Article Text
PDF
Abstract
Background Optimal oncological care nowadays requires discussing every patient in a multidisciplinary team meeting (MDTM). The number of patients to be discussed is rising rapidly due to the increasing incidence and prevalence of cancer and the emergence of new multidisciplinary treatment options. This puts MDTMs under considerable time pressure. The aim of this study is therefore to identify the facilitators and barriers with regard to performing an efficient, competent and high-quality MDTM.
Methods Semistructured interviews were conducted with Dutch medical specialists and residents participating in oncological MDTMs. Purposive sampling was used to maximise variation in participants’ professional and demographic characteristics (eg, sex, medical specialist vs resident, specialty, type and location of affiliated hospital). Interview data were systematically analysed according to the principles of thematic content analysis.
Results Sixteen medical specialists and 19 residents were interviewed. All interviewees agreed that attending and preparing MDTMs is time-consuming and indicated the need for optimal execution in order to ensure that MDTMs remain feasible in the near future. Four themes emerged that are relevant to achieving an optimal MDTM: (1) organisational aspects; (2) participants’ responsibilities and requirements; (3) competences, behaviour and team dynamics and (4) meeting content. Good organisation, a sound structure and functioning information and communication technology facilitate high-quality MDTMs. Multidisciplinary collaboration and adequate communication are essential competences for participants; a lack thereof and the existence of a hierarchy are hindering factors.
Conclusion Conducting an efficient, competent and high-quality oncological MDTM is facilitated and hindered by many factors. Being aware of these factors provides opportunities for optimising MDTMs, which are under pressure due to the increase in the number of patients to discuss.
- Clinical Decision-Making
- Communication
- Qualitative research
- Teamwork
- Medical education
Data availability statement
Data are available on reasonable request. The datasets used and/or analysed during the current study are available from the corresponding author on 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: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Multidisciplinary collaboration takes place throughout healthcare by discussing patient cases in multidisciplinary team meetings (MDTMs). Oncological MDTMs in particular suffer from time pressure and optimal execution is not evident.
WHAT THIS STUDY ADDS
This study identified facilitating and hindering factors for the performance of an efficient, competent and high-quality MDTM.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
These findings form the starting point for determining the improvements needed to make MDTMs future-proof.
Introduction
In recent decades, the organisation of care for patients with cancer has changed significantly worldwide. There has been a shift from different specialties all working within their own silo, to integrated multidisciplinary care,1 which is particularly reflected in the emergence of oncological multidisciplinary team meetings (MDTMs).2 In these—often weekly—meetings, the outcome of diagnostic procedures is discussed with the intention of arriving at a final diagnosis and treatment plan. Medical specialists from all the involved specialties, including a surgical, medical and radiation oncologist, radiologist, nuclear radiologist and pathologist are present at MDTMs.3 In addition, in teaching hospitals, residents (defined as qualified doctors in training to become medical specialists) from all these specialties are also present.3 In many countries, there are also administrative support and clinical nurse practitioner (CNS) present.3 4
Discussing a patient in the MDTM is a central point in the treatment trajectory, as it contributes to adequate tumour staging, improves decision making about the most appropriate treatment plan, enhances communication between involved departments and contributes to efficient planning.4–6 Limited evidence even suggests an improvement in patients’ overall survival after being discussed in an MDTM.5 Most national guidelines worldwide, therefore, recommend that all patients with cancer should be discussed at least once in an MDTM.7–9 In some cases, a patient is discussed several times: preoperatively to discuss diagnostics and treatment plan, postoperatively to determine the pathological tumour-node-metastasis stage and follow-up plan, in the event of recurrent disease and in the palliative phase.7 Sometimes a patient is the subject of discussion in different MDTMs (eg, in a local and a regional MDTM).10 Although patients can be discussed in general MDTMs dealing with different tumour types, it is increasingly common to discuss cases in tumour-type specific MDTMs.3 The duration of the meetings varies, usually between 1 and 2 hours, with an average of 2 min discussion time per patient.11 Furthermore, the number of different patients to be discussed is growing rapidly due to the ever-increasing incidence and prevalence of cancer and the increasing number of multidisciplinary treatment options.12 13
