Confirmation of COVID-19 infection status & reporting of [LC] symptoms in a population-based birth cohort: No evidence of a nocebo effect, 2024

[SNT Gatchaman]

Confirmation of COVID-19 infection status and reporting of Long COVID symptoms in a population-based birth cohort: No evidence of a nocebo effect
Catherine IA Macleod-Hall; Marcus R Munafò; Maddy L Dyer

Some patients with COVID-19 develop symptoms after the acute infection, known as ‘Long COVID’. We examined whether or not confirmation of COVID-19 infection status could act as a nocebo, using data from questionnaires distributed to the Avon Longitudinal Study of Parents and Children cohort. We examined associations between confirmation of COVID-19 infection status (confirmed by a positive test vs unconfirmed) and reporting of Long COVID symptoms. We explored the roles of sex and anxiety as potential moderators.

There was no clear evidence of a strong association between confirmation of COVID-19 infection status and the Long COVID composite score, physical or psychological symptoms or duration of symptoms. There was no clear evidence of moderation by sex or anxiety. We therefore found no evidence of a nocebo effect.

Our data suggest that this psychological mechanism does not play a role in the medical symptomatology experienced by patients with Long COVID.

Link | Paywall (Journal of Health Psychology)

 

[SNT Gatchaman]

I'm leaving references in some of the quotes.

The aetiology underlying Long COVID is currently unknown. Biological theories have been proposed. [...] However, it is currently unclear whether any of these factors are causal, and if so, what the mechanisms are that they act through. Long COVID has been likened to myalgic encephalomyelitis/chronic fatigue syndrome (ME/ CFS) (Hunt et al., 2022; Poenaru et al., 2021; Wong and Weitzer, 2021), which exists under the umbrella of medically unexplained symptoms despite the cumulating evidence for biological mechanisms (Cortes Rivera et al., 2019).

Psychosocial mechanisms may also contribute to the aetiology of Long COVID. Sykes et al. (2021) argue that the biopsychosocial effects of COVID-19 may lead to physical and psychological symptoms akin to post-traumatic syndromes. However, some are concerned about proposed psychological mechanisms of Long COVID, arguing that they could perpetuate the idea of medical symptomatology in the absence of biological cause, as with the preexisting debate about ME/CFS (Hunt et al., 2022). For example, some patients report that their concerns have been dismissed as symptoms related to anxiety and stress (medical gaslighting) (Yong, 2021). In the case of ME/CFS, discourse focused on psychological mechanisms can be a source of conflict between patients and their physicians, and can lead to harmful interventions (Geraghty and Blease, 2019). It is therefore important to determine the presence or absence of nocebo effects in Long COVID, to inform these discussions and provide guidance for clinicians.

The nocebo effect is ‘the induction or worsening of symptoms induced by sham or active therapies’. Psychological mechanisms (e.g. negative expectations) and neurobiological mechanisms (e.g. the hypothalamus–pituitary–adrenal axis) can produce adverse events. Negative expectations of long-term symptoms following an acute COVID-19 infection could increase anticipatory anxiety and, in turn, contribute to the symptoms experienced. Amanzio et al. (2020) suggest that ‘the COVID-19 era is an unavoidable breeding ground for the possible nocebo effect’. Like Long COVID, nocebo effects appear to be more common in women and in people with anxiety and depression.

This study explored the possible influence of nocebo effects on Long COVID via the following aims: to what degree (1) confirmation of COVID-19 infection status (confirmed vs unconfirmed) is associated with Long COVID symptoms (i.e. duration of COVID-19 symptoms, physical symptoms and psychological symptoms), (2) sex and anxiety are possible moderators of any association between confirmation of COVID-19 infection status and reporting of Long COVID symptoms and (3) COVID-19 infection status is associated with reporting of Long COVID symptoms.

We hypothesised that among participants with confirmed (vs unconfirmed) COVID-19 infection status: (1) number of Long COVID symptoms (a composite score of physical and psychological symptoms) would be higher, (2) number of physical Long COVID symptoms would be higher, (3) anxiety and depression would be higher, and wellbeing would be lower and (4) presence of ongoing COVID-19 (symptoms >4 weeks) and post-COVID-19 syndrome (symptoms >12 weeks) would be greater. We also hypothesised that (5) the magnitude of any associations would be greater among participants who are female and participants who reported having anxiety at the start of the pandemic and (6) number of Long COVID symptoms would be higher among participants who reported having had COVID-19 compared to participants who reported not having had COVID-19.

