Article Text
Abstract
As many drugs in paediatrics are used off-label, prescribers face a lack of evidence-based dosing guidelines. A Dutch framework was developed to provide dosing guidelines based on best available evidence from registration data, investigator-initiated research, professional guidelines, clinical experience and consensus. This has clarified the scientific grounds of drug use for children and encouraged uniformity in prescribing habits in the Netherlands. The developed framework and the current content of the Dutch Paediatric Formulary could be used as basis for similar initiatives worldwide, preferably in a concerted effort to ultimately provide children with effective and safe drug therapy.
- Evidence Based Medicine
- Pharmacology
- Therapeutics
Statistics from Altmetric.com
Introduction
Drugs are granted a marketing authorisation after a comprehensive review and approval by Medicines Evaluation Boards, such as the Food and Drug Administration (USA) and European Medicines Agency (Europe). The most important part of this review is the analysis of benefits (efficacy) versus risks (safety) of use. Only if the benefits exceed the risks, a drug is granted a marketing licence (Product Monograph in the USA, Summary of Product Characteristics (SmPC) in Europe). Historically, most drugs are only licensed for use in adults, although these same drugs are used by children as well. This suggests that the high quality and safety standards applied to assess drug use in adults have not been applied to drug use in children, leading to off-label drug use in children.1 This situation has improved since the European Union Paediatric Regulation became effective in 2007.2 Aimed at ensuring that paediatric use is addressed in the authorisation application, all new drugs need to have a Paediatric Investigation Plan, before a marketing authorisation is granted, in a comparable way as the US Paediatric Research Equity Act.
Still, the paediatric community is left with many drugs for which paediatric use is not approved. This applies to 40% of all prescriptions at a general practitioner's office and up to 90% of prescriptions for hospitalised children.3–5 This leads to suboptimal treatment and exposes children to unknown risks of overdosing and potential adverse effects, and also to underdosing and lack of efficacy.6 Furthermore, prescribers are left in a no man's land of insufficient information, and have difficulty to decide on an adequate dose advice for their patients.7 ,8
Thus, current knowledge on paediatric pharmacotherapy is empirical, practice based and seldom systematically collected and disseminated. This has resulted in a variety of dosing handbooks, with substantial differences in the recommendations on dosing and administration routes, regionally, nationally and internationally.9 ,10
Founded in 2005, the Dutch Expertise Network on Paediatric Pharmacotherapy aimed to develop a national paediatric formulary, based on best available evidence. The aim of this paper is to present the unique development of a knowledge-based paediatric formulary for the Netherlands.11
Key features
Aim
The overall aim was to develop an openly accessible, web-based formulary containing best evidence-based, transparent and up-to-date drug-specific information, which was acceptable to paediatricians, hospital pharmacists and general practitioners.
Editorial board
Shared paediatric care in the Netherlands is organised around eight university paediatric hospitals. Representatives, preferably both a paediatrician and a pharmacist, from all eight clinics were invited to sit on the editorial board. In addition, specialists in all paediatric fields were invited, as well as general practitioners, clinical pharmacologists and community pharmacists.
Resources and funding
The work was done by a team of a coordinating paediatrician (0.2 fulltime equivalent (fte)), project manager (0.8 fte) and pharmacist (1.0 fte). The 35 members of the editorial board were not paid, but expenses made to attend the full committee meetings were reimbursed. The overall budget on an annual basis was €250 000 in the first 2 years, and currently €220 000. A grant from the Ministry of Health helped to set up the centre of excellence and to develop the first version of the paediatric formulary (2005–2010). A second grant from the Ministry of Health and a grant from the Dutch healthcare insurers and the Federation of Dutch Medical Specialists enabled us to retrieve additional evidence for our dose recommendations and balance the risks and benefits for the paediatric population, as is done for adults within the marketing authorisation process (2010–2012). As of 2013, a co-finance structure by professional associations and the branch organisation of healthcare insurers was set up to cover the costs of the maintenance of the formulary.
Scope
Scientific information on paediatric drug prescription is scarce in general, but high-quality comparative studies on different drugs for single paediatric conditions are even rarer. Thus, it is almost impossible to recommend one drug over the other based on solid comparative effectiveness and safety studies. Therefore, the editorial board decided to focus on providing dose recommendations and additional drug information only and leave the choice of most appropriate drug to the discretion of the prescriber.
