Article Text
PDF
Abstract
Introduction A designathon is a three-stage participatory activity informed by design thinking. There is a growing literature on designathons in health. This study synthesised designathons’ effectiveness and implementation-related factors to address health challenges.
Methods We searched Cochrane Library, Embase, PubMed, Scopus and the ClinicalTrials.gov registry for articles containing primary data on designathons for health from their dates of inception to 29 November 2022. We retrieved additional studies from citation searching and a complementary open call. We synthesised data on designathons’ effectiveness (ie, engagement, outputs and implementation), required resources and implementation-related factors (ie, resources, facilitators, barriers, strengths and limitations). We assessed the risk of bias using a checklist adapted from Joanna Briggs Institute Critical Appraisal tools.
Results In total, 4973 citations were identified, and 42 studies were included. In total, 26 studies (62%) were from high-income countries. The median number of total participants was 49, divided into a median of 8 teams. The duration of the intensive collaboration phase ranged from 3 hours to 7 days. Common evaluation criteria were feasibility, innovation and impact. Idea and prototype outputs included mobile phone applications, educational programmes and medical devices. Interventions developed from a designathon was estimated to be highly cost-effective. The most common facilitators were interdisciplinary participants and high-quality mentorship. The most common barriers were suboptimal execution of the events, difficulties in balancing interdisciplinary participants across teams and limited support for participants along the process. There were limited data on required resources and further implementation of solutions after designathons.
Conclusion Given designathons’ adaptability in terms of budget, mode of delivery, type of output and involvement of diverse participants, including end users, designathons can be implemented in a wide range of contexts to address various health issues.
PROSPERO registration number CRD42023389685.
- Systematic review
- Health policies and all other topics
- Public Health
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
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
There is evidence of the implementation of designathons across different regions, settings, populations, budgets and modes of delivery. However, no systematic review on designathon in health was identified.
WHAT THIS STUDY ADDS
We found that designathons can be implemented in a wide range of circumstances to address various health issues, advance health outcomes and increase the sustainability of solutions due to their adaptability in terms of budget and mode of delivery and ability to engage a variety of participants’ professions and end users.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Given that designathons facilitate collaboration between community members and experts from diverse fields to develop innovative and impactful solutions, this approach has been used to address complex health challenges in various populations, including marginalised populations and settings, including those with limited resources.
Introduction
A designathon is a time-based three-stage participatory activity informed by design thinking that includes preparation with end users, an intensive period of collaborative teamwork (typically from a few hours to a few days) and follow-up activities for implementation and research (including the evaluation of solutions by topic experts or community partners).1 This approach is a type of crowdsourcing activity where solutions are generated by a large group or ‘crowd’ instead of relying on internal experts (ie, a ‘top-down’ approach).1
Designathons are a participatory approach that directly engages multiple stakeholders, including end users, and increases public awareness of specific health challenges (eg, multiple cultural settings). This approach fosters a sense of community empowerment and collective responsibility, leading to more sustainable and impactful outcomes in health-related initiatives.2 This aligns with the WHO framework on integrated people-centred care. This framework revolves around developing health services or products that co-ordinate around people’s needs, respect their preferences and actively involve end users in the development to empower individuals and communities.3 Additionally, the versatility of designathons extends to developing social innovations which operate within a framework that is deeply ingrained within the community. They draw on the diverse strengths of local individuals and institutions, synergising social transformation and health advancement.4
Designathons share commonalities with hackathons, co-creation sessions and design sprints in terms of problem solving5 and a specifically structured process. However, designathons differ from those approaches listed above by explicitly integrating design thinking or human-centred design principles. This creative problem-solving approach involves inspiration, ideation and implementation stages, with the aims to develop desirable, feasible and viable solutions.6
Designathons are increasingly used to engage end-user communities in multiple settings to develop social innovations.7 8 For instance, more than 26 000 individuals attended an online designathon hosted by Germany’s government to develop solutions to address the COVID-19 pandemic.9 Additionally, designathons have been used by the National Academies of Sciences, Engineering, and Medicine,10 WHO11 and the United States National Institute of Health.12
Despite the expansive literature on designathons in programmes and research, there are limited reviews on this topic.13–15 The data gathered from this review will guide future designathons and contribute to a concomitant WHO and the Special Programme for Research and Training in Tropical Diseases (TDR) practical guide. We aim to synthesise data from designathons used in health research in the peer-reviewed literature to understand their effectiveness (ie, engagement, outputs and implementation), required resources and implementation-related factors (ie, resources, facilitators, barriers, strengths and limitations).
