Article Text
PDF
Abstract
Objectives Monitoring instructions related to adverse drug reactions (ADRs) are not always clearly described in clinical practice guidelines (CPGs) and not always easily applicable in daily clinical practice. The aim of this study was to assess the clarity of presentation and the applicability of ADR-related monitoring instructions in CPGs for children and adolescents treated with antipsychotic drugs.
Setting Guidelines from different countries were selected, and monitoring instructions for 13 ADR-related parameters were assessed.
Primary and secondary outcome measures To assess the clarity and the applicability of the sections concerning monitoring instructions in each CPG, the Appraisal of Guidelines for Research and Evaluation instrument was used. To assess the clarity and the applicability of the monitoring instructions for each ADR-related parameter, the Systematic Information for Monitoring score was used.
Results Six CPGs were included. Overall, the presentation of the monitoring instructions in the different CPGs was clear; three CPGs scored >75%. All CPGs scored lower on applicability, as, for example, the barriers and facilitators were poorly described. The number of ADR-related parameters included in the CPGs varied between 8 and 13. Why and what to monitor was always described for each parameter. When to start monitoring was also often described (90.2%), but when to stop monitoring was less frequently described (37.4%).
Conclusions The CPGs differed on the parameters that needed to be monitored. Overall, the monitoring instructions were clearly presented, but improvement in their applicability is required. By improving the monitoring instructions, CPGs can provide better guidance on monitoring ADRs in daily clinical practice.
- child & adolescent psychiatry
- adverse events
- protocols & guidelines
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information. The guidelines included in this study are freely available on the internet.
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
Strengths and limitations of this study
To the best of our knowledge, this is the first study that assessed the clarity and applicability of adverse drug reaction-related monitoring instructions in guidelines for children and adolescents treated with antipsychotic drugs.
The Appraisal of Guidelines for Research and Evaluation instrument and Systematic Information for Monitoring scores were used to assess the content and quality of the monitoring instructions in the clinical practice guidelines.
Scoring was performed by two reviewers independently, and discrepancies were discussed and resolved by consensus.
Scoring of the clinical practice guidelines and individual monitoring instructions remains partly subjective.
By assessing the clinical practice guidelines, it becomes clear whether and on which topics these clinical practice guidelines need to be improved for the use in daily clinical practice.
Introduction
Antipsychotic drugs are widely prescribed on-label and off-label to children and adolescents (hereafter referred to as children) to treat psychiatric disorders and symptoms, including attention deficit/hyperactivity disorder, irritability related to autism, mood disorders, anxiety disorders and tics.1 2 Evidence for the efficacy of antipsychotic drugs in this young and vulnerable population is not always available, while these drugs often cause bothersome, and sometimes severe, adverse drug reactions (ADRs).3 ADRs associated with antipsychotic drug treatment in children include, for example, weight gain, abnormal blood glucose levels, tachycardia, gynaecomastia, sexual dysfunction and movement disorders.3–5 Adequate monitoring of individual children is important when considering treatment initiation, for the early identification of the development of ADRs and to evaluate and, when needed, adjust the antipsychotic drug treatment to balance efficacy and safety.
Multiple clinical practice guidelines (CPGs) worldwide provide guidance to healthcare professionals on how to monitor for ADRs in children treated with antipsychotic drugs.6–11 These ADRs can be monitored through related parameters, including physical (weight, height, body mass index (BMI), waist circumference, blood pressure, pulse and ECG, laboratory (glucose, glycated haemoglobin (HbA1c), lipids and prolactin), and observational (extrapyramidal and prolactin-related, for example, gynaecomastia) parameters. There are differences between the CPGs in, for example, which ADR-related parameters should be monitored as well as the timing and frequency of monitoring. Regardless of these differences in the content of the instructions, all instructions aim to provide guidance to improve monitoring practices. Nevertheless, previous studies have shown that the monitoring of children treated with antipsychotic drugs is suboptimal and improved only marginally after the introduction of monitoring instructions provided in the CPGs.12–14
To enable the implementation of the monitoring instructions provided in the CPGs in daily clinical practice, first, the quality of the CPG is important, for example, the clarity of presentation. Second, each monitoring instruction included in the CPG has to be easily identifiable, clear, unambiguous and easy to apply. Each instruction should define why it is necessary to monitor, what to monitor, when to start, when to stop, how frequently to monitor, what to look for or what the critical values of the parameter are and how to respond to the monitoring results.15 Clear and easily applicable CPGs could enhance monitoring in daily practice and thereby contribute to the safety of antipsychotic drug use in children. However, previous studies have shown that the monitoring instructions are not always clearly described in the CPGs and that the instructions are not always easily applicable in daily clinical practice.16 17 This could lead to suboptimal monitoring frequencies and, consequently, to unidentified ADRs. Therefore, the aim of this study was to assess the clarity of presentation and the applicability of ADR-related monitoring instructions in CPGs for children treated with antipsychotic drugs.
