because although the hypothetical BWFU questioning commenced at KHR6000, respondents had earlier had an actual opportunity to pay KHR5000. 172
Reliability of willingness-to-pay data may be affected where a good has no demonstrated personal significance, or minimal perceived need, to the respondent.11 Although it is assumed that the demonstration of vision improvement with spectacles and subsequent decision of the respondent to buy the appliance, both prior to data collection during this study, overcomes this, the authors recognise this assumption may not hold firm for all respondents. Also, related to this, there is an assumption that current and future willingness to pay would be identical, which ignores the fact that the subsequent experience of spectacle wear by first-time purchasers may alter their intentions. More reliable analysis of willingness to pay in terms of actual need for spectacles and vision improvement associated with their use may be supported by data of unaided and aided visual acuity, which were not obtained during this study. Further, given the seasonal variation in disposable income of this largely agrarian sample, hypothetical willingness-to-pay information gathered during the study period Br J Ophthalmol 2008;92:170–174. doi:10.1136/bjo.2007.122192
Downloaded from bjo.bmj.com on 6 February 2008
Global issues Table 2 Associations with future maximum willingness to pay for spectacles being less than KHR1500 (US$0.38) Category Gender Male Female Age (years) ,60 60+ Domicile Ta Ong Svay Teap Kor Vision status Not visually impaired Visually impaired* Occupation{ Paid employment Subsistence farming Unemployed Proxy of household wealth{ No vehicle Bicycle Moto/car Household monthly income (KHR)1 ,10 000 10 000–49 999 50 000+ Previous eye examination No Yes Previous spectacle wear No Yes Income earner Yes No Total
Total n = 252 valid responses
Not willing to pay KHR1500 n (%)
Willing to pay KHR1500 or more n (%)
Odds ratio of not being willing to pay KHR1500 OR (95% CI)
99 153
15 (15.2) 44 (28.8)
84 (84.8) 109 (71.2)
1.0 2.26 (1.19 to 4.30)
183 69
30 (16.4) 29 (36.2)
153 (83.6) 40 (63.8)
1.0 3.70 (2.00 to 6.84)
74 86 92
2 (2.7) 10 (11.6) 47 (51.1)
72 (97.3) 76 (88.4) 45 (48.9)
1.0 4.74 (1.12 to 19.8) 37.6 (9.57 to 146.3)
187 65
32 (17.1) 27 (41.5)
155 (82.9) 38 (58.5)
1.0 3.44 (1.85 to 6.40)
72 110 63
4 (5.6) 25 (22.7) 30 (47.6)
68 (94.4) 85 (77.3) 33 (52.4)
1.0 5.00 (1.73 to 14.4) 22.7 (7.73 to 66.2)
19 127 104
8 (42.1) 46 (36.2) 4 (3.8)
11 (57.9) 81 (63.8) 100 (96.2)
14.6 (3.92 to 54.0) 14.2 (5.10 to 39.4) 1.0
37 57 136
22 (59.5) 10 (17.5) 8 (5.9)
15 (40.5) 47 (82.5) 128 (94.1)
29.5 (11.2 to 77.9) 5.21 (2.09 to 13.0) 1.0
204 48
51 (25.0) 8 (16.7)
153 (75.0) 40 (83.3)
1.67 (0.74 to 3.72) 1.0
137 115
38 (27.7) 21 (18.3)
99 (72.3) 94 (81.7)
1.72 (0.94 to 3.13) 1.0
105 147 252
19 (18.1) 40 (27.2) 59 (23.4)
86 (81.9) 107 (72.8) 193 (76.6)
1.0 1.69 (0.92 to 3.11)
*Unaided vision worse than 6/18 in the better eye; {7 ‘‘other occupation’’ not considered; {2 missing; 122 unknown values.
may not necessarily be an accurate predictor of actual willingness to pay at some other time during the harvest cycle.