Discussing a large number of patients in an MDTM requires a substantial investment in terms of effort and time on the part of the medical specialists and residents involved and accentuates the need to perform MDTMs efficiently. However, this is not self-evident: MDTMs have simply been introduced in cancer care without the development of formal training programmes.14 15 Participants are expected to possess competences such as multidisciplinary collaboration and communication, while Fahim et al16 found that the lack of such competences impairs the decision-making process in MDTMs.16 Residents are expected to learn to participate in MDTMs according to the master-apprentice principle, in other words they learn on the job.17 This seems insufficient given that their ‘masters’ might not have the necessary skills, nor will it necessarily make them into excellent models for future residents.18
In summary, MDTMs are under pressure and optimal execution is not evident. Existing templates on how to perform an MDTM are based on the Calman-Hine report (1995), which described principles about how to organise and structure high-quality multidisciplinary care.1 Following this, in 2010 the British National Cancer Action Team (NCAT) came with a report with 86 recommendations, divided into 5 domains (ie, the team, infrastructure for meetings, meeting organisation and logistics, patient-centred clinical decision-making and team governance) on how to effectively set up and implement an MDTM.19 However, these recommendations are based on a survey with 2000 multidisciplinary team members in the UK, and not based on an interventional study on MDTM-effectiveness.
Therefore, there is no clear definition for the execution of an optimal MDTM. However, the experiences of MDTM participants can provide us the necessary insights into factors that contribute to the MDTM quality. The aim of this study is therefore to identify what, according to medical specialists and residents, the facilitating and hindering factors are for the performance of an efficient, competent and high-quality MDTM. This should be the starting point for determining the improvements needed to make MDTMs future-proof. This study focuses on oncological MDTMs. However, MDTMs are executed throughout healthcare and therefore this study serves as an example for conducting optimal multidisciplinary collaboration in general.
Methods
See online supplemental file 1 for a more detailed description on the methods that were used to execute this study.
Supplemental material
Study design
Between May 2018 and May 2019, a qualitative semistructured telephone interview study was conducted following the Standards for Reporting Qualitative Research (online supplemental file 2).
Supplemental material
Participants
Participants were required to participate in oncological MDTMs on a regular (eg, weekly) basis. In order to maximise variation in participants’ professional and demographic characteristics, we purposively sampled20 interviewees based on five criteria: (1) sex; (2) medical specialist vs residents; (3) specialty (surgical, medical and radiation oncology, radiology, nuclear radiology and pathology); (4) type of hospital (peripheral or academic) and (5) region of hospital (coded to A-B-C-D, based on the provinces in the Netherlands). Of note: since the CNS and administrator are not standard (actively involved) MDTM members, they were not included in this study. Interviewees were approached by email by two researchers (JEWW and IMED) to participate in our study.
Data collection
The primary researcher (JEWW) conducted semistructured interviews. JEWW is a medical oncologist who has been attending two MDTMs per week for 5 years and received interview training prior to the study from an experienced researcher in the field of qualitative research (GH). Interviews were conducted using a topic guide, which was evaluated and adjusted if necessary after each interview. The main topics that guided question development were: MDTM quality, atmosphere and competences, and MDTM improvements and the future (online supplemental file 3). These topics emerged from an extensive systematic literature search into quality factors for MDTMs prior to the interview study.19 21
Supplemental material
All interviews were audiotaped and transcribed verbatim. Interviews had a median duration of 38.7 min and lasted between 27 and 72 min. The transcripts were loaded and stored on the secure servers at the hospital where the researchers work, using ATLAS.ti software V.8.0, a software program for detailed coding in qualitative data analysis.
