Article Text

Investigating physician leadership competencies in rural and remote areas of the province of Aceh, Indonesia
Free
  1. Fury Maulina1,2,
  2. Mubasysyir Hasanbasri3,
  3. Fedde Scheele4,5,
  4. Jamiu O Busari6,7
  1. 1 School of Health Professions Education, Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, the Netherlands
  2. 2 Department of Public Health, Faculty of Medicine, Universitas Malikussaleh, Lhokseumawe, Aceh, Indonesia
  3. 3 Department of Biostatistics, Epidemiology, and Population Health, Faculty of Medicine, Public Health and Nursing, Gadjah Mada University, Yogyakarta, Indonesia
  4. 4 Athena Institute for Transdisciplinary Research, Faculty of Science, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands
  5. 5 Research in Education, Amsterdam UMC Locatie VUmc, Amsterdam, the Netherlands
  6. 6 Educational Development and Research, Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, the Netherlands
  7. 7 Department of Pediatrics, Dr Horacio E Oduber Hospital, Oranjestad, Aruba
  1. Correspondence to Fury Maulina, School of Health Professions Education, Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht 6229 ER, Netherlands; f.maulina{at}maastrichtuniversity.nl

Abstract

Backgrounds Globally, the most rural healthcare systems are lagging behind those of urban healthcare systems. Especially in rural and remote areas, the essential resources to provide principal health services are inadequate. It is purported that physicians have an important role in healthcare systems. Unfortunately, there is a paucity of studies on physician leadership development in Asia, especially on how to enhance physician leadership competencies in rural and remote low-resource settings. This study aimed to investigate doctors’ perceptions of existing and needed physician leadership competencies based on their experiences in primary care settings in low-resource rural and remote areas are in Indonesia.

Methods We performed a qualitative study with a phenomenological approach. Eighteen primary care doctors, who worked in rural and remote areas of Aceh, Indonesia, purposively selected, were interviewed. Prior to the interview, participants were asked to select the top-five skills they deemed most essential for their work based on the five domains of the ‘Lead Self’, ‘Engage Others’, ‘Achieve Results’, ‘Develop Coalitions’ and ‘Systems Transformation’ (LEADS) framework. We then performed a thematic analysis of the interview transcripts.

Results We identified the following qualities a good physician leader in low-resource rural and remote settings should possess: (1) cultural sensitivity skills; (2) a strong character that includes courage and determination; and (3) creativity and flexibility skills.

Conclusions Local cultural and infrastructural factors create a need for several different competencies within the LEADS framework. A profound amount of cultural sensitivity was considered the most important in addition to the ability to be resilient, versatile and ready for creative problem-solving.

  • medical leadership
  • doctor
  • competencies
  • primary care
  • health system

Data availability statement

No data are available.

Statistics from Altmetric.com

Request Permissions

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

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • Formalised leadership development is scarce in low-income and middle-income countries, particularly in rural and remote settings. Meanwhile, doctors working in these settings are frequently thrust into leadership roles and are expected to lead community health activities.

WHAT THIS STUDY ADDS

  • We discovered that the leadership competencies of rural and remote primary care doctors must fit with the unique features, culture and dynamics of rural/remote settings. Being resilient, adaptable and prepared for creative problem-solving are important physician attributes needed in these environments.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • Due to the inherent constraints and complexity in rural and remote places, there is a need for leadership training in the medical curricula that is structured, and tailored to fit the local contexts.

Background

Globally, inhabitants of rural and remote communities have poorer health status than urban dwellers. Poor access to healthcare is a significant contributor to these conditions, particularly in low-income or low- and middle-income countries (LMICs).1 Health inequalities exist between urban and rural populations.2 Other reports, too, have determined that several countries face comparable issues because health resources are concentrated in cities, while most of the population lives in rural areas.1 3 What’s more, rural and remote healthcare systems face critical issues in service delivery, human resources and communication.4 5 Given the unique context of practice in a rural community, it stands to reason that there should be a list of domains and competencies that are especially important to sustained competence in that setting. Physicians, therefore, who want to practise in low-resource settings will eventually need additional competencies.6 Indeed, one study showed that constraints in rural and remote settings affected leadership and resource stewardship. Further, the quality of primary care services in rural settings depends on physicians’ leadership competence.7 As such, a healthcare organisation’s success depends on its ability to cultivate competent physician leadership.8 It also suggests that the concept of physician leadership would be relevant for rural and remote areas as it provides a way to improve health services in vulnerable settings9 and they are ready to handle and take on unusual responsibilities.10

