Incomplete assessment and premature discharge of patient following heart attack

Names have been removed to protect privacy. Identifying letters are assigned in alphabetical order and bear no relationship to the person's actual name.

Introduction

An independent cardiologist was commissioned by the Ministry of Health and a District Health Board to investigate cardiology services at a public hospital following the death of a patient on 4 January 2000, less than three weeks after being discharged from the hospital. The circumstances of this patient's death were the subject of a report on the "20/20" television programme which raised serious concerns about cardiology services at the hospital.

The inquiry began on 30 March 2000 and was completed by 7 April 2000. The terms of reference for the inquiry were to determine whether:

• clinical decisions concerning diagnosis and treatment of patients with cardiac disease at the hospital are consistent with other secondary hospitals in New Zealand; and
• referral patterns and rates for angiography and coronary artery bypass surgery at the hospital are consistent with other secondary hospitals in New Zealand.

On 11 April 2000, the Chief Medical Advisor of the Ministry of Health reported:

"The review has shown quite clearly that cardiac services provided at [the hospital] are appropriate and safe considering [the hospital] is a secondary provider. ... Whilst we don't want to diminish what has happened, we believe that the public [of the city] can have absolute confidence in the cardiac services provided by [the hospital]. However, this does not mean that patients interviewed by media do not have complaints. Their concerns should be dealt with by the Health and Disability Commissioner."

Mr A's case was one that was highlighted by the media programme. The complaint, as expressed by his widow, the information obtained during the investigation, and the Commissioner's findings are set out in this report.

Complaint

Mrs B complained about the care her late husband, Mr A, received from the District Health Board following his admission to the hospital on 23 December 1998. The complaint has been summarised as follows:

• Mr A was not monitored for long enough and the length of his stay in hospital was not sufficient.
• The necessary repeat blood tests were not done.
• Mr A was not given adequate information on discharge as the doctor who was meant to see him prior to discharge failed to turn up.
• On discharge Mr A was told that a letter telling him when he was to have more tests would follow. This letter was never received.
• Mr A's hospital discharge summary was not completed and his general practitioner was never informed of his admission to or discharge from the hospital. Had this been done his general practitioner would have been alerted to the fact that a follow-up blood test had not been done and an exercise tolerance test appointment not made.
• Mr A died at home nine days later from 'coronary artery occlusion due to atheroma'. Had the hospital provided Mr A with services of an appropriate standard, he may not have died.

Investigation process

The complaint was received on 18 May 1999 and an investigation was commenced on 18 June 1999. Information was obtained from the following:

Mrs B    Complainant, wife of the consumer
Dr C    Provider / Registrar of the hospital
Dr D    Provider / Medical Consultant at the hospital
Dr E    Provider / House Surgeon at the hospital
Dr F    Provider / Clinical Director Medical Services of the hospital
Ms G    Provider / Staff Nurse at the hospital
Ms H    Provider / Charge Nurse Manager at the hospital
Ms I    Provider / Director of Nursing at the hospital
Mr J    Provider / Employer / Chief Executive Officer of the DHB

Mr A's records were obtained from Mrs B, his general practitioner, the hospital, and the Accident Compensation Corporation (ACC). Independent expert advice was obtained from a cardiologist, Dr Richard Luke, and a nurse, Ms Debbie Penlington.

Information gathered during investigation

Background

On 23 December 1998, at approximately 6.20pm, 51-year-old Mr A experienced a sudden onset of chest pain that lasted several minutes, followed by a sudden loss of consciousness lasting not more than 10 minutes. His wife, Mrs B, witnessed the event. During this episode Mr A lost control of his bladder and bowel functions. An ambulance was called.

Mr A regained consciousness but then suffered a further episode of chest pain lasting five minutes associated with shortness of breath, sweats and nausea. He again lost consciousness, but had regained it by the time the ambulance arrived. Single lead electrocardiograph (ECG) taken by the ambulance officers showed some ST depression indicative of cardiac ischaemia (insufficient blood flow to the heart muscle).

Admission to the hospital

Mr A arrived by ambulance at the hospital's Emergency Department at 7.42pm. He was seen by Dr C, the acute admitting registrar.

The hospital provides immediate care for heart patients in its region. It does not provide a tertiary service such as coronary surgery and angioplasty. These services are provided at another public hospital.

In the hospital notes Dr C recorded events leading to Mr A's admission to the Emergency Department and noted that changes on the ECG taken by ambulance officers had disappeared when Mr A became pain free. Dr C also recorded that Mr A had previously been well, was a smoker, and that he had a strong family history of ischaemic heart disease.

Dr C examined Mr A and found nothing remarkable. Mr A's vital signs were within normal range and there was no evidence of cardiac ischaemia or heart failure. Chest x-ray showed a collapse of his right lower lung and widened mediastinum which caused Dr C to consider an aortic dissection (a tear in the wall of the aorta, the major blood vessel leaving the heart) as a possible diagnosis. His final decision was to keep Mr A in hospital, treat his angina, and conduct further investigations (including an ECG and blood tests) to exclude the possibility of an acute myocardial infarction (heart attack). During this assessment Dr C noted that Mr A would need an exercise tolerance test (stress ECG).

Dr C phoned Dr D, the admitting consultant physician on call that night, and discussed Mr A's case with him. Dr D advised me that his impression from that discussion was that Mr A's pain was cardiac in origin and that he advised Dr C to manage him accordingly. It was agreed that Mr A would be admitted to hospital to exclude the possibility of an acute myocardial infarction.

Overnight in the ward Mr A was monitored on telemetry (an electronic device that transmits the patient's heart rhythm and rate to a central monitor). A staff nurse who was on duty that night recorded that Mr A slept well and did not complain of chest pain.

The ward round

The following morning, 24 December 1998, Mr A was reviewed and physically examined by Dr D on the post-intake ward round. Dr C and the house surgeon, Dr E, were also present. Dr F, Clinical Director of Medical Services at the hospital, advised me that on such a round "all new admissions are reviewed and a decision is made concerning investigation, treatment and the length of stay in hospital".

Medical notes taken by Dr E during the ward round record that Mr A's vital signs were stable, his pulse rate 52-56 beats per minute, he was afebrile (did not have an elevated temperature), his heart sounds were normal and he had no heart murmur. Dr E recorded that Mr A had had one episode of chest pain through the night, which he (Mr A) had not reported to the nursing staff. The pain rapidly resolved itself.

Dr D advised me that the blood tests and ECG taken the night before did not show evidence of myocardial damage (heart attack). The clinical picture was in keeping with angina. He did not consider the angina to be unstable and therefore commenced Mr A on atenolol 25mg daily, soluble aspirin 150mg daily and Nitrolingual (glyceryl trinitrate (GTN)) spray as required for chest pain. Dr D advised Mr A to stop smoking and indicated that he would require an outpatient exercise tolerance test some time in the future to assess the severity of his ischaemic heart disease.

Decision to discharge

Dr C advised me that a decision to discharge a patient is made on a ward round. Although this decision is made on the morning round, the actual physical discharge of a patient does not occur until a Discharge Summary and associated documentation is completed and the patient is given the necessary information. Dr C said that on the morning ward round on 24 December 1998, Dr D made the decision to discharge Mr A and said that he could go home. Dr C said there was no ambiguity in Dr D's statement and it was not contingent on the outcome of any further test results. The follow-up plan was to organise an outpatient exercise tolerance test (ETT).

Dr C advised me that in his opinion, as a registrar, he would have kept Mr A in hospital for longer. He said that in cases of chest pain, patients are normally kept in hospital for at least 24-48 hours and three sets of tests (blood tests and ECGs) are normally done before a patient is discharged. However, Dr C said that he accepted the decision of Dr D, who was the more experienced physician and his consultant, and who ultimately was responsible for Mr A's care. Dr C stated:

"He [Dr D] clearly understood all the details of the case and had fully assessed the patient and available investigations. His decision to discharge the patient that day without further inpatient investigations did not seem inappropriate to me at the time. There was no evidence clinically that an acute ischaemic event (i.e. infarction) had taken place nor that the patient had unstable angina. [Dr D] had considered the case fully and had decided that [Mr A] could be further treated and investigated on an outpatient basis."

Dr C said that on the ward round Dr D gave Mr A brief but appropriate advice including giving up smoking, taking it easy and, if he had further chest pain, to see his general practitioner. Although he could not recall exactly what Dr D told Mr A, Dr D normally counselled his patients appropriately about angina, communicated well with them, and was a careful physician.

