Search results

The literature search identified 269 articles. A further two studies were identified from reference lists. After removal of duplicates and title/abstract screening, seven articles were evaluated as full text. The main reasons for excluding articles were no qualitative element to the research, no patient involvement and not meeting our definition of long COVID (we were interested in studies relating to symptoms over 4 weeks’ duration). Out of the seven fully evaluated articles, one study was excluded because it did not use qualitative methods or contain data on direct patient experience. A second study that was initially included was later excluded after it was withdrawn from prepublication by the authors. A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram depicting the study selection process is provided in figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

PRISMA flow diagram of study selection. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Characteristics of included studies

Five studies were included in the thematic synthesis (table 1).1 16–19 Three studies conducted focus groups or interviews with patients from the UK and two studies, from the Patient-Led Research group, conducted international surveys with most responses coming from the USA and the UK. Sample sizes varied from 24 interviews to 3762 survey respondents and were generally weighted towards white (83.8%), female participants (75%). The number of patients included in the studies in which information was gathered through surveys was much larger than those using interviews and focus groups as data collection methods. However, while representing fewer patients, the latter method offers the opportunity of collecting more in-depth data and for interaction among participants and/or with the interviewer. All studies focused on adults with an age range of 20–68 years in the four studies that reported participants’ ages; one study did not report the number of participants or their ages.1

Methodological quality

Studies were of variable methodological quality. Three met most of the criteria on the CASP checklist (table 2) and thus were considered of high quality, and two met fewer criteria. No studies were excluded on the basis of quality as all were considered to offer valuable content despite the limitations identified.

All five studies recruited participants through social media and/or online support groups. While this is understandable given the need to quickly access participants for whom no established groups or organisations existed, this convenience sampling may have resulted in bias.20 People who are active on social media or online support groups are likely to differ from the general population (eg, younger age) and may be more vocal about their experiences. Three included studies acknowledged skewed sample characteristics including mainly white ethnicity, over-representation of women and a generally younger age group.16 18 19 Limited demographic information was provided on participants, particularly in Maxwell,1 making it difficult to determine which population groups may have been missed by these studies.

None of the studies discussed potential biases arising from the relationship between researchers and study participants. This is despite people with lived experience of long COVID symptoms being among the study authors, or performing data analysis in some studies.16 17 19 This participatory research approach can be considered to represent both a strength and a weakness. Having authors and researchers with experience of long COVID analyse data is beneficial in bringing lived experience to the interpretation of data. However, it may also introduce bias for the same reason.

Several other quality issues were noted. In the study by Kingstone et al,17 participants received a compensation voucher for their time, which may have influenced decisions on whether to participate. Ladds et al18 only fully transcribed the first 10 out of the 55 interviews (the remaining interviews were partially transcribed). This was due to the urgency of the work and limited resources plus a perceived lack of need to duplicate previously discovered themes. This may have introduced bias. Finally, Maxwell1 reported very limited methodological details, making it difficult to determine how the research was conducted or the number of people involved in the focus group.

Review findings

The initial stages of thematic analysis resulted in the generation of 138 descriptive themes. These were then refined into 54 subthemes, which were attributed to 11 higher order themes using an iterative process, with continuous discussion between reviewers. Further review and refinement of themes resulted in three overarching analytical themes: (1) symptoms and self-directed management of long COVID; (2) emotional aspects of living with long COVID; and (3) healthcare experiences associated with long COVID. Figure 2 shows the relationship between the final three themes and the initial 11 higher order themes. Full details of descriptive themes and subthemes are available in online supplemental file 3.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Map of analytical and descriptive themes from the analysis.

Symptoms and self-directed management of long COVID

Evidence from all the included studies1 16–19 showed that people with long COVID experience a wider range of symptoms than the three symptoms officially recognised as acute COVID-19: high temperature, new continuous cough and change or loss of sense of smell or taste. One individual stated:

From week four I started to get chest pains and then breathlessness, gradually other symptoms developed including dry mouth, sore tongue, joint pains, fatigue, rash and tachycardia.1

The symptoms experienced by patients with long COVID varied in severity from relatively mild to potentially life-threatening symptoms that required hospital admission.16–19 Symptoms also fluctuated over time with new symptoms appearing at different stages of the illness and in different parts of the body.1 17–19 Each symptom was experienced for a prolonged but variable length of time, with a cumulative effect in many cases.1 16 18

People identified a disconnect between their lived experiences, official advice and public perception of the illness. It was felt that the public perceived the illness as a binary condition1 17—either mild and easily treated at home or serious and requiring hospitalisation—with no variation or allowances made for ongoing symptoms.

