Article Text
PDF
Abstract
Introduction There is a global interest in cancer immunotherapy. Clinical trials have found that one group, immune checkpoint inhibitors (ICIs), has demonstrated clinical benefits across various cancers. However, research focused on the experiences of people affected by cancer who have undergone this treatment using qualitative methodology is currently limited. Moreover, little is known about the experiences and education needs of the healthcare staff supporting the people receiving these immunotherapies. This study therefore seeks to explore the experiences of using ICIs by both the people affected by cancer and the healthcare professionals who support those people, and use the findings to make recommendations for ICI supportive care guidance development, cancer immunotherapy education materials for healthcare professionals, cancer policy and further research.
Methods and analysis Patient participants (n=up to 30) will be recruited within the UK. The sample will incorporate a range of perspectives, sociodemographic factors, diagnoses and ICI treatments, yet share some common experiences. Healthcare professionals (n=up to 15) involved in supporting people receiving immunotherapy will also be recruited from across the UK. Data will be generated through in-depth, semistructured interviews. Reflexive thematic analysis will be used to obtain thorough understanding of individual’s perspectives on, and experiences of, immunotherapy. Study dates are as follows: December 2019–March 2022.
Ethics and dissemination The research will be performed in accordance with the UK Policy for Health and Social Care Research and Cardiff University’s Research Integrity and Governance Code of Practice (2018). The study received ethical approval from the West Midlands and Black Country Research Ethics Committee in October 2019. Health Research Authority and Health and Care Research Wales approvals were confirmed in December 2019. All participants will provide informed consent. Findings will be published in peer-reviewed journals, non-academic platforms, the Macmillan Cancer Support website, disseminated at relevant national and international conferences and presented via a webinar. The study is listed on the National Institute for Health Research (NIHR) Clinical Research Network Central Portfolio.
- qualitative research
- oncology
- education & training (see medical education & training)
- immunology
- adult oncology
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: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- qualitative research
- oncology
- education & training (see medical education & training)
- immunology
- adult oncology
Strengths and limitations of this study
Few qualitative studies have explored people’s experiences of immunotherapy and its associated supportive care, with no studies exploring these experiences known to be reported in the UK context.
This original qualitative study has been designed to build on existing knowledge derived predominantly from clinical trials and capture rich, detailed insight into aspects of individual’s experiences of cancer immunotherapy in the UK to develop suggestions for improving person-centred care from those receiving, prescribing and supporting treatments.
This work samples healthcare professionals from within and outside oncology, and as a result explores the unique experiences of professionals who are expected to have knowledge and experience of managing immune checkpoint inhibitor toxicities and providing safe and effective person-centred supportive care.
The sample size is appropriate given that qualitative research does not search for a representative sample but to give breadth, depth and rich information for analysis. However, there is the possibility of selection bias in that as participants are self-selecting they may be particularly motivated to participate in the study.
The study coincided with the global COVID-19 pandemic and the introduction of physical distancing measures in the UK may have an impact on the data generated.
