The impact of the COVID-19 pandemic on the physical activity environment in English primary schools: a multi-perspective qualitative analysis

Article history

The contractual start date was in April 2021. This article began editorial review in February 2023 and was accepted for publication in July 2023. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The PHR editors and production house have tried to ensure the accuracy of the authors’ article and would like to thank the reviewers for their constructive comments on this article document. However, they do not accept liability for damages or losses arising from material published in this article.

Copyright statement

Copyright © 2024 House et al. This work was produced by House et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2024 House et al.

Participants and procedure

The Active-6 project is a repeated cross-sectional natural experiment examining the impact of the COVID-19 pandemic on the physical activity of 10- to 11-year-old children and their parents/carers in England. 28–30,37–39 To measure differences over time, accelerometer, questionnaire and qualitative data were collected in two waves (Wave 1: July–December 2021; Wave 2: January–July 2022), which were then compared with baseline data collected in 2017–8 during the B-Proact1v project (Wave 0). 28,40 The schools that took part in Active-6 were state primary schools in the wider Bristol area, England, recruited from those that took part in B-Proact1v. Fifty schools participated in B-Proact1v and 28 continued into Active-6, with a range of inner-city, suburban, rural and small town schools; size in terms of classes per year group and pupil numbers; local authority, academy and faith schools; and high/medium/low deprivation based on percentage of pupils receiving free school meals and school postcode Index of Multiple Deprivation (IMD) score.

This study is drawn from qualitative data collected in Waves 1 and 2 of Active-6 (see Figure 1). Participant groups were (1) children aged 10–11 years (Year 6) who had worn accelerometers for Active-6, (2) parents or carers of the child participants who had worn accelerometers for Active-6 and (3) primary school staff from the participating schools. Eligibility criteria for parents and children were that they had worn an accelerometer as part of the Active-6 project and had consented to being recontacted, while school staff needed to be a member of a school supporting the Active-6 project. Parents and children were approached via the contact information provided during the Active-6 sign-up process, whereas school staff were approached directly via e-mail. Parents and children who participated in the qualitative aspects of Active-6 were not related and there was no ‘complete’ data set that included a child, their parent and their schoolteacher. Due to recruitment challenges, parents and school staff were convenience sampled, whereas children were purposively sampled using their accelerometer data and individual and school demographic information. Semistructured interviews were conducted with parents and school staff and focus groups were conducted with children. In total, 12 focus groups were facilitated (Wave 1 = 6, Wave 2 = 6) with 92 children from 12 schools. The number of children in these focus groups was on average 8 and ranged from 5 to 10, with no repeat children between waves. Forty parents from 15 schools participated in 43 one-to-one semistructured interviews (Wave 1 = 21, Wave 2 = 22; 3 parents were interviewed in both waves). Lastly, 18 one-to-one semistructured interviews with 13 members of school staff from 12 schools were conducted (Wave 1 = 9, Wave 2 = 9; 5 school staff were interviewed in both waves). Information power was used to derive sample size, whereby the study’s aim, the extent of participants’ specific knowledge and experiences in relation to our research question, theoretical background of the study, dialogue quality and the adopted cross-case analysis were reflected on and discussed within the research team throughout data collection. 41 Tables 1–3 display participant demographic information, and participants for each group came from a range of study schools, which is explored further in the equality, diversity and inclusion (EDI) section below.

In Wave 1, parents were interviewed remotely between September and December 2021 and school staff between November and December 2021 (also remotely) by RW, TR and BT. Child focus groups were conducted in December 2021 in person in schools, facilitated by RW, TR, BT or DH. In Wave 2, parent interviews were conducted remotely via Zoom or telephone by RW between February and July 2022, and school staff interviews between May and July 2022 by RW (eight remotely via telephone or Zoom and one in-person). Child focus groups were facilitated by RW, DH and KS between May and June 2022 (see Figure 1 for data collection waves alongside school COVID-19 measures). Parent interviews ranged from 27 to 75 minutes in duration, school contact interviews from 33 to 59 minutes and child focus groups from 33 to 61 minutes.

