The Halifax Examiner is providing all COVID-19 coverage for free.
The Nova Scotia Health Authority has mapped out “COVID clusters” by postal code, and that information is being mistakenly used such that people within those postal codes are being denied medical attention.
I’ve obtained the list of 10 postal codes, which are described by the first half of the code — the two letters and one number. Each code therefore describes a large geographic area that can include hundreds or even thousands of households.
I asked the Nova Scotia Health Authority about the COVID cluster list, and received this response:
We are aware of Nova Scotia Health Authority information on social media that identifies some Nova Scotia postal code areas or “community clusters” where public health has identified higher rates or spread of COVID-19.
NSHA is using enhanced precautions during the COVID-19 pandemic, and we know it is possible that people may have and be transmitting COVID-19 without having symptoms or without knowing they have been exposed themselves. The document that was shared on social media is one part of a risk assessment tool, intended to help our staff identify if extra precautions are needed to keep the patient and staff safe. People in the noted areas are still receiving care.
Here is how we use this information:
- Staff at hospitals, clinics and primary care settings may identify people presenting for care, by their postal code, as being from one of the parts of the province where public health is currently seeing an increase of COVID-19.
- Given the possibility that a person in an area where transmission rates are high may have been exposed and not know it, staff can then use appropriate precautions to protect the patient and those around them during their visit.
- In the case of surgical patients, where it is safe and appropriate to do so, a decision may be made to delay surgery until the patient’s COVID status is known.
- If someone presents at an emergency department or is admitted for care and comes from an area where there is a community cluster, they would be tested for COVID-19.
- If they are being seen at another clinic, they will be referred to an assessment centre for testing.
This information is updated as soon as we receive new information from Public Health Services.
We appreciate the concern that may result from information like this being shared without context. Whether you’re in an area identified as a community cluster or not, all Nova Scotians should monitor for symptoms of COVID-19 and follow the precautions in place.
As presented, the COVID cluster list appears to be intended to increase screening for the disease among people in those geographic areas.
The NSHA’s website gives instructions to health care providers about how to screen patients for COVID-19, ending with:
5. If no to screening risk factors, assess for community / facility clusters:
• Does the patient live in a geographic location with known community cluster or work in a facility cluster? See the list of COVID-19 community and facility clusters [link is to a password protected page].
6. If no, proceed with booking appointment (phone, in person, etc.) based on urgency and importance.
7. If yes, proceed with booking the appointment, and use the suggested script (Community/Facility Cluster- Information for Primary Health Care Providers) that advises patients they will need to don a mask upon arriving at the practice.
I can understand how that made sense to whoever decided to create the list, but there are lots of problems with this approach — even if it works as intended.
But before we get to that, understand that the list is not being used as intended, at least by some health care providers. People who live in the so-called COVID cluster zones tell me that they have been denied health care services.
The following was posted on Facebook; it is just one example of several I’ve come across:
I called my Doctor’s office to see if the nurse was in…
Receptionist: Dr. ___ Office
Me: Is the nurse in this morning?
Receptionist: Yes she is what did you need to see her for?
Me: I just need to get my needle
Receptionist: What is your name?
Receptionist: Can you spell your last name for me?
Me: [spelled it]
Receptionist: And your date of birth?
Me: [gave my date of birth]
Receptionist: Ok I’m going to put you on a brief hold
Receptionist: Right now we’re not seeing any patients from that postal code area as it’s a high risk area for Covid.
Me: Excuse me? No it’s not.
Receptionist: Is this your postal code [reads postal code]?
Receptionist: Based on your postal code, you’re listed in a high risk Covid area, and we’re not taken any patient from this area at the time.
