(Preprint) Excess risk and clusters of symptoms after COVID-19 in a large Norwegian cohort by Caspersen et al.

Moved post

Excess risk and clusters of symptoms after COVID-19 in a large Norwegian cohort
https://www.medrxiv.org/content/10.1101/2021.10.15.21265038v1

(Paragraphs mine for legibility)

We aimed to calculate the excess risk and identify patterns of 22 symptoms up to 12 months after COVID-19 infection. We followed more than 70,000 participants in an ongoing cohort study, the Norwegian Mother, Father and Child Cohort Study (MoBa) during the COVID-19 pandemic. Infected and non-infected cohort participants registered presence of 22 different symptoms in March 2021.

One year after the initial infection, 13 of 22 symptoms were associated with SARS-CoV-2 infection, based on relative risks between infected and uninfected subjects. For instance, 17.4% of SARS-CoV-2 infected cohort participants reported fatigue that persist 12 months after infection, compared to new occurrence of fatigue that had lasted less than 12 months in 3.8% of non-infected subjects (excess risk 13.6%).

The adjusted relative risk for fatigue was 4.8 (95 % CI 3.5 to 6.7). Two main underlying factors explained 50% of the variance in the 13 symptoms. Brain fog, poor memory, dizziness, heart palpitations, and fatigue had high loadings on the first factor, while shortness-of breath and cough had high loadings on the second factor. Lack of taste and smell showed low to moderate correlation to other symptoms. Anxiety, depression and mood swings were not strongly related to COVID-19.

Our results suggest that there are clusters of symptoms after COVID-19 due to different mechanisms and question whether it is meaningful to describe long COVID as one syndrome.​

 

Moved post

The media here today is covering a preprint from the Norwegian Institute of Public Health on Long Covid. The paper is titled: Excess risk and clusters of symptoms after COVID-19 in a large Norwegian cohort and is based on another project where the institute is following 70,000 citizens by sending them a survey twice a month.


This is a google translated summary from the institute's website:

• Among people who had not been diagnosed with SARS-CoV-2 (corona), 21 percent reported at least one new symptom that had occurred during the last twelve months.

• Among people who had been diagnosed with SARS-CoV-2 in the spring of 2020, just over half (56 per cent) reported at least one new symptom that had occurred after the time of infection twelve months earlier.

• Of the 22 different ailments examined, the researchers found that 13 of these were reported more frequently among people who had been diagnosed with SARS-CoV-2 than among people who had not been diagnosed with SARS-CoV-2.

• One year after the first wave of infection in the spring of 2020, 44 percent of those who had been diagnosed with SARS-CoV-2 reported no symptoms. Among all those infected, 16.5 per cent reported one symptom, 10.6 per cent two symptoms, 14.7 per cent three to four symptoms and 14.2 per cent five or more symptoms.
  
  
• For example, 17.4 per cent of those infected reported experiencing fatigue ("fatigue") twelve months after infection, while the new incidence of fatigue among non-infected was 3.8 per cent. The relative risk of fatigue was 4.6 (95% confidence interval 3.2–6.3).
  
  
• Among those infected in the spring of 2020, approximately 1 in 6 reported that their sense of taste or smell had still changed one year after infection (16.9 per cent). Interestingly, prolonged change in taste or odor did not necessarily occur with other ailments.
• In total, 18.2 per cent of those who had detected SARS-CoV-2 reported reduced memory twelve months later. The corresponding figure for persons who had not been diagnosed with SARS-CoV-2 was 3.6 per cent, ie a difference in incidence (unadjusted risk difference) of 14.6 percentage points.
  
  
• The incidence of anxiety and depression did not appear to be particularly linked to being infected with the virus itself, but the risk increased with the severity of the disease and may therefore be a consequence of having experienced a serious illness and possible hospitalization

The researchers say they identified two main groups of reported symptoms linked to the airways and to the brain and nervous system. They therefore think this can be several different conditions that probably won't be possible to collect in one specific syndrome.

I'm a bit surprised to see such a paper from this institute. They're behind the GET Cochrane report and there are representatives there promoting Lightning Process and having been dismissive of both ME and Long Covid. I wonder what they make of this paper.