Knowledge-based dosing recommendations: practice-based evidence supplements evidence-based practice
The Dutch Paediatric Formulary is a best-evidence formulary. This does not mean that every dose recommendation of the formulary is supported by high-quality clinical studies. It is almost impossible to make a paediatric risk-benefit assessment based on the same standards that are mandatory when assessing adult drug use. We are aware that many of our dosing recommendations therefore still bear a varying degree of uncertainty. ‘Best-evidence’ means that we do know the scientific background that supports paediatric use. This also implies that, in the face of low-quality evidence, expert opinion or a consensus of a group of experts is important12 in creating practical dosing advices, which can be implemented in clinical practice. Clinical findings, daily practices and cultural differences as well as applicability of research findings in a (Dutch) clinical setting are of utmost importance in developing and implementing a monograph. It is a common misconception that a recommendation supported by low-quality evidence implies a recommendation against use. The Dutch Paediatric Formulary deliberately chose to give insights in the limited amount of evidence available and create awareness rather than to reject paediatric use because of limited evidence. When the editorial board feels the need for a dose recommendation, but sound evidence is lacking, this is mentioned within the monograph.
Balancing licensed use and off-label use
Treating children solely with licensed drugs would seriously jeopardise treatment options. Off-label prescribing in paediatrics is therefore not a matter a free choice, but a bare necessity.
Evidence on drug use in children is mainly provided by academia and does not find its way into marketing authorisations. Therefore, it is important to realise that off-label is not the same as off-evidence,13 although this is suggested from a regulatory point of view. It is worth mentioning that on-label does not always mean evidence-based.14
Unbiased and transparent
Since the proposed drug monograph and risk analysis, which are based on available scientific evidence, are reviewed by the joint medical and pharmaceutical expertise of multidisciplinary, national and editorial board, the introduction of systematic bias, due to personal or commercial interests is prevented. Pharmaceutical industry representatives are excluded from the review process. The listing of literature references encourages users of the formulary to retrieve, judge and compare the evidence used and to control the actions taken by the editorial board.
Development from consensus based to knowledge-based dosing recommendations
Phase 1: Consensus based
Although we aimed to compile a completely evidence-based formulary based on systematic reviews, this was not feasible within the initial 2 years of funding. For pragmatic reasons, the editorial board decided to start with a consensus-based formulary: it would take years to start from scratch and systematically review all relevant drugs. The board was of the opinion that initially a complete, consensus-based formulary was preferred over an incomplete, best-evidence formulary. After deliberation on contents and gaining support for a national formulary, the 2007 version of the paediatric formulary of the Wilhelmina Children's Hospital (WKZ) in Utrecht15 was adopted as the initial reference for dosing recommendations in the new collaborative national formulary. This formulary was best known in the Netherlands and was broadly used by the paediatricians and pharmacists in the Netherlands (E-Supplement 1: contents of a drug monograph).
Supplemental material
Phase 2: Knowledge based
From this point onwards, the editorial board stated the ambition to retrieve the knowledge base for each dose recommendation and additional paediatric drug information in the formulary. With 550 drugs included in the first edition of the formulary, there was an urgent need to differentiate between the drugs for which the knowledge base was unknown and the drugs for which some evidence for paediatric use was already available. Drugs without any licence for paediatric use were prioritised. Overall, only 41% of the drugs listed in the formulary were licensed for children. For each drug on the priority list, a pharmacist searched and analysed the available scientific literature and guidelines following the Standard Operating Procedure (E-Supplement 2) and assessed the risks and benefits of use. The evidence was summarised in a drug monograph by the pharmacist, reviewed by the multidisciplinary editorial board, and if deemed necessary other experts were consulted. In order to be able to recapture at all times the scientific grounds for the drug monograph, all references and written notes were archived and extensive minutes of the expert committee review meeting were written and kept on file. If no evidence could be found at all, the dose recommendation of the WKZ Formulary,15 sometimes amended after board discussion, was maintained as ‘consensus dosing guideline’. After approval by majority of the editorial board, the monograph was published in the paediatric formulary (figure 1).
Supplemental material
Initially, we started documenting and summarising all the evidence appraised in evidence tables. Very soon, this state-of-the-art review strategy appeared to be very time consuming, and slowed down the development of the paediatric monographs. Therefore, the editorial board decided to omit this strategy and to give priority to the straightforward development of drug monographs.
When the first updating cycle in 2012 proved to be less time consuming than the initial review process (because the amount of ‘new’ evidence being limited), it was decided to comply again with the original documentation strategy for the appraisal of literature: all reviewed studies are now documented in risk-analysis documents, which summarise the parameters of the study, the observed effect and conclusions of the authors. Most important, the recommendations of the reviewed study for the drug monograph of the paediatric formulary are listed.
