“I feel better, but my mind isn’t the same.” Four years after the height of the COVID-19 pandemic, such comments are still heard regularly in many medical practices in South Africa. What began as a respiratory virus seems to have left a lingering mark on some people who were infected.

In South Africa, more than 4 million cases of COVID-19 were confirmed. For some people, the physical recovery was just the beginning. Ongoing fatigue, poor concentration, and mood changes due to lasting viral effects have affected work, relationships and quality of life.

Our team of specialist psychiatrists, clinical immunologists and laboratory scientists at the University of Cape Town set out to understand why some people continue to experience fatigue, anxiety and memory loss long after recovering from COVID-19. We wanted to know whether the body’s early immune and cardiovascular responses to the virus – that is, how the body fought the virus – could help predict who might go on to develop these persistent symptoms. They are often referred to as long COVID, or a major component of long COVID.

Most studies on long COVID have come from high-income countries in Europe or North America. African populations have been underrepresented, despite clear differences in age, health status and environmental exposures that may influence both the course of infection and recovery. We felt that this was an important gap: the neuropsychiatric effects of long COVID in an African context. Also, understanding whether the same biological risk factors apply is crucial for designing appropriate health services and ensuring equitable research representation. Local data matters as it demands policy makers take the problem seriously.

Our recent research revealed an alarming picture. More than half of the participants in our study group of people in Cape Town who had been infected with the coronavirus (and mostly had been in hospital with COVID-19) had at least one neuropsychiatric symptom more than six months after infection. The symptoms included fatigue, concentration or memory difficulties. Many still had these symptoms up to two years later.

Unfortunately, none of the blood biomarkers we measured during acute infection, including those linked to inflammation, cardiovascular stress and the entry system for the SARS-CoV-2 virus, the renin-angiotensin system, could predict who would develop long-term cognitive or mental health problems. Biomarkers are biological signs of what’s happening in the body.

Read more: We recorded acts of kindness as COVID-19 unfolded in South Africa: What we learnt

What we did

We followed 97 people in Cape Town who had tested positive for SARS-CoV-2 during the first three waves of infection, before vaccines were widely available. Most were hospitalised with moderate to severe disease, while others had mild or even asymptomatic infections. The first three waves were June to August 2020, November 2020 to February 2021, and May to September 2021.

Blood samples were collected during the peak of illness and again several months later (between 6 and 24 months). Using advanced protein profiling, we measured 96 molecules linked to inflammation, cardiovascular stress and the renin–angiotensin system, a network of hormones that regulates blood pressure and has been implicated in COVID-19’s effects on the body.

At least six months after infection (and up to 24 months for some), participants completed telephone interviews using validated questionnaires to assess anxiety, fatigue and cognitive function. These included the Telephonic Montreal Cognitive Assessment, a screening tool for memory and attention problems. While we did not interview a control group, we know the background rates for age.

Our findings

The results were striking:

Persistent symptoms were common: More than half had at least one ongoing symptom more than six months after infection.

Memory and thinking problems: 44% showed measurable cognitive or memory impairment, while 53% reported memory problems.

Fatigue: 55% reported moderate to severe fatigue.

Anxiety and distress: One in four had high anxiety levels, and the number taking psychiatric medication nearly doubled after infection.

Severity didn’t predict outcome: These problems affected people across all levels of illness, including those who were never hospitalised.

No predictive biomarkers: None of the inflammatory, cardiovascular, or renin–angiotensin markers measured during acute illness or recovery were linked to persistent symptoms. In short, the usual blood tests taken during infection didn’t help identify who would go on to develop long COVID symptoms.

All these reported symptoms were new to the individuals and only came after COVID.

The lack of a clear biological predictor suggests that long COVID’s mental and cognitive effects aren’t easily explained by inflammation alone. Some scientists propose that the virus may directly injure brain cells or persist in the nervous system. Others believe that subtle, ongoing immune activation in the brain, known as microglial activation, could be involved.

In South Africa, there’s also another layer: the social and economic aftermath of the pandemic. Unemployment, inequality and limited access to healthcare have magnified stress and vulnerability. Research during lockdowns showed increased rates of domestic violence and psychological distress, particularly among people of lower socioeconomic status. These factors likely amplify the mental toll of COVID-19 and its long-term effects.

Read more: Jobs in South Africa: the labour market is recovering from COVID – but unskilled and less educated people are still being left behind

How our findings fit into the global picture

Our results align with international evidence showing high rates of cognitive and emotional symptoms months after infection. A global review estimated that around 20% of people experience some form of persistent fatigue, depression, anxiety, or memory deficit six months after COVID-19 infection.

However, ours is one of the first studies from Africa to combine early serum biomarker profiling with long-term mental and cognitive follow-up. Despite demographic and healthcare differences, we found the same pattern seen elsewhere: a high burden of long-term symptoms, but no clear laboratory signature to predict them.

This highlights the need to look beyond blood tests, to the broader biopsychosocial picture. This is a holistic approach to understanding health that considers the interaction of biological, psychological and social factors. Long COVID isn’t purely a biological problem; it’s a human one, shaped by both physiology and lived experience.

What should be done

Our study underscores the need for an integrated approach to long COVID care:

  • Recognise long COVID as real and disabling. Fatigue, “brain fog” and anxiety may not show up on lab results but can profoundly affect daily life.

  • Screen in primary care. Simple questionnaires can help clinicians identify people who need mental health or cognitive support.

  • Expand multidisciplinary care. Collaboration between doctors, psychiatrists and occupational therapists can help patients rebuild cognitive and emotional resilience.

  • Invest in follow-up services. Dedicated long COVID or rehabilitation clinics could improve recovery outcomes.

  • Sustain prevention. Vaccination and infection control remain the best defence against long COVID, and a notable limitation of our study is that infections occurred prior to the availability of widespread vaccination. Other studies have found a beneficial effect of vaccination in reducing the prevalence of long COVID at the population level.

For many, recovery from COVID-19 isn’t about surviving the virus; it’s about reclaiming normal life.

Participants in our study often described feeling “not themselves” months later: forgetting words mid-sentence, losing focus at work, or feeling constantly tired. These are not minor inconveniences; they affect livelihoods and self-esteem.

Listening to these stories reminded us that long COVID is not just a medical condition. It’s a lived experience that deserves empathy, research, and sustained public health attention.

This article is republished from The Conversation, a nonprofit, independent news organization bringing you facts and trustworthy analysis to help you make sense of our complex world. It was written by: Jonny Peter, University of Cape Town

Read more:

Jonny Peter received funding from the South African Medical Research Council for the research work described in this article.