Article Text

Patient perceptions and experiences with medication-related activities in the emergency department: a qualitative study
  1. Birgitte Zahl-Holmstad1,2,
  2. Beate Hennie Garcia1,2,
  3. Tine Johnsgård1,2,
  4. Eirik Hugaas Ofstad3,4,
  5. Elin Christina Lehnbom2,5,
  6. Kristian Svendsen2,
  7. Torsten Risør4,6,
  8. Renata Vesela Holis1,
  9. Renate Elenjord1,2
  1. 1Hospital Pharmacy of North Norway Trust, Tromsø, Norway
  2. 2Department of Pharmacy, UiT The Arctic University of Norway, Tromsø, Norway
  3. 3Department of Medicine, Nordland Hospital Trust, Bodø, Norway
  4. 4Department of Community Medicine, UiT The Arctic University of Norway, Tromsø, Norway
  5. 5Department of Health and Caring Sciences, Linnaeus University, Kalmar, Sweden
  6. 6Department of Public Health, University of Copenhagen, København, Denmark
  1. Correspondence to Birgitte Zahl-Holmstad; birgitte.zahl-holmstad{at}sykehusapotek-nord.no

Abstract

Background Emergency department (ED) pharmacists reduce medication errors and improve quality of medication use. Patient perceptions and experiences with ED pharmacists have not been studied. The aim of this study was to explore patients’ perceptions of and experiences with medication-related activities in the ED, with and without an ED pharmacist present.

Methods We conducted 24 semistructured individual interviews with patients admitted to one ED in Norway, 12 before and 12 during an intervention, where pharmacists performed medication-related tasks close to patients and in collaboration with ED staff. Interviews were transcribed and analysed applying thematic analysis.

Results From our five developed themes, we identified that: (1) Our informants had low awareness and few expectations of the ED pharmacist, both with and without the pharmacist present. However, they were positive to the ED pharmacist. (2) Our informants expressed a variation of trust in the healthcare system, healthcare professionals and electronic systems, though the majority expressed a high level of trust. They believed that their medication list was automatically updated and assumed to get the correct medication. (3) Some informants felt responsible to have an overview of their medication use, while others expressed low interest in taking responsibility regarding their medication. (4) Some informants did not want involvement from healthcare professionals in medication administration, while others expressed no problems with giving up control. (5) Medication information was important for all informants to feel confident in medication use, but the need for information differed.

Conclusion Despite being positive to pharmacists, it did not seem important to our informants who performed the medication-related tasks, as long as they received the help they needed. The degree of trust, responsibility, control and information varied among ED patients. These dimensions can be applied by healthcare professionals to tailor medication-related activities to patients’ individual needs.

  • Emergency department
  • Pharmacists
  • Qualitative research
  • Medication safety
  • Patient Preference

Data availability statement

No data are available.

http://creativecommons.org/licenses/by-nc/4.0/

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

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

  • Emergency department (ED) pharmacists reduce medication errors and improve quality of medication use. However, patient perceptions and experiences with the ED pharmacist have not been studied.

WHAT THIS STUDY ADDS

  • The patients were positive to the ED pharmacist, but had low awareness and few expectations of the pharmacist’s role. We found that the four dimensions trust, feeling of responsibility, need for control and information need varied among ED patients. These dimensions can be applied by healthcare professionals to tailor medication-related activities to the patient’s individual needs.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • Future studies should focus on developing standard questions for the assessment of the four dimensions, as well as identifying how the healthcare professionals in practice can tailor medication-related activities in a busy ED in the everyday life.

