Reports into breast cancer misdiagnoses fail to address equipment or staff
Remarkably, neither were investigated in the three reports produced by the Department of Health and the Health Service Executive yesterday.
Author of one of the reports, Dr Ann O’Doherty, tasked with reviewing the mammography service from a medical point of view, said images produced by the mammogram machine were of poor quality and hampered diagnosis, a problem she put down to the processing methods rather than the actual machine.
But she didn’t get to test the machine because it was decommissioned when the cancer services at Portlaoise were suspended after the problems there became public last August.
She wasn’t told how old the machine was although she guessed about 15 years old — about five years past normal retirement age — and she received no information about the film processing system either.
Her report states: “It is difficult to provide a definitive report... A comprehensive review would necessitate a visit to the unit, inspection and quality-assurance measurements of the X-ray and film processing equipment and an interview with relevant staff.”
The limited terms of reference laid down by the HSE for her review precluded her from doing any of that. They were the same terms of reference that prevented her from speaking with Dr Peter Naughton, the consultant surgeon who first raised concerns about the service in 2005 — ironically, because he found too many women were being wrongly diagnosed with cancer and subsequently cleared rather than the other way around.
He wrote to her about her review, stressing his failed attempts to get improved facilities and specialist staff, but neither she nor the other report authors could interview him.
Neither could they interview Dr Visalatchee Moodley — the radiologist with the highest number of mistaken diagnoses, who was placed on administrative leave when the service was suspended.
Dr Moodley wanted input into the reviews and raised issues through her legal representatives but she has returned to work at the hospital without getting a chance to formally air her views on what went wrong.
The lack of involvement by key staff is a theme repeated throughout the three reports. Ann Doherty of the HSE’s National Hospitals Office was tasked with reporting on the administrative steps taken in the lead-up to the decision to suspend services at Portlaoise and arguably she should have been able to show whether that decision was taken quickly enough with good enough reason and sufficient consultation.
But her report shows the decision to designate Portlaoise as a regional cancer centre — a decision since revoked — was taken in opposition to best medical advice, that medical staff did “not regularly participate in management” at the hospital, did not have any clear channels of communication with management and were not around the table when the decision to suspend services was made last August.
Although she does not state it so baldly, it is clear from her report that the environment, atmosphere, structures and procedures were not conducive to medical staff’s concerns or criticisms, constructive or otherwise, being relayed to hospital management in any form that could or would be taken seriously.
Staff weren’t the only ones kept out of the loop. When the problems at Portlaoise came to a head last August, the needs of patients potentially affected by the suspected misdiagnoses were overlooked in the ensuing panic.
The third report published yesterday, that of former Dublin city manager, John Fitzgerald, examined the way the crisis was handled concluded there was “a fundamental weakness in the management and governance” of the situation.
John Fitzgerald said all patients whose cases were being reviewed should have been written to at the beginning of the process but this did not happen. Most heard what was going on from media reports.
Worse, women were kept in the dark about an additional ultrasound reviews they might need until a HSE official unexpectedly spilt the beans at an Oireachtas Health Committee meeting.
Yesterday, Health Minister Mary Harney said lessons had been learned from the episode and a draft “serious incident protocol” was being adopted with immediate effect.
When finalised, it will establish the procedures to be followed in informing patients in similar situations in future, in releasing information to the general public, arranging reviews and providing counselling and other supports to those affected.
“Patients’ interests will come first,” Ms Harney said.
Given the sorry state of hospital facilities, staffing and management revealed by the reports, many commentators would say that would indeed be a first.
An examination of the circumstances that led to the HSE’s decision in August 2007 to suspend breast radiology services at Portlaoise Hospital, initiate a clinical review of the symptomatic breast radiology service, and place a consultant radiologist on administrative leave.
Ann Doherty, national director, corporate planning and control processes, HSE (not to be confused with Ann O’Doherty of St Vincent’s who wrote another of the three reports), Barry O’Brien, assistant national director, human resources, HSE South and Mary Shore, director of quality, risk and consumer affairs, St Vincent’s University Hospital, Dublin.
This report did not attempt to make findings. Instead, it concentrated on establishing the facts that led to the three HSE decisions outlined above.
The circumstances that led to the first two decisions — suspending breast radiology services and ordering a review — were the same.
In late June 2007, the clinical nurse specialist in Oncology/Breastcare expressed concern to the director of nursing regarding the time patients had to wait for diagnosis.
