Post-acute non-specific symptoms following COVID-19 vaccination: a Danish population-based study, 2025, Jensen et al

[forestglip]

Post-acute non-specific symptoms following COVID-19 vaccination: a Danish population-based study

Christina Bisgaard Jensen, Kristoffer Torp Hansen, Bodil Hammer Bech, Stefan Nygaard Hansen, Henrik Nielsen, Charlotte Ulrikka Rask, Per Fink, Thomas Meinertz Dantoft, Torben Jørgensen, Jeremy A Labrecque, Sanne Møller Thysen, Dorte Rytter

[Line breaks added]

Background
Post-acute non-specific symptoms such as headaches, muscular pain, and fatigue are frequently reported following COVID-19 vaccination, raising concerns about adverse events. As vaccine-hesitant individuals may be more likely to experience symptoms of vaccines, this study investigates whether COVID-19 vaccination increases the odds of post-acute non-specific symptoms, stratified by COVID-19 vaccine hesitancy.

Methods
This longitudinal study uses repeated measurements from the BiCoVac cohort—a random sample of 913 116 Danish citizens aged 16–65 years. Data were collected through national registers and questionnaires (May 2021–June 2022). Non-specific symptoms were identified by using the 25-item Bodily Distress Syndrome checklist and analysed by using logistic and linear regression models comparing vaccinated and unvaccinated individuals. Models were fitted with robust standard errors and inverse probability of selection weights.

Results
A total of 132 776 observations were collected from 61 316 COVID-19 vaccine-unconcerned individuals, 52 325 observations from 25 272 COVID-19 vaccine-sceptical individuals, and 10 242 observations from 5064 COVID-19 vaccine-concerned individuals.

Among the vaccine-unconcerned individuals, those vaccinated had lower odds of most symptoms compared with unvaccinated individuals >4 weeks post-vaccination [e.g. muscular pain: odds ratio (OR) = 0.56, 95% confidence interval (CI): 0.47–0.68]. A similar tendency was found for vaccine-sceptical individuals, although with attenuated estimates.

Conversely, among vaccine-concerned individuals, we observed a tendency for higher odds of most symptoms >4 weeks post-vaccination compared with unvaccinated individuals (e.g. muscular pain: OR = 1.23, 95% CI: 0.91–1.67).

Conclusion
For most individuals aged 16–65 years, we found no evidence of post-acute symptoms after COVID-19 vaccination. Vaccinated vaccine-concerned individuals were, however, more likely to report most symptoms, potentially due to reporting bias or nocebo effects.

Web | International Journal of Epidemiology | Paywall

 

[Utsikt]

For most individuals aged 16–65 years, we found no evidence of post-acute symptoms after COVID-19 vaccination. Vaccinated vaccine-concerned individuals were, however, more likely to report most symptoms, potentially due to reporting bias or nocebo effects.

I think reporting bias is fair. The nocebo effect is questionable.

I think it’s plausible that some of the people that were concerned about vaccines had a legitimate reason for it, e.g. a history of reacting badly to vaccines or poorer health. Without knowing if they controlled for that, it’s difficult to assess the claims made by the authors.

The use of the BDS questionnaire implies a certain bias among the authors.

 

[forestglip]

I think reporting bias is fair. The nocebo effect is questionable.

Why is nocebo effect questionable as a potential explanation?

They both seem fine to me as speculation, but I agree, they should have also included a third option of vaccine-concerned people having a reason to be concerned about their own bodies.

 

[Utsikt]

Why is nocebo effect questionable as a potential explanation?

To invoke nocebo based on this kind of data is pure speculation. By definition, nocebo is what’s left after every other explanation has been disproven, and the only remaining factor for explaining the reported symptoms is people’s expectations of experiencing symptoms.

I find the entire concept pf placebo to be very questionable and I suspect that the appearnce of the existence of nocebo is mostly a result of various biases and confounders that have not been accounted for.

 

[forestglip]

To invoke nocebo based on this kind of data is pure speculation.

So is invoking reporting bias, no? Or suggesting that they had more adverse effects for a legitimate health reason?

By definition, nocebo is what’s left after every other explanation has been disproven, and the only remaining factor for explaining the reported symptoms is people’s expectations of experiencing symptoms.

Based on what definition? Wikipedia says "A nocebo effect is said to occur when a patient's expectations for a treatment cause the treatment to have a worse effect than it otherwise would have." It's just a possible explanation, like other possible explanations.

Hypothetically, the placebo effect could be nailed down to a specific mechanism in the brain. Then it'd just be another biological phenomenon. The only difference is that currently the mechanism isn't known. I don't think you need to exhaust every other possible explanation before even suggesting an effect where the mechanism is not known.

