Article Text

Original research
Exploring support, experiences and needs of older women and health professionals to inform a self-management package for urinary incontinence: a qualitative study
  1. Yu Fu1,
  2. Cath Jackson2,
  3. Andrea Nelson3,
  4. Heather Iles-Smith4,5,
  5. Linda McGowan6
  1. 1 Department of Primary Care and Mental Health, University of Liverpool, Liverpool, UK
  2. 2 Valid Research Ltd, Wetherby, UK
  3. 3 School of Health and Life Sciences, Glasgow Caledonian University, Glasgow, UK
  4. 4 School of Health and Society, University of Salford, Salford, UK
  5. 5 Corporate Nursing, Northern Care Alliance NHS Group, Salford, UK
  6. 6 School of Healthcare, University of Leeds, Leeds, UK
  1. Correspondence to Dr Yu Fu; yu.fu{at}


Objectives Many women attempt to manage urinary incontinence (UI) independently with variable success while health professionals may be unaware of their needs. This study aimed to (1) understand older women’s experiences of UI, their self-management strategies and support needs; (2) explore health professionals’ experiences of supporting women and providing relevant services and (3) combine their experiences contribute to development of a theory-based and evidence-based self-management package for UI.

Design Qualitative semi-structured interviews were conducted with 11 older women with UI and 11 specialist health professionals. Data were analysed independently using the framework approach, then synthesised in a triangulation matrix to identify implications for content and delivery of the self-management package.

Setting Community centres, community continence clinic and urogynaecology centre of a local teaching hospital in northern England.

Participants Women aged 55 years and over who self-reported symptoms of UI and health professionals delivering UI services.

Results Three overarching themes emerged. Older women see UI as a ‘fact of life’ but many struggle with it: women typically considered UI as part of ageing yet expressed annoyance, distress, embarrassment and had made significant lifestyle changes. Access to information and limited high-quality professional support: health professionals provided specialist UI care and information. Yet less than half of women accessed specialist services, those who had, highly valued these services. ‘Trial and error’ with different self-management strategies: women had tried or were using different strategies (continence pads, pelvic floor exercises, bladder management and training, fluid management and medication), with mixed success. Health professionals provided evidence-based, personalised support and motivation.

Conclusions Findings informed the content of the self-management package that focused on providing facts, acknowledging challenges of living with/self-managing UI, sharing others’ experiences, using motivational strategies and self-management tools. Delivery preferences were independent use by women or working through the package with a health professional.

  • urinary incontinences
  • qualitative research
  • urogynaecology

Data availability statement

Data are available upon reasonable request. The data that support the findings of this study are available from the corresponding author upon reasonable request.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • In-depth exploration of both women’s and health professionals’ experiences of UI, support and self-management facilitated triangulation of perspectives.

  • Employing a ‘target population-centred’ approach to intervention development driven by the views and preferences of the people the intervention serves is more likely to be acceptable and effective.

  • Women participants all spoke English, their views and experiences and the resultant self-management tool might not represent those who do not speak English.


Urinary incontinence (UI), ‘the complaint of involuntary loss of urine’, places physical, psychosocial and economic burdens on individuals and their families.1 2 In the UK, over 14 million people are affected by bladder control problems.3 The prevalence is estimated as up to 40% for women over 21 years and increases with age.4 Women with UI can experience feelings of isolation and embarrassment, making it an underreported and undertreated condition. Only a quarter of affected women seek medical help, of which, less than half receive treatment.5 There are associations between untreated UI and generally poor quality of life, falls and fractures, and anxiety and depression, leading to increased morbidity and healthcare costs.6–8

Despite clinical guidance on the treatment and management of UI from international and specialty organisations,9–11 UI remains poorly prioritised in routine healthcare practice.12 Associated stigma, perceiving UI as inevitable, poor awareness by health professionals and low expectations of treatments may render older women silent on the topic. Many women affected try to cope on their own with variable success.13–17

Self-management has been recommended for women with UI,18 to help quality of life and increase confidence.16 17 This refers to actions taken by individuals to manage the symptoms, treatment, physical and psychosocial consequences and lifestyle changes inherent in living with a condition.19 However, there are gaps between women’s experiences of managing UI and support provided by health professionals for self-management. Evidence suggests that women do not have sufficient knowledge or understanding of self-management principles and poor access to services,20 while health professionals are unaware of the needs of older women and how to best support them.18 The COVID-19 pandemic in 2020/2021 likely exacerbated these challenges by creating difficulties attending appointments in person and closure of some support services.

