A Survey of Medicine Use in Children and Adolescents in Austria

Home , ATC code J, ATC code N, ATC code R, Monobactam, Sultiame

A survey of medicine use in children and adolescents in Austria

Doctoral thesis at the Medical University of Vienna for obtaining the academic degree

Doctor of Medical Science

Submitted by

Mag. pharm. Elisabeth Rauch

Supervisor:

Univ. Prof. Dr. Christoph Male MSc

Department of Paediatrics and Adolescent Medicine Waehringer Guertel 18-20 A-1090 Vienna

Vienna, date of submission: 10/2016

DECLARATION

This thesis project was realised at the Department of Paediatrics and Adolescent Medicine, Medical University of Vienna. Study concept was prepared by Univ. Prof. Dr. Christoph Male MSc and Elisabeth Rauch. Data collection, analysis and interpretation as well as preparation of the original manuscript was done by Elisabeth Rauch under the supervision of Univ. Prof. Dr. Christoph Male MSc (Department of Paediatrics and Adolescent Medicine, Medical University of Vienna) and the co-supervisors Univ. Prof. Dr. Harald Herkner (Department of Emergency Medicine, Medical University of Vienna) and PD Dr. Florian Lagler (Paracelsus Medical University and Clinical Research Center Salzburg).

Data amalgamating and statistical analysis was done with support of Ao. Univ. Prof. Mag. Dr. Walter Gall and his students at the Center for Medical Statistics, Informatics and Intelligent Systems (CeMSIIS), Section for Medical Information Management of the Medical University of Vienna. Further statistical support was given by Univ. Prof. Dr. Harald Herkner.

The project was financially supported by the Austrian Paediatric Society.

TABLE OF CONTENTS

LIST OF FIGURES ...... vi

LIST OF TABLES ...... vii

ABSTRACT ...... viii

ZUSAMMENFASSUNG ...... ix

ABBREVIATIONS ...... xi

ACKNOWLEDGEMENTS ...... xii

1. GENERAL INTRODUCTION ...... 13 1.1. Children as therapeutic orphans ...... 13 1.2. Legal requirements and their impact on paediatric practice ...... 15 1.2.1. The EU Paediatric Regulation ...... 15 1.2.2. Survey of all paediatric uses of medicinal products in Europe ...... 16 1.3. Evidence of medicine use in children ...... 16 1.4. Status quo of paediatric medicine use in Austria ...... 18 1.5. Rationale ...... 18

2. OBJECTIVES OF THIS THESIS ...... 20 2.1. Primary objective ...... 20 2.2. Secondary objectives ...... 20

3. METHODS ...... 21 3.1. ATC Coding...... 21 3.2. Data sources ...... 22 3.3. Paediatric medicine use in primary care setting ...... 22 3.3.1. Data collection ...... 22 3.3.2. Data preparation and coding ...... 24 3.3.3. Challenges and limitations ...... 24 3.3.4. Data analysis ...... 24 3.4. Paediatric medicine use in hospital care setting ...... 25 3.4.1. Data collection ...... 25 3.4.2. Data preparation and coding ...... 26 3.4.3. Challenges and limitations ...... 28 3.4.4. Data analysis ...... 28 3.5. Comparison of paediatric medicine use in primary versus hospital care setting ...... 29 3.5.1. Data preparation and coding ...... 29 3.5.2. Data analysis ...... 30 3.6. Evaluation of off-label use ...... 30 3.6.1. Definition of off label use ...... 31 3.6.2. Evaluation of off-label use (according to pre-defined criteria) ...... 32 3.7. Qualitative assessment of use ...... 32 3.7.1. Interviews with selected primary care paediatricians in Vienna ...... 32

iii

3.7.2. Analysis of qualitative prescription data regarding approval status ...... 36 3.7.3. Analysis of qualitative prescription data in comparison to quantitative primary care data ...... 37 3.7.4. Variability of prescription habits ...... 37

4. RESULTS ...... 38 4.1. Prescription frequency of medicines for children and adolescents in primary care setting ...... 38 4.1.1. Study population ...... 38 4.1.2. Prescription frequency on ATC level 1 (anatomical class) ...... 40 4.1.3. Prescription frequency on ATC level 2 (therapeutic class) ...... 45 4.1.4. Prescription frequency on ATC level 5 (chemical substance) ...... 49 4.1.5. Prescription frequency of anti-bacterials for systemic use in primary care ...... 54 4.2. Prescription frequency of medicines for children and adolescents in hospital care setting ...... 58 4.2.1. Study population / data coverage ...... 58 4.2.2. Prescription frequency on ATC level 1 (anatomical class) ...... 59 4.2.3. Prescription frequency on ATC level 2 (therapeutic class) ...... 63 4.2.4. Prescription frequency on ATC level 5 (chemical substance) ...... 67 4.2.5. Prescription frequency of anti-bacterials for systemic use in hospital care ...... 71 4.3. Prescription frequency of medicines in primary care versus hospital care setting ...... 74 4.3.1. Comparison on ATC level 1 (anatomical class) ...... 74 4.3.2. Comparison on ATC level 2 (therapeutic class) ...... 76 4.3.3. Comparison on ATC level 5 (chemical substance) ...... 79 4.3.4. Comparison of prescription frequency of anti-bacterials for systemic use...... 80 4.4. Evaluation of licence status of medicines used in children and adolescents ...... 81 4.4.1. Off-label use in primary care setting ...... 81 4.4.2. Off-label use in hospital care setting ...... 82 4.5. Qualitative assessment of medicine use in children and adolescents ...... 85 4.5.1. Qualitative assessment of prescription habits by interview ...... 85 4.5.2. Evaluation of off-label use (based on interviews) ...... 88 4.5.3. Comparison of qualitative data based on interviews to quantitative data obtained from reimbursement institutions ...... 89 4.5.4. Variability of prescription habits ...... 90 4.6. List of “Essential Medicines for Children and Adolescents in Austria” ...... 92

5. DISCUSSION ...... 93 5.1. Patterns of medicine use ...... 93 5.1.1. Primary care setting ...... 93 5.1.2. Hospital care setting ...... 96 5.1.3. Comparison of medicine use between primary care and hospital settings ...... 97 5.2. Qualitative information on prescription patterns based on interviews ...... 99 5.3. Available evidence for medicine use in children and adolescents ...... 100 5.3.1. Definition of off-label use ...... 100 5.3.2. Off-label use in Austria ...... 101 5.3.3. Use of pain medication for children and adolescents...... 103 5.4. Strengths and limitations of the survey ...... 104 5.5. Essential medicines for Children and Adolescents in Austria ...... 106

6. CONCLUSIONS ...... 107

7. REFERENCES ...... 108

8. APPENDICES ...... 111 8.1. Eighty percent most frequently prescribed chemical substances (ATC level 5) in primary care, stratified by age group (n=66) ...... 111 8.2. Eighty percent most frequently prescribed chemical substances (ATC level 5) in hospital care setting (n=114) ...... 113 8.3. Eighty percent most frequently prescribed chemical substances (ATC level 5) used in primary and hospital care each (n=150) ...... 116 8.4. Essential list of medicines including qualitative information from interviews with selected paediatricians (n=196) ...... 121 8.5. Licence status of eighty percent of most frequently prescribed medicines in primary care ...... 127 8.6. Licence status of eighty percent of most frequently prescribed medicines in hospital care ...... 130

CURRICULUM VITAE...... 133

LIST OF FIGURES

Figure 1 Description of ATC classification for clarithromycin ...... 22 Figure 2 Abstract of SPC Inotyol® ...... 32 Figure 3 Questionnaire for interviews with primary care physicians (example) ...... 34 Figure 4 Prescription frequencies for age cohort 0-19 years ...... 41 Figure 5 Prescription frequencies per ATC code level 1 stratified by age cohorts ...... 43 Figure 6 Prescription frequencies by anatomical class (ATC code level 1) ...... 44 Figure 7 Overall top 10 prescription frequencies per ATC code level 2 ...... 46 Figure 8 Top 10 most frequently prescribed therapeutic class groups (ATC code level 2) overall, frequencies per age cohorts ...... 47 Figure 9 Top 10 most frequently prescribed therapeutic class groups (ATC code level 2) overall, frequencies per geographical region ...... 49 Figure 10 Overall prescription frequencies of anti-bacterials for systemic use ...... 55 Figure 11 Prescription frequencies of antibacterials for systemic use, stratified by age cohorts ...... 56 Figure 12 Prescription frequencies of anti-bacterials for systemic use, stratified by geographical region ...... 57 Figure 13 Overall prescription frequencies to children in Austrian hospital ...... 60 Figure 14 Prescription frequencies stratified by care type ...... 62 Figure 15 Top 10 overall prescription frequencies per ATC code level 2 ...... 64 Figure 16 Top 10 prescription frequencies per ATC code level 2, stratified by hospital care type ...... 66 Figure 17 Overall prescription frequencies of anti-bacterials for systemic use (J01) ...... 71 Figure 18 Prescription frequencies of antibacterials for systemic use (J01), ...... 72 Figure 19 Prescription frequencies in primary versus hospital care setting ...... 75 Figure 20 Top 10 therapeutic classes (ATC level 2) per care system ...... 78 Figure 21 Prescription frequencies of anti-bacterials for systemic use in primary vs hospital care ...... 80