Methods
We followed guidance from the Cochrane Collaboration to conduct the systematic review and reported our findings according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement.16 We registered the protocol in PROSPERO (CRD42023389685).
Eligibility criteria
Studies were included if they met the following eligibility criteria: (1) contained primary data regarding designathons for health (ie, time-limited event), involved a range of stakeholders working in teams to design solutions to a specific problem, solutions generated were judged or evaluated and the solutions might be implemented or tested after the event); (2) focused on health outcomes related to the attainment of either physical, mental and/or social well-being of human beings or the healthcare system and/or policy; and (3) the event explicitly mentioned design thinking or human-centred design or included participant activities focused on common design thinking components (ie, empathise, define, ideate, prototype and evaluate). Studies were excluded if they were duplicates of other studies.
Search strategy and study selection
A health sciences librarian (JLC) searched the following four databases from their inception to 29 November 2022: the Cochrane Library, Embase (Elsevier), PubMed and Scopus (Elsevier). The ClinicalTrials.gov registry was also searched. The searches included a combination of subject headings and keywords for the main concept of designathons. Four separate approaches were combined with “OR” terms for designathons or agile methodology; user-centred design or rapid prototyping events; hackathons, makeathons and other similar events; and community participatory ideation events. The Scopus search had an additional search string related to health and well-being since the database includes disciplines beyond health and medicine. Conference abstracts were filtered from the Embase search. No other filters or restrictions were placed on the searches. The complete and reproducible search strategy can be found in online supplemental table 1.
Supplemental material
The librarian (JLC) placed all references in Endnote V.X9 (Philadelphia, Pennsylvania, USA) and removed duplicates. When screening the title and abstract, we used a two-stage automation approach using supervised clustering through ICF’s Document Classification and Topic Extraction Resource (DoCTER) (https://www.icf-docter.com/).
To begin the automation process, the team independently screened 300 references to identify seed studies as training data for DoCTER. All identified seed studies were used for supervised clustering, the first stage of automation. In this stage, all references were given an ensemble score ranging from zero (deemed irrelevant) to six (deemed relevant in all six models). Those with a score of four or higher were imported in Covidence (Veritas Health Innovation, Melbourne, Australia, available at www.covidence.org) for screening. Each abstract was screened against eligibility criteria by two researchers independently (WT, TR, EC, CL, AMO, AOB, UN), with the third reviewer (JJO) resolving any conflicts that arose. In the second stage, those studies included after the first stage screening were used as training data for machine learning through DoCTER. DoCTER gave a probability score ranging from zero to one for each unscreened study. A higher score indicated a higher relevance ranking. Each title and abstract were again screened by two researchers independently (TR, EC, CL, AMO, AOB, UN) in Covidence until relevance dropped off, at a probability score of 0.4. A third reviewer (WT) resolved conflicts that arose during the screening. We also screened additional potentially relevant peer-reviewed studies we received from a complementary open call for examples of designathons. This was conducted between 16 January 2023 and 7 March 2023 and received 27 submissions. Data, including outcomes shown in table 1, from each eligible full-text article was extracted by two researchers independently (TR, EC, CL, AMO, AOB, UN) using a survey WT and EC created. Data were subsequently exported to a spreadsheet, and the third reviewer (WT) resolved any conflicts.
The definition of the outcomes included
We used descriptive statistics, including percentages for categorical variables and median and mean for continuous variables, to summarise designathon characteristics, such as region of study, designathon participants and final outputs of designathons. We conducted a content analysis to synthesise results related to the outputs, barriers, facilitators, strengths and limitations to implementing specific designathons and more generally using designathon approaches to collect data.