Implementation, improvement and use
The Dutch Paediatric Formulary was developed using a bottom-up approach. While stakeholders (like the Dutch government and the paediatric and pharmacist societies) endorsed the initiative, pharmacists and paediatricians from the hospitals were represented in the editorial board and were involved in the actual development of the formulary. This resulted in good support for the formulary by university paediatric hospitals and paediatric wards.
After the launch in 2008, the Departments of Paediatric Pharmacy of the eight academic paediatric hospitals acknowledged the formulary as the primary formal reference for paediatric dosing, which had a ripple effect on the use of the formulary also in the regional clinics. The formulary was also adopted as the official dosing guideline reference by the Dutch Paediatric Association, the Royal Dutch Pharmacy Association and the Dutch Association of Hospital Pharmacists. Another contributing factor was the restriction in Dutch legislation to off-label drug use: off-label drug use is only acceptable when it is recommended by official guidelines of professional organisations.
Numbers of visitors of the website and visited monographs are tracked by Google analytics. Two thousand five hundred unique visitors per day visited the website in the early years, and the number has risen to >6000 unique visitors currently. Eighty-two per cent of the visitors are returning visitors.
Comparison with other paediatric formularies
Other European countries have addressed paediatric drug use by developing a national paediatric formulary. The British National Formulary for Children (BNF-C)16 is the best known. In 2004, Italy launched the Guida all'uso dei farmaci per i bambini.17 The Medical Committee from Spanish National Paediatric Society (CM-AEP) developed and launched the Spanish Pediamecum18 in 2012. WHO launched the WHO Model Formulary for Children19 in 2010, based on the WHO essential medicines list; it does not include all drugs currently used in developed countries. The main difference between the Dutch Paediatric Formulary and these paediatric formularies is the comprehensiveness of the Dutch monographs and the transparency of the evidence used to develop the drug monographs. In fact, The Spanish handbook, the WHO formulary and the Italian guide refer mainly to other (outdated editions) paediatric dosing handbooks, such as the BNF-C, while the BNF-C does not reference its information sources nor provides references within the monograph. Therefore, it is difficult to find the underpinning evidence for dose recommendations and to know whether the SmPC text was the main source or whether other academic publications were used to support paediatric use, as done in the Dutch Paediatric Formulary.
Work in progress
Maintenance: All monographs are periodically reviewed and renewed every 5 years, considering all new available evidence. The original search is repeated, new evidence is added to the risk-analysis document and appraised, an updated monograph is drafted and reviewed by the editorial board.
Quality assurance: As of 2011, every change to the formulary is documented within the content management system of the formulary capturing date, the rationale for change, author and references.
Quality control: every change to the formulary must be authorised by second person before publishing.
Development of a dosing calculator: this project aimed at the development and implementation of a calculation interface which integrates the dosing recommendations of the formulary with clinical patient variables, thus resulting in an individual recommended dose. In concordance with European legislation marking has been applied for to ensure technical quality and safety of the calculator.
Dose adjustments in children with impaired renal function: this project aimed at elucidating available evidence and incorporating recommendations for dose adjustments in children with impaired renal function in the monographs.
Disseminating knowledge to parents and children: the information in the formulary is geared towards healthcare professionals and not easy to understand for the lay public. Therefore, we collaborated with the Royal Dutch Pharmacy Association to add specific paediatric information leaflets to their website http://www.apotheek.nl.
Summary
We here present the development and successful implementation of a national, best-evidence formulary for children. The Dutch Paediatric Formulary is a proof of concept in creating a knowledge-based paediatric formulary. Knowledge-based means that we know the scientific backgrounds that support paediatric use and it guarantees that a risk-benefit assessment for paediatric use has been performed. We emphasise again that a best-evidence formulary does not necessarily imply high-level evidence for every dosing recommendation. The development has resulted in the revision of many consensus-based dose recommendations, clarified the scientific grounds of drug use for children and ensured uniformity in prescribing habits in the Netherlands. The formulary has also been of great value in timely translating scientific research knowledge to daily practice. We believe that the Dutch approach in creating a knowledge-based paediatric formulary was successful and may serve as an example for other countries that struggle with paediatric drug dosing information.