Dr E's notes were recorded at the bedside at approximately 10.00am. Although he has not specifically noted "for discharge" in the records, Dr E advised me that he understood, at the time, that Mr A was going home.

Dr D denied that he made a decision or authorised the discharge of Mr A on the 24 December 1998 ward round. He said the discharge was dependent on the results of the repeat ECG and blood tests. Dr D said he could not have made that decision because the results of the repeat tests were not available to him on the ward round. He said that the results of repeat tests are checked and the patient is reassessed later in the day by the house surgeon or registrar. At this stage decisions about further management are made.

Ms G, who was looking after Mr A that morning, advised me that she approached the doctors on the ward round and enquired about Mr A's discharge. She said both Dr D and Dr C informed her that he would be discharged.

In his response to my provisional opinion, Dr D repeated his assertion that he did not order Mr A's discharge. He claims that Mr A was not to be discharged without a further clinical decision made in the light of repeat blood tests (which Dr D says he would have expected to have been ordered). Dr D thinks that a decision to discharge was probably made that afternoon by a doctor or doctors other than him.

There is plainly an important disagreement about relevant facts. I return to this later in my opinion when I discuss the position of the various individuals involved in relation to the requirements of the Code of Rights.

Repeat tests

Dr D stated that it is standard medical practice to repeat blood tests (namely Troponin-T and CPK) and ECG before the decision is made to discharge a patient who has presented with cardiac chest pain. This is done even if the initial tests are normal.

Mr A had a single set of cardiac enzymes taken on the evening of 23 December when he was admitted and a single Troponin-T estimation, which was normal. An ECG was also done that evening. Repeats of these tests were not ordered. Although Mr A's Troponin-T level was within normal range (less than 0.10 ug/L), a note at the base of his laboratory report stated:

"A Troponin-T result of 0.1 ug/L or above strongly suggests myocardial damage. However, a negative result does NOT exclude myocardial damage for at least 12 hours after onset of chest pain. Patients with no TnT elevation may still have chest pain of cardiac origin and warrant further assessment."

Nurses' involvement in Mr A's discharge

Mrs B said that when she came to the hospital at 2.00pm on 24 December 1998, her husband informed her that he had been told by Dr D that morning that he could go home. Her husband also told her that a doctor was due to see him at 3.00pm after which he would be able to leave.

Ms G advised me that nursing staff on the morning shift usually hand over the care of their patients to the afternoon shift staff at approximately 2.45pm. Her understanding of her role as a nurse was to ensure the patient's family was informed of his discharge either by the patient or her, and that any intravenous luer was removed. At 2.10pm that day she recorded the following in the nursing notes:

" ... Telemetry removed, for discharge this afternoon, no complaints of pain, intravenous luer removed."

Ms G's entry is the last written entry in Mr A's clinical and nursing notes dated 24 December 1998.

Ms G advised that her normal practice was to ask patients being discharged to remain on the ward to receive discharge papers and prescriptions, and any information regarding their condition, treatment and follow-up by the doctor. She also advised that, at the time of Mr A's discharge, it was common for some patients to have their discharge papers sent to them by mail at a later date. In such cases medical staff would notify nursing staff of the arrangement so the patient could be sent home. Her understanding regarding Mr A's discharge was that a doctor would return to complete the discharge documentation and give it to Mr A.

Ms H, Charge Nurse Manager of the ward at that time, advised me that she was on duty on 24 December 1998, but was not present on the ward round and had no recollection of Mr A. She said that once the decision is made to discharge a patient, medical staff inform either the charge nurse or the nurse looking after the patient. In December 1998 it was customary for the charge nurse, or the nurse or the doctor looking after the patient, to write a "D" beside the patient's name on the white board, indicating that the patient was for discharge. Ms H advised that patients still waiting for discharge papers on completion of the morning shift are "generally" handed over to the afternoon staff.

Ms H advised that medical staff are responsible for writing the Discharge Summary and handing it to the patient. The ward clerk sends requests for outpatient appointments as per instructions on the Discharge Summary. Occasionally, a doctor may indicate to nursing staff that the patient can go home without a Discharge Summary and that one will be faxed to the patient's general practitioner the following day and a copy mailed to the patient. This task is carried out by the ward clerk.

Regarding Mr A, Ms H advised that her understanding was that the afternoon on-call house surgeon was contacted by nursing staff and indicated that Mr A could go home without the Discharge Summary. This was not recorded in the nursing notes. Ms H could not provide the name of the nurse who made the call, or the name of the doctor who was contacted.

Ms G advised that towards the end of her shift, because Mr A had not yet been discharged, she "would have verbally handed over to the entire afternoon nursing staff that [Mr A] was waiting to see [Dr C] to receive his discharge papers. [Mr A] would then have been allocated to an afternoon nurse." She had no recollection of contacting Dr C to remind him that Mr A was waiting for his discharge papers.

Ms G informed me that after the group verbal hand-over, afternoon nurses are allocated a patient load based on room numbers. This includes patients waiting for their discharge documentation. The patient allocation is usually done by the charge nurse or the duty co-ordinator. She advised that when there is a shortage of beds for new admissions, independent, self-caring patients waiting for discharge papers are asked to wait in the ward lounge. Such patients are still assigned to a nurse and their name is written on the board to show that they are in the lounge rather than in a bed space. She said that on 24 December 1998 there was no shortage of beds. Mr A's bed space would have been available to him until he left the ward.

Ms G advised that it remains current practice at the hospital for the afternoon staff to be advised through the group verbal hand-over of any patients waiting for discharge papers. She said that it is the afternoon nurse's responsibility to ensure the patient has received the necessary documentation before he or she leaves the ward.

Mrs B said that she and her husband were asked by a nurse to go to the waiting room and wait for a doctor there. When the doctor had not arrived by 4.40pm, her husband went to the nurses' station to ask when the doctor would be arriving. A nurse told him that the doctor had left the hospital and that he must have forgotten to come and see him. The nurse apparently told him to go home and that a letter would be sent advising him when he was to have further tests. Because Mrs B remained in the waiting room, she did not see the nurse to whom her husband spoke.

Mrs B said that before her husband left the nursing station, the nurse gave him a prescription for atenolol, aspirin tablets and Nitrolingual (GTN) spray. Records show that the prescription was written and signed by Dr E, and dated 24 December 1998. Dr E had no recollection of writing the script. He also said that "it is quite common for prescriptions to be written in preparation of discharge". Mr and Mrs B left the ward at about 5.00pm and uplifted the prescription from the hospital's pharmacy before going home.

Ms I, Director of Nursing at the hospital, advised me that the "Charge Nurse Manager (CNM) and various consultants manage patients' care on a ward. The CNM assumes responsibility for co-ordination of all processes and alerting the medical staff as required. The nursing care of each patient is managed on a patient allocation basis i.e. not primary nursing. The person maintaining oversight is the CNM." The Charge Nurse Manager or the person delegated as co-ordinator on the shift is responsible for:

• co-ordination of patient admission to the ward
• co-ordination of patient discharge from the ward
• the allocation of nursing staff on each shift
• overall co-ordination of patient care on the ward.

Ms I advised that in December 1998 there was a documented procedure relating to nursing practice in regard to discharge planning. She said that "this required patients to receive information as pre-discharge education, that patients received a discharge summary, prescription and outpatient appointment on discharge. They were also to be provided with information for self-care on discharge. These expectations were generic and relied on nurses to ensure that the expectations were achieved."

Doctors' involvement in physical discharge of Mr A

Dr C and Dr E had no recollection of an undertaking being given to see Mr A at 3.00pm. They were not aware that Mr A was expecting to see them at that time. Neither Dr C nor Dr E could recall being contacted by ward staff regarding Mr A's discharge. Both Dr C and Dr E stated that Mr A should not have been sent home before he was reviewed by one of them. Mr A left the hospital without proper medical clearance.

Dr E advised that "if a patient is waiting a long period for the discharge procedure, it is normal that the nurse responsible for the patient calls the doctor to question when they will be attending. I do not recall ever receiving such a call from any of the ward staff informing me that Mr A still needed to be seen." He said that there was a doctor on call 24 hours a day and that he would have been in the hospital at that time. Normally he would have seen a patient being discharged, by 5.00pm. Dr E could not explain why this did not happen on this occasion or why he did not liaise with Dr C regarding physical discharge of Mr A.

Dr D advised that normally he is not contacted by a house surgeon or his registrar following their review of the patient and regarding their decision whether to discharge a patient. He said that the house surgeon and the registrar would normally review a patient on their return from lunch and decide who is to be discharged. Dr D said that, as a consultant, he is normally off site by that time of the day.

