Last week, clinical epidemiologist Ziyad Al-Aly and his team at Washington University and the St. Louis Veterans Affairs (VA) health system published a new paper demonstrating that Covid-19 has a higher long-term health burden than the flu. The new study, published in The Lancet, follows past research by Al-Aly and colleagues that has explored different facets of Long Covid using health records from the VA system. Betsy Ladyzhets from The Sick Times talked to Al-Aly about his new paper and other areas of Long Covid research.
In the new paper, the research team analyzed health records from 81,000 people hospitalized for Covid-19 between March 2020 and June 2022, then compared their health issues to those of 11,000 people hospitalized for the flu between 2015 and 2019. Unlike other studies that have compared Covid-19 and the flu, Al-Aly and his colleagues focused on long-term outcomes: they tracked patients over 18 months following their hospitalizations.
While the flu does contribute to health issues in the months and years after infection, Covid-19 does more damage to almost every organ system, the researchers found. Al-Aly describes this distinction between the two viruses: flu disrupts the respiratory system, while Covid-19 is multisystemic, disrupting many parts of the body. It’s not appropriate to equate “Long Flu” and “Long Covid” — as other outlets have done in covering this study — he said, as the research clearly finds Covid-19 is “more formidable.”
Read on for our conversation about the new paper, the dangers of reinfection, future research questions, why clinical trials are a top priority, and more. This interview has been lightly edited and condensed for clarity.
Betsy Ladyzhets: What would you say are the main new findings from this study?
Ziyad Al-Aly: We wanted to compare Covid and the flu. There’s a lot of studies that have been done on Covid versus flu, but they don’t [look at long-term] outcomes… What we wanted to do was Long Covid versus flu. And what we found, which was not surprising to us, was that Covid produces a post-acute burden, in the post-acute phase of disease, and that is generally termed Long Covid.
But what we also found, is that if you track people with the flu over a year and a half past the initial infection, you’ll find they also are coming back to the clinic, to the hospital, with a lot of problems — that is also a significant burden of disease. Comparatively speaking, still, Covid is a mightier foe. Covid is a much more formidable virus and produces much more or higher health loss, in both the acute phase and the post-acute phase.
I think one of the “aha moments” here is that the burden of health loss in the post-acute phase eclipses — or is much higher — than the burden of health loss that you get in the acute phase. So we think one of the take-home messages from this paper is that conceptualizing Covid and flu as acute illnesses really will miss the larger portion of the health burden that comes from this disease, which is happening in the post-acute phase.
BL: I noticed in the conclusion, that there’s a line where you describe flu is more of a respiratory virus, and Covid is more of a multisystemic virus. I’m curious if you could talk more about why you make that characterization, and how that is demonstrated by the data that you have.
ZA: This is really important. We followed these patients over the span of 18 months. And we looked at about 94 different outcomes: outcomes in the heart, outcomes in the pulmonary or respiratory system, outcomes in the GI tract, outcomes in the nervous system, etc. We looked at all organ systems. And we had these two groups [of people who had Covid and who had the flu], we were comparing and contrasting.
Comparatively speaking, we found that Covid was hitting the heart harder, hitting the brain harder, hitting the kidneys harder, and hitting the metabolic system harder. But the flu was hitting the pulmonary system harder. There was more risk in the pulmonary system from the flu than from Covid, and there was more risk — relatively speaking — in all other organ systems with Covid than with the flu.
So one of the key conclusions is that true to how we think about it and define it, the flu is more of a respiratory virus. We’ve known flu for more than 100 years, and we generally think of it appropriately and rightly so as a predominantly respiratory virus. But this [paper] also dovetails very nicely with all the prior thinking and research that has been done by us and others, that Covid is a multisystemic virus…
The virus that causes Covid is SARS-CoV-2. And the “R,” in SARS, which stands for respiratory — that’s a misnomer. It’s simply not accurate. The “R” in SARS should be changed to a [letter for] “multisystem” or something. When you compare them, the flu is a much more formidable respiratory pathogen, but Covid is a much more formidable multisystemic pathogen.
BL: I wanted to ask also about the process for this [research], because I know this is now one in quite a few papers that you and your colleagues have done with the VA health system data. So what are some of the things that you’ve learned about how to best work with this data source, and also in choosing all the outcomes that you looked at?