Quality assessment
We assessed the risk of bias using a checklist adapted from Joanna Briggs Institute Critical Appraisal tools.17 Risk of bias was independently examined by two reviewers (WT and EC), and any discrepancy was resolved by a third reviewer (JJO). However, this did not influence the inclusion of the studies in this review (online supplemental table 2).
Patient and public involvement
It was not appropriate or possible to involve patients or the public in the design, or conduct, or reporting, or dissemination plans of our research.
Results
Overall, 4846 citations were identified via electronic database searching, with 31 publications included in this review. We retrieved 127 publications from additional searches (ie, citation searching and the complementary open call), with a further 11 publications included in this review. In total, we included 42 studies in this review,7 8 18–57 with 39 studies (93%) focusing on developing social innovations.7 8 18–20 22–42 44–53 55–57 The PRISMA flowchart is presented in online supplemental figure 1. We included 26 publications (62%) from high-income countries, 13 (31%) from middle-income countries, 2 (5%) global publications and 1 publication (2%) with an unclear location. In total, 35 publications (83%) presented data on in-person designathons and 30 (71%) used ‘hackathon’ to describe designathons. In total, 26 publications (62%) were case reports, 9 (21%) were mixed-method studies, 4 (10%) were qualitative studies, 1 (2%) was a quasi-experimental study, 1 (2%) was a modelling and economic evaluation study and 1 (2%) was a randomised controlled trial (table 2).
Overview of the studies included (n=42)
Engagement
In total, 41 studies (98%) reported the number of designathon participants.7 8 18–53 55–57 The median number of individuals participating in designathons was 49 (IQR 34–90, range 6–434), and the median number of teams was 8 (IQR 5–14.5, range 2–40). In total, 30 studies (71%) included participants who were a mixture of health and non-health professionals.7 19–22 24 26 28–42 44 47 50–54 57 In total, 37 studies (88%) reported the duration of the intensive collaboration part, ranging from 3 hours to 7 days (online supplemental table 3).7 8 18 21–24 26–29 31–53 55–57
Resources
Only two studies (5%) reported data on the cost of organising a designathon.29 36 A 2-day designathon organised in India to develop solutions for cancer care reported that each participant costs around £16.36 Meanwhile, a 2-day hackathon in the USA aimed at generating solutions for unmet clinical needs costs approximately US$40 000. One study (2%) examined the cost-effectiveness of an intervention developed through crowdsourcing, designathon and randomised controlled trial (RCT).57 The intervention was estimated to be highly cost-effective, with incremental cost-effectiveness ratios ranging from US$171 to US$204 per disability-adjusted life years averted for a one-off event.57
Outputs
In total, 28 studies (67%) reported the number of final outputs developed during the event.7 8 18–20 22–27 29 30 35 36 38–40 42 44–47 50 52 55 57 The median number of outputs was 8 (IQR 6–13, range 2–40). The final output from designathons were ideas (n=20, 48%)7 8 18 19 21–23 25 26 32 40 41 47–51 54–56 and prototypes (n=20, 48%).20 24 27–31 33 35 36 38 39 42–46 52 53 57 The final outputs reported from 34 studies (81%) included the following: (1) educational programmes (n=7, 17%), such as wellness and healthy choice programmes18 21–23 25 43 54; (2) mobile phone applications (n=7, 17%), such as management of opioids and support mental health24 26 27 31 38 39 45; (3) medical devices (n=4, 10%), such as breast pumps and humanitarian surgical care20 32 42 46; and (4) other outputs (n=16, 38%), such as telehealth service, HIV self-testing platforms and campaign packages (figure 1 and online supplemental table 4).7 8 19 29 30 33 36 40 44 47 48 50 52 55–57 Regarding evaluation, 24 studies (57%) reported evaluation criteria, with the common criteria being feasibility, innovation and impact.7 8 19 20 22 24 26 27 29 33 34 36 38 40 42 43 45 49 50 52 53 56 57 Eight studies (19%) reported using other methods, such as feedback from the judging panel and audience and votes, to evaluate the output generated during the designathon (figure 1 and online supplemental table 4).21 25 26 31 44 46 54 58 In total, 34 studies (81%) reported having judging panels as follows: designathon participants (n=2, 5%),25 26 external expert (n=19, 45%),7 24 28–30 33–35 37 40 41 44–46 48–51 56 more than one profession (n=10, 24%)8 18 20 22 23 27 36 38 43 47 58 and others (n=6, 14%) (eg, steering committees and faculty mentors) (figure 1 and online supplemental table 4).19 21 34 52 54 57
Summary of outputs from designathon included.