In this context, the LEADS framework, as a conceptual framework, is of significant relevance as it describes the capabilities physicians and other healthcare providers need to be effective leaders in healthcare. The ‘LEADS’ acronym reflects the following five leadership domains: ‘Lead Self’, ‘Engage Others’, ‘Achieve Results’, ‘Develop Coalitions’ and ‘Systems Transformation’. LEADS has proven to be versatile enough to be helpful in a wide variety of organisations and settings.11 As such, it might also offer insight into certain capacities that physicians in remote areas of LMICs need to perform their tasks properly.

Formalised leadership training for doctors is scarce,12 13 for example, in Indonesia. As a result, rural municipalities are forced to hire doctors who lack these qualifications.7 Research on how to engage doctors in rural and remote areas is scant14 and little is known about how to improve physician leadership in these areas,9 including Asian countries. Although several articles have described healthcare problems in LMICs, these rarely touched on leadership in this context. The LEADS framework, while providing a very comprehensive guide for leadership development, does not provide points of focus for the specific context of rural and remote areas in LMICs. As an LMIC, the Indonesian context offers an opportunity to study doctors’ perceptions of physician leadership during practice in rural and remote places. Hence, we sought to fill this knowledge gap by focusing on a rural and remote part of Indonesia. Our research question was: Which specific leadership competencies are needed in rural and remote areas of Indonesia as an LMIC?

Methods

Study setting

Indonesian healthcare system has grown to meet the needs of its population. However, significant inequalities between urban and rural areas in access to care and state of health persist up to this day.15 Aceh, one of provinces which have many underserved villages, is divided into 18 districts and 5 autonomous cities. Of the estimated 17 000 villages dispersed throughout Indonesia, 2244 are located in the province of Aceh. Most of these villages, moreover, have been classified as poor.16

With respect to health services, Indonesia offers primary, secondary and tertiary care. The largest share of the country’s total healthcare (86%), however, is offered in primary care facilities, called Puskesmas (Pusat Kesehatan Masyarakat). As such, primary care is offered in every subdistrict and aims to meet both community and individual health needs by promoting and preventing disease, and providing curative and rehabilitative services.17 As one of the health system leaders, primary care physicians play the role of gatekeeper and guarantee long-term health management.18 In Aceh, the total number of Puskesmas was 359 in 2020,19 with 196 units located in rural (55%), 69 units in remote (19%), and 33 units in very remote areas (9%).20 Access to primary care services in remote and very remote areas is sometimes difficult due to inadequate road systems, especially during the rainy season. Serious concerns of primary care staffing, a poor health infrastructure and limited access to basic laboratory and emergency drugs are complicating matters even more.

Study design

We performed a qualitative study with a phenomenological approach, for which we held semistructured interviews. After being primed with information about the LEADS framework, we invited participants to elaborate on their responses based on their work experiences in rural and remote places. To elucidate certain issues or specific points, we also asked follow-up questions. We designed the interview guide using the five pillars of the LEADS framework11 (see online supplemental appendix 1). The author translated the interview guide into Indonesian and then back-translated it into English to ensure consistency of meaning across all questions. Data were collected during a one-and-a-half-month time span.

