Preprint Fatigue >12 weeks after coronavirus disease COVID is associated with reduced reward sensitivity during effort-based decision making, 2024, Scholing+

[SNT Gatchaman]

Fatigue >12 weeks after coronavirus disease COVID is associated with reduced reward sensitivity during effort-based decision making
Judith M. Scholing; Britt I.H.M Lambregts; Ruben van den Bosch; Esther Aarts; Marieke E. van der Schaaf

BACKGROUND
Fatigue and depressive mood is inherent to acute disease, but a substantial group of people report persisting disabling fatigue and depressive symptoms long after a COVID infection. Acute infections have shown to change decisions to engage in effortful and rewarding activities, but it is currently unclear whether fatigue and depressive symptoms similarly affect decision making during acute and persistent phases of a COVID infection. Here, we investigated whether fatigue and depressive mood are associated with altered weighting of effort and reward in decision-making during different timepoints after COVID infection.

METHODS
We conducted an online cross-sectional study between March 2021 and March 2022. 242 participants 18-65 years) with COVID <4 weeks ago (n=62), COVID >12 weeks ago (n=81), or no prior COVID (self-reported) (n=90) performed an effort based decision-making task, in which they decided whether they wanted to exert physical effort (ticking boxes on screen, 5 levels) for reward (money to be gained in a voucher-lottery, 5 levels). State fatigue and depressive mood was measured by the Profile of Mood States (POMS) prior to the task. We used multilevel binomial regression analysis to test whether fatigue and depressive mood were related to acceptance rates for effort and reward levels and whether this differed between the groups.

RESULTS
Compared with no COVID and COVID <4 weeks groups, the COVID >12 weeks group reported higher state fatigue scores (mean±SD: 20±7 vs. 14±7 and 12±6 POMS-score, respectively; both p<0.001) and was less sensitive to rewards (Reward*Group: OR: 0.35 (95%CI 0.20, 0.62), p<0.001 and OR: 0.38 (95%CI 0.20, 0.72), p=0.003). In the COVID >12 weeks group, fatigue was more negatively associated with reward sensitivity compared with the COVID <4 weeks group (Reward*Fatigue*Group: OR 0.47 (95%CI 0.25, 1.13), p=0.022) and the no COVID group (Reward*Fatigue*Group: OR 0.48 (95%CI 4.01, 0.92), p=0.029). No group differences were observed for the relationship between fatigue and effort sensitivity. No group differences were observed for the relationship between depressive mood and effort or reward sensitivity. Higher age, lower BMI, unhealthy lifestyle, and worrying during the acute phase of COVID each predicted decreased reward sensitivity in the >12 weeks group (Age*Reward: OR 0.30 (95%CI 0.19, 0.48), p<0.001; BMI*Reward: OR 1.43 (95%CI 1.01, 2.00), p=0.047); Lifestyle*Reward: OR 1.50 (95%CI 1.06, 2.14), p=0.022; Worrying*Reward: OR 0.59 (95%CI 0.38, 0.94), p=0.025, respectively).

CONCLUSIONS
The finding that fatigue is related to lower reward sensitivity >12 weeks after COVID, suggesting potential reward deficits in post-covid fatigue. These findings are in line with previous observations that long-term inflammation can induce dysregulations in neural reward processing, which should be further investigated in future studies.

HIGHLIGHTS:
• We tested if fatigue and mood were related to altered decision making post COVID

• Participants post-COVID >12wks ago were more fatigued and less reward sensitive

• Post-COVID-related fatigue was associated with reduced reward sensitivity

• Post-COVID-related depressive mood was not associated with altered decision making

• Higher age, unhealthy lifestyle, and worrying predicted reward deficits

Link | PDF (Preprint: BioRxiv) [Open Access]

 

[SNT Gatchaman]

We hypothesized that fatigue in both the <4 weeks and >12 weeks group is more strongly associated with increased effort sensitivity compared to the no COVID group. Additionally, we hypothesize that depressive mood is more strongly associated with decreased reward sensitivity only in the >12 weeks group compared to the <4 weeks and no COVID groups