Implementation
In 20 studies (48%), ideas or prototypes generated during designathons were implemented within 12 months.7 8 21–25 32 36 39–42 47 49 52 54–57 Seven studies (17%) reported testing outputs in the real world using an online survey,54 a start-up business,32 research studies8 25 56 and randomised controlled trial.52 57
Facilitators of organising a designathon
In total, 30 (71%) studies reported facilitators of organising a designathon.7 8 18 19 21–24 26 29 32 33 35–38 40 42 44 47–49 51 53 55 56 The most common facilitators reported were (1) diverse interdisciplinary participants (n=19, 45%)7 19 21–23 26 29 32 33 35–38 42 44 47 48 51 53; (2) high-quality mentoring/coaching for designathon participants (n=13, 31%)7 21 22 29 32 33 35 36 38 40 42 49 53; and (3) identifying participants with contextual knowledge and/or familiarity with the target problem (n=10, 24%).22 27 29 32 36 38 42 51 55 56 Other facilitators reported included clear communication plans (n=7, 17%),7 18 24 29 33 37 44 conducive workshop environment (n=7, 17%),8 23 32 35–37 44 team-building activities (n=6, 14%),19 23 24 36 38 53 partnership with organisations and stakeholders (n=4, 10%),8 28 35 39 virtual workshop options (n=3, 7%)21 22 53 and the collocation of designathons within larger events (n=2, 5%) (figure 2).19 49
Facilitators and barriers to a successful designathon.
Barriers to organising a designathon
In total, 18 studies (43%) highlighted various barriers encountered during organisation designathon events.7 19 21 22 25 29–31 33 36 37 39 41–45 53 These barriers can be classified into three main themes: (1) suboptimal execution of the designathon events (n=13, 31%)7 19 22 25 30 33 36 41–45 53; (2) difficulties in engaging diverse participants (n=8, 19%)22 25 29 31 33 36 39 42; and (3) limited onsite mentorship (n=3, 7%).36 42 53 59 Execution barriers included limited funding, technology-related issues (eg, bandwidth and software) during the event, insufficient planning of the intensive collaboration period and limited post-event follow-up. For example, while organising designathon events within existing conferences could be convenient, careful planning in advance was necessary to ensure sufficient time allocation. Participating in an intense in-person event lasting 2 or 3 days was difficult for participants who typically had demanding schedules (eg, medical professionals and policymakers). The engagement barriers primarily revolved around recruiting interdisciplinary participants, balancing each profession across teams and effectively collaborating with team members holding divergent perspectives. The challenges of providing mentorship of good quality included unclear communication or limited and unstructured post-event support, which affects the long-term sustainability of projects (figure 2).
Strengths of using a designathon
In total, 34 studies (81%) explicitly discussed the strengths of the designathon approach.7 8 18 19 22–29 31–53 56 57 The strengths emerged into 10 themes as follows: (1) engaging an interdisciplinary set of participants (n=25, 60%,)7 19 22 24 26 27 29 32–36 39–44 47 49 51 53 56 57; (2) encouraging creativity and innovation (n=25, 60%)7 8 18 22–24 26–29 32 33 35 36 39–43 47 49 53 56 57; (3) creating ways for end users to lead in the development (n=20, 48%)7 8 22–25 27 32 33 35 36 38–40 42 47 50–53 57; (4) networking, connections and collaborations (n=18, 43%)7 8 19 22 23 25–27 32 33 39–44 49 53; (5) being faster (n=5, 12%)7 24 33 47 53; (6) being cheaper (n=2, 5%)24 39; (7) free and open for all to participate; (8) accelerated the development of ideas; (9) engaged the broader society in a topical area; and (10) increased the likelihood of community engagement for a proposed solution (figure 3).
