Background/Case Studies: Previous studies have shown that donors deferred for low haemoglobin levels are less likely to return to donate than those who complete a successful donation. However, no studies have investigated differences in return dynamics at fixed sites versus mobile blood drives. We analysed the time to return for deferred and completed donors across fixed and mobile collection sites and identified the predictors of return for these groups.
Study
Design/Methods: We analysed all visits made at fixed and mobile blood drives from 2017 to 2022. We used Kaplan-Meier curves to characterise the time to return for donors after completing a donation, being deferred for low haemoglobin levels, or being deferred for other reasons. We compared return times at mobile and fixed sites for both first-time and repeat donors. We developed a LASSO-penalised Cox proportional hazards model using 5-fold cross-validation to investigate the association between donors’ return times and covariates such as age, sex, race, and donation history.
Results/Findings: Out of 4,286,365 visits made by 596,176 donors, 48% were at fixed sites, and 13% resulted in a deferral. Haemoglobin deferrals occurred at 4% of visits to a fixed site and 6% of visits to a mobile drive. After a completed donation, the median time to return from when a donor is eligible was 70 days at fixed sites and 106 days at mobile drives. After a haemoglobin deferral, the median return time was longer: 138 days at fixed sites and 190 days at mobile drives. The lasso-penalised Cox proportional hazards model had an average cross-validated concordance index of 0.67. Male donors (hazard ratio [HR] of 1.21), black South African donors (HR of 1.02) and donating at a fixed site (HR of 1.03) were associated with a shorter time to return. Female donors (HR of 0.97), first-time donors (hazard ratio of 0.84) and haemoglobin deferrals (HR of 0.9) were associated with longer return times. Interaction terms indicated that first-time donors returned later at fixed sites vs mobile (HR of 0.96), and haemoglobin deferred donors returned earlier at fixed sites vs mobile (HR of 1.04). Conclusions: The results suggest that donors at mobile sites exhibited a slower return rate than donors at fixed sites following both a completed donation and a haemoglobin deferral, with the difference being more pronounced for first-time donors. This finding suggests that modifications to deferral policies, such as recommending extended deferral periods for donors with low ferritin levels, may result in a disproportionate decrease in blood collections from mobile blood drives. These insights can aid in forecasting the impact of changes to donor eligibility and deferral policies on blood centre operations, thereby informing decision-making processes and strategic planning efforts.
Importance of research: The research will inform the scrutiny of donor deferral policies. It is innovative due to the focus on overlooked mobile donors, and rigorous modelling techniques. Data from South Africa was chosen because it is a large and diverse cohort, thus increasing the generalizability of the results. Our findings contribute to a better understanding of the dynamics of donor return behavior in different blood collection contexts and inform strategies to improve donor retention and blood supply management.