Introduction

Medication errors (MEs) and adverse drug events (ADEs) are common among medication users1 2 and may lead to hospitalisations, patient harm and increased costs.3–5 Care transitions are particularly vulnerable to MEs,6 hence the emergency department (ED) is an important place to detect MEs and ADEs. It is crucial to identify and resolve medication discrepancies as well as suboptimal and inappropriate medication use to reduce MEs and ADEs. This can be done through medication reconciliation and medication review. Studies show that employing clinical pharmacists in the ED can reduce MEs and improve quality of medication use.7

Although the ED pharmacist is an integrated team member in countries such as USA and Australia,8 9 this is not the case worldwide. In Norway, only a few hospitals have ED pharmacists employed, and research on this field is scarce. Consequently, the ED physician has the responsibility to perform medication-related tasks such as medication reconciliation, prescribing/deprescribing and patient counselling, in collaboration with other team members. An intervention study was conducted in Norway to examine the impact of introducing clinical pharmacists as part of the ED interdisciplinary team on various outcomes.10

When planning new interventions, it is crucial to involve stakeholders affected by the intervention to maximise the potential of it.11 Patients are often key stakeholders11 and their unique insight, knowledge and experiences can be used to improve a service.12 It is also important to identify how different stakeholders experience the intervention and collect advices for amelioration of the intervention.13 Involving stakeholders can optimise the intervention before, during and after implementation and also contribute to successful implementation.11 13 14

Research on patients’ experiences and perceptions of the ED visit in general, reveals that good communication, sufficient information, having one’s emotional needs met, and feeling respected are important aspects for the patients.15–19 Patients are mostly satisfied with their ED visits; appreciating the physicians and nurses and having trust in them to provide care.16 18–20 When it comes to clinical pharmacist services, studies show high patient satisfaction in both primary and secondary care settings.21–24 Although some studies show that the role of the pharmacist in clinical care is unclear to patients, patients feel safer and more involved in their own treatment, as pharmacists provide necessary information about their medications.25–27 However, patient experiences and perceptions with meeting a clinical pharmacist in the ED have not yet been studied.

The aim of this study was to explore patients’ perceptions of and experiences with medication-related activities in the ED before and during an ED pharmacist intervention.

Methods

Study design, setting, data collection and recruitment

The intervention was carried out in three EDs, where the pharmacists were present in the EDs during the busiest hours of the day. The intervention was pragmatic, meaning that the ED pharmacists performed medication-related tasks close to and in contact with the patient based on the patient’s individual need.10 These types of activities included medication reconciliation, medication review and medication information. In addition, the pharmacists collaborated closely with other healthcare professionals (HCPs) in the EDs as needed, for example, by engaging in discussions about patients, providing medication therapy recommendations or answering questions regarding medication management.

We conducted semistructured individual interviews with patients acutely admitted to the ED in one intervention study site. At this ED about 12 000 patients pass through annually. Interviews were conducted both before the ED pharmacists were introduced (January–September 2020) and during the period with ED pharmacists present (November 2021–January 2022). We completed 24 interviews, 12 in each period.

We developed two interview guides, one for each period, which were reviewed by the research group, including a layperson (see online supplemental file 1). The interview guide for the first period concentrated on how patients experienced the medication focus in the ED including medication-related activities, and how they perceived the role of a future pharmacist. In the second period, the interview guide still focused on what happened with medications during their ED visit, but also on the pharmacist’s role in medication-related activities in the ED. The interview guides were dynamic and adjusted consecutively during both periods.

Supplemental material

We recruited patients who used prescription medications regularly. During the first period, patients were recruited at the observation ward connected to the ED, where patients are admitted when their condition must be assessed and observed more closely before discharge or transfer to another ward. Admission to the observation ward demands cognitively adequate patients who are able to take care of themselves. During the second period, we recruited patients who the ED pharmacist had been involved with. Patients were asked to participate in the study the day after admission while their ED experience was still fresh. Before approaching the patient, we consulted the ward nurse to ensure the patient was cognitively adequate and able to participate in an interview. We recruited patients until we had gained sufficient information power28 and richness in our data material.29

The interviews took place in a quiet and private room in the hospital. The interviewer (BZ-H) was a female clinical pharmacist with 7 years of experience from a hospital setting. The interviews were audiorecorded and transcribed non-verbatim.

Data analysis

The data were analysed applying thematic analysis inspired by the principles of Braun and Clarke29–31 and carried out in three phases (see figure 1). Interviews in the first period were preliminary analysed before the interviews in the second period were conducted (data analysis phase 1). After the interviews in the second period were conducted, the data analysis started with listening to the audio recordings thoroughly before transcription (data analysis phase 2). As there were clearly similarities in the data material in the two periods, the 12 interviews from the second period were incorporated in the original tentative analysis. Finally, the data analysis was completed with all the 24 interviews included (data analysis phase 3). The data analysis was an iterative process. Online supplemental file 2 provides an example of codes and subthemes in a theme.