Methods
Selection of the clinical practice guidelines
A search for CPGs that included ADR-related monitoring instructions for children treated with antipsychotic drugs was performed by using the literature database PubMed, the guideline-specific database of the Guidelines International Network (GIN) and the general search engine Google. The search terms for the CPGs were related to psychiatric symptoms and disorders, as well as antipsychotic drugs (online supplemental table 1).
Supplemental material
The CPGs had to meet five criteria to be selected. First, the CPG had to be available in Dutch, English or German so that the reviewers could understand it. Second, the publication had to be titled as a guideline, or there had to be a statement to the effect that this publication was a guideline. When identified through Google, the CPG had to be linked to a website of a national or international association for child and adolescent psychiatry or a national healthcare organisation. Third, the CPG had to include a section on antipsychotic drug treatment. Fourth, the CPG had to be focused on children (<18 years) or include at least one separate chapter on antipsychotic drug treatment in children. Finally, the full CPG had to be available in the public domain. The GIN database was not freely accessible and was, therefore, used to list published guidelines that were subsequently searched for on PubMed and Google.
A maximum of one CPG per country was included. When several CPGs emerged for the same country, those prioritised for this study were CPGs from child and adolescent psychiatry associations, CPGs for antipsychotic drug treatment instead of specific psychiatric disorders and CPGs with the most extensive sections in terms of follow-up and monitoring. There was one exception to the non-inclusion of more than one CPG for one country, namely when an organisation had published more than one CPG on the treatment and follow-up of children prescribed antipsychotic drugs, and these CPGs referred to each other. The selected CPGs could have been revised, and the most recent versions were selected. To determine which CPGs should be included, three authors (LM, JWA and ERH) discussed all selected CPGs.
Selection of the monitoring instructions
A monitoring instruction was defined as an instruction on measuring a physical, laboratory or observational ADR-related parameter before or during antipsychotic drug treatment. In total, 13 ADR-related parameters were included, based on the cardiometabolic, endocrine and extrapyramidal ADRs that can be caused by antipsychotic drugs.13 14 The physical parameters included were weight, height, BMI, waist circumference, blood pressure, pulse and ECG. The laboratory parameters included were glucose, HbA1c, lipids and prolactin. The observational parameters included were extrapyramidal symptoms (eg, parkinsonism and akathisia) and prolactin-related symptoms (eg, gynaecomastia, galactorrhoea and sexual dysfunction).
All monitoring instructions for children treated with antipsychotic drugs were obtained from the included CPGs by reading them, and the sections concerning the treatment, risks, pretreatment advice and follow-up were carefully examined. In addition, terms relating to the ADR-related parameters, monitoring of the ADR-related parameters and drug safety were searched for in the entire CPGs. General instructions on psychotropic medications were excluded; antipsychotics had to be explicitly mentioned.
Clarity and applicability of the clinical practice guidelines
To assess the clarity of presentation and applicability of the complete sections concerning monitoring instructions in each CPG, eligible parts of the Appraisal of Guidelines for Research and Evaluation (AGREE) II instrument and its complement the AGREE-Recommendations Excellence (AGREE-REX) instrument were selected.18 19 These instruments were designed by the AGREE Research Trust and are intended to help guideline users and developers to assess the methodological quality of guidelines.18
The two domains 4 and 5 of the AGREE-II instrument, with seven items in total, were considered eligible and relevant and therefore included for this study:
Clarity of presentation
The recommendations are specific and unambiguous.