Supplemental material

Participants

The study sample comprised 18 doctors who worked in primary care in rural and remote areas. Participants were purposively selected due to geographical and logistical constraints. The province of Aceh is made up of 18 districts (2 of which are located on islands) and access to each of these islands is difficult and only either by air or sea. Furthermore, to meet a respondent who works in a remote location on this island, we must travel for 8 hours (round trip by car). Since our aim of our study was to obtain both a diverse and representative overview of leadership needs and practice in the different districts, we had to weigh our objectives against the practical and local challenges on the ground. To address this in part, we chose to purposefully sample participants from each district to ensure that we obtained a representative perspective on physician leadership in the province. Respondents were doctors who had been in practice for at least a year. We chose this time frame because it represents the average period employees need to adjust to the workplace.21

Data collection

After finding respondents and keeping in contact with them, assessing their eligibility and receiving their consent, we gave participants an overview of leadership attributes based on the five pillars of the LEADS framework, that is, the priming process. Respondents were asked to use this framework to reflect on their own experiences and to select from among this list the top-five skills they deemed most essential for their work in rural and remote settings (ranking). We subsequently interviewed the participants, transcribed the interviews, analysed the transcripts and discussed the findings within the author team. The lead author conducted the 18 individual interviews with respondents by visiting them in person.

Analysis

After participants had ranked the leadership attributes they considered most essential, we categorised each item (attribute) according to the number of responses received as follows: ‘low’ (0–5 responses); ‘moderate’ (6–10 responses) and ‘high’ need (over 10 responses). This step was used to set the stage for further discourse on leadership and to get a first impression of perceived needs.

The interviews were recorded and transcribed verbatim; transcription was done manually using Microsoft Word. To analyse the interview transcriptions, we performed a thematic analysis (see online supplemental appendix 2), which is a manual data analysis technique that spanned the following phases: (1) generating initial codes; (2) searching for themes; (3) reviewing themes; (4) defining and naming themes; and (5) writing up.22 From our raw data, we collected quotes that were representative of each theme and translated these from Indonesian into English.

Reflexivity

FS is a practising gynaecologist and a professor in health systems innovation and education in Amsterdam; he is an expert in qualitative methods within transdisciplinary research projects. JB is a general paediatrician, educationalist and associate professor of medical education in Maastricht; his area of expertise is in equity and leadership development in postgraduate medicine. MH is a medical doctor and associate professor of public health at Gadjah Mada University and teaches primary healthcare policy and management in Indonesia. FM is a general practitioner from Aceh, and her research interests centre on physician leadership in underserved areas and healthcare systems in rural and remote settings.

The authors were aware of the potential for bias in qualitative studies and took this into account during this study, especially since the lead researcher (FM) hails from the research setting under investigation. Despite the potential for bias, it was believed that, if existent, it would be effectively mitigated by the variegated composition of the research team. For example, FS and JB’s experiences as physician leaders and researchers in the Netherlands provided an external perspective on the study design, analysis and interpretation of data. They used a (Western) European lens to critically question and clarify misconceptions that emerged. As a senior researcher from Indonesia, MH, in turn, guided the process by delivering vital and relevant insider input based on his knowledge and experience of the local and cultural dynamics.

Results

Demographic findings

The study included 18 respondents from 18 districts throughout Aceh province (see table 1).

Table 1

Participants' characteristics (N=18)

We conducted 18 interviews that lasted 23–90 min (50.72 min on average). Sixteen of the 18 respondents held leadership positions at their current workplace, such as head of primary care (4/18), chief of quality assurance (7/18), and chief of individual services in primary care (5/18). Finally, 11 doctors worked in rural settings and seven worked in remote areas.

The leadership attributes perceived as most essential

In this section, we analysed and clustered attributes that respondents flagged as having a ‘high’ need, that is, more than 10 responses for further investigation in the interviews, and presented them in table 2. For a complete overview of needs, see online supplemental appendix 3.

Table 2

Leadership attributes that received the highest scores (N=18)

The identified needs as represented in table 2 were subsequently used for the in depth interviews with the respondents. Based on the interviews conducted, we identified a couple of themes that described the qualities of a good physician leader in an LMIC setting like Aceh.

Qualities of a good physician leader commensurate with LEADS

Being culturally sensitive

The doctors we interviewed faced unique obstacles because each community where they worked had its own culture. One respondent, for instance, explained how they accommodated the community’s wish not to receive primary care on Wednesdays:

It is a local custom, like a mystical one, that the community will not go to primary care on Wednesdays [Participant 18].