All adult participants provided written informed consent, parents consented to their child participation in the focus groups and children provided additional written assent. 42 As an appreciation of their time, parents and school staff were given a £10 gift voucher, and the children received a small incentive when they took part in Active-6 (a frisbee or kit bag). Ethical approval was gained from the School for Policy Studies Ethics Committee at the University of Bristol, UK (Ref SPSREC/20-21/150). This project is funded by the National Institute for Health and Care Research (Public Health Research Programme NIHR131847) and the project was listed on the research registry (www.researchregistry.com/browse-the-registry#home/registrationdetails/604b4760d539c90020642be6/).

Study materials

Topic guides were developed by the research team to facilitate discussions in the semistructured interviews and focus groups. An independent parent patient and public involvement (PPI) member of the study management group provided feedback on all topic guides. A flexible, iterative process of reflection was used to adjust topic guides throughout the data collection process. We felt that each interview and focus group ran well and added valuable information related to our research question. In Wave 1, these focused on changes to parent and child physical activity behaviour over the COVID-19 pandemic, any factors that may have influenced any changes, and school environment changes in this time period that might have influenced activity levels among Year 6 pupils. Building on the data collected in Wave 1, Wave 2 topic guides explored parent and child changes in physical activity and screen-viewing behaviour from January 2022 onwards, factors that had influenced possible changes and the school perspective regarding changes to and influences upon activity levels among Year 6 pupils. All topic guides are presented in Report Supplementary Material 1 and details on the researchers and analysis in Report Supplementary Material 2.

Data analysis

Data were analysed using the framework method and organised using the NVivo 1.0 program (QSR International, Warrington, UK). 43 This process consisted of seven stages: (1) verbatim transcription of interview/focus group audio recordings (using an encrypted Dictaphone) by a university approved transcription service; (2) data familiarisation through reading and re-reading transcripts; (3) coding, undertaken by three members of the team (Wave 1 RW, BT, TR, DH or KS; Wave 2 RW, DH and KS) who each coded two transcripts from each participant group each wave using inductive and deductive codes. This process allowed the team to discuss and deliberate codes and our subjectivity in interpretation, leading to consensus;29 (4) collectively developing a working analytical framework using inductive and deductive codes; (5) applying the analytical framework to all transcripts; (6) charting raw data into the framework matrix using NVivo, which was then manually summarised by RW and LH; and (7) interpreting the data. At the interpretation stage, we combined school, child and parent perspectives, enabling triangulation of diverse experiences. All adult participants were given the opportunity to read and amend their transcripts prior to analysis, but none opted to do this.

This study was based in critical realism, a philosophical meta-theory that argues a world exists independently of human beings (ontological realism), but our perceptions and understanding of this world are derived through perceptions mediated by language, culture and human practices (epistemological relativism). 45 This allowed us to synthesise multiple perspectives in the context of the other, wider mixed-methods findings within the Active-6 project. This also means that we recognise the subjectivity of our qualitative analysis and that our interpretations are culturally situated.

Patient and public involvement

Patient and public involvement has been central to the Active-6 project. Year 6 children, teachers and school staff in a range of roles have been engaged in our research design, study materials and dissemination plans, in two-way feedback between participants and the research team. This has included parent members of study governance groups, running child PPI group sessions at schools to review data collection methods and dissemination materials, and sharing early school-level results with schools and participating families. This engagement has provided valuable feedback to Active-6, enabling us to improve and adapt the study as it rolled out.

Equality, diversity and inclusion

Equality, diversity and inclusion were considered in participant recruitment for this study. A range of schools in terms of location (urban, suburban, town, rural), size (number of Year 6 classes) and deprivation (school postcode IMD) were included in each participant group. As recruitment was taking place each wave, participant demographics were monitored and certain schools and groups were targeted to increase their inclusion in the study. However, inclusion and representation of diverse ethnic backgrounds, lower socioeconomic status (SES) groups and male parents are limited, in part due to the challenges of conducting research during a pandemic and ongoing school and family disruptions. School staff who facilitated the accelerometer data collection were invited to participate in an interview, with attention paid to securing a range of job roles and a gender balance, which was better achieved in Wave 2 (see Table 1). Parents were categorised as low, medium or high MVPA level based on their accelerometer measured weekday MVPA in comparison to their school group. Their IMD score (based on home postcode), age, ethnicity and highest level of educational qualification were all noted. The majority of participants were female, white British, higher qualified, higher IMD and active. Intentional sampling helped to achieve a greater balance in Wave 2 regarding parent gender, but not in terms of participation of lower SES parents (see Table 2). Child activity levels were generated in the same manner, and even ratios of children with low/medium/high MVPA from schools situated in an even range of urban/rural and high/low deprivation areas were invited to attend a focus group (see Table 3). The demographics of the sample may mean that those schools and families facing greatest challenges are not represented in this study, and therefore work that includes these experiences is needed to ensure policy implications are relevant and suitable for all.

