Article Text
PDF
Abstract
Background A considerable proportion of drugs administered to children are not authorised for this purpose, and consequently off-label use is common in paediatric care. Our aims were to quantify systematically the number of drugs authorised in Switzerland for use in children based on their current summary of product characteristics (SmPC) and to assess the quality of this information.
Methods We used natural language processing to screen all Swiss SmPCs, available in German language in the open-source drug database, for information about use in children. Based on the SmPCs of the most frequently used drugs in Swiss children’s hospitals, 10 search terms were defined to retrieve this information.
Results Of the analysed 4214 drugs corresponding to 1553 active substances, 2322 (55.1%) drugs were authorised for use in children. In only 639 (15.2%) SmPCs, information about authorisation for children was found in the section ‘Therapeutic indications’. 320 (13.8%) SmPCs of drugs authorised for use in children contained only verbal age indications such as ‘children’ and ‘adolescents’ without a clear definition of the age or an age range.
Conclusions Most Swiss SmPCs contain information about children, but only a minority refer to an official indication. Even if some SmPCs clearly indicate that use in children is authorised, a clear statement of the age at which the drug may be administered is missing. Standardisation of information about use in children in SmPCs is needed.
- Paediatrics
- Therapeutics
- Pharmacology
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Off-label use is common in paediatric care, as many drugs used in children were not studied in this population and consequently summaries of product characteristics (SmPCs) often lack information about use in children.
Development and availability of drugs for children started to increase since the European Paediatric Regulation (1901/2006).
Off-patent drugs, which have only been licensed for adults, are practically not affected by the European Paediatric Regulation.
WHAT THIS STUDY ADDS
Almost half of the drugs available on the Swiss market are not authorised for use in children.
The minority of SmPCs contain a clear indication or contraindication for use in children.
Evidence-based paediatric data were found in only one-third of the Swiss SmPCs.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
In absence of information in SmPCs about children, professionals prescribing drugs for children need to consult other sources providing evidence- and expert-based off-label dosing recommendations.
Introduction
Clinical studies are essential to prove safety and efficacy of drugs and to minimise the risk of treatment failure or adverse drug reactions (ADRs).1 2 For drug use in children, the European Paediatric Regulation (1901/2006) enforces pharmaceutical companies to demonstrate the quality, safety and efficacy of medicines by submitting a paediatric investigational plan (PIP). This applies to new drug products, indications, formulations or routes of administration.3 A PIP is now a prerequisite to obtain marketing authorisation and should lead to drug labelling with evidence-based dosing recommendations for children.3
However, analyses on the impact of the new regulation on the paediatric drug market revealed that although more medicines authorised for children have entered the market, there is frequently a lack of age-appropriate dosage forms and formulations.4 5 While paediatric research and drug development have increased, many PIPs have been waived, deferred, or delayed and dose-finding studies and drug approvals are often limited to adults.6–10
Accordingly, summaries of product characteristics (SmPCs) often do not contain approved dosing recommendations for children. As these drug products without information about use in children are often the only treatment available for certain conditions or age groups, off-label use remains common in paediatric care.11–13 The reported off-label prescription rates in paediatric care vary from 3.2% to 95%, depending on the study setting,13 with younger age generally being associated with higher off-label prescription rates.14–16 A prospective study in a Swiss children’s hospital revealed that about half of the prescriptions are not covered by the terms of the drugs’ marketing authorisation and are therefore not in accordance with the SmPC.17 A recent analysis of the impact of the paediatric regulation in Switzerland concluded that in addition to regulatory efforts to reduce off-label use in paediatrics, authorities should place more emphasis on improving the safety in off-label prescribing.9 In the absence of approved dosing recommendations for children, initiatives aim to establish standardised off-label treatments using evidence from scientific literature and expert knowledge from clinical practice.18
According to the Swiss Agency for Therapeutic Products (Swissmedic), information on use in children is based on (1) data submitted by the pharmaceutical companies and (2) on the guideline E11 of the International Council for Harmonisation (ICH).19 20 This guideline addresses the safe and ethical study of medicinal products in the paediatric population and provides suggestions for age classification of paediatric subgroups or for paediatric formulations. Our aim was to quantify systematically the number of drugs authorised by Swissmedic for use in children based on their current SmPC. Moreover, we aimed to assess and evaluate the quality of the information in Swiss SmPCs about the drug’s use in children.
