Cognitive Impairment After COVID-19: What the Research Really Shows

Many media outlets have reported on a study published in a journal from The Lancet group, one of the most respected sources of scientific medical information. The topic is especially relevant: researchers observed a significant decline in cognitive abilities after COVID-19 infection, based on a large sample of 81,337 people. However, the dramatic claims about the irreversibility and severity of these changes, which have filled news headlines, are likely exaggerated. The comparisons made by the researchers also have certain limitations, and the authors themselves do not draw definitive or harsh conclusions. Let’s break down what the study actually found and what may be the result of journalistic hype.

How the Study Was Conducted

Enough time has passed since the start of the pandemic to allow for a comprehensive assessment of the long-term effects of COVID-19 on the human body. The most statistically powerful studies in this area are so-called longitudinal, prospective studies, where participants are observed over time and their condition is periodically assessed to allow for proper “before and after” comparisons. However, these studies are very resource-intensive, especially if researchers aim to collect a large, population-wide sample.

To partially address this, researchers often use statistical approaches such as regression models. These models, based on available population data, can predict the state of a dependent variable (for example, cognitive ability) based on one or more independent variables (such as socioeconomic status, gender, age, etc.).

British researchers set out to study the impact of “Long COVID”—the lingering effects of the disease—on cognitive abilities as broadly as possible. This research was needed as reports accumulated that COVID-19 could lead to a subacute or chronic phase, causing a range of psychological and neurological issues for months after recovery.

From January to December 2020, scientists conducted a nationwide study, inviting participants to complete a large questionnaire with nine cognitive tests and a personal data survey, which included questions about COVID-19 infection, symptom severity, hospitalization, and whether they required mechanical ventilation. To minimize participant bias, they were not told the purpose of the testing. The invitation, including one aired on the popular BBC2 documentary series “Horizon,” simply offered a free “Great British Intelligence Test” to discover their strongest cognitive abilities.

Over 12 months, 86,285 responses were collected, with 81,337 included in the final analysis. The sample was predominantly British (93%), with an average age of 46.75 years. In the survey, 12,689 people reported suspected COVID-19 infection of varying severity. For 386 people, infection was confirmed by a biological test, including 86% of those hospitalized and placed on mechanical ventilation.

What the Tests Measured

The cognitive tests assessed a wide range of abilities, from basic (like working memory) to more complex functions such as reasoning, multi-component analysis, spatial planning, and others. These tests are not equivalent to an IQ test, which only partially reflects intelligence and mainly measures the ability to solve IQ test problems.

The authors note that the observed cognitive deficit was significant, especially for those who required mechanical ventilation. Their overall score dropped by 0.47, which is greater than the average 10-year decline in cognitive function between ages 20 and 70 in this population. This figure was even higher than for people who had suffered a stroke (480 people; 0.24) or had learning disabilities (998 people; 0.38).

In classic IQ testing, a 0.47 drop is roughly equivalent to a 7-point difference in IQ. Is this difference statistically significant? Yes, definitely. Does it have a major impact on a person’s cognitive abilities? Probably not. The most affected were complex cognitive functions, consistent with previous studies on post-COVID “brain fog,” which makes it hard for people to concentrate, find the right words, or build logical chains of thought.

Key Findings and Limitations

Cognitive deficit in people with suspected and confirmed COVID-19:

• People who reported recovering from COVID-19 performed worse on cognitive tests. The deficit increased with the intensity of treatment received for breathing difficulties.
• Among those who did not receive medical care, the deficit was more pronounced in those with a confirmed infection than in those who only suspected they had COVID-19.

Importantly, the authors do not claim these changes are irreversible. They found that the severity of the deficit did not depend on the time since infection, age, gender, handedness, native language, education level, country of residence, professional status, or income. The main factors were the severity of respiratory failure, hospitalization, and mechanical ventilation. Even those who were ill at home experienced a notable cognitive decline—greater than the average for their age group (0.23 overall score).

One limitation was that only 275 respondents had “before and after” data, as an additional questionnaire was introduced in December and some people reported getting sick after their initial participation. For the rest, “before” data were estimated using a regression model (generalized linear modeling, GLM) trained on a sample of 269,264 people. No differences were found between intelligence “before” and “after,” supporting the validity of the findings from the 275 real cases.

The authors themselves discuss the limitations of their work. The cross-sectional nature of the study allowed for the inclusion of many potentially confounding variables. Future prospective studies are needed to confirm (or refute) the impact of COVID-19 on cognitive health and to determine how long the deficit lasts depending on the severity of respiratory symptoms.

Additionally, the results were based on self-reports, meaning the authors did not have access to clinical information for all participants. The study also did not aim to determine the biological basis of COVID-19-related cognitive deficits in terms of neural systems or psychological mechanisms. Further research, including neuroimaging data, is needed.