Wednesday, December 10, 2014

Angel (Musings on communication errors in TM)

Updated: 13 Dec. 2014

This month's blog derives from news items appearing in TraQ's monthly newsletter involving Jehovah's Witnesses in NZ and the UK, and the availability of online legal summaries.

In particular, the blog features a case from NZ in which communication failures led to a woman's death. Her death likely could have been prevented if the surgeon had known that she was a Jehovah's Witness and had refused to be treated with blood and blood products.

The case is all the more tragic because her surgery was routine, elective laparoscopic cholecystectomy. Even with 'open surgery,' transfusion may be required but seldom is. For many years, transfusion services have done only a type and screen, just in case blood is needed.

So, although involving a Jehovah's Witness, the blog is about communication and how critical it is to patient safety.

The title derives from a 1997 much-covered song by Canada's Sarah McLachlan.

UK SHOT Reports always feature cases that emphasize the importance of communication to transfusion safety. Musings on communication failures and examples from SHOT follow the NZ report.

Below is my edited version of key case details as described in the report. I've kept the NZ spelling (with the diphthong 'ae' used in many former British colonies,though not so much in Canada with our proximity to the USA).

1. Ms A was seen by a surgeon at an outpatient clinic (Hospital 1). A surgeon confirmed gallstones and she was put on a waiting list for an elective laparoscopic cholecystectomy.

2. Later Ms A attended a nurse-led pre-admission clinic, where she confirmed that she did not consent to the use of blood and blood products.

3. Ms A was admitted for surgery. Surgeon Dr C and anaesthetist Dr D met with her to discuss the operation and to complete the process of obtaining informed consent.

4. When the surgery began Dr C was unaware of the patient's views on blood transfusion. The matter was not raised during the surgical 'Time Out', when any issues of concern are brought to the attention of the OR team. (See below for info on surgical 'timeouts'.)

5. Surgery began at 9 am. Because of difficulties, at 9.50 am, the laparoscopy was converted to open surgery. Ms A's gallbladder was removed and the operation ended at 11.15 am.

6. Bleeding occurred during surgery, but not enough to cause concern. Ms A was transferred to the Recovery Unit at 11.25 am.

7. There were concerns about Ms A's condition from about noon. Initial measures were unsuccessful and it was thought that she was probably bleeding internally.

8. Dr C instructed that Ms A was to be transfused, at which point he was advised of her blood product refusal.

9. Dr C determined that further surgery was needed to identify and address the cause of the bleeding. Ms A, still partially sedated, confirmed that she would not accept blood.

10. Permission was sought from Ms A's mother to override Ms A's directive but she said she could not do this.

11. Ms A was returned to the OR and surgery began at 2.55pm. Because no obvious bleeding point was identified, Dr C determined that the best course of action was to pack the liver bed and close the abdomen, so that Ms A could be transferred to a facility better able to manage her condition.

12. Arrangements were made to transfer Ms A by helicopter to Hospital 2. When the helicopter crew arrived, it was decided that transfer was inappropriate due to likely hypoxic brain injury. Ms A was confirmed dead at 6.59 pm.

For more information, and key findings, I encourage you to read the full report below. 62 pages is a lot but many are appendices, plus it's fascinating. Documented tidbits include
  • At 2.29 pm, prior to the second surgery, Ms A's Hb was 45 g/L
  • At 3.26 pm, post second surgery, Ms A's Hb was 11 g/L 
A physician at Hospital 2, where she was to be transferred, said about the second surgery:
"Pre-operatively her haemoglobin level was 45, which in a Jehovah's Witness who refused blood product transfusion I felt was life threatening. Post-operatively her haemoglobin was 11 which is almost incompatible with survival.

At some stage I suggested using concentrated factor VII, which if used early enough before severe dilutional anaemia has occurred might control the bleeding. I thought [Ms A] was almost certainly going to die..."
Recommendations and Follow-up actions are on pp 42-3 of the Report and include mandating that 
  • Those involved review their practices and apologise to Ms A's family in writing.
  • Appropriate medical colleges be sent a copy of the report, and advised of the names of Drs C and D.
The communication failures in the NZ case highlight a long standing issue in transfusion medicine. Communication errors are common causes of adverse events.

Patients with special transfusion needs such as those requiring irradiated or CMV-negative blood components are particularly at risk when communication fails.The spectrum of communication deficiencies includes:
  • Physicians failing to communicate with nurses, technologists, pharmacists, and other health professionals and vice versa
  • Attending physicians failing to communicate with residents and interns
  • Staff from one unit failing to communicate with those from others
  • Staff on one shift failing to communicate with those on the next shift
  • Documentation failing to accompany patients from facility to facility
  • Health personnel failing to listen carefully to patients
Common tools to prevent communication errors include mistake-proofingaka as the politically incorrect 'idiot-proofing'.