In some instances, patients were waiting “between four to eight months for an outcome”, according to the report.
The clinical nurse specialist also reported eight instances where diagnosis differed in reports from Portlaoise Hospital and St Vincent’s University Hospital, where some mammograms were sent for second opinions.
After the clinical nurse specialist’s concerns went through the necessary channels, a meeting was arranged to discuss the issues on August 28.
At that meeting, the network manager took the decision to suspend the breast radiology service. No one raised objections.
At the same meeting, concerns regarding the eight false positive tests were discussed. The network manager took the decision to initiate a review of the breast radiology service as a result.
The third decision — to place the consultant radiologist on leave — was not taken at that meeting, as the network manager “felt it was not appropriate to discuss an individual’s performance at such a meeting”.
Instead, the matter was discussed the following day between the network manager, the assistant national director for quality, risk and customer care, and the national director of the National Hospitals Office.
All three expressed reservations as to whether administrative leave was warranted. But on August 30, the national director “as the senior decision maker, made a judgment call, on the balance of the available information, to place a consultant radiologist on administrative leave, in the interests of patient safety”.
No recommendations made, but several worrying facts were documented in the report. One such fact, pertaining to another hospital, leapt from the pages.
Oncology services in the midlands were essentially spread across three hospitals: Tullamore, Mullingar and Portlaoise.
Following a report of a failure to diagnose fractures from a cervical scan taken at Tullamore Hospital in January 2006, a risk management review was undertaken.
Completed in January 2007, it recommended that a clinical performance review be carried out. But by August 2007, this clinical review had not taken place.
“It is particularly alarming that clear warning signals of problems at Portlaoise were not acted upon and that, for instance, the recommendations of a risk management review into an earlier ‘incident’ was not acted upon” — Labour health spokeswoman, Jan O’Sullivan.
The clear pattern emerging from this document was in keeping with the other two reports: communications failures, mismanagement and other glaring problems within the HSE.
The report examined how the HSE arrived at its decisions to suspend breast radiology services at Portlaoise, initiate a clinical review of those services, and place a consultant radiologist on leave.
It found there were no multi-disciplinary team meetings in relation to patients attending the symptomatic breast services in Portlaoise.
In other words, there was no triple assessment, which is considered international best practice.
Under triple assessment, the radiologist who takes the mammogram, the surgeon who removes the lump and the pathologist who examines it under lab conditions, must all agree on whether it is benign or cancerous.
The procedure is designed to minimise the risk of misdiagnosis of patients.
The report also found that no consultants were invited to the August 28 meeting at which it was decided to suspend the breast radiology services and initiate the review.
The report found there were “differing opinions as to why consultant medical staff do not regularly participate in management” at the hospital.
It also found some hospital management were unclear as to whom consultants and other staff should report to. There were wider issues, too.
The national cancer services strategy published in 1996 established Tullamore as the lead hospital for cancer treatment in the midlands region, with services also to be provided at Mullingar and Portlaoise.
The report found this was not ideal for the provision of breast disease services.
“Acute hospital services in the HSE Midland area are described to the review group as a single hospital on three sites — Midland Regional Hospital Tullamore; Midland Regional Hospital Mullingar and Midland Regional Hospital Portlaoise.
“The model for the provision of symptomatic breast disease services in the former Midland Health Board is at variance with the National Cancer Strategy.
“The review group was informed that this model led to problems with recruitment and fragmentation of services.”
Perhaps most seriously was the lack of a “formal reporting relationship” between the quality and risk management division and hospital management.
The quality and risk division had recommended in January 2007 that a clinical performance review be carried out at Tullamore.
This stemmed from a report of failure to diagnose fractures from a cervical scan. But seven months later, in August 2007, this performance review “remained outstanding” — in other words, had not been done.
It reports on a review carried out of 3037 mammograms taken by seven radiologists at Portlaoise Hospital between November 2003 and August 2007, plus 607 breast ultrasounds carried out at the hospital between August 2005 and August 2007.
That review was ordered after concerns were expressed by a member of the hospital’s nursing staff over the length of time some women were waiting to receive results because mammograms taken at the hospital were being referred to St Vincent’s Hospital in Dublin for a second opinion.
Second opinions were being sought since 2005 after a consultant surgeon at Portlaoise raised concerns about the quality of mammograms taken there. He had noticed that mammograms were wrongly being read as indicating cancer when in fact, the women involved were healthy .