Data analysis
The data were analysed through thematic analysis, where the unit of analysis was the recorded interview. In thematic analysis, researchers become familiar with the data by reading and rereading the data, generating initial codes, finding overarching themes and revising those themes.22 Three researchers (JEWW, RvdM and AO-B) were involved in reviewing and analysing the interview transcripts. RvdM and AO-B had different backgrounds than JEWW to ensure different reflexive positions (RvdM is a student of biomedicine, AO-B a health scientist). Relevant data were identified and structured using open, axial and selective coding. Coding is the interpretive process in which conceptual labels are given to the data.23 Data sufficiency was reached after 35 interviews, that is, new data no longer provided additional insights relative to the research question.24 During the iterative analysis process, researchers regularly shared and discussed the meaning and uniqueness of generated open codes. Throughout the analysis JEWW grouped codes belonging to the same concept into categories and finally identified themes from the data in consultation with other research members involved (IMED, GH, RHAV). Data analysis was supported using a qualitative analysis software program (ATLAS.ti V.8.0).
Patient and public involvement
Patients were not involved in this study. We thank the interviewees for participating in this study and will send them a copy of this publication to inform them of the results.
Results
Thirty-five individual semistructured telephone interviews with 16 medical specialists and 19 residents were analysed. Interviewees were evenly divided between medical specialties and sex. The distribution of the interviewees across the regions was slightly skewed. More residents were located in academic hospitals (n=16) than in peripheral hospitals (n=3), reflecting the teaching role of academic hospitals. However, the distribution of medical specialists was equal: academic hospital (n=7) vs peripheral hospital (n=9) (table 1).
Characteristics of participants
All interviewees agreed that attending and preparing MDTMs is very time-consuming and indicated that they should be performed as efficiently as possible in order for them to remain feasible in the near future.
The analysis resulted in the emergence of four themes: (1) organisational aspects; (2) participants’ responsibilities and requirements; (3) competences, behaviour and team dynamics and (4) meeting content (online supplemental table 2). Furthermore, online supplemental table 2 also lists the associated eleven categories, 55 facilitators and 45 barriers that were identified. Figure 1 includes associated quotes.
Supplemental material
Quotes related to themes impacting the quality of oncologic multidisciplinary team meetings.
Theme 1: organisational aspects
Five categories were identified within this theme: (1) conditions for information and communication technology (ICT), logistics and administrative support; (2) planning and preparation conditions; (3) conditions for structure; (4) prerequisites for minutes and (5) evaluation needs.
Conditions for ICT, logistics and administrative support
According to the interviewees, the basic conditions for performing a high-quality MDTM are having a U-shaped arrangement in which participants can see each other, where radiological images are projected on large screens and where administrative support is present. Residents, in contrast to specialists, describe that they regularly have to take a seat in the back row, which makes it very difficult for them to make an active contribution to the discussion. Although video conferencing with other hospitals is seen as a benefit when it comes to easy participation, due to less travel time, connectivity issues have been raised.
Planning and preparation conditions
A clear preference was expressed for scheduling MDTMs during working hours. Some interviewees reported feeling less energetic or easily distracted during MDTMs that take place at lunchtime or outside, both before and after regular working hours.
Interviewees indicated that a good discussion is only possible if the participants have prepared the patient case. This assumes a number of preconditions: the preparation time is planned, all necessary information (eg, results of radiological and pathological procedures, patient medical history and preferences) is available on time and there is a deadline for adding a patient case to the application form. In addition, this application form must state which diagnostic specialists (eg, radiologist, nuclear radiologist and pathologist) are required to present diagnostic results, so that they do not prepare cases unnecessarily. Furthermore, radiologists (in training) in particular endorse the importance of a clear question on the application form that must be answered in the MDTM, so that they can specifically show those parts of the radiology results that contribute to the discussion, instead of naming all abnormalities.
Conditions for structure
All interviewees mentioned that structure is an important condition for an efficient MDT: this includes maintaining a fixed order in which participants speak so that everyone’s opinion is heard, a fixed order in which patients are discussed and presenting a patient case in a structured manner.
Barriers to a structured MDTM are disturbances during the meeting (eg, participants making calls or walking in and out of the room), or long meetings (defined by most interviewees as longer than 2 hours), causing attention to decline. Interviewees indicate that, in particular, patients at the bottom of the patient list may receive less attention due to time constraints caused by an excessively long list, or a disproportionate distribution of time between patient cases.