Our project set out to develop and evaluate a theory-informed, tailored self-management package for older women living with UI, guided by the Medical Research Council (MRC) framework for developing and evaluating complex health interventions.21 Our systematic review22 and feasibility trial of the package have been published.23 We present here a qualitative study, which was carried out in parallel with the systematic review22 during the intervention development phase. We employed a ‘target population-centred’ approach24 to our intervention development whereby the intervention is based on the perspectives of the people who will use it, namely, women with UI and relevant health professionals.


The study aims were to (1) understand older women’s experiences of living with UI, their self-management strategies and support needs; (2) explore health professionals’ experiences of supporting women and providing services to manage UI and (3) use this combined insight to contribute to the development of a theory-base and evidence-based self-management package.



A qualitative study using semi-structured face-to-face interviews, guided by the standards for reporting qualitative research checklist.

Patient and public involvement

An advisory group was set up prior to commencing the project. Three older women with UI and one specialist nurse working in a local continence clinic formed the group. Bi-annual meetings and regular phone/email contact led by YF ensured their ongoing input into all stages of the project: reviewing sampling and recruitment strategies, interview topic guides, preliminary research findings and developing dissemination strategies. The initial draft of the self-management package (informed by the systematic review,22 qualitative interviews and a consensus process) was co-developed with the group through frequent face-to-face meetings and phone/email discussions.

Setting and participants

The study setting was a northern UK city. We aimed to interview 10–15 women aged 55 years and over, who self-reported symptoms of UI and could communicate in English, and 8–10 health professionals delivering UI services. Exclusion criteria for the women were cognitive impaired or UI caused by neurological disease.

To recruit women, flyers were posted in the community centre of a local forum for older people. To recruit health professionals, study information was circulated by the clinical leads of a community continence clinic and urogynaecology centre of a local teaching hospital. Consecutive sampling was used for women, and maximum variation sampling was used to ensure a mix of health professions.

Women and health professionals interested in taking part either telephoned or emailed the lead researcher (YF), who then arranged a time and place for the interview.

Data collection

Face-to-face interviews were conducted in women’s homes and at health professionals’ workplace. Two researchers (YF and Jill Edwards) experienced in qualitative methods and health services research collected the data. Written informed consent was collected before each interview.

Interviews with women explored their specific condition and symptoms, methods and skills they use to manage their UI, experiences of coping with UI and expectations of effective self-management and desired support. Health professional interviews discussed common problems relating to UI, skills they use to support patients’ self-management, their experiences of supporting patients and expectations of delivering effective self-management. Topics were informed by the findings of the systematic review22 with advisory group input. They were expected to elicit important detail for sections of the self-management package. Interview guides ensured consistency, although the format was flexible to enable participants to generate naturalistic data on what they viewed as important. Both guides (see online supplemental files 1 and 2) were piloted, and adaptations made to improve clarity. Women’s interviews lasted between 28 min and 95 min and health professionals’ between 14 min and 75 min.

Supplemental material

Supplemental material

Data analysis

Interviews were audio-recorded and transcribed verbatim, and data were analysed using framework analysis,25 facilitated by Microsoft Excel V.2010. Analysis was led by CJ with input from other researchers (YF, LM and AN) to ensure rigour.

Interviews from the two participant groups were first analysed independently. A thematic framework for each group was developed based on the topic guide and emergent themes from subsamples of transcripts (three women and two health professionals), then piloted with another transcript per group, before finalising. The frameworks were then applied to the interview data. Summaries of participant responses and verbatim quotes were entered. Charted data were then reviewed and interrogated to compare and contrast views, seek patterns, connections and explanations within the data; and descriptive findings were produced for both data sets. These were reviewed and refined through full team discussion, with agreement on three overarching themes. Finally, a matrix26 was used to triangulate the key findings for each theme and their implications for the content of the self-management package. This was presented to the advisory group for their input.


