Predicting blood donor arrival

BACKGROUND:
Keeping waiting time at blood donation
short is important for making donation a good experience
for the donors and hence to motivate for repeat donations.
At the Blood Bank of Oslo, fixed appointments are used,
and few donors arrive without appointments. On average,
59 percent of scheduled donors arrive, but day-to-day
variations are large. Methods for predicting the number of
donors that will arrive on a given day would be valuable
in reducing waiting times.
STUDY DESIGN AND METHODS:
Information about
candidate explanatory variables was collected for all
appointments made in a 971-day period (179,121
appointments). A logistic regression model for the
prediction of blood donor arrival was fitted.
RESULTS:
Among 18 explanatory variables, the most
important were the time from appointment making to
appointment date; the contact medium used; the donor
age and total number of donations; and the number of noshows,
arrivals, and deferrals during the preceding
2 years. Compared to taking only the average arrival rate
into account, prediction intervals were reduced by
43 percent.
CONCLUSION:
Statistical modeling can provide useful estimates of blood donor arrival, allowing for better planning of donation sessions.