16 November 2022: Haemovigilance & Pharmacovigilance Panel - B

The questioning went to the SABRE system from the SHOT perspective and the overlap with MHRA. The “SHOT or NOT” document was displayed (cool name). It is intended for those designated as “reporters” – people usually within health boards. The categories of reporting were reviewed. It is a process linked to “continuing excellence”. The two leading causes of harm are delayed transfusion and over-transfusion. There is still an issue about people knowing what to report and what not to repot to SHOT.

Specific to the matter of whether or not to transfuse, there should be an “indication for transfusion”; why to do it and what product to use. Of 35 or 36 reports of deaths from last year, 27 were connected to transfusion-related circulatory overload. So, understanding the justification for transfusing at all is important.

Next was the issue of mandating reporting. There is little requirement on hospitals to report incidents. SHOT has tried to introduce a requirement, but it is not supported by legislation. There is a link to someone being “professionally mandated” to report and that may be connected to clinical negligence arrangements. The idea of mandating of reporting was supported. It was noted that there is a large level of under-reporting. 2021 was the first year when there were reports from all hospitals and health boards in the UK. It simply isn’t credible for there to have been no near misses in a hospital in a whole year. Similar countries also recognise under-reporting. One suggests that only one in 300 near miss incidents may be reported. Certainly, there is a tendency not to report near misses, but it is obviously not helpful for real harm to be caused before change happens. Patterns such as behaviours and system errors need to be recognised. Desiring a mandated system would be faced by difficulties such as the mechanics of doing it. The current system is too flimsy and nebulous. It might help to have a designated person to decide if a report should be made, but again that still leaves some discretion with the Trust. If it is not a mandated requirement, the cuts to the NHS would likely result in this type of function being lost as not a priority since it is not a legal issue. It was noted as going against the grain to impose another legal requirement on doctors, but it would help, if it could be made effective. Having someone statutorily responsible would be a good start.

It was noted that a transfusion incident is likely to produce a quick problem, but for the matters of the Inquiry where the problems emerge later, that might not be so readily amenable to the SABRE or SHOT type of incident responding. The BSQRs (?) actually do mandate reporting, but not the near misses. It was seen an important to see a culture of reporting, system-wide. Making it as simple to report as possible would also help. There may be a problem with having only one person for a whole hospital, but maybe there could be a small number of people to delegate to. Another possible problem was the fear of being caught doing something wrong, so it may lead to sanitised record keeping. People would not want to contribute to a lowering of the esteem their lab was held in. Making reporting of near misses a professional responsibility through the GMC or a job role obligation were other considerations.

There are local arrangements for picking up on problems, including near misses. Some are pilot projects. They usually involve a “root cause analysis investigation”. That may miss issues related to problems occurring between different systems interacting. There is also the human factor.

The possibility of patient reporting was raised. It did not happen, but there was no reason why it could not. It is probably because their agencies do not have a public presence. There are many members of the public at conferences, but that is not the same thing. There is not a portal for patients to report, but such a report would not be unwelcomed. SABRE does allow reports from individuals and patient reporting is welcome. The Yellow Card Scheme is available to the public, too.

In 2021 there were 4,088 reports altogether with 2,871 going to SHOT, 1,523 went to MHRA, and 1,045 had reports in common. There was a brief discussion over why some reports would be withdrawn, such as them being incomplete. Counsel then went through the data for categories of reports in 2021. For the period 1996 to 2021 when there have been 27,008 reports in total, around 6,000 relate to incorrect blood component transfusion. Transfusion-transmitted infection was another category of incidents. For the whole period (1996-2021) there were reported to be 42 infections involving 35 donors resulting in 12 HBV infections and 12 Hep E infections. Transfusion issues make up the highest number of reports.

The panel reflected that the document illustrated that they were doing reasonably well as regards those infective agents they know about. (What about the ones they don’t know about?) Hep E poses problems because it can be transmitted from food. Nevertheless, there is still a huge amount of transfusion error going on, a huge number of near misses. The largest number by far were those near misses and yet these were still understood to be heavily under-reported.

The vigilance system was seen to be vulnerable in three ways. Firstly, people report what they choose to report, and they do so when they can. Secondly, the agencies themselves who receive the report are quite small with few human resources. Thirdly, the capacity to disseminate reports back to trusts is a weakness. The surveys the agencies send out do not attract as many responses as they would like. From the big picture, there have been year-on-year improvements, but it is still at too small a stage and impact. Blood is overall safe, but there is still the matter of under-reporting. Allied to this is the under-recognition of adverse events connected to blood use in treatment, such as breathing problems. Infection-related incidents may be missed due to the delay in their manifestation. People may simply forget they had a transfusion and so not make the connection. Also, a lot of blood is used in intensive care but the patients there are often not able to know what is happening around them and to them. It is recommended that patients are informed of having been given a blood component after they recover, but that may or may not happen. Some viruses come about by different routes, not just blood, so it is not always possible to say if a transfusion was the cause. There is also the difference for some people who require blood and have lower life prospects due to, say oncology factors, whereas otherwise healthy people will return to what seems like full health and nothing emerges to be reported. A single reporting portal may help, since some already busy clinicians have to use several systems, even if it is about the single issue. Equally, for linking events to patients, the MHRA use electronic systems such as the Clinical Practice Research Datalink to connect with longer-term clinical events.

It was noted that there were difficulties in making sure transfusions were recorded in primary care records held by GPs. In fact, it was admitted that the documentation for blood use was poor, particularly because it is inconsistent, especially when dealing with paper records. The changeover to electronic records is slowly progressing. That requires scanning in and attaching the older paper records, which is not happening consistently. And if a patient moves between wards, even a recent transfusion can be missed out of being included in the patient record. It was noted that the NHS does not have a good track record in adopting new electronic record-keeping systems. There need to be common standards between providers and sections of the healthcare system. Electronic system interoperability is still a major issue.

The importance was stressed of going back to the basis of making sure the original product was as safe as possible, that testing was as rigorous as it can be, and distribution works effectively and efficiently.