Introduction
Globally, the rapidly evolving field of cancer immunotherapy1 is substantially transforming outcomes for some people with advanced solid and haematological cancers. As populations age and grow, cancer detection improves and treatments advance, and more people will live with cancer.2 3 This, together with the increasing use of some immunotherapies earlier in disease trajectories as the standard of care, means more people will receive these treatments as part of their management pathway. However, immunotherapies have risks. Indeed, there is real potential for treatment-related adverse events, some of which can be severe and even life-threatening.4–6
Arguably, immune checkpoint inhibitors (ICIs), one type of cancer immunotherapy, constitute one of the most important developments in cancer therapeutics in recent years, bringing enhanced survival hope to patients with advanced cancer and transforming the standard of care.7 8 ICIs include the anticytotoxic T lymphocyte antigen-4 (anti-CTLA-4; eg, ipilimumab), antiprogrammed cell death-1 (anti-PD-1; eg, nivolumab and pembrolizumab) and antiprogrammed cell death-1 ligand-1 (anti-PD-L1; eg, atezolizumab and durvalumab) and monoclonal antibodies, which revive antitumour immune responses and restore anticancer immunity by targeting immune checkpoints and blocking specific proteins in cancer cells which turn the immune system off.7 9
Clinical trials of ICIs in people with advanced cancers, including, for example, non-small cell lung cancer, metastatic and unresectable melanoma and recurrent or metastatic head and neck cancer, have demonstrated clinical benefits.10–20 Indeed, when evaluated against traditional comparator treatments, for example, chemotherapy, consistent improvements in progression-free, treatment-free and overall survival have been reported in both treatment-naive and previously treated patients.10–17 Furthermore, pembrolizumab and nivolumab maintained or even improved quality-of-life (QoL) to a greater degree than comparators.18–20
ICIs are usually delivered intravenously within an oncology day hospital setting in treatment cycles ranging between 2 and 6 weeks and lasting for up to 2 years. As the targeting of immune cells generates an autoimmune response, immune-related adverse events (irAEs) are not uncommon. When used alone (eg, ipilimumab) and in combination (eg, ipilimumab and nivolumab), ICIs have also produced severe and unique treatment-related adverse events,4–6 which are very different to those associated with traditional cancer therapies and can generate a considerable negative impact on individual’s QoL.21 22 Indeed, patients have reported a range of irAEs including endocrine, gastrointestinal, respiratory, dermatological and musculoskeletal problems.23 24 Furthermore, by the end of 2018, in excess of 13 000 cancer immunotherapy irAEs in 18 countries had been reported, with more than two-thirds of recorded cases connected with ICIs.25
Compared with the effects of some chemotherapy regimens, irAEs may be relatively minor, manageable and reversible with timely administration of immune-modulating interventions such as corticosteroids. However, irAEs can also be unpredictable, severe and challenging to manage, arise months after treatment initiation,26 persist once treatment has ended and even arise several months and years after treatment has been completed.27–29 Furthermore, while uncommon, fatalities due to the toxic effects of ICIs have been reported.30 Indeed, while recognising the limitations of the WHO pharmacovigilance database (Vigilyze), a comprehensive analysis of entries between 2009 and 2018 identified 613 fatalities due to ICI irAEs. Most frequently, deaths were due to colitis (70% of the anti-CTLA-4 deaths (n=193) and 37% of combination therapy (n=87)) and pneumonitis (35% of the anti-PD-1/anti-PD-L1 monotherapy deaths (n=333)).30
Prioritising the enhancement of peoples’ experiences of care, treatment and support, together with meeting individual’s needs during treatments and recovery are central to the cancer policy commitments of the UK’s central and devolved governments.31–33 In phase III cancer trials, patient-reported outcomes, notably health-related QoL, have provided invaluable insights into treatment impact on individuals (see, eg, previous studies34–36). Health-related QoL assessment has been prominent in multiple ground-breaking international phase III randomised controlled trials of checkpoint inhibitors.18 37–40 Yet, notwithstanding the positive results from many randomised trials, treatment experiences of patient populations in real-world settings, as opposed to trial settings, may be different. Certainly, given the potentially prolonged nature of immunotherapy treatment delivery, together with the possibility of unique immune-related adverse events in the short, medium and long terms, there is potential for a substantial burden of treatment and ‘collateral damage’ which may adversely impact on individual’s lives, health and well-being.
Although the use of ICIs in clinical practice is in its infancy, several ICIs have now been approved for treating a range of cancers and are used across cancer centres in the UK and beyond. The emergence of exciting, new and ‘cutting edge’ ICI therapies as standard care outside of clinical trials has been shown to engender hope and optimism among people with advanced cancers.24 26 41 42 In addition, these perceptions may outweigh much of the perceived risk of undergoing treatment. Some recipients feel sufficiently well and motivated to resume a degree of normalcy in their everyday lives.23 However, findings from international studies have also highlighted the lasting and profound existential-related, social-related, financial-related, treatment-related and disease-related uncertainities, adverse effects on physical and emotional health and well-being, and a perceived need among some for enhanced informational support and guidance.23 24 26 41 42
Shared decision-making, where clinicians support patients to share responsibility for decisions, based on the best available evidence and considering the strengths and risks of various treatment options,43 positively affects patients’ treatment experiences44 and 45QoL in cancer care. However, it has been noted that existing cancer decision-making pathways for some patients focus on clinical management of disease rather than patients’ preferences and priorities for cancer treatment.46 Limited literature examines the immunotherapy decision-making process, for patients across all tumour sites and at various stages of disease. In one study, very few patients communicated a good understanding of immunotherapy, particularly the potential effectiveness of treatment and the possibility of experiencing treatment-related side effects.41 Patients receiving immunotherapy and their informal carers have also experienced uncertainty related to communication and treatment decision-making.24 47 Indeed, patients reported feeling hampered by a lack of clear information,23 and carers experienced unclear communication regarding immunotherapy treatment.47 Findings from a recent study48 indicated that most recommendations for ICI treatment were made by physicians, though patients generally preferred to have the final say regarding treatment commencement. Further enquiry is therefore required to explore both patients’ and healthcare professionals’ experiences of immunotherapy treatment decision-making in the context of the UK.
