One of the lingering mysteries of the pandemic is Long Covid, the term used for a range of symptoms, from shortness of breath and fatigue to cognitive deficits, that persists many weeks or months after someone has been infected with SARS-CoV-2. With research suggesting that up to 30% of people who have had Covid-19 are impacted by Long Covid, governments have dedicated special clinics and major research funding for the condition. 

The subject has also been divisive, with social media full of both support groups and skeptics. A recent French study showing some correlation between a person’s belief they had Covid-19 and their reporting of Long Covid symptoms was used by skeptics to claim that the condition doesn’t exist (a conclusion the study doesn’t support). Another new study, however, suggests that vaccinations reduce the risk of Long Covid, even among those who get jabbed after they’ve had Covid-19.

Bloomberg Opinion’s Therese Raphael spoke to Bloomberg Intelligence senior pharmaceutical analyst Sam Fazeli about the study and how we should be thinking about Long Covid now.

Therese Raphael: When the vaccines were first being rolled out, some people with Long Covid — long haulers, as they are often called in the U.S. — reported their symptoms worsening after vaccination. Has this recent study, posted on medRxiv last week and not yet peer reviewed, put some of these concerns to rest?

Sam Fazeli: This is a welcome study in an area where there has been disagreement, with at least one paper showing no impact from vaccination on risk of Long Covid. With this study, the body of evidence moves in favor of vaccination lowering, but not eliminating, the risk of Long Covid.

There are two key observations from it. First, they found a seven-to-10-fold reduction in the risk of Long Covid in infected people who had received one dose of any of the three vaccines authorized in the U.S. (from Moderna Inc., Pfizer Inc. or Johnson & Johnson). The authors defined Long Covid as two or more ongoing symptoms. The effect fell to 4.5 times if they defined Long Covid as “any symptom.”

The second interesting observation was that vaccinating individuals even after they got Covid led to a reduction — by 1.3 to 5.3 times — in the risk of getting Long Covid, again depending on the definition of the condition and the interval between infection and vaccination.

Like all such studies that use retrospective analysis, however, there are important caveats. The authors list at least six issues that could have affected the data. We should also note that the size of the group that received a vaccine dose (2,392) before infection was much smaller than the unvaccinated group (220,460), making it tough to control for all variables that could have impacted the outcome, such as an imbalance in the ethnic background of the subjects. As with some other studies, the sample contained more women than men, which could have impacted the findings, too.

TR: An interesting finding is that those who were vaccinated even after being diagnosed with Covid-19 were three times less likely to report multiple Long Covid symptoms compared to the unvaccinated. Why might that be the case?

SF:  There are two theories. One is that there is a “hidden reservoir” of the virus that is causing the chronic issues and that vaccination elicits an immune response that clears the infection. Another, less likely, theory is that by inducing a specific immune response, the vaccine dampens an immune reaction that is responsible for some of the symptoms. The data from this study showed a clear correlation between the time after the infection that a vaccine dose was administered and the impact on Long Covid symptoms, providing more support for the hidden reservoir hypothesis.

TR: The current recommendation from the U.S. Centers for Disease Control is that those who are diagnosed with Covid-19 defer being vaccinated until after they have recovered from acute illness. Do the findings of this study suggest we ought to reconsider that guidance and administer vaccines even sooner after infection?

SF: The best effect of the post-infection vaccination in this study was seen if the dose was given within zero to four weeks after infection and the least impact was in those vaccinated eight to 12 weeks after an infection. But the group that had their vaccine soon after their infection had already mounted a strong enough immune reaction and may have had milder disease. So are we really seeing an effect of the vaccine or is there something about people who resolve their disease faster, or both?

TR: As you mentioned, this study has limitations. One is that it uses data collected before the emergence of the delta variant. We’re obviously living in the delta era. How does that make you look at the results?

SF: The key difference between delta and alpha, the previous dominant variant of SARS-CoV-2, is its higher infectiousness and viral load. If Long Covid is indeed due to a hidden reservoir of virus, then the delta variant may have a better chance of establishing such a hideout.

Also, the study did not look at the impact of time between vaccination and infection. It’s possible that an infection several months after vaccination, when antibody levels are lower, could have a better chance of causing Long Covid than one closer to the vaccination date.

TR: The U.S. cohort would have had one of the three FDA-approved vaccines and not the AstraZeneca Plc shot that many in Britain have had, but the study doesn’t seem to have differentiated between them. Might the particular vaccine given be a factor here?

SF: It’s possible. This is another shortcoming of the study, but frankly, given the relatively small number of subjects vaccinated prior to a diagnosis, they would not have been able to draw many conclusions anyhow.

On the one hand, the latest U.K. data show that you have a higher risk of infection if you were vaccinated with the AstraZeneca jab than with the Pfizer-BioNTech one. This could translate to a higher risk of Long Covid for those vaccinated with Astra’s vaccine. On the other hand, if claims that Astra’s vaccine induces a stronger cellular immune response — a T-Cell response — are true, then the risk of Long Covid may be reduced, especially if it’s caused by pockets of viral infection in the body. But I have yet to see a rigorously conducted study that shows one vaccine is better at inducing T-cells than another or that the different levels have any clinical impact.

TR: Presumably if two shots help keep Long Covid at bay, a booster shot will be an added plus. Are we looking at the end of Long Covid, possibly — at least for those who don’t already have it?

SF: Remember that this study required people to have had at least one vaccine shot before catching Covid. We actually don’t know what percentage had their full vaccination schedule. If they had focused on people who’d had both doses, the effect may have been even better. And, yes, theoretically, a third shot should have a significant impact, if only by reducing the risk of infection.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Therese Raphael is a columnist for Bloomberg Opinion. She was editorial page editor of the Wall Street Journal Europe.

Sam Fazeli is senior pharmaceuticals analyst for Bloomberg Intelligence and director of research for EMEA.

More stories like this are available on

©2021 Bloomberg L.P.