Error proofing is  ubiquitous in society, e.g., beeping alerts when keys are left in cars or headlights are left on. Non-communication transfusion-related examples include
  • Colour-coded ABO typing sera, 
  • Pretransfusion nursing checklists 
  • Cross-checking work done by others
  • Eliminating identification errors via technology, e.g., barcodes, RFID
Preventing communication errors between TM professionals is more onerous because it involves complex, long term strategies such as
  • Developing standard operating procedures and tools (forms, letters, patient cards) to facilitate intradepartmental, interdepartmental, and inter-facility communication
  • Implementing methods to train and retrain health professionals to value effective communication and teamwork
  • Fostering a culture that eliminates communication barriers such as hierarchies within and between professions, and boundaries between departments
  • Developing information management systems to facilitate information transfer
Below is a small selection of communication issues documented by SHOT over the past decade (most recent first). (I've sometimes edited the original text.)

1. SHOT 2012 has 50 'hits' for 'communication'
One example (p.16, under 'Human factors in hospital practice'):
  • The errors described in this SHOT report consistently demonstrate failures in communication and handover that lead to adverse incidents, some life-threatening, in transfusion practice.
  • Failures of 'handover' (communication errors) may occur as the patient travels between wards and departments within a hospital, between clinicians in different hospitals, and between hospitals and community settings. 
  • Why? Often it's because of the human tendency to assume that someone else is responsible
2. SHOT 2013 has 48 'hits' for 'communication'
One example (p. 42 under 'ABO incompatible transfusions n=12'):
  • This shows the importance of communication between clinicians and laboratory staff in an emergency. There was no historical record available for the patient and laboratory staff issued FFP based on the misleading grouping result.
3. SHOT 2003 has 21 'hits' for 'communication'

Fewer 'hits' mean nil because the report includes 107 cases in which patients with special needs were transfused with the wrong blood. Of these, 81 involved patients at risk of GVHD for whom there was a failure to provide irradiated components.

The following three examples from SHOT 2003 (p. 23) illustrate the issues (italics not in original):
Case 10. Lack of awareness of guidelines puts patient at risk. A 66 year old male patient received fludarabine for chronic lymphatic leukaemia. The ward staff were unaware of the indication for irradiated blood components and so the laboratory was not informed.

Over a 5 month period the patient received 13 units of unirradiated red cells. 
Case 11. Failure of communication in shared care. A 14 year old male was admitted for an open lung biopsy following which he bled and required transfusion. He had previously received a stem cell transplant in another hospital in the same Trust, but there was no facility to link the two transfusion laboratory computer systems and the requester was not aware of the previous history.

Non-irradiated red cells were given. 
Case 12. No notice taken of an informed patient. An elderly male patient was admitted to hospital A with an ischaemic foot. He informed the ward staff that he required regular transfusion with 'special blood' at hospital B.

The ward confirmed with the transfusion laboratory at hospital B that he had an anti-ANWJ but this information was not passed on to the laboratory at hospital A who were undertaking pretransfusion testing.

The antibody screen was negative and 3 units of red cells were issued electronically and transfused. The patient had a rise in temperature and a raised bilirubin, and died 8 days later from bronchopneumonia.
As documented by SHOT, communication failures continue to happen because they involve humans, and 'to err is human'.

Sarah McLachlan's 'Angel' (often mistitled 'In the arms of an angel') has been used so often as a song of comfort that it's almost become a cliche. To me it fits a blog that describes a series of communication failures that resulted in a tragedy that need not have happened.This performance with iconic guitarist Santana is a 'oner'.
  •  Angel (by Sarah McLachlan with Carlos Santana)
In the arms of the angel fly away from here
from this dark cold hotel room and the endlessness that you fear.
You are pulled from the wreckage of your silent reverie.
You're in the arms of the angel, may you find some comfort here.
You're in the arms of the angel, may you find some comfort here.
The song's origin is not at all what it's come to symbolize: Sarah says it was inspired by articles about musicians turning to heroin to cope with the pressures of the music industry and subsequently overdosing.

1. 'Universal Protocol for Preventing Wrong Site, Wrong Procedure, and Wrong Person Surgery', including
1. A pre-procedure verification process
2. Surgical site marking
3. Surgical "time out" immediately prior to starting the procedure
2. A Report by NZ's Health and Disability Commissioner (30 June 2014)
Nelson Marlborough District Health Board
General Surgeon, Dr C
Anaesthetist, Dr D 
3. News item: Jehovah's Witness dies after refusing blood transfusion (20 Oct. 2014)