To the best of our knowledge, there has been no published empirical investigation of people’s experiences of the ICI treatment journey from a UK perspective. Moreover, the experiences of healthcare professionals’ who deliver cancer immunotherapy and support people receiving these treatments, both within and outside specialist oncology settings in the UK, appear to be absent from the literature, as are their education and training needs. The paucity of existing research within the UK context is a concern, for it is a barrier to the effective planning and delivery of high-quality safe and effective person-centred care and support across the cancer immunotherapy treatment journey and beyond, particularly as patients experiencing irAEs may present to acute medical or emergency department admissions rather than oncology services.
Further investigation of patients’ experiences of immunotherapy treatment and support, as well as healthcare professionals’ experiences of associated care delivery and training needs, is therefore imperative to identify gaps in knowledge, improve understanding and enhance patients’ health outcomes and experiences across care settings, and strengthen healthcare professionals’ cancer immunotherapy education and training. Furthermore, data from a UK perspective are required to ensure contextually relevant immunotherapy education and training interventions are developed.
Aim and research questions
This study aims to investigate people’s experiences of ICI treatment and associated supportive care and healthcare professionals’ experiences of delivering and caring for people receiving this treatment. Specifically, it seeks to address the following research questions:
What are the decision-making experiences of people receiving ICI immunotherapy treatment?
What are people’s experiences of ICI immunotherapy treatment? What are their expectations, concerns, information and support needs?
What are healthcare professionals’ experiences of caring for and supporting patients receiving ICI immunotherapy for cancer?
What are healthcare professionals’ cancer immunotherapy education, training and support needs?
Methods
Design
To obtain a thorough understanding of individual’s perspectives on and experiences of cancer immunotherapy as standard care, there is a need to generate rich data that has the power to account for and explain context and complexity. Thus, an exploratory, qualitative approach comprising in-depth interviews and reflexive thematic, interpretive analysis49 50 will be used. The use of qualitative research will facilitate in-depth exploration of individual’s personal and unique views, capturing rich and detailed insight into hitherto unexplored aspects of individual’s experiences. Furthermore, qualitative research is valuable in the investigation of situations that are not yet fully understood,51 complex and sensitive.
Research setting and study participants
A purposive sample of up to 30 people affected by cancer who are being treated with ICIs will be identified from two oncology treatment centres in the UK. The oncology treatment centres were selected based on convenience sampling, notably existing academic and clinical collaborations and networks between Cardiff University and an National Health Service (NHS) University Health Board and a University Hospital Trust, where it was known that immunotherapies were widely delivered, and where patients could be recruited reflecting the above variables. The sites reflect both urban and rural sociogeographic contexts and different treatment settings. Based on expert knowledge, it was suggested that these characteristics are largely consistent across the immunotherapy treatment context in the other UK nations.
Purposive sampling will be utilised to represent a range of sociodemographic (eg, age and gender), cancer diagnoses (eg, lung, melanoma, head and neck, and renal) and treatment-related variables (eg, ICIs used as first-line and second-line treatments). Purposive sampling and recruitment by clinicians during routine consultations will ensure representation of people affected by cancer regarding their performance status and experience of irAEs.