In some remote and very remote areas, cultural beliefs were so strong that the communities did not allow doctors to visit them, because they did not believe in modern medical treatment and the community preferred traditional healers:

In very remote villages, some people still believe in mystical things. Also, they have a higher regard for traditional healers than for doctors [Participant 5].

I visited a sick woman who had recently experienced a miscarriage at her home. She did curettage at the traditional healer. I got information that the traditional healer used cogon grass afterwards. It was used to treat patients who miscarried and inserted into the female reproductive system [Participant 15].

Doctors also needed to understand how cultural values influenced their impact. More specifically, factors such as knowledge and daily habits could influence how people responded to healthcare services since beliefs, habits and culture were so firmly ingrained:

People do not want to use the toilet; they still perform open defecation. Also, they do not mind defecating in the river since they have had this habit for as long as they can remember [Participant 12].

To overcome such cultural issues in the communities, doctors experienced a need to develop specific skills, of which communication and listening skills were indispensable. Adaptive communication was considered key.

Some people in remote areas are still illiterate. We must pay more attention to them, explain things slowly, frequently and in simple words [Participant 3].

Doctors who had the ability to communicate in the local language, for instance, were more readily accepted:

It is easy for me to approach them [society] since I can speak their native tongue. I have a strong emotional bond with them. Ultimately, they want to take part in primary care activities [Participant 6].

Doctors stressed they had to invest in listening; people in rural and remote areas were more likely to tell their stories and they were proud and happy if a doctor would listen to them, because, in their view, doctors had a high hierarchical position:

We need to listen more, drop our egos, and adapt to their culture so that when we provide therapy or education, they are willing to accept it [Participant 3].

Another way to address cultural issues was through good collaboration with members of the community. Good networking abilities provided doctors with access to potential resources that could help them to complete their tasks:

In the evening, I referred the patient to a hospital that was over 40 kilometres away from primary care. Because of the heavy rain, the ambulance was unable to pass the road, which was coated thick with muck. As the ambulance got stuck in the mud, the community helped us to pull the ambulance by using their personal vehicle (owned by one of the community members) [Participant 14].

Especially in Aceh province, where culture is heavily influenced by Islamic values, good collaboration with local religious leaders was considered key. Because these leaders have a high status in the community, their opinions carry the weight of a community decision. Participants therefore deemed it crucial to involve religious leaders in all primary care activities:

People refuse to be vaccinated [because of forbidden and permitted issues]. So, we sought to educate not only the public, but also the religious leaders. These leaders then wanted to assist us in educating the community. Finally, most individuals wanted to embrace the vaccine, albeit a tiny percentage still refused [Participant 16].

Finally, participants argued that doctors are likely to have more impact on health-related decision-making when they are capable of building bridges between religious beliefs and medical practices. Consider the following experience shared by one of the respondents:

When I educate the community about the importance of vaccination [the people in that area are quite religious], I must discuss vaccines from both health and religious perspectives. I noticed people were responding positively [Participant 13].

Having a strong character (perseverance)

A second important quality was that doctors who work in limited settings must demonstrate character.

To reach the farthest village, we must travel by river in a traditional boat for over three hours. The boat was not available daily. It was quite dangerous if it rained severely. A phone signal was quite difficult to come by in this area. Before travelling to this place, I posted a status update on social media to let my family know I was there [Participant 14].

We made a home care visit. The challenge comes from having to go up- and downhill and walk for long distances. I also frequently ride a motorcycle and once had a miscarriage after I had fallen from a motorcycle [Participant 1].

Other conditions, too, such as firm cultural beliefs and deep-rooted traditions, could complicate working in a rural and remote community. Consequently, doctors needed to be patient and understand that building trust required a long time investment:

We did not build trust in a day or two. Doctors built it when they treated patients in primary care and when doctors involved themselves in the community (outside of working hours). Doctors also need to be patient and make sure that their societies understand their explanations. Take, for instance, the vaccination programme: the people initially resisted, but now they widely accept vaccination [Participant 2].