Prerequisites for minutes
Interviewees indicated that good minutes are an important aspect of a high-quality MDTM. Prerequisites for the minutes are that they are taken by experienced administrative support during the meeting, that they are visible on the screens and can be corrected immediately when necessary, that they answer the question and in any event contain a conclusion and treatment plan and alternative treatment options if available. In addition, the minutes should be comprehensible to healthcare providers outside the core team.
Evaluation needs
There was disagreement among the interviewees about the added value of evaluating the functioning of the MDTM; some indicated that a critical assessment offers scope for applying improvements, while others described such an evaluation as time-consuming and predicted that it was unlikely to lead to structural improvements. After comparing the different categories of participants (ie, gender, medical specialist vs resident, medical specialty, type of hospital), we could not identify a category that was specifically in favour or against the evaluation of MDTMs. It was striking that some interviewees were unable to formulate their opinion on this point.
Theme 2: participants’ responsibilities and requirements
Two categories were identified within this theme: (1) the chairperson’s responsibilities and requirements and (2) team member requirements.
Chairperson’s responsibilities and requirements
All interviewees agreed that the presence of a designated chairperson is indispensable to efficient MDTM discussion. The chairperson is responsible for deciding not to discuss a case if preparation is insufficient, structuring the discussions, ensuring that all participants get speaking turns, checking the minutes, summarising each case and drawing the final conclusion. Furthermore, interviewees indicated that intervening in conflicts is also the task of the chair. Young chairpersons (defined as less than 5 years of experience as a medical specialist) in particular indicated that they find it difficult to actually do this, citing a lack of authority.
Team member requirements
According to the interviewees, a high-quality MDTM can only be guaranteed if all core team members (ie, surgical, medical, and radiation oncologist, radiologist, nuclear radiologist, and pathologist) have sufficient up-to-date tumour-specific knowledge and if at least one member from each core specialty is present. A fixed composition of the team is preferred, so that team members become familiar with each other. In the absence of a participant, it is their own responsibility to provide a suitable replacement. Insufficient preparation and absence of a participant who knows the patient personally (to implement patient’s preferences in discussions) are considered as the main obstacles to a smooth discussion.
Theme 3: competences, behaviour and team dynamics
Two categories were identified within this theme: (1) required competences and behaviour of participants and (2) team dynamics and hierarchy.
Required competences and behaviour of participants
Interviewees endorse the importance of the presence of competent MDTM participants. By this they mean: participants are confident, are aware of any gaps in their knowledge and are not afraid to name them, listen to other participants and allow them to finish without interruption and are open to feedback on their performance. In particular, residents who describe themselves as shy indicated that they did not feel free to speak during discussions due to their shyness or introversion. Others (ie, residents or specialists who reported feeling free to talk themselves) noticed that dominant behaviour hinders other participants from providing input in discussions. Engaging in other activities (eg, checking mobile phones or answering emails) is considered to be inappropriate behaviour on the part of participants.
Team dynamics and behaviour
The importance of good team dynamics was emphasised by the interviewees. This means that there is an open and friendly atmosphere and a professional and non-personal attitude towards each other. In addition, participants are able to reach democratic consensus, vulnerability among participants is allowed and participants are loyal and willing to help and trust each other, but are also able to address undesirable behaviour.
However, good team dynamics can be hindered by hierarchy: young, less experienced participants in particular indicate that they feel weighed down by this. Relationships of authority (eg, experienced physician from peripheral hospital vs young consultant from academic centre) might also play a negative role.
Interviewees stated that collaboration is more difficult if there are conflicts between participants, if participants are irritated or unfriendly, if mutual respect is lacking or if subgroups are formed that work together, counteracting the team dynamics. An atmosphere that is too informal is however also seen as a barrier, as it can lead to inefficiency.
Theme 4: meeting requirements
Two categories were identified within this theme: (1) discussion needs; (2) acknowledge educational aspects of MDTMs.