In view of the COVID-19 physical distancing requirements to reduce infection risk and following Health Research Authority guidance,52 if interested, individuals who meet the inclusion criteria (table 1) will be provided with a study information pack comprising a letter of invitation, participant information sheet and expression of interest researcher contact form, featuring the primary researcher’s phone number and email address so as to enable the potential participant to respond directly to the researcher. Individuals who decide to participate will be asked to contact the primary researcher directly either by email or telephone as detailed on the expression of interest researcher contact form. The researcher will subsequently reply to the individual directly either by e-mail or by telephone to address any further questions.
Inclusion/exclusion criteria for people affected by cancer
If the individual is still interested in participating, study documents (participant information sheet, consent form and letter of invitation) will again be emailed to ensure patients have the correct information, and consent will be accepted via electronic completion and signature as recommended by the latest NHS Health Research Authority guidance52 updated in response to COVID-19. Following this, a mutually convenient day, time, place and format (by telephone or via a secure video-conferencing platform, eg, Microsoft Teams), for the interview will be agreed.
Following pilot interviews and to ensure our participants have appropriate insights, recruitment of up to 15 registered nurses, pharmacists and physicians from oncology services, primary and secondary care (acute oncology) with direct experience of caring for and supporting people receiving cancer immunotherapy will proceed using a combination of purposive and snowball sampling. The sample will include a range of healthcare professionals including clinical nurse specialists (oncology and immunotherapy), oncologists, advanced nurse practitioners, nurse consultants, pharmacists and primary care practitioners. Services outside specialist oncology centres are considered important as patients often present to these services for toxicity management and late onset irAEs, including those which arise post-treatment.
Although the sample size of 15 relevant healthcare professionals is not large, it will be the in-depth, semistructured interview that will enable the team to generate rich data, providing meaningful insights into these experiences across various professional groups. In-depth qualitative interviews have been considered to be an appropriate methodology when looking to generate rich, meaningful data based on experiences, and the success of this methodology in doing so requires close proximity to the human experience under study.53 The term ‘information power’ was conceptualised54 as a way to guide adequate sample size for qualitative research: the more relevant information that the sample holds, the fewer participants required. For this study, it is the relevance of healthcare professionals’ experience in supporting patients undergoing immunotherapy within the sample that is considered the key marker of meaning making within the qualitative approach.53 54
Owing to COVID-19 physical distancing regulations, physicians, registered nurses and pharmacists supporting people receiving immunotherapy will initially be recruited via targeted online social media, specifically Twitter, using existing project networks, and advertising within society newsletters and bulletins including the UK Oncology Nursing Society. Interested healthcare professionals will contact the researcher directly by email. If willing to participate and eligible based on the inclusion criteria (table 2), a convenient time and preferred interview format (telephone or secure, university approved and encrypted video-conferencing software such as Microsoft Teams) will be arranged. To facilitate snowball sampling, interview participants will be asked if they could identify known healthcare professionals (physicians, nurses and pharmacists) who are directly involved in supporting patients undergoing immunotherapy, forward the project flyer to these individuals and ask them to contact the project researcher by email or telephone if they are interested in taking part. Individuals who contact the researcher via this sampling method will be also screened using the inclusion/exclusion criteria documented in table 2. All documents will be emailed and consent will be accepted via electronic completion and signature.
Inclusion/exclusion criteria for healthcare professionals
Data collection
All data will be generated through in-depth, semistructured, digitally, audio-recorded interviews. In view of COVID-19 physical distancing requirements, to prevent infection and following Health Research Authority guidance,52 interviews will be conducted either by telephone or secure video conferencing software (only the audio of video interviews will be recorded to protect participant’s anonymity). Semistructured interviews allow core topics to be raised for discussion, while leaving space and scope for the identification and exploration of unforeseen issues that may emerge.55 56 All interviews will take place in accordance with participants’ preferences. It is anticipated that interviews will last up to an hour, although they could be longer.
Before commencing interviews, information about the study will be confirmed. Participants will have an opportunity to ask questions and will be informed of their right to withdraw at any time without reason or prejudice. If a companion is present during an interview, this will be respected and facilitated, as this may be advantageous in terms of support. Companions’ informed consent will be sought and obtained electronically.