Figure 1

The data analysis was carried out in three phases and was an iterative process. This figure shows the different phases and steps of the data analysis process. *Steps of thematic analysis, as described by Braun and Clarke.30 31

The main researcher (BZ-H) coded the data and developed the themes (steps 1–3), while the research group participated in reading, discussing, reviewing, naming and writing the themes (steps 1, 4–6). NVivo V.12 and MindManager 2020 were used in the analysis process, in addition to pen and paper.

Patient and public involvement

A patient representative is part of the project group for the main research project.10 This layperson contributed to the development of the interview guides and the consent form for the patients, thus improving readability and comprehensibility. The layperson participated in analysing the data by reading some of the transcribed interviews and engaging in discussions about the findings. Due to COVID-19 restrictions, it was not possible to include the layperson in the data collection.

Results

The age of informants ranged from 23 to 85 years old and half of them were females (see table 1). For some informants, this was their first visit to the ED while other informants had visited several times. Eight of the 24 informants had some sort of professional affiliation with the healthcare system, for example, by being a nurse or a medical secretary. The length of the interviews ranged from 7 to 83 min, with an average of 34 min.

Table 1

Characteristics of informants during interview period 1 (January–September 2020) and interview period 2 (November 2021–January 2022)

We developed five themes representing our informants’ experiences and perceptions regarding medication-related activities in the ED (see figure 2). The themes are presented with illustrative quotes. We identified few differences between the first and second period, and these differences were only detected in the first theme ‘Low awareness and few expectations of the ED pharmacist’. Consequently, we do not differentiate between the two periods in the presentation of the remaining four themes.

Figure 2

The five themes identified from thematic analysis of interviews with patients about medication-related activities in the ED with and without an ED pharmacist present. ED, emergency department.

Low awareness and few expectations of the ED pharmacist

When asked about medication focus and medication-related activities in the ED, informants had few expectations of pharmacists in the first period and little to say about pharmacists in the second period.

In the first period, the informants were positive towards the intervention with an ED pharmacist when it was explained to them, but had few ideas on how the pharmacist could contribute in the ED. Most of them were familiar with the profession and knew that pharmacists have knowledge about medications, but it was difficult to suggest possible work tasks in the ED.

I have heard about pharmacists, but I know little about them. Irene

No, I don’t know [which other tasks the pharmacist can contribute with]. But I do know that pharmacists have a lot of knowledge about medications. Steven

Other informants did not know what a pharmacist was and consequently did not have any input. Anne stated that pharmacists ‘do a heck of a good job—they are good at drawing blood’, so it seems like she thought about a phlebotomist when she was asked about the pharmacist.

In the second period, the informants who remembered having met a pharmacist found talking to them okay. Some experienced that the pharmacist was more thorough than the physicians when they talked about the medications, and that they received some medication information.

I thought it was nice. Because there is a different focus on medications. It was not just the physician asking “have there been any changes [since last time]?”. […] [The pharmacist] was more thorough and focused on getting it [the medication list] correct. Lisa

Some informants expressed surprise about meeting a pharmacist in the ED and a few of them were not even aware of meeting one.

I thought it was very… I was surprised. I am not used to meeting someone who focuses solely on medications. Jim

Yes, she asked me what [medication] I was using. […] I think she was a nurse. […] No, I did not [talk to a pharmacist]. Brenda

Despite that the informants did not have many thoughts about the pharmacist’s role in the ED, they had no objections or negative opinions about the service the pharmacists provided.

Trust and mistrust in HCPs and the healthcare system

In general, most informants expressed a high level of trust in the healthcare system and HCPs. They relied on HCPs at the hospital knowing which medication(s) they used at home and they expected to get correct medication(s) during hospitalisation.