Discharge Summary and follow-up information

Dr C advised that it is a house surgeon's job to complete discharge documentation. This is normally done immediately after the ward round or shortly afterwards. However, it was part of his job as a registrar to assist the house surgeon where necessary. Dr C assumed that Dr E was going to attend to discharge documentation, especially as the instructions were relayed to and written down by him on the ward round. As nothing further was said, Dr C assumed that Dr E had written the Discharge Summary, arranged the exercise tolerance test and provided Mr A with necessary explanations.

Dr C and Dr E advised that the hand-written copy of the Discharge Summary is normally completed by a house surgeon at the time of discharge and given to the patient before he or she leaves hospital. A copy of the summary is sent to the patient's general practitioner. A formal version of the summary is dictated by the registrar some days later.

Dr F, Clinical Director Medical Services at the hospital, confirmed that a copy of the written Discharge Summary should have been given to Mr A at the time of his discharge and a copy sent to Mr A's general practitioner. This did not occur. Dr F acknowledged to Mrs B: "There was a serious failure of process which seems to have been due to communication ... This is unsatisfactory and I will endeavour to ensure that this does not happen again."

Subsequent events

Having been told on the ward round that Mr A had a possible case of angina, Mrs B said she and her husband did not consider the condition to be life threatening. She said her husband was sent home without appropriate instructions. He was not told to rest or not to do heavy physical work. Feeling well and secure on the medications he was prescribed, Mr A embarked on helping his sons build a fence. He died suddenly on 2 January 1999, nine days after being discharged from the hospital. According to the Coroner's report, Mr A died from coronary artery occlusion due to atheroma.

Dr C advised me that he first became aware that Mr A's Discharge Summary had not been completed a few days after Mr A died. Dr F informed Dr C of the death and the fact that Mr A's Discharge Summary had not been completed. (Hospital notes record that on 8 January 1999, six days after Mr A died, a friend of the family phoned the Outpatients Clinic at the hospital to say that Mr A had died. Mr A was scheduled for an aural toilet at the Clinic on 14 April 1999.) Dr C completed the Discharge Summary on 8 March 1999 after a request from Dr F.

Mr J, Chief Executive Officer of the District Health Board, advised that Mr A's records were found on Ward 6 on 7 January 1999 in a state of waiting for medical staff to complete the process. As the regular ward clerk was on holiday and not there to remind medical staff, the omission was not identified until the Medical Records Department requested Mr A's records on 7 January.

Dr C advised that Mr A's file was not on the desk in the registrar's office where it is normally placed by clerical staff. Had it been there he would have attended to it. Dr C said that although he "would have liked the paperwork to have been completed so that the family would have been more reassured that the treatment of [Mr A] had been medically correct, the final tragic outcome, [Mr A's] death which was essentially the unforeseeable outcome of ischaemic heart disease, would not have changed".

Dr D stated:

"While certain aspects of management (such as repeating of blood tests, arranging follow-up and writing discharge summary) fell short of the standard of practice at [the hospital], it is very difficult, if not impossible to predict if this could have prevented his death. With the benefit of hindsight, we know he remained well for nine days after discharge from hospital ... The length of time between his admission and his death makes it very unlikely that further stay in hospital could have prevented his death. Had he remained in hospital for a few days longer and remained well during this period (as we know he did after he returned home), his treatment on discharge would not have been different from the one he in fact received."

Orientation of new medical staff to the hospital

Dr E stated that at the time of Mr A's admission it was his first week on Dr D's team and his first run in a New Zealand hospital. Dr C had joined hospital staff ten days earlier.

Dr C stated that in December 1998 there was no formal orientation programme for new medical staff at the hospital. Responsibilities of registrars and house surgeons were discussed during the brief morning hand-over meetings. Dr F advised that clinical issues and patient management were also discussed at these meetings.

Dr D advised me:

"I talk to my new team at the beginning of their run about what my expectations are of their duties, advising them to contact me with matters pertaining to patient management and about any matters they are unsure about. ... I teach my registrars and house surgeons during my Monday to Friday rounds. House surgeons and registrars are given written guidelines by the Department of Medicine at the beginning of their run, outlining the expectations of the Department."

Dr C recalled being told at that time that house surgeons were responsible for completing the written Discharge Summary and handing it to the patient. Registrars, on the other hand, were responsible for completing the more formal, dictated summary, within several days of discharge. He said registrars are responsible for overseeing this process, and although they often help their house surgeons with giving patients their discharge documentation and prescriptions and arranging their outpatient tests, they would normally not see the patient again unless there was a specific concern raised at the ward round or subsequently raised by the house surgeon, nursing staff, or the patient.

The hospital provided no documentation to me regarding the following:1

• details of orientation provided to new medical staff at the time of Mr A's admission to the hospital
• job description and responsibilities of registrars and house surgeons in the discharge of patients, as it was in December 1998
• clinical pathways relating to the management of patients presenting with cardiac chest pain.

The Customer Liaison Advisor of the hospital, advised that "there is now a dedicated Registered Medical Officer [RMO] Unit co-ordinator who works closely with junior medical staff and the Learning and Development team to meet the training and educational needs of junior medical staff". She said that the RMO Project, which has been in place since early 2000, was designed to "monitor and improve the training, information and safety needs of junior medical staff". She said that the Learning and Development Department, and the RMO Unit co-ordinator "work closely with the RMO representatives to provide a comprehensive orientation to working at [the hospital]".

Independent advice to Commissioner

Cardiology advice

Dr Richard Luke, an independent cardiologist, provided the following expert advice:

"Medical History: The symptoms resulting in hospital admission are well documented. Severe chest pain developed on relatively minor exertion and was followed within minutes by loss of consciousness associated with faecal incontinence. Upon recovery of consciousness, further severe chest pain was noted associated with sweatiness and pallor. This pain persisted for between 5 and 8 minutes - resolving prior to ambulance arrival.

The initial electrocardiogram (ECG) strip recorded by ambulance staff shows a definite abnormality consistent with myocardial ischaemia (restricted blood flow to the heart muscle). While the limitations of a single lead ECG are noted in the statement by [Dr C], the fact that a second ECG strip recorded 2 minutes later shows resolution of the abnormality offers strong evidence that the initial diagnostic information was valid.

The patient arrived at [the hospital] at 1942 hrs and was pain free with the 12 lead ECG showing no abnormality. The initial blood tests taken within 1 hour of symptom onset revealed no evidence of heart muscle damage / heart attack. He was observed until discharge late on the following afternoon. No further ECGs or blood tests were completed. I understand that [Mr A] died suddenly 9 days later with no intervening symptoms.

OPINION: This man had numerous coronary risk factors and those caring for him seem to accept that his chest pain was related to ischaemic heart disease. Presentation with new onset ischaemic pain, recurring after the initial episode is, by definition, unstable angina.

The associated syncope and ECG evidence of ischaemia - both of which are markers of high risk, should have heightened concern about this man.

Appropriate management should have included repeat measurement of Troponin-T - a marker of heart muscle damage - at least 6 to 8 hours after symptom onset. Troponin-T is an important additional marker of risk in the setting of unstable angina and it is well known that there is a delay following heart muscle damage until appearance in the blood of abnormal levels. The first blood sample was taken too early in the course of the illness to provide this information. A second ECG, prior to discharge would also have provided potentially useful diagnostic information.

Any abnormality of these delayed test results would have provided a clear indication for hospital admission and Heparin (blood thinning or anticoagulant) therapy for 48 - 72 hours.

If normal then early exercise testing is indicated to seek evidence of underlying coronary disease. In an ideal world this screening exercise test would be completed prior to discharge but in some hospitals excessive workloads may not allow this and thus an early outpatient test would be arranged.

[Mr A] almost certainly had unstable angina with high-risk features and was discharged without adequate assessment. The available documents indicate that those involved with his care accept that additional tests should have been completed thus there is no dispute that the standard of care was inadequate.

No one has accepted responsibility for the failure to complete an adequate assessment nor for the premature discharge. Both [Dr D] and [Dr C] seem to have believed that 'usual practice' would dictate management beyond the initial ward round. This is clearly a dangerous assumption in the absence of written clinical guidelines or pathways - particularly when relieving junior doctors, who may be unfamiliar with 'usual practice' are involved with patient care.

While I am reluctant to apportion blame, it falls to the Consultant responsible for a patient's care to ensure that the appropriate standards are followed. The entry into the medical record from [Dr D's] ward round implies that early discharge was intended and no mention is made of the additional tests required to make an accurate diagnosis.

