Smartphones, social Media and Adolescent mental well-being: the impact of school policies Restricting dayTime use—protocol for a natural experimental observational study using mixed methods at secondary schools in England (SMART Schools Study)

Introduction Smartphone and social media use is prevalent during adolescence, with high levels of use associated with lower levels of mental well-being. Secondary schools in the UK have introduced policies that restrict daytime use of smartphones and social media, but there is no evaluation on the impact of these policies on adolescent mental well-being. The SMART Schools Study aims to determine the impact of daytime restrictions of smartphone and social media use on indicators of adolescent mental well-being, anxiety, depression, physical activity, sleep, classroom behaviour, attainment and addictive social media use. Methods and analysis This is a natural experimental observational study using mixed methods. Secondary schools within a 100 mile radius of the recruiting centre in the West Midlands (UK) have been categorised into two groups: Schools that restrict (intervention) and permit (comparator) daytime use of smartphones. We aim to recruit 30 schools (20 restrictive, 10 permissive) and 1170 pupils aged 12–13 and 14–15 years. We will collect data on mental well-being, anxiety and depressive symptoms, phone and social media use, sleep and physical activity from pupil surveys, and accelerometers. Policy implementation measures and data on individual pupil factors will be collected through school staff surveys, and website/policy analysis. Six case study schools will explore individual, school and family/home factors that influence relationships between school smartphone policies, smartphone/social media use, and mental well-being. Economic evaluation will be completed through a cost–consequence analysis from an education sector perspective. Ethics and dissemination Ethical approval was obtained from the University of Birmingham’s Research Ethics Committee (ERN_22-0723). Parents/carers of pupil participants can complete a form to opt their child out of the study. Pupil, school staff and parent/carer participants are asked to complete online/written consent (or assent). Findings will be disseminated through policy briefings, resources for schools, social media, reports, and open access publications. Trial registration number ISRCTN77948572.

ABSTRACT Introduction Smartphone and social media use is prevalent during adolescence, with high levels of use associated with lower levels of mental well-being. Secondary schools in the UK have introduced policies that restrict daytime use of smartphones and social media, but there is no evaluation on the impact of these policies on adolescent mental well-being. The SMART Schools Study aims to determine the impact of daytime restrictions of smartphone and social media use on indicators of adolescent mental well-being, anxiety, depression, physical activity, sleep, classroom behaviour, attainment and addictive social media use. Methods and analysis This is a natural experimental observational study using mixed methods. Secondary schools within a 100 mile radius of the recruiting centre in the West Midlands (UK) have been categorised into two groups: Schools that restrict (intervention) and permit (comparator) daytime use of smartphones. We aim to recruit 30 schools (20 restrictive, 10 permissive) and 1170 pupils aged 12-13 and 14-15 years. We will collect data on mental well-being, anxiety and depressive symptoms, phone and social media use, sleep and physical activity from pupil surveys, and accelerometers. Policy implementation measures and data on individual pupil factors will be collected through school staff surveys, and website/policy analysis. Six case study schools will explore individual, school and family/home factors that influence relationships between school smartphone policies, smartphone/social media use, and mental wellbeing. Economic evaluation will be completed through a cost-consequence analysis from an education sector perspective. Ethics and dissemination Ethical approval was obtained from the University of Birmingham's Research Ethics Committee (ERN_22-0723). Parents/carers of pupil

INTRODUCTION
Globally, mental disorders (eg, anxiety and depression) are the leading cause of disability

STRENGTHS AND LIMITATIONS OF THIS STUDY
⇒ This is a natural experimental observational study exploring the impact of school phone policies that restrict daytime use of smartphones and social media on adolescent mental well-being and associated behaviours, and the cost implications of these policies from an education sector perspective. ⇒ Pupils will self-report data on time spent on their smartphones and social media by using data from their iOS or Android apps. ⇒ Pupil outcomes will be collected from an online survey that includes validated measures for mental well-being, anxiety and depressive symptoms, addictive social media use, motives for social media use, health-related quality of life and demographic data; and from accelerometers measuring physical activity and sleep. ⇒ Qualitative research will explore how individual, school and family/home factors influence relationships between school phone policies, phone/media use and mental well-being. ⇒ This is a cross-sectional study and is limited to data collected in England.