Theme 1: the return to school

When schools re-opened to all students after the first national lockdown in June 2020, it was apparent to school staff that children’s physical, social and academic development had been impacted. We have reported elsewhere that parents who had children at home with them during lockdowns particularly noticed that not only had children lost their curriculum PE sessions, but they lost any active travel to and from school, active play during breaktimes, active after-school clubs and spontaneous after-school park visits. 29 Although strategies to promote physical activity at home during school closures were developed, teachers expressed difficulty in creating exciting virtual PE sessions using the child’s home environment and children expressed that PE at home under lockdowns was boring. 29 It became particularly challenging for teachers to influence and engage children who were not motivated to take part and did not have a parent to encourage them, as well as children who had issues accessing the necessary technology or enough space to be able to participate. 29

Across the parent and school staff interviews, it was felt that the return to the school environment increased children’s physical activity and the opportunity to develop physical skills, even while social distancing measures were in place.

When they went back to school they were doing more physical activity, because they were at school and so they still had their playtimes and they had just being up and about, and naturally doing more throughout their day than being at home and home-schooling. Walking to school, walking back from school … You know, all those little parts.

Parent 17, Female, School ID 75, Wave 1

The positive impact of attending school on physical activity was discussed through ‘keyworker’ children – children whose parents worked in key services that continued through lockdowns so attended school in-person. Several teachers described how during lockdowns, due to a prioritisation of well-being at school, COVID-19 mitigation measures, good weather and the smaller number of children in school, they were in fact able to do more active and outdoor activities than under pre-pandemic school conditions, and keyworker parents noted this in our interviews. PE for keyworker children in school was largely enjoyed, but some noted technical difficulties that at times made the sessions challenging, such as participating in a ‘blended’ lesson in a classroom while classmates at home joined virtually, or being limited to the classroom space.

It was warm, it was spring, it was summer, we went out and let [keyworker children] play for longer. […] If anything, they probably had more physical activity when they were in school. […] So COVID-19 made us kind of freer to go out. The children at home, we knew, would not be doing very much physical activity

School Contact 6, Year 6 Teacher, School ID 71, Wave 1

I was in school. Playtime was amazing then. [… We] used to play tennis, because we used to be able to get, like, balls and tennis rackets out

Focus Group 3, School ID 95, Wave 1

We had to do [PE] in class, which was annoying because we did not get to move around as much and run around

Focus Group 5, School ID 72, Wave 1

In September 2020 when the school term began, most schools undertook some form of ‘recovery curriculum’ to support the development of the pupils’ social, physical and emotional skills and reacclimatise them to the school environment (see Figure 1). Underlying this approach was recognition of the uneven impact school closures had on pupils, and how some children had fallen far behind expected development levels. The recovery curriculum, then, meant pushing academic priorities aside in the short term to bring children up to a base-level skill set to cope with the school environment, while understanding that these measures were a necessary pathway to be able to once again focus on academic outcomes. Being physically active in school at this time played a key role in this adjustment, with schools using physical activities, sports and games to practise teamworking, increase stamina and assist conflict resolution. The COVID-19 risk levels at this time being relatively high also encouraged staff to spend more time in outdoor activities.

