Monday, January 5, 2015

Blood donation, in China

Here's a paper on incentivizing blood donation, with data from China.

Solving Shortage in a Priceless Market: Insights from Blood Donation
by Tianshu Sun, Susan Feng Lu, and Ginger Zhe Jin
November 2014

Blood shortage is common in many countries but it cannot be solved by price adjustment given the WHO recommendation for 100% unpaid voluntary donation. In this paper, we evaluate two non-price methods that blood banks often use to address shortage. The first method is informing existing donors of the current shortage via mobile message and encouraging them to donate voluntarily. The second method is asking the patient’s family or friends to donate in a family replacement program. Using 472,342 individual donation records from a large Chinese blood bank, we show that both methods are effective in addressing blood shortage in the short run but the two methods target different audiences and therefore have different implications for total blood supply. 

In the conclusion, the authors note that shortage messages reach existing donors, while family replacement programs primarily draw in new donors.

"Overall, the back-of-envelope calculation suggests that shortage message can be used in places
where the donor population is large and the shortage is small. In comparison, FR could be more
useful when the donor population is small and the shortage is severe. In this sense, our data suggests a more optimistic picture for FR than the WHO recommendation. However, in a society with a low donation rate (which could be the reason for severe shortage to begin with), most FR donors will be no-history FR donors and the FR treatment may discourage voluntary donation in
the long run by generating either distrust or crowd-out. Although a broader introduction of FR
can bring more blood supply in the short run, it may exacerbate shortage problem in the long run.
Like the WHO, we reach a cautionary conclusion for the FR program but for a reason different
from quality concerns. "

1 comment:

Cal Yh Z said...

I read this paper. The regression model intended to isolate FR method's selection effect is sophisticated. But I concern how its proxy ratio will affect its final result. (As the authors have mentioned, the coefficients are measured in probabilities and the SE are small. Even small measurement error will impose great influence on the result?) Another concern is about crowd-out effect of FR (page 5). As I know, blood can be regenerated pretty quickly after donation, which won't affect potential later donation for family. The reasoning "saving one's blood for family later" might not be true?