Table 3 Multivariate model of future maximum willingness to pay for spectacles being less than KHR1500 (US$0.38) Characteristic
n
Gender 99 Male 153 Female Age (years) 183 ,60 69 60+ Domicile 74 Ta Ong 86 Svay Teap 92 Kor Household monthly income (KHR) 136 50 000+ 57 10 000–49 999 37 ,10 000 Income earner Yes No
105 147
Odds ratio (95% CI) 1.0 1.9 (0.8 to 4.7) 1.0 6.3 (2.5 to 16.2) 1.0 2.9 (0.5 to 15.9) 33.4 (6.6 to 169.8)
1.0 6.9 (2.5 to 19.0) 11.0 (3.5 to 34.2) 1.0 2.3 (1.0 to 5.7)
Br J Ophthalmol 2008;92:170–174. doi:10.1136/bjo.2007.122192
A financially self-sustaining programme achieving equitable utilisation of spectacles within a blindness prevention programme needs to be anchored in the realities in which it operates. The costs of inventory, service provision, monitoring, equipment acquisition and maintenance need to be balanced against considerations such as volume and unit cost to consumers. As is generally the case, the Kampong Cham spectacle programme has developed without an appreciation of these. For example, the spectacle price determined by the refractionist for each participant at the three different health centres did not take into account either the cost of the spectacles or the willingness to pay of the health-centre attendees. This resulted in revenue loss at one health centre and overall bare cost-recovery on the cost of inventory for the three sites combined, not including transport and equipment costs or wages of the refractionist. Analysis of the Kampong Cham willingness-to-pay data indicates an opportunity to increase revenue. Improved consideration of the cost of different spectacle stock, and planning based on knowledge of willingness to pay within different provincial locations, could improve cost recovery within the programme. In so doing, a subsidisation system could be set up for those consumers unable to pay the full price, the current reliance on external donations of inventory could be decreased, and local 173
Downloaded from bjo.bmj.com on 6 February 2008
Global issues Table 4 Revenue and profit/loss: actual and potential for spectacles sales at three health centres in Kampong Cham Health centre
No. of spectacles dispensed Total revenue from spectacles sold on the day (KHR) Cost to the spectacle service of spectacles dispensed (KHR) Profit/(loss) (KHR) Mean profit/(loss) per pair (KHR) Revenue if nominal price/pair were raised to KHR6000{ (KHR) (% increase) Profit/(loss) (KHR) (% increase) Mean profit/(loss) per pair (KHR)
All locations
Kor
Svay Teap
Ta Ong
299* 910 800 860 000 50 800 170 1 166 000 (28.0) 306 000 (502.4) 1023
100 150 300 334 000 (183 700) (1837) 226 000 (50.4) (108 000) (41.2) (1080)
106 315 000 300 000 15 000 142 427 500 (35.7) 127 500 (750.0) 1203
93 445 500 226 000 219 500 2360 512 500 (15.0) 286 500 (30.5) 3080
*82 distance and 217 near spectacles dispensed to the 293 participants (six of whom received both distance and near corrections). {Assumptions: (1) that the service use locally purchased spectacles in each location rather than source all from the supplier providing the cheapest appliances of acceptable standard (KHR1500); (2) that those not willing to pay KHR6000 and those not giving valid willingness-to-pay responses, would pay as they had done so on the day of the survey; (3) that those who received their spectacles at no cost would do so if appropriate according to assumption (2).
autonomy of the programme could be increased. Profit could be used to provide additional outreach refraction services and overcome barriers to utilisation experienced by vulnerable groups, further improving equity of the service. Assuming stated willingness to pay converts to actual, a price increase from KHR5000 to 6000 would increase profit (table 4). Such a price elevation in isolation will, however, likely have a detrimental effect on the programme’s equity of utilisation, unless strategies are adopted to accommodate the paying capacity of those who cannot meet the new price. This could involve a subsidisation protocol that includes a lower price that recovers the cost of the spectacles (KHR1500) from those who can afford them, and no charge for those unable to provide any payment. The characteristics of the 23.4% of people unwilling to pay KHR1500 (tables 2 and 3) could be used to determine who should be the beneficiaries of subsidisation. Such a strategy could also be enhanced by introducing a range of spectacles of differing cost price. A tiered pricing structure could then be devised for the service, with the patients self-selecting their spectacles of choice based on personal preference and capacity to pay. Despite the initial training and ongoing supervision required for such a pricing apparatus, this system would likely be more time-efficient and financially effective than the current ad hoc decision-making associated with discretionary reduction of price for those claiming inability to pay. Multivariate analysis suggests that those aged 60 years or older, with a monthly household income of less than KHR10 000 (or perhaps of less than KHR50 000), especially if they do not contribute to this income, should be considered for subsidisation in this Cambodian province. Residents in less affluent areas, for example Kor, may also need to be considered. This operational research to inform creation of a financially self-sustaining system of spectacle provision has recognised weaknesses. However, it illustrates the benefits of such activities. An apparently acceptable change in pricing would likely generate more revenue. Profits could be further increased with a more considered approach to stock spectacle purchase by the service. The enhanced profit may then be used to support an informed standardised subsidisation protocol, ensuring financial accessibility for all and financial sustainability for the provision of spectacles within a blindness-prevention programme.