 

Merged thread

ABSTRACT
Physical, psychological and cognitive symptoms have been reported as post-acute sequelae for COVID-19 patients but are also common in the general, uninfected population. We aimed to calculate the excess risk and identify patterns of 22 symptoms up to 12 months after COVID-19 infection. We followed more than 70,000 participants in an ongoing cohort study, the Norwegian Mother, Father and Child Cohort Study (MoBa) during the COVID-19 pandemic. Infected and non-infected cohort participants registered presence of 22 different symptoms in March 2021. One year after the initial infection, 13 of 22 symptoms were associated with SARS-CoV-2 infection, based on relative risks between infected and uninfected subjects. For instance, 17.4% of SARS-CoV-2 infected cohort participants reported fatigue that persist 12 months after infection, compared to new occurrence of fatigue that had lasted less than 12 months in 3.8% of non-infected subjects (excess risk 13.6%). The adjusted relative risk for fatigue was 4.8 (95 % CI 3.5 to 6.7). Two main underlying factors explained 50% of the variance in the 13 symptoms. Brain fog, poor memory, dizziness, heart palpitations, and fatigue had high loadings on the first factor, while shortness-of breath and cough had high loadings on the second factor. Lack of taste and smell showed low to moderate correlation to other symptoms. Anxiety, depression and mood swings were not strongly related to COVID-19. Our results suggest that there are clusters of symptoms after COVID-19 due to different mechanisms and question whether it is meaningful to describe long COVID as one syndrome.

Full text at: https://www.medrxiv.org/content/10.1101/2021.10.15.21265038v1#disqus_thread

 

This is not a new study, just noticed it on Twitter, but it's quite interesting and I think it deserves its own thread.

A good Twitter summary is available here:
https://twitter.com/user/status/1480463466439888899

 

Possible a POTS cluster, maybe with ME/CFS in there too hidden by the lack of attention to PEM as symptom.

 

The authors used a Norwegian cohort of 70.000 people being followed up for a study on pregnancy during the pandemic. 774 (1.0%) got infected with the coronavirus. A COVID-19 diagnosis was obtained from registry data based on PCR confirmed SARS-CoV-2 infection. The researchers matched the infected with a control group that didn't get infected with corona and followed both of them up for approximately one year. 22 symptoms were assessed.

As expected, infected subjects had increased risk of 13 of the 22 symptoms when compared to uninfected subjects. Interestingly, the risk ratio for symptoms such as fatigue, poor memory and brain fog were higher than for anxiety and depression, suggesting that the former are not the result of the latter. These findings do not fit well with psychosomatic theories of long covid that have been proposed.

The authors also did a factor analysis which pointed towards the existence of two groups that many had previously speculated about. The first group seems to be people with lingering complaints following the infection. Typical symptoms here are shortness of breath, cough, reduced lung function, chest pain. The other group looks more like the long covid group that resembles ME/CFS or POTS with symptoms such as fatigue, brain fog, dizziness, heart palpitations etc.

The authors also looked at severe versus mild patients and although both were associated with increased symptoms, pretty much all of these symptoms were more frequently reported in those patients who got a severe instead of mild covid infection.

 

Twitter thread also has headaches in this list.

Edited to add:

 

ScienceNorway has an article about the study
New study on Long Covid: Brain fog, poor memory and shortness of breath one year after infection

quote:
In a preprint published on medRxiv, the researchers write that as their results suggest that different people experience clusters of symptoms, they “question whether it is meaningful to describe long COVID as one syndrome”.

“Our findings suggest that this is a about a number of different ailments that most likely cannot be seen as one syndrome, as they manifest quite differently within different people”, researcher Lill Trogstad says in a press release from the Norwegian Institute of Public Health.

She stresses that the findings must be confirmed by other studies in order to know how representative they are.

 

Also the newspaper Morgenbladet had an article about the study, with comments from among other prof. Wyller. It's paywalled, so here's a summary:

First the article introduces the findings. The data was collected when most people were unvaccinated, so next step is to examine long term symptoms in a vaccinated population.

Another Long Covid researcher, Arne Søraas, says the data from the study are in line with what they've found. He believes long term symptoms after corona infection can lead to a big public health problem.