ZA: First of all, we spend a lot of time thinking about, what are the most relevant questions to ask? What are knowledge gaps that are really important to fill? … And then, we focus a lot on trying to make sure that we design our studies in a way that are scientifically rigorous and defensible with the peer review process, and yield results that are reproducible.
We tend to come out, generally, first with these findings, but then they replicate. One of the major tests of validity in science is, “Do these findings replicate?” We came out with a paper two years ago, [finding] that Covid was associated with increased risk of diabetes — now that has been replicated more than 20 times. Now we’re finding the reinfection study is being replicated, the Canadians released some data the other day saying that reinfection is bad for you, it increases the risk of Long Covid.
BL: Yeah, I saw that, I was going to ask you about it!
ZA: [When we published our study], nobody believed us! People were like, you’re the VA, [your dataset is] old white folks, it doesn’t apply to me. I kept arguing that you cannot and you should not reduce the vast heterogeneity of about 11 million people at the VA to a single number. Yes, the average age at the VA is 60. But there are literally hundreds of thousands of people in their twenties and thirties and forties. And yes, 90% of the VA is male, but we have more than 600,000 females in our cohort. And we can literally fill 10 Taylor Swift stadiums, with our female cohort alone…
What we found throughout the history of our Covid research, and even prior to starting to do Covid and Long Covid research — generally our findings, using VA data, will be replicated with other datasets.
BL: With that Canadian study, in particular, I’m curious, what did you see as takeaways from that? And what else — or what other studies do you want to see done on this reinfection question, both with the health records that we have, and also other kinds of studies that would dig into the immunology of this stuff more?
ZA: I think the most important finding from the Canadian study is that reinfection really matters. A lot of people want us to trivialize reinfection, and to think of it as a “nothing-burger” that is inconsequential. But that’s not true, reinfection is indeed consequential and contributes additional risk of Long Covid. It would be good to have even more replication studies because people will still not believe it until they have ten studies saying the same thing.
And then, I really would like us to not even wait for those [additional] studies to move forward. How can we better protect the public? I think the conversation about protecting the larger public has stalled because a lot of people think that reinfection is not consequential. There’s really no public health messaging about trying to reduce one’s risk… Most public health leaders, you don’t see them talking about this.
This needs to change, and I don’t want us to wait until we have ten studies on reinfection to act. I asked the question the other day, “How many Long Covid cases could we have prevented if we believed the study that we published a year and a half ago? If we took that seriously?” We don’t need a million studies to all tell us the same thing, and then act on it in, like, 2030 — and then we’re all sick in 2030! Because we didn’t pay attention to reinfection. Because we wanted, like, “one more study, one more study,” and then it’s 2030, and 90% of the population is sick. And [public officials] are like, “We should take precautions now.”
From my perspective, the evidence is very clear. Everything I see in the literature, including that Canadian study, suggests that this is happening. And the world needs to move in a direction where we are acting on this information, and not stuck, waiting for Godot.
One of the take-home messages from this paper is that conceptualizing Covid and flu as acute illnesses really will miss the larger portion of the health burden that comes from this disease, which is happening in the post-acute phase.
ziyad al-aly
BL: Another thing that I’ve been thinking about, with this challenge of evidence generation or these requests for, “you need to keep proving that this is a problem,” is the challenge with testing, and with even tracking who has Covid these days. I saw, on Twitter, a critique of that Canadian study that said, “many of the people they included didn’t have a positive Covid test.” And I was thinking about — obviously, one thing I’ve learned from patient researchers, like PLRC, is that many people [with Long Covid] are not able to access PCR tests. Maybe they couldn’t in early 2020, maybe they can’t now.
And so there’s this challenge of wanting to continue to document what’s happening. Or at least for people who continue to get sick, wanting to continue to be able to track them in the medical system. Even to be included in the studies that you guys do, you need some kind of initial evidence of Covid. So how do you think about looking for that, or continuing to track that, as testing becomes less available? Are there new ways we can thinking about for, not even replication [of studies], but just like, looking at the impact of third, and fourth, and fifth infections?
ZA: This is a very important question. First of all, for clinical care and for diagnosis of Long Covid, and disability benefits and other things, diagnosis of Covid should not be required. If you require that, you really will disenfranchise a whole bunch of people who are most vulnerable. There’s a reason why they didn’t get tested: they didn’t have access to testing, they didn’t have the money for it, didn’t have the transportation to go and get [tests] from CVS… That’s a health equity problem, a disparity problem.