LIST OF TABLES

Table 1 List of ATC codes (level 1) and their terms ...... 21 Table 2 Master analysis table of hospital prescription data ...... 27 Table 3 Indications per therapeutic field assessed during interviews ...... 35 Table 4 Prescription data for primary care setting in Austria overall ...... 38 Table 5 Prescription data for primary care setting (Vienna vs other federal states of Austria) ...... 39 Table 6 Prescription frequencies per age cohort (ATC level 1) ...... 43 Table 7 Top 10 most frequently prescribed therapeutic class groups (ATC level 2) in different age cohorts ...... 48 Table 8 Five most frequently prescribed substance groups per each anatomical class (ATC level 1) in Austria ...... 52 Table 9 Prescription frequencies of anti-bacterials for systemic use, stratified by age cohorts ...... 56 Table 10 Prescription data for hospital care setting ...... 58 Table 11 Prescription frequencies, overall and per hospital ...... 61 Table 12 Prescription frequencies / ATC level 2, overall and per hospital ...... 65 Table 13 Five most frequently prescribed chemical substances (ATC level 5) per ATC group in Austrian hospitals, overall ...... 69 Table 14 Mode of administration of systemic anti-bacterials in hospital ...... 73 Table 15 Top 10 therapeutic classes (ATC level 2) per care setting ...... 77 Table 16 Percent off-label use in primary care overall and for anatomical classes by age groups ...... 81 Table 17 Percent of off-label use in hospital care by anatomical class and age ...... 83 Table 18 Medicines most often mentioned with indications for their use ...... 86 Table 19 Diseases for which paediatricians refer to specialist or hospital rather than initiate medical treatment themselves ...... 87 Table 20 Use of drugs mentioned by child psychiatrists, including recommended indication and age of use and approval status ...... 88 Table 21 Frequency of off-label use of drugs according to interviews with general paediatricians ...... 89 Table 22 Variability in prescription habits according to interviews with paediatricians ...... 91

vii

ABSTRACT

Most medicines have not been evaluated for use in children and are widely used off-label in this patient population. Data supporting medicine use in children stem from various sources of heterogeneous quality. There exist no representative data on medicine use in children in Austria. The aim of this survey was to identify all medicines used in children and adolescents in Austria, to evaluate their extent of off-label use, and to define a list of around 200 medicines most frequently used in this patient population. Reimbursement data of medicines prescribed to children and adolescents were collected from Austrian health insurances and data of medicines dispensed to paediatric wards were collected from hospital pharmacies. By this, we derived prescription frequencies for both, paediatric primary and hospital care settings. We analysed these data by Anatomical Therapeutic Chemical (ATC) classification system, patients´ age (only primary care setting), region of prescription (Vienna versus rest of Austria) and type of hospital care setting (secondary versus tertiary care). As anti-bacterials for systemic use turned out to be most frequently prescribed in both, primary and secondary care setting, we analysed this class of medicines by its subgroups, patients´ age, type of hospital care setting and region of prescription. Furthermore, we evaluated the extent of off-label medicine use in children and adolescents and obtained qualitative information on prescription patterns by performing interviews with physicians in paediatric general primary care and psychiatric care. We could find out that in primary care setting anti-infectives (mainly antibiotics for systemic use), medicines for the respiratory system (medicines for obstructive airway diseases, nasal preparations, cough and cold preparations, anti-histamins for systemic use) and nervous system (analgesics, antiepileptics, psychostimulants) were most frequently prescribed. In hospitals, highest prescription frequencies occurred for anti-infectives (mainly systemic antibiotics), medicines for the nervous system (analgesics, anaesthetics) and alimentary tract (various groups, mainly antacids, anti-emetics). Off-label use is decreasing with age from 52% to 11% of all prescriptions in primary care setting, and from 54% to 18% in hospital setting. Primary care paediatricians prescribe medicines for a panel of about 20 to 30 diagnoses. Depending on the diagnosis, prescription habits considerably vary among the physicians. Most heterogeneous prescriptions occur for common infections like cough, non-bacterial otitis or rhinitis. The results of this survey provide important epidemiological information specific for Austria and also serves as basis for a future project to develop an on-line information platform for medicines used in children and adolescents.

viii

ZUSAMMENFASSUNG

Die meisten Arzneimittel (AM) wurden nicht für Kinder entwickelt und werden außerhalb der Zulassung (off-label) eingesetzt. Informationen zur AM-Therapie bei Kindern stammen aus verschiedenen Quellen von heterogener Qualität. Es gibt keine repräsentativen Daten zum AM Einsatz bei Kindern in Österreich. Ziel dieses Surveys war, alle bei Kindern in Österreich eingesetzten AM zu identifizieren, das Ausmaß des Off-label Gebrauchs dieser AM zu evaluieren und eine Liste von ca. 200 AM zu definieren, die am häufigsten verschrieben werden. Daten zur Rückvergütung von AM durch den Hauptverband der Österreichischen Sozialversicherungsträger, sowie Abgabedaten von AM von Krankenhausapotheken an die Kinderstationen dienten uns als Quellen zur Erfassung von Verschreibehäufigkeiten im niedergelassenen Bereich und im Krankenhaus. Wir analysierten die Daten anhand des anatomisch-therapeutisch-chemischen (ATC) Klassifikationssystems, nach Alter der Patienten (nur niedergelassener Bereich), geographischer Lage der Verschreibung (Wien versus restliches Österreich) und Versorgungsstatus des Krankenhauses (sekundärer versus tertiärer Bereich). Da sich Antibiotika zum systemischen Gebrauch als die am häufigsten verschriebene AM-Gruppe, sowohl im niedergelassenen, als auch im Krankenhausbereich, herausstellte, analysierten wir diese Gruppe nach ihren Subgruppen, Alter der Patienten, Versorgungstyp des Krankenhauses und geographische Region der Verschreibung. Weiters evaluierten wir das Ausmaß des Off-label Gebrauchs von AM für Kinder und Jugendliche und holten im Rahmen von Interviews qualitative Informationen zum Verschreibeverhalten von Kinderärzten und Kinderpsychiatern im niedergelassenen Bereich ein. Im niedergelassenen Bereich wurden Anti-Infektiva (vor allem Antibiotika zur systemischen Anwendung), AM für den Respirationstrakt (Mittel bei obstruktiven Atemwegserkrankungen, Rhinologika, Husten- und Erkältungspräparate, Antihistaminika zur systemischen Anwendung) und AM für das Nervensystem (Analgetika, Antiepileptika, Psychostimulantien) am häufigsten verschrieben. Im Krankenhaus waren die höchsten Verschreibungsraten bei Anti-Infektiva (vor allem Antibiotika zur systemischen Anwendung), AM für das Nervensystem (Analgetika, Anästhetika) und den Stoffwechsel (verschiedene Gruppen, vor allem Antazida, Anti-Emetika). Der Off-label Einsatz von AM sinkt mit steigendem Lebensalter der Patienten von 50% auf 11% aller Verschreibungen im niedergelassenen Bereich, und von 54% auf 18% im Spitalsbereich. Kinderärzte im niedergelassenen Bereich ohne Spezialisierung verschreiben AM für ca. 20-30 Diagnosen. Abhängig von der Diagnose kann unter den Kinderärzten ein sehr unterschiedliches Verschreibeverhalten beobachtet

ix werden. Besonders heterogen sind Verschreibungen für banal Infekte wie Husten, nicht- bakterielle Otitis oder Rhinitis. Der Survey liefert wichtige epidemiologische Daten spezifisch für Österreich. Er dient auch als Grundlage eines zukünftigen Projektes zur Entwicklung einer Informationsplattform für Kinder-AM in Österreich.

ABBREVIATIONS

ABDA Bundesvereinigung Deutscher Apothekenverbände e.V. (Federal Union of German Associations of Pharmacists) abs. absolute ADHS Attention-deficit hyperactivity syndrome ADR Adverse Drug Reaction ATC Anatomical Therapeutic Chemical Classification System BAP British Association of Psychopharmacology BASG Bundesamt für Sicherheit im Gesundheitswesen (Austrian Federal Office for Safety in Health Care) DEKO Datenaustausch – Erstattungskodex EMA European Medicines Agency ESAC European Surveillance of Antibiotic Consumption HVB Hauptverband der Österreichischen Sozialversicherungsträger (Main Association of Austrian Social Security Institutions) MRSA Methicillin resistant staphylococcus aureus N Number NSAID Non-steroidal anti-inflammatory drug ÖGARI Österreichische Gesellschaft für Anästhesiologie, Reanimation und Intensivmedizin rel. relative SPC Summary of Product Characteristics WHO World Health Organisation

ACKNOWLEDGEMENTS

First of all, I would like to thank my supervisor, Univ. Prof. Dr. Christoph Male MSc for his inspiration and overall guidance throughout this thesis project. Moreover, many thanks to my co-supervisors, PD Dr. Florian Lagler and Univ. Prof. Dr. Harald Herkner for their advisory support during planning of the survey and for revising the manuscript.

I would like to express my gratitude to Dr. Robert Sauermann from the “Hauptverband der Österreichischen Sozialversicherungsträger”, for providing prescription data of paediatric primary care setting, as well as to Mag. pharm. Dr. Silvia Hetz from the Austrian Association of Hospital Pharmacists and all my contact persons in the 13 hospital pharmacies that made hospital prescription data for children and adolescents available to us.

Furthermore, many thanks to DDr. Peter Voitl from the Austrian Paediatric Society (regional organisation: Vienna) for his advice and help to realise the interviews with Viennese primary care paediatricians, and of course to all paediatricians who took their time for the interviews.

I would like to thank Univ. Prof. Dr. Walter Gall and his students, Gergely Mentsik, Katharina Salzlechner and Sonja Walczak, for their IT support in regards to pooling prescription data from different data sources into one data-base and for statistical data analysis.

Finally, special thanks to my beloved husband and children for their great encouragement, thoughtfulness and support that enabled me to realise this thesis project.

xii

1. GENERAL INTRODUCTION

1.1. Children as therapeutic orphans

For decades, medicines have not been systematically evaluated for use in children who have thus been termed “therapeutic orphans” [1]. The majority of medicines required to treat sick children are currently used outside the licenced indications (off-label). There were and still are various reasons for this status – ethical considerations, methodological challenges of performing clinical trials in children, and also the limited financial potential of marketing drugs in children.

To treat children with medicines developed for adults implies that data on quality, efficacy and safety are from adults. Such implicit extrapolation is usually not justified because of physiologic, pathophysiologic and pharmacologic differences between adults and children of various ages that affect the pharmacokinetics and pharmacodynamics of medicines [2-4]. This applies to all ways of absorption as well as to altered body distribution, metabolisation and excretion of medicines. To give an example, intra-gastric pH values are higher during neonatal period; hence, doses of medicines like penicillin G or phenobarbital must be adapted due to different bioavailability. For some medicines there are age-dependent differences in efficacy caused by different receptor density. There is no linear relationship between age, body size or weight and the appropriate dosing of a drug. There may be adverse effects to medicines only occurring in children, for example if they are related to growth or development. Moreover, many diseases or disease manifestations are specific to children, for example clinical complications in neonates, congenital abnormalities, genetic diseases or malignancies in children. Finally, child-appropriate medicine formulations are required to allow safe and reliable administration for all age groups.

