Don't Stop Me Now: the Impact of Waiting to Donate Blood
Motivation:Blood donation is an ideal setting to study how costs affect prosocial behavior. The behavior itself is costly, so we know we are dealing with donors that are willing to incur in costs to perform an altruistic action. What we don’t know is whether they react negatively to additional costs, or if they completely absorb whatever costs the system imposes on them. We focus on the time cost of donation. There are three steps in the donation process: registration, clinical triage and blood collection, if the donor is approved for donation.
Aims: In this paper I analyze the impact of time spent to donate blood. I estimate how the time spent affects the probability and timing of a second donation by new donors. Furthermore, I estimate how waiting times affect donors’ likelihood of switching donation site.
Data: The dataset is a panel of all new donors enrolled in the Lisbon center of the Portuguese Blood Institute, between January 2008 and December 2012 (N=61189). It contains information regarding health (triage and lab test results), the donation (site, staff, blood drive, waiting times) and sociodemographic variables. We study the behavior of donors at all blood donor centers and all blood drives held in the region, over the sample period. Moreover, the triage outcomes allow us to study how waiting times affect differently potential donors that are accepted or denied the possibility to donate blood.
Methods: I use a probit regression model to study how the time spent at each stage of the donation process affects the probability of a new donor returning to donate blood, in the future. Furthermore, I use a proportional hazards model to estimate the impact of these time costs on the duration of the period until the donor returns to donate blood.
Results: I find that higher waiting times make it less likely for first-time donors to donate again, controlling for both donor-specific variables and and donation site fixed effects. An increase of 10 minutes in waiting time until triage results in a decrease in the probability of donor retention of 0.6%. The effect is highly nonlinear, with the most affected donors being the ones in the upper quartile of the distribution. For approved donors, time until triage is more relevant than the time spent waiting for the actual blood collection, after being cleared for donation. Furthermore, new donors at blood drives react more negatively to waiting times than new donors at blood donor centers. This seems to indicate that the latter are more willing to incur in costs of donation, not only the time cost but also the transportation cost to a blood donor center.