Eleven women and 11 health professionals were interviewed, after which data saturation was achieved for both groups with no new perspectives emerging.

Women aged 56–87 years (mean: 78.7 years) had experienced UI mean of 13.5 years. Most were white British, suffered stress or urge UI, were retired and had coexisting medical conditions. Over half lived alone; two were living in residential/sheltered accommodations (table 1).

Table 1

Women participant characteristics (n=11)

Four health professionals (table 2) worked within community continence clinics and seven worked in a urogynaecology centre. All had completed specific UI training.

Table 2

Health professional participant characteristics (n=11)

Three over-arching themes emerged: (1) older women see UI as challenging and a ‘fact of life’, (2) access to information and limited high-quality professional support and (3) ‘trial and error’ with different self-management strategies.

The third theme is most extensive given our focus on developing the self-management package. We present only women’s accounts for theme 1, and both women’s and health professionals’ views on the other two themes. If only women or health professionals spoke of a specific issue, this is evident because only those data are included.

Theme 1. Older women see UI as challenging and a ‘fact of life’

The women had lived with UI from 12 months to 50 years and a third for over 25 years. Symptoms had worsened for those with a lengthy experience of UI. From their accounts, it was evident that most had urge UI, a small minority had stress UI, while a few had mixed UI. Urge UI was depicted as desperately needing the toilet, often without warning. While women with stress UI, spoke of having ‘an accident’ after they sneeze, cough, laugh or do anything ‘jerky’. Both types were unpredictable, could result in ‘an accident’ and happen several times a day. Furthermore, most women were waking to urinate one to three times most nights. Again this had increased in frequency over time and could result in ‘an accident’.

Many women attributed UI to ageing, seeing it as ‘a fact of life’ (quote 1, table 3). Other suggested causes were health issues, for example, a throat issue that prompted coughing or medication such as taking water tablets for swollen legs. Less common attributions were having children, hormones while approaching menopause or bladder issues. Two women believed historical toilet habits were a potential cause (quote 2, table 3).

Table 3

Illustrative quotes for women

There were two contrasting views on living with UI. Most women were outwardly bothered by their UI, describing it as annoying, embarrassing, upsetting and inconvenient (quote 3, table 3). They made significant changes to their lives often giving up enjoyable activities such as going for walks, exercise classes, theatre or concerts, or modifying activities: shortening their walks, not having a drink or having a single room on holiday instead of sharing with a friend. Related to this was the significant worry about finding public toilets in time (quote 4, table 3).

A minority spoke of seeing their UI as ‘part of life’ (P6, stress UI) typically attributing this to being lucky and determined. They tended to say that they were not worried, had learnt to live with it and did not dwell on it too much (quote 5, table 3). Often this was because they had other issues or health conditions, for example, bereavement, a neurological condition or skin infection, that concerned them more.

There were also mixed accounts of discussing UI with others. A third of women spoke to friends and family, but not in any detail. The people they spoke to were often someone they knew who also had UI or who was seen as knowledgeable about UI, for example, a friend’s daughter who was a gynaecologist. Another third of women kept it private. Indeed, two women had not discussed UI with their husbands. UI was seen as ‘not a pleasant topic’, ‘embarrassing and a ‘British thing’ not to talk about bodily functions (quote 6, table 3). Furthermore, talking about it ‘doesn’t make it go away’ (P6, stress UI).

Finally women were typically modest in their aspirations for the UI, commenting that it was unlikely they would be cured. Instead they hoped to have better control and not to have to think or worry about having an accident on a daily basis and to find a way to live with it (quotes 7 and 8, table 3).

Theme 2. Access to information and limited high-quality professional support

Women did not appear to have sought or received much written information about UI. Comments included not knowing where to look, receiving no or limited information from their general practitioner (GP) and limited helpful advice on the internet (quote 9, table 3). Conversely health professionals working in the continence clinic and urogynaecology department alike reported providing a range of leaflets produced in-house, by local UI support groups, the NHS and British Society of Urogynaecology.