We found that when the children came back, they didn’t even know how to sit on the chair properly, they’d been sat on the sofa doing their work for the last six months. They just had their brother or sister to deal with for the last six months, they didn’t have these 30-odd children. […] For the first term, we completely redesigned our curriculum. For the first weeks there was no maths and English. […] We focused on our five golden threads. A lot around team-building, physical activity, conflict resolution. […] It was all around the social skills and their building up stamina to be able to sit down and do an English lesson. […] Then, when we were going into the fourth, fifth, sixth week of the first term, we started dripping in snappy maths and English lessons. Again, very physical [active] maths and English lessons. […] Then, we started weaving in the more classic maths and English lesson of sitting down and doing a worksheet

School Contact 7, Deputy Head Teacher, School ID 57, Wave 1

Theme 2: over-pressured staff and environment

Lost learning and skill development among children over the pandemic created exceptional pressure for students to academically ‘catch up’, which staff explained exacerbated pre-existing issues within the highly demanding state primary school environment, such as limited resources and overburdened workloads. Although school staff described an ability to prioritise social, emotional and physical well-being and recovery during the first reopening of schools in September 2020 (Theme 1), by September 2021 pressures from governing bodies impacted the extent to which they could diverge from the expected curriculum, including a waning of the use of physical activity to assist social, academic and physical development. This pressure for pupils to academically catch up was exacerbated by an increased need to support children who were not yet reaccustomed to learning in the school environment.

The government has made it clear that they would like pupils to be back on track … I think that’s just filtered through our Trust and it’s filtered through the Heads and it’s filtered through to classes

School Contact 1, Head Teacher, School ID 44, Wave 2

I feel like I’m needing to do more for them than I would have done historically with the same aged cohorts. Just a little bit more spoon-feeding and I don’t even mean academically. I just mean helping them listen, helping them achieve what I’ve asked them to do, or helping them understand what the structure of the activity is

School Contact 9, Year 6 Teacher, School ID 61, Wave 2

These post-lockdown school pressures impacted upon the time for and quality of structured and unstructured physical activity. For the Year 6 cohort in this study, who took the first Standard Assessment Test (SAT) exams since the pandemic began (in May 2022), teachers described how PE lessons and other additional physical activities competed for time against ‘higher priority’ core subjects which are assessed. This is counter to the experience of many schools during the recovery curriculum, where outdoor and active time helped children’s behaviour and ability to concentrate in the classroom.

At the moment, I don’t think many [classes] are doing [The Daily Mile] because the curriculum is so tight that we can’t literally fit it in

School Contact 5, Year 6 Teacher, School ID 25, Wave 1

Before SATs we didn’t do as much PE as we did. I think it was because we were all getting ready for SATs. […] the teachers are always like, ‘We’ll do PE if we have time.’ It’s like PE is for the spare time. It’s not really that important to them

Focus Group 5, School ID 75, Wave 2

Many Year 6 teachers described a lack of confidence in delivering quality PE at the Year 6 level. At this age, pupils can be experienced in a sport or activity if they attend a club outside of school. Teachers are aware these pupils are about to go to secondary school where PE will be taught by a subject specialist, and several teachers described how they felt the PE provision in the postgraduate certificate of education (PGCE) teaching qualification was inadequate and left them feeling ill-equipped to teach. Staff described how this had been an issue before the pandemic but was exacerbated by post-lockdown academic pressures. Teachers described feeling inclined to cancel PE or preferring for it to be delivered by external providers, to use PE time to complete administrative tasks. Using timetabled PE to catch up on other work further reduces PE training and continuing professional development (CPD) opportunities for staff, and improving delivery of PE lessons requires teachers to find time in their already over-burdened workload or, more often, to do so outside of work hours.

When you are a primary school teacher, obviously you deliver all of the provision. We can’t all be brilliant at everything. So, the most obvious thing to say is there are some teachers who don’t feel that they are good at delivering PE, particularly as you go higher up the school, where actually there are children who might do the sport externally and might actually end up knowing more than you do. It can sometimes feel like you are trying your best, but maybe what you are delivering is not as good as it could be

School Contact 9, Year 6 Teacher, School ID 61, Wave 2

Occasionally we’ll have a PE staff meeting but they’re few and far between. I don’t think many people do [the CPD courses that are available] because of time. When you go back to it, it’s down to time

School Contact 4, Year 6 Teacher, School ID 71, Wave 2

Playground support staff were often insufficient as staff retention was challenging at this time. This impacted upon the quality of the playground support staff, where the high turnover required ongoing training in facilitating games and activities. With fewer staff in the playground and lower staff-to-children ratios, fewer physical activities were supported during break and lunch times.

