How realistic/reliable is this long covid game?

February 21, 2024 9:03 AM

A member of our game-playing and activity group has recently pitched this long covid simulator/game. How reliable and realistic are the assumptions made by the game's design and are the data it's based on current and correct?
posted by anonymous to Science & Nature (16 answers total)

One thing that jumps to front of mind is that the game mechanisms don't seem to have any stand-in for how vaccination status affects risk of developing long covid specifically, which there's been a good bit of research on. It's certainly a very overly simplistic representation of risk, which the designer admits was intentional for effect: Finally, while you could create a comprehensive simulation for individual risk of contracting Long COVID (I have thought of doing that many times over the past four years), the potential for ‘lucky’ players to come away with a diminished impression of the risk of Long COVID is, in my mind, too high.
posted by superfluousm at 9:15 AM on February 21

They cite this page; I think the way the describe it (a cumulative, 15% chance of Long COVID each time you have COVID) is at least slightly overstating the actual study. The structure of the game seems reasonable, though.

However, they apparently didn't realize this was a cumulative chance originally, and had to update the game.

Furthermore, I don't know what sort of group you're in, but this is less a "game" than it is a pedagogical tool; I don't think it's meant to be intrinsically fun or rewarding - is your group having a lot of concerns about COVID/COVID-related behaviour that need to be addressed?
posted by sagc at 9:16 AM on February 21 [2 favorites]

If it's from a Dec 2023 paper that's based on 2023 survey data, then, that feels pretty current.

That paper uses a common definition for long COVID, which is "experiencing symptoms 3+ months after infection." There is a really big spectrum of things that fit that definition - from "still coughing" to "long-term disability" - so I think that definition tends to be a very broad catch-all that likely overestimates the incidence of long COVID, or at least conflates some manifestations that are exponentially different in severity.
posted by entropone at 9:18 AM on February 21 [9 favorites]

The game asserts that every case of long covid is permanent.

In the study,~40% of the people with extended covid symptoms did eventually recover.
posted by BungaDunga at 9:23 AM on February 21 [14 favorites]

How reliable and realistic are the assumptions made by the game's design

I think this is not realistic: "it would be absurd for the villagers not to wear the ward against the curse"

That is only true if the cost of wearing a respirator is zero. It isn't. There is a small monetary cost; I don't think many people would be limited by that. However, there is a higher social cost. Interpersonal communication becomes significantly more difficult, particularly with people with hearing loss. Comfort is difficult. Partaking in common social activities requiring eating or drinking becomes difficult. Activity requiring physical exertion becomes more difficult. Public art performances become more difficult. These are all costs that add up, making the "ward" not truly zero cost.

I don't think all people who decide not to wear a respirator 100% of the time in the presence of other people are doing so because they believe it's useless. At least some people (including me) do so because they believe the cost of long COVID is less than the cost of social isolation.
posted by saeculorum at 9:44 AM on February 21 [14 favorites]

Also, even 40% of folks recovering from long COVID seems low to me. I'm not qualified to evaluate medical studies, but here's one I remembered reading in the news.

individuals reporting non-recovery decreased to 18.5% (16.2% to 21.1%) at 12 months, 19.2% (16.3% to 22.5%) at 18 months, and 17.2% (14.0% to 20.8%) at 24 months. A similar decrease occurred at 24 months in the severity of health impairment with 10.4% (8.0% to 13.5%) having mild, 3.9% (2.5% to 6.0%) having moderate, and 1.9% (1.0% to 3.5%) having severe impairment (1% missing EQ-VAS).

posted by saeculorum at 10:05 AM on February 21 [2 favorites]

The other thing that isn't realistic is that respirators are not foolproof and perfectly effective against contracting covid and certainly do not repel the curse for all those who choose to wear it. They are very effective but (unlike a magic ward) not particularly good at preventing infection from people you live with who either can't or don't wear them.
posted by BungaDunga at 10:35 AM on February 21 [1 favorite]

The game's page admits that it is wrongly mixing up the cumulative probability of ending up with long covid after three infections with the individual probability of getting long-covid on your third infection. That is a really fundamental error.

It is kind of like pretending the chance of getting a six at least once in ten rolls of a dice is actually the chance of getting a six on your tenth roll. They are just completely different things. (In the dice example, one is very likely (84%) and the other is very unlikely (17%).)

Instead of correcting the error he has justified it on the grounds that players wouldn't come away from the game with his intended message. It is hard to take him seriously.
posted by Klipspringer at 11:04 AM on February 21 [11 favorites]

The underlying Statistics Canada survey and stats seem robust to me, an epidemiologist, with the caveats above around how long COVID was defined.

This game does not accurately represent the percentages from the Statistics Canada report in its gameplay. It instead inflates the percentage of individuals with the curse in each round in order to advance a specific goal (behavior change among students) per the design rationale section. There are also assumptions about reinfection (that 100% of the population is infected yearly), uniform distribution of infections (that everyone has the same likelihood of infection/reinfection), and uniform distributions of long COVID (that everyone has the same likelihood of long COVID). These assumptions are also described in the design rationale section.

If you have the same goal on behavior change, this game may be interesting for your group. As a stats pedant who doesn't like games with too much dice-rolling, I would not enjoy playing the game.
posted by quadrilaterals at 11:19 AM on February 21 [7 favorites]

I mean, this isn’t about the mechanics of the game but referring to a chronic illness as a curse is kind of icky and not super realistic. Someone being cursed has connotations of being punished for trespassing or other bad behaviors. As someone coming up on year 4 of Long COVID I can tell you the worst thing I did was have a kid who got assigned to the wrong classroom, or maybe my wife brought it home from her job or maybe one of us got it because we dared to go grocery shopping.

