Very interesting! I’d be interested in seeing some microCOVID-style expected value estimates here. Like, approximately how much does this reduce your risk: from what to what? And how does this translate to EV? How valuable is it to avoid such illness?
My guess is that it’s only worth a few dollars at most, in which case it isn’t something I’d want to bother with. But if the benefit gets into the $25-50+ range, I’d be thinking about it for my next flight.
The short version is that the best evidence I can find suggests somewhere between 20% and 100% increased risk of infection and most likely getting more sick but I’m unsure how to quantify that.
mice infected [with MuHV-4] intranasally at the onset of resting phase [...], exhibited 10-fold higher viral replication than mice infected just before their active phase
I don’t entirely follow the details of the paper but the authors do other cell-level tests with HSV-1 and influenza A that seem to imply we should see similar results if we infected mice with them.
We found that viral infection [using VSV] at ZT12 [start of active period] resulted in a 40% mortality rate, as expected with an LD50 dose. The same dose of virus administered at ZT0 [end of active period] led to a 95% mortality rate (Fig. 1A)
It’s not quite the same thing since this also involves circadian disruption but this study[3] finds that:
The risk of severe COVID-19 was higher in health workers (adjusted odds ratio: 2.32 [95% CI: 1·33, 4·05]; shift workers (2·06 [1·72, 2·47]); and in health workers who worked [night] shifts (7·56 [3·86, 14·79]).
Translation: Working night shifts doubled your risk of severe COVID (similar to being a healthcare worker), and being a healthcare worker who worked night shifts put you at 7.5x the standard risk.
Shift workers’ incidence rate of ILI/ARI was 1.20 (95% confidence interval (CI): 1.01, 1.43) times higher than that of non–shift workers, and for severe ILI/ARI episodes, shift workers’ incidence rate was 1.22 (95% CI: 1.01, 1.49) times higher. The mean duration of an ILI/ARI episode did not differ (ratio between means = 1.02, 95% CI: 0.87, 1.19).
Meaning shift workers got sick 20% more often, although I don’t like that the confidence interval just barely doesn’t include 1.
None of this evidence is amazing, but the studies all seem to at least agree on the direction of the effect.
R.S. Edgar, A. Stangherlin, A.D. Nagy, M.P. Nicoll, S. Efstathiou, J.S. O’Neill, & A.B. Reddy, Cell autonomous regulation of herpes and influenza virus infection by the circadian clock, Proc. Natl. Acad. Sci. U.S.A. 113 (36) 10085-10090, https://doi.org/10.1073/pnas.1601895113 (2016).
A.V. Rowlands, C. Gillies, Y. Chudasama, M.J. Davies, N. Islam, D.E. Kloecker, C. Lawson, M. Pareek, C. Razieh, F. Zaccardi, T. Yates, K. Khunti. Association of working shifts, inside and outside of healthcare, with risk of severe COVID-19: An observational study. medRxiv 2020.12.16.20248243; doi: https://doi.org/10.1101/2020.12.16.20248243
Loef, B., van Baarle, D., van der Beek, A. J., Sanders, E. A. M., Bruijning-Verhagen, P., & Proper, K. I. (2019). Shift work and respiratory infections in health-care workers. American Journal of Epidemiology, 188(3), 509-517. https://doi.org/10.1093/aje/kwy258
Claude estimated for me a 1-in-100 to 1-in-1,000 chance of getting sick on a flight as a baseline. Let’s say 1-in-500. Ballparking the value of avoiding some sort of illness at $500, the baseline cost is $1. So then, if booking a later flight leads to a 20-100% decreased risk of illness, it doesn’t seem like it’d save more than a dollar or so in EV.
If you value avoiding illness at something higher like $5,000, then the expected savings are more like $10 or so, which maybe becomes worth it, although if you value avoiding illness that highly might be worth wearing a P100 mask or something, in which case the value of a later flight goes down to something more like $1, I’d guesstimate.
I might be unusually unlucky, but I get sick on more like 1 in 10 flights. Price-wise I don’t think middle-of-the-night flights are worth it for more mundane reasons (being awake all night has a 100% chance of ruining my next day), so for me it’s entirely about scheduling.
