It’s very, very unlikely that MRI/CT scanning or nuclear medicine procedures are performed in the maternity unit.
At the cost of how many lives?
Coincidences happen, and it’s actually quite plausible that this is a coincidence. Which does not mean, of course, that this is not worth checking out. But even given that, heck, even if they had found a definitive cause, actions have consequences. Shutting down a hospital can have very severe consequences.
So this might be a “space-filler” article that sounds weird but isn’t actually anything big or concerning? Apart from perhaps questioning the thoroughness of the hospital’s investigation.
It might be real, and the fact that it might be real probably is enough to justify an article on it.
For more perspective, here’s a New Yorker article on cancer clusters by surgeon and public health researcher Atul Gawande.
He states that in one year, between three and four thousand purported cancer clusters were reported to Massachusetts authorities.
Gawande does note that while neighborhood cancer clusters overwhelmingly don’t yield any convincing explanations, occupational clusters have a better chance of having a common cause. The question surrounding the current example is whether a total of five cases of multiple different types of benign brain tumors occurring over an unknown period of time in one occupational group forms a true cluster with a discernible cause.
The article clearly said the tumors are benign, not malignant. That doesn’t rule radiation out as a cause, but it does rule in a bunch of non-environmental factors.
Any sudden cluster of disease needs to be investigated fully, but let’s not go closing the hospital until this has been investigated more thoroughly. Maybe start by examining former employers, particularly those who worked there a long time.
Who, by the way, practiced surgery at the hospital in question.
It is harder to calculate the probability of this happening at random than you think. How many hospitals are there in the world? How many floors are there on average in each hospital? How many nurses are there on average on each of those floors? How many different diseases are there that they may have that we can categorize? For each of those diseases, what proportion of all nurses have that disease? Unless we know how many there are of each of those numbers, we can’t tell how rare the fact that a certain number of nurses on a certain floor somewhere all have some disease really is.
For instance, suppose you write down the day and month of birth of each of your 22 best friends and yourself. You notice that there are two people in that list with the same birthday (counting just the day and month, not the year). You ask someone else to do the same thing for a group of 23 which doesn’t overlap with your 23 people.. They look at the 23 birthdays. They discover that in their group, it’s also true that two of them have the same birthday. You think, “How could this possibly happen? Two groups of 23 with two of the same birthdays. Wouldn’t this be incredibly rare?”. No, in fact in any group of 23 people, the chances are just over one half that two people share a birthday. So what are all the numbers that I listed above?
It was you that said that. For some silly reason I did not scroll up to check and misremembered Beck. Oops on me.
I apologize for the misattribution. You deserve full credit for this very salient point. So it’s now been made 4 times. 
Also, look for cases in other classes of employees on the floor- and the duration of effected nurses’ employment on the unit.
Someone mentioned the idea that the nurses could be related.
If I had to investigate this, that’s the first thing I’d look at.
In the 80s, I was part of the Candy Striper program, back when there was such a thing, and I remember being surprised at how many people were related to each other. There were mother-daughter pairs or triplets, with two daughters, in nursing, one NP mom with a doctor-daughter, and a lawyer-daughter, and the lawyer was the hospital’s legal consultant.
There were young doctors just out of med school doing internship where their father’s or sometimes mother’s had privileges, and were planning on joining them in practice, in things like orthopod, or OB or pediatric surgeons.
I’d get confused when two people I thought were unrelated (different last names, not much resemblance), and same gender (1983) would be talking like a married couple about their weekend, and then mention they were cousins, and another cousin was on CCU.
It was like the hospital bred it’s own workers.
On 3 separate occasions, I ran into people I thought I was tight with, and they kinda blew me off, so I said something the next time I saw her, and then learned I’d run into her twin sister the time before, who did not, in fact, know me from anyone.
Given how recessive genes can express themselves, so that family traits are not necessarily seen only in close relatives like parent-child, or siblings, having lots of extended family in one place could produce similar health problems like you see in mysterious “sick building” clusters.
And if this hospital is like the one where I worked, the nurses may all be related, but distantly enough not to have it at the forefront of their minds.
It’s worth noting that different cancers have different odds. Two of the nurses had meningiomas (the most common type of brain tumor, and benign), which have odds of occurrence of 1 in 100 in the general population, but closer to 1 in 50 middle-aged women.
Those are the only two types I saw specifically reported.
As many have noted, events that look like clusters are statistically inevitable, and absent finding an environmental cause, or repeatability, you can’t make much (or anything) of them.
Menstrual synchrony was thought to be a real thing for decades just because the first major study lucked into a women’s dorm that just coincidentally had a high number of women on the same cycle.
And I was BUMMED! I was looking forward to the follow-up on which woman the other women tended to sync to, and why?
I think this is overstating meningioma incidence. This paper found a meningioma incidence of about 17 cases per 100,000 women aged 50-54, or about 1 case per 5800 women. I’ve seen other figures suggesting overall incidence may be as high as 97 cases per 100,000 (or close to 1 in 1000), but that still wouldn’t get anywhere near 1 in 50 in the most susceptible populations (excluding those with genetic predisposition).
Yeah, sorry, I misread something. Thank you for catching that.
Or there may be some other common factor to at least a few - this group all shares some out of work thing socially that increases risk, for example.
Point that rare things are almost always going to happen given a big enough pool is well made. And correlation is not causation cannot be dismissed.
You’d also want to check to see if they live in the same neighborhood.
“The law of very large numbers.”
Sometimes when a curious thing happens, and someone says “What are the odds?” I say (or used to-- kinda given up) “100%.”
Yeah, if the family thing didn’t pan out, but considering the enormous number of things that have been found just in the last 50 years or so to have a genetic cause, I’m still going with that.
The next thing that occurs to me is not that there’s an A → B but B → A
In other words, people who develop benign brain tumors frequently go into nursing as a profession. Why? I have no idea, but 5 is too small a group for any kind of extrapolation. You need to gather a better sample of people with this tumor, and see what they ALL have in common.
Maybe they all go into nursing because the tumors affect their hearing in such a way that background noise is unpleasant, so they look for jobs in quiet settings. Maybe there are lots of librarians, too.
Or maybe you discover that they don’t gravitate toward any particular profession, but when they all work in the same large company, they tend to gravitate to the same department or office.
I’m spitballing-- just saying the tumors could cause the profession, or some other common thing could cause both the tumor and the profession (like being related). Immediately assuming that one thing caused all the tumors is too little, too soon.
Same here, it’s a textbook unsolved medical mystery. I’ve visited some parts of the facility, including some deep inside with special gowning procedures, and I don’t doubt that some ultraexotic molecules exist there.