View commentaries on this research

This is a plain English summary of an original research article

Only 1 in 3 people were fully recovered from COVID-19 a year after they left hospital. Being female, having obesity or being on a ventilator were each linked with ongoing symptoms, months after people were discharged (long-COVID).

Nearly 1 million people in the UK have so far been admitted to hospital because of COVID-19. The long-term effects of the infection in this group are only just coming to light.

Researchers explored the impact of being hospitalised for COVID-19 on people’s mental and physical health, and on their employment. They looked at characteristics such as age and sex to see which were associated with worse recovery. They also assessed whether inflammation in the blood may be a potential target for treatment.

This study described, for the first time, 4 different patterns of COVID-19 recovery. It found, for example, that some people with long-COVID had higher levels of inflammation. The researchers say that targeting treatment to specific clinical problems, such as treating inflammation in people with higher levels of inflammation, are promising approaches to aid recovery.

More information about long-COVID is available on the NHS website.

What’s the issue?

Previous research has shown that many people who receive critical care in hospital for serious illnesses have prolonged physical and mental ill health. This can last for years after they are discharged.

As of September 2022, more than 990,000 people in the UK had been admitted to hospital because of COVID-19. Many have reported long-term effects. This research therefore followed-up this group of people. Researchers assessed people’s long-term health (at 5 months and 1 year) after they were discharged from hospital. They looked at people’s physical and mental health, their employment, and took blood samples.

The researchers analysed patterns of recovery and whether factors such as age or sex made a difference. They also measured levels of inflammation in the blood of people who had been hospitalised for COVID-19.

What’s new?

The study was carried out at 83 NHS hospitals across the UK. Participants were at least 18 years old (around 59 years on average) and had all been discharged from hospital after short-term treatment for COVID-19. Most (64%) were male and the majority (around 70%) were White.

The researchers described people’s health 5 months after they were discharged from hospital, and again at 1 year. People filled out questionnaires about their recovery and remaining physical and mental symptoms, their quality of life and any changes to their occupation. Researchers assessed physical function, breathing tests, and blood tests.

1,077 people discharged March - November 2020 were assessed at 5 months and the researchers published their findings in a first paper. The researchers continued to assess people being discharged until April 2021. In a second paper, they report their findings on a total of 2,320 people assessed at both 5 and 12 months after they were discharged. The team took blood samples to measure levels of inflammation in this group.

The study found that:

  • few people felt fully recovered at 5 months (26%) and 12 months (29%)
  • anxiety did not reduce over time (24% people were feeling anxious at 5 months; 22% at 12 months) and nor did depression (27% people were feeling depressed at 5 months; 25% at 12 months)
  • 1 in 5 (19%) people changed their job or stopped working because of their health at 5 months (this was not reported at 1 year)
  • 1 in 5 (20%) had developed a new disability at 5 months.

Being female and having obesity were linked with ongoing problems at 1 year. People who had been on a ventilator in hospital (to aid breathing) were also more likely to have long-term problems. The most common symptoms at 1 year were fatigue, aching muscles, physically slowing down, poor sleep and breathlessness.

The research team found that the severity of physical and mental health symptoms were closely linked, but severe cognitive impairment (brain fog) was present in people with moderate other symptoms. They grouped people according to these symptoms into 4 categories: very severe, severe, moderate plus cognitive impairment (of memory and attention), and mild.

  • People with the most severe symptoms at discharge had more chance of having a new disability and of stopping working or changing their job at 5 months.
  • Those with very severe or severe symptoms were more likely to have obesity, reduced exercise capacity, more symptoms and more whole-body inflammation (higher levels of a marker called serum C-reactive protein).
  • Those in the very severe group and those in the moderate plus cognitive impairment group had increased levels of an inflammatory marker called interleukin-6 not seen in the other groups.

Why is this important?

This was the first large study to look at the impact of COVID-19 hospitalisation on longer-term health. It found that few people (less than a third) admitted to hospital with COVID-19 had fully recovered 5 months after they were discharged. Those who had not recovered at 5 months were unlikely to do so by 1 year.

The study shows that people discharged from hospital after COVID-19 need careful assessment of their symptoms and proactive care. The different patterns of recovery suggest that different groups might need different care and treatments. For example, people with mild symptoms when they are discharged might need less intense follow-up, and those with persistent inflammation might benefit from anti-inflammatory treatments.

Almost 1 in 5 people in the study stopped working or changed their job because of their health. This highlights the impact that long-COVID could have on the economy.

What’s next?

More research is needed to see if different processes are causing the 4 different groups of symptoms. The researchers would also like to test whether anti-inflammatory drugs can help people with high levels of inflammation feel better.

The current study included a diverse population. Further research could explore whether patterns of recovery vary in people of different ethnicities. It should also be repeated in people who have been vaccinated and those who have not been in hospital.

Understanding the time it takes for people to recover will help the NHS plan its provision of long-COVID clinics. The researchers suggest that, to reduce health inequalities, the NHS should reach out to people who were hospitalised and are not fully recovered, rather than waiting for them to seek help. Further research could explore whether the number of ongoing symptoms after COVID-19 could identify people in greatest need of support.

