3 radiologists warned Central Health about mammography accreditation concerns in 2019
A former Central Health radiologist says she warned the provincial government and the regional health authority about problems related to mammography in 2019, but felt her concerns were largely ignored.
Dr. Jane Rendell, who worked in Gander for six years until she was sidelined in September of 2019 due to ongoing disputes, was one of three radiologists to send a letter to the Department of Health and Community Services on Oct. 10, 2019.
They warned that the breast screening program at the Central Newfoundland Regional Hospital in Grand Falls-Windsor was not accredited by the Canadian Association of Radiologists (CAR), and therefore was not undergoing the stringent checks that other facilities were obligated to go through.
They also outlined 15 other concerns with Central Health's radiology department, including the number of out-of-province radiologists being brought in for as little as one week at a time to fill gaps.
"They just ignored us," Rendell said of the regional health authority and provincial health department. "Basically, I think we were considered troublemakers because we had problems and we would raise them. They didn't like doctors doing that. They wanted us to just shut up and do what we were supposed to do."
In the last two weeks, Central Health has discovered about 3,000 mammograms were read on substandard screens and must be re-examined. Central Health says all the images were taken in Grand Falls-Windsor and Gander.
After that discovery, the province's three other health authorities reviewed their own procedures and found about 14,000 mammograms from other regions needed to be re-examined because they were read on three megapixel screens instead of five megapixel screens.
None of the health authorities have explained how this happened, or where any of the mammograms were actually analyzed after being taken. Rendell isn't certain if it has to do with the problems outlined in the letter she signed alongside Dr. Shelley Dalton and Dr. Paula Kennedy, but assumes there is some overlap.
Central Health did not respond to a request for comment before publishing time.
Health Minister Tom Osborne, who took over the portfolio in July 2022, said he was not aware of the letter.
"That's new information for me today," he told CBC News on Thursday. "I can certainly speak to the CEO of Central Health regarding this. My understanding from Central Health was this occurrence was brought to the CEO very recently."
How does accreditation work?
Eight facilities in Newfoundland and Labrador have accredited programs. Those mammography units are inspected annually to ensure all aspects are up to the standard set by CAR, which administers the accreditation program.
But the process is a little more complicated than that. If a facility is listed as being accredited on the CAR website, it does not necessarily mean all the units and radiologists in the building are accredited.
This was raised as an issue in Newfoundland and Labrador in 2005, when the Canadian association called on the province to make accreditation mandatory for all mammography units, and even asked women to avoid places that were not fully accredited.
Rendell says the breast screening program in Gander — in which women who have never had cancer receive regular annual checkups — is accredited by the CAR and undergoes stringent oversight. But she said diagnostic mammography, which examines women who have a history of cancer or are believed to have cancer, is not accredited and therefore units and personnel are not inspected by the Canadian association.
In their letter in 2019, Rendell, Dalton and Kennedy called for all programs to be regulated the same.
"The only way to ensure that the standard of care in mammography is being provided to all women of Central Health is through the audits and peer review that occur routinely in the breast screening program," they wrote.
'They're standards for a reason'
There has been some debate about the significance of the wide-ranging review of mammograms.
Dr. Angela Pickles, the head of medical imaging with Eastern Health, said there is little difference between a three megapixel screen and a five megapixel screen. She said the two are "so close that the human eye struggles to differentiate any small occurrences at this level."
Dr. David Jacobs, the head of the Ontario Association of Radiologists said a 5 megapixel screen is the best technology available, but might not make that much of a difference in most cases. While a small abnormality might be crystal clear on a five megapixel screen, it would still be visible on a three.
Rendell, however, said that small difference can make a huge impact.
"When you do mammography, you're looking for one thing. It's kind of like Where's Waldo. You are looking for breast cancer only. That's the only serious disease you're going to see in the breast. If you miss it, it's a big deal," she said.
"If you are missing microcalcifications and things because your monitor is not good enough, I'm sorry, but that's just not up to the proper standard, and they're standards for a reason."
At home testing?
Central Health's refusal to say where the images were analyzed has led to questions about mammograms being read at home, especially during the pandemic.
Rendell and Jacobs said it is fairly common for radiologists to read them at home — but it's also expensive. Jacobs said a proper set up requires four monitors, which comes in at over $40,000. Jacobs said three megapixel screens are much cheaper.
Rendell believes that's a big factor in the current predicament.
"I think what it was is these people had threes in their home and decided they were going to read mammo on them anyway," she said. "It's an informed theory. I've worked at radiology for a long time."
Whatever the case, Rendell believes more discrepancies will be found, even if it's just due to human error.
"I still think you have to have the best equipment because you're going to have human error anyways, even with the best equipment. So if you start with subpar equipment, your human error is going to be higher. And it's not fair to the patient. The patient is the person at the bottom of this."