Off-label use of medicines in children carries risks such as ineffective dosing, overdosing, lack of efficacy, and unknown safety profiles in children. A classic example was the “grey baby syndrome”, resulting from chloramphenicol given to neonates in doses downscaled from adults. Because of immature metabolisation, these doses lead to accumulation and life- threatening toxic effects [5]. Child-appropriate formulations are usually lacking and the use of extemporaneous formulations, such as crushed tablets, pharmacy-prepared liquid formulations, carries risks of contamination and unprecise dosing.

Within the last 20 years, there has been upcoming awareness of the lacking evidence for medicine use in children with several studies analysing the situation, and resulting in various initiatives to solve the problem, both on regulatory and on academic level.

Horen et al. prospectively investigated the relationship of off-label drug use and occurrence of adverse drug reactions (ADR) in the outpatient setting [6]. The study was part of a collaboration initiative to study drug utilisation in children in different locations, to analyse existing diagnoses and therapeutic differences, and to detect prescription problems and propose improvements. Resident paediatricians in the region Haute Garonne in France prospectively provided prescription data from children and adolescents and description of potential adverse drug reactions within 7 days. They found a significant relationship between off-label use and the risk of ADR, particularly if drug was given despite contraindication or for another indication than defined in the summary of product characteristics (SPC).

Following a request by the European Commission, the European Medicines Agency (EMA) published a report on evidence of harm from medicines utilised in children outside their labelling recommendations as described in the SPC [7]. There is empirical evidence that adverse drug reactions in children are more frequent with drugs used off-label than with licenced drugs. Moreover, obviously there is underreporting of adverse drug reactions for medications used off-label or unlicensed.

However, according to a literature review done by Pandolfini et al. [8], off-label medicine use in children often goes along with evidence-based drug therapy recommendations, hence it is not generally inappropriate practice. Nevertheless, the authors suggest that evidence for use and drug licences should be harmonised, and that existing information on the optimal drug use should be made available to everybody who is working in the paediatric health care system.

To overcome problems of non-evidence-based drug use in children, legal requirements were put into place in the European Union almost 10 years ago as well as initiatives on national and/or academic level were started in some countries to provide up-to-date evidence-based information on paediatric drug use to the practising physician, pharmacist and other healthcare professionals. Developments of the latter however depend of financial resources and are therefore at different stages of progress throughout Europe.

1.2. Legal requirements and their impact on paediatric practice

In order to overcome the hurdles and deficits described above, changes of legislation were initiated in USA (1997) and in Europe (2006).

1.2.1. The EU Paediatric Regulation

The EU Regulation on Medicinal Products for Paediatric Use (EC Nr. 1901 & 1902/2006) came into force in 2007 [9]. It aims to ensure high quality, ethical research into medicines for children, to increase availability of authorized medicines for children, and to increase the availability of authorised medicines for children. These targets ought to be achieved without unnecessary studies in children and without delaying the authorization process for medication in adults.

The regulation implemented obligations and incentives for the pharmaceutical industry to develop and authorise new medicines also for children. As key obligation of the paediatric regulation, applicants must submit a paediatric investigational plan (PIP) for all medicines to be aimed for marketing authorisation in Europe. The PIP defines clinical studies to be performed with the medicinal product in children and its time schedule for submission. There may be a waiver granted in case the substance is of no use for children or a deferral if results in clinical studies with adults are a prerequisite for clinical investigations in children. As incentive, the Paediatric Regulation defines a six-month patent extension in case of compliance with an agreed PIP and inclusion of the study results into the product information.

Further measures of the regulation deal with already existing results of clinical studies and their submission to the regulatory authorities in order to update product information of medicines on the market (articles 45, 46). Additionally, in article 42 all member states were requested to submit data on existing use of medicinal products in the paediatric population until January 2009. The results were described in the ‘Survey of all paediatric uses of medicinal products in Europe’ [10] and are used as basis to update an ‘Inventory of Paediatric Needs’ (article 43) that had been established based on paediatric expert opinion [11]. The inventory lists the key deficits of data currently available for all medicines in the relevant therapeutic areas. It is used as guidance to set priorities when devising paediatric developmental plans by the Paediatric Committee.

1.2.2. Survey of all paediatric uses of medicinal products in Europe

Aim of the survey performed by the European Medicines Agency (EMA) was to reflect ´real world´ use of medication in children and adolescents in Europe. Authors emphasised the need to collect data on all existing uses of medicinal products in the paediatric population, i.e. not only medication authorised for the use in children, but also medication that – due to missing alternatives – needs to be given in an off-label or unlicensed way.

Unfortunately, data were submitted by only 20 EU member states and two non-EU member states (Norway, Iceland) representing less than 50% of the EU paediatric population. They were of very heterogeneous quality and quantity in regards to application details (e.g. various methods of quantification), so the results could not be compared between countries. Consequently, interpretation of data and conclusions drawn from the survey are limited.

1.3. Evidence of medicine use in children

In order to assess the evidence of medicine use in children and adolescents, information on the current status of child specific drug treatment is needed. This includes epidemiological studies on drug use in children and adolescents in different countries and clinical settings.

Sturkenboom et al. [12] performed a cohort study in three European countries (Italy, Netherlands, United Kingdom) using population based data on primary care prescriptions. Patient databases of the years 2000 to 2005 were retrospectively analysed for prevalence of drug prescriptions in regards to WHO Anatomical Therapeutic Chemical (ATC) classification system, children´s age, and mode of drug use (acute or recurrent prescription). By comparing prescription habits in the different countries they could identify suboptimal use or undesirable prescribing practices. In consequence, they emphasise the need for education of practitioners for drugs with high prescription rates but little evidence for use (e.g. cold and cough medication). High variability of licence status of drugs in the three countries was observed which the authors interpreted as the major reason for high inter-country differences of off-label use. Olsson et al. focussed on out-patient medicine use in children and adolescents in Sweden in 2007 [13]. They observed highest prescription rates for drugs of the respiratory system and anti-infectives. Off-label use was most common for dermatologicals and sex hormones, as well as for cardiovascular, psychiatric medicines and non-steroidal anti-inflammatory

16 substances. In consequence, the authors suggest further paediatric research especially in these therapeutic fields. Several surveys were performed to analyse paediatric drug use in hospitalised paediatric patients. Magalhaes et al. did a systematic literature review on publications between 1994 and 2013 to describe off-label and unlicensed drug use in the paediatric hospital setting [14]. In consequence, they state that off-label use is diversely defined throughout the literature and that therefore the possibility to compare off-label use status between the different studies and different countries is limited. In conclusion, they claim an international study with common evaluation criteria of off-label and unlicensed use. From their analyses they detect the highest off-label and/or unlicensed prescription rates in systemic hormonal preparations excluding sex hormones, in medications for the alimentary tract and metabolism and for the respiratory system. Off-label or unlicensed use decreases with age and is highest in intensive care units.

Irrespective of the definition of off-label or unlicensed use, we see that studies on the use of paediatric medication have been performed in different countries and care settings reflecting widely varying prescription patterns. Additionally, as result of the paediatric regulation, marketing authorisations and product informations have been updated and extended since 2007 for drug use in children and adolescents. On the one hand this is due to increased activity by regulatory authorities who have analysed study reports of existing clinical trials [15]. On the other hand the number of well-designed paediatric studies has increased within the last 10 years and new or up-dated marketing authorisations were therefore given for drug use in children [16].

Nevertheless, for many substances there is still limited authorised information available for children, and most use in children is based on scarce data from academic studies, expert opinion, or practical experience only. Data on dosing, efficacy, safety, and on age- appropriate formulations of medicines for children are of heterogeneous quality and come from various sources. Such information is not readily available and it is challenging for practitioners to keep an overview of the best evidence and up-to-date information.

To address these problems, a number of formularies and databases of medicines for children have been generated compiling such information for practical use, as for example the WHO Essential Medicines List for Children. Based on a resolution of 2007, the WHO has published the ‘Model List of Essential Medicines for children’, listing around 300 medicines in about 30 therapeutic areas, considered the most relevant medicines used worldwide in children and adolescents. Databases in some European countries provide relevant information on the use

17 of medicines in children, such as the ‚British National Formulary for Children‘ (www.bnf.org), and internet-based databases in the Netherlands (www.kinderformularium.nl) and Switzerland (www.kinderdosierungen.ch). Some of these databases however are open for use only for professionals in the respective country, hence are either not accessible from abroad or are only provided in country-specific language and with regional regulatory information for application, dosing, etc.

1.4. Status quo of paediatric medicine use in Austria

In 2009, Prandstetter et al. retrospectively analysed records from an Austrian neonatal intensive care unit over a duration of 3 months showing that off-label medicine use in this setting is similar to the results of studies from other countries [17]. In the course of a symposium in Leoben/Styria in 2012, a poll was performed among 27 Styrian primary care paediatricians to define the 20 medicines that they most frequently prescribe in their daily routine [18].

Apart from these rather small investigations, there is no information available on the current medicine use in children and adolescents in Austria. There exists no representative survey, neither for primary nor for hospital care setting. Presently, there is no systematic information on prescription frequencies for children and adolescents, neither in regards to their indications nor for other factors like patients´ age or geographical region of prescription. Moreover, no information exists on the extent of off-label use in the paediatric population in Austria.

1.5. Rationale

Currently, there exist no quantitative and qualitative data on medicines use in children and adolescents in Austria. Thus, there is a need for a representative survey investigating medicine use in both paediatric primary and hospital care setting. Such a survey will provide important epidemiologic data, reflecting the medicines available in Austria, the national licence status for paediatric use, and the specific prescription pattern of Austrian paediatricians and other medical practitioners. The data will be useful for regulatory purposes, health economic planning, and for evaluation of treatment strategies from a paediatric perspective.

Obtaining representative and up-to-date information on the current use of medicines in children and adolescents in Austria will also be the basis for any measures to improve the information base for medicine use.