In investigating the concerns about the delays, the director of nursing found a more worrying problem — that some women who had been given the all-clear at Portlaoise were subsequently discovered to have cancer. These findings were alerted to hospital management and prompted the review.
It was compiled by Dr Ann O’Doherty (not to be confused with Ann Doherty, acting head of the National Hospitals Office) who was also one of the three external consultants assigned to carry out the review. Dr O’Doherty is a consultant radiologist at St Vincent’s Hospital in Dublin and clinical director of the BreastCheck screening unit located on the hospital campus.
The review of the ultrasounds found that they had been correctly interpreted and there were no missed tumours.
However, 235 of the 3,037 mammograms were found to require further investigation because they were unclear.
Of those 235, there were 14 women who had previously been given the all-clear and were now found to require biopsies.
Nine of those women were found to have cancer and the delay in them receiving that diagnosis ranged from four and a half months to two years and nine months.
One of those women, considered high risk because of a family history of breast cancer, WAS noted to have an abnormality at Portlaoise in 2006 and again in 2007 but was still not properly diagnosed because she was not referred for biopsy.
The quality of the mammograms was a major concern. Dr O’Doherty described them as “patchy”, a problem she put down to the processing of the films rather than a fault in the mammogram machine itself although she did not get to examine the machine because it was decommissioned prior to the review.
She also criticised the quality of the reports provided by the radiologists who carried out the mammograms to other medical staff such as oncologists, surgeons and pathologists, who had to decide the course of treatment, if any, required for the patients.
The reports were “difficult to interpret” and “lacked clarity, specificity and helpful conclusions”.
Triple assessment, the recommended practice whereby all diagnoses must be discussed and agreed by a trio of experts including a radiologist, oncologist/surgeon and pathologist, was not in use at Portlaoise Hospital.
Image guided biopsy, the recommended practice whereby biopsies are carried out with the assistance of ultrasound or x-ray to minimise the need for surgery, was also not in use. The number of false all-clears was not excessive when compared to international rates in hospitals of a similar kind.
“The safety, quality and standard of many aspects of the breast imaging service at the Midland Regional Hospital, Portlaoise over the period between November 2003 and August 2007 fell well below achievable best breast imaging practice and this has resulted in a significant and avoidable delay in the diagnosis of breast cancer.”
In reviewing the mammography services at Portlaoise, Dr O’Doherty found they lacked many features essential for a well-run and reliable cancer centre.
Mammographies are printed on film, much the same as an X-ray, and they have to be processed cleanly and clearly to be read correctly but that basic function was too often botched.
Some films were of such poor quality that it wasn’t sufficient to just have them re-read — they had to be scrapped and repeat mammograms carried out yet women were wrongly given the all-clear or unnecessarily referred for biopsy from those films.
Radiologists who took the mammograms could have flagged problems with the images in the accompanying notes they prepared for the doctors of each patient but the quality of those notes was also often poor.
Dr O’Doherty herself found them difficult to interpret. She said abnormalities were not properly reported and sometimes, open surgical biopsy — as opposed to the less invasive needle biopsy — was suggested on what were in fact healthy women. Those were the technical and professional flaws and bad as they were, they were exacerbated by a poor system of checks and safeguards in the hospital which did not employ the ‘triple assessment’ procedure.
Significantly, Dr O’Doherty could not recommend that mammography equipment be upgraded, that staff be retrained or procedures be tightened at Portlaoise because breast cancer services were suspended there last August and will not be returning.
She did, however, emphasise the anxiety and stress caused to women whose mammograms were reviewed, both those who were cleared and those who received the bad news that they had in fact got cancer.
Wherever cancer services were carried out, women needed to be confident that the care they received was of the best possible standard. “A mechanism for measuring standards in each centre and comparing practice between centres should be devised rapidly,” she said.
“Implementation of these standards would, in my view, significantly reduce the likelihood of such an occurrence of sub-standard care in the future.”
WHAT IT WAS: A review of the management of all the events starting from, and subsequent to, the decisions by the Health Service Executive to suspend the breast radiology service at the Midland Regional Hospital Portlaoise, place a consultant radiologist, Dr Visa Moodley, on administrative leave, and initiate a clinical review of breast cancer services at the hospital.