The different options for management of the condition or health issue are clearly presented.
Key recommendations are easily identifiable.
Applicability
The guideline describes facilitators and barriers to its application.
The guideline provides advice and/or tools on how the recommendations can be put into practice.
The potential resource implications of applying the recommendations have been considered.
The guideline presents monitoring and/or auditing criteria.
Furthermore, three domains of the AGREE-REX instrument, with seven items in total, were considered eligible and relevant and therefore included for this study:
Clinical applicability
Evidence.
Applicability to target users.
Applicability to patients/populations.
Values and preferences
Values and preferences of target users.
Values and preferences of patients/populations.
Implementability
Purpose.
Local application and adoption.
For each included CPG, all items were scored based on a seven-point scale, ranging from 1 (strongly disagree) to 7 (strongly agree) for the AGREE-II instrument, and 1 (lowest quality) to 7 (highest quality) for the AGREE-REX instrument.
Clarity and applicability of the monitoring instructions
To assess the clarity of presentation and applicability of the monitoring instructions for each ADR-related parameter, the Systematic Information for Monitoring (SIM) score was used.15 With this score, the monitoring instructions were assessed based on six domains of information, namely: ‘what to monitor’, ‘when to start monitoring’, ‘when to stop monitoring’, ‘how frequently to monitor’, ‘what to look for/critical values of the parameter’ and ‘how to respond’. Each domain of information was allotted a score of either 0 (not described/not clearly described) or 1 (clearly described), resulting in a total score of between 0 and 6 (online supplemental table 2). The seventh domain, ‘why to monitor’, was assessed separately. Four domains of the SIM score were considered to be essential for the clarity and applicability of a monitoring instruction, namely ‘what to monitor’, ‘how frequently to monitor’, ‘what to look for/critical values’ and ‘how to respond’.15
The AGREE and SIM scores were determined by two authors independently (JWA and LM) and discrepancies were discussed and resolved by consensus. Final inconsistencies were discussed with the other authors until consensus was reached.
Data analysis
To assess the clarity and applicability of the complete sections concerning monitoring instructions in each CPG, the AGREE scores were calculated. Final scores for each domain were calculated as a percentage of the maximum score, using the following formula:
Maximum possible score=7 (strongly agree/higher quality) × number of items
Minimum possible score=1 (strongly disagree/lower quality) × number of items
In addition, the monitoring instructions of the 13 ADR-related parameters (see section Selection of the monitoring instructions) were assessed separately. The number of monitoring instructions was calculated for each CPG, it was determined which instructions were most often missing, and whether the reason for the advice to monitor was included. To assess the clarity and applicability of each monitoring instruction, the SIM scores were calculated. The instructions that were considered to be clear and applicable were those with a SIM score ≥4 that included at least the four essential domains ‘what to monitor’, ‘how frequently to monitor’, ‘what to look for/critical value’ and ‘how to respond’.
Results
In total, CPGs from six different countries that were retrieved through PubMed and Google searches were included after the selection criteria were applied (online supplemental table 3). Three CPGs from the Canadian Alliance for Monitoring Effectiveness and Safety of Antipsychotics in Children (CAMESA) were included, as the CAMESA had published three CPGs on monitoring and managing antipsychotic drug safety. These CPGs included one on monitoring the safety of second-generation antipsychotic drugs in children, one on managing metabolic complications and one on managing extrapyramidal side effects.11 20 21 Hereafter, these three CAMESA guidelines will be referred to and assessed as being one CPG.
The years of publication of the most recent versions of the CPGs were between 2011 and 2020. The scope of four CPGs involved monitoring for the safety of antipsychotic drugs in children, and the scope of two CPGs was the treatment of schizophrenia, of which one was a guideline for adults but included a chapter regarding children.