Present treatment strategies can be evaluated in regards to licence status, medical literature and other paediatric medicine data-bases. The survey will allow to identify the 200 medicines most frequently used in children and adolescents and to develop an information platform providing actual evidence-based information. Presently, there exists no national medicine information system specifically providing evidence for use in children and adolescents. Existing medicine information systems, for example Austria Codex or MEDIS®, contain information on licensed indications only, which by far do not cover all therapeutic needs of children. International data bases are mostly not generally accessible, not in German language, and not appropriate for all users, particularly primary care paediatricians. Thus, there is an urgent need for an Austrian information platform specific for medicines used in children and adolescents that compiles evidence-based information from various sources, keeps it up-to-date, and readily accessible to practitioners.

2. OBJECTIVES OF THIS THESIS

2.1. Primary objective

The primary objective of this thesis was the assessment of medicine use in children and adolescents in Austria in 2014. Data on semi-quantitative frequencies of medicines prescribed in primary care and medicines dispensed in hospitals were collected and analysed by Anatomical Therapeutic Chemical (ATC) classification system, age, type of hospital care setting, and geographical region. Based on frequency analyses, lists of essential medicines used in children and adolescents were defined for primary care and hospital care setting in Austria.

2.2. Secondary objectives

Secondary objectives were to - compare prescription frequencies of medicines used in children and adolescents in the primary and hospital care setting; - evaluate the extent of off-label medicine use in children and adolescents in Austria per ATC classification system and age; - analyse the use of systemic anti-bacterials as this group of medicines was most often prescribed in both, primary and hospital care setting; - describe prescription patterns in paediatric general primary care and psychiatric care.

3. METHODS

3.1. ATC Coding

ATC coding is a widely used tool developed by the WHO Collaborating Centre for Drug Statistics Methodology (http://www.whocc.no) to exchange and compare data on medicine use. It is structured by 5 different levels. The first level, defined by a letter (described in table 1) identifies the anatomical main group of the substance.

Table 1 List of ATC codes (level 1) and their terms

Term ATC code (level 1) A Alimentary tract and metabolism B Blood and blood forming organs C Cardiovascular system D Dermatologicals G Genito urinary system and sex hormones H Systemic hormonal preparation, excl. sex hormones and insulins J Anti-infectives for systemic use L Antineoplastic and immunomodulating agents M Musculoskeletal system N Nervous system P Antiparasitic products, insecticides, and repellents R Respiratory system S Sensory organs V Various

Second levels, named by a letter (level 1) plus 2 digits (e.g. 01, 02, etc.) characterises the pharmacological/therapeutic subgroup. The 3rd and 4th levels are chemical/pharmacological/therapeutic subgroups and the 5th level is the chemical substance. An example for ATC coding is given for clarithromycin in Figure 1:

J ANTIINFECTIVES FOR SYSTEMIC USE J01 ANTIBACTERIALS FOR SYSTEMIC USE J01F MACROLIDES, LINCOSAMIDES AND STREPTOGRAMINS J01FA Macrolides

ATC code Name

J01FA09 clarithromycin

Figure 1 Description of ATC classification for clarithromycin (homepage of WHO Collaborating Centre for Drug Statistics Methodology, http://www.whocc.no/atc_ddd_index)

3.2. Data sources

In order to obtain prescription data of medicines for children and adolescents in Austria in 2014 we used two types of data sources:

- Primary care setting: the Main Association of Austrian Social Security Institutions (“Hauptverband der Österreichischen Sozialversicherungsträger”) for outpatient prescriptions. - Hospital setting: hospital pharmacies for data regarding dispensings of pharmaceuticals to paediatric wards.

3.3. Paediatric medicine use in primary care setting

3.3.1. Data collection

The ‘Main Association of Austrian Social Security Institutions (”Hauptverband der Österreichischen Sozialversicherungsträger”, HVB) kindly provided information on medicines that had been reimbursed for patients aged 0 – 19 in the year 2014. Information was listed at level 5 of Anatomical Therapeutic Chemical (ATC) classification system. Quantities of medicinal products without ATC code (e.g. dietary supplements, bandaging material, diverse homeopathics, etc.) were provided as ‘products without ATC code’ (“Heilmittel ohne ATC Code”) – without further specification of the kind of active substance. Extemporaneous preparations were listed as such (“Magistrale Zubereitungen”); quantities of these substances were known but active substances were not available in the database.

The data were provided as summary data of prescription frequencies (correcter: reimbursement frequencies) of individual medicines but no individual patient data was available. Summary data were provided for the entire cohort of patients aged 0-19 as well as for 6 different age strata (0-2, 3-4, 5-6, 7-9, 10-14, 15-19 years). For geographical region, data stratified by the 9 federal states of Austria. No information was available on the indication for prescription or patient demographics other than age.

The HVB covers medical insurance for about 95% of the Austrian population. It acts as umbrella organisation for 19 Austrian insurances, such as 9 district health insurance funds (“Gebietskrankenkassen”), the Insurance Institution for the Austrian Railways & Mining Industry (“Eisenbahner”), the Insurance Institution for Public Service Wage and Salary Earners (“Versicherungsanstalt der Öffentlich Bediensteten”), the Social Security Institution for Trade and Industry (“Sozialversicherung der Gewerblichen Wirtschaft”), 6 occupational health insurance funds (“Betriebskrankenkassen”) and the Social Security Institution for Farmers (“Sozialversicherungsanstalt der Bauern”).

Excluded from this coverage are people without social insurance and those who are insured within the so called “Krankenfürsorgeanstalten” (KFA) that are not associated within the Main Association of Austrian Social Security Institutions.

Data were available for patients aged 0-19 in primary care setting for - all prescriptions with a price above patient´s contribution fee (€ 5.40 in 2014, that means health insurance selling price of € 4.90). Medicines with a lower price are usually not claimed from the insurance, and are thus not contained in the insurance data base. - prescriptions to patients exempted from patient´s contribution fee.

The data include prescriptions from paediatricians as well as other physicians who also treat paediatric patients, e.g. general practitioners, ophthalmologists or gynecologists. However, no individual information on the prescribing physician nor regarding the proportional distribution of each specialist group was available.

3.3.2. Data preparation and coding

For our project, primary care prescription data were obtained from HVB as excel files indicating the number of prescriptions as per ATC code level 5, i.e. defining the chemical substance but not the name and/or application form of the drug. Lists were provided as “total count for Austria” and stratified by age groups and federal states. For some prescriptions, age group and/or federal state could not be identified. These cases were only taken into the overall analyses for Austria.

Medicines without ATC code (=”Heilmittel ohne ATC”) and extemporaneous preparations (“Magistrale Zubereitungen”) were listed as extra categories in the excel file by HVB. Whereas the former summarises the number of medicines not listed as medicines in the Austria Codex, i.e. medical devices, food supplements, bandaging material, etc., the latter gives information concerning the number of extemporaneous preparations but not concerning the (active) ingredients of these preparations. These preparations and medicines without ATC code were coded as “X” and “Y” on ATC level 1, as “X00” and “Y00” on ATC level 2, and as “X00XX00” and “Y00YY00” on ATC level 5, respectively. As information on active substances cannot be retrieved from these groups we delineated these numbers separately and did not include them into further analyses.

In order to obtain comparability with data presented in literature [12] and for simpler presentation we reduced age categories from six to three groups and geographical regions from nine to two groups.

3.3.3. Challenges and limitations

Reimbursement data from HVB are taken as proxy for prescription frequencies. We are aware that this does not include information on substances that are privately paid for by patients. Furthermore, the number of prescriptions by physicians or reimbursed is not equivalent to the number of medications actually taken by patients.

3.3.4. Data analysis

Frequency analyses were performed for the overall data set (i.e. for prescription frequencies in Austria overall) and stratified by three age groups (0-2, 3-9, 10-19 years) and geographical region (capital city (Vienna) versus 8 other federal states).

Analyses were performed for ATC levels 1 (anatomical main group), 2 (therapeutic main group), and 5 (chemical substance). On ATC level 2, the ten most frequently prescribed medicine groups are described. Within each ATC level 1 (anatomical main group) the five most frequently used ATC levels 5 (chemical substance) are described. Chemical substances (ATC level 5) among 80% of medicines most frequently used in the total study population (0-19 years) are assigned to the list of essential medicines.

Frequencies are described in percent in relation to overall prescriptions and as number of prescriptions per 1000 insured patients aged 0-19 years.

For ATC code J01 (anti-bacterials for systemic use), frequencies were analysed on ATC level 5. Based on data from literature [19], systemic antibiotics were divided into the following subgroups (ATC codes in brackets): tetracyclins (J01AA), broad spectrum penicillins (J01CA, J01CR), narrow spectrum penicillins (J01CE, J01CF), first (J01DB), second (J01DC) and third (J01DD) generation cephalosporins, other ß-lactam antibiotics (J01DE, J01DF, J01DH, J01DI), sulfonamides/trimethoprim (J01EA, J01EE), macrolides (J01FA), lincosamides (J01FF), and other systemic antibiotics (J01GB, J01MA, J01XA, J01XB, J01XC, J01XD, J01XE, J01XX). Similar to analysis of all other treatment groups, prescription frequencies were described per 1000 insured children, in total, and stratified by 3 age cohorts and by Vienna region versus rest of Austria, respectively.

SAS Enterprise guide was used for data preparation and grouping, data analysis and descriptive statistics were done in SPSS version 17.

3.4. Paediatric medicine use in hospital care setting

3.4.1. Data collection

For data collection, all hospital pharmacies registered within the Association of the Austrian Hospital Pharmacists (n=47) were contacted by email and asked to provide listings of medicines that had been dispensed to paediatric wards during the year 2014. In case of no reply, pharmacies of the biggest hospitals were again contacted via telephone.

The number of hospital beds for children and adolescents was taken as proxy for the size of paediatric wards per hospital and in consequence for the extent of coverage by our data (sources: www.spitalskompass.at and www.docs4you.at) as well as for calculation of prescription frequencies per hospital bed.

Data were collected on level of drug name (indicating pharmaceutical number of medicines and/or ATC code if available) and contained information concerning - quantity of packs dispensed from pharmacy to the ward - type of ward - proprietary medicinal product or extemporaneous preparation.