So, COVID-19, it’s either a mild infection or you die? No. But no one is prepared to think about us.17

The literature showed that people believed they would require a short recovery period and would be back at work in 2 weeks, a belief mirrored by employers and the public.1 16–19 The lived experience, for some, was quite different:

After nearly 6 months I have started to feel some improvement, although doing anything remotely physical results in a flare up of symptoms…1

I had to take two weeks off, had to work from home for four, but had to return for two weeks with fever as my employer would not give me more time […].16

This discordance between expectations and experience seemed to have a direct effect on the mental and emotional state of those experiencing prolonged illness,1 18 19 often leading to uncertainty about what to do about their symptoms.1 17 18 People described needing to adjust their lifestyle, including pacing themselves and setting realistic goals, in order to self-manage their symptoms.1 17 18 One study highlighted specific methods used by a number of patients attempting to self-care, such as taking supplements or trying therapeutic massage.17

Many people turned to social media and support groups (online or face to face) for support and found them to be a valuable way to share experiences, knowledge and resources with others in a similar situation.17–19 This communication helped to validate patient experiences and provided reassurance they were not alone in their struggle with long-term symptoms.

At least I know I'm not alone. And I think people who actually have had the disease tend to know a little bit more about it… I actually think that the support group has given more knowledge than the doctors have.18

However, there were also reports of stigma, anxiety and depression17 19 triggered by knowledge garnered from these online groups.

… Internet support groups, yeah on the Facebook groups that I'm on, I mean to be honest, I try not to read that group too much because it depresses me, makes me a bit anxious.17

Emotional aspects of living with long COVID

For many patients, there was a feeling that their self-identity was affected by long COVID. People reported an impact on how they viewed themselves, before and after their illness.16 18 There was a feeling they had to reconsider who they were and what they could do within the context of family and work.16–18 The phrase ‘compared with how I used to be’ was used by multiple participants in Kingstone et al’s study,17 while Ladds et al18 commented on the concept of a ‘spoiled identity’ where an identity as previously ‘healthy, independent and successful’ was perceived to be threatened.

Interviews by Ladds et al18 with doctors and other clinicians who had experienced long COVID showed that many were worried about the impact of cognitive deficits on their ability to perform their jobs.

[T]he medicolegal aspect is huge… and it’s scary to not be able to recognise potentially where you have deficits because if you can’t recognise them then that’s an unknown unknown in what can you do with that.18

There was a sense of stigma associated with long COVID, with people experiencing a sense of shame and blame (internally generated stigma) and expressing fears that employers and others in the community may stigmatise them for having long COVID (externally generated stigma).1 18 19 Family members were considered to be affected by long COVID and were seen as also requiring support.1 17 One interview participant described the impact her symptoms had on her family and how she felt they did not believe her:

I think, at first, they just thought, ‘Oh, for god’s sake, she’s napping again’. I feel like I constantly have to explain. I’m just exhausted and I just want to know why I’m so exhausted …17

Patients described experiencing a range of emotions as part of their illness journey.1 16–18 Anxiety was often related to multiple aspects of the illness including uncertainty about the cause of symptoms, concern that they may never recover completely and anxiety due to not being believed by healthcare professionals, family and friends.

… I was really frightened, terrified and just thought I might die on a couple of occasions … maybe not ‘I’m going to die right now’, but definitely ‘I’m never going to get better from this’ kind of feeling.17

Patients also expressed a strong desire to find acceptance and understanding about their experiences of long COVID, both among healthcare professionals and family and friends.

… one of my friends did say after quite a while, ‘I’m not being awful, but do you think a lot of it’s in his mind?’ and I said ‘no’. I was quite upset about that…17

Similarly, there was a widespread perception that healthcare professionals doubted patients’ descriptions of long COVID,1 ignored patient concerns,17 misdiagnosed symptoms19 or were dismissive of patient experiences.19 This lack of knowledge affected people’s feelings around their healthcare experiences.17

Healthcare experiences

Across most of the studies, participants expressed concerns relating to the lack of knowledge, information and understanding about long COVID among healthcare professionals.1 17–19 While the reason behind this lack of knowledge was understood, there was a general feeling that there needed to be acknowledgement of this gap within the healthcare community.