Strengths and limitations of using a designathon.
Limitations of using a designathon
In total, 24 studies (57%) explicitly discussed the limitations of the designathon approach.7 8 18–20 22 26–29 33 35 36 39 41–44 46 47 49 50 52 57 The limitations emerged into seven themes as follows: (1) the generalisability of findings generated from designathons (n=9, 21%)7 8 20 26 33 39 41 43 50; (2) the intensive time commitment required to participate (n=7, 17%)7 22 33 39 41 46 49; (3) challenges associated with assessing impact (n=6, 10%)7 27 29 44 47 49; (4) location if the event is held in-person (n=5, 12%)20 27 29 35 44; (5) participant stigma or sampling bias (n=3, 7%)8 36 39; (6) cost (n=2, 5%)18 49; and (7) concerns regarding long-term potential, such as low idea feasibility, or insufficient support or opportunities to sustain initial idea investments (n=4, 10%).8 28 29 42 Other limitations identified in individual articles included concerns that designathons encouraged competition rather than collaboration, participants focused on earning a prize rather than facilitating true impact, ideas were not always feasible and participants might not be knowledgeable of the communities or settings for which they were developing interventions or prototypes, limiting the applicability of ideas generated during designathons (figure 3).
Discussion
This systematic review comprehensively synthesised data from designathons for health. Our review highlights key characteristics of designathons in health research in terms of effectiveness (ie, engagement, outputs and implementation), resources and implementation-related factors (ie, facilitators, barriers, strengths and limitations). We found that designathons were an adaptable approach accommodating a wide range of participants (eg, marginalised populations) and had versatile delivery modes (ie, in-person, online and hybrid) and time frames. This approach offered flexibility in terms of participant scalability and required resources. By engaging interdisciplinary participants and end users, designathons encouraged creativity and innovation.
The participatory nature of designathons has the potential to enhance implementation success and solution creativity and sustainability since the solutions better align with the needs of the end users. Designathons provide a supportive environment that facilitates intensive collaboration between interdisciplinary participants and end users. With the guidance and support from mentors, the workshop experiences are enhanced, leading to higher quality solutions generated, thereby enabling the development of user-led innovations. Additionally, designathons have successfully engaged marginalised groups who faced specific challenges and generated sustainable solutions for them.21 54 57 The diverse ideas and prototypes generated during designathons emphasise their effectiveness in generating solutions to specific contexts and health challenges. This aligns with the larger literature on crowdsourcing.60 Designathons are inherently action-oriented and have been successfully implemented in various settings (ranging from local community organisations,25 within conferences19 to international platforms30 53 and from low-income,7 8 55 56 middle-income52 57 to high-income countries26 29 61). Designathons are flexible to adapt to constantly changing real-world settings and constraints, including number of participants, delivery modes and cost. For instance, designathons successfully transitioned to online formats during the COVID-19 pandemic,21 53 54 which expanded stakeholder involvement and reduced organising costs. Additionally, the large number of studies focusing on using designathons to develop social innovations suggests that this may be a productive area for further designathons.
While there are several advantages to using designathons to improve health outcomes, previous literature highlighted notable challenges that may limit their utility. First, several studies pointed out concerns regarding the generalisability of ideas resulting from designathons7 8 20 26 33 39 41 43 50 because designathon participants are often tasked with identifying solutions to specific problems in a designated context or population. However, since designathons enable out-of-the-box thinking, using designathons could solve persistent problems in the local context. This suggests the importance of ongoing evaluation of prototypes and ideas for broader applications and their potential for scale up. Second, a key critique of numerous designations has been the lack of further development of ideas and/or prototypes after the conclusion of the formal event. This may reflect some designathons having a prize structure that does not strongly link to implementation or research. However, there are several ways that designathons have formally included follow-up, including mentorship following the intensive event, recognition within the regional or global (eg, Social Innovation in Health Initiative) or other networks or prizes money explicitly to support implementation and research to assess effectiveness.7 52 Further studies on implementing the designathon solutions after the event are warranted. Third, costs were identified as a disadvantage for organisers and participants who had limited budgets.7 32 However, using local resources and volunteer mentors was reported as a cost deduction solution,7 18 and an intervention developed through a designathon was estimated to be cost-effective.57 Additionally, designathons were successfully implemented in low-income and middle-income countries, such as Nigeria and China.7 8 52 55–57 Given that only two studies provided detailed cost information and only one provided cost-effectiveness output, further research on the required resources and cost-effectiveness of designathons is needed.