Sometimes I do training with the lunchtime staff and I train them in what things they could do at lunchtime with the children, so different sports, different games they can play. Whether it’s because of COVID or not, we have such a high turnover of staff and quite often there’s not a lot of staff that turn up for work. […] If we’ve only got a couple of staff outside, they can’t then offer the activities. [… We then have an issue with] whether we can keep upskilling them because every year we keep getting different ones and it keeps changing all the time

School Contact 3, PE Coordinator, School ID 72, Wave 2

School environment pressures also impacted the number and variety of extracurricular clubs, which were often dependent on staff skills and availability. However, the teaching workload required teachers to use their spare time in the evenings or weekends to find the capacity to run these clubs, which meant needing to catch up on core academic workloads. For these reasons, it required staff who were highly motivated to organise these clubs, without which the number and variety of clubs might have been reduced.

It’s having enough adults on the staff team, that aren’t at complete breaking point, that are willing to help us [the PE lead and assistant]. […] Ideally, because we’ve got so many children who are engaged in clubs, we would take every child to an event with us but it’s having those adults who are willing to do it

School Contact 6, PE Coordinator, School ID 74, Wave 2

Theme 3: the uneven impact of the pandemic

This theme describes how the lockdowns and COVID-19 measures had an uneven impact on schools and their pupils. The post-lockdown physical activity environment is characterised by variation and is explored through three subthemes: retained pandemic policies; impact on physical activity culture and different children need different things.

Study implications

The key findings and implications of this study are summarised in Table 5. The study’s key finding is that the post-lockdown primary school environment and staff body are characterised as over-pressured, with staff expected to deliver on core academic work while trying to meet complex post-lockdown pupil needs. This is impacting upon the child physical activity environment, although this impact is characterised by variety and divergence of experience. Schools would benefit from upstream policy changes to alleviate the pressure placed on them through school governance systems. This might enable schools to retain and continue the physical activity and well-being-centred approach seen during recovery curriculums. This implication has been included in an Active-6 study policy briefing which has been developed in partnership with key policy and practitioner stakeholders.

Several additional implications have arisen in this study. Many Year 6 class teachers feel unconfident in delivering quality PE lessons. COVID-19 social distancing policies have been retained to varying degrees within schools, at times in the interests of pupils and at others in the interests of the over-pressured school. These policies should be evaluated for their impact on child physical activity. Schools’ physical activity cultures have changed in varying ways since lockdowns. Context-specific research is necessary to understand how these cultures are created and shaped, and future intervention work should make school culture, ethos and context central in their implementation and evaluation. Lastly, the COVID-19 lockdowns had an uneven impact on children’s physical activity, and some groups risk getting left behind. Strategies to reach all pupils are necessary even more so than before the pandemic, and schools should be supported in meeting these now more divergent levels of child ability and need.

Strengths, limitations and future research

This study has several strengths. It combines school, parent and child perspectives, enabling triangulation of opinions and experiences. Furthermore, data were collected at two time points in a period of rapid change, both in COVID-19 policy and mitigation strategies, particularly within schools, but also in embodied experiences of the pandemic and restrictions. This has provided rich data on a complex issue, supporting an analysis that has been able to consider how experiences have changed over time.

Active-6 was limited to the school sample from the comparator baseline B-Proact1v study. At the time of data collection (both quantitative and qualitative) schools were under great pressure, negotiating ongoing COVID-19 complications and outbreaks while supporting our work. This likely limited the numbers of schools able to participate in Active-6. Although a range of schools participated in this qualitative study (see Methods), these factors again may have contributed to our participant sample from within these schools. It was challenging to recruit staff for interviews, particularly SLT staff, and the parent interview sample is predominantly female, active and of higher SES. Therefore, the experiences of male, less active and lower SES parents are limited, and this must be considered in interpretation of the parent findings.

This study suggests that the impact of COVID-19 on child physical activity is uneven, affecting some children more than others. Future work is therefore needed to explore the details of this potential diverging experience, to understand which sorts of schools followed which paths through the pandemic and how children have been differently affected, particularly over the longer term. Importantly, in light of the varying experiences this study has highlighted, future work to support schools to improve child physical activity and their physical activity environments must be context specific: sensitive to staff abilities and capacities, the SLT, school space and environment, facilities, wider neighbourhoods and environments, demographics, cultures and indicators of deprivation such as numbers of pupils receiving free school meals.

