How big is the risk of developing Long Covid after a corona infection? According to researchers, the frequency of this very poorly defined secondary disease is probably overestimated. The reason for this is methodologically deficient studies that often do not have a suitable control group and, moreover, often include very different clinical pictures that do not necessarily have a causal connection to Covid-19. The consequences of a possible misjudgment can be unjustified fears among the population and an inappropriate distribution of research funds.
The Long Covid clinical picture includes a variety of symptoms, from changes in the sense of taste and smell to sleep disorders and shortness of breath to general fatigue. Several scientific associations have now formulated working definitions of the disease. However, these definitions are usually very broad and include numerous clinical pictures that can also occur independently of Covid-19. None of the previous definitions requires a causal connection to a previous infection.
“Due to overly broad definitions, missing or unsuitable control groups and other methodological weaknesses, many scientific publications overestimate the frequency of Long Covid,” writes a team led by Tracy Høeg from the University of California in San Francisco. “For example, the US Centers for Disease Control and Prevention (CDC) recently announced that almost one in five American adults suffers from Long Covid after a Covid 19 infection.” Høeg and her team believe these numbers are significantly too high.
“According to all previous definitions, any symptom that occurs after a confirmed or suspected Covid-19 infection can be described as Long Covid, regardless of its cause,” criticize the researchers. “This broad definition has led to over 200 symptoms being associated with Long Covid in the scientific literature.” However, many of these symptoms are not specific to Covid-19, but also occur after other viral diseases, for example after a severe one Pneumonia or after treatment in the intensive care unit for other reasons.
According to Høeg and her team, most of the studies published so far do not allow any conclusions to be drawn about how many of the cases described would have occurred without Covid-19. This would require suitable control groups. Only then can it be compared whether corresponding symptoms occur more frequently in people who have had Covid-19 than in people who are of a similar age and have similar previous illnesses but have not yet been infected with Covid-19. “According to a large review, of 294 studies evaluated, only 22 had a control group,” write Høeg and her team. The studies with a control group showed that symptoms typically associated with Long Covid occurred only slightly more frequently in the group of infected people than in the control group.
Another problem is the so-called sampling bias: “In the early phase of the pandemic, when tests were not yet widespread, people with severe symptoms were more likely to be included in studies, while infected people with mild symptoms were underrepresented,” explain the researchers. There may also have been biases in later phases because people who suffer from persistent symptoms are more motivated to take part in long-Covid studies than people whose symptoms quickly disappear. It should also be noted that many studies come from times when the Delta variant was dominant. However, according to previous findings, this led to long-term effects more often than the omicron variant.
Negative consequences for society and science
“Misestimating the frequency of Long Covid can fuel unnecessary fears in the population and lead to funds for research and healthcare being misdirected,” write Høeg and her team. “There is also a risk that other diseases that could actually be treated will be misdiagnosed as Long Covid. And those who actually suffer from chronic consequences of Covid-19 can also receive less help due to the inadequate differentiation.”
Both in research and in clinical practice, it is therefore important to make the diagnosis of Long Covid only after other causes have been ruled out. The authors also call for clearer definitions and suggest describing specific symptoms instead of the general term Long Covid, for example “loss of smell after Covid-19”. In studies, attention should be paid to a representative selection of cases and controls and the physical and mental health status before the infection should also be taken into account.
Take it seriously, but don't overestimate it
Clara Lehmann from the Post-Covid Outpatient Clinic at the University Hospital of Cologne, who was not involved in the publication, also agrees with this. “It is extremely important to restore an objective basis and not allow ourselves to be guided by emotions, political pressures or personal suffering,” she says. “I would like to emphasize at this point that long or post-Covid is not a pipe dream. In some patients we see severe inflammatory reactions even several weeks after the infection. But this does not apply to the majority of patients. We need to take this condition seriously, but this requires an honest scientific determination of the true risk of the disease.”
Source: Tracy Høeg (University of California San Francisco, USA) et al., BMJ Evidence-Based Medicine, doi: 10.1136/bmjebm-2023-112338