Discussion needs
For an efficient discussion, the interviewees emphasised that taking into account patient preferences and disease-specific characteristics is crucial to formulating a treatment plan, leaving room to deviate from guidelines and devise an alternative treatment plan. In addition, it has been agreed by the interviewees of the clinical specialties that MDTMs can also serve as a forum for discussing the possibility of participating in clinical trials.
The discussion appears to be less efficient if a participant elaborates too much or shares irrelevant information, if arguments are repeated multiple times, if participants (predominantly pathologists were named) only read their report aloud instead of adding value to it, or if only one tumour-specific expert is present, leading to one-sided input. In addition, some interviewees mentioned that their attention decreases if no active input is asked from them.
Acknowledge educational aspects of MDTMs
Interviewees mentioned the importance of acknowledging MDTMs as a learning instrument for both medical and non-medical competences (eg, collaboration, communication). For learning purposes, a case should be reintroduced if the treatment received deviates from the MDTM advice. Furthermore, attention should be paid to non-medical competence training between participants. Residents indicate that they can learn from active participation in MDTMs, although time pressure prevents them from asking questions. Some medical specialists said that participation by too many residents slows the progress of the meeting and felt that too much focus on education is a waste of time.
Discussion
With this interview study, we identified four themes that are important to performance of an efficient, competent and high-quality oncological MDTM: (1) organisational aspects; (2) participants’ responsibilities and requirements; (3) competences, behaviour and team dynamics and (4) meeting requirements. These findings are in line with the themes from the NCAT report, which makes our results more valid and sustainable. All interviewees agreed that MDTMs require a considerable time investment and that the workload has increased substantially in recent years and will continue to increase due to the large number of patients that need to be discussed. They; therefore, indicated that current MDTM execution needs to be improved in order to ensure that the implementation of MDTMs remains feasible in the near future.
Some of these suggested improvements seem quite easily achievable at first glance, however can still be challenging within the background of a demanding healthcare system with high workload and staff shortages.25 26 This concerns improvements such as structuring the meeting, having all core members including a designated chairperson present, scheduling sufficient preparation time, and ensuring that all necessary information is available.
MDTMs often take place in consultation with experts from a nearby hospital or through partnerships where several hospitals participate in one regional MDTM. To enable this kind of consultation and collaboration, a well-functioning video connection with display of radiological imaging, visibility of members on the other side of the connection and medical record display is essential.27 28 However, making optimal use of ICT seems to be a challenge since different hospitals use different systems and sharing information is not therefore self-evident.29 Furthermore, Janssen et al29 found that ICT is underused for providing feedback and real-time data collection.29 However, it should be noted that the COVID-pandemic has accelerated improvements in ICT-capabilities.30 Further future improvements should focus on expanding opportunities of the use of artificial intelligence and computerised decision support systems.31 32
Another challenge in improving MDTMs lies in optimising communication and collaboration between medical specialists and residents of different specialties and recognising their educational value in this respect. It starts with recognising the importance of these core competences, as team dynamics are negatively affected if they are absent.16 33 Residents and young medical specialists in particular reported being hindered by existing hierarchical cultures, making it even more difficult to speak freely. This phenomenon has been identified in several other studies.33 34 To improve multidisciplinary collaboration, MDTM simulation training was suggested by some interviewees. However, the need for this was not supported by all interviewees. Further research is needed to determine the added value of simulation or competence training in general. In addition, this research should include analysing the role of the CNS. The CNS plays an invisible role, but is considered important for the presentation of patient-centred information.35–37
A striking finding from the interviews wat that patients at the bottom of the patient list sometimes receive less attention. This phenomenon was explained by the large number of patients on the list, as well as by disproportionate distribution of time between patient cases. However, it may also be related to cognitive fatigue: a literature review of 2019 described behavioural pitfalls that are associated with prolonged periods of cognitive activity within oncological MDTMs. Decreased rational thinking, reduced attention and more impulsive and riskier decisions were mentioned.25 A lack of attention due to the long duration of MDTMs was also recognised in our interview study. Named solutions to improve cognitive fatigue were taking a short break with stretching in between, take food or drinks (ie, glucose and caffeine) or perform cognitive exercises.25
Interviewees disagreed whether structural evaluation of MDTMs would be of added value for improving quality compared with the time investment such evaluation would require. Several evaluation tools can be identified in literature.4 21 Most of them require an observer that scores predefined quality items such as attendance of core members and availability of all required patient data.21 These evaluation tools were thought the be useful in guiding the evaluation process. Whether they actually optimise MDTM functioning is not yet proven and needs to be further investigated.