It will never be known whether or not a more thorough assessment would have saved this patient's life but [it is] possible that subsequent events could have been modified by more aggressive investigation of a patient who was almost certainly at high risk.

Responses to questions not yet addressed:

• Were appropriate follow-up arrangements made for [Mr A's] discharge and was he given adequate instructions before being discharged?

It is not clear from the medical record whether or not an appointment for exercise testing was ever going to be provided. There is no record of what instructions were given to the patient beyond smoking cessation. Other instructions may have been provided but not documented. The failure to provide discharge information to the GP is unfortunate but not surprising given the complete lack of discharge planning.

Conclusion:

I believe that [Mr A] did not receive an appropriate standard of care from the staff at [the hospital]. The failure to adequately assess his condition may have influenced the ultimate outcome.

There were clearly systems failures, which contributed to the poor standard of care. It is possible that staffing and workload issues at this hospital also played a part.

Many hospitals have moved towards clinical pathways for the management of chest pain and other acute conditions. These may now be in place at [the hospital]. Written guidelines ensure that the appropriate steps in assessment and management are completed and make it less likely that substandard care will occur."

Nursing advice

Ms Debbie Penlington, an independent nurse advisor, provided the following expert advice:

"Response to Specific Questions

1. What specific and relevant standards apply in this case and did nursing staff follow them?

I have been unable to locate any national or international best practice / evidence based practice guidelines on the nursing management of angina that were available prior to or at the time of [Mr A's] hospitalization. However there are a number of nursing textbooks which were published at the time and which outline the management of angina i.e. the nurses involved were not being asked to provide nursing management of an uncommon disorder. Resources such as these books could have been used in one of two ways, by individuals to guide their practice or as the basis for developing either ward or hospital guidelines for nursing staff.

In October 1996 the Nursing Council of New Zealand approved a set of 11 competencies for safe nursing practice. These were incorporated as Standard 10 of the standards for Registration of Comprehensive Nurses i.e. these are the expectations of a nurse at the time they gain nursing registration in New Zealand. Some would argue that these competencies are irrelevant however as these have been proposed as the basis for competency based practicing certificates for nurses (subject to a change in legislation) I believe they can be used as a benchmark and are relevant in the case of [Mr A].

Of particular relevance to [Mr A's] situation are standard 7 Ethical Accountability, standard 8 Health Promotion and standard 9 Interprofessional Care.

Standard 7 Ethical Accountability

The relevant performance criteria are that the nurse

• Facilitates the client's access to appropriate therapies or interventions and respects the client's rights to choose among alternatives
• Appropriately challenges health care practices which could compromise client safety, privacy and dignity.

Standard 8 Health Education

The relevant performance criteria are that the nurse

• Recognises the potential for health teaching in nursing interventions
• Uses formal and informal methods for teaching appropriate to the individual or group activities
• Prepares the client and / or others for continued health care
• Evaluates client learning and understanding about health practices.

Standard 9 Interprofessional Care

The relevant performance criteria are that the nurse

• Promotes a nursing perspective within the interprofessional activities of the health care team
• Co ordinates care to maximize health for the client
• Collaborates, consults and refers to maximize health gains
• Documents appropriate nursing information and communicates this to other team members
• Accurately documents assessments of client's health status and decisions made about prescribed interventions, treatments, medications and referrals/ follow up
• Collaborates and consults with, and provides accurate information to client, family and other health professionals about the prescribed interventions or treatments and/ or medications.

2. Was the service provided by Nursing Staff at [the hospital] provided with reasonable care and skill?

Based on the nursing documentation provided and the statements from the nurses involved there appear to be some deficits in the care [Mr A] received from the nurses during his admission at [the hospital].

In particular, there is no documented evidence that the nurses challenged the lack of effective discharge planning for [Mr A] or collaborated / consulted with medical colleagues in an effort to promote his discharge planning.

There is no evidence of an assessment of [Mr A's] understanding of his condition nor of any teaching being carried out to address any learning needs identified. This was particularly relevant on the 24th December prior to discharge. Relevant teaching may have included:

• Disease process
• How to describe the characteristics of chest pain e.g. onset (fast /slow), severity, radiation, location, duration, associated symptoms (e.g. anxiousness), alleviating factors (e.g. oxygen, change of position), precipitating factors (e.g. bowel movements)
• Management of chest pain including the need for prompt treatment
• How to avoid or control possible precipitating factors e.g. digestion of a heavy meal
• Health promotion e.g. diet /exercise
• Management of risk factors e.g. smoking / passive smoking
• Medication issues e.g. administration, storage, side effects, and contraindications
• Stress reduction.

Where a patient's learning needs exceed the nurse's knowledge / experience the nurse is responsible for referring the patient to another health care professional to ensure their learning needs are met. It may be argued that when patients are in the ward for a short time frame it is not possible to address learning needs such as those stated above, however I would argue that the short time these patients have contact with nurses/health professionals makes it even more important that learning needs are addressed.

3. What is the role and responsibility of a charge nurse in the patient discharge process?

Where hospital discharge guidelines / policies are available the charge nurse is responsible for ensuring they are implemented within his or her area.

In the absence of hospital guidelines or policies the charge nurse is responsible for establishing and implementing a process / guideline / policy within their area which clearly defines the accountabilities of staff located in the ward i.e. nurses and ward clerks and the interface with other staff contributing to the process.

The charge nurse is also responsible for ensuring that new staff are informed of the discharge process / guideline / policy (e.g. during their orientation) and for monitoring the effectiveness of the process which can be achieved in a variety of ways including random audit, talking with patients / relatives and the monitoring of incidents / complaints.

The charge nurse's role in the discharge process is determined by the process / guidelines / policy that is in place in the area at the time. In some organisations the charge nurse may co ordinate the process and be the direct link with the medical staff in other organisations this may be delegated.

4. What responsibility, if any, did Charge Nurse [Ms H] have in the discharge of [Mr A]?

As Charge Nurse [Ms H] states she was not on the ward round on the 24/12/98 her role in the discharge process would have been to oversee the process.

In the a.m. [Ms G] was responsible for discharge planning then as the p.m. shift co-ordinator S/N [Ms K] was responsible for allocating a staff nurse to [Mr A], this staff nurse would have then ensured that discharge requirements were being addressed.

5. At what point do you consider the hospital's responsibility for the care of [Mr A] ceased?

Ward staff responsibility for the care of [Mr A] ceased when [Mr A] left the ward, the hospital responsibility ceased when he left the hospital grounds.

6. Should a nurse have been allocated to [Mr A] between [Ms G] going off duty and [Mr A's] departure from the ward at 5pm? If so, who is responsible to ensure that such an allocation was made?

Yes a nurse should definitely have been allocated to [Mr A] until his departure from the ward. This nurse should have introduced him / herself to [Mr A] and acknowledged that they were responsible for him until he left the ward. The person responsible for allocating staff for the afternoon shift was the person responsible for ensuring this allocation was made (this may vary according to the area, in some clinical areas the charge nurse allocates workload for the p.m. shift, in other areas it is the p.m. co-ordinator). In this case it was S/N [Ms K].

7. What is the usual practice when a patient wants to go home or is informed he/ she can go home but has not been formerly discharged by medical staff?

The nurse would contact the relevant medical staff and arrange for them to visit the patient and complete the discharge process. The nurse would document all action he / she took to facilitate the discharge process. They would also keep the patient informed of progress toward their discharge. A nurse would be assigned to the patient until discharge was completed and they left the ward.

8. What documentation would you expect to be completed by nursing staff in these circumstances?

I would expect the nurse to document all action they had taken in the care of the patient. I would also expect the nurse to document any care that was ordered or indicated but which they did not carry out, including the rationale for not doing it and who they informed of this. All entries should be dated, signed and include the staff member's designation.

9. Was [Ms G's] documentation adequate and did it reflect adequate discharge planning and follow-up of [Mr A]?

No I do not believe it was adequate. On 23/12/98 there is no overview of the nursing management plan for [Mr A] e.g. frequency of observations to be carried out. Also there is no statement of instructions given to [Mr A] e.g. the importance of reporting pain promptly and being able to clearly describe its characteristics (this could be particularly valuable in differentiating stable and unstable angina). The night nurse states that [Mr A] had no pain overnight however [Dr C] subsequently documents that [Mr A] has one bout of pain overnight. It is possible that [Mr A] did not realize the importance of accurately reporting his pain.