Interviews will be conducted by trained researchers (SA, SJ,TW) and will commence with open questions to develop rapport. Loose interview guides (please see online supplemental file) for patients and healthcare professionals, developed by the research team, derived from the literature review, practice and personal knowledge and scrutinised by the project management team’s Patient and Public Involvement (PPI) member, will act as aide memoires. The interview guides will use mainly open-ended questions.57 To elicit further responses, enrich the description and illuminate experiences, prompts will be made and clarifications sought when necessary. Prior to closing the interview, participants will be given the opportunity to reflect and add any additional relevant information to ensure important aspects not included in the interview guide are addressed.
Supplemental material
Data analysis
Data collection and analysis will occur simultaneously. All interviews will be fully transcribed verbatim by a university-approved external transcriber. Files will be securely sent via the Cardiff University FastFile application. Transcribed data will be analysed using the framework for reflexive thematic analysis devised by Braun and Clarke.49 50 This inductive, systematic, analytic approach involves searching across the dataset for repeated patterns of meaning: data familiarisation, noting early analytical observations; generating initial codes, collating codes and relevant data extracts; identifying meaningful relationships between initial codes and developing themes; refining, defining and naming themes and subthemes in relation to the research aim and objectives. This approach to thematic analysis was chosen to enhance the transparency and detail of the analysis and to best ensure that key patterns and areas of participants’ experiences are captured, relative to the study research questions. The approach also provides researchers with flexibility regarding theoretical approaches that can be applied to the data. Table 3 documents how Braun and Clarke’s49 50 six-step approach to thematic analysis will be applied in this study:
Application of Braun and Clarke49 50 six-step approach to thematic analysis
Within a constructionist, reflexive approach to thematic analysis,50 the process of generating codes, categories and themes is influenced by the researcher. Indeed, the development and identification of themes is based on the researchers’ interpretations and positionality (ie, experience, background, characteristics and assumptions). It is therefore important within this analytic approach to outline the process of theme generation. Furthermore, to ensure rigour, all core research team members (SA, SJ, TW) will contribute to data analysis to ensure consistency in interpretation and a reflective research diary detailing the researcher’s role and impact on the study will be maintained.
Patient and public involvement
Following National Institute for Health Research (NIHR) guidance for involving people in research,58 59 this study has been developed through consultation and active collaboration with members of Swansea University Patient Experience and Evaluation in Research (PEER) and Cardiff University PPI group, who have been personally affected by cancer. Initially, members of the PEER group provided constructive feedback on the research idea and question, study population and design. PEER group representation on the project management team ensures continuing, active patient and public involvement at all stages of the research, including dissemination. Examples to date include offering feedback on the research protocol and ethics applications, acceptability of participant information sheets and interview schedules and contributing to project team meetings. PPI membership of the project’s advisory group will contribute to strategic decisions including disseminating findings and routes to impact.
Ethics and dissemination
The proposed research will be performed in accordance with the UK Policy for Health and Social Care Research and Cardiff University Research Integrity and Governance Code of Practice (2018). The project was reviewed by the West Midlands and Black Country Research Ethics Committee in October 2019 and received a favourable opinion (REC ref: 19/WM/0299). The study is listed on the NIHR Clinical Research Network Central Portfolio Management System (study ID: 43946).
The dignity, rights, safety and well-being of participants will be the primary ethical considerations. Potential participants will be given sufficient time to read and consider study information and ask any questions. All participants who decide to participate will be asked to provide informed consent prior to taking part in the audio-recorded interview. Consent will be accepted via electronic completion and signature, and following Seymour,60 the researcher will use phrases such as ‘would it be okay if I asked about …’ to reaffirm consent during the interview. The participant’s willingness to continue will also be checked at regular intervals.
Data collected from participants will be stored securely at Cardiff University. Completed consent forms will be stored on the university’s password-protected secure server in a location which is only accessible to the chief investigator (TW) and project researcher (SJ). The interview audio-recordings will be uploaded onto the password-protected secure server at Cardiff University. Only the core research team (SA, SJ,TW) will be able to access the recordings. Once interview transcripts have been checked against corresponding audio-recordings, all identifiable information will be redacted and pseudonyms ascribed to all participants. All data will be securely stored after this study in accordance with Cardiff University policies, the General Data Protection Regulation (2016/679) and the Data Protection Act 2018.
