Thread by @awgaffney

A new study at @JAMAInternalMed finds that protracted symptoms are not independently associated with evidence of prior SARS-CoV-2 infection ascertained serologically (except loss of sense of smell), but is associated with belief in having had COVID.

jamanetwork.com/journals/jamai…

Although that is the top line finding, as important to my mind was the finding that among those who believed they had COVID, there was no correlation between antibody positivity and any protracted symptom (except for loss of sense of smell).

In August, I wrote:
"The lack of difference in symptom prevalence between those with and without serological evidence of a prior SARS-CoV-2 infection in such studies suggests that for some, 'long COVID symptoms' could arise from processes other than a prior SARS-CoV-2 infection"

...in a discussion of some earlier studies. The new JAMA IM study, not without limitations, provides some more support for that assertion.

Link to my article: amjmed.com/article/S0002-…

One thing to respond to: yes, antibody tests are imperfect, and antibodies can wane. However, the lion's share of longitudinal studies have found that majority remain + for IgG over time.

E.g. below, 91% IgG+ at baseline remained positive at >=6 months:

thelancet.com/journals/lance…

In any event, none of this has any ramifications for the "realness" of the illness, suffering, or symptoms — it merely supports the proposition that, as with many chronic symptoms and illnesses, monocausality is rare. There may be numerous etiologies at work.