Dispensing lists were provided by hospital pharmacies as excel lists. Depending on internal hospital information technology programs, these listings had variable structure and contained heterogeneous information. To harmonise the data sets for the purpose of a combined analysis, data were entered into SAS Enterprise Guide and data sets were recoded and restructured to generate a joint master data base containing all relevant information from all pharmacies, as described in chapter 3.4.2.

3.4.2. Data preparation and coding

To obtain a uniform master list of medicines provided by all hospitals the following measures were taken:

- Data concerning enteral (baby) nutrition, diagnostics, care products (washing lotions, body lotions, lip sticks, …) were excluded from analysis

- ATC code D08 (Antiseptics and disinfectants) was filtered upfront in SAS Enterprise Guide as this group of therapeutics would mask further analyses for frequency of use of other medicines. This filter is reversible and this group may be included on demand.

- In case of missing ATC code in the original list, ATC code was either added automatically by matching the name of the preparation with the 2014´ index “Datenaustausch – Erstattungskodex” (DEKO) showing preparation name and ATC code. If ATC code was missing in the “DEKO”, a manual check-up was performed by searching alternative sources for ATC information, such as Austria Codex, BASG Arzneispezialitätenregister, ABDA database, “Rote Liste” and/or “Gelbe Liste”. If in

consequence ATC code could be assigned it was documented manually in the database. If ATC still could not be defined it was marked as “WATC”. - Extemporaneous preparations were identified as such indicating ATC code if available. - Negative numbers in dispensing listings indicate a back transfer into stock by the pharmacist; these amounts were excluded from analysis.

Table 2 describes the master analysis table that was used for further data analysis in SPSS version 17.

Table 2 Master analysis table of hospital prescription data

Column Description

ATC LVL 1 ATC code level 1 ATC LVL 2 ATC-Code level 2 ATC BEZ LVL 2 Name of ATC-Codes level 2 ATC LVL 5 ATC-Code level 5 ATC BEZ LVL 5 Name of ATC-code level 5 SUM LVL 5 Delivery to this ward (counted as no. of packs = no. of prescriptions) APPLIKATION*) Type of application form (parenteral, enteral, n.a.)*) INFO Information on extemporaneous preparation status IDST ID of the ward in the hospital IDKH_BEZ Name of the hospital IDB_BEZ Name of the federal state IDKH ID of the hospital IDB ID of the federal state *) only applicable for group “J01” (systemic antibiotics)

For frequency analyses of data, further up-front filters were put into place to focus on data that are relevant for medical analysis. Apart from excluding medicines categorised in ATC level D08 (antiseptics and disinfectants, see above), we excluded the following B05 subgroups (blood and blood forming organs):

 B05BA (solutions for parenteral nutrition)  B05BB (solutions affecting the electrolyte balance)  B05C (irrigating solutions)  B05D (peritoneal dialytics)

 B05X (i.v. solution additives)  B05Z (hemodialytics and hemofiltrates)

Quantities of extemporaneous preparations were described in units, such as “mg”, “g”, “ml”, “l”, etc. by some hospital pharmacies. These numbers were equated with “1” to show the qualitative use of these preparations but to put quantities into representative relation to all other preparations listed as “packs” or similar terms. Again, we are aware of this limitation of quantitative assessment of these data.

Medicines without ATC code (WATC) and extemporaneous preparations without further information on active ingredients were coded as “Z” on ATC level 1, as “Z00” on ATC level 2, and as “Z00ZZ00” on ATC level 5, respectively. As information on active ingredients cannot be retrieved from these groups we delineated these numbers quantitatively but did not enclose them into further analyses.

For sub-analysis of systemic antibiotics (ATC code “J01”), we differentiated enteral and parenteral forms of application. For some preparations, application form could not be identified (e.g. due to availability of both forms) and was therefore marked “not available”.

3.4.3. Challenges and limitations

Data on medicines dispensed to paediatric wards were used as proxy for prescription frequencies, as information on prescriptions to individual patients is not available. For relative quantification of the available information, we chose the following pragmatic definition: one package of dispensed medicine reflects one prescription. This definition obviously involves considerable uncertainty and is a limitation of the data available. Due to the lack of individual patient data, the data set does not contain information on patient characteristics (e.g. age) or indication for prescription.

3.4.4. Data analysis

Frequency analyses were performed for the overall data set on ATC levels 1 (anatomical main group), 2 (therapeutic sub group), and 5 (chemical substance). On ATC level 2, the 10 most frequently prescribed groups were described. Within each ATC level 1 the five most frequently used ATC levels 5 were described. Chemical substances (ATC level 5) among the

80% most frequently used were used to define a list of essential medicines for hospital setting.

Dispensing (i.e. prescription) frequencies were described as prescriptions per day and as prescriptions per hospital bed per day. Data were normalised to individual hospital size, i.e. frequencies of dispensings per hospital per ATC group in relation to overall dispensings per hospital, and average values and range between hospitals are described per ATC group.

For sub-group analyses, the data set was stratified by care setting (tertiary versus secondary care hospitals). Data are described as dispensing (i.e. prescription) frequencies per ATC group divided by overall dispensing per sub-group including the range of relative frequencies by hospital. Chi-2 test was used to compare frequencies between hospitals and hospital care setting. Random variability was described by 95% confidence intervals for proportions.

As antibacterials for systemic use (ATC code “J01”) were most commonly used, frequencies of this drug group were analysed at ATC level 5. Based on data from literature [19], systemic antibiotics were divided into the following subgroups (ATC codes in brackets): broad spectrum penicillins (J01CA, J01CR), narrow spectrum penicillins (J01CE, J01CF), first (J01DB), second (J01DC) and third (J01DD) generation cephalosporins, other ß-lactam antibiotics (J01DE, J01DF, J01DH, J01DI), sulfonamides/trimethoprim (J01EA, J01EE), macrolides (J01FA), lincosamides (J01FF), tetracyclins (J01AA), and other systemic antibiotics (J01GB, J01MA, J01XA, J01XB, J01XC, J01XD, J01XE, J01XX). Prescription frequencies were analysed per percent of total prescriptions “J01”, in total, stratified by care setting (secondary versus tertiary care) and by geographical region (Vienna region versus rest of Austria). Furthermore, frequency of enteral versus parenteral route of administration was compared and described.

3.5. Comparison of paediatric medicine use in primary versus hospital care setting

3.5.1. Data preparation and coding

The two data-sets were combined and up-front filters set as described in chapter 3.4.2., i.e. medicines listed in ATC group D08 (antiseptics and disinfectants) and certain sub-groups of

B05 (blood and blood forming organs) were not included into data analysis as these data would mask results of frequency analysis of pharmacologically more relevant medicine groups for our purpose.

For both data derived from hospital and from primary care data-base, medicines without ATC code and extemporaneous preparations without further information on active ingredients were coded as “Z” on ATC level 1, as “Z00” on ATC level 2, and as “Z00ZZ00” on ATC level 5, respectively. As qualitative information cannot be retrieved from these groups we delineated these numbers quantitatively but did not enclose them into further analysis.

To allow comparison between primary care and hospital settings, relative frequencies were calculated by percentage of prescriptions in relation to the overall prescriptions in the respective setting. Thereby, differences in distribution between medicine groups were compared.

3.5.2. Data analysis

Frequency analyses were performed for the overall data set on ATC levels 1 (anatomical main group) and 2 (therapeutic sub group). On ATC level 2, the 10 most frequently prescribed groups were identified and compared per care setting type. For ATC level 5, chemical substances representing 80% of prescriptions are listed for each setting and compared.

Prescription frequencies of anti-bacterials for systemic use (ATC code “J01”) were analysed at ATC level 5 according to subgroup analysis described before. Data were described as percent of total prescriptions in each care setting.

3.6. Evaluation of off-label use

In order to investigate evidence of medicine use identified in the data sets in primary care and hospital setting, respectively, the 80% most frequently used medicines (described as chemical substance, i.e. ATC level 5) were listed. The same was performed in stratified fashion for 5 different age cohorts (0-2, 3-4, 5-6, 7-9, 10-19 years) in primary care data-base. For hospital setting, as with analyses described previously, certain blood products (drugs with ATC codes B05BA, B05BB, B05C, B05D, B05X, B05Z9), and antiseptics and

30 disinfectants (D08) were excluded from analysis as these drug groups more likely have hygienic or nutritional than therapeutic impact and would mask results for therapeutic use of drugs. Medicines without ATC code and extemporaneous preparations (“X00XX00”, “Y00YY00”, “Z00ZZ00”) were numbered among the 80% but excluded from analysis of off-label use.

3.6.1. Definition of off label use

Age was used as the only criterion to define off-label use of medicines, for lack of information on indication for use, dosage, relevant co-prescribing or concomitant diseases in our data set. For this purpose, we compared age of use with the approved age in the summary of product characteristics (SPC) of medicines in the specific chemical substance group (ATC level 5). The Austria Codex was used as reference.

Medicine use as per our data set was considered off-label in age groups below the approved age noted in all SPC´s within the specific ATC group. If the lowest approved age in the SPC was within the age group indicated for use the medicine use was considered on-label. Medicines with statements like “There is not enough experience in use below xx years of age” or “Use is not recommended in patients below xx years” in the SPC were counted as off-label.

For lists of primary care setting, medicine use was coded for each substance group (ATC level 5) as follows: - medicine was used on-label in regards to age group, - medicine was used off-label in regards to age group, - medicine was not prescribed in this age group.

For hospital care setting, medicine use was coded for each substance group (ATC level 5) as follows, as patient’s age was not known: - medicine is approved for children aged 1 month and above, i.e. use is off-label for children aged 0-1 month, - medicine is approved for children aged 2 years and above, i.e. use is off-label for children aged 0-2 years, - medicine is approved for patients over 19 years of age, i.e. use is off-label for children aged 0-19 years.

The reason to define these categories was to differentiate off-label use in various age groups and to estimate amount of off-label use also in the neonatology setting.