Methods
We have previously identified the 40 most frequently used active substances in Swiss children’s hospitals.18 Based on the SmPCs of those, we selected and refined search terms to find information about use in children aged <18 years in all available Swiss SmPCs. We applied natural language processing (NLP) to screen the 4452 Swiss SmPCs for sentences including the defined search terms (figure 1). We performed a full-text search on 19 January 2022 using the open-source drug database (https://ch.oddb.org), which contains SmPCs that pharmaceutical companies are required to publish in a public domain. The outcome of the search was a comma-separated values (CSV) file with one hit sentence per row. In addition, the row contained the drug name, the active substance and the section of the SmPC in which the hit sentence appeared.
Flow chart depicting the individual steps of the analysis, the classification categories and the 10 search terms in German language (in the top right circle). English translation of search terms: Jugendliche/n—Adolescents, Kind—Child, Kinder/n—Children, Kleinkinder—Toddlers, Neugeborene—Newborns, Pädiatrie—Paediatrics, Säuglinge—Infants. ATC, Anatomical Therapeutic Chemical; SmPC, summary of product characteristics.
We removed duplicate sentences and hits in the sections ‘Composition’, ‘Pharmaceutical form and quantity of active substance per unit’, and ‘Effects on ability to drive’ containing inapplicable information. After further screening, we excluded (1) the sections ‘Pregnancy and lactation’, ‘Overdose’, ‘Preclinical data’, ‘Interactions’ and ‘Other information’ which were not relevant to our analysis and we excluded (2) SmPCs of drugs that were no longer on the market or whose marketing status was unknown (ie, the drug is approved in Switzerland, but not (yet) on the market).
The number of rows to be analysed in the CSV file was reduced by merging all hit sentences of the same section into one row with the open-source data analysis tool Pandas (V.1.3.4) using Python (V.3.8.12) (online supplemental table 1). Further analyses were performed in Microsoft Excel 2013.
Supplemental material
In Switzerland, each SmPC either belongs to one single drug product (eg, Comirnaty 10 μg/dose concentrate for dispersion or for injection) or describes multiple drug products with the same brand name (eg, Valium describes tablets 5 mg, 10 mg and sterile injection 10 mg/2 mL). Even if one SmPC described multiple drug products, the information was handled as a single drug.
For each of the identified drugs, we analysed the retrieved sentences per section for information about children and classified them stepwise into different categories. In doing so we followed the order in table 1.
Description of the classification categories applied to the information about children in SmPCs
If a drug was authorised for use in children, based on the presence of a paediatric indication, a paediatric dosing recommendation in the section ‘Posology and method of administration’ or sentences in other sections indicating the use of the drug in children, we recorded the respective age or age range and how this information is stated. One SmPC may contain different age information for which a drug is approved, depending on the pharmaceutical form or dose strength, but also on indication of the different drug products. If different age information was available, the youngest age stated was considered in our analysis. Unspecific age information was translated into the recommended age categories of the ICH.19
Furthermore, we searched for ADRs that may occur specifically in the paediatric population in the section ‘Undesirable effects’. In addition, we examined whether the drug is contraindicated in children, not recommended, or can only be administered to children on medical prescription and we extracted this information together with the concerned age category. We systematically screened the sections ‘Properties/effects’ and ‘Pharmacokinetics’ for clinical or pharmacokinetic data of children and extracted number and type of paediatric studies. Finally, we classified drugs according to the Anatomical Therapeutic Chemical (ATC) Code stated in the SmPC.21
Classification as described in table 1 was performed for all rows by one researcher. To test for reliability, a random selection of 10% of the rows was independently classified by two researchers. Disagreements were resolved in a consensus discussion. In the framework of this descriptive analysis, the proportion of drugs approved for use in children was calculated relating the number of drugs with an SmPC indicating that use in children is authorised to the total number of drugs analysed.
Results
The primary NLP search of 4452 SmPCs describing 16 767 drug products approved in Switzerland identified 51 008 hit sentences including the search terms. After deleting 21 147 duplicate sentences and 75 sentences of sections containing inapplicable information, 29 786 hit sentences in 4362 SmPCs corresponding to 1570 active substances remained. As shown in figure 2, hit sentences were most often retrieved from the sections ‘Posology and method of administration’, ‘Other information’ or ‘Warnings and precautions’. 3195 of the 4117 sentences in the section ‘Other information’ were stating ‘Keep out of reach of children’. After excluding this and other sections not relevant for further analysis and 103 SmPCs of approved drugs not available on the Swiss market, 16 945 sentences in 4214 SmPCs of drugs corresponding to 1553 different active substances remained for the systematic analysis and classification.