You may be interested to read

This Alert is based on:

• 1 year follow-up: Evans RA, on behalf of the PHOSP-COVID Collaborative Group. Clinical characteristics with inflammation profiling of long COVID and association with 1-year recovery following hospitalisation in the UK: a prospective observational study. Lancet Respiratory Medicine 2022;10:8

• 5 month follow-up: Evans RA, on behalf of the PHOSP-COVID Collaborative Group. Physical, cognitive, and mental health impacts of COVID-19 after hospitalisation (PHOSP-COVID): a UK multicentre, prospective cohort study. Lancet Respiratory Medicine 2021;9:11

Information from the charity, Asthma + Lung UK.

Information on support for long-COVID from the charity Long Covid Support.

Our NIHR research collection on 'Researching long-COVID: addressing a new global health challenge'.

Funding: This study was funded by the UK Research and Innovation (UKRI) and the NIHR.

Conflicts of Interest: One member of the research group has received unrestricted funding from a technology company. One author received fees and funding from a pharmaceutical company.

Disclaimer: NIHR Alerts are not a substitute for professional medical advice. They provide information about research which is funded or supported by the NIHR. Please note that views expressed in NIHR Alerts are those of the author(s) and reviewer(s) at the time of publication. They do not necessarily reflect the views of the NHS, the NIHR or the Department of Health and Social Care.

  • Share via:
  • Print article


Study author

This research needs to be put into context – we performed the study before anyone was vaccinated, and people had older viral variants. Although vaccination reduces the chances of developing long-COVID, data from the Office of National Statistics have shown that even people who are triple vaccinated can have ongoing problems at 12 weeks.

We were surprised and disappointed that so few people who had been in hospital with COVID-19 were fully recovered by 1 year. Most people who hadn’t recovered at 5 months remained not recovered by 1 year.

The study has given insights into the different types of long-COVID, and potential therapeutics and strategies that might benefit people with this long-lasting condition.

Rachael Evans, Associate Professor (Clinical) and Honorary Consultant Respiratory Physician, University of Leicester

Long Covid Support

This large study describes the long-term impacts of COVID-19 on hospitalised patients, even for those not receiving intensive care. The clustering of patient types, and indicators such as inflammatory markers are likely to be similar in non-hospitalised patients, so this study has wider validity. In our support group, as in published literature, we also see persistent symptoms of a broadly similar nature in non-hospitalised patients. While a lower percentage of those not hospitalised have long-term negative outcomes, the absolute number affected is vastly higher with the Office of National Statistics estimating 2 million in the UK with long-COVID.

We support the call for trials of anti-inflammatory treatments. As this study reinforces the growing body of literature that being female is a risk factor, we also call for trials of hormone treatments. While obesity was found to be a risk factor, most people with long-COVID are not obese. Moreover, long-COVID causes exercise intolerance. We, therefore, urge caution in promoting weight loss as a widely applicable strategy.

That those unrecovered at 5 months are still unrecovered by 12 months highlights the need for early intervention to prevent long-term disability and loss of workforce capacity. We urgently need more research into the underlying disease processes and treatment for long-COVID.

Margaret O’Hara, Founding Trustee, Long Covid Support

Lung Physician

This paper describes patients hospitalised in the first wave of COVID-19 and is a timely reminder of how devastating the first 6 months of the pandemic were. Both patients and members of the healthcare community struggled, not just in the acute phase, but in the months following recovery.

The variant(s) circulating now have different in-hospital and recovery characteristics. This may make it difficult to generalise the results of this study to future healthcare planning. For example, we currently see few patients in hospital in the high-severity recovery groups that this paper describes.

Having said this, a key take-home message is that recovery from COVID-19 is variable, with several distinct characteristics, and this persists regardless of the severity of the initial illness. We therefore need a patient-centred individualised data-driven approach to recovery.

Tom Bewick, Clinician, University Hospitals of Derby and Burton

This reviewer works at University Hospitals of Derby and Burton NHS Foundation Trust, which participated in this trial, but he was not directly involved.

Nurse with Lived Experience

As a person with long-COVID, this research validates my experience and gives me more confidence when communicating with healthcare professionals. As a nurse, it gives me evidence-based knowledge, which I can use to educate patients or to influence service design and provision.

This paper contributes to a growing evidence base that a COVID-19 infection can leave many people with long-term, multisystem complications. Care after COVID-19, especially in long-COVID clinics, should be multidisciplinary and adopt a holistic approach.

The high levels of anxiety, depression and post-traumatic stress disorder demonstrate the need for mental health support. The research suggests that mental health problems might be related to inflammatory changes in the brain, and trauma from hospitalisation. Currently, some people are being diagnosed with anxiety and depression and, after that, their physical symptoms are attributed to mental health conditions. Some are then discharged with neither physical care nor mental health support. I hope this paper will lead to better understanding of the post-COVID syndrome.

I am worried that the exclusion of non-hospitalised patients from the study could lead to inequality in accessing services. Many people were very unwell at home, and they had the same illness. Access to ongoing post-COVID care should be based on symptoms and individual needs, not on whether a patient was hospitalised.

Nora Dimitrova, Public Contributor and Nurse, University Hospital Southampton, Research and Development

Back to top