Open access
in adolescents (age range: [10][11][12][13][14][15][16][17][18][19]. 1 2 In the UK, one fifth (20%) of adolescents are reported to have a mental health disorder, 3 mostly anxiety and depression. 2 Half of all mental health disorders start before the age of 14, 4 5 and if left untreated, mental health problems are highly likely to persist well into adulthood. 2 6 Poor mental well-being also negatively affects other aspects of adolescents' lives, including cognitive, social and physical behaviours. 6 7 For example, poor mental well-being is associated with higher rates of disruptive behaviour, school absence and lower educational attainment. 8 9 Sleep problems are also common among adolescents diagnosed with anxiety and depression, and evidence suggests a bidirectional relationship between sleep disturbances and mental health problems. 10 Levels of physical activity also decline significantly during adolescence, and this coincides with increased onset of mental health problems. 4 11 Hence, adolescence is a crucial period for mental health interventions, 7 and there is a pressing need to improve and develop approaches to mental health prevention and intervention. 6 Smartphone and social media use is prevalent during adolescence, and accounts for the majority of their overall screen time. 12 13 In the UK, most adolescents (98%) own a smartphone, and are reported to be active users of social media (93%), 14 with comparable trends reported in other Western populations. 15 Samples in the USA and the UK (2021-2022) estimate that the time adolescents spend on smartphones and social media ranges between one and a half hours and eight and a half hours per day, 12-14 16 17 with most adolescents spending between 1 and 3 hours per day. 14 16 17 Problematic social media use is also prevalent, with 12% of adolescents in England reported to exhibit addictive use behaviours. 18 In moderation, smartphone and social media use (eg, <2 hours per day) can be advantageous for mental wellbeing and mental health, [19][20][21][22][23][24] as well as other associated health and behavioural outcomes (eg, sleep, physical activity, classroom behaviour and attainment). However, at higher levels of use, the reverse effect tends to be seen, with increasing time spent on smartphones and social media associated with decreasing levels of mental well-being and higher levels of anxiety and depression. 17 19-21 23 25 26 Poor academic performance, disruptive classroom behaviour and less time spent in physical activity and sleep are also more likely in adolescents who spend a greater proportion of time on smartphones and/ or social media. 19-23 25 26 Reducing the time adolescents spend on smartphones and social media is thus a plausible intervention to improve mental well-being, possibly operating through improving the related behavioural outcomes (eg, physical activity, sleep, academic performance, classroom behaviour). However, uncertainties in the strength of associations between smartphone/social media and mental well-being exist, and this is mainly due to reliance on self-reported use. 15 27-29 Furthermore, individual (eg, gender, age, socioeconomic status) 13 16 30-33 and family/home (eg, parental usage and attitudes toward technology) [34][35][36][37] factors are also likely to impact on relationships between smartphone/social media use and mental well-being.
There is considerable evidence that school-based interventions can have beneficial effects on adolescent mental well-being and associated behavioural outcomes (eg, sleep, physical activity, classroom behaviour and attainment). [38][39][40][41] Whole-school environment interventions that promote lifestyles conducive to good health are reported to have a more pronounced effect on mental well-being than individual approaches targeting knowledge and beliefs. 38 42 A whole-school approach targets physical and social influences of health, and through the alignment of school policies, values and practices with effective school leadership, the whole school approach seeks to promote a set of values, attitudes and behaviours that encourage the development and maintenance of positive physical, social, cognitive and emotional habits. 38 40 Evidence suggests that whole school policies related to health and well-being can: (1) reduce overall screen time; (2) positively influence mental well-being; and (3) improve physical activity, sleep, educational attainment and reduce disruptive classroom behaviour. [43][44][45] Therefore, whole school policies aiming to influence smartphone and social media use have the potential to positively impact on adolescents' mental well-being.
School phone policies that restrict daytime phone/ media use are an example of a current whole-school environment intervention. In the UK, Australia, Sweden, Czech Republic and elsewhere, many schools have introduced school policies that restrict daytime use of smartphones in order to reduce classroom disruptive behaviour and cyberbullying incidents and improve attainment. 44 46 47 We suggest that these policies have the potential to lower the overall time adolescents spend on the smartphones/social media, which may improve mental well-being and associated behavioural outcomes. However, there is currently no evaluation of the effect of school smartphone policies on mental well-being and there is limited evidence on how smartphone policies are implemented in schools. 48 The SMART Schools Study aims to determine the impact of school daytime restrictions of smartphone and social media use on adolescent mental well-being (primary outcome), anxiety, depression, physical activity, sleep, classroom behaviour, attainment and addictive social media use. We will do this by comparing impacts in two different secondary school phone policy contexts: (1) schools that do not permit smartphone use during recreational time in the school day (intervention); and (2) schools that permit smartphone use during recreational time (breaks/lunchtimes) (comparator). We will also explore how variation in school-based, individual and family/home factors influences the relationship between school phone policies, smartphone and social media use and mental well-being. We will conduct an economic evaluation in the form of a cost-consequence analysis from an education sector perspective.