On 24/12/98 [Ms G] states that observations were stable but it is difficult to determine if she is referring to the observations she then records (temperature and pulse) or other observations she has taken.

[The hospital] did have a procedure documented at the time on nursing practice in relation to discharge planning ([Ms I], 12/8/01), [Ms G] does not record any action she took in [Mr A's] discharge planning.

10. What documentation would be expected of a nurse who allegedly contacted a doctor and informed [Mr A] that he could go home?

I would expect the nurse to record whom she spoke to (name and position), the date and time of the conversation, any instructions the Doctor had given in relation to the specific patient and what action she (the nurse) had taken following the conversation.

11. Who was ultimately responsible for [Mr A's] discharge and in your opinion was this authority exercised appropriately?

The medical (not nursing) staff were ultimately responsible for [Mr A's] discharge and I do not believe this authority was exercised appropriately as listed below:

It is unclear which member of the medical team, if any, authorized the discharge (see documents J, K, L, N, S, nursing entry 24/12/98).

Repeat blood and ECG testing indicated prior to discharge were not actioned.

[Mr A] was not clinically reviewed prior to discharge by a member of the medical team, which they described as their normal practice (see documents I, J, K, L, S for statements related to normal practice).

Aspects of the treatment plan which were documented e.g. the ETT as an outpatient were not actioned.

A written discharge summary was not completed at the time of discharge with a copy being given to [Mr A] and one forwarded to his GP. I do not consider it is the nurse's responsibility to routinely remind the medical staff to initiate the clinical review and relevant documentation as indicated in document H that records a discussion between [investigation staff] and [Dr C]. However, if a nurse noted that the process for her patient was not being actioned, I would expect her to follow up until it was actioned and [Mr A] received no post discharge instructions from the medical team e.g. related to level of activity.

Another alternative, formally delegating the responsibility for discharging [Mr A] to a member of the nursing staff was not taken. This would have required a documented statement of the parameters within which [Mr A] could have been discharged.

12. Are there any other issues that arise from [the hospital] staff responses and other information provided?

I was concerned that no nurse appeared to be given or accept responsibility for [Mr A] during the early p.m. shift on the 24/12/98. Both Charge Nurse [Ms H] and [Ms G] clearly state the procedure for this, which was not followed."

Independent advice to ACC

Mrs B lodged a claim for medical misadventure cover with ACC. ACC obtained independent expert advice from an interventional cardiologist, Dr Christopher Nunn, who stated:

"Comment: According to the Hospital record [Mr A] had no prior history of ischaemic heart disease. The onset of chest pain was abrupt and recurrent and associated with syncope of sufficient severity to result in bowel and urinary incontinence. This would suggest that a major haemodynamic event had occurred. The presence of ST segment changes on the ECG rhythm strip obtained by the ambulance would confirm cardiac ischaemia as the cause of the patient's symptoms and collapse. This presentation is by definition unstable angina. The syncope could represent either severe ischaemia or a cardiac arrhythmia secondary to ischaemia. [Dr D] in his own report confirmed the diagnosis of angina however he felt this was stable. I believe this diagnosis is incorrect given the nature of the patient's presentation. There was nothing to suggest any degree of stability given the patient's recurrent pain, complete syncope and presence of ECG changes.

The importance of a diagnosis of unstable angina lies in its management. This has evolved over the recent years and may vary to some extent between Hospitals and indeed between countries. I have therefore sought opinions of two other senior cardiologists in constructing this opinion. Unstable angina is generally managed with in-hospital monitoring over a period two to three days. Intravenous or subcutaneous Heparin is administered for part of this period. Exercise testing may then be done either during hospitalisation or shortly thereafter. As part of the management plan at least two serial estimations of cardiac enzyme levels are undertaken with particular attention focused on Troponin levels. The management strategy used for this patient appeared to include none of these components. In particular, a second blood assay for Troponin levels was not taken and this may have identified a high risk patient requiring both prolonged monitoring, heparinisation and probably in-hospital exercise testing.

There was agreement among the cardiologists I consulted that the very minimum level of care for this patient included serial blood assays for cardiac enzymes and monitoring with heparinisation for at least 48 hours.

It can never be determined what potential impact this management strategy may have had on this patient given the fact that he died nine days later, however, the presence of elevated Troponin levels (had they been checked) would have alerted the medical team of the high risk nature of the patient's presentation. Intravenous or subcutaneous Heparin therapy is known to reduce patients' risk of death or myocardial infarction following an episode of unstable angina. Had an exercise test been done which was strongly positive consideration may also have been given towards further investigation including cardiac catheterisation. Revascularisation may therefore have been considered which could also have had an impact on the patient's prognosis.

Specific questions:

1. Has the claimant suffered physical injury as a result of medical treatment?

The treatment per se has not led to physical injury but has resulted in the underlying disease process progressing resulting in the patient's death.

2. On the balance of probabilities, can [Mr A's] death be attributed to the failure of his medical team at [the hospital] to treat or further investigate his cardiac condition, i.e. can a causal link be established?

I believe that [Mr A's] death can be attributed to failure to appropriately manage and diagnose his underlying medical condition.

3. Has medical error occurred as defined by the ACC Act?

In my opinion medical error has occurred in this situation. The medical team has failed to observe a standard care and skill reasonably expected in the circumstances. As stated earlier the minimum of care for this patient would entail 48 hours observation with monitoring and Heparin. The clinical circumstances of the patient's presentation along with the observed ECG changes does not allow for a diagnosis of stable angina to be made. This was clearly highly unstable."

By letter dated 29 November 2000, ACC accepted Mrs B's claim in respect of the failure by medical staff at the hospital to diagnose and appropriately treat her husband's unstable angina resulting in his death on the basis of medical error. The error was stated to be the wrong diagnosis as a result of which correct management was not implemented, resulting in Mr A's death.

On 28 February 2001 Dr C and Dr D lodged applications for review of the ACC decision. Prior to the review hearing, after ACC had obtained further independent expert advice from ACC's cardiologist, revoked its decision of medical error against Dr C. ACC's cardiologist stated:

"The late [Mr A], apparently previously asymptomatic, presented to [the hospital] in the evening of 23/12/98, having experienced an episode of severe upper chest pain accompanied by loss of consciousness and faecal incontinence. The time interval between that attack and his admission to hospital is not recorded in the notes available. I assume that I have been given all of the medical records. He had a further attack of chest pain accompanied by pallor, sweating and faintness before going to hospital and apparently lasting five minutes or so. It is reported that an ECG was recorded by the ambulance personnel which showed ischaemic ECG changes but that is not in the notes provided. [Dr C] admitted him and the Emergency Department who assessed him before that specified that the ECG recorded then was either normal or showed no acute changes. He was admitted at 8 p.m. and had a single set of blood tests including creatine kinase and Troponin T to determine whether cardiac damage had occurred. It is again not specified what the interval was between his first episode of chest pain and the taking of blood.

It appears that there was no further ECG recorded. He had a further episode of chest pain during the night. The next day, he was assessed by the consultant, [Dr D]. The brief notes specified the third episode of chest pain during the night and outline a plan to continue treating him with a Beta Blocking drug (atenolol) and aspirin and to supply him with a Nitrolingual spray and to arrange an exercise test as an outpatient. He was advised to stop smoking. During the period until 1410 hours on 24/12/98, he had telemetric ECG monitoring which was then discontinued and he was discharged. There appears to have been no discharge note prepared for the patient and the definitive summary was not sent to his GP until March 1999.

I infer from the comments by the deceased's widow that he was reasonably well at home and participated in the construction of a fence with his sons and then died suddenly nine days after discharge from hospital.

It is not clear whether there was an autopsy undertaken but the Coroner's conclusion was that death was due to 'coronary artery occlusion due to atheroma'. If this conclusion was based on an autopsy finding, it would have been possible to judge whether he had had myocardial infarction and roughly how long before death that may have been. There is nothing in the records I have received to suggest that information is available.

The preliminary diagnosis here must have been unstable angina based on the information available. The occurrence of a first episode of chest pain, admittedly with some exercise but subsequently two further episodes within a few hours at rest, make that diagnosis firm and the duration of pain does not affect that conclusion. Unstable angina is a retrospective diagnosis once it has been proved with blood tests taken at an appropriate time, that there has not been any heart muscle damage. I must presume that the taking of the single set of blood tests here was too early for any conclusion to be reached in that regard even though they were normal. A second set of tests should have been carried out and a repeat ECG and neither of these actions was undertaken before a decision was made, apparently by [Dr D], to manage things as an outpatient.