Clearly, the intent of creating a COVID cluster list was not to limit health care in those zones, but here it is being used exactly in that way. So I asked the province’s chief medical officer of health, Dr. Robert Strang, about it at today’s COVID-19 briefing. Here’s our exchange:
Bousquet: The Nova Scotia Health Authority has been sending to doctors, or making available to doctors and other health care professionals, geographic information by postal code about COVID clusters… The purpose of the cluster information is to increase screening and better protect health care workers — that’s obvious. But residents are telling me in those zones that they’re being denied doctor visits. The NSHA itself says surgeries will be delayed until further testing goes on for residents in those districts. People are worried they’re being redlined because, frankly, the Prestons [are on the list]; they’re saying they’re being redlined out of health care. Can you speak to them about that? And also, health professionals and staff who might be misinterpreting how to be using that information.
Strang: That’s a complex question. This was one of the challenges and you’re one of the people, Tim, who keep pressing for more local level information and one of the things you just described is one of the things we’ve always been concerned about — the information is released inappropriately and is used without the right context.
So first of all, that information was provided internally within the Nova Scotia Health Authority. I’m aware that it was actually released publicly. The fact that you have your hands on it makes it concerning. It wasn’t intended for public release. The Health Authority is looking into that. The purpose of releasing that data was to allow frontline staff in our local hospitals and especially in the Dartmouth General Hospital to know if people come in from certain communities, there’s a higher possibility they may have COVID and so they can use appropriate personal protective equipment in those patient encounters. That was the purpose of that information. It is not at all intended to label or the term you used redline anyone from any communities, and it’s certainly not to be used as a way to deny people health care. In fact, it’s used for the opposite. People coming from those communities may have a greater need for health care because of potential exposure to COVID and that information was to be used to make sure that they could be provided what they need in the safest manner possible.
A couple of things here. First, for the record, beyond asking for more information about the jail, homeless shelters, and long-term care facilities, I have not asked for more local information generally. Other reporters have, however, and Strang declined to give them that information.
But it was Strang himself who was the first to release local level information about COVID clusters, at the daily COVID briefing on April 7. As El Jones reported:
In the province’s previous COVID-19 daily press conferences, we have heard about the importance of preserving privacy, of not stigmatizing communities, and of the public not lashing out at businesses or other places where the virus has been identified.
All of that seemed to go out of the window on Tuesday, when both Dr. Robert Strang and Premier Stephen McNeil took aim at Black communities that have been identified as having infections.
When it was largely white people with money to travel who were returning to the province, we heard about confidentiality. When it was a St. Patrick’s Day party, we were carefully told that the information was only being made public to help identify cases and that we should not lay blame.
But now that Black communities also have virus spread, the tone has changed.
In the beginning of the briefing, Dr. Strang spoke about communities that are currently dealing with infection. While he mentioned Elmsdale and Enfield, the majority of his comments were reserved for the Preston area:
We have established temporary assessment centres in communities where there’s concerns about increased activity. We opened one up in Elmsdale and Enfield on Sunday. Today we are opening up a temporary assessment centre in North Preston. There are concerns about increased disease activity as well as the patterns of how the disease is being spread and underlying issues in these communities of North Preston, East Preston, and Cherry Brook that we need to understand more, and going in with a temporary assessment centre is part of working with the community to help understand those issues. We’re working very closely with community leadership and community leaders in these communities. They share the concerns and have their own concerns and we’re working very collaboratively to understand this issue.
These communities are very concerning to me because they have the potential if we don’t get out front and support these communities, we already know that we have cases in the broader HRM area that are linked back to these communities. We have cases among health care workers that have gone in through their work unknowingly and transmitting disease in some of our health care facilities. That is very concerning about how this could impact the broader community.
As I said, leaders in these communities are equally concerned and are working closely with us. Unfortunately, there are groups in these communities that are — because of their willful not following the requirements to minimize social gatherings, to stop unnecessary social interactions are a key piece of what is driving the disease within these communities. And that makes me very concerned. But it’s not just these subgroups in these communities.
Until only days ago, all the cases in the province were linked to travel. Yet in Strang’s comments, it has become Black people in their own communities who are responsible for spreading disease. Black people are spoken about as the ones who contaminate everyone else. The implication here is that if Black people are not contained, we go out and get white people sick.
As I’ll explain below, I’m not publishing the postal zone list, but I’ve made the exception of mentioning the Prestons because Strang — not I — already let that cat out of the bag.