I’ve been in such a bad shape, that I haven’t been able to talk to anyone in a few days. In these situations, you just take whatever [medications] they give you and assume that they are doing the right thing and know what [medications] you are supposed to have. Margot

The informants also had a high level of trust in electronic systems and medical records. They thought that the hospital and its employees knew about every medication they used and had ever used. They had the impression that all information from primary care to secondary care was updated digitally and that their medication list would automatically be up to date.

They get the information they need through their systems. And they showed me a list, a medication list, but I don’t use all of them [the medications]. So, I felt that they had good control. Anne

However, some informants said things that indicated that they did not have full confidence in HCPs and the healthcare system. Sarah feared she would not be taken seriously if she asked questions about her medications and potential side effects, while Tina did not feel that the HCPs trusted her. Both Edwin and Christine lost their trust in the healthcare system due to earlier mistakes; Edwin’s wife was misdiagnosed and Christine experienced a ME.

It was the hospital’s fault that the medication list was wrong in the first place. […] According to the list, I got 75 µg levothyroxine every day for four weeks. It was supposed to be 25 μg levothyroxine two days a week. […] So, I guess that was the reason I hallucinated and thought that I was going to die. I think it’s the most terrifying experience I’ve ever had. Christine

Other informants had also experienced unfortunate episodes, but despite that, they still had confidence in the HCPs and the healthcare system.

I was given ibuprofen. […] I called the nurse and asked if it was a good idea to use ibuprofen as a painkiller. […] With the symptoms I had, stomach pain and stomach bleeding. […] But they had taken my haemoglobin, right, and it was ok. So maybe they just did not think about it [my other symptoms]. Betty

The trust and mistrust the informants have, may influence their ability to give up responsibility and control during their hospitalisation.

Taking responsibility for one’s own medications

When it came to taking responsibility for one’s own medications, there were various views among the informants on how important that was. Some of the informants expressed a feeling of obligation to know what medications to take, why and how.

I think patients owe it to themselves to know what [medications] they are putting in their mouth. Irene

Others took quite the opposite approach, by presenting a lack of responsibility and taking things easy.

I don’t use any critical medication, so if I skip a dose one day, it is not the end of the world. I don’t know much about side effects and such, but that’s my own fault. I throw out the consumer medicine information leaflet. Robert

Matt showed an absence of responsibility regarding his medications when he didn’t know what or why he took it, and asked the interviewer if he actually needed the medications.

Some informants refrained responsibility regarding medicines during the hospitalisation by being obedient and did what the physician told them or took whatever the nurse gave them. Some expressed the opinion that you show that you are responsible by taking the medications as the physician has prescribed.

I think it’s really important that you take the medication as prescribed. Anything else is sloppiness. You should do as your physician says. Edwin

Need for control over medication administration

The informants had different need for control when it came to medication administration. When admitted to the hospital, patients are forced to hand over control for medication administration to HCPs which can lead to patients feeling submissive or disempowered.

It’s the feeling of losing control over your routines, because this is their [HCPs’] dominion. I haven’t been given any wrong medications, it’s just their arrogance. "We’re in charge here, boy." George

It can be a challenge for someone to give up control. Some of the informants wanted to remain partly in control by taking the medications they had brought with them to the hospital, and not have nurses involved in the administration process.

Because my INR was 2, she [the nurse] asked, “how many tablets of warfarin should you have then?” I said “give me two”, but she wanted to ask another nurse. But they must not interfere with my regimen, I know this myself. I’ve been doing this for three years. I was so annoyed. Jim

Some informants had a strong need to double-check that the medication or information they were given were correct. Sarah had experienced getting the wrong dose from the nurse but had luckily caught the error before she took the dose. She had felt very uncomfortable to correct the nurse.

Others expressed no problems with giving up control and just took the medication they were given, even though they thought it could be wrong.

This morning there was a tablet I didn’t recognize. And I don’t know what it was. […] But I didn’t ask. I just took the tablets from the medicine cup. Mary

Information is important, but the need for it differs

There was a big difference in how much information the different informants wanted and needed. Some thought that there was an information deficiency in the ED and described that they got almost no information, while others said they got enough information. No one expressed that they received too much information.