Competing interests: None. Patient consent: Written informed consent (or verbal consent if there were literacy barriers) was obtained from each participant.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Acknowledgements: The authors gratefully acknowledge the assistance of the Cambodian Optometry Association and IRIS Cambodia.
21.
Funding: This work was in part financially supported by the Australian Federal government through the Co-operative Research Centres Program (Vision CRC).
22.
174
Dandona R, Dandona L. Refractive error blindness. Bull World Health Organ 2001;79:237–43. Coleman AL, Yu F, Keeler E, et al. Treatment of uncorrected refractive error improves vision-specific quality of life. J Am Geriatr Soc 2006;54:883–90. Van der Veen F, Fransen L, Drugs for STD management in developing countries: choice, procurement, cost, and financing. Sex Transm Infect 1998;74(Suppl 1):166–74S. Peterson S, Assey V, Forsberg BC, et al. Coverage and cost of iodized oil capsule distribution in Tanzania. Health Policy Plan 1999;14:390–9. Haak H, Hogerzeil HV. Essential drugs for ration kits in developing countries. Health Policy Plan 1995;10:40–9. Soucat A, Levy-Bruhl D, De Bethune X, et al. Affordability, cost-effectiveness and efficiency of primary health care: the Bamako Initiative experience in Benin and Guinea. Int J Health Plann Manage 1997;12(Suppl 1):81–108S. Frick KD, Lynch M, West S, et al. Household willingness to pay for azithromycin treatment for trachoma control in the United Republic of Tanzania. Bull World Health Organ 2003;81:101–7. Onwujekwe O, Chima R, Shu E, et al. Hypothetical and actual willingness to pay for insecticide-treated nets in five Nigerian communities. Trop Med Int Health 2001;6:545–53. Guyatt HL, Ochola SA, Snow RW. Too poor to pay: charging for insecticide-treated bednets in highland Kenya. Trop Med Int Health 2002;7:846–50. Shrestha MK, Thakur J, Gurung CK, et al. Willingness to pay for cataract surgery in Kathmandu valley. Br J Ophthalmol 2004;88:319–20. Lewallen S, Geneau R, Mahande M, et al. Willingness to pay for cataract surgery in two regions of Tanzania. Br J Ophthalmol 2006;90:11–3. Ramke J, du Toit R, Palagyi A, et al. Correction of refractive error and presbyopia in Timor-Leste. Br J Ophthalmol 2007;91:860–6. National Institute of Statistics. Cambodia Inter-Censal Population Survey (CIPS). Phnom Penh: National Institute of Statistics, Ministry of Planning, Kingdom of Cambodia, 2004. The World Bank. World Development Indicators 2005. Washington, DC: The World Bank, 2005. Onwujekwe O, Uzochukwu B. Stated and actual altruistic willingness to pay for insecticide-treated nets in Nigeria: validity of open-ended and binary with follow-up questions. Health Econ 2004;13:477–92. Onwujekwe O, Hanson K, Fox-Rushby J. Do divergences between stated and actual willingness to pay signify the existence of bias in contingent valuation surveys? Soc Sci Med 2005;60:525–36. Onwujekwe O. Searching for a better willingness to pay elicitation method in rural Nigeria: the binary question with follow-up method versus the bidding game technique. Health Econ 2001;10:147–58. Blumenschein K, Johannesson M. Hypothetical versus real willingness to pay in the health care sector: results from a field experiment. J Health Econ 2001;20:441–58. Smith RD. The discrete-choice willingness-to-pay question format in health economics: should we adopt environmental guidelines? Med Decis Making 2000;20:194–206. Jones-Lee MW, Hammerton M, Philips PR. The value of safety: results of a national sample survey. Econ J 1985;95:49–72. Cameron TA, Quiggin J. Estimation using contingent valuation data from a ‘‘Dichotomous choice with follow-up’’ questionnaire. J Environ Econ Manage 1994;27:218–34. Herriges JA, Shogren JF. Starting point bias in dichotomous choice valuation with follow-up questioning. J Environ Econ Manage 1996;30:112–31.
Br J Ophthalmol 2008;92:170–174. doi:10.1136/bjo.2007.122192