Participants may benefit from knowing that their experiences will be used to help inform service developments at individual and organisational levels and healthcare professional education, and thus potentially enhance the quality of person-centred care for people receiving cancer immunotherapy. The study findings may also facilitate knowledge transfer across different treatment sites, potentially benefiting wider patient groups. Knowledge transfer processes include outreach and collaboration with the study team developing collaborations with cancer centres to enhance the practitioner-focused relevance of the educational and training recommendations arising from analysis of data.
Although the research involves people affected by cancer, some of whom may have advanced disease, their participation in the study is unlikely to cause physical harm. However, there is an element of risk related to emotional distress. If this occurs, the researcher will stop recording immediately. Only if the participant is certain that they would like to resume, the interviewer will continue. After interviews, a ‘debrief’ space will mean that all participants will have the opportunity to talk to the researcher about the interview. For patient interviews, if any upsetting or unsettling feelings arose or are disclosed at this point, they will be signposted to local cancer support services. With permission, the person’s cancer key worker61 and consultant will be informed. For healthcare professional interviews, if any distress arises participants will be asked if they would like information about local NHS employee well-being support systems. Any signposting events will be logged in the study file.
Findings will be reported in relevant, peer-reviewed and professional journals using accepted reporting criteria such as COnsolidated criteria for REporting Qualitative research (COREQ-32)62 to ensure transparency. A policy briefing highlighting key findings will be prepared and webinars facilitated to disseminate findings to participants. The findings will be presented at relevant local, national and international conferences and healthcare professional education meetings.
Discussion
There are a number of factors to consider relative to the study design. For instance, there are potential implications for the sample characteristic of using purposive and snowball sampling, as participating healthcare professionals are more likely to be self-selecting, engaged participants, and may even have more years of experience and seniority. However, it is feasible that this approach will ensure that the intended study population is recruited to produce data which will enable us to address the research questions. As snowball sampling will likely lead to professionals identifying colleagues and known networks, this will potentially lead to some degree of bias in terms of work setting, level of experience. Patient participants will also likely to be affected by self-selection bias, even though they are to be identified and approached by clinicians. Self-selection bias in this instance might possibly be related to performance status, overall health and stage of disease, with patients who are generally more well and able to participate electing to participate in the interviews.
Differences in recruitment strategy will also potentially affect analysis of data. The healthcare professionals’ population is feasibly more likely to represent a UK-wide perspective compared with patients, who are likely to be resident within travelling distance of the two oncology treatment centres, situated in Wales. The possible implications for the analysis are that findings from healthcare professionals are potentially more applicable to a UK-wide context. Experiences of immunotherapy treatment, the decision-making processes, associated treatment-related side effects are, however, likely to be translatable across regions within the UK. Within the analysis of data, the limitations of the variable recruitment strategies and resulting differences between population samples will be discussed.
The findings from this research will provide novel insights and in-depth, contextualised knowledge of cancer immunotherapy decision-making, the impact of treatments on people’s everyday lives, their needs and concerns and how they feel they might best be prepared and supported during treatment and beyond. It will explore healthcare professionals’ confidence and preparedness to provide safe, effective, proactive immunotherapy care and identify their support, education and training needs. Understanding people’s experiences may ultimately assist in the co-design of appropriate, effective supportive interventions to optimise the delivery of safe, effective and person-centred immunotherapy care. Furthermore, study findings may be used to inform and support the co-production of educational materials related to cancer immunotherapy and associated supportive care for healthcare professionals.
Current status of the study
Data collection commenced at the end of May 2020 and is ongoing.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @sjennings_01, @@tesswatts15
Contributors SJ and TW: wrote the first draft of the manuscript. SA, JBu, AB, JBo, DFe and DFi: co-applicants of the grant that funds this study. TW, JBu and JBo: conceived and designed the study. JBu: Patient and Public Involvement representative who contributed to the study design. Every author contributed to the drafting of this manuscript.
Funding This study was supported by Macmillan Cancer Support (FO = 7165318).
Competing interests The research team includes individuals from various organisations, including Cardiff University, Swansea University, Hywel Dda University Health Board and Swansea Bay University Health Board. DFe reports an honorarium from Roche.
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.