I want to know everything. Yes. I am a curios person. […] It has to do with me, so I want to know what they [the medications] do to my body, what they [health care professionals] fill my body with and how they [the medications] affect me. Eric

Simply put, I don’t need to know a lot about my medications. […] If they work, and I can go outside and chop wood or mow my lawn without experiencing discomfort afterwards, then everything is fine. Then I don’t need to know anything else. Freddy

Despite the differences in information needs, most of the informants agreed that there was minimal information provided about medication during the whole hospitalisation, both in the ED but also at the wards. The informants thought it was important to get information to feel confident in their medication use. They wanted information about how the medication works, why it is important for them to take it, what side effects they can expect, and how to deal with side effects. Most of the informants wanted a combination of oral and written information. That was because they could be too sick to comprehend and remember information provided, thus they needed the written information to refer to later.

Discussion

This is, to our knowledge, the first study that explores patients’ perceptions of and experiences with medication-related activities in the ED with and without a pharmacist present. We have identified low awareness and few expectations of the ED pharmacist among patients admitted to the ED. Our results show that trust, responsibility, control and information are important aspects to our informants when it comes to medication-related activities.

When we asked the informants about their perceptions of and experiences with meeting a pharmacist in the ED, they did not have much to say. However, the informants were positive towards the pharmacist, which is also supported by other studies.26 27 The fact that they had no particular opinions about the pharmacist’s contribution in the ED or which tasks the pharmacist should perform, can have several explanations. First, it is difficult for patients to differentiate between the different HCPs involved in care and their roles,25 32 consequently, it is challenging to have an opinion about the pharmacist in particular. It seems that for the informants it was not important who performed the different tasks in the ED, as long as they felt taken care of. When admitted to the ED, they were seriously and acutely ill, and were just grateful to get help. Second, there was little knowledge among our informants about what a clinical pharmacist is, and which tasks a clinical pharmacist can perform in the ED. This may be true for most patients, as studies report that patients are often surprised when meeting a pharmacist in a hospital setting, compared with in a community pharmacy, and do not know what to ask or expect from pharmacists in hospitals.25 26 The informants that had the most to say about the pharmacist’s contribution, were in fact HCPs themselves. This strengthens the assumption that you need to have knowledge about the different types of HCPs and what they do, in addition to the healthcare system, to have opinions and perceptions. Third, the informants may not be aware of what the different medication-related tasks in the ED are. As a consequence, they do not recognise the importance of tasks such as medication reconciliation, choosing the correct antibiotic and identifying non-adherence in the ED. In that way, they are not able to have opinions and perceptions about the value or need of an ED pharmacist. Taking these points into consideration, the questions are; ‘Do patients need to know which medication-related activities that are performed in the ED?’, and ‘Do patients need to know whether it is a physician, a nurse or a pharmacist who does them?’ From our results, it seems that the most important for patients is to feel safe and cared for. Consequently, we need to make sure that there is a high level of medication safety in the ED so that patients receive the correct medication and high standard care.

Our results suggest that medication-related activities in the ED must be tailored to meet the patient’s individual needs, as four out of five themes concerned dimensions important for our informants regarding medication-related activities. These four dimensions; trust, responsibility, control and information can also be looked upon as different dimensions of patient characteristics. Most of our informants expressed a high level of trust in the HCPs, even though some had experienced errors in the past. This may seem like a paradox, but maybe they do not have a choice. They are in a vulnerable situation with an acute health problem and need help. In addition, they do not know the system that well and may have different degrees of health literacy, which may influence their understanding of consequences regarding unfortunate incidents. The level of trust may also be linked to the degree of responsibility and need for control the informants felt. To give up control to HCPs, they must have some trust, and this is in line with a study that found that level of engagement in medication administration are influenced by patient’s trust in the system.33 This may explain why some of our informants were controlling or double-checking medication given by the nurse—they just could not let go of the control. Knowledge is essential for involvement18 and it makes patients empowered in decision making.34 For patients visiting the ED, studies have shown that sufficient information is important15 17 18 26 and patients’ need for it regarding medication is not met during their hospital stay.33–35 Our study confirms these findings, in addition to emphasising the importance of giving patients tailored information based on their individual needs and condition.33–35