METHODS AND ANALYSIS
This is a natural experimental observational study using mixed methods, taking place between April 2022 and July 2024. Quantitative and qualitative data will be collected from all schools in the sample (n=30) to compare outcomes between restrictive (intervention) and permissive (comparator) school policies and to complete an economic evaluation. Qualitative data will be collected from six case study schools to understand the contextual factors that could influence relationships between school policies, smartphone and social media use and mental well-being.
Intervention: school smartphone policies that restrict daytime use The intervention and data capture are directed by our logic model (figure 1) which integrates multiple theories and evidence. First, we adhere to displacement theories 19 to propose that reducing the time adolescents spend on phones/media (ie, restricting school time use) is optimal for mental well-being. Overuse can displace other mental well-being promoting activities (eg, sleep and physical activity) and very low use can deprive adolescents of interactions that support mental well-being (eg, affect and relationships). 19 21 23 Second, psychological motives drive phone/media use; motives related to enhancement and social interactions promote mental well-being and motives related to coping and conformity (eg, Fear of Missing Out) are associated with problematic use (addiction) and poor mental well-being. 49 50 Hence the school policy and ethos have the potential to influence adolescents' motives for using phones/media, which may impact on mental well-being. Third, the ecological model of social influence proposes three agents that shape well-being and technology use, including the school environment, home/ family and individual factors. 51 Finally, policy enactment and implementation process models 39 52 53 identify that school policy implementation effects will be shaped by social processes (eg, training, leadership, compliance, administration, family--school interactions).
Informed by our patient and public involvement (PPI) activities and school smartphone policy analysis, the components of the intervention (ie, restrictive school smartphone policies) are presented in table 1. In table 2 we have outlined variations in school smartphone policies and our classification of these variations into two school policy groups: restrictive (intervention) and permissive (comparator).

Study setting
The sampling frame comprises UK secondary schools (ages 11-19) located within a 100 mile radius of the recruiting centre in the West Midlands. Sixty-four local authorities are included in the sampling frame from the West Midlands, East Midlands, East, South East, South West and North West. The schools are situated in regions Open access of high and low levels of deprivation and in areas that have high 54 and low proportions of Black, Asian and Minority Ethnic (BAME) groups. 55 56 Schools other than state-funded mainstream schools (special schools, pupil referral units and independent schools) were excluded because it was expected that there would be additional influences on mental well-being. Schools that did not have an accessible smartphone policy and/or had different smartphone policies for the year groups 8 and 10, and schools with missing data that were required for the propensity score estimation (see sampling and participants) were also excluded (n=10,810). A total of 1345 secondary schools are included in our sampling frame; 1220 schools with policies classified as restrictive (intervention) and 125 as permissive (comparator).