I mean I get that it’s a game, but it’s a game whose goal is to change attitudes about COVID specifically, so ideally they’d have put a little more care in the implications of their framing.
posted by Gygesringtone at 2:03 PM on February 21 [10 favorites]

I think you've gotten good feedback on your question, but besides the exaggerated risk (that the creator acknowledges) I question the pedagogical value of a game like this - I could see it working with little kids, but it feels infantilizing for adults. If someone in your activity group wants people to wear masks they'll likely get more mileage by simply explaining where they're coming from and asking if the group is willing to accommodate them (whether it's by masking or holding events outside, etc.) than trying to manipulate them with this game. If someone in my social circle suggested this game, I'd suggest we skip it and just try to have a group conversation where people respectfully shared their needs and everyone brainstormed potential solutions.
posted by coffeecat at 3:51 PM on February 21 [3 favorites]

I could see it working with little kids

I'd have an issue with little kids, or even middle elementary kids, having to play this game. I was going to list all the reasons but I'll illustrate with a story. There are four kids - two pairs of siblings - in my former after school program who lost a father to pre-vaccine Covid. Out of those four kids, precisely none were perfect at mask-wearing even when they tried, and as of now, aren't wearing masks. Making any of those kids - or the ones who lost grandparents or aunties or went through teachers treating them like they were radioactive, which did happen here when schools reopened - play this game would be incredibly cruel.
posted by warriorqueen at 6:20 PM on February 21 [7 favorites]

As mentioned above the cost for "warding against the curse" is heavily underplayed. It's not zero, both in a social sense and likely an actual lifelong employment monetary sense. Ignoring the benefits of not wearing a respirator makes this feel like a "abstinence is the best way to prevent pregnancy, why would you do anything else?" analogy, and I suspect it will be just as effective as that message has been over the years.
posted by true at 7:26 PM on February 21 [3 favorites]

This is really interesting and we need more tools like this. So many people don't realise that the various health problems they are currently experiencing are, in fact, some flavour of long covid from a previous infection - difficulties with cognitive function, going from problematic blood sugars to frank diabetes, muscle and joint pain, etc. Nobody is making the connection and everybody says/believes, Oh I had covid and I have fully recovered. I agree that the 'curse' framing is a shame-based idea and an ableist one. But you have to start somewhere.

Are you on Twitter? This is a different approach but with a similar aim - an imagined simulation (it's not real; the poster is just imagining doing this in a classroom) to get people to understand that covid is airborne and has risks beyond the acute infection. It is, imo, an excellent idea.

Re: the numbers - I don't think the risks of LC are additive in quite that way (even though they are pulled directly from this study). From the results you can see it's different by time period/variant and of course vaccination didn't exist in the early stages, and now the variants and much different and many of us have been vaccinated - some of us many times! For an exercise like this I'd want to be conservative and use estimates of LC risk on the lower end of the spectrum so as not to seem alarmist (I say this as a committed indoor respirator-wearer - and novid). This is a single study but there are other studies with different estimates of risk and different estimates of cumulative risk. For example, the linked Canadian government report states, "As of June 2023, 19% of Canadian adults infected reported ever experiencing long-term symptoms", which is much lower than the game estimate of 40%. Also, it's important not to conflate 'ever experienced' with 'LC at time point X'. I mean, yes, it could go to 40% or even beyond, but I'm not sure we have enough evidence of this exact number yet.

However the idea that the risk of LC increases with each subsequent infection is, I believe, substantiated. Plus some of the folks who get LC in the first round will not be eligible to get it again as they already/still have it by round 2.
posted by lulu68 at 1:42 AM on February 22

So many people don't realise that the various health problems they are currently experiencing are, in fact, some flavour of long covid from a previous infection

I just don't think there is enough evidence to prove this is true. IMO, suggesting that 19% of people who have gotten COVID have 'some symptoms' of long COVID is just so far out there to be not even sort of true. According to this, I should know dozens of people experiencing long COVID, but it's something I only hear about mostly on Metafilter. Many of the conditions of long COVID are legitimately debilitating, they would be pretty obvious to friends or relations.

That's not to say it's not real, just that it's instance is far lower. Maybe that means that pre-existing conditions, chronic conditions, and maybe age are big factors in getting it, or something else that hasn't been identified yet.

Further, the US employment participation rate (the number of people who choose to work) has stayed pretty stable, only down about 1%, so if 20% of the population have it, it's not effecting their ability to choose to work.

IMO: A realistic version of long COVID would involve 2 dice, and anyone who roles double 1s would get long COVID. After that, the rules could be the same.
posted by The_Vegetables at 7:21 AM on February 22 [3 favorites]

I don't comment anymore generally but I do lurk sometimes, and I felt strongly enough about this to say something. Leaving aside what others have covered re: that the statistics are extremely inaccurate, especially for the levels of risk post-Omicron and boosters and antivirals and hybrid immunity in 2024, this is going to make a lot of students feel horrible. Not just anxious, healthy students scared by an inflated measure of risk, but chronically ill students and people (like me!), and students with long covid and other post-viral illness. We presently exist in a landscape where even the kindest and most allied people treat them like bogeymen, cautionary tales, worst case scenarios. There is very little to tell chronically ill people, again, especially people with post-viral illness, their lives are not haunted or curses. Please don't do this.
posted by colorblock sock at 9:03 PM on February 23 [6 favorites]