How Long Covid Affects Work and Job Changes in England

This study is a secondary analysis of pre‐existing interview data collected within the REACT Study, a broader population‐based cohort described elsewhere [18, 19, 20]. Figure 1 presents an overview of REACT components and the process by which the final sample was derived (N = 60). A comprehensive account of the interview study methodology has been reported previously [21]. Briefly, interview participants were drawn from the REACT‐LC sample based on eligibility criteria. To be considered, individuals must have experienced persistent symptoms lasting 12 weeks or more following symptomatic Covid‐19 (regardless of diagnosis status), attended an assessment clinic and provided biological samples and/or completed a Health and Wellbeing survey, and consented to be recontacted for an interview. To capture a wide range of experiences, the sampling approach purposively selected individuals to reflect diversity in age, gender, ethnicity and symptom severity. For the present analysis, no additional sampling was conducted, and data saturation did not guide participant selection. This approach aligns with established guidance for secondary analyses, where the analytic focus is on re‐examining existing data rather than generating new data [22, 23].

Interviews used a semi‐structured topic guide that was developed in line with previous literature, a pilot interview study [24], and feedback from the REACT Public Advisory Group (PAG). It comprised questions related to participants' history of Covid‐19, pre‐pandemic and pandemic life experiences, persistent Covid‐19 symptoms, perceived trajectories of these and management strategies [25]. Interviews were conducted online either via MS Teams or Zoom, between January 2022 and January 2023. Each interview lasted around an hour and was recorded for professional transcription.

We selected framework analysis as our analytical approach due to its suitability to manage a relatively large dataset and its allowance to incorporate both a priori and emerging codes [26]. NVivo 14 software [26] was used to organise the data, assist coding and document our analytic decisions. We followed the five stages of framework analysis outlined by Ritchie and Spencer [27]: (1) Familiarised with all transcripts and field notes; (2) Developed initial codes using a combined deductive‐inductive approach shaped by our research question and prior literature, while allowing new codes to emerge [28]; (3) Iteratively refined our codebook by consolidating codes into broader thematic categories and then applied it to the remaining transcripts; (4) Developed a framework matrix to summarise each participant's employment experiences; and (5) Finalised themes by identifying patterns within and across participants.

Coding was primarily conducted independently by the lead author, with co‐authors providing input on the final themes. Interrater reliability was not formally assessed; instead, rigour was supported through regular presentation of preliminary analyses and interpretations to the wider research team for discussion and critical feedback [28]. Illustrative quotes were selected based on representativeness of patterns and themes, clarity of expression, and diversity of perspectives, including negative or minority cases [29, 30, 31]. The lead author carried out the initial quote selection, with feedback from PAG co‐authors prompting some alternative choices to ensure better alignment with the criteria.

We reflected on our positionality as researchers and the potential impact of our own experiences, assumptions and biases in the data collection, analysis and interpretation. Individual reflections and group discussions were recorded in our research log and allowed us to consider our insider/outsider perspectives, track our analytic decisions and iteratively adapt our methods.

This theme captures participants' perspectives about how their persistent Covid‐19 symptoms intersected with their ability to meet work expectations. The symptoms most commonly described as disrupting work were physical fatigue, diminished cognitive capacities (e.g., brain fog, poor memory and mental fatigue), breathlessness and pain. Symptoms manifested in diverse combinations, severities and timelines, shaping participants' capabilities to carry out work to varying degrees. Some participants reported they felt unable to deliver outputs to their previous standard, as Vivian (46–55 years, FT) illustrated: 'I'm not functioning at the same levels as I used to. Stuff that I used to knock off in a couple of hours is now taking me weeks'. A sense of diminished capacity to meet work demands was echoed by others, not only in terms of time taken to complete tasks but also in the need to repeatedly revisit and relearn material long familiar to them. Elaine (46–55 years, FT), who worked as an exhibition's curator, described: I've done my job for 30 years and I'm very competent at it, but I couldn't understand what some of the words were meaning. I had to reread about stuff. I was thinking I'd forgotten about things. It was more than just being out of touch […] It's almost having to recognise things again and relearn things which I struggled with a lot.

Others reported that the severity of their symptoms, combined with the demands of their roles, made it impossible altogether to sustain their previous employment. Jasleen (56–65 years, FT), who worked teaching vulnerable people, illustrated: 'I couldn't entertain work now, at all, because it would just be‐ especially the kind of work I do. The work I do, it gives out a lot of energy'. Several other participants also used 'energy' to describe finite functional reserves necessary to meet work demands, which were often depleted by their symptoms.

Beyond cognitively demanding roles, work disruptions were also pronounced among participants employed in public‐facing positions, where physical demands, immediacy of response and greater reinfection risks exacerbated challenges to continue working. This was often reported among healthcare workers and those in retail or service positions. Vinay (46–55 years, FT), who worked in a hospital, reflected on this, noting that 'it was frightening because being on a ward is not something that you could just think of not doing, moving around and being active'.