Sampling and participants
To improve the comparability of the two school groups, stratified sampling based on propensity scores was employed. 57 We obtained routine data from the Department for Education on the following school characteristics: region, school type, urban or rural, total pupil roll size, Income Deprivation Affecting Children Index (IDACI), 58 inclusion of a sixth form, selective or nonselective admissions policy, religious affiliation and the proportion of pupils with the following characteristics: male, from BAME groups, English as an additional language, eligible for free school meals and special education needs. Propensity scores were calculated using restrictive or permissive school smartphone policies as the outcome and school characteristics as explanatory variables. Propensity score terciles were then used to create three groups with subsequent division by restrictive or permissive policy type, resulting in six distinct sampling groups. Schools in each group have been randomly ordered and are being invited sequentially to participate, aiming to recruit six to seven schools from each restrictive tercile and three to four from each permissive tercile to achieve a sample size of 20 schools with restrictive and 10 schools with permissive policies.
In participating schools, two classes of mixed ability year 8 (age 12-13) and year 10 (age 14-15) pupils are recruited. We are focusing on adolescents in this age range because of the age of onset and prevalence of mental illness, 5 and the prevalence of smartphone and social media use by adolescents aged 12-15. 7 14 59 Within this age range we may also be able to observe potential differences in smartphone/social media use relationships with wellbeing. 33 For example, year 8 pupils are likely to be newer users of phones/media and physical activity levels begin to decline at this age, particularly among girls. 7 13 In year 10, pupils are more likely to be established phone/media users, mental well-being tends to be lower, and this age group are approaching the peak onset of mental health conditions. 4 In addition to adolescent participants, the form teacher for each class recruited (or an equivalent teacher responsible for the class) and a member of senior The degree to which pupils and teachers adhere to the school policy, and parents/carers are in support of the policy varies across schools *MATs: non-profit companies that manage more than one academy. 76 MAT, multiacademy trusts; SLT, senior leadership team. Open access leadership team (SLT) who is responsible for the school smartphone policy are recruited from each school.

Sample size calculation
To account for the imbalance of schools in our sample that have permissive (n=125) and restrictive policies (n=1220), we are recruiting schools using a 2:1 (restrictive: permissive) ratio. The primary outcome of mental well-being will be measured using the Warwick-Edinburgh Mental Well-Being Scale (WEMWBS; score range=14-70). 60 To detect a mean difference in score of three points (considered the minimum clinically important difference 60 between the two school groups), assuming a SD of 6.8 61 and an ICC of 0.1 (a conservative estimate 62 ), with 90% power and 5% significance, we require 20 schools in the restrictive and 10 schools in the permissive smartphone policy groups, with an average cluster size of 39 (1170 pupil participants in total; 780 in the restrictive, and 390 in the permissive policy groups).
In each participating class, we aim to recruit a minimum of 19-20 pupils (67% if estimated class size n=30). In studies with multiple layers of clustering (here classes within schools), it is conservative to treat clusters within clusters as one larger cluster, which is the approach used here. 63 Recruitment School recruitment commenced in September 2022 and will continue until December 2023. A study advertisement has been emailed to all schools in the sampling frame. Following our propensity sampling approach, schools are then invited by post and email with a telephone follow-up. In participating schools, a School Liaison Member (SLM) of staff is identified and a school-university contract outlining expected commitments signed. Subsequently, a member of SLT responsible for the smartphone policy, and pupils and teachers from the year 8 and year 10 classes are recruited. £600 compensation is allocated to each school and a £5 voucher per pupil participant.
Parents/carers are provided with written detailed information about the study, what their child's participation will involve and how their child's data will be processed. Schools are asked to assist in the distribution of this information to parents/carers in different formats (eg, email, post, text messages, website, and so on).

Data collection
Data collection methods include self-administered surveys for pupils, teachers and a member of the SLT; accelerometer measured physical activity and sleep for pupils; and document analysis. All online surveys are completed using university-approved online survey software (REDCap). Table 3 provides an overview of outcomes, measures and timepoints of evaluation.

Pupil online survey
Pupils complete an online survey at one time point (in school time) that includes validated measures for mental well-being, anxiety and depressive symptoms, addictive social media use, motives for social media use, healthrelated quality of life and demographic variables (eg, age, gender) (table 3). Within the survey, pupils are asked to self-report data on their physical activity levels, sleep and phone/media use, and by using data from their iOS or Android apps, they additionally self-report data on time spent on their smartphone and social media apps (table 3). Within the survey, pupils are also asked to report on their knowledge and understanding of the school smartphone policy and compliance with the school smartphone policy. The survey is completed on encrypted tablets, using a portable Wi-Fi hub owned by the research team.
Secondary measure of mental well-being A second online survey (in school time) to measure mental well-being (table 3) is completed 4-8 weeks after pupils have completed the initial online survey. The SLM is sent an email link for pupils to complete.

Teacher online survey
Data are collected from the form teacher (or an equivalent teacher responsible for the class) for each participating pupil on: pupil attainment; classroom behaviour; and whether pupils are eligible for free school meals, have a special educational need or have English as an additional language (table 3). Within the survey, teachers are also asked to report on their knowledge and understanding of the school smartphone policy, compliance with the school smartphone policy and to report on the time they spend implementing the school smartphone policy (table 3). Teachers are sent the online survey by email.