One approach to tailoring medication-related activities in the ED could be to identify the patient’s level in each of the four identified dimensions. These dimensions can be applied by HCPs by assessing the ED patients’ views with regard to medications and medication management on arrival. In this way, medication-related activities can be individualised based on the patient’s preference. This is illustrated in figure 3, showing how four fictive patients may have different views on the four dimensions and contains two examples on how HCPs can customise the medication-related activities. One important thing to keep in mind is that the patient’s condition may affect the preferred level of involvement in medication management18 32 33 and the same patient may have different needs during various hospital visits. Consequently, the patient’s condition, views and needs must be identified at each ED visit. When the HCPs take the patients’ preferences into account tailoring the medication-related activities, the patient will feel that they are being heard and taken seriously. This will in turn make them feel empowered and become more involved in their own health. As the ED is a busy environment where time is scarce, figure 3 can also illustrate how to prioritise patients for involvement and medication information. Using resources on involving patients with no feeling of responsibility, no need for control, or with no information need is probably not an appropriate use of time, at least not during the ED stay. This time could rather be spent on patients who are sceptical, who want to take responsibility, maintain control or have a great information need.

Figure 3

Illustration of how four fictive patients may have various views on the four dimensions “trust in the health care professionals and health care system”, “feeling of responsibility regarding their own medication”, “need for control over medication administration” and “information need”. Two patient examples are elaborated on the right side on how health care professionals can tailor the medication-related activities to specific patients.

Future studies should investigate whether there is a relationship between the four dimensions trust, responsibility, control and information in order to tailor patient care more specifically. In addition, developing standard questions for the assessment of the four dimensions will be crucial, as well as identifying how the HCPs in practice can customise medication-related activities in a busy ED in the everyday life.

Strengths and limitations

The main strength of this study is the inclusion of patients both before and during the ED intervention. This enabled us to study the experience and perceptions with the ED pharmacist, both in patients who had experienced the pharmacist and those who had not. The main limitation is that we included patients visiting only one ED, which limits the generalisability of our results. However, we do not believe that this ED and its patients are very different from other EDs in Norway. Nor the study population, as our informants had a wide age and gender distribution with varied work background and different ED experiences. Consequently, our results may also be valuable for other EDs. Another limitation is the background of the interviewer and researcher (BZ-H) who is a pharmacist and also introduced herself as a pharmacist. This may have influenced thoughts and statements by our informants, as well as the analyses.

Conclusion

This study shows that patients in one ED in Norway are positive to including the pharmacist in the ED team, but seem to have low awareness and few expectations of the ED pharmacist regarding medication-related activities. It did not seem important for our informants which HCP that performed the medication-related tasks, as long as they got the help they needed. We also found that the degree of trust, feeling of responsibility, need for control and need for information vary among patients. This highlights the importance for HCPs to assess these four dimensions on arrival to the ED. Medication-related activities should be tailored to the patient’s preferences and needs, and also adapted to a busy ED setting. More research is needed to further develop this into standardised practice and patient care.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and the study is approved by the Data Protection Officer at Nordland hospital (No 28-19). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors would like to thank all study participants and hospital staff, including the ED pharmacists, for their contribution to this study. A special thanks to our patient representative Anne Lise Brygfjeld (ALB) for her contribution in the development of the interview guides and consent forms, in addition to participation in data analysis.

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 BZ-H (the study guarantor), RE, BHG, ECL and EHO were involved in study design. Data collection was performed by BZ-H, with support from BHG and EHO. Data analysis was performed by BZ-H, in collaboration with RE, BHG, ECL, EHO and TJ. BZ-H drafted the initial manuscript, while all of the authors contributed in further writing and editing of the manuscript. All authors read and approved the final manuscript. The patient representative, ALB, contributed in the development of the interview guides and consent forms, in addition to participation in data analysis.

  • Funding The study is funded by the North Norway Regional Health Authority grant number HNF1483-19.

  • Disclaimer The sponsor has no part in collection or analysis of the data, nor writing and reporting study conclusions.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

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

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