Being creative and flexible

A third quality that was frequently mentioned by participants was that doctors must be creative and adapt their thinking to find the solution. Geographical obstacles, for instance, proved a serious issue that called for creative solutions:

There were two rural areas on the river’s bank that were difficult to reach. On a raft, we once transported a woman who was about to give birth [Participant 10].

Likewise, limited healthcare resources such as materials and anything else required to provide healthcare services called on doctors’ creativity:

A child was brought to primary care because of maggots lodged in the child’s head. His head swollen, and we found about 30 maggots. I then treated him using simple tools available in primary care. I treated the patient for 20 days until the wound healed [Participant 1].

Further restraints, such as a poor internet connection, caused doctors to experience difficulties in expanding their knowledge and competence:

The mobile phone signal is poor in this village. As a result, I must either set the phone near the window or travel to a neighbouring village in the next sub-district to gain a signal strong enough to be able to join the online seminars [Participant 18].

Despite these many difficulties, doctors could still develop programmes aimed to ease problems, simply by being creative:

We found that traditional healers assisted many mothers in giving birth by providing a ‘one-package service’, which included helping with childbirth, washing the mother and baby’s clothes and taking care of the baby. As a result, we approached traditional healers and the community. When they were helping deliver, a midwife had to accompany the traditional healer. This programme was running well. We can see this from the decline in maternal and infant mortality rates [Participant 12].

Another respondent developed a programme in response to access obstacles, which was even rewarded by the government:

As some areas are hard to reach because of steep hills and rivers [people have to use a boat to cross the river], we then created a programme called ‘Saweu Ureung Saket’ [English: visiting a sick person]. So, if someone gets sick, we visit the patient(s) and provide health services for them. This programme was included in the 45 national-level innovation programmes of the Ministry of Health of the Republic of Indonesia [Participant 10].

Besides developing programmes for communities, respondents also built creative programmes for their staff:

To boost literacy, foster enthusiasm for learning, and practise public speaking, we created ‘Senin Pintar’ [English: Smart Monday]. So, on Mondays, the primary care staff have to present one learning material. They can choose any material linked to health services or even general information, such as cleaning a wound or even how to turn the computer on or off [Participant 6].

A final quality flagged by respondents was that doctors must be flexible and willing to compromise or adapt to the viewpoints of the local people:

Because most of the inhabitants here work as farmers and gardeners, all of their activities take place in the morning. As a result, we select the best moment to implement a health promotion campaign. We have to be flexible. We then adjusted the schedule to their working hours [Participant 4].

Ability to perform multiple tasks (versatility)

First of all, we observed that doctors in rural or remote areas performed a dual role in their workplace, namely that of clinician and leader. When only one or two doctors were available in primary care, the doctor would assume many responsibilities. Doctors, moreover, held a special position among other health professionals; people respected and considered their opinions more than those of other medical personnel. As a result, doctors had to be involved in all primary care activities:

I double-checked the medical records and ensured the integrated service posts in some villages. I had to go to the community, also do health promotion, be involved in environmental health activities, nutrition education, and much more. Although there was someone in charge, I remained involved [Participant 1].

Discussion

Summary of findings

In this study, we explored the leadership competencies that are needed in a rural and remote area of Indonesia as an LMIC. The LEADS framework evidently had face validity as the results from the interviews bore a strong degree of similarity to LEADS principles. We found that the ‘well-prepared doctor’ had to be culturally sensitive with specific knowledge of local cultures and habits, preferably speaks local languages and is immersed in, and connected to the communities served and their religious leaders. In addition, the well-prepared doctor must have developed a strong character to cope with limited resources and infrastructure. They also must be patient and understand that building trust among rural and remote community with strong cultural views and deep-rooted traditions takes time. Because of the hard tasks and often inadequate infrastructure, the well-prepared doctor is flexible and adaptive, but above all, can be a creative problem-solver.