Acknowledgements

A special thank you to the Active-6 schools, the school staff, child participants and parents/carers for their involvement in the study.

Thank you also to previous team members: Prof. Charlie Foster, who was co-applicant and in the Study Management Group; Byron Tibbetts, who was Project Manager from April to December 2021; Tom Reid, who was Fieldworker April 2021–January 2022; Lara Hollander, who helped to develop the Wave 2 qualitative frameworks; Christine O’Shea, who was summer Intern in 2022; and Tabitha Pring, who was a casual fieldworker for Wave 2 data collection.

Contributions of authors

Danielle House (https://orcid.org/0000-0001-6171-9922) (Study Manager) conducted analysis and led on the writing of the paper.

Robert Walker (https://orcid.org/0000-0001-9901-5285) (Qualitative Research Associate) supported all analysis and writing of the paper.

Ruth Salway (https://orcid.org/0000-0002-3242-3951) (Senior Research Associate, Statistics) edited the paper for intellectual content.

Lydia Emm-Collison (https://orcid.org/0000-0002-5493-3223) (Lecturer in Physical Activity and Behaviour Change) edited the paper for intellectual content.

Katie Breheny (https://orcid.org/0000-0001-6886-4049) (Senior Research Associate, Health Economics) edited the paper for intellectual content.

Kate Sansum (https://orcid.org/0000-0003-3392-6750) (Fieldworker) edited the paper for intellectual content.

Sarah Churchward (https://orcid.org/0009-0007-7765-3682) (PPI member of Study Management Group) edited the paper for intellectual content.

Joanna G Williams (https://orcid.org/0000-0002-4737-1760) (Consultant in Public Health) edited the paper for intellectual content.

Frank de Vocht (https://orcid.org/0000-0003-3631-627X) (Professor in Epidemiology and Public Health) edited the paper for intellectual content.

Russell Jago (https://orcid.org/0000-0002-3394-0176) (Principal Investigator, Professor of Physical Activity and Public Health) oversaw all aspects of study design and interpretation, and edited the paper for intellectual content.

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are available in the toolkit on the NIHR Journals Library report publication page at https://doi.org/10.3310/KLML4701.

Primary conflicts of interest: Russell Jago, Katie Breheny and Frank de Vocht are partly funded by the National Institute for Health Research (NIHR) Applied Research Collaboration West (NIHR ARC West) at University Hospitals Bristol NHS Foundation Trust and the University of Bristol. Russell Jago is partly funded by the National Institute for Health and Care Research Bristol Biomedical Research Centre, and was a member of the PHR Prioritisation Group (11 October 2019–12 October 2021) and a member of the NIHR PHR – Research Funding Board (1 June 2014–12 October 2021). Frank de Vocht has been on the NIHR Public Health Research Funding Board since 8 October 2019.

Ethics statement

Ethical approval was gained from the School of Policy Studies Ethics Committee at the University of Bristol, UK (Ref SPSREC/20-21/150), on 9 March 2021. The project was listed on the Research Registry.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to anonymised data may be granted following review.

Information governance statement

The University of Bristol is committed to handling all personal information in line with the UK Data Protection Act (2018) and the General Data Protection Regulation (EU GDPR) 2016/679. Under the Data Protection legislation, the University of Bristol is the Data Controller, and you can find out more about how we handle personal data, including how to exercise your individual rights and the contact details for our Data Protection Officer here (www.bristol.ac.uk/secretary/data-protection/).

Funding

This article presents independent research funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme as award number NIHR131847. The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, the PHR programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, the PHR programme or the Department of Health and Social Care. Study data were collected and managed using REDCap hosted at the University of Bristol. The sponsor of this study is University of Bristol, Research and Enterprise Development, One Cathedral Square, Bristol, BS1 5DD, UK www.bristol.ac.uk/red/.

Article history

The contractual start date was in April 2021. This article began editorial review in February 2023 and was accepted for publication in July 2023. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The PHR editors and production house have tried to ensure the accuracy of the authors’ article and would like to thank the reviewers for their constructive comments on this article document. However, they do not accept liability for damages or losses arising from material published in this article.