It may be a quantum leap to suggest that the error in not reaching a full and proper diagnosis before leaving hospital caused [Mr A's] subsequent death nine days later. Coronary heart disease is an unpredictable disorder and even if he had not had myocardial infarction at the time of his admission (the lack of information does not allow one to draw a conclusion here) it may not have been possible to undertake the appropriate steps which could have led to a better outcome within the available timeframe. That does not exclude the fact that there was medical error here if only in failing to make a full and proper diagnosis. The ultimate responsibility lies in the hands of the senior consultant. Taken at face value and on the basis of the records presented, I cannot see that Dr C failed to fulfil his responsibilities and the house physician who wrote the notes on the day of discharge at the instruction of [Dr D] was merely following those conclusions and directions from his senior.

Apart from the medical error, there appears to have been substantial systems failures in the lack of communication at the time of discharge and that is another issue which again conceivably may have contributed to an undesirable outcome and those failed communication systems cannot be condoned.

My conclusion is that medical error did occur here and that it was the responsibility of the senior physician to ensure that did not occur and there was also an important failure or omission in the systems of communication by [the hospital] (but not unique to that organisation) to pass on information at the time of the patient's discharge. It would be going too far to suggest that the issues raised here necessarily led to the patient's death but one cannot deny that the scenario did nothing to support the best possible outcome."

In a decision dated 17 October 2001, the ACC reviewer granted the application for review and quashed ACC's finding of medical error and negligence on the part of Dr D, essentially for reasons related to reasonable forseeability and causation.

Mrs B has lodged an appeal to the District Court against the review findings in respect of Dr D and has requested ACC to review its decision in respect of Dr C. The appeal in respect of Dr D due to be heard in the District Court has been adjourned pending the outcome of the decision of the ACC review hearing. The ACC review in relation to Dr D and Dr C is pending.

Code of Health and Disability Services Consumers' Rights

The following Rights in the Code of Health and Disability Services Consumers' Rights are applicable to this complaint:

RIGHT 4

Right to Services of an Appropriate Standard

1) Every consumer has the right to have services provided with reasonable care and skill.

2) Every consumer has the right to have services provided that comply with legal, professional, ethical, and other relevant standards.

...

5) Every consumer has the right to co-operation among providers to ensure quality and continuity of services.

Opinion: Breach - The public hospital

Rights 4(1) and 4(5)

Direct liability

I have concluded, in light of the information provided by the hospital, independent advice received from my expert advisors, and the evidence provided by the expert advisors to ACC, that the hospital breached Rights 4(1) and 4(5) of the Code in two respects. First, in relation to the diagnosis and treatment of Mr A's condition; second, in connection with his discharge on 24 December 1998. The breaches in each case lay in failing to offer treatment to Mr A of the appropriate standard of care and skill.

Diagnosis and treatment

Mr A, who was admitted with chest pain and had numerous coronary risk factors, and whose pain recurred after the initial episode, ought to have been treated as a case of unstable angina.  This is the opinion of all independent experts who have assessed the file (my expert advisor, and the advisors to ACC).  Despite this, a second blood test - which acceptable practice would dictate - was not obtained.  The patient was discharged the day after admission in circumstances where (a) there is disagreement between the doctors as to whether a decision to discharge was made; (b) the patient waited until nearly 5 pm, was not seen by a doctor, was sent home without a Discharge Summary, and was not given relevant information for self care; and (c) no relevant outpatient appointment (for exercise tolerance test) had been arranged.

I have accepted the advice of my independent expert that there was both an incomplete assessment and a premature discharge.  The allocation of individual responsibility for that state of affairs is dealt with later in this Opinion.  It is enough to say here that there is an underlying institutional responsibility, and that, to quote the independent expert, "[Mr A] did not receive an appropriate standard of care from the staff at [the hospital]".

In its response to my provisional opinion indicating a direct liability for breach, the District Health Board, through its Chief Executive Officer, stated that the lack of written policies for management of patients such as Mr A should not obscure the fact that such policies existed and were generally followed.  I am prepared to accept that this may well be so, but the fact remains that any such policy was not fully implemented in relation to Mr A and that this was due to a failure by some or all of the relevant personnel.

I am aware that Dr D says he was seeking to exclude myocardial infarction, which is a worse condition than unstable angina.  Dr D says his management plan at the time was to manage as for angina, and that this would equally provide protection for unstable angina, myocardial infarction and (though this was unlikely to be present in Mr A) dissecting aortic aneurism.  I acknowledge that, on the assumption that Dr D did not intend Mr A to be discharged on 24 December, his management was acceptable.  All this can be accepted, but the fact still remains that, notwithstanding this plan, Mr A was discharged without completion of the diagnosis and without adequate instruction for self-care and plans for appropriate treatment and testing.  The effective reason for discharge appears to have been miscommunication amongst the doctors for which, for reasons given below, I find Dr D responsible.  The present point is that there was a breach by the hospital of its duty to provide services to Mr A with reasonable care and skill.  This was a serious breach, resulting as it did in the premature discharge of a patient, whom experts agree presented with symptoms of unstable angina.

Discharge

The hospital acknowledges that failings occurred in relation to the discharge of Mr A.  It says, however, that the events should be judged by 1998 and not 2002 standards.  I agree with this, but it makes no difference given the hospital's acknowledgment that even the process said to be in place in 1998 was not followed.  Dr F, Manager of Clinical Services at the hospital, acknowledged at the earliest opportunity (letter of 31 March 1999 to Mrs B), that in relation to the discharge of Mr A there was "a serious failure of process which seems to have been due to communication between two house surgeons one of whom was a reliever".

The failings in relation to the discharge include:

(a)   failing to ensure that Mr A was seen by a doctor before his discharge and given appropriate information and an outpatient appointment, and allowing him to leave the hospital;

(b)   failing to ensure that an afternoon nurse was assigned to Mr A before he left the ward;

(c)   failing to properly document the discharge plan and to provide a discharge summary.

The failings, as I explain below, came about because the registrar and house surgeon formed the view that Mr A's discharge had been ordered by Dr D.  They may well have been wrong in reaching that view, as explained below, but that is not the important point for present purposes.  The fact is that, on the assumption that Mr A was to be discharged, there were failings in the discharge process.  These failings, I am satisfied, are principally the responsibility of the house surgeon, Dr E, and the nurse who ought to have been responsible for Mr A on the afternoon shift.  Had these processes been followed, then the risks to which Mr A was exposed by being discharged might to some extent have been mitigated.

A question arose in the investigation as to the responsibility of the registrar, as opposed to the house surgeon, in completing the physical discharge of patients.  The usual allocation of responsibility was not entirely clear, although I am prepared to accept that in the normal course, the expectation would be that the house surgeon would prepare the discharge summary and see the patient prior to discharge.  The hospital has stated that, although no written policy has been located as to allocation of responsibility for discharge as between house surgeon and registrar, there was a clear process in place that "seems to have worked well in the vast majority of cases" (letter, 15 April 2002).  That may well be so, but it does not mean that the failings in relation to Mr A's discharge are excusable.  The purpose of polices and processes is to avoid events such as the ones that occurred in this case.  It is the fact that the failings occurred that constitutes the breach of the duty to provide treatment with reasonable care and skill.  That there may have been a process in place that avoided this in most cases does not alter the fact that the failings occurred in this case.  In making the finding of institutional breach it is not necessary for me to apportion blame to the individuals concerned; certainly not in the face of the hospital's admission that "for reasons that are unclear, this process does not appear to have been followed in [Mr A's] case" (letter, 15 April 2002).

In my opinion Mr A was not provided with adequate continuity of care.  His discharge process was fragmented, reflecting a lack of communication and co-operation amongst medical, nursing and clerical staff.  Because his discharge summary was not completed, his GP was never advised of Mr A's admission to the hospital and what post-discharge follow-up was required.

Accordingly, I find the hospital in breach of Rights 4(1) and 4(5) of the Code in relation to the treatment and discharge of Mr A.  It breached its duties of care and co-ordination through the actions of the various medical personnel involved.  I return at the end of this Opinion to the consequences of that finding.

Opinion: Breach - Dr D

Rights 4(1) and 4(5)

The allocation of responsibility to particular individuals is made difficult in this case by disagreement over an important fact.  At the ward round on December 24 it appears that Drs C and E, and Mr A, each gained the impression that Dr D had said that Mr A was to be discharged that day.  Dr D is adamant, however, that he gave no such instruction.  He says that a decision on discharge was still to be made, and that it would have been made in light of the results of the repeat blood tests (which in fact had not been ordered but which Dr D says he assumed would have been initiated before the morning ward round, albeit that the results would not have been available at the ward round).