Fatigue was, in contrast to our hypothesis, not related to effort sensitivity.

in the current study, we separated the decision phase from the execution phase to rule out potential fatigue fluctuations due to task execution, which makes it less comparable to these studies in healthy individuals. Nevertheless, although effort sensitivity did not differentiate between the groups, th e findings with respect to reward sensitivity still indicate that the fatigue experienced in the >12 weeks phase, as compared to fatigue in the acute phase, may be driven by different biological mechanisms, possibly altered dopamine transmission due to an extended period of raised inflammation levels.

This study had several limitations. First, the population in the >12 weeks group was a mix of recovered individuals and individuals with a diagnosis of POST -COVID and we do not know whether they had a diagnosis for POST-COVID. We can therefore not conclude that the reduction in reward sensitivity is a characteristic of POST-COVID syndrome or merely fatigue after COVID.

we used a measure of state fatigue (i.e. fatigue at the moment of testing) rather than trait fatigue (i.e. fatigue in the past two weeks) in our analysis.

In conclusion, we found evidence that fatigue, but not depressive mood, occurring >12 weeks after COVID infection is associated with reward deficits compared to people in the acute phase of COVID or who did not have COVID.

 

[bobbler]

Am I the only one thinking it’s a bit strange we have another one suddenly deciding to do this new fangled angle of ‘effort-based’ when the decision for this can’t have been preceded by NIH results (too close) and I don’t know how they’d have known about its inclusion before their release.

When was the effort preference idea and task even introduced by Wallit into the NIH thing? Do we know? Was everything signed off from the start or bits introduced over those years?

And that I’m guessing these are just general psychosomaticists not people who done effort preference for a living and found a new illness to try.

it just seems too obscure a new angle to have two suddenly deciding to use as their new big thing bith at the same time without it being ‘put there’


Though useful that hopefully both failing to find much and then struggling to scratch around for something to scrape from it as a claim might put of those who are just looking for a low-hanging fruit new way of doing money for old rope realise this isn’t the ‘elsewhere’ to look for easy inferences even if these tasks are ready-made seemingly

 

[bobbler]

So they separated out the decision phase from the execution phase

and us this what they think means there is now such a thing as having reward sensitivity without effort sensitivity ie no difference in effort?

I’ll take a closer look when I can but intrigued on what they’ve assumed they can tell this From - looks like fishing for finding anything once they realised there weren’t the differences they wanted anywhere

but I haven’t looked at the hypothesis design and nulls to see if this was specifically included and if these were signed off before etc

Was it something like the decision screen of whether you play high or low effort vs the number of times high effort selected they are talking about?

Even if not it’s very useful we have that NIH full data forevermore now to inform analysis of all future tests like this even if they don’t - because I think it’s given a very good understanding of where the different elements lie snd patterns etc to cross-compare even across studies?

 

[SNT Gatchaman]

They don't seem to reference Walitt et al or mention EEfRT as far as I could see on a quick skim.

 

[SNT Gatchaman]

suggesting potential reward deficits in post-covid fatigue

I guess you could call PEM a "reward deficit".

(See also the thread on "Do you recover AND improve from exercise?")

 

[Louie41]

I too was struck by how this seems to be related to "effort preference." Also, I wondered how they could separate out fatigue from "depressive symptoms" since they are both characteristic of acute illness. More psychologizing, I guess, to no useful purpose..

 

[Hutan]

Oh, where to start with this one?

Characterisation of the groups
The >12 week group included both people who had recovered and people who still had symptoms. '12 week' doesn't mean having persisting symptoms for 12 weeks. It just means that the people had a Covid-19 infection more than 12 weeks ago. The authors don't know how many people had Long Covid, let alone any finessing about whether people had ME/CFS and/or, for example, lung damage.

