Objectives: To identify ways to improve the operation of blood-screening programs and to decrease the inappropriate use of blood by evaluating blood-transfusion practices and blood-banking services in a Kenyan hospital.
Design: Prospective cohort.
Setting: The study was conducted in a rural district hospital in western Kenya between September 1990 and July 1991.
Methods: We collected data on all transfusion requests (blood donation, grouping, HIV screening) and blood recipients (age, sex, diagnosis, and for a 3-month period on the pediatric, maternity, and female wards, admission hemoglobin and outcome).
Results: During the 11-month study period, 799 patients received 927 transfusions: 67% were children < 15 years of age, 27% were adult women and 6% were adult men. Transfusions were often delayed due to reliance on patient-recruited donors. Patients who received blood donated on or after the date of request waited longer for transfusion (median, 3 days) than patients who received blood that had been banked and screened before the request (median, 1 day). Patient-recruited donors had a higher HIV-seropositivity rate than volunteer donors (13.4 and 4.6%, respectively; chi 2 test, P < 0.001). Overall, 47% of pediatric transfusions were classified as inappropriate: 23% did not meet the criteria of having hemoglobin < 5.0 g/dl and clinical evidence of respiratory distress, and 27% were transfused 2 or more days after requested. Among adults, 68% received one unit of blood or less.
Conclusions: Improved laboratory services, reduction of unnecessary transfusions, and increased recruitment of volunteer donors are critical for improving the appropriate and timely use of blood and reducing transfusion-associated HIV transmission.
PIP: Between September 1990 and July 1991, health workers and/or laboratory personnel at Siaya District Hospital in rural western Kenya (about 60 km northwest of Kisumu) gathered data on 799 patients who received 927 blood transfusions, including blood donation, grouping, and HIV screening. Most blood recipients were children (under 15 years old). Only 6% of all recipients were men. Just 30% of transfusions were performed the day of request. Blood donors recruited when it was most needed for survival. Their blood tended to be available 3 days after the request. The volunteer donated blood tended to be available for transfusion the day of request, however, because it had already been banked and screened. Patient-recruited donors were more likely to be HIV infected than volunteer donors (13.4% vs. 4.6%; relative risk = 2.91; p .001). 47% of the pediatric transfusions should not have taken place because 23% of these children did not suffer respiratory distress and their hemoglobin levels were greater than t gm/dl and because 27% received the transfusion 2 days after the day of request. 90% of all adult transfusions were inappropriate (i.e., transfusion of no more than 1 unit of blood or received the transfusion 2 days after the day of request). 30% of blood units that had been banked and screened at the time of request were not transfused until at least 2 days after request. These findings identified those areas which must be targeted to improve the appropriate and timely use of blood and reducing transfusion-induced HIV transmission: reduction of inappropriate transfusions, increased recruitment of volunteer donors, and improved laboratory services.