Clinical practice guidelines
For the clarity of presentation according to the criteria of the AGREE II instrument, three CPGs scored >75% (table 1). In most CPGs, the recommendations were specific and unambiguous (overall mean percentage: 75%), and the CPGs included easily identifiable tables listing the parameters that should be monitored (77.8%). However, the different options for the management of the condition or health issue were less clearly presented in three CPGs (Women’s and Children’s Health Network (WCHN), National Institute for Health and Care Excellence (NICE) and American Academy of Child and Adolescent Psychiatry (AACAP); overall mean percentage: 50%). This item, on management of the condition, included responses to abnormal test results, which were lacking, unclear or incomplete in these CPGs. All CPGs scored lower on applicability compared with the clarity of presentation. Especially the item ‘potential resource implications of applying the recommendations’ scored low (22.2%). This item included the cost information, which was extensively described in one CPG (NICE), but was lacking or insufficiently described in the other CPGs.
AGREE II
For the clinical applicability according to the criteria of the AGREE-REX instrument, all CPGs scored >65% (table 2). The evidence was not always clearly described (63.9%), as, for example, the consistency of results, bias of the included studies, directness of the evidence and magnitude of the benefits and harms were not included or not completely described in all CPGs. Most CPGs scored low on the item concerning values and preferences of the target users and patients/populations (47.2% and 33.3%, respectively). The method by which the values and preferences were assessed in the CAMESA guideline was the most clearly and explicitly described, as the evidence had been discussed by experts and consensus reached and focus group sessions that involved families of children with mental health disorders had been held.11 Regarding the implementability of the CPGs, all scored low on local application and adoption (22.2%), as, for example, the change required from current practice, relevant factors for successful dissemination and resource considerations needed to implement the recommendations were lacking or poorly described.
AGREE-REX
Monitoring instructions
The number of ADR-related parameters included in the CPGs varied between 8 (Accare) and 13 (German Association for Psychiatry, Psychotherapy and Psychosomatics (DGPPN); tables 3 and 4). Monitoring instructions for the parameters weight, BMI, blood glucose, lipids and prolactin were included in all CPGs (table 3). Monitoring instructions for the physical parameters pulse and the performance of an ECG were most often missing, namely in 50% (WCHN, CAMESA and Accare) of the CPGs. Although the CAMESA guideline stated that the performance of an ECG was beyond the scope of the current guideline, a reference to an article with guidance on ECG monitoring was provided.11 Monitoring instructions for waist circumference (WCHN and Accare) and HbA1c (WCHN and CAMESA) were missing in two of the CPGs, and monitoring instructions for height, blood pressure and the two observational parameters extrapyramidal symptoms (Accare) and prolactin-related symptoms (NICE) were missing in one CPG each. All CPGs described ‘why to monitor’ by explaining the ADRs that could be caused by antipsychotic drugs.
Scoring of monitoring instructions for each adverse drug reaction-related parameter
Scoring of monitoring instructions for each clinical practice guideline
Although the Accare guideline included the lowest number of monitoring instructions for ADR-related parameters (n=8), all instructions that were included were considered to be clear and applicable, as they had a total SIM score of ≥4 and included the four essential domains (table 4). For two CPGs (NICE and AACAP), none of the monitoring instructions were considered clear and applicable. The domain ‘what to monitor’ was clearly described for all monitoring instructions in the different CPGs, whereas there were differences between the other domains. Overall, when to start monitoring was clearly described (90.2%). All CPGs advised healthcare professionals to start monitoring blood glucose and lipids at baseline, except for the Dutch guideline, which recommended to start monitoring only when there were risk factors present, for example, a high BMI or familial hypercholesterolaemia. Four CPGs (WCHN, CAMESA, DGPPN and AACAP), did not clearly spell out when to stop monitoring, while the other two CPGs (Accare and NICE) advised monitoring for the duration of the treatment (overall mean percentage: 37.4%). Although the frequency of monitoring was described for most parameters (80.6%), these frequencies differed between the CPGs, as recommendations to monitor the laboratory parameters varied from half-yearly, yearly, an advice depending on the type of antipsychotic drug, to no advice on how to monitor beyond 1 year of antipsychotic drug treatment because of a lack of long-term evidence. Descriptions of what to look for or critical values (reference values) were missing for all laboratory parameters in three CPGs (DGPPN, NICE and AACAP; overall mean percentage: 68.7%), and how to respond if there were abnormalities in test results was not described for most parameters in these same three CPGs (58.0%).