This study had several limitations. First, the population in the >12 weeks group was a mix of recovered individuals and individuals with a diagnosis of POST-COVID and we do not know whether they had a diagnosis for POST-COVID. We can therefore not conclude that the reduction in reward sensitivity is a characteristic of POST-COVID syndrome or merely fatigue after COVID.

They do go on to say that 80% of the >12 week group said that they were still recovering from their Covid-19 infection, but, it's pretty loose.

The authors suggest their finding might help with identifying treatments to help people with persistent symptoms following a Covid-19 infection. But, if they haven't even bothered to separate out the person who has fatigue due to breathing problems as a result of damaged lungs from the person who has ME/CFS, well, ...good luck finding a treatment.

Confounding due to non-homogeneous selection
The >12 week group was older and richer and more female than the other two groups, substantially so. The mean age of the No Covid group was 28 years; the mean age of the >12 week group was 43 years. The mean income of the No Covid group was 38,000 euros; the mean income of the >12 week group was 56,000 euros. The percentage of males in the No Covid group was 21%; the percentage of males in the >12 week group was 13%. If we think about the task, which required manual dexterity with a mouse to rapidly click on boxes on a screen, clearly, it's the sort of task that gamers would have an advantage in - older females are not generally the main gamer demographic, and manual dexterity typically reduces with age.

The task seems to have offered a single 5 euros up as the prize, across all of the participants. Each task was worth cents, up to the 5 euro limit, towards a potential win, and then the participant was part of the lottery with the hundreds of others to win the portion of the 5 euros that they had accumulated. The authors aren't very clear how it worked and I haven't spent much time picking through things. Suffice to say though, the reward was paltry. People were unlikely to have been particularly motivated by the reward. Rich people were less likely to have been motivated by the monetary reward.

Arbitrary removal of some of the participants

75 of the 308 participants were excluded because they 1) accepted more than 90% or less than 10% of the offers in the decision phase (n= 67), 2) they pressed only one button in >70% of the decision trials (n= 3)

I think the problems with this study make it pretty worthless, in terms of making any conclusions about people with persisting symptoms following Covid-19 infection. I think this study would reward effort in terms of spending time finding problems with it. But, uh...

 

[Hutan]

That said, I note:

no group differences were observed in emotional well-being or limitations due to emotional problems.

They also show that the COVID <4 weeks and COVID >12 weeks groups reported overall more physical and fatigue-related symptoms compared to participants who did not have COVID, while symptoms of feeling depressed, having a low motivation, feeling stressed/anxious did not differ between the three groups.

The three groups also experienced a similar amount of pandemic-related stress

 

[Turtle]

At least there were 75 participants smarter than the researchers by boycotting this way of testing. BPS still runs rampant here.
When LC came up fysiotherapists, psychologists (CBT) and occupational therapist lined up to "help". No clue about LC and the "help" they could offer.
Those "therapies" are not paid for by insurance anymore, because they were not effective.

 

[Evergreen]

A little compare-and-contrast with the EEfRT in the NIH study by Walitt et al:

Unlike Walitt et al, Scholing et al calibrated:

During calibration, we measured how many boxes the participant was able to tick with their computer mouse or on their tablet’s touch screen within 10 seconds. This was repeated three times, and the highest number of boxes ticked (maximum number of boxes ticked (MBT)) was used to determine the five levels of effort (i.e. 10, 30, 50, 70, and 90% of MBT). During familiarization, participants performed all five levels of effort three times to get familiar with each effort level. To validate the experienced task load, participants filled in the NASA Task Load index after each effort level (Hart, 1986).

Ability did not differ between groups in the calibration phase (not assessed by Walitt et al):

The mean (standard deviation (SD)) maximum number of clicks during the calibration phase across all groups was 24±7 and did not differ between the groups (F=1.78, df=2, p=0.171), indicating that the groups did, on average, not differ in their ability to click the boxes.

