Two Sample Policy

Abstract

Background and Objectives: Pretransfusion testing has undergone remarkable changes over the past few decades starting from manual methods to semi-automation to full automation, but sample collection from the right person is out of scope of blood center. Through this study, we tried to address this issue by introducing two-sample policy in our institute. The objective of this study is to emphasize the importance of using two different samples for type and screen and crossmatch. Methods: This is a retrospective study extending over a period of 3.5 years (January 2015–June 2018) in a tertiary care health center. Retrospective data collections were done for the period post implementation of two-sample policy in the institute. As per the two-sample policy, two samples of the same patient would be collected, one at the time of type and screen and other at the time of blood component arrangement request. In case of any discordance between the two samples, incident is generated. Results: Of total 110,949 requests, 85,326 blood/blood components were issued, 56879 of which were for packed red blood cells (PRBCs) and a total of 34,922 PRBC units were released in the study duration. A total of 22/(0.06%) errors were recorded as grouping discrepancy between type and screen sample and sample received for blood component arrangement request. Incidentally, all 22 cases were for the arrangement of PRBC. Out of 22 errors, 20 were near misses and were detected in immunohematology laboratory at the time of compatibility testing and 2/22 (9.09%) errors were actual no harm events. Conclusion: The problem of wrong blood in tube (WBIT) is still a challenge faced by many laboratory persons. The issue can be resolved by training dedicated staff for sample collection. Although this policy has many drawbacks, such as added costs of sample collection and double testing, inconvenience to the patient, delay in providing blood, and increased type O blood use, we would like to emphasize that it would be helpful in error reduction.