Discussion
The clarity and applicability of ADR-related monitoring instructions in CPGs for children treated with antipsychotic drugs varied. Overall, the purpose and the presentation of the monitoring instructions in the CPGs were clear. However, the applicability could be improved, as, for example, the barriers, facilitators and cost implications were poorly described. In addition, recommendations on how to apply these instructions locally were missing or insufficiently described in all CPGs, as, for example, the changes required in current practice and relevant factors for successful dissemination were most often lacking. The applicability of the CPGs to healthcare professionals and children was more clearly presented than the description of the preferences of these two groups. Not only were there differences between the CPGs, but differences were also apparent in the completeness of ADR-related monitoring instructions of different parameters included in the same CPG. Although the number of parameters included varied between CPGs, all CPGs included instructions on weight, BMI, blood glucose, lipids and prolactin. Overall, what to monitor, when to start and the frequency of monitoring were most often described, while it was not always clear when to stop monitoring, what the critical values were or how to respond to abnormal test results. In particular, the applicability of the CPGs and of the individual monitoring instructions need to be improved for use in daily clinical practice.
Previous studies have also shown that monitoring instructions need improvement.16 17 22–24 Brouwer et al assessed the applicability of monitoring instructions in CPGs for elderly patients treated with antipsychotic drugs.23 The number of instructions and the monitoring frequencies also differed between these guidelines. In addition, the critical values and how to respond to abnormal test results were insufficiently described, in line with several CPGs included in the current study, while the CPGs for elderly patients were clearer regarding when to stop monitoring. However, not only the monitoring instructions of antipsychotic drugs in CPGs need improvement. A study by Nederlof et al regarding monitoring instructions for patients using lithium for the treatment of bipolar disorder and a study by Chiappini et al regarding symptomatic management of fever in children indicated that the clarity of presentation was good in most CPGs, but the applicability could be improved, which is also in line with the results of the current study.16 22 Moreover, the monitoring instructions in, for example, the summary of product characteristics also do not always provide adequate information, that is, easily applicable in daily clinical practice.24
The preferences of children, adolescents or their caregivers were poorly incorporated in the development process of most CPGs, or the extent to which the children, adolescents or their caregivers were involved remained unclear. Since CPGs provide recommendations and instructions to optimise patient care, it is essential to consider the preferences of patients. Previous studies have shown that the involvement of patient representatives is important because this can, for example, influence the scope of the CPG, encourage the use of plain language, emphasise the importance in real life and lead to incorporation of patient-relevant topics and outcomes.25 26 Via involvement of children, adolescents and their caregivers in the development process of monitoring instructions, the barriers to monitoring could also be discussed and possible solutions included in the CPGs. Barriers associated with children, adolescents or their caregivers could be a lack of knowledge, parents who resist or forget to obtain tests, or refusal by the child to take tests because of, for example, a fear of needles.27
The differences between the CPGs could be caused by several factors. First, the scope of the CPGs differed, as four CPGs focused on the safety of antipsychotic drug use in children, and two focused on schizophrenia. When the scope is broader and includes the overall therapy for a disorder, the focus on the monitoring instructions in the CPG could be less extensive, and this topic might be discussed in less detail. Second, five CPGs focused on children, while one CPG (DGPPN) focused on adults and included a section on children. Third, the year of last publication ranged from 2011 to 2020, and three CPGs had never been revised since the first publication. The quality of CPGs increased over time, which might result in higher quality in recent or frequently updated CPGs.28 This increase in quality over time is not in line with the findings of the current study because, although the CAMESA guideline was published in 2011 and could improve in several domains, overall, this guideline scored high and could potentially be used as an example to improve other CPGs. Fourth, one CPG (Accare) was written for local use but published on a national website for child and adolescent psychiatry so that it could be used by other healthcare professionals.6 By whom the CPG is developed could influence the clarity and applicability, as, for example, CPGs developed by international organisations seem to score high in those two domains, and these international organisations include a variety of expertise leading to a better understanding of, for example, implementation barriers.28 Finally, several other factors influence the development and content of a CPG, for example, differences in clinical practice between countries.