 

[SNT Gatchaman]

Contrary to hypotheses 1 and 2, we found that confirmed (vs unconfirmed) COVID-19 infection was not associated with a greater incidence of Long COVID symptoms.

Overall, the results suggest that confirmation of a COVID-19 diagnosis does not exert nocebo effects that contribute to the development of Long COVID. This contrasts with a cross-sectional study of French adults, which found that self-reported COVID-19 infection was associated with more persistent physical symptoms, whereas COVID-19 infection confirmed by serology was associated only with anosmia (Matta et al., 2022).

Contrary to hypotheses 3 and 4, there was no clear evidence of moderation by sex or anxiety. This contrasts with a previous study which found that certainty of being infected with COVID-19 was associated with reporting more COVID-19 symptoms, and that this effect was greater in participants with anxiety (Daniali and Flaten, 2022).

Finally, contrary to hypothesis 6, those who reported a previous COVID-19 infection reported fewer physical symptoms than those who did not.

Our study found no clear evidence of a nocebo effect on the development of Long COVID, adding to our understanding of the pathophysiology of this poorly understood condition and providing reassurance to clinicians interacting with patients with COVID-19. There have historically been concerns about nocebo effects arising when diagnosing conditions such as ME/CFS (Huibers and Wessely, 2006). Whilst our study focused on the effect of acute COVID-19 infection diagnosis, rather than Long COVID diagnosis, it lends evidence against this argument.

Confirmation of COVID-19 infection by positive test was not associated with increased reporting of Long COVID symptoms. Furthermore, there was no clear evidence of moderation by either sex or anxiety. There was no evidence that a COVID-19 diagnosis exerted nocebo effects contributing to the aetiology of Long COVID.

 

[dave30th]

Interesting that this is coming from Univ of Bristol--home of the methologically and ethically challenged Professor Crawley. I wonder what she thinks of this study??

 

[rvallee]

There have historically been concerns about nocebo effects arising when diagnosing conditions such as ME/CFS (Huibers and Wessely, 2006).

Hum, not quite. Claims. Baseless assertions. Used to support entire treatment models and derived services. It has gone way, way beyond concerns, it has been operationalized for decades based on nothing but 19th century vibes.

It started with beliefs, not really concerns, but it has snowballed since then and launched avalanches that consumed enough people to make a mid-sized nation.

Although a problem remains here that you can't prove a negative. But the belief in nocebo has never been proven, and yet it is fully belief to the degree where they let people die for those beliefs. The real issue is baseless beliefs having been turned into real-life decisions, without evidence, in fact against all evidence.

 

[Hutan]

The nocebo effect is ‘the induction or worsening of symptoms induced by sham or active therapies’.

Isn't that an odd definition? Surely 'the induction or worsening of symptoms induced by ...an active therapy' is the production of side effects. Some of those side effects may well be entirely real and explicable.

Like Long COVID, nocebo effects appear to be more common in women and in people with anxiety and depression.

There was no clear evidence of moderation by sex or anxiety. We therefore found no evidence of a nocebo effect.

I think this idea that nocebo effects are more common in women is probably rubbish. I'm not sure about the logic in that last quote. They seem to be saying that the fact that being a woman didn't moderate the associations was evidence that there was no nocebo effect. That seems pretty circular.


While it's great that the authors concluded that

this psychological mechanism does not play a role in the medical symptomatology experienced by patients with Long COVID

I'm not sure about their basis for doing that.

There was no clear evidence of a strong association between confirmation of COVID-19 infection status and the Long COVID composite score, physical or psychological symptoms or duration of symptoms....
...we found that confirmed (vs unconfirmed) COVID-19 infection was not associated with a greater incidence of Long COVID symptoms.

Other authors have concluded that Long Covid is psychosomatic for the same reason as these authors conclude that Long Covid isn't psychosomatic. That is, that Long Covid is occurring in people who don't have confirmed Covid-19 infections.


I think the problems with access to and accuracy of diagnostic testing makes studies attempting to conclude anything about psychosomaticism as the cause of Long Covid on the basis of reported Covid-19 infection status pretty worthless.