We are convinced that optimising MDTM organisation, ICT and team dynamics improves the quality of MDTMs. However, it is questionable whether such interventions actually relieve the perceived time pressure sufficiently and additional measures should be investigated. Time pressure is predominantly caused by the number of patients to be discussed in an MDTM. Streamlining is a possible solution to reduce the number of patients to discuss.10 21 With streamlining patient cases are classified as standard or complex. Standard cases can be selected for discussion in a smaller group of medical specialists and do not need to be discussed in a large regional MDTM.10 38 Streamlining support was investigated by Hoinville et al38; they conducted a national survey of 1220 MDT members in the UK and found that 60% of respondents were in favour of streamlining, while 25% expressed concern about ensuring optimal care for patients that are not discussed.38 Applying streamlining in order to reduce the time pressure on MDTMs needs further research.
Limitations
Our findings should be interpreted in light of several limitations. First, our interview study was conducted exclusively in the Netherlands. In other countries MDTM organisation and ICT capabilities may differ, while other cultural aspects (eg, hierarchy) can have a different impact on the quality of MDTMs. Nevertheless, since MDTMs are common practice, we believe that our general findings are relevant worldwide.
Second, we indicated that MDTMs are performed throughout healthcare. However, it is assumed that the quality factors found in this study also apply to MDTMs other than oncological MDTMs.
Third, we conducted telephone interviews rather than face-to-face interviews. This may have given the interviews a different depth or dynamic. However, the primary researcher was aware of this potential disadvantage and maintained a non-directive and open attitude at all times. By using telephone interviews, we increased the chances of making an appointment and possibly even interviewees’ willingness to participate, as they have busy schedules.
Fourth, we only interviewed medical specialists and residents, as they actively contribute to the MDTM discussion. However, it would be valuable to also include insights from the CNS or administrative supporter with regard to quality improvements for MDTMs. Further research is needed.
Fifth, after an extensive literature search we found that there is a lack of a clear definition for a ‘high-quality MDTM’, since ‘quality’ is a subjective concept.21 Therefore, we focused in our exploratory interview study on factors that reasonably contribute to the execution of an optimal MDTM. This makes our findings less specific and measurable. However, the long list of facilitators and barriers that we found offers an important insight into how MDTMs can be improved to strive for the highest possible quality.
Lastly, interview findings may be biased by the medical background of the interviewer: this entailed a risk of steering the direction of the interview or interpretation of the data. However, this was mitigated by extensive interview training and having the data analysed by multiple researchers from different backgrounds who reached consensus on the final themes and categories.
Conclusion
Conducting an efficient, competent and high-quality oncological MDTM is facilitated and hindered by many factors. Good organisation, a sound structure and functioning ICT are preconditions. Attention should be paid to (training in) multidisciplinary collaboration and communication competences to optimise team dynamics. Future research should focus on additional options to further reduce time pressure on MDTMs, for example, by streamlining cases.
Data availability statement
Data are available on reasonable request. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by CMO Arnhem – Nijmegen: registration number ECSW-LT-2022-3-11-40903. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The assistance of A. Oude-Bos (AO-B) in coding the data of this study was greatly appreciated.
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
Contributors JEWW identified the research question, the topic guide and the research design which was corrected and checked by IMED, RvdM, GH, JJMvdH, RHAV and VEPPL. JEWW received extensive interview training from GH; JEWW and IMED invited interviewees to participate. JEWW performed semistructured telephone interviews. JEWW, RvdM and AO-B analysed interview transcripts. The codebook was developed and refined and categories and themes emerged in consultation with IMED, GH and RHAV. JEWW wrote the manuscript. All authors reviewed the manuscript. Guarantor, IMED.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
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.