Reflection based on existing literature

Patients’ culture influences doctor-patient communication.23 Doctors who learn and practise cultural sensitivity are better equipped to minimise such barriers between physicians and their communities, increase understanding of diversity, and resolve miscommunications and mistrust.24 In Aceh’s context, doctors were more likely to influence people’s health-related decisions when they were capable of building bridges between religious and modern health perspectives. More specifically, our study showed that an Indonesian doctor in a remote area was most effective when they complemented general cultural competencies with specific knowledge of the local culture and religion.

It is also important that doctors realise that informal communication is more important than formal communication. Traditional wedding receptions and harvest season celebrations, for instance, are ideal opportunities to connect with people in these societies. The doctors in our study also emphasised the competence of listening. Being immersed in the local communities may be a part of deep listening, which composer Oliveros has defined as listening which ‘digs below the surface of what is heard, unlocking layer after layer of imagination, meaning, and memory down to the cellular level of human experience’.25 In addition to this, doctors in low-resource settings require social support. To achieve desired outcomes in rural or remote primary care, community participation is critical.26 27 Co-creation and community-based participation are methods to ensure that people have a sense of belonging to primary care and believe that they can fully contribute and take part in health programmes.27 This suggests that the success of the health agenda depends on how effective doctors are at incorporating their communities into all agendas in primary care.

Compared with urban places, doctors in low-resource settings face many challenges during their medical practice. Hence, they need to develop creative problem-solving skills. Requiring a complete break with conventional thinking, creativity typically starts with a divergence of ideas and is followed by attempts to attain convergence (agreement).28 Several scholars have argued that, considering the complex nature of the health system, leaders in healthcare should offer the appropriate conditions to help cultivate creative behaviours.29

Some developed countries, such as Canada, have introduced the LEADS framework which represents a common language and set of expectations for leaders within the health system. Similar initiatives have been launched in Australia (called ‘Health LEADS Australia’) and the UK (known as the ‘Healthcare Leadership Model’). Consequently, their medical school curricula have adopted the LEADS standards as a guide to develop physicians and to help shape the professional development of practising physicians.30 This study contributes new insights that may expand the comprehensive LEADS framework to include specific points of interest for the curricula in LMICs.

Our study has several strengths. First, we performed the research in all districts of Aceh province in order to obtain a variety of perspectives from respondents from various workplaces. We interviewed participants to collect detailed and in-depth data based on respondents’ experiences. Since we did not restrict the interviews to the questions in the interview guide, our data were a rich source of human experiences. As with any study, however, we also identified limitations. The findings should be interpreted and applied within the context, as we based them on a relatively small sample of participants from Aceh province (one doctor per district). As such, they may not represent the entire range of geographic diversity and, consequently, they may not be entirely generalisable to other LMIC contexts.

We recommend that the leadership (development) should also be studied in other LMICs with different cultures so that generalisable features and specific needs for certain contexts can be identified. Further, we recommend the need for additional studies to explore how physician leadership can be developed in specific settings, such as conflict areas and areas prone to natural disasters.

Conclusions

In conclusion, we used a phenomenological qualitative approach, to investigate doctors’ perceptions of existing and needed physician leadership competencies in primary care settings in low-resource, rural and remote areas of Indonesia. The participants used the LEADS framework to describe their work experiences, which showed that doctors in remote and rural settings already apply physician leadership concepts in their work. Despite a heavy workload, they deliver a large range of clinical and non-clinical services. Furthermore, the cultural and infrastructural challenges in rural and remote LMIC contexts, demand that special attention be given to three specific domains within the LEADS framework: cultural sensitivity, character building and creative problem-solving combined with flexibility.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by Medical and Health Research Ethics Committee (MHREC), Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada - Dr. Sardjito General Hospital, Yogyakarta, IndonesiaRef. No. : KE/FK/0609/EC/2021. Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors wish to thank participants for this study.

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.

  • 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 @jobusar

  • Correction notice This article has been corrected since it first published. Affiliation has been updated in 'Correspondence to' section.

  • Contributors FM and JB developed study idea. FM recruited participants and performed data collection. All authors performed data interpretation and analysis. JB, FS and MH provided manuscript revisions. All authors contributed to the edits to the manuscript. All authors provided their final approval for publishing of the manuscript and are responsible for the overall content.

  • 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.

  • 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.