It should be very clear for everyone that, to establish a diagnosis of Long Covid, you only need a plausible history of infection, you do not need to require testing. And if you do, know that you are actively participating in inequities, you’re actively disenfranchising a bunch of people. The people who don’t have access to testing are really the most vulnerable among us.
Now, from the research perspective, the methodology perspective, I think measuring exposure is going to be a challenge, because as you pointed out, testing behavior has dramatically changed over time. And now for the most part, a lot of people have largely abandoned testing… A lot of people think of it as a cold. They’re not willing to test, they don’t want to take an antiviral, they don’t want to do anything about it.
So we need to evolve our approach to really capture exposure in a different way. There is some early conversation: this may not be feasible at the individual level, but using population-level data, we can computate exposure [to Covid]. In the same way as we computate exposure to air pollution. We don’t put a monitor on your person to monitor your exposure to air pollution. We say, you live in this ZIP code, and this ZIP code, on average, is exposed to this much air pollution. We know from satellite data, from air monitoring stations, that ZIP codes are exposed to a certain amount of air pollution per year. So we can assign you, [as a resident of the ZIP code], exposure to air pollution. I’ve done a lot of pollution research, and can say, this is an established science.
I think we need to evolve our concept for infectious diseases in a similar way. That would mean we need a sort of exposure index for populations, where we’d be able to assign exposure at a population level and try to associate it with the burden of disease, and then see if additional waves of reinfection clearly result in additional burden of disease — and continue to tackle it over time.
Bottom line, it’s become more challenging to assess and measure exposure scientifically. Relying on individual testing is going to be fraught with biases, precisely because the testing behavior of Americans has dramatically changed over time… We need to borrow lessons learned from pollution research and other areas of research, to try to develop exposure metrics or indices to assess exposure at the population level.
BL: Have you looked at the CDC’s new wastewater dashboard? Because I think this also relates to what you’re talking about.
ZA: Yes, one of the inputs for that exposure index I’m talking about, one of the inputs is wastewater surveillance systems. Other inputs are still individual data, but aggregated at the population level and corrected for testing behavior. It’s really an index of multiple variables.
BL: I also wanted to ask, I saw you spoke at a meeting in Australia recently about the urgent need for more research, and particularly for clinical trials. I was curious how that was received, it seemed like a pretty high-profile event. And also, your takeaways from talking about this — what do you see right now as the most urgent priorities for clinical trials and other continued research?
ZA: Governments have to commit to funding Long Covid for the long haul… There needs to be a long-term commitment. Covid produces Long Covid, other infections produce infection-associated chronic illnesses. There needs to be research, and we’re not going to be done with it today or tomorrow or the day after. That applies to the U.S. government, and I pretty much said exactly the same thing to everybody I met with in Australia.
I feel this is an “all hands on deck” situation. RECOVER is not going to solve all the problems in the world. It’s not. The U.K. government has to do its part, the Australian government has to do its part, the U.S. government has to do its part, and we have to commit long-term funding for Long Covid and infection-associated illnesses. That includes better epidemiology, better understanding of the underlying mechanisms, and really, really importantly — because people are hurting and they want treatment yesterday — most important, to put on top of that list, clinical trials.
We definitely still need to understand the biology. But on top of that priority list, number one should be clinical trials to help us understand what works for these patients or what doesn’t work, so we can offer treatments to help… And we also need to advance the conversation in the direction that, there needs to be in place funding for infection-associated chronic illnesses… We need a line item in the budget, on a yearly basis, here [in the U.S.] and elsewhere. Each government has to pitch in.
I focus on clinical trials because I feel that really should be the number one item, because that’s gonna impact patient care and patient quality of life immediately… Governments need to recognize that this is a marathon, this is not a sprint. This isn’t like a, “Well, I’ll fund something for one or two years and that will be that.” This like HIV, Alzheimer’s disease, and heart disease.
There will always be heart disease, there will always be cancer, we need to continue to research it. There will always be viruses and infections, and then chronic disease from infection that we need to solve. We need funding in perpetuity to address the challenge of infection-associated chronic illnesses. Those are not going to cease to exist… There’s always going to be chronic disease as a result of infection, and our job is to really address them: to figure out why they happen, how they happen, how to best prevent them and how to best treat them.
All articles by The Sick Times are available for other outlets to republish free of charge. We request that you credit us and link back to our website.
Leave a Reply