When asked by the journalist if he's worried about creating fear and overdramatic stories in the media, one of the researchers behind the study, Per Magnus from the Norwegian Directorate of Public Health answers no, and that they have to inform researchers and the public what they've uncovered.

Then the journalist refers to previous coverage about Long Covid, and if one should be more careful when talking about long term consequences, in case fear in itself could evoke such symptoms. One who warned about fear being a cause of illness, was Vegard Bruun Bratholm Wyller, known as a researcher in chronic fatigue. He has worked for including also social and psychological factors when understanding the illness, and is now completing his own study on Long Covid.

He says that similar long term effects are well known after many other infections, and is critical to the researcher' suggestion that Long Covid might be different syndromes.

Wyller says: Studies of long term effects after infections have repeatedly shown that there is a poor correlation between patient's symptoms and physical findings. One should therefore be very careful in drawing conclusions about disease mechanisms based on symptom reporting.

 

Using an existing longitudinal cohort is a great way to do this, I hope we'll see more of it. Hopefully people ask better questions next time, missing on PEM means they had no prior knowledge of LC and so only asked the questions they could think of, missing one of the most important ones.

The weird insistence in the article above about "fear and panic" is frankly ridiculous at this point. People have been dying in droves for 2 years now and most people don't give much of a damn beyond how it inconveniences them. It shows how this nonsense stays alive long after it's been fully invalidated. Wyller being clueless, as usual. Or just selling his junk, as usual.

 

Per Magnus from the Norwegian Institute of Public Health and also one of the researchers of the study said on TV today that he believed research into Long Covid could provide insight in other conditions, like ME. We're quite behind in this country when it comes to both Long Covid and ME, but I think that statement may be a sign of some movement forward.

On the other hand two other people from the same institute retweeted an article yesterday about Lightning Process and ME, so who knows..

 

Not that it’s important for the findings, but the cohort is not part of a study on pregnacy during the pandemic especially.

It’s from MoBa, an ongoing, long-term project ment to follow up families over two generations. MoBa - mor barn (mother child). Even if fathers also are included.

It started in 1999, and is done by FHI (Norwegian institute of public health). More then 100 000 children have been part of this study since before they where born, 90 000 pregnant women and 70 000 fathers.


https://www.fhi.no/studier/moba/hva-er-moba/

English project page:
https://www.fhi.no/en/studies/moba/

 

LongCOVID never was supposed to be considered a single syndrome. If anyone thinks this, it is obvious they haven't bothered to listen to the people who invented the term in the first place.

 

Comments

For me, the most striking thing about this well-done study is that it finds two biologically coherent groups of symptoms among those with long covid:

• one a post-infectious ME/CFS and POTS-like cluster, e.g fatigue, brain fog, memory issues and palpitations.
  
• the other a chest-based cluster: e.g problems breathing and a cough.

About the study

Factor analysis
This is the key bit.

Crucially, researchers have quite a lot of scope in how they do factor analysis so that different researchers can derive different factors from the same set of data. However, in this case, the researchers followed an obvious path and reported that taking different approaches still led to the same results.

The two factors, "chest" and "post-infectious" were cleanly separated from one another (symptoms were cleanly in one factor or the other) and the factors were biologically coherent. I've seen plenty of MEcfs studies where the factors were a mishmash of disparate symptoms (or whatever data was being analysed).

In short, the analysis seems credible and delivers biologically plausible factors, so the researchers may well be onto something.

It needs replication, of course!

Weaknesses
There are a few weaknesses.
- The Cohort had a narrow age range from 40-54.
-the study only looked at unvaccinated people (because very few people had been vaccinated in the time covered by the study).
- the largest cohort, Wave 2, only had follow-are up data for 1-6 months should. That's pretty short.

Also, the factors only explain half of the variance in the data. Actually, the two factors are pretty impressive. But there will have been additional factors, perhaps 7-8, with each factor explaining less variance than the one before it. (For comparison, postinfectious factor explains 33% of variance and the chest factor 17%.)

Overall, though, the findings look very interesting. The two factors could be looked at the classic postinfectious group and a separate group we have symptoms specific to Covid.