Acknowledgments
We thank all members of the Editorial Board of The Dutch Paediatric Pharmacotherapy expertise Network who contributed to the development of the Dutch Paediatric Formulary (http://www.kinderformularium.nl). J. N. van den Anker MD PhD, professor of clinical pharmacology, neonatologist, clinical pharmacologist, UKBB Basel, Switzerland, Children’s National Medical Center, Washington DC, USA and Erasmus MC-Sophia Rotterdam. D. Bastiaans PharmD PhD, hospital pharmacist, Radboud UMC, Nijmegen. M. C. G. Beeren MD, paediatrician, St Annaziekenhuis, Geldrop. L. A. tenBerg-Lammers PharmD, hospital pharmacist, AMC/EKZ Amsterdam. M. S. Bolhuis PharmD PhD, hospital pharmacist, UMCG/Beatrix Kinderziekenhuis, Groningen. G. E. van den Bosch MD PhD, physician, Erasmus MC-Sophia, Rotterdam. N. Bouman MD PhD, paediatric psychiatrist, Amarant groep, Tilburg. J. Bouquet MD PhD, paediatric gastroenterologist NP (this person has died). P. L. M. De Bruyne MD PhD, general paediatrician, clinical pharmacologist, UZ Gent, Belgie. M. Dalinghaus MD PhD, paediatric cardiologist, Erasmus MC-Sophia, Rotterdam. P. H. Dijk MD PhD, neonatologist, UMCG Beatrix Kinderziekenhuis, Groningen. G. J. Driessen MD PhD, paediatrician infectious diseases, Erasmus MC-Sophia, Rotterdam. R. A. Doedens MD PhD, paediatrician infectious diseases, Martiniziekenhuis, Groningen. E. M. Dorresteijn MD PhD, paediatric nephrologist, Erasmus MC Sophia, Rotterdam. F. K. Engels PharmD PhD, hospital pharmacist, Erasmus MC-Sophia, Rotterdamdrs. M. Ezinga PharmD, hospital pharmacist, AHZ, Den Haag. R. B. Flint PharmD, hospital pharmacist, Radboud UMC Nijmegen. H. Folmer MD, general practitioner, N. P. M. Groeneweg MD PhD, paediatric gastroenterologist, Maasstadziekenhuis, Rotterdam. T. R. de Haan MD PhD, neonatologist, AMC/EKZ Amsterdam. L. M. Hanff PharmD PhD, hospital pharmacist Erasmus MC-Sophia, Rotterdam. I. G. A. Holsappel PharmD, pharmacist, KNMP, Den Haag. M. de Hoop-Sommen PharmD, pharmacist KNMP, Den Haag. A. Horikx PharmD, pharmacist, KNMP, Den Haag. K. Huizing MD, paediatric intensivist, MUMC, Maastricht. F. E. Jansen MD PhD, paediatric neurologist, UMCU/WKZ, Utrecht. H. M. Janssens MD PhD, paediatric pulmonologist, Erasmus MC-Sophia Rotterdam. E. M. Kemper PharmD PhD, hospital pharmacist, clinical pharmacologist AMC/EKZ Amsterdam. K. Langenberg MD, paediatrician, clinical pharmacologist i.o., Ziekenhuis Gelderse Vallei, Ede. M. J. van Ledden MD, paediatrician, Ziekenhuis Dirksland, Dirksland. T. B. Y. Liem PharmD PhD, hospital pharmacist, UMCU/WKZ, Utrecht. M. Lilien MD PhD, paediatric nephrologist, UMCU/WKZ, Utrecht. D. M. W. M. te Loo MD PhD, paediatric haematologist, clinical pharmacologist Radboud UMC, Nijmegen. M. N. Lub-de Hooge PharmD PhD, hospital pharmacist, UMCG/Beatrix Kinderziekenhuis, Groningen. K. P. Mohan-Gayadien PharmD, pharmacist, Vlietland Ziekenhuis, Schiedam. B. Molenbuur, MD, paediatric anaesthesiologist, UMCG Beatrix Kinderziekenhuis Groningen. S. Natsch PharmD PhD, hospital pharmacist, clinical pharmacologist Radboud UMC, Nijmegen. A. Oranje MD PhD, paediatric dermatologist Dermicis Skin Clinic Alkmaar, Hair clinic Breda and Kinderhuid.nl Teledermatology Rotterdam, the Netherlands. L. van Onzenoort MD PhD, paediatrician, clinical pharmacologist i.o. Radboud UMC, Nijmegen. S. G. M. A. Pasmans MD PhD, paediatric dermatologist/immunologist, Erasmus MC-Sophia Rotterdam. R. Rodrigues Pereira MD, paediatrician, mc Kinderplein, Rotterdam. C. M. A. Rademaker PharmD PhD, hospital pharmacist, UMCU/WKZ, Utrecht. P. C. van Rijn-Bikker PharmD PhD, hospital