Quality management in the transfusion service: case studies in process improvement.

BACKGROUND: Laboratory-based quality improvement (QI) initiatives can improve clinical outcomes and patient safety. STUDY DESIGN AND METHODS: We present three cases of QI that impact processes from the transfusion service (TS) laboratory to the patient's bedside. RESULTS: Case 1 was event discovery reporting (EDR). We were able to reduce our biologic product deviation reports from 41 (17%) of 238 EDRs to only 19 (7%) of 272 (p < 0.01) EDRs after implementation of a QI workflow process. Case 2 was antibody evaluation before elective surgery. We implemented process improvement strategies: 1) surgical safety checklist with confirmation of type-and-screen completion and antibody evaluation before patients can proceed to surgery; 2) specimen retention policy of 30 days to allow advance testing; and 3) daily review to identify specimens needed on day of surgery. After intervention, only 7 (0.3%) of 2298 patients required antibody evaluation on day of surgery, compared to 65 (0.75%) of 8656 patients (p < 0.01) before intervention. Case 3 was wrong blood in tube (WBIT). We have a two-specimen requirement for blood type verification before transfusion. To determine whether trauma patients should be exempted, we reviewed WBIT errors. Six WBIT errors were from the emergency department (an error rate of 1:400) and nine WBIT specimens were institution-wide. Three patients were transfused after correction of the WBIT error. Based on this analysis, our institution agreed that no clinical units shall be exempted from our policy. CONCLUSION: Successful QI in the TS improves processes that promote efficiency, effectiveness, and patient safety.

Impact of cytomegalovirus (CMV) antibody reflex testing in the transfusion service on management of CMV-seronegative blood inventory.

BACKGROUND: Our goal is to minimize unnecessary cytomegalovirus (CMV)-seronegative blood transfusion to preserve the CMV-seronegative blood inventory for patients who are identified as CMV seronegative. STUDY DESIGN AND METHODS: We implemented a CMV antibody reflex testing protocol for patients who require CMV-compatible blood but in whom a CMV serostatus is unknown (coded as CMVT in our computer system). A solid-phase red blood cell (RBC) adherence antibody detection system was validated to detect CMV antibodies in plasma samples (received for ABO/Rh type and RBC antibody screen) with acceptable sensitivity and specificity. We evaluated the impact of this CMV antibody reflex testing on the management of RBC and platelet (PLT) inventory for patients requiring CMV-compatible blood. RESULTS: Over a 16-month period, implementation of CMV antibody reflex testing identified 361 (34%) of 1063 previously CMV-untested patients who required CMV-compatible blood and who were CMV seronegative. We observed a 75% decrease in the number of CMVT patients in our data base from 190 per month before implementation to 57 at 16 months postimplementation. Consequently we reevaluated the percentage in our blood inventory of CMV-seronegative units required while potentially saving 1234 CMV-seronegative blood products (835 RBCs and 399 PLTs) each month. CONCLUSION: A strategy of performing CMV antibody reflex testing in the transfusion service allows more effective blood inventory management and control in maintaining a CMV-seronegative blood inventory dedicated for patients who truly require it.

Implementation of a two-specimen requirement for verification of ABO/Rh for blood transfusion.

BACKGROUND: This study presents our implementation of a two-specimen requirement with no prior record of ABO/Rh to verify patients' blood type before transfusion. MATERIALS AND METHODS: Blood type verification was introduced, discussed, approved, and implemented over a 12-month period (May 2007 to May 2008). Potential barriers and impact on benchmark indicators were identified and tracked. RESULTS: Inpatient identification and/or specimen labeling for nursing and laboratory phlebotomists baseline corrected error rates were 1:467 and 1:5555, respectively. This study therefore sought and obtained approval to initiate a new policy of blood type verification before blood transfusion. Compliance in turnaround time (TAT) before and after implementation for completion of STAT type and screen/crossmatch within 60 minutes worsened marginally, from 90% to 80%. The impact on use of O-, uncrossmatched blood was found to be manageable. Seven (of 25 total) recorded electronic complaints were received after implementation. The corrected error rate for nurse phlebotomy draws after implementation was 1:630. CONCLUSION: Despite the lack of an instigating event, verification of blood type before blood transfusion was successfully implemented. An impact on resources and benchmark indicators such as TAT can be anticipated and managed. Further process improvement efforts will be needed to ensure safety (e.g., at time of blood transfusion) for patients receiving blood transfusions. ABO/Rh verification may be necessary even after future implementation of bar coding and/or RFID chips, because human errors continue to occur even with systems improvements.