Despite our relatively broad definition of off-label use, this still remained quite difficult to clearly define for all medicines. To illustrate this statement, an example is described in figure 2 (abstract / chapter 4, SPC of Inotyol® ointment): amongst the indications listed is the use for diaper dermatitis in babies. However, the chapter “dosing and mode of application” states “no clinical data for use in children and adolescents”.

4. KLINISCHE ANGABEN 4.1 Anwendungsgebiete – Ekzeme: akute und chronische Ekzeme – Wunden, Abschürfungen, Schrunden – Verbrennungen leichten Grades (z.B. Sonnenbrand) – Sonstige Hautreizungen: Rötungen, Frostbeulen, Risse, offene Hautblasen – Windeldermatitiden ohne Superinfektion 4.2 Dosierung und Art der Anwendung Ein- bis zweimal täglich Salbe zum Auftragen auf die Haut. Inotyol-Salbe messerrückendick auftragen und verbinden. Bei Hautschäden in den Falten der Leistengegend, des Gesäßes oder unter den Brüsten, die Salbe auf Gazestreifen aufstreichen und in die Hautfalten legen. Kinder und Jugendliche: Es liegen keine klinischen Daten zur Anwendung vor. Figure 2 Abstract of SPC Inotyol®

3.6.2. Evaluation of off-label use (according to pre-defined criteria)

Off-label use was evaluated overall and per therapeutic field. Percent off-label use was calculated by dividing the number of medicines prescribed off-label by the overall number of medicines prescribed in the respective age group.

3.7. Qualitative assessment of use

3.7.1. Interviews with selected primary care paediatricians in Vienna

For assessment of prescription patterns in paediatrics, we performed interviews with 10 primary care paediatricians and 4 child psychiatrists in Vienna. Thereby we could (i) obtain qualitative information on medical indications for prescriptions to children and adolescents in the paediatric primary care setting, (ii) compare qualitative information obtained in interviews

32 with quantitative data from HVB, and (iii) evaluate prescription patterns in the primary care setting.

Interviews with primary care paediatricians: Due to the small number of interviews, performed only in Vienna, and due to the fact that interviews were exclusively done on a voluntary basis, we do not claim representativeness of data obtained. However, they can be interpreted as “spotlights” and may - as a reflection of real life treatment of children and adolescents - be useful for the development of an evidence-based information platform on medicine use in children and adolescents.

To identify interviewees, we presented our project at the periodically held meeting of the Viennese section of the Austrian Paediatric Society (ÖGKJ) and via newsletter announcement. Of the almost 200 primary care paediatricians in Vienna, 10 agreed to be interviewed. They were visited in their offices or, alternatively, called by phone.

The interviews were structured in 8 different therapy fields: - Respiratory diseases - Ear-nose-throat diseases - Infections - Gastro-intestinal diseases - Dermatological diseases - Eye diseases - Uro-genital diseases - Unspecific symptoms/diseases, e.g. fever, pain, nausea

In each of these therapeutic areas, the most frequent diagnoses were listed and physicians were asked: “Which medicine do you normally prescribe in case of …?” Furthermore, paediatricians were asked to provide information whether specific medicines are prescribed for different age groups (0-2, 3-6, 7-9, 10-19 years) and whether the substance was prescribed as proprietary medicinal product or extemporaneous preparation from the pharmacy. In case the physician did not prescribe medication in a mentioned indication, the options “not answered” or “transfer to specialist or hospital” were provided. Finally, for each therapeutic area, the physician was asked if he/she regularly prescribes medication for indications not listed in the questionnaire.

An example of the questionnaire is given in figure 3.

2. HNO-Erkrankungen:

2. 1. Welche Arzneimittel (AM) verschreiben Sie bei Rhinitis acuta? 0-2 Jahre 3-6 Jahre 7-9 Jahre 10-19 Jahre Arzneimittel 1: als Fertigarzneimittel als magistrale Zubereitung Arzneimittel 2: als Fertigarzneimittel als magistrale Zubereitung Arzneimittel 3: als Fertigarzneimittel als magistrale Zubereitung Arzneimittel 4: als Fertigarzneimittel als magistrale Zubereitung Figure 3 Questionnaire for interviews with primary care physicians (example)

The structure of the questionnaire and selection of diagnoses were developed with the support of representatives of the Austrian Paediatric Society. Table 3 shows indications that were assessed during the interviews. Additionally, doctors were asked if he/she regularly treats children for other indications than those described in table 3.

Table 3 Indications per therapeutic field assessed during interviews

Therapeutic field Diagnosis Respiratory diseases Cough Obstructive bronchitis Acute asthma, Chronic asthma Laryngitis (Pseudo-Krupp) Eye-nose-throat diseases Rhinitis acuta Allergic Rhinitis Secretory otitis media (non-bacterial) Pharyngitis Infections Bacterial otitis media Bacterial throat infection Bacterial sinusitis Pneumonia Bacterial bronchitis Urinary tract infection Vulvitis / Balanitis Impetigo Erysipel Borreliosis Bacterial eye infections Bacterial maxillodental infections Mycotic infection / mouth-pharynx Mycotic infections / skin Herpes simplex infection Varicella Influenza Gastro-intestinal diseases Acute gastroenteritis Chronic obstipation Chronic diarrhoea Gastritis/Ulcus Dermatological diseases Atopic dermatitis Seborrhoic dermatitis Psoriasis

Acne vulgaris

Urticaria Hyperkeratosis Unspecific symptoms Fever Acute pain / Chronic pain Nausea / emesis Hypertension Eye diseases Allergic conjunctivitis Uro-genital Enuresis nocturna Dysmenorrhoea Contraception

Responses from the interviews were summarised in one database and analysed in total and per therapeutic field. Preparations mentioned by the physicians in the interviews were allocated to the corresponding ATC code, level 5 (chemical substance), retrieved from Austria Codex. Preparations without coding in Austria Codex, but in the “Warenverzeichnis” 3 or 4 (supplemental index for goods that are exclusively traded in Austrian pharmacies but do not have the status “pharmaceutical drug”, i.e. medicinal products, dietary supplements, etc.) were labelled with “WATC” and not taken into analysis of approval status. Extemporaneous preparations were coded as “MAG” and separately taken into analysis of approval status.

We analysed the medicines mentioned in the interviews by their frequency of use in specific indications. Thus, we could identify medicines that should be prioritised for research in paediatric use due to their broad use.

We also identified indications that were commented by the paediatricians with “transfer to a specialist or hospital”, i.e. disorders that might not be treated by primary care paediatricians, maybe due to lacking information on suitable treatments for children.

Interviews with child psychiatrists: For interviews with child psychiatrists, we filtered medicines with ATC code N (nervous system) from the prescription lists provided by HVB. Apart from analgesics and anti-epileptic medicines used for the treatment of schizophrenia (risperidone), attention-deficit hyperactivity syndrome (methylphenidate, atomoxetine) and depression (sertraline) were left. Analog to the questionnaires used for general paediatricians, we developed a questionnaire as basis for the interview. However, due to the small number of interviews on the one hand, and of medicines used in this field (primary care child psychiatrist) on the other hand, interviews were conducted very individually to obtain the best possible qualitative information. Again, we identified the medicines most widely used with qualitative information on their spectrum of indications in daily practise.

3.7.2. Analysis of qualitative prescription data regarding approval status

This analysis was done for data from interviews with general paediatricians only. Information on medicine prescription was coded as follows:

- medicine was used on-label in regards to indication and age group, - medicine was used off-label regarding either indication and/or age group, - medicine was not prescribed in this indication or age group or no information was provided by the physician

A medicine was considered off-label if indication for use (mentioned in the interview) was not noted in the summary of product characteristics (SPC) or if the lowest approved age noted in the SPC was above the age group indicated for use in the interview. If the lowest approved age in the SPC was within the age group indicated for use in the interview the medicine was considered on-label. The Austria Codex was considered as reference source for SPC.

Extemporaneous preparations were allocated off-label status. However, as these frequently are preparations of registered products (e.g. ointments including steroids, nasal drops, etc.), we analysed off-label status with and without extemporaneous preparations in relation to the number of prescribed medicines in total, per therapeutic field and per age group.

3.7.3. Analysis of qualitative prescription data in comparison to quantitative primary care data

Qualitative prescription data retrieved from interviews were compared to the 80% most frequently used chemical substances (ATC level 5) for the Vienna region from the primary care data-base provided by the HVB. Percentage of consistency was calculated by dividing matching medicine groups by number of 80% most frequently used chemical substances.

Furthermore, frequencies of extemporaneous preparations were compared between the two data-bases. As the latter did not show information on active ingredients this comparison could only be performed for total amount and not for individual therapeutic use of extemporaneous preparations.

3.7.4. Variability of prescription habits

For every diagnosis, medicines prescribed by the interviewees were analysed on ATC levels 2 (therapeutic class) and 5 (chemical substance). By assessing the number of different medicines used per indication, the variability of prescription habits among paediatricians could be analysed.

4. RESULTS

4.1. Prescription frequency of medicines for children and adolescents in primary care setting

4.1.1. Study population

Prescription data of the year 2014 were available for 1.665.952 children who qualified for benefit by social security. This corresponds to approximately 20% of total number of medically insured people in Austria in 2014.

In total 4.636.209 prescriptions were reimbursed for patients aged 0-19. The average number of prescriptions per child decreased with age, with a mean of 3.6, 2.9 and 2.5 prescriptions per child per year for the children aged 0-2, 3-9, and 10-19 years, respectively (table 4).

Table 4 Prescription data for primary care setting in Austria overall

Average N of N of insured N of Age group prescriptions 95%CI children prescriptions per child Austria (overall): 0-19 years 1.665.951 4.636.209 2,8 2,78 – 2,79 0-2 years 238.354 860.034 3,6 3,60 - 3,62 3-9 years 562.752 1.655.397 2,9 2,94 – 2,95 10-19 years 864.845 2.120.778 2,5 2,45 – 2,46

Table 5 lists the number of insured children and prescriptions in Vienna and the other Austrian federal states. The relative frequency of prescriptions in Vienna was significantly (p=0.017) higher compared to the rest of Austria. However, it is important to consider that in 2014 the number of exemptions from prescription charge per child was almost 2.5 fold higher in Vienna than in the other federal states (0.14 versus 0.06 exemptions per insured child).