Well yeah, I feel like there’s a lack of knowledge. And I really wasn’t able to get any answers, I know, you know this is obviously a novel illness. But just even for one doctor to look into it a bit and come back to me, didn’t happen.17

The absence of knowledge and information about long COVID symptoms was reported to create anxiety and confusion for patients.1 17–19 Ladds et al18 found that this confusion was intensified by the lack of medical knowledge, understanding and guidance from healthcare professionals. There were also reports of conflicting or inconsistent advice from health professionals.18

Some professionals did recognise the limitations of their own knowledge17 18 and referred patients to online support groups. Focus group participants suggested they would rather be told that the professional did not have the knowledge required to address their illness, if that was the case.17 The importance of finding a general practitioner (GP) who was understanding, empathetic and who provided support to those experiencing long COVID is highlighted in this quote:

I have to say it was a really powerful experience speaking to the GPs … the two more recent ones, actually just the experience of being heard and feeling like somebody got it and was being kind about it, but you know it was okay that they couldn’t do anything, I just kind of needed to know that I wasn’t losing it really and it was real what I was experiencing, I think so that was really helpful.17

Along with this perceived lack of knowledge, multiple perceived barriers to healthcare access were reported1 17 18, along with a perception among participants that health services and doctors were too busy dealing with cases of acute COVID-19 to have capacity to deal with anything else, including patients with long-term symptoms.1 18 This perception appeared strengthened by the difficulties people experienced when trying to access primary care, especially if they were seeking a face-to-face consultation.

I think the message to avoid hospital and the GP unless you had specific symptoms was very unhelpful, particularly as I didn’t have, and never have had, a cough or fever.1

In general, study participants found accessing care to be ‘complex, difficult and exhausting’.18 This led to patients describing how they felt they had to manipulate the inflexible algorithm-driven systems in order to receive care, which led to feelings of guilt and anger.18 Some patients described creative solutions they had come up with to help them access healthcare, while others reported resorting to private healthcare to access tests.18 Many patients felt they needed to conduct their own research and construct their own care pathways, taking the lead in arranging consultations with specialists and circumventing bottlenecks in the system.18 This was reported as a route often employed by medical professionals who themselves were suffering from long COVID.18

There was also a perceived lack of support within the system.1 17 18 Some individuals described how NHS111 (a national telehealth helpline in the UK) had directed them to their GP who then directed them back to NHS111.18 There was what appeared to be a lack of guidance for those who did not need to be admitted to hospital but were no longer in the acute phase of the illness.1 18 19

Patients who felt they had received satisfactory care and access to healthcare were generally those who had been offered follow-up appointments and who felt their healthcare providers listened to them and gave them ongoing support, even if that was in the form of a video or telephone call.17–19

Telemedicine was widely used to facilitate interactions with healthcare services.1 17–19 However, it was generally perceived by patients to have limitations.1 17 Remote consulting with primary care was viewed by some patients as potentially limiting direct access to GPs, disrupting continuity of care (people often could not see the same GP every time) and making the communication of symptoms more challenging.1 17 18 Some patients felt that strict adherence to protocols for telemedicine-delivered care affected patient safety or led to mismanagement of their care.

… I remembered ringing my GP from the floor on my lounge laying on my front and kind of saying I’m really short of breath, you know, do you think I should try an inhaler do I need to go back to A&E and I was kind of told well you don’t really sound too out of breath over the phone… I really felt at that point right if you could see me you would see that I am really like broken.18

A positive view expressed in relation to telemedicine was that it increased accessibility of primary care during periods of societal restrictions aimed at controlling the spread of COVID-19.

My doctor was available via messaging, telephone, and telemedicine. She also contracted COVID-19 so she shared her experience with recovery and it helped me stay calm that I was on the right track.19

When asked to describe desirable features of healthcare services or service delivery for patients with long COVID, research participants asked for face-to-face assessments1 17 and talked about the need for ‘one-stop clinics’ with multidisciplinary teams who could look at their wide-ranging symptoms and treat them holistically.1 17 18 A case manager to oversee individual patients and ensure that all aspects of their care was considered was suggested, along with meaningful referral pathways and criteria.1

What would be most helpful is if all main hospitals could have a COVID clinic that had experts from respiratory, cardiology, rheumatology, neurology, physiotherapy etc, so you could go along for half a day and see people from these different departments, they can refer you for tests and you can get a plan in place, we are having such a range of symptoms that GPs are struggling to know what to do with you.1

Other participants spoke about wanting to be listened to, to be believed and understood and to be offered practical advice on coping.1