Ability did not differ between groups in the execution phase either (unlike in Walitt et al):

During the execution phase, the mean±SD success rate of the trials was 97±6%, and mean±SD total rewards obtained during the execution phase was 46±14 apples, corresponding with €5,- + €1.84±0.56 worth in gift cards, all of which did not differ between the groups either (F=0.571, df=2, p=0.571), indicating that all participants were highly successful in executing the accepted offers.

Scholing et al's interpretation of their task-that-was-not-EEfRT findings is that fatigue was associated with reduced reward sensitivity, but not reduced effort sensitivity. Walitt et al's interpretation of their EEfRT findings was that ME/CFS was associated with reduced effort sensitivity, but not reward sensitivity.

Other interesting tidbits:
Selection in Scholing et al (patients were not diagnosed with anything or seen by a doctor):

In total, we recruited 308 Dutch-speaking participants aged 18-65 years old through social media (e.g. post-COVID patient groups), flyers and the university’s online recruitment database

SF36 Physical Function subscale scores for Scholing et al's subgroups:
no COVID group: 91±13
COVID less than 4 weeks prior: 74±24
COVID more than 12 weeks prior: 51±27

Medium intensive exercising, pre-pandemic (h per week) p=0.001 (no difference in high intensive exercise)
no COVID group: 2.90±1.73
COVID less than 4 weeks prior: 3.19±1.56
COVID more than 12 weeks prior: 3.94±1.67

Alcohol use, pre-pandemic (times per month) p=0.035
no COVID group:3.7±4.6
COVID less than 4 weeks prior: 5.7±6.0
COVID more than 12 weeks prior:3.8±4.7

Edited to fix spelling.

 

[rvallee]

Post-COVID-related fatigue was associated with reduced reward sensitivity

Fatigue was, in contrast to our hypothesis, not related to effort sensitivity.

Well, thanks for muddying this up. Because of course everyone has the same baseline of those to begin with. And for a mere chance of a few bucks. Dance, monkey, dance for a few cents. Come on, show some enthusiasm here! Let's see that big smile of yours and you may get enough to buy a... stick of gum or whatever. It's not as if this is a valid way of testing for anything but whatever. Up? Also down. And left. And mauve. Also, umami and a bit of the 3rd track of Pink Floyd's 4th album.

I don't see how it makes sense to have people studying illness when they clearly have no relevant concept of what illness is or does. It's completely alien to them. Their model of illness is one where people can basically function completely normally with zero impairment in performance or quality of life if they just don't let it bother them. Only disease matters. Illness is a state of mind with zero physical properties or relevance in the real world.

I don't think we'd regress significantly if things just went back to 16th century medical practices. It would just be a lateral move. The only real difference is the rest of society having made significant technological progress, allowing social safety nets and basic needs to be more easily fulfilled. Their understanding of illness has not progressed one bit since then. The only actual difference is everything else, they are completely stuck in time.

 

[Arvo]

When was the effort preference idea and task even introduced by Wallit into the NIH thing? Do we know? Was everything signed off from the start or bits introduced over those years?

The "effort preference" term appears to be a newly coined label -quite possibly inspired by Van der Schaaf, Knoop & co's writings- put on Walitt's old belief on fibromyalgia (and ME) that they are "interoceptive disorders" by which he means a disorder of subjective perception.

Walitt got to do the thing he's been doing for at least the past 15 years (I'm out of the topic, but he opened his first unit on "interoceptive disorders" in 2010 before he pursued the topic and got to educate his colleagues at the NIH have-you-tried-yoga department.).

They don't seem to reference Walitt et al or mention EEfRT as far as I could see on a quick skim.

It could be a strategic choice for the moment. Either way, Walitt referenced them - and they said the same thing just in other words.

Walitt and Van der Schaaf similarities discussed (by me) here, here and here.

Walitt also used work by Chalder and White on this topic. (It's quite the rage among the CBT gang.)

 

[Arnie Pye]

All the mentions of "effort" make me think that doctors assume the patient isn't trying hard enough.