Sections in the SmPCs containing the hit sentences. The numbers within/beside the bars refer to the number/percentage (%) of hit sentences in the corresponding drug section. SmPC, summary of product characteristics.
Authorisation status for children and age information
2322 (55.1%) of the drugs analysed were authorised for use in children. Among those, 643 were authorised for adolescents, 980 for children, 385 for infants and toddlers, 234 for term newborns and 26 for preterm newborns (figure 3A). The age categories are interdependent, for example, if a drug was authorised for use in infants, it was also authorised for use in children and adolescents. Only in 639 (15.2%) SmPCs, the information about authorisation for children was found in the section ‘Therapeutic indications’. In the remaining, a concrete classification (authorised yes or no) was only possible in the context of information from the section ‘Posology and method of administration’ or from other sections.
Classification of the drugs into the official age categories19 (A) and into the ATC index21 on the first ATC level (B). The age categories are interdependent, for example, if a drug was authorised for use in infants, it was also authorised for use in children and adolescents. This does not apply to individual drug products which are principally intended for a defined age group within the paediatric population (eg, Synagis, for preterm newborns up to the age of <2 years). The numbers within or beside the stacked bars refer to the percentage (%) of drugs that are not authorised for use in children. ATC, Anatomical Therapeutic Chemical; n.a., not applicable.
We used the ATC Code to classify the drugs. The lowest proportion of drugs authorised for use in children was found in the anatomical groups C: Cardiovascular system and J: Antineoplastic and immunomodulating agents (figure 3B).
The age information was often specified in days, months or years, but there were also SmPCs providing unspecific age information using words, like ‘children’ or ‘adolescents’. 320 (13.8%) of the authorised SmPCs contained such verbal age information, but the age or an age range was not clearly stated. The verbal age information provided in those SmPCs was not strictly used according to the age classification by name and life span (eg, children; 2–11 years). Whether the term ‘children’ is meant to include or exclude infants and newborns could not be determined unambiguously (eg, it is recommended to treat children with 45 mg/m2 as long as no severe side effects occur).
Of 2322 SmPCs of drugs authorised for use in children, 2189 contained paediatric dosing recommendations in the section ‘Posology and method of administration’.
Information about ADRs and restrictions in children
In 799 (19.0%) of the analysed drugs, we found information about children in sentences of the section ‘Undesirable effects’. 685 of these 799 drugs were authorised for use in children. In 734 of these 799, ADRs in children were mentioned. In the remaining 65, either paediatric safety studies were cited or it was stated that no such studies were conducted.
The ADRs in children were often comparable or the same as in adults, while some SmPCs indicated differences in frequency and/or in type and severity of the ADR. Sometimes only assumptions were made; such as ‘it is assumed that the reported ADRs could also occur in children’.
The restrictions for use in children examined in all sections of the SmPCs were as follows; 1004 (23.8%) of the analysed drugs were either contraindicated and/or not recommended in a paediatric age category. Importantly, in 436 of the corresponding SmPCs, this information was provided in a section other than ‘Contraindications’. 230 (5.5%) SmPCs stated that the drugs may only be administered to children on medical prescription, up to a certain age limit or in a specific condition (eg, HIV-infected children).
Source of paediatric data
1480 (35.1%) SmPCs contained information about paediatric studies or data in the sections ‘Properties/effects’ or ‘Pharmacokinetics’. Of these SmPCs, 275 indicated that the data were comparable to those for adults. The percentage and type of paediatric studies and data reported in these SmPCs are listed in figure 4. Information about children in sentences of the section ‘Pharmacokinetics’ was found in 1248 (29.6%) of the analysed drugs, among those were 384 not authorised for use in children.
The percentage and type of paediatric studies and data reported in 1480 of the 4214 drugs analysed. PD, pharmacodynamics; PIP, paediatric investigational plan; PK, pharmacokinetic.
Most frequently used active substances in children’s hospitals
In Switzerland, 299 (7.1%) drugs containing the 40 most frequently used active substances in Swiss children’s hospitals were available on the market.18 By analysing their SmPCs, we learned that 281 of these drugs were authorised for use in children, but 129 of these were not authorised for use in children under 6 years of age. Among them were 18 drugs containing 11 of the 40 active substances, which were not authorised at all for children <18 years of age.
