Associations of health status with subsequent blood donor behavior—An alternative perspective on the Healthy Donor Effect from Donor InSight

INTRODUCTION In donor health research, the 'Healthy Donor Effect' (HDE) often biases study results and hampers their interpretation. This refers to the fact that donors are a selected 'healthier' subset of a population due to both donor selection procedures and self-selection. Donors with long versus short donor careers, or with high versus low donation intensities are often compared to avoid this HDE, but underlying health differences might also cause these differences in behaviour. Our aim was to estimate to what extent a donor´s perceived health status associates with donation cessation and intensity. METHODS All active whole blood donors participating in Donor InSight (2007-2009; 11,107 male; 12,616 female) were included in this prospective cohort study. We performed Cox survival and Poisson regression analyses to assess whether self-reported health status, medication use, disease diagnosed by a physician and recently having consulted a general practitioner (GP) or specialist were associated with (time to) donation cessation and donation intensity. RESULTS At the end of 2013, 44% of the donors in this study had stopped donating. Donors in self-rated good health had a 15% lower risk to stop donating compared to donors in perceived poorer health. Medication use, disease diagnoses and consulting a GP were associated with a 20-40% increased risk to stop donating and a lower donation intensity, when adjusting for age, number of donations and new donor status. Both men and women reporting good health made on average 10% more donations. CONCLUSION Donors with a "good" health status were less likely to stop donating blood and tended to donate blood more often than donors with perceived poorer health status. This implies that the HDE is an important source of selection bias in studies on donor health and this includes studies where comparisons within donors are made. This HDE should be adjusted for appropriately when assessing health effects of donation and donors' health status may provide estimates of future donation behavior.