I have reviewed the information provided to me by all relevant parties, and their responses to my provisional opinion.  It is not possible for me to conclude that one version of events is correct and that the other is mistaken.  The circumstantial evidence supports the conclusion that each of the three doctors is reporting the events to the best of their recollection.  In particular, I attach significance to the fact that Drs E and C each formed the view that discharge had been approved and were quite clear about this.  In the case of Dr E this is an "admission against interest" in the sense that Dr E then has to explain the fact that he failed to attend to any of the tasks associated with the discharge save for writing a prescription for medication.  In the end, Mr A was discharged without seeing any doctor and without appropriate documentation.

As to Dr D, I accept his explanation that he did not order the discharge of Mr A.  I accept that, in light of his treatment plan initiated the previous evening, he would not have ordered discharge without a review of the repeat blood results, whether by him or another doctor.  That said, it is inescapable that the discussion at the ward round created the impression in the minds of all others present that Mr A was to be discharged that day.  The likely explanation for this is probably that some fundamental points were left unsaid because of Dr D's belief that a "usual practice" was being followed - namely, the ordering of repeat blood tests and their subsequent review with a view to dischargeif then appropriate.  It is conceivable that, with this unsaid, a discussion about discharge could have been interpreted differently by the two sets of participants.  On the one hand, when speaking of a pending discharge, Dr D may have been looking ahead to a time when the repeat tests were available and indicated to the doctors no sign of myocardial infarction.  On the other hand, Drs C and E were aware that no further blood test results were awaited, and they may reasonably have assumed that discharge (with medication and an outpatient exercise test to be arranged) was to follow without more.  This explains their subsequent actions and explanations (albeit that in the case of Dr E he subsequently failed to see Mr A before his discharge - a failure that, significantly, he does not seek to explain by contending that he did not think that Mr A was to be discharged).

I think the account I have offered is the most likely reconciliation of the differing interpretations of events on 24 December.  It raises the following questions, however.

Was there a "usual practice" for repeat blood tests upon which Dr D could reasonably rely?  It would seem not, for Dr C advised me that on Mr A's admission on 23 December he had not ordered repeat blood tests, in light of his belief that Mr A would go to the Acute Assessment Ward and that they would be ordered there.  But Mr A did not go to that ward, and no further tests were in fact ordered.

Dr D asserts that the essence of the problem was the omission by Dr C to have ordered these repeat blood tests.  In his response to my provisional opinion, Dr D comments on the independent expert's critical observation that each of Drs C and D "seem to have believed that 'usual practice' would dictate management beyond the initial ward round".  Dr D states: "[T]he need for repeat tests was not only a well known standard procedure in the course of treating what was a common cardiac complaint, but was also clearly documented in the notes made by [Dr C] on the night of 23 December, the need for cardiac enzymes to be repeated is also clearly stated in the result of the initial tests issued by the laboratory."

It is, I think, enough to point out that "usual practice" was not in fact followed, and that whatever the reason, this is something for which Dr D as consultant must ultimately bear responsibility.  That there may have been a combination of errors does not alter that fact.  I accept the advice of my independent advisor that the consultant responsible for a patient's care must ensure that appropriate standards are followed.  Here I believe that the failings of Dr D were (a) in not establishing, at the ward round on 24 December, whether repeat blood tests had been carried out and that results were awaited; (b) not making it clear that there could be no discharge of the patient without the additional tests that were necessary to make an accurate diagnosis; and (c) allowing, as a result, a state of affairs to arise in which an erroneous assumption was made by his medical team that Mr A was to be discharged that day without more tests.  I accept my advisor's conclusion that reliance on "usual practice" was "a dangerous assumption in the absence of written clinical guidelines or pathways, particularly when relieving junior doctors, who may be unfamiliar with 'usual practice', are involved with patient care".

It follows that I have concluded that Dr D is in breach of Right 4(1), the duty to treat with reasonable care and skill, and Right 4(5), the duty of co-operation among providers to ensure quality and continuity of care.

Opinion: No breach - Dr C

Rights 4(1) and 4(5)

Dr C was the acute admitting registrar who first saw Mr A on admission.  Dr C formed the impression that Dr D had ordered the discharge of Mr A at the ward round on 24 December, having diagnosed angina and indicated the need for an outpatient ETT and a prescription for aspirin and atenolol.  Dr C says that although he would have kept Mr A in hospital to repeat the blood tests and ECG, he agreed that discharge was appropriate management as the "overall clinical picture was that of angina".

As to this, it is apparent that Dr C appreciated the need for repeat blood tests to ensure an appropriate diagnosis.  However, in the face of what he perceived as a direction from Dr D that the patient was to be discharged - and his own assessment that this was "appropriate management" ­- he raised no objection.  While the independent expert criticises Dr C along with Dr D for making the assumption that "usual practice" would dictate management beyond the ward round, I am satisfied that Dr C cannot be implicated in this particular criticism.  It is plain from Dr C's responses to my provisional opinion that he did not envisage the discharge to be conditional on any further investigation or result of any blood test (indeed Dr C would have known that no results were awaited).  So Dr C's case is essentially that, although he himself would have preferred to retain the patient for repeat tests and ECG, he was not prepared to contest the directions that he believed had been given by Dr D for discharge because he felt they were appropriate.

On balance, I have concluded that Dr C acted reasonably in this regard.  In the light of a perceived indication of his consultant's treatment plan, I do not think Dr C should have acted differently.  In saying this, I should make it clear that I also accept that Dr D did not intend Mr A to be discharged without further consideration in light of the repeat blood tests.  My finding in relation to Dr D was that, as the responsible consultant, he breached the standard of due care and skill by relying on usual practice and allowing the confusion to develop.  Against that background, it is not inconsistent to assess the culpability of Dr C on the basis that he (Dr C) could reasonably have formed the view that Dr D was happy for Mr A to be discharged without further tests.

On this view, the question then becomes whether Dr C is in breach for not having ordered the repeat blood tests on the previous evening.  In his statement to me, he accepts the need for such tests, but explains that he did not himself order them on 23 December 1998 because he thought they would be ordered on the Acute Assessment Ward to which Mr A would be transferred.  In this state of affairs I think the real issue arose when, the following morning, it became apparent to Dr C that there had not been a second set of tests (because Mr A did not go to that ward).  But it is at that point that Dr C formed, albeit possibly wrongly, the impression that Dr D wished Mr A to be discharged that day without further blood tests.  If it was reasonable for him to accept that course of action - and I have determined above that it was reasonable, since Dr D was the consultant - then there can be no finding of a breach of Right 4(1) by Dr C on that occasion.

In this regard I have taken into account the fact that Dr C told me that, in his view, Mr A did not, clinically, present with myocardial infarction or unstable angina.  The advice of the independent experts would suggest Dr C was wrong in his view.  Further, Dr D has explained that from his point of view Mr A was being treated while in hospital as an angina patient but in such a manner that he would equally be protected if his condition was unstable angina, myocardial infarction, or aortic aneurysm.  This suggests that Dr C's own assessment may not have been shared by Dr D.  But I have formed the view that when, at the ward round, Dr C gained the impression from Dr D that Mr A was to be discharged, he acted reasonably in implementing that decision given that it was a decision he agreed with.  (The real concern, of course, is that it appears that a discharge of Mr A was not really Dr D's intention at all, but that confusion arose because of unstated assumptions about a "usual practice".)

The next area to explore in relation to Dr C arises from the fact that a doctor did not see Mr A prior to his discharge and that no discharge summary was prepared.  In this regard, I accept that usual practice is for discharge summaries to be prepared by the house surgeon when there is no further clinical decision to be taken.  Accepting, as I do, that Drs C and E each believed Mr A was to be discharged with a prescription of medication and arrangements for an ETT, it was reasonable for Dr C to assume that Dr E would undertake the necessary work.  As to the prescription, it is apparent that Dr E duly completed it and that this was within his sphere of competence.

Accordingly I find there to have been no breach by Dr C in respect of Rights 4(1) and 4(5) of the Code.

Opinion: Breach - Dr E

Right 4(1)

Dr E was the house surgeon on duty.  His first involvement with Mr A was attendance at the ward round on December 24.  Although Dr E shared the view of Dr C that Mr A was to be discharged, Dr E did not, as would have been normal practice, prepare the discharge summary; nor did he ensure arrangements were made for the ETT.  He left for the day without seeing Mr A.  It seems he did write a prescription, although he has no recollection of doing so.  Dr E accepts that normal practice would have required that he see Mr A.  His explanation for his failure was that he had a "system I used to ensure that I saw all patients for discharge".  This, says Dr E, was not just his system but also one used by other house surgeons.  It consisted of using the whiteboard at the nurses' station.  Dr E believes that the reason he did not see Mr A was that someone had removed Mr A's name from the whiteboard.