Few women had accessed professional support, with only half having consulted their GP, typically when symptoms started or worsened. Reasons included: feeling embarrassed, ‘it is not life threatening’ (P8, stress UI), not wanting to bother the GP, they were managing it OK and would see the GP if it got worse, perceiving there was insufficient time in an appointment to fully discuss the issues, and assuming the GP could not help (quote 10, table 3).

Those who had seen their GP reported that the doctor had provided leaflets, recommended pelvic floor exercises, pads, prescribed medication and signposted to specialist services. There was a clear desire among women to access someone who could spend time discussing their condition in detail (quote 11, table 3).

Even fewer women had attended specialist services, either at the continence clinic or urogynaecology department where time to talk was available. Indeed, several health professionals observed that older women often assumed that these services were not available to them because of their age, seeing their UI as ‘normal’ and nothing could be done (quote 1, table 4).

Table 4

Illustrative quotes for health professionals

Health professionals described multiple referral pathways to the continence clinic, most commonly via GPs, but also social workers, neighbourhood teams and other specialist services such as podiatry, memory and psychological (quote 2, table 4). People could also self-refer, which a healthcare assistant saw as an important option for those who reluctantly seek help from anyone (quote 3, table 4). Referrals to the urogynaecology department were from GPs or the continence clinic. Both services offered a detailed assessment, used to inform appropriate treatment typically including a self-management plan.

Women’s assessment of this specialist support from both these services was very positive. The nurses at the clinics were unanimously favourably reviewed and seen as well placed to have the time and expertise to support people as ‘they’ve heard so many people’s stories’ (P6, stress UI). Indeed, a medical consultant and nurse specialist both recognised the value of providing an opportunity for women to talk about their UI (quote 4, table 4).

Theme 3. ‘Trial and error’ with different self-management strategies

Women described using a mix of self-management strategies, with varying degrees of adherence and success. Relatedly, the consensus among health professionals was that UI self-management is challenging, and while some women are receptive and motivated to make lifestyle changes, others are not commonly linked to the misperception that nothing can be done as UI is ‘normal’ at their age. The professionals were using multiple strategies to motivate women: managing expectations, jointly setting realistic goals, encouraging them to take ownership, monitoring progress, providing feedback on small improvements and sharing success stories (quote 5, table 4). A good relationship with women was seen as critical to success. The continence clinic team tried to provide continuity and ensure that women saw the same professional each time. Two nurses who had UI themselves shared this to demonstrate that they knew how it felt (quote 6, table 4).

Using continence pads

Continence pads were the most common self-management tool. One continence clinic nurse, estimated that this accounted for 90% of her work, while another commented that 80% of clinic patients were using pads. All but one of the women were using pads, with varied motives. Some wore a pad ‘just to be on the safe side’ (P1, stress UI), usually when going out or at night. For others, the pad was more of a necessity to ‘stop the embarrassment of wetting my knickers’ (P2, stress UI).

Overall pads were perceived as effective although several women mentioned discomfort due to skin soreness, pad movement and being bulky and rough. This prompted some to restrict their use or, deliberately wear ‘tight pants’, tights or trousers to prevent pads movement. A common concern was embarrassment associated with the pads smelling, being visible and not having anywhere to immediately change and dispose of a soiled pad when away from home (quote 12, table 3).

The continence clinic team perceived that some women used pads because they were an easier option than other self-management strategies. While pads were available from the clinic for as long as the women needed them, there was a clear view that pads would not be given to people who only attend to access free pads or ‘who just cannot be bothered to do self-management’ (HP3, advisor, continence clinic). Indeed, only two women accessed pads free of charge, at the clinic. The rest bought them, often recommended by their GP or on TV adverts. Half did not see cost as an issue, because money was not a concern, they did not need to buy many or they perceived them to be cheap. However, three women were affected by the cost, seeing them as ‘quite expensive’ (P3, urge UI), particularly if buying them on a regular basis or when needing the larger, specialist ones. Strategies to save cost were putting toilet paper on top of the pad or using a scrunched-up hanky (quote 13, table 3).

Understanding the correct type/size of pad to use as well as how to wear it was seen as important by a physiotherapist and consultant. Two women concurred that it would be useful to have this information, explained clearly, and in large writing on packaging.

