2022 Audit of Blood Sample Collection & Labelling

Key objectives

• To collect information on the quality of practice of labelling transfusion samples

• To determine whether bedside electronic identification systems had an impact on mislabelling

• To assess the incidence of Wrong Blood in Tube (WBIT)

• To explore reasons for sample rejection

• To provide information for a project of targeted review and improvement at selected sites

What did we audit?

• All NHS Trusts and independent hospitals in England were invited to participate in the audit. Trusts and hospitals in Wales, Northern Ireland and Scotland were also invited to participate, as were independent laboratories.

• Data were submitted by Trusts as a whole and by individual hospitals. Therefore, the term “sites” is used throughout this report to refer to either Trust or hospital.

• A letter explaining the purpose of the audit, the proposed timescale and the proposed dataset to be collected was sent via email to Chairs of HTCs, Trust Transfusion Laboratory Managers, Transfusion Practitioners and Consultant Haematologists with responsibility for blood transfusion. For independent hospitals a letter was sent to the hospital manager.

• Sites were asked to provide data on all blood samples sent for group and save or group and crossmatch which were rejected for labelling errors in the month of October 2022. They were also asked to provide the total number of samples that were sent for group and save or group and crossmatch in order to calculate local and national rates of mislabelling.

• We additionally asked for details of the number of incidents formally investigated in the hospital during the audit period because they were “wrong blood in tube” events.

Who took part?

• 191 sites contributed data

• 21511 rejected samples were analysed

What did we find?

• Full report (PDF 1478KB)
• Key points (PDF 209KB)

Main findings 

• 23584 rejected samples were reported by 179 sites in 1 month.

• Sample rejection rate of 4.4% represents a 50% increase compared to 2012 audit (2.99%).

• Reported number of wrong blood in tube (WBIT) incidents increased almost 3 fold compared to 2012 (92 in 1 month compared to 99 in 3 months).
  WBITs may be underreported to SHOT

• Sites with electronic sample labelling systems reported 50% fewer mislabelled samples, but no fewer (and based on few sites, potentially more) WBIT incidents.

• Electronic systems are not a substitute for positive patient identification.

• Staff at sites with electronic systems still need training in appropriate hand labelling for areas or scenarios where the electronic system is not available.

• It is often impossible to identify the individual taking a rejected sample – which represents a missed opportunity for feedback.

• 14.9% of samples were rejected for missing signatures on sample or form. Unless this represents a reliable way to identify the blood-taker, it may be questioned whether this adds to patient safety.

• Sites commonly collect data on reasons for mislabelling but not in a format readily enabling automated analysis.

Our recommendations

• All hospital transfusion teams should ensure that induction and refresher training on sample labelling and requests is made available and is appropriate to Trust guidelines and policies. It is suggested that this training should be targeted to the areas where rejection rates are highest, as indicated by the results from this audit or from local reporting/knowledge.

• Transfusion teams should assess environmental and human factors in clinical areas with high mislabelling rates, to identify systemic factors contributing to poor practice and understand any workarounds.

• The identity of staff responsible for taking samples/completing request forms should be readily identifiable both from the request/sample itself and from any electronic or audit records held. Electronic systems and processes should be designed to collect sufficient information to be able to confirm the identity and job role of the sample taker.

• Data collected routinely about mislabelled and rejected samples should be sufficient to allow meaningful reports to be easily generated. The systems used should be capable of producing summary reports automatically. We recommend that hospitals use these data to regularly measure their mislabelling/WBIT rates in order to benchmark their progress.

• The number of WBITs reported during the audit period is at odds with the annual number reported in recent SHOT reports. Transfusion teams should report all cases of WBIT to SHOT to support safety initiatives nationally

• While electronic requesting and labelling cannot eliminate all problems, the improvement in sample labelling quality is clearly demonstrable. This is recommended as the gold standard that should be aimed for. However, transfusion teams need to continually review how these systems are being used in practice to ensure workarounds and corner-cutting measures are not being taken. Paper request forms and hand labelling should still form part of mandatory training to cover system downtime, particularly where electronic systems are the only method in regular use and staff do not normally complete manual/paper requests.

• Positive patient identification remains fundamental at all stages in the transfusion process and its importance must continue to be emphasised, particularly when electronic bedside identification systems are implemented. Patients themselves should be encouraged to check the labelling of their samples, where appropriate.

• We recommend that sites review their local policies on sample rejection, particularly in relation to discrepancies in fields such as signatures on both sample and form (eg. in fields other than the core identifiers), to ensure that they benefit patient safety.