After development and publication of a CPG, the CPG has to be disseminated, adopted and incorporated into daily clinical practice. A review by Fischer et al provided information on barriers to guideline implementation.29 The barriers described were related to the CPGs, for example, access to the guidelines, poor lay-out, lack of evidence, plausibility of recommendations, lack of applicability and complexity.29 As shown in the current study, these barriers related to CPGs could also emerge in daily clinical practice when clear ADR-related monitoring instructions for children treated with antipsychotic drugs are required. Other barriers described include personal factors related to the physicians’ knowledge and attitude, for example, a lack of awareness, familiarity, skills or agreement with the guideline, or external factors, including a lack of resources or collaboration.29 Before a healthcare professional can adhere to a CPG, he or she must be aware of this guideline. A study by McLaren et al has shown that most child psychiatrists reported being aware of the CPGs for antipsychotic drug monitoring,27 while a study by Mangurian et al has shown that a large proportion of the primary care providers seem to be unaware of the consensus guidelines.30 Nevertheless, previous studies have demonstrated that monitoring rates were low and remained low after implementation of monitoring guidelines.12 14 As far as we know, no studies have been conducted that evaluated whether (a) more clear and applicable CPGs lead to better CPG adherence, and (b) whether good adherence to such clinical guidelines for the monitoring of ADR-related parameters indeed leads to better clinical outcomes such as less ADRs or better recognition and management thereof. Since awareness might not be the largest barrier for all healthcare professionals, the barriers other than awareness should also be investigated, for example, barriers related to the adoption, implementation and applicability in daily clinical practice. However, several barriers do not stand alone but could be related to each other. For example, when CPGs are evidence based and include well-founded advice, healthcare professionals might be more likely to concur and adopt the monitoring instructions, and if a CPG is easy to follow and apply in daily clinical practice, adhering to the monitoring instructions will be less time consuming. Therefore, clear and easily applicable CPGs might also decrease other barriers to monitoring.
A strength of this study was that only those CPGs including ADR-related monitoring instructions for children treated with antipsychotic drugs were examined. In addition, the AGREE-instrument and SIM-scores were used to assess the content and quality of the monitoring instructions in the CPGs. A limitation is that we selected six CPGs based on language (Dutch, English, German) and public availability. However, our main goal was to assess clarity and applicability in some widely used CPGs and not to identify all nor the best CPGs. Furthermore, a limitation is the possible subjectivity in scoring these CPGs. However, the scoring of the CPGs and individual monitoring instructions was performed by two reviewers independently, and discrepancies were discussed and resolved by consensus. Information could have been missed, but that would have meant that it had been overlooked by two reviewers and might also not be clear for daily clinical practice. The summary of product characteristics (SmPCs) of approved drugs is another important source of information for prescribing and monitoring drugs.24 We did not take these into account in the present study, since CPGs are more patient and treatment oriented and include relevant product-oriented information such as available from SmPCs. The applicability of monitoring instructions included in SmPCs is generally lower than that included in CPGs.24
Conclusion
The CPGs differed on the parameters that needed to be monitored and in the content of the monitoring instructions. Overall, the monitoring instructions in CPGs for children treated with antipsychotic drugs were clearly presented, while the applicability needed improvement. More information is required on how to put the recommendations into (local) practice, what the facilitators and barriers are and potential resource implications of applying these recommendations. Furthermore, the CPGs did not all clearly describe when to stop monitoring, what to look for or the critical values of the parameters and how to respond to abnormal test results. By improving the monitoring instructions, CPGs can provide better guidance so that monitoring practices can improve in daily clinical practice, and ADRs can be identified in a timely fashion.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information. The guidelines included in this study are freely available on the internet.
Ethics statements
Patient consent for publication
Acknowledgments
The authors would like to thank Karin Egberts for her help with the German guideline.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @robheerdink
Contributors LM, JWA, EvdB, TCGE and ERH contributed to the plan and design of the study. LM, JWA and ERH analysed the data. LM drafted the manuscript and was in charge of study planning. LM, JWA, EvdB, TCGE and ERH contributed to the interpretation of the results and critical revision of the manuscript for important intellectual content and approved the final version of the manuscript. LM and ERH are the guarantors of this paper.
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.
Disclaimer This research was solely funded by Utrecht University.
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.