Out of 9 individual variables found significant in univariate analysis, 5 variables, namely: Age (AOR = 0.93, P < 0.001), total chest (AOR = 1.001, P=0.001), hospital stay (AOR = 1.55, P < 0.001), >30 h ventilator time (AOR = 2.29, P = 0.006), and patients groups as per transfusion criteria
(P < 0.001) were significant in multivariate analysis which indicated that younger age, higher total chest, more time in hospital stay days, and >30 h ventilator time were increasing risk in the patients.
Transfusion guidelines have been available for decades, and both laboratory (1) and hospital accreditation standards (2-4) require or recommend monitoring of transfusion practice according to the institution's defined transfusion criteria
. Since guidelines for transfusion have been available, there has been continual effort to move transfusion practice gradually closer to existing guidelines, while guidelines have also become more refined.
With these potential or actual adverse effects taken together with the large variability in observed prescriptions, it has been determined that when faced with "liberal" or "defensive" traditional transfusion criteria
, more restrictive transfusion practices are adopted along with an enhanced usage of alternative treatments.