Table 5 Prescription data for primary care setting (Vienna vs other federal states of Austria)

Average N of N of insured N of Age group prescriptions 95%CI children prescriptions per child Vienna: 0-19 years 331.664 1.209.218 3,6 3,64 – 3,66 0-2 years 53.654 273.244 5,1 5,05 – 5,13 3-9 years 118.073 444.456 3,8 3,75 – 3,78 10-19 years 159.936 491.518 3,1 3,06 – 3,09

Austrian federal states other than Vienna: 0-19 years 1.320.955 3.409.950 2,6 2,58 – 2,59 0-2 years 182.602 583.674 3,2 3,19 – 3,21 3-9 years 439.759 1.205.510 2,7 2,73 – 2,75 10-19 years 698.594 1.620.766 2,3 2,32 – 2,33

4.1.2. Prescription frequency on ATC level 1 (anatomical class)

Overall prescription frequency: Figure 4 shows overall prescription frequency on ATC level 1 (anatomical class). Upon itemisation by ATC level 1, the highest prescription frequencies were observed for anti- infective drugs with almost 700 prescriptions per 1000 children, followed by drugs for the respiratory system and the nervous system (400 and 220 prescriptions per 1000 children, respectively) (figure 4). These numbers correspond to 25%, 14% and 8%, respectively, of all drugs reimbursed in primary care in the respective ATC groups (J, R, N). Dermatologicals (D) and drugs for the alimentary system (A) ranged at 200 prescriptions per 1000 children each, medicines for the musculoskeletal system (M) at 100 per 1000 children. Drugs for blood and cardiovascular system (B, C), as well as genito-urinary system drugs, hormones, immunomodulating agents and drugs for the sensory organs (G, H, L, S) were prescribed with a frequency of less than 50 per 1000 children.

Extemporaneous preparations (“X”) and preparations without ATC code (“Y”) were prescribed at an overall frequency of 10% and 17% of total prescriptions, respectively.

Figure 4 Prescription frequencies for age cohort 0-19 years;

A= Alimentary tract and metabolism, B= Blood and blood forming organs, C= Cardiovascular system, D= Dermatologicals, G= Genito urinary system and sex hormones, H= Systemic hormonal preparation, excl. sex hormones and insulins, J= Anti-infectives for systemic use, L= Antineoplastic and immunomodulating agents, M= Musculoskeletal system, N= Nervous system, P= Antiparasitic products, insecticides, and repellents, R= Respiratory system, S= Sensory organs, V= Various, X= Extemporaneous preparations, Y= Preparations without ATC code.

Age stratified prescription frequency: Prescription frequencies per age cohort are illustrated in figure 5 and listed in table 6. For children aged from 0 - 2 years, the most frequent prescriptions were for anti-infective medicines (J), medicines for the respiratory (R) and the alimentary tract / metabolism system (A) with a frequency of 790, 340 and 300 prescriptions per 1000 children, respectively, followed by 220 and 130 prescriptions per 1000 children for dermatologic drugs (D) and drugs of the nervous system (N). Drugs from all other ATC groups were prescribed with a frequency of less than 100 per 1000 children in this age cohort. For children aged 3-9, anti-infective medicines were prescribed most often (830 prescriptions per 1000 children, 28% of prescriptions in this age cohort), followed by drugs for respiratory and nervous system with 460 and 150 prescriptions per 1000 children, respectively. Compared to children aged 0-2 and 10-19 years, the number of prescriptions in dermatologic indications in this cohort was lowest (100 prescription per 1000 children versus 220 and 250 prescriptions per 1000 children aged 0-2 and 10-19). Drugs from all other ATC groups were prescribed with a prevalence rate of less than 100 per 1000 children in this age cohort. For adolescents (10-19 years), as in the age cohorts described above, anti-infective drugs were most frequently used, followed by drugs for respiratory system and drugs for nervous system diseases (580, 360 and 300 prescriptions per 1000 patients). Dermatologic drugs and medication for alimentary tract and metabolism were prescribed more than 200 times per 1000 children; drugs of the musculoskeletal system were prescribed 140 times per 1000. Analysing the most frequently prescribed groups prescriptions in drug groups “blood” (B), “dermatologicals” (D), “musculoskeletal” (M), and “nervous system” (N) reached their peak levels in adolescents whereas “respiratory drugs” (R) and anti-infectives (J) were most often prescribed to 3-9 year old children. Drugs for the alimentary system (A) were most often prescribed to the youngest.

Figure 5 Prescription frequencies per ATC code level 1 stratified by age cohorts (0-2, 3-9, 10-19 years); key as per figure 4

Table 6 Prescription frequencies per age cohort (ATC level 1)

0-2 YEARS 3-9 YEARS 10-19 YEARS Prescriptions Percent of Prescriptions Percent of Prescriptions Percent of ATC Code Level 1 per 1000 total per 1000 total per 1000 total children prescriptions children prescriptions children prescriptions Total % Total % Total % A (Alimentary tract and metabolism) 295 8 135 5 203 8

B (Blood and blood forming organs) 23 1 18 1 64 3

C (Cardiovascular system) 3 0 9 0 30 1

D (Dermatologicals) 215 6 102 3 249 10

G (Genito urinary system and sex hormones) 15 0 17 1 31 1 H (Systemic hormonal preparation, excl. sex 52 1 39 1 36 1 hormones and insulins) J (Antiinfectives for systemic use) 787 22 831 28 579 24 L (Antineoplastic and immunomodulating 11 0 29 1 32 1 agents) M (Musculoskeletal system) 64 2 90 3 143 6

N (Nervous system) 127 4 149 5 300 12 P (Antiparasitic products, insecticides, and 4 0 8 0 10 0 repellents) R (Respiratory system) 338 9 461 16 362 15

S (Sensory organs) 30 1 46 2 46 2

V (Various) 1 0 9 0 13 1

X (Extemporaneous preparations) 658 18 237 8 202 8

Y (Preparations without ATC code) 988 27 763 26 154 6

OVERALL 3608 100 2942 100 2452 100 43

Prescription frequency stratified by geographic region: Figure 6 shows prescriptions frequencies of medicines in Vienna compared to all other Austrian federal states. Analysis in regards to geographic region of prescription was done for the overall study population and for the three age cohorts. Comparison of age groups revealed the same prescription pattern (data stratified by age cohorts not shown). Medicines of ATC groups A (alimentary), D (dermatological), M (musculoskeletal), N (nervous system), R (respiratory) and S (sensory organs) were more frequently used in Vienna compared to all other federal states (figure 6). However, this might result from the fact that the price for many preparations of these ATC groups is below the prescription charge and that in Vienna the number of people exempted from prescription charge is higher in Vienna than in the rest of Austria. For ATC groups J (anti-infectives), B (blood), C (cardiovascular), G (genito-urinary), H (hormones), L (anti-neoplastic and immunomodulatory), and P (antiparasitic) there were no significant differences between Vienna and residual Austria.

Figure 6 Prescription frequencies by anatomical class (ATC code level 1) for Vienna compared to all other federal states (0-19 years); key as per figure 4

4.1.3. Prescription frequency on ATC level 2 (therapeutic class)

Overall 10 most frequent therapeutic classes: For ATC level 2, the 10 most frequently used medicine groups (i.e. therapeutic classes) are displayed in figure 7. Anti-bacterials for systemic use (ATC code J01) ranged on top, followed by drugs for obstructive airway diseases (R03) and anti- inflammatory / antirheumatic agents (M01). Almost one quarter of all prescriptions (24%) amounted to anti-bacterials for systemic use (J01), 6% to medication against obstructive airway diseases (R03), and 4% to anti-inflammatory and antirheumatic agents (M01). This corresponds to 670, 160, and 100 prescriptions per 1000 children, respectively. The top 10 drug groups on ATC level 2 represented 51% of all prescriptions (without counting extemporaneous preparations and preparations without ATC code).

In contrast to systemic anti-bacterials which form the major fraction of anti-infectives for systemic use (J), four different drug groups for the treatment of respiratory system disorders were widely used in primary care setting. Apart from medicines against obstructive airway diseases (R03), nasal preparations (R01), cough and cold preparations (R05), and antihistamins for systemic use (R06), were among the 10 most frequently prescribed substance classes. Similarly, medicines for the nervous system, such as psychoanaleptics (N06), anti-epileptic drugs (N03), and analgesics (N02), contributed to the top 10 of most frequently prescribed drug groups. Finally, anti-acne preparations (D10) were identified among the top 10 drug groups used in primary care setting.

Similar to the analysis on ATC level 1, preparations without ATC code, i.e. either extemporaneous preparations or medicinal products, homeopathics, etc. without ATC code, appeared among the most frequently used drug groups.

Figure 7 Overall top 10 prescription frequencies per ATC code level 2;

J01= Anti-bacterials for systemic use, R03= Drugs for obstructive airway diseases, M01= Anti-inflammatory / antirheumatic agents, R01= Nasal preparations, D10= Anti-acne preparations, N06= Psychoanaleptics, R05= Cough and cold preparations, R06= Antihistamins for systemic use, N03= Antiepileptics, N02= Analgesics, X00= Extemporaneous preparations, Y00= Preparations without ATC code.

10 most frequent therapeutic classes per age group: Figure 8 shows the 10 most frequently prescribed therapeutic classes in the overall cohort (0-19 years) presented by age groups 0-2, 3-9 and 0-19 years. Table 7 shows the 10 most frequently prescribed therapeutic classes in each age group (0-2, 3-9 and 10-19 years).

Anti-bacterials for systemic use (J01) were most frequently prescribed for children at the age of 3-9 (28% versus 21% and 23% of total prescriptions in children 0-2 and 10-19 years, respectively) (figure 8). As well, medication for respiratory diseases, hence medication to treat obstructive airway diseases (R03), cough and cold preparations (R05), nasal preparations (R01), and antihistamins for systemic use (R06) were most often used in children between 3-9 years of age compared to the other 2 age cohorts. The prescription of anti-inflammatory / antirheumatic products slightly increased with age, anti-acne drugs and psychoanaleptics were hardly present in children below 10 years of age On the contrary, analgesics (N02) were most frequently used in the youngest.