There is no evidence upon which I could accept this explanation, and in any event it is not an excuse.  Whether or not there was a system, the fact is that Dr E's failure to attend to a patient scheduled for discharge and to complete a discharge summary and schedule appropriate tests is a lapse from the standards of due care and skill.  Other than explaining the nature of the system, pointing to the possibility of a name being erased from the board, and pointing to the failure of nursing staff to remind him, Dr E has advanced nothing to justify this lapse.  As a result of this lapse Mr A left the hospital without medical clearance, and without necessary documentation and information regarding his follow-up care.  In consequence, follow-up tests were not arranged and Mr A's GP was not made aware of his admission to hospital or any follow-up care required.

Accordingly, in my opinion Dr E breached Right 4(1) of the Code.

Opinion: No breach - Ms G

Right 4(2)

Ms G was on duty until 3.30pm.  She was not present at the ward round when Mr A was being dealt with by Drs D, C and E, but shortly afterwards she enquired whether he was to be discharged and received the answer that he was.

In my provisional opinion I expressed the view, based on the advice of my expert advisor as to nursing practice and standards, that Ms G's subsequent performance of her duties was in breach of Right 4(2) in:

(a)   not providing information to Mr A on heart disease, how to manage cardiac pain, and associated topics

(b)   not documenting an overview of the nursing management plan for Mr A, and not following the hospital's procedure relating to nursing practice regarding discharge planning.

I have reconsidered this finding in light of the response provided by Ms G.  I accept that, in the circumstances of this case, where Ms G was expecting that Mr A would be seen by a doctor before discharge, it was reasonable for her to consider that the information in (a) above would be transmitted by the doctor, if it had not been already.  I am inclined to think that, in this case, the advice of my expert advisor is a counsel of perfection.  It is plain that the real lapse on 24 December was brought about by the failure of Dr E to see Mr A, and the failure to ensure that a nurse coming on to the afternoon shift was allocated responsibility for Mr A.

In relation to the standard of documentation of the nursing management of Mr A, I accept that a number of the criticisms made by my nursing advisor relate to periods for which Ms G was not responsible (eg, no overview of nursing management plan on 23 December 1998, nor statement to Mr A by the night nurse on the importance of reporting pain promptly).

In all the circumstances I am satisfied that Ms G did not breach Right 4(2) of the Code.

Opinion: No breach - Charge Nurse Ms H

Rights 4(1) and 4(5)

Ms H was the Charge Nurse Manager of the ward in December 1998.  In my provisional opinion I expressed the view that Ms H was in breach of Rights 4(1) and 4(5), principally for failure to ensure adherence to nursing standards relating to the discharge of Mr A.  The failings suggested lay principally in the lack of documentation making it impossible to say which nurse had been assigned to Mr A for the afternoon shift, and which nurse had approved Mr A's departure from the hospital.

At the time of writing my provisional opinion I had not had the benefit of seeing the Discharge Planning nursing/midwifery standard that was current in 1999.

On consideration of this document and in light of Ms H's comments on my provisional opinion I accept that, although there was a significant failure to meet these standards, this does not in itself amount to a breach by Ms H of Right 4(1).  There was a serious lapse in relation to Mr A on 24 December 1998, but there is no evidence that this was due to a nursing system failure as opposed to the isolated failings of individuals on that day.  Nor is there evidence that inadequacy in care was brought to Ms H's attention on that day.

Ms H admits that Mr A was inappropriately discharged.  She says this was due to a failure by medical staff to return to the patient, compounded now by incomplete documentation.  The incomplete documentation she alludes to is the lack of documentation of the afternoon nurse who assumed responsibility, and of the nurse who gave approval for Mr A to leave the hospital without being seen by a doctor.  Ms H sees this as an issue of documentation rather than substance, and believes that no nurse would have sent Mr A home without a clearance from a doctor.  Unfortunately it is impossible to tell whether a clearance from a doctor was in fact obtained, and the lack of documentation is part of the problem.

On balance, I am prepared to accept that Ms H is not responsible for the lapse of standards that occurred while Mr A was in the ward.  According to systems then in place, a nurse should have been assigned to care for Mr A on the afternoon shift (and her name recorded on the ward whiteboard), a doctor should have seen Mr A, and the discharge procedure should have included written documentation.

It is not clear whether a nurse was assigned to Mr A.  Mrs B says that no nurse came to see him.  However, the critical fact is that Mr A was not seen by a doctor, and the documentation for his discharge was inadequate.  These are failures for which the hospital must bear direct responsibility.  I find no breach by Charge Nurse Ms H.

Opinion: Breach - The public hospital

Vicarious liability

In addition to any direct liability, employers may also be vicariously liable under section 72(2) of the Health and Disability Commissioner Act.

In my opinion the hospital is vicariously responsible for the breaches of the Code of Rights committed by Dr D and Dr E.  An employer has a defence if it can show that it took such steps as were reasonably practicable to prevent the employee from committing the act or omission that breached the Code.  In my opinion this defence is not made out on the facts of this case.  The lapse from reasonable care that I found to have been committed by Dr D came about as a result of confusion generated by unwise reliance on a "general practice".  The breach by Dr E came about because of an unexplained omission to see a patient.  It seems that in each case there were no documented guidelines and policies that might have prevented such breaches occurring.

Accordingly, I find the hospital vicariously liable for the breaches of the Code by Dr D and Dr E.

Other comments

I have made no finding of fact in respect of the nursing care provided to Mr A.  I could not establish that no nursing handover took place or that no afternoon nurse was allocated to Mr A's care.  What has been established is that there was no documentation in the nursing notes for the afternoon shift and that Mr A left the ward without appropriate discharge information and follow-up.

It is fundamental to good nursing practice to provide nursing care until a patient leaves the ward.  It is not appropriate to go off duty without ensuring that all discharge matters have been completed and documented or that a nurse coming on the following shift has taken over responsibility.

Actions taken

I am informed that since December 1998, as a result of Mrs B's complaint, the hospital / the District Health Board has taken the following actions to minimise the likelihood of a similar adverse event in the future:

• Introduced a change from largely service and profession-based orientation programmes to centralised corporate orientation of staff through the Human Resources Department, with individual service areas retaining their responsibility for much of the clinical orientation of its new staff.
• Created a position of a Registered Medical Officer (RMO) Unit co-ordinator who, in conjunction with the Learning and Development team provides orientation to new junior medical staff at The hospital, and aims to meet their training and educational needs.
• Developed an "RMO Project" which is designed to monitor and improve information to, and training needs of, junior medical staff.
• Issued (in March 2000) a revised policy regarding patient discharge planning.
• Issued (in April 2001) a revised policy regarding patient discharge processes at the hospital.

Recommendations

I recommend that:

• The District Health Board, Dr D and Dr E apologise in writing to Mrs B for their breaches of the Code.  These apologies should be sent to my Office and will be forwarded to Mrs B.
• The District Health Board, in light of this report, review its current guidelines for the management of patients with cardiac chest pain, consider developing clinical pathways for the management of patients with cardiac chest pain, and make any necessary changes to current practice to minimise the likelihood of similar adverse events in the future.
• The District Health Board, in light of this report, review its current procedure regarding co-ordination of patient care (assessment, discharge planning and follow-up) in light of this report, and put in place measures to minimise the likelihood of a similar adverse event in the future.
• The District Health Board, in light of this report, review its training and orientation procedures for new and existing staff, to ensure that patient discharge planning and discharge processes are clearly understood by medical, nursing, clerical and other relevant staff.

Other actions

• In light of the review of the cardiac services at the hospital by the Ministry of Health in April 2000, a copy of this opinion will be sent to the Director-General of Health.
• A copy of this opinion will be sent to the Medical Council of New Zealand and the Nursing Council of New Zealand.
• A copy of this opinion, with personal identifying features removed, will be sent to the Royal Australasian College of Physicians, the New Zealand Resident Doctors' Association, the Deputy Director-General, Clinical Services (for distribution to the Chief Medical Advisors of all District Health Boards), and the Nursing Council of New Zealand (for distribution to appropriate nursing organisations), and placed on the Health and Disability Commissioner website, www.hdc.org.nz, for educational purposes.