Additionally, participants described work disruptions related to fluctuations of their symptoms through time, non‐linear recovery trajectories, and uncertainty on the overall course of the condition. Some of the participants also perceived that work had sometimes caused a relapse or exacerbation of their symptoms, which was often associated with a quick return to work or 'overdoing it'. For instance, Tasha (46–55 years, FT) who held a position within a charity organization, described each time she would try to return to work as 'things were getting better', her symptoms would 'go backwards' and she had to take additional time off, which affected her emotionally and financially: I was trying to work. So, I would work and I'd be off, I'd work and I'd be off, and I'd just be going through this yoyo existence. I'd manage a couple of months of work and then I'd have to be off again for a month […] so income was still really sporadic.

This theme details the consequences of participants' 'struggles to get through' work, including the emotional, physical and social tolls involved in balancing job demands with persistent symptoms. Some participants reported losing self‐confidence in their abilities to perform previous work tasks, which led to feelings of frustration, low mood and anxiety. As Dolores (36–45 years, PT) illustrated: 'I'm a bit of a person who doesn't want to be seen as not doing something correctly, I like to do a job to the best of my ability. And feeling that you're not necessarily doing that to the best of your ability is frustrating'. Similarly, participants reported facing or fearing an increase in their work mistakes, so some verified their outputs multiple times which often led to additional stress and more time needed to complete a task. Sabina (46–55 years, FT), who's work involved dealing with numbers, explained: I could not concentrate, I had to keep on doing things over and over again, just to make sure that I was doing it right. Which kind of stressed you out, because you think, "Oh my God, I've got such a responsibility here".

Changes in participants' capabilities challenged their established roles within their work teams. Colleen (36–45 years, FT), a project manager operating at the interface of environmental policy and science, remarked: It was noticeable to my peers as well that I wasn't quite as sharp as I usually am. You know, I'm the one that never says anything inappropriate at a meeting, but it wasn't that I was inappropriate. I'm the people person. I'm the communicator. I'm the one that will explain this policy, you know, in simple terms, and I just … yes, it was difficult. It was hard work.

In the face of these challenges, several participants felt that leaving their employment was not an option for them, mostly due to a combination of financial circumstances, and wishing to maintain their professional identity and sense of purpose. This resulted in participants attempting to reconcile work expectations with their symptoms and daily life, which required them to plan, prioritize and iteratively appraise their activities. Danielle (46–55 years, FT) explained:I put a lot of emphasis on work because I want to keep my job, and I want to be able to at least have some income, but work means […] there's no energy left, at the end of the day, to do that [laundry and other home chores].

Limited energy and prioritization of work reduced participants' opportunities to engage in other valued activities, such as recreation and leisure, which previous research suggests may contribute to long‐term decline in physical and mental well‐being [33] and in work ability [34]. This also increased reliance on social networks for instrumental support with home and caregiving tasks, potentially generating additional challenges for women and migrant individuals, who are more often expected to assume primary responsibility for these roles [35, 36].

Taken together, challenges in balancing employment with fluctuating symptoms and non‐linear trajectories placed participants, such as Elaine (46–55 years, FT), in a state of uncertainty regarding their identity and career prospects: What are we now, 6 months into returning and I'm still only doing 2 days a week. I wondered will I ever get back up? Hopefully I can get back up to things, but I don't think I'll probably return to be, maybe I will, I don't know, the person I was before.

This theme describes participants' perspectives of individual and organizational strategies for managing persistent symptoms at work, as well as perceived barriers or enablers influencing their implementation.

The purposive sampling approach used to recruit participants to the REACT interview study enabled us to explore a wide variety of employment experiences among people with persistent Covid‐19 symptoms, including individuals with heterogeneous demographic backgrounds, diverse employment types, and varying access to long Covid support groups and clinical services.

The cross‐sectional study design of our study provides limited scope to explore the long‐term effect of long Covid in employment. This should be addressed in future studies as the condition is characterized by potential relapses and knowledge about the condition continues to emerge in the face of changing socio‐economic conditions that may affect opportunities for sustainable employment. Although our sampling approach allowed us to recruit participants from several regions of England and yielded a substantial sample by qualitative standards [56, 57], the purposive and self‐selective nature of recruitment may have introduced selection bias and constrained representativeness [58]. As such, our findings should be considered with caution and triangulated with longitudinal evidence and studies from different political and economic contexts. Moreover, we consider that understanding the employment experiences of people with long Covid also requires the consideration of the perspectives from other relevant stakeholders not included in this analysis, such as carers and family members, line managers, co‐workers, occupational therapists and clinicians.