Our systematic review has several strengths. First, all peer-reviewed publications from different regions and a complementary open call were comprehensively and systematically reviewed. Second, our review provided a comprehensive understanding of designathons in the context of health. Few studies have systematically explored this topic. Finally, this process was done in collaboration with the WHO/ TDR to inform a practical guide. Our study should be considered in light of limitations. First, due to the nature of designathons, which focus on developing ideas for a particular problem rather than generating scientific-based evidence, we might miss reports of designathon as part of other events, such as crowdsourcing and randomised controlled trials. We mitigated this limitation by including data from the crowdsourced open call for designathons to identify any reports we might miss.
Moving forward, we need to establish a shared understanding of applications of designathons in health contexts, outlining core components and the process involved (ie, plan, implementation and evaluation). Guidance, tips and lessons learnt from previous designathons should be incorporated into future designathons to ensure consistent reporting, enhance equity and assist when comparing between studies. Key metrics for evaluation should include testing prototypes with end users and evaluating medical or public health outcomes. External evaluations of designathons are necessary to ensure rigour.62
In conclusion, designathons emerge as a promising and effective tool to develop innovative and impactful solutions to address complex health challenges in various populations and settings due to their ability to facilitate collaboration between the community and experts from diverse fields. Given its adaptability (ie, budget and mode of delivery) and ability to engage a variety of participants’ professions and end users, designathons can be implemented in a wide range of circumstances to address various health issues. By harnessing the strengths of designathons, we can unleash new opportunities and empower communities to significantly impact healthcare outcomes. Future work can explore the cost-effectiveness and implementation of designathons to promote diversity and inclusion in health intervention development and implementation.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
Acknowledgments
The authors thank Dr Beatrice Halpaap and Dr Meredith Labarda, present staff members of Social Innovation in Health Initiative by WHO/TDR, and Dr Noel Shaskan, lead of Solve’s Health Community Lead by Massachusetts Institute of Technology, for reviewing the manuscript before submission. The work received support from TDR, the Special Programme for Research and Training in Tropical Diseases, co-sponsored by UNICEF, UNDP, the World Bank and WHO. TDR was able to conduct its work thanks to the commitment and support from a variety of funders.
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
WT and EC are joint first authors.
Handling editor Seye Abimbola
Twitter @DrJasonJOng
Contributors JDT and JJO conceived the idea. JLC did the search strategy. WT, EC, TR, CL, AOB, AMO and UN did the screening and extracted the data. All authors contributed to the first draft of the manuscript and the interpretation of the results and had final responsibility for the decision to submit for publication. WT and EC assessed the risk of bias and revised the manuscript. JJO accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
Funding JJO was funded by an Australian National Health and Medical Research Council Investigator Grant (GNT1193955). TDR’s effort was supported by a career development award from the National Institute of Mental Health (K08MH118965). TDR was supported by authors' long-term core contributors from national governments and international institutions, as well as designated funding for specific projects within authors' current priorities. For the full list of TDR donors, please visit TDR’s website (https://www.who.int/tdr/about/funding/en/). TDR received additional funding from Sida, the Swedish International Development Cooperation Agency, to support Social Innovation in Health Initiative. AMO was funded by the Philippine Council for Health Research and Development, Department of Science and Technology (RGAO-2022-0304). JDT was funded by National Health Institute (K24AI143471, R01CA271033, R01AI158826 and UG1HD113156-01). The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the funders.
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.