Second, I quoted El Jones’ report at length because Strang said outright that:
We have cases among health care workers [who live in the Prestons] that have gone in through their work unknowingly and transmitting disease in some of our health care facilities.
So my follow-up question today was this, with his response:
Bousquet: Perhaps related, you spoke earlier in the briefing about Northwood staff being spread across a wide geographic area and coming into the facility at Northwood possibly, if I understood you correctly, possibly bringing COVID into the facility because they come from so many diverse communities. I wonder two things: if that was sort of a backhanded reference to people from these clusters working at Northwood, but also, why isn’t the reverse a concern? That people who are working at Northwood may be carrying COVID back to their communities?
Strang: First of all, it’s not intended as a reflection on any community. My comment was simply a recognition that when you have a large facility with hundreds of staff coming from a broad range of communities throughout the HRM area and beyond, as opposed to a smaller facility in a more rural part of Nova Scotia, there’s a greater likelihood that that broader mix of communities of COVID being brought into the long-term care facility. That’s the only intention of my remarks, so please don’t take it as a reflection negatively on any community.
The reality is we know in long-term care facilities and our acute care facilities that there are strict infection control guidelines and protective equipment guidelines that are based on the type of care that is delivered and the type of facility, that if adhered to greatly minimizes the chance of a health care worker actually being exposed and infected. So the real risk in long-term care or even in hospitals is people coming into the facility who now that we know about asymptomatic spread, and that’s one of the reasons why, we’ve been able to move on masking all frontline health care providers. There’s a much, much lower risk of introduction into a community from a health care worker from a facility.
Does this make sense? As of today, there are 73 staff members at Northwood who have tested positive for COVID-19. Are we to believe that each and every one of those 73 caught the disease in their home communities and then just happened to be working in the same building, and none of them were infected at the building? If so, shouldn’t there be similar large number of workers at other large workplaces — hospitals, the police department, the military base, whatever — who test positive for COVID-19?
Regardless, the point of my first question was not to challenge Strang, but rather to lob him a softball. I said: hey, people in the communities are misunderstanding the postal code thing, and some health care workers are also misunderstanding it, why don’t you straighten them out? Instead of doing so, he in effect faulted me for being a reporter.
After I spoke to Strang, I called Françoise Baylis, a bioethicist who studies public policy, to see what she thought.
“He doesn’t just get to wash his hands of that,” Baylis told me. “If you create a tool that is being misused and or misunderstood, it doesn’t matter if you intended otherwise. If you learn that it is being used in the wrong way, you absolutely have a moral obligation to address it.”
“My starting point,” Baylis said, “is do you have accurate information about what communities are affected, and is it targeted or is it just an unfortunate — unexpectable, don’t get me wrong — fallout? In other words, they didn’t pay attention to what they were doing. So, I’m trying to figure out the difference between ‘it’s wrong and it needs to be corrected and it was unintentional or stupid’ versus ‘this is actually overtly, as opposed to covertly racist.'”
But Baylis said that even if the COVID cluster information were being used as intended, she still has problems with it.
“What’s the criteria for being a cluster?” she asked. We discussed the various possibilities: reaching, say, 100 people testing positive for COVID-19 in a postal zone makes it a cluster. Or maybe it’s a certain percentage of positive test results from people who live in the zone. Or is it simply arbitrary?
“What puts one postal zone on the list while another isn’t on the list?” she asked. “And how does a zone get taken off the list?”
I asked the Nova Scotia Health Authority how it defines a “COVID cluster,” and how the list is updated. Thursday evening, long past normal working hours, spokesperson Carla Adams responded via email, “Our public health team has deferred to the office of the Chief Medical Officer of Health to respond. I’ve shared the request with them.”
When and if Strang responds, I’ll update this story. If he doesn’t respond, I’ll ask again at the next COVID briefing in which I am given the opportunity to ask a question.
But Baylis’s concerns didn’t stop at the definitional. She wondered what the purpose of the COVID cluster list is.