Discussion
Our analysis of current SmPCs aimed to quantify systematically the number of drugs authorised for use in children in Switzerland and to evaluate how the information about use in children was presented.
55% of the drugs available on the Swiss market were authorised for use in children. This rate is higher than in Europe, where 33% of the drugs have paediatric authorisation.22 In the USA, 51% of drugs are licensed for paediatric use.23 Nevertheless, our analysis revealed that only 15% of available drugs were authorised for children under 2 years of age. This low proportion was consistent with previous estimates on the high proportion of off-label use in this age category.15 24 However, it was rather unexpected that 43% of the drugs containing the 40 most frequently used active substances in Swiss children’s hospitals18 were not authorised for use in children under 6 years of age, including the paediatric population (aged 3–6 years) with the highest drug prescription rate.24
We identified information about children in most Swiss SmPCs, but only the minority referred to a therapeutic indication. Importantly, whether a drug was authorised for children aged <18 years could often only be determined by the presence of a paediatric dosing information in the section ‘Posology and method of administration’. The lack of paediatric indications was striking. A situation analysis of the PIP in Switzerland revealed that this may be the intention of pharmaceutical companies, as covering paediatric use in sections other than ‘Therapeutic indications’ allows them to avoid the processes associated with submitting extended use applications, including respective price negotiations.9 In line with findings by Wimmer et al, there seems to be no specific rule in which section information about children is to be listed.25 Even when SmPCs clearly indicated that a particular drug is authorised and can be used in children, they often omitted a clear statement of the age at which the drug may be administered or were providing arbitrary age categories. This reflects previous findings by Mühlbauer et al in Germany.24
An international survey of experts showed that a paediatric drug monograph should state contraindications due to ADRs without restricting to the age of the patient treated, but including the information on how frequently ADRs are expected to occur in the paediatric population.26 Our analysis of Swiss SmPCs revealed that contraindications due to ADRs are stated in most of them, but only one-third of the drugs authorised for children specified the frequency of ADRs in the paediatric population. A comparison of the Swiss and the US SmPC using the example of Zithromax suggests that ADRs occurring in the paediatric population are underreported in Swiss SmPCs.27 28
Information in Swiss SmPCs about children frequently concerned warnings and precautions. A quarter of all drugs are either contraindicated and/or not recommended in a paediatric age category. Our findings are in line with Wimmer et al who further concluded that the high rate of contraindications and warnings may result from a lack of data rather than from differences in pharmacokinetics.25
We acknowledge some limitations. First, in our NLP search, sentences and subtitles were extracted from the SmPCs if they contained the search terms. However, tables in the SmPCs were not searched. Second, our data could contain information from drugs that were authorised in Switzerland but are no longer on the market. Finally, SmPCs may describe more than one formulation (eg, mouthwash and spray) with the same brand name, so the information on the age at which the formulation can be used or the contraindications may differ. The analysis of these sentences could therefore not be performed unambiguously.
In conclusion, the quality of information in Swiss SmPCs was limited with regard to use in children. In the indication sections, information about authorisation for use in children was only found in 15% of the SmPCs. In Switzerland, it is not clearly regulated that in the absence of a paediatric indication, the presence of dosing information officially means that the drug is authorised for the age category concerned. However, in our analysis, we counted the SmPCs with paediatric dosing recommendations as drugs authorised for use in children. In 14% of the SmPCs with authorisation for children, no clear paediatric age category was provided. Overall, our analysis revealed that 45% of all available drugs in Switzerland were not authorised for use in children. Standardisation of information about use in children in SmPCs with an official paediatric indication and clear statement of the age is urgently needed in order to support professionals in providing the right dose for the right age of their paediatric patient.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
Acknowledgments
The authors would like to thank Zeno Davatz for his support in using the ODDB.
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 RT planned and performed the presented analysis, drafted the initial manuscript and approved the final manuscript as submitted. She is responsible for the overall content as the guarantor. HMzS supervises the thesis of RT in paediatric pharmacology. She initiated and planned the presented analysis and performed the reliability test. She reviewed and revised the manuscript and approved the final manuscript as submitted. SA planned the analysis and performed the reliability test. He reviewed and revised the manuscript. CB supervises the thesis of RT in paediatric pharmacology. He reviewed and revised the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.