pharmacist, AMC/EKZ Amsterdam. A. van Royen-Kerkhof MD PHD, paediatric immunologist, UMCU/WKZ, Utrecht. S. D. T. Sassen PharmD, pharmacist NKFK 2010–2012. B. A. Semmekrot MD PhD, neonatologist, Canisius-Wilhelmina Ziekenhuis, Nijmegen. N. E. Schalij-Delfos MD PhD, paediatric ophthalmologist, LUMC, Leiden. A. M. J. W. Scheepers-Hoek PharmD PhD, hospital pharmacist, MUMC, Maastricht. M. Schreuder MD PhD, paediatric nephrologist, Radboud UMC/Amalia Children’s Hospital, Nijmegen. M. Shams PharmD, pharmacist, NKFK 2007–2010. I. H. van der Sijs PharmD PhD, hospital pharmacist, Erasmus MC-Sophia, Rotterdam. S. Simons MD PhD, neonatologist, Erasmus MC-Sophia Rotterdam. M. Sonnenberg PharmD, pharmacist KNMP, Den Haag. R. N. Sukhai MD PHD, kinderarts-nefroloog, LUMC, Leiden. A. I. Veldkamp PharmD PhD, hospital pharmacist VUMC, Amsterdam. M. Verduijn PharmD, pharmacist, NHG, Utrecht. F. G. A. Versteegh MD PhD, paediatrician Groene Hart Ziekenhuis, Gouda, the Netherlands and Ghent University Hospital, Department of Paediatrics, Ghent, Belgium. M. M. van Weissenbruch MD PhD, neonatologist, VUMC, Amsterdam. M. N. Weitering MD, physician, Erasmus MC-Sophia, Rotterdam. drs. B. Wensveen PharmD, pharmacist, KNMP, Den Haag. M. Williams MD PhD, paediatrician metabolic diseases, Erasmus MC-Sophia Rotterdam. B. C. M. Witjes PharmD, hospital pharmacist, VUMC, Amsterdam. J. Zwaveling PharmD PhD, hospital pharmacist, LUMC, Leiden.
Footnotes
Collaborators On behalf of the Editorial board the Dutch Paediatric Pharmacotherapy Expertise Network NKFK (Nederlands Kenniscentrum voor Farmacotherapie bij Kinderen).
Contributors TMvdZ, has coordinated the development of the Dutch Paediatric Formulary, drafted the initial manuscript and approved the final manuscript as submitted. SNdW has been involved as director of the NKFK since 2012, reviewed and revised the manuscript and approved the final manuscript as submitted. YL is a hospital pharmacist at UMC Utrecht-Wilhelmina Children’s Hospital. The comprehensive paediatric formulary of the Wilhelmina Children's Hospital (WKZ) was adopted—with permission—as the initial reference for dosing recommendations. He has reviewed and revised the manuscript, and approved the final manuscript as submitted. MO and MdH are the founders of the Dutch Paediatric Pharmacotherapy Expertise Network NKFK and initiated the development of the Dutch Paediatric Formulary. They reviewed and revised the manuscript and approved the final manuscript as submitted.
Funding Dutch Ministry of Health (2005–2010, 2012–2013, 2016). Fund for Innovative actions of the branch organisation for insurance companies (Innovatiefonds Zorgverzekeraars Nederland) (2011–2012). Fund for qualitative improvement of the federation of medical specialists (Stichting Kwaliteitsgelden Medisch Specialisten (SKMS) (2011–2012). Branch organisation for insurance companies Zorgverzekeraars Nederland (ZN) (2012–2016). Dutch Paediatric Association (NVK) (2012–2016). Royal Dutch Pharmacy Association (KNMP) (2012–2016). Dutch Association of Hospital Pharmacists (NVZA) (2012–2016). The Netherlands organisation for Health Research and Development (ZonMw): grant for development of the dosing calculator, grant for development of patient information on paediatric use. Dutch Kidney Foundation (Nierstichting Nederland): grant for renal project.
Competing interests TMvdZ is manager of the Dutch Paediatric Pharmacotherapy Expertise Network; SNdW is director of Dutch Paediatric Pharmacotherapy Expertise Network.
Provenance and peer review Commissioned; externally peer reviewed.