Pelvic floor exercises

No women were currently doing pelvic floor exercises on a regular basis, although some had previously performed them when their stress UI had first started or during worsening symptoms. Multiple barriers to performing these exercises were reported, often linked to a lack of understanding; for example, uncertainty about technique, perceived lack of muscle strength or finding the exercises too tiring. Other barriers were forgetting and not having time. Women reported variable success and effectiveness of the exercises, further affecting motivation (quote 14, table 3).

These reasons were all familiar to health professionals (quote 7, table 4). All were teaching pelvic floor exercises to women with stress UI, believing them to be effective for most women when performed correctly over time. However, they recognised the challenge of staying motivated when improvement might take 3 months.

Bladder management and training

Two-thirds of women were using their own strategies to manage their bladder and UI, including going to the toilet before leaving home, before going into meetings/concerts and going as many times as possible in the evening to try to stay dry (quotes 15 and 16, table 3). Other approaches were to only go when your bladder feels full or to go every 2 hours ‘like they do with patients in care homes’ (P6. stress UI).

Several health professionals were advising women with urge UI to train their bladder (ie, allow it to fill up before going to the toilet) and to complete a fluid chart. The physiotherapist described following the guideline of 6 weeks of bladder retraining prior to considering using medication. The healthcare assistant explained high variability regarding success dependant on how long women has had a particular habit. She considered a reduction in the number of times someone goes, from 12 to 11 times a day, as important progress.

Managing fluid intake

Two-thirds of women had changed their drinking habits to try to manage their UI. Strategies mentioned were drinking less or no caffeine, particularly coffee, drinking less or nothing after 18:00 and not drinking before going out to a new place (quote 17, table 3).

Health professionals provided verbal and written advice on fluid intake: to drink 1.5–2 litres a day, drink water, reduce/avoid caffeine, fizzy drinks and alcohol and avoid drinking 1–2 hours before bed if leaking at night. Conversely, sometimes the advice was to drink more fluids to change the urine consistency and avoid infections. Fluid management was seen by health professionals as effective. Indeed, most women who had tried this confirmed that these changes had worked. A physiotherapist commented that some women find it difficult to break long-established drinking habits (quote 8, table 4).

Taking medication

Medication (anticholinergics and antimuscarinics) was seen as an option by the health professionals for patients with urge UI, when self-management had not worked. Half the women had been prescribed medication (eg, oxybutanin, solifenacin, mirabegron, etc), usually by their GP. The key barrier to adherence, identified by several women and health professionals, was ‘pretty horrible’ (P7, urge UI) side effects, for example, dry mouth, tight chest, blurred vision, bad indigestion, constipation and headaches. To try to address this, women attending the continence clinic would be prescribed a low dose, spray or patch, which might be less effective but has fewer side effects.

There were mixed views among women on the effectiveness of the medication, ranging from completely ineffective to highly successful (quotes 18 and 19, table 3). The nurses in the continence clinic acknowledged that medication works for some patients and not others.

Triangulating women’s and health professionals’ views to inform the content and delivery of the self-management package

Table 5 summarises the key findings for each theme for women and health professionals, with the implications for, and final content of, the self-management package (full table in online supplemental file 3). Both groups focused on providing facts, acknowledging the challenges of living with/self-managing UI, sharing others’ experiences and providing motivational and self-management strategies. Women and health professionals had mixed views on how they would prefer to use a self-management package. Some saw it as a tool for women to use independently, while others believed it should be used with health professional support.

Supplemental material

Table 5

Triangulating women’s and health professionals’ views to inform the content and delivery of the self-management package


This qualitative study offers valuable insight into women’s lived experience of UI, their self-management challenges and needs for support, alongside health professionals’ perspectives on providing this support. Together with a systematic review,22 these findings informed the content and delivery of a theory and evidence-based self-management package for UI.

