Long-term return behavior of Chinese whole blood donors

Transfusion. 2013 Sep;53(9):1985-91. doi: 10.1111/trf.12142. Epub 2013 Mar 5.

Abstract

Background: It is important to understand donor return behavior to maintain sufficient numbers of blood donors in developing countries where blood supplies are often inadequate.

Study design and methods: A total of 54,267 whole blood (WB) donors who donated between January 1 and March 31, 2008, at the five blood centers in China were followed for 2.5 years. Logistic regression was conducted to identify factors associated with their return behavior. A recurrent-event Cox proportional-hazard model was used to evaluate the overall effect of demographic variables and return behavior among first-time donors.

Results: Donors with previous donation history were more likely to return and the number of previous returns was positively associated with future return (odds ratios, 3.31, 4.82, and 8.16 for one, two to three, and more than three times compared to none). Thirty-four percent of donors (first-time donor, 21%; repeat donor, 54%) made at least one return donation, with 14% returning in the first 9 months. The multivariable logistic regression model for all WB donors and the Cox proportional hazard model for first-time donors showed consistent predictors for return: female sex, older age (≥ 25 years), larger volume (300 or 400 mL), and donating in satellite collection site.

Conclusion: Encouraging first-time donors to make multiple donations is important for keeping adequate blood supply. The finding that first-time and repeat donors shared the same predictors for return indicates that retention strategies on repeat donors may be effective on first-time donors. Studies on motivators and barriers to return are needed, so that successful retention strategies can be tailored.

MeSH terms

  • Adult
  • Blood Donors / psychology*
  • Blood Donors / statistics & numerical data*
  • China
  • Female
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Proportional Hazards Models