Regional
Implementing a check sample (second sample) policy in Groote Schuur Hospital, Cape Town, South Africa
Hemovigilance remains an important pillar of modern blood banking. In the era of risk-based process controls, there is an increased need to develop a pre-emptive, rather than reactive, approach to risk management. This involves implementing measures to detect and prevent probable adverse events before they occur.
However, some risks may go undetected before they occur. One area of such concern is wrong blood in tube (WBIT) errors where a patient is either misidentified at the bedside or incorrect identification labels are placed on samples and request forms. In such cases, the laboratory is unable to detect the discrepancy in the absence of a known record.
In order to address this risk, the Western Cape Blood Service (WCBS) turned to observing best practice in other nations with regards to the prevention of misdirected blood transfusions due to patient misidentification.
Most commonly, blood banks implement a second sample policy whereby two samples are required for patients with no previous record on the blood bank information system. Other mitigation measures revolve around the identification and sampling procedures and are implemented outside of the blood bank 3,4.
For a second sample policy to be effective, both samples should be taken independently and the patient must be positively identified on both occasions 5,6. This reduces the probability that a WBIT error can go undetected before transfusion as a blood group discrepancy between the samples can be noted at the blood bank 1, 2.
Figure 1. WBIT Reduction as GSH
Initially, there were some concerns regarding the feasibility of implementing such a policy locally, given the fact that these policies are currently employed by more developed nations where the local healthcare systems are better funded and equipped to deal with such a policy.
In April 2023, the WCBS initiated the first pilot at Groote Schuur Hospital in Cape Town, South Africa, in an attempt to implement a second sample policy within the context and constraints of the local healthcare system.
Due to financial and staffing constraints at local state hospitals, such a policy could not be implemented as a mandatory pre-requisite to crossmatching. Instead implementation focused on educating stakeholders on the purpose of the policy, and the associated risks it is meant to address, while encouraging co-operation, but not strictly enforcing it.
The education of stakeholders involved a combination of educational seminars along with direct communication when these additional samples were requested.
In its original implementation, the policy required the blood bank to check the system for a known record at the point of receipt. If no record was found, the requesting physician would be contacted and then asked about the urgency of the transfusion request and whether or not a second sample could be taken. The benefits and purpose of the second sample policy would also be explained at this stage if the physician was unaware. In the absence of any clinical urgency, crossmatching would be done upon completion of testing on the second sample.
The result was an overall decrease in the WBIT error rate from 1-in-2000 samples at Groote Schuur Hospital before implementation to 1-in-18000 for samples received from Groote Schuur Hospital in the current year as of December 2024.
Figure 2. Compounding Effect on Improving EXIM Rate and Turnaround Times Met.
As an additional benefit, the electronic crossmatch (EXM) rate at Groote Schuur Hospital increased from 57% to 64% improving the turnaround time for red cell product requests by reducing the need to perform serological crossmatching.
After the successful implementation at Groote Schuur Hospital, the policy has been adopted at all hospitals with on-site blood banks within the Cape Metro with similar success.
With input from blood bank staff and stakeholders, the procedure for fulfilling the policy has also been adjusted to ease implementation for blood bank staff as well as for stakeholders. Currently the patient record is looked up at the point of contact and a second sample is immediately requested if no record is found. Whenever possible, stakeholders are encouraged to have a second individual take the second sample to further validate that the patient is positively identified on a second occasion.
Figure 3. Example Process Revision Following Input from Staff and Stakeholders
These changes have enabled the policy to be expanded to off-site hospitals and several of these hospitals have already implemented the policy. Routine logistical runs to these sites have been adjusted to accommodate the second sample requests and allow more efficient and cost-effective delivery of red cell products as a result.
When including samples received from hospitals where the policy has not yet been implemented, the overall WBIT error rate has gone down from 1-in-1500 samples to 1-in-2600 across all WCBS sites, an overall reduction of 57% across the province since implementation. This number is expected to rise as the policy is implemented at more hospitals. The reduction in WBIT errors is most likely due to the raised stakeholder awareness regarding the prevention of WBIT errors rather than a direct effect of the second sample policy itself.
Ultimately, the key to successful implementation of the policy at the WCBS has been proper engagement and education of stakeholders with opportunities for feedback to ensure that the benefits of the policy are not outweighed by negative impacts to service delivery and operational burden to stakeholders.
References
1. Glisch C, Jawa Z, Brener A, Carpenter E, Gottschall J, Treml A, Karafin MS. Evaluation of a two-sample process for prevention of ABO mistransfusions in a high-volume academic hospital. BMJ Open Qual. 2018 Jul 15;7(3):e000270. doi: 10.1136/bmjoq-2017-000270. PMID: 30057954; PMCID: PMC6059318. 2. Thomas W, Davies J, Asamoah A, Scott-Molloy C, Sansom V, Kerr J. Two samples for blood transfusion: single centre experience. Transfus Med. 2014 Aug;24(4):209-12. doi: 10.1111/tme.12122. Epub 2014 Apr 30. PMID: 24779632. 3. Cottrell S, Watson D, Eyre TA, Brunskill SJ, Dorée C, Murphy MF. Interventions to reduce wrong blood in tube errors in transfusion: a systematic review. Transfus Med Rev. 2013 Oct;27(4):197-205. doi: 10.1016/j.tmrv.2013.08.003. Epub 2013 Sep 25. PMID: 24075096. 4. Dunbar NM, Kaufman RM; WBIT Study Investigators, The Biomedical Excellence for Safer Transfusion (BEST) Collaborative. Factors associated with wrong blood in tube errors: An international case series - The BEST collaborative study. Transfusion. 2022 Jan;62(1):44-50. doi: 10.1111/trf.16716. Epub 2021 Nov 2. PMID: 34726274. 5. Obaidallah N, Downie H, Colavecchia C, Callum J, Lin Y. Implementation of a blood bank generated tube for second blood group determination: Challenges, yield, and cost. Transfusion. 2022 Apr;62(4):784-790. doi: 10.1111/trf.16838. Epub 2022 Feb 25. PMID: 35213739; PMCID: PMC9304256. 6. American Association of Blood Banks . Standards for blood banks and transfusion services. 32nd ed. Bethesda, MD: American Association of Blood Banks; 2020.