SLT online survey
The SLT member responsible for the school smartphone policy is asked to complete survey questions on the SLT member's role; the school timetable and school policies; features of the school smartphone policy; perceived time spent by school staff developing and implementing the school smartphone policy; knowledge and understanding of the school smartphone policy; and compliance with the school smartphone policy. The SLT member is sent the online survey by email.

Accelerometers
Pupils are asked to wear a wrist worn GENEActiv accelerometer watch for 24 hours a day for the subsequent 7 days after completing the survey. 64 Watches are worn on the non-dominant wrist during all activities, including waterbased activities. Accelerometers are initialised to collect data in 100 Hz. Data will be analysed in R to produce physical activity and sleep outcomes (table 3).

Document analysis
School policy documents and handbooks related to smartphones, social media, pupil behaviour, mental health/ well-being, e-safety/IT policy and the wider school aims and ethos are collected. Documents related to the school Open access

Open access
timetable are also collected so that time spent in physical education, and time allocated for breaks and lunch can be quantified for each school.
Case study schools Six case study schools are being purposively sampled from the 30 participating schools in relation to two school characteristics: (1) smartphone policy type and duration to ensure a balance of restrictive (n=3) versus permissive policies (n=3), and a range of length of policy implementation in the six case studies; and (2) schools from low, medium and high area of deprivation, measured by the IDACI. An additional £300 compensation is allocated to case study schools. Across the case study schools, up to 36 focus groups (FGs) will be completed with adolescents (n=12), school staff (n=12) and parents/carers (n=12). Data collection is taking place following the school level data collection (ie, surveys, accelerometers and document analysis) and following obtaining written consent (or assent). In each case study school four to six FGs are completed: adolescents (n=2), school staff (n=1-2) and parents/carers (n=1-2). Each FG comprises four to six members and aims to balance gender and ethnicity (where possible). For adolescents, one FG is completed with year 8 (age [12][13] and one with year 10 (age 14-15) pupils. School staff FGs include SLT, school governors, teachers and support staff (admin, caretakers, teaching assistants). Parents/carers of pupils from within the school (excluding sixth form, age [16][17][18][19] are recruited.
FGs are led by research staff employing established elicitation and semistructured interview techniques. 65 66 Each FG lasts approximately 60 min and takes place in school or online. Data are collected from voice transcription.

Analysis of pupil outcomes
The primary analysis will examine the association between pupil mental well-being and school policy type (restrictive/permissive). Multilevel linear models will be developed, accounting for repeated measures, clustering of classes and schools and adjusting for the school-level variables included in the propensity score estimations alongside relevant individual-level sociodemographic variables. Secondary analyses will use the same modelling approach investigating differences in the secondary (behavioural and mental health) and intermediary (smartphone and social media use) outcomes between school policy groups. Differential association between school policy and the primary, secondary and intermediary outcomes will also be explored across: (1) socioeconomic position; (2) gender; and (3) ethnicity by including relevant interaction terms in the developed models.
Analysis of school policies, documents and websites School policies, documents (eg, school handbook) and relevant website content will be analysed using document analysis. 67 This will provide an overview of smartphone policy content, rationale and communication and how schools support pupils to use their smartphones and social media, as well as their mental well-being. We will adopt a comparative approach to compare restrictive and permissive policies.