 

[Utsikt]

So is invoking reporting bias, no? Or suggesting that they had more adverse effects for a legitimate health reason?

Fair point, but I’d still argue that those are more reasonable and probable explanations, so not mentioning them ik the abstract in favour of nocebo implies a preference among the authors.

Based on what definition? Wikipedia says "A nocebo effect is said to occur when a patient's expectations for a treatment cause the treatment to have a worse effect than it otherwise would have." It's just a possible explanation, like other possible explanations.

The key point is that the adverse effect is caused by expectations. We have no way of determining what someone’s expectations are because we don’t know how thoughts work on a physical level so we’d have to rely on subjective reports that by themselves are influenced by a whole host of confounders and biases so the reports quickly become too unreliable.

This means that the nocebo effect is unverifiable and incalculable for the foreseeable future.

I don't think you need to exhaust every other possible explanation before even suggesting an effect where the mechanism is not known.

So you’d be fine with me suggesting that it was the being Nocebus from a parallell universe that caused it? I see that as equally scientifically valid as suggesting it’s the nocebo effect.

 

[forestglip]

Fair point, but I’d still argue that those are more reasonable and probable explanations, so not mentioning them ik the abstract in favour of nocebo implies a preference among the authors.

Well they did mention one of the two, but I agree, excluding the other suggests like they might prefer these explanations.

So you’d be fine with me suggesting that it was the being Nocebus from a parallell universe that caused it? I see that as equally scientifically valid as suggesting it’s the nocebo effect.

Unlike nocebus from a parallel universe which has no existing evidence or reason for believing in, a placebo effect is biologically plausible.

Thoughts can make my heart race or my skin flush. So the idea that certain thoughts/feelings like "I am receiving a drug" can lead to other thoughts/feelings like "I feel good/bad" seems like a simple, straightforward hypothesis to consider when a group gets better even though they haven't taken any drug (along with all the other things that should be considered, like some type of confounding or bias).

 

[Utsikt]

Unlike nocebus from a parallel universe which has no existing evidence or reason for believing in, a placebo effect is biologically plausible.

Thoughts can make my heart race or my skin flush. So the idea that certain thoughts/feelings like "I am receiving a drug" can lead to other thoughts/feelings like "I feel good/bad" seems like a simple, straightforward hypothesis to consider when a group gets better even though they haven't taken any drug (along with all the other things that should be considered, like some type of confounding or bias).

There is a long way to go from getting people to say say «I feel better/worse» to having the mental state of expecting an adverse effect when receiving the vaccine in itself causing long-lasting symptoms.

We know how easy it is to get people to report changes that are not there objectively. So you’d still be stuck with the issues of bias and confounders when people report symptoms.

 

[forestglip]

We know how easy it is to get people to report changes that are not there objectively.

Maybe we're working with different ideas of what placebo means. I would consider if someone swore that they felt feverish after taking no medication, even if a thermometer does not indicate any higher temperature, would be a placebo effect. The effect would just be on the subjective feeling of being hot, instead of an objective measurement of temperature.

 

[Utsikt]

Maybe we're working with different ideas of what placebo means.

Looks like it.

I would consider if someone swore that they felt feverish after taking no medication, even if a thermometer does not indicate any higher temperature, would be a placebo effect. The effect would just be on the subjective feeling of being hot, instead of an objective measurement of temperature.

The placebo and nocebo folks appear to be pretty adamant that just expecting to get a fever could by itself result in you getting a fever. And similarly, that just expecting that your fever will reduce could by itself result in your fever reducing.

I have no doubt that someone could get me to report feeling better or worse after doing something to me.

 

[forestglip]

The placebo and nocebo folks appear to be pretty adamant that just expecting to get a fever could by itself result in you getting a fever. And similarly, that just expecting that your fever will reduce could by itself result in your fever reducing.

I have no doubt that someone could get me to report feeling better or worse after doing something to me.

Yeah I just think there may or may not be certain subjective and objective measures that can be influenced by expectations. I certainly think it's unlikely that all objective outcomes could be influenced by expectations.

If a patient got a placebo and was told it was a drug designed to increase heart rate, I wouldn't be surprised if their heart rate increased just based on that, or vice versa. But I'd be more skeptical about the idea that a sugar pill could clear up someone's foot fungus, for example, unless they start maybe living a healthier lifestyle overall.

 

[Utsikt]

If a patient got a placebo and was told it was a drug designed to increase heart rate, I wouldn't be surprised if their heart rate increased just based on that, or vice versa.

Sure, but that might just as well be due to excitement or other mental states. You’d have a hard time proving it’s caused by the mental state of expectation.