LAUSR

A recent article by Stubbs et al. in Transfusion suggested that remuneration could be offered to platelet donors. This suggestion was addressed in a recent Editorial by Dodd et al. in Transfusion “in the context of recent unpublished data on donor motivation and hospital perceptions about paid platelet donors.” Dodd et al. concluded that “donors are motivated by altruism and convenience and not financial rewards.” However, the survey data that Dodd et al. cite in support of this conclusion might give a misleading view of the efficacy of offering remuneration to platelet donors. First, the preferences that persons state that they would have in hypothetical situations presented to them in surveys do not necessarily predict the actions that they will perform when they are actually faced with the choice that they were hypothetically presented with. It has been suggested that the difference can be “largely explained by a model where people derive utility from a positive selfimage associated with morally commendable behavior.” This model has received support from work performed by Johansson-Stenman and Svedsäter. They informed one group of experimental participants that the financial reward that they would receive for participation depended on how they chose to allocate money between their financial reward for participation, receiving a restaurant voucher, or a donation to the WWF. Another group was offered a fixed fee for participation. The former group was less likely to allocate money to the WWF than the latter. There was no significant difference between the amount of money each group allocated to the restaurant voucher. These results support the hypothesis that persons would in hypothetical situations overstate their willingness to incur costs to act in accord with their ethical views. Such “hypothetical bias” toward morally commendable behavior has also been exhibited in surveys concerning blood donation. In the United Kingdom, the National Health Service Blood and Transplant (NHSBT) commissioned an iterative survey of blood donors (consisting of a pilot survey and a main survey) to determine the effects that possible service changes would have on the frequency of blood donation. The survey was designed so that donors' stated preferences for alternative service changes (measured by the frequency by which they stated that they would donor blood under them) could be anchored to their revealed preferences for donation frequency over the prior 12 months. It was found that male respondents exhibited significant hypothetical bias toward donation. Their revealed preference data showed an average number of 2.3 donations p.a. in the pilot survey and 2.24 p.a. in the main survey; the discrepancy between this data and their stated preferences was 0.71 in the pilot survey and 0.55 in the main survey. Second, the survey that Dodd et al. cite to address the effects that the introduction of incentives will hypothetically have on current, lapsed, and potential donors' motivations to donate is marred by selection bias. The set of persons identified as “potential donors” in this survey included only those who had “donated red blood cells within the previous 6 months” and who had “donated 2 or more red cell procedures in each of the previous 2 years.” That only 18% of potential platelet donor respondents said that incentives would lead them to donate platelets where the only “potential platelet donors” surveyed were recent blood donors tells us nothing about the effect that incentives would have on the willingness of people in general to donate. Dodd et al. also cite a recent review by Chell et al. that showed that the data on the efficacy of cash payment for blood donation were inconclusive. However, Chell et al. also noted that “overall” the use of incentives resulted in a “net benefit” and that they were “particularly relevant for younger individuals or those early in their donor career.” This supports the view of Stubbs et al. that platelet donors should be paid. Given the importance and complexity of this issue, we should secure further data on the effects of remuneration. Since persons' stated preferences might differ from their revealed preferences, the best way to do so would be through limited trials of paying for platelets.