“Is it because there’s a lack of resources?” she asked. “There are 10 zones on the list. If we get more resources does the list grow to 11 or 12 zones, or do one or two of the zones on the list get taken off? How do you decide who gets on the list, and how do you decide who gets off?”
As we talked, I was reminded of a common boneheaded media mistake that non-thinking reporters make all the time: confusing relative and absolute risk. Everyone has seen the headlines along the lines of “Drinking coffee increases your chance of male pattern baldness by 20%” or whatever.
To explain the problem, let me make up a completely bogus scenario. Let’s say that on average four people in Canada die every year from a horrible tonsil-eating fungus called a-ieatteeth. And using computer modelling and testing on rats, a researcher discovers that if everyone in Canada had three cups of green tea a day, five people in Canada would die from a-ieatteeth. So the researcher sends out a press release and the next day the headline in The We Reprint Press Releases Daily screams “There’s a 25% increase in the chance of dying from a-ieatteeth if you drink three cups of green tea a day!” Well, that headline would be correct. But so would a different headline: “99.99999% of Canadians won’t die from a-ieatteeth even if they drink green tea.” But which headline sells the most papers, right? Don’t pay attention to such headlines. They confuse relative and absolute risk.
It’s maybe not quite on that scale of ridiculousness, but something like that seems to be happening with the NSHA’s COVID cluster list. There are only 348 known active cases of COVID across the entire province, and 256 of those are connected to Northwood: 198 residents and 58 staff. So there are only 92 non-Northwood-related known active cases in communities across Nova Scotia. Therefore, the postal zones, containing thousands of residents, are simply too big to provide anything in the way of meaningful information about risk to health care workers. If there are 20 people out of 6,000 in Postal Code A that have COVID versus 10 people out of 6,000 in Postal Code B, should there really be different protocols for dealing with the two postal codes? If you live in Postal Code A, the relative risk is that you are twice as likely to have COVID than someone in Postal Code B. But as the health care provider experiences it, the absolute risk is that the patient from Postal Code A has a 0.33% chance of having COVID, while the patient from Postal Code B has a 0.17% chance of having COVID. Or seen another way: both patients have more than a 99.5% chance of not having COVID. Surely, if it’s worth taking extra precautionary measures for one, it’s worth taking them for the other.
Whatever the calculation that defines a postal code COVID cluster, it feels like a formula that’s looking for a reason to exist.
Anyway, back to Baylis. She told me that at the heart of her concern is an issue she’s been studying and writing about throughout the pandemic: “we don’t deliver health care, or access to it, by geography.”
“One hundred people in that zone have COVID, so someone else living there gets a certain level of screening and care, but because only 98 people in my zone have COVID, I don’t get it.”
Baylis has looked at the COVID response cross-jurisdictionally. “My aunt might get on a ventilator in one province, but not in another province. More to the point, she might get one, and I might not, just because I live in Nova Scotia.
“We think we have national heath care with one set of standards, but we don’t.”
“Ask for the triage protocol,” she told me — that is, should hospitals be overwhelmed with COVID patients, how will it be decided who will live and who will die? “They won’t give it to you,” she said. “There’s a lack of transparency.”
I thought long about publishing the COVID cluster list, and weighed the positives and negatives. On the one hand: transparency, and maybe it would help explain the issues raised in this article. But on the other hand, I worried that some readers might look at the list and conclude that because they don’t live in one of the supposed “clusters,” they don’t have anything to worry about, and that would lead to them lowering their guard and actually becoming more likely to spread the disease.
I asked Baylis whether she thought it was a good idea to publish the list. “I don’t think you should,” she said. “I think everyone should think about the implications of whether they’re considered to be in a cluster.”
The Halifax Examiner is an advertising-free, subscriber-supported news site. Your subscription makes this work possible; please subscribe.
Some people have asked that we additionally allow for one-time donations from readers, so we’ve created that opportunity, via the PayPal button below. We also accept e-transfers, cheques, and donations with your credit card; please contact iris “at” halifaxexaminer “dot” ca for details.