Similar to previous studies,27 28 most women interviewed considered UI to be normal at their age and still a stigma reported in this study attached to UI and embarrassment about disclosing UI. Reasons include childbirth, the ageing process and lack of knowledge, which were not only reported among people living with UI but also caregivers who experienced the extra burden and affected the quality of life in previous studies.29 Women in this study described their symptoms as upsetting, embarrassing and inconvenient, leading to changes in their behaviours and routines. This prevented many women from talking to peers or family members and seeking support. However, our study found that all women expressed a strong desire to access specialist services to facilitate detailed discussion about their condition. Women’s lack of awareness of referral pathways often led to poor access to specialist services. This underlines the importance of health professionals using a proactive approach when consulting and assessing patients and the need for training on handling sensitive topics. Some regions of the UK support NHS self-referral services, although these are patchy and are not consistent in their processes. Publicising these services through public campaigns may increase patients’ self-referral and promote knowledge of local resources and health services.

This study explored a range of self-management strategies and participants reported varied outcomes from these strategies. With limited access to existing services, women were familiar with the concepts of some self-management strategies provided by health professionals, including pelvic floor muscle training, bladder training and control fluid intake that are recommended by guidance.30 31 However, no participants consistently performed these activities with barriers reported, including forgetting, lack of time or limited progress. Instead, in line with other studies,32 women commonly used continence pads without rigorous evidence.11 33 This suggests low level of motivation to self-manage UI and the potential psychological dependence on pads developed in women with UI34 despite the cost and discomfort. There is a need to develop and maintain motivation in conjunction with self-management with optimal interventions, which may require empowerment approaches underpinned by motivational interviewing, concordance skills and cognitive behavioural therapy from health professionals.35

Guided by the MRC framework for developing and evaluating complex interventions,36 the identified challenges and needs for self-management of UI in this study together with the systematic review published22 formed the content of a self-management package for UI in older women that was proven to be feasible and acceptable.23 Given there are only a few self-management programmes developed either focusing on a single type of UI37 or a specific self-management strategy, for example, pelvic floor muscle training,38 this study provided an evidence base, from both women and health professionals’ perspectives, of the development of multifaceted self-management intervention that factored in awareness building, description of self-management behaviours and promotion of confidence and motivation in older women.

The key strength of this study was including both women and health professionals, facilitating systematic triangulation of views and experiences of both user groups to inform the development of the self-management package. This ‘target population-centred’ approach24 to intervention development using the views of the people who will be using the SM package should increase the likelihood of acceptability and feasibility. A limitation is that we recruited women from local voluntary organisations who could converse in English. We did not purposively sample on ethnicity, and only have one participant from a black, Asian and minority ethnic background. It is uncertain if participants’ accounts would be different to women recruited in other ways, with a different ethnic background or those who do not speak English. However, we achieved data saturation, captured a good diversity of views that were consistent with the existing literature.27 28 32 This, and the rigour of the study design and conduct, give us confidence in our findings and generalisability (as a qualitative concept)25 to other older women living with UI.


Older women do not actively seek early help for their UI and UI management. Barriers to self-management of UI identified include relying on pads, lack of instruction and guidance, lack of motivation and poor adherence. These factors are important to be taken into account and addressed in developing future evidence-based intervention aiming to raise awareness, improve self-management behaviour and build confidence and motivation in older women with UI. The self-management package for UI that was developed from this work has since been evaluated in a feasibility trial with 50 women living with UI.23 The package was highly acceptable and showed a positive trend in reducing the UI severity, symptoms and anxiety. Women further reported higher subjective improvement and self-efficacy. The next step would be to support the implementation of this package in routine practice.

Data availability statement

Data are available upon reasonable request. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants. Ethical approval was granted from the University of Leeds (HREC16-016) and the UK Health Research Authority (18/HRA/0142).


We are grateful to the women and health professionals who participated in this study. We thank Jill Edwards of the University of Leeds for conducting some interviews, and Janet Ashworth and Claire Hitchings of Valid Research for coding the interview data.


Supplementary materials


  • Twitter @maggieyf, @hiles_smith

  • Contributors YF, AN and LM led the study design. YF led recruitment (supported by HI-S) and data collections contributed by another researcher. CJ led data analysis contributed by YF, HI-S and LM. YF and CJ led writing of the manuscript contributed by others. All authors contributed to interpretation and approved the final version of the manuscript. YF is responsible for the overall content as the guarantor.

  • Funding This work was supported by the Leeds Benevolent Society for Single Ladies, a Charitable Incorporated Organisation (registered charity number: 1155794).

  • 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.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.