It was requested by Health Minister Mary Harney following a public outcry at the manner in which it emerged 97 women who had undergone ultrasounds at the MRHP Portlaoise Hospital were being recalled for further investigation.
Details of the recall were announced unexpectedly by a HSE official at the Oireachtas Committee on Health and Children on November 22. Up to that point, the focus had been on a review of more than 3,000 mammograms read at the hospital. Following the ultrasound recall announcement, it emerged neither the women themselves, nor Ms Harney nor HSE chief Prof Brendan Drumm, were aware of this second review.
On foot of this embarrassing admission, Ms Harney wrote to the chair of the board of the HSE requesting an immediate review of management of events from the time the HSE took the decision to suspend the breast radiology service at MRHP. The review was to focus, in particular, on the manner in which the HSE dealt with patients throughout the period; the governance and management of all aspects of the process; and the communication arrangements with patients, internally within the HSE, and with the Department of Health and Children. The report is entitled Management, Governance and Communications issues arising from the Review of Breast Radiology Services at Midland Regional Hospital Portlaoise.
John Fitzgerald, former Dublin city manager, previously tasked with planning the regeneration of parts of Limerick city.
The Department of Health was given details at least twice in relation to the ultrasound review — once in a briefing note to the minister’s advisor on September 28 and again on November 6 in material for a ministerial statement sent by the HSE to the department. This is contrary to claims by Ms Harney that neither she nor her department were aware of the ultrasound review until November 21.
* There was a lack of urgency in the response from corporate and area HSE to the review process prior to the revelations at the Oireachtas Committee on November 22. It was only after November 22 that a level of urgency was introduced “despite the fact it was known these patients were in a much lower risk category” Mr Fitzgerald said.
* Mr Fitzgerald found “a fundamental weakness in the management and governance” of the review process from the outset. He was critical of the Review Facilitation Group set up to coordinate and manage the process. It met on three occasions which Mr Fitzgerald said did “not reflect the kind of urgency and level of oversight that should have been accorded a matter of this importance”. The group did not sign off terms of reference or protocols for the review process. It did not exert control over the integrity of the communications process either with patients, with the department, or internally. Its focus was on the mammography and not the ultrasound review which ultimately created the situation where not even the Health Minister nor the head of the HSE was aware this second review was underway.
* The consequences of this lack of overall management and governance meant communication throughout the period was inconsistent, and sometimes contradictory. Different people in the system seemed to have different understandings of what was going on, contributing to confused communications.
* There was a lack of clarity about the nature and status of the terms of reference and methodology. Had these been clearly established and signed off at the outset by a group or individual that was clearly in charge of the process as a whole, then much subsequent confusion could have been avoided.
* There were too many people involved from different levels and areas within the HSE without clarity about their roles and responsibilities. Throughout the process there was insufficient clarity about whom, if anyone, had the most complete or up-to-date information, or who was responsible for providing definitive information on the status of the overall review.
* The decision-making process was fragmented, with insufficient clarity about decisions, who was making them, why they were being made, or when they were signed off.
* There seemed to be an issue about prioritising, and particularly the need for dedicated resources to be devoted exclusively to manage critical incidents such as this.
* There was a delay in retrieving approximately half of the 600 ultrasounds under review which raised concerns about the local administrative management.
* There was inconsistency and lack of clarity in the information provided in the press releases. The “drip feed” of patient numbers could only have heightened anxiety and uncertainty for those patients potentially affected.
* All patients whose cases were under review should have been written to at the outset, but were not. Instead it was decided to contact those who needed to be recalled.
* Mr Fitzgerald said he had concerns about the quality of the communication process between the HSE and the department, partly due to the multiplicity of communication channels in operation involving different people within the HSE, and different people within the department. He believes a more shared approach between the department and the HSE could have avoided much of the negative portrayal of the health services that followed the November 22 revelations.
“In summary, the communications difficulties that arose cannot be separated from the weakness of management and governance in the process. In the midst of the intense activity surrounding the review, the needs of the patients potentially affected receded,” Mr Fitzgerald said.
“While, as already noted, there were significant and competing pressures on many of the people who were involved in this process, in the final analysis my assessment is that fundamentally the problems arose from systemic weaknesses of governance, management, and communication for dealing with critical situations such as arose at MRHP in late August 2007. These are the issues that need to be tackled to avoid a recurrence.”
CONNECT WITH US TODAY
Be the first to know the latest news and updates