Assuming each sickness makes me sick for 10 days, would I rather pick a less convenient schedule or be sick for (on average) an additional 5 hours? It might depend on the trip but I usually don’t do much on my traveling-home days anyway, so it seems like I should at least travel home mid-day, even though I might be willing to travel out late to maximize vacation days (trade 5 hours of sickness for one vacation day).
Very interesting! I’d be interested in seeing some microCOVID-style expected value estimates here. Like, approximately how much does this reduce your risk: from what to what? And how does this translate to EV? How valuable is it to avoid such illness?
My guess is that it’s only worth a few dollars at most, in which case it isn’t something I’d want to bother with. But if the benefit gets into the $25-50+ range, I’d be thinking about it for my next flight.
The short version is that the best evidence I can find suggests somewhere between 20% and 100% increased risk of infection and most likely getting more sick but I’m unsure how to quantify that.
This study on mice[1] found that:
I don’t entirely follow the details of the paper but the authors do other cell-level tests with HSV-1 and influenza A that seem to imply we should see similar results if we infected mice with them.
This study[2] (also mice) found that:
It’s not quite the same thing since this also involves circadian disruption but this study[3] finds that:
Translation: Working night shifts doubled your risk of severe COVID (similar to being a healthcare worker), and being a healthcare worker who worked night shifts put you at 7.5x the standard risk.
And this study[4] finds that:
Meaning shift workers got sick 20% more often, although I don’t like that the confidence interval just barely doesn’t include 1.
None of this evidence is amazing, but the studies all seem to at least agree on the direction of the effect.
R.S. Edgar, A. Stangherlin, A.D. Nagy, M.P. Nicoll, S. Efstathiou, J.S. O’Neill, & A.B. Reddy, Cell autonomous regulation of herpes and influenza virus infection by the circadian clock, Proc. Natl. Acad. Sci. U.S.A. 113 (36) 10085-10090, https://doi.org/10.1073/pnas.1601895113 (2016).
K. Gagnidze, K.H. Hajdarovic, M. Moskalenko, I.N. Karatsoreos, B.S. McEwen, & K. Bulloch, Nuclear receptor REV-ERBα mediates circadian sensitivity to mortality in murine vesicular stomatitis virus-induced encephalitis, Proc. Natl. Acad. Sci. U.S.A. 113 (20) 5730-5735, https://doi.org/10.1073/pnas.1520489113 (2016).
A.V. Rowlands, C. Gillies, Y. Chudasama, M.J. Davies, N. Islam, D.E. Kloecker, C. Lawson, M. Pareek, C. Razieh, F. Zaccardi, T. Yates, K. Khunti. Association of working shifts, inside and outside of healthcare, with risk of severe COVID-19: An observational study. medRxiv 2020.12.16.20248243; doi: https://doi.org/10.1101/2020.12.16.20248243
Loef, B., van Baarle, D., van der Beek, A. J., Sanders, E. A. M., Bruijning-Verhagen, P., & Proper, K. I. (2019). Shift work and respiratory infections in health-care workers. American Journal of Epidemiology, 188(3), 509-517. https://doi.org/10.1093/aje/kwy258
Claude estimated for me a 1-in-100 to 1-in-1,000 chance of getting sick on a flight as a baseline. Let’s say 1-in-500. Ballparking the value of avoiding some sort of illness at $500, the baseline cost is $1. So then, if booking a later flight leads to a 20-100% decreased risk of illness, it doesn’t seem like it’d save more than a dollar or so in EV.
If you value avoiding illness at something higher like $5,000, then the expected savings are more like $10 or so, which maybe becomes worth it, although if you value avoiding illness that highly might be worth wearing a P100 mask or something, in which case the value of a later flight goes down to something more like $1, I’d guesstimate.
I might be unusually unlucky, but I get sick on more like 1 in 10 flights. Price-wise I don’t think middle-of-the-night flights are worth it for more mundane reasons (being awake all night has a 100% chance of ruining my next day), so for me it’s entirely about scheduling.
Assuming each sickness makes me sick for 10 days, would I rather pick a less convenient schedule or be sick for (on average) an additional 5 hours? It might depend on the trip but I usually don’t do much on my traveling-home days anyway, so it seems like I should at least travel home mid-day, even though I might be willing to travel out late to maximize vacation days (trade 5 hours of sickness for one vacation day).