Economic analysis
In view of the multiple outcomes of interest, complex nature of school budgets and emergent nature of economic evaluation of school-based interventions, a cost-consequence analysis will be conducted through the relevant data collected (table 3). This approach has been previously employed for school-based interventions. 68 In addition to the WEMWBS and CHU-9D outcomes, secondary behavioural and health outcomes outlined in the intervention logic model will be included ( figure 1). An exploratory cost-utility analysis from the payer (school) perspective will be conducted to compare incremental education costs and incremental Quality-Adjusted Life-Years associated with restrictive daytime smartphone use.
Qualitative/case study analysis Given the purpose of the case study method to understand complexity and situatedness, 69   Open access seems appropriate to analyse the case study data taking a dual approach. Thus, each case study will be analysed individually using a thematic analysis, 70 and then multiple case analysis will be adopted. 71 Following this approach, the findings will then be reported in detailed 'vertical' case reports on single schools, and in multiple case thematic analysis. 71 Patient and public involvement During research plan development, a teacher member of the investigator team advised on the study. We also consulted with adolescents and teachers through an online survey (teachers n=40) and five FG consultations (teachers n=11; adolescents n=20). In addition, we undertook analysis of school websites to determine the nature of school smartphone policies. Information gathered through these activities helped to inform and refine the study design, including categorisation of school policy types, research questions, data collection methods, primary and secondary outcomes and the logic model ( figure 1). The approach to PPI during the SMART Schools Study has been constructed based on National Institute for Health Research guidelines. 72 73 We are engaging with two groups throughout the study: (1) adolescents (from secondary schools); and (2) adults (from schools/ teachers, parents/families, local/national health organisations and policy). These groups are directly impacted by the research and are representative of key stakeholders who would act on the findings. We have one PPI lead and a PPI coapplicant, who are leading on PPI activities. Throughout the study there will be four online meetings per group which will focus on: (1) designing and managing study procedures; (2) undertaking the research; and (3) dissemination. The results and conclusions from each PPI group meeting will be reported using Guidance for Reporting Involvement of Patients and the Public (GRIPP2) checklist 74 for reporting PPI in research. For participation in the study, parents/carers of pupil participants are not asked for active consent but are given the opportunity to complete and return a form to opt their child out of taking part in the study. Pupil, teacher, SLT and parent/carer participants are asked to provide online or written consent (or assent).

ETHICS AND DISSEMINATION Ethical and regulatory considerations
We have also developed a safeguarding protocol for conducting research on mental health with adolescents. This safeguarding protocol has been developed in the context of our measure for depression (PHQ-9) that detects whether adolescents in our sample may have had self-harm or suicidal thoughts in the past 2 weeks. The protocol outlines key procedures during recruitment, data collection and debriefing periods to safeguard pupils, and include steps related to confidentiality and establishing efficient, secure and effective communication pathways between the research team and school leaders.
Study sponsorship is provided by the University of Birmingham, with provision of research related costs supported by the National Institute for Health Research (NIHR). Data management and storage is compliant with the University of Birmingham's policies and procedures. Participant data from online surveys and other sources will be pseudo-anonymised, stored on a University of Birmingham secure server and retained for 10 years through the University's Research Data Archive. Data from this study will be owned by the University of Birmingham.
Study oversight will be guided by an independently chaired Study Steering Committee (SSC) and a Data Management and Ethics Committee (DMEC). A study monitoring plan has been developed and agreed on by the SSC and DMEC. The current protocol has been reviewed and agreed on by all members of the SSC and DMEC.
Dissemination Dissemination activities will be co-produced with our PPI participants. We have planned the following dissemination outputs: policy briefings and research summaries for agencies on the impact of school smartphone policies on adolescent mental well-being; guidelines and resources for schools on the characteristics of school smartphone and social media policy implementation that positively influence mental well-being; blogs, podcasts, videos and infographics to raise awareness and understanding of relationships between smartphones and social media and mental well-being; an NIHR public report that summarises the main project findings; and peer review and open access publications focused on the main study findings.
Open access DMEC and members of our PPI groups for their advice on the study protocol and conduct.
Contributors VG and MP are chief investigators and obtained funding for the study. VG and MP led on protocol development with contribution from all named authors in the following ways: PA advised on study design and conduct; MM and PP contributed to MH assessments and analysis; SF lead on physical activity and sleep data collection and analysis; HA-J lead on economic evaluation; AJS advised on the study design and statistical analysis plans; BM developed the statistical analysis plan; KJ advised on school recruitment and data collection procedures and MW and GE lead on PPI. GW, VG and MP drafted the manuscripts and all authors reviewed and approved the final draft.
Funding This project is funded by the NIHR Public Health Research programme (NIHR131396). The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.
Disclaimer This project is funded by the NIHR Public Health Research programme (NIHR131396). This paper presents independent research supported by the NIHR Birmingham Biomedical Research Centre at the University Hospitals Birmingham NHS Foundation Trust and the University of Birmingham. The views expressed are those of the author(s) and not necessarily those of the NIHR, NHS or the Department of Health and Social Care.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting or dissemination plans of this research. Refer to the Methods section for further details.
Patient consent for publication Not applicable.
Provenance and peer review Not commissioned; peer reviewed for ethical and funding approval prior to submission.
Open access This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/ licenses/by/4.0/.