42% of prescriptions for children aged less than 3 years were preparations covered within the 10 most frequent therapeutic classes, this number rose with higher age (52% for children at 3-9 years, 59% for adolescents at 10-19 years). Hence, variability of prescriptions at therapeutic class level (ATC level 2) decreased with patients´ age.

Figure 8 Top 10 most frequently prescribed therapeutic class groups (ATC code level 2) overall, frequencies per age cohorts (0-2, 3-9, 10-19 years); key as per figure 7

Table 7 Top 10 most frequently prescribed therapeutic class groups (ATC level 2) in different age cohorts

Prescriptions per 1000 children Therapeutic class (ATC Level 2) (% of total prescriptions in this age group) 0-2 years:

Antibacterials for systemic use (J01) 750 (20.8)

Drugs for obstructive airway diseases (R03) 168 (4.7)

Antidiarrheals, intestinal antiinflammatory/antiinfective agents (A07) 98 (2.8)

Analgesics (N02) 84 (2.3)

Stomatological preparations (A01) 80 (2.2)

Nasal preparations (R01) 77 (2.1)

Antifungals for dermatological use (D01) 70 (2.0)

Cough and cold preparations (R05) 64 (1.8)

Other dermatological preparations (D11) 63 (1.7)

Anti-inflammatory and anti-rheumatic agents (M01) 63 (1.7)

TOTAL: 1517 (42.1)

3-9 years:

Antibacterials for systemic use (J01) 818 (27.8)

Drugs for obstructive airway diseases (R03) 194 (6.6)

Cough and cold preparations (R05) 95 (3.2)

Nasal preparations (R01) 90 (3.1)

Antiinflammatory and antirheumatic agents (M01) 86 (2.9)

Antihistamins for systemic use (R06) 76 (2.6)

Antiepileptics (N03) 58 (2.0)

Antidiarrheals, intestinal antiinflammatory/antiinfective agents (A07) 45 (1.5)

Analgesics (N02) 40 (1.4)

Ophthalmologicals (S01) 36 (1.2)

TOTAL: 1538 (52.3)

10-19 years:

Antibacterials for systemic use (J01) 555 (22.6)

Anti-acne preparations (D10) 155 (6.3)

Drugs for obstructive airway diseases (R03) 142 (5.8)

Anti-inflammatory and anti-rheumatic agents (M01) 120 (4.9)

Psychoanaleptics (N06) 117 (4.8)

Nasal preparations (R01) 85 (3.5)

Antiepileptics (N03) 78 (3.2)

Antihistamins for systemic use (R06) 68 (2.8)

Drugs for acid related disorders (A02) 63 (2.6)

Psycholeptics (N05) 52 (2.1)

TOTAL: 1435 (58.6)

Prescription frequency stratified by geographic region: Figure 9 shows prescription frequencies of therapeutic groups in Vienna compared to all other Austrian federal states. Likewise to analysis on ATC group level 1, differences between Vienna and all other Austrian federal states were seen mainly in those therapeutic groups that comprise medicines with prices below the prescription charge, i.e. anti-inflammatory/antirheumatic agents (M01), nasal preparations (R01) and cough and cold preparations (R05). The other top 10 therapeutic groups were similar to each other in both cohorts.

Figure 9 Top 10 most frequently prescribed therapeutic class groups (ATC code level 2) overall, frequencies per geographical region Vienna compared to all other federal states (0-19 years); key as per figure 7

4.1.4. Prescription frequency on ATC level 5 (chemical substance)

Table 8 shows the 5 most frequently prescribed substance groups (ATC level 5) per each anatomical class for all children overall, and per age group. Among anti-infectives for systemic use (ATC code J), the most frequently used preparations were penicillins (amoxicillin and enzyme inhibitor, phenoxymethylpenicillin), 2nd generation cephalosporines (cefaclor), and macrolides (clarithromycin, azithromycin). As already described above (chapter 4.1.3.), peak of use was for children aged 3-9, except for cefaclor which was applied more frequently in younger children (0-2 years).

Among the respiratory system group (ATC code R), drugs against obstructive airway diseases were most often prescribed, both for systemic (montelukast) and for inhalative (salbutamol, fluticasone) use. Furthermore, antihistamines, like desloratadine, and sympathomimetics and corticosteroids (mometasone) were most commonly prescribed. Likewise, these medications were most frequently administered to 3-9 year old children. Within the nervous system group (ATC code N), the analgesic drug paracetamol was found among the most regularly prescribed medications, mainly in children from 0-2 years. Moreover, the centrally acting sympathomimetic (methylphenidate) and the antiepileptic drug valproic acid, as well as the antidepressant sertraline could be found among the top 5 of this anatomical class group. The former were more frequently prescribed in the age cohort 3-9 years, whereas use of the latter increased with age (10-19 years). In anatomical group A (alimentary tract and metabolism), anti-diarrheal preparations like lactic acid producing organisms and oral rehydration electrolyte formulations were most often prescribed in children aged 0-2. Fast acting insulin (insulin aspart) and the proton pump inhibitor pantoprazole ranked among the most frequently prescribed preparations in this ATC group, however these drugs were mainly prescribed for older children (10-19 years). The same applied for anti-acne drugs, both systemically (isotretinoin) or topically administered (clindamycine, adapalen, acelaic acid) within ATC group D (dermatologicals). Besides, topical antimycotics and other dermatological preparations belonged to the preparations most frequently prescribed within this ATC group, the latter mainly for children up to 2 years of age. Among preparations structured within ATC group M (musculoskeletal system), non-steroidal antiphlogistic medications (mefenamic acid, ibuprofen, naproxen, diclofenac) were most frequently prescribed, peak level of use was at the age between 3 and 9. Drugs less often prescribed were antithrombotic agents (enoxaparin, dalteparin), iron preparations and blood coagulation factor, listed in the blood and blood forming organs group (ATC group B) as well as ACE inhibitors (enalapril, ramipril), calcium channel blockers (amlodipine), and adrenalin (epinephrine) as cardiovascular preparations. Drugs mainly prescribed for genitourinary indications comprised oxybutynin and estriol preparations, the former in age groups 3-9 and 10-19, the latter particularly in younger children. Amongst systemic hormonal preparations (excl. sex hormones and insulin) prednisolone, desmopressin, levothyroxine and somatropin were the most commonly prescribed drugs. Anti-allergic and anti-infective eye drop preparations were most frequently applied within the sensory organ group (ATC group S), immunosuppressive drugs (azathioprine, tacrolimus, adalimumab), methotrexate and other immunostimulants within ATC group L (antineoplastic and immonomodulating agents), and permethrin, metronidazole, pyrantel and mebendazole against parasites (ATC group P).

There were some differences in most often prescribed chemical substances between Vienna and the rest of Austria which may partly but not entirely be a result of the uneven distribution of patients exempted from prescription charges (data not shown).

Table 8 Five most frequently prescribed substance groups per each anatomical class (ATC level 1) in Austria

0 - 19 years 0 - 2 years 3 - 9 years 10 - 19 years Drug class and name Prescriptions per % prescriptions Prescriptions per % prescriptions Prescriptions per % prescriptions Prescriptions per % prescriptions 1000 children (0-19) 1000 children (0-2) 1000 children (3-9) 1000 children (10-19) Alimentary tract and metabolism (A) Pantoprazole 18 0,7 0 0,0 1 0,0 34 0,0 Insulin aspart 16 0,6 1 0,0 7 0,2 27 0,2 Oral rehydration salt formulations 16 0,6 47 1,3 20 0,7 4 0,5 Lactic acid producing organisms 14 0,5 17 0,5 14 0,5 12 0,4 Miconazole 11 0,4 66 1,8 3 0,1 1 0,1 SUB-TOTAL 2,7 3,6 1,5 1,2 Blood and blood forming organs (B) Enoxaparin 15 0,6 1 0,0 1 0,0 29 0,0 Coagulation factor VIII 5 0,2 1 0,0 4 0,1 6 0,1 Dalteparin 3 0,1 0 0,0 0 0,0 7 0,0 Elektrolytes 3 0,1 5 0,1 3 0,1 2 0,1 Various (iron) combinations 2 0,1 4 0,1 3 0,1 1 0,1 SUB-TOTAL 1,0 0,3 0,4 0,3 Cardiovascular system (C) Epinephrine 4 0,1 1 0,0 4 0,1 4 0,1 Enalapril 2 0,1 0 0,0 0 0,0 3 0,0 Amlodipine 1 0,1 0 0,0 1 0,0 2 0,0 Diosmin, combinations 1 0,0 0 0,0 0 0,0 2 0,0 Ramipril 1 0,0 0 0,0 0 0,0 2 0,0 SUB-TOTAL 0,3 0,0 0,2 0,2 Dermatologicals (D) Isotretinoin 35 1,3 0 0,0 0 0,0 67 0,0 Other dermatological preparations 18 0,6 49 1,4 19 0,7 9 0,5 Clindamycin 11 0,4 0 0,0 0 0,0 21 0,0 Azelaic acid 10 0,4 0 0,0 0 0,0 19 0,0 Adapalene 10 0,3 0 0,0 0 0,0 18 0,0 SUB-TOTAL 3,0 1,4 0,7 0,5 Genitourinary system and sex hormones (G) Oxybutynin 4 0,1 0 0,0 8 0,3 2 0,2 Estriol 4 0,1 13 0,4 4 0,1 1 0,1 Trospium 3 0,1 0 0,0 2 0,1 4 0,0 Clindamycin 2 0,1 0 0,0 0 0,0 4 0,0 Dydrogesterone 2 0,1 0 0,0 0 0,0 4 0,0 SUB-TOTAL 0,5 0,4 0,5 0,4 Systemic hormonal preparation, excl. sex hormones and insulins (H) Prednisolone 14 0,5 44 1,2 16 0,5 5 0,4 Desmopressin 7 0,3 0 0,0 13 0,5 5 0,4 Somatropin 7 0,2 0 0,0 5 0,2 10 0,1 Levothyroxine sodium 4 0,2 0 0,0 1 0,0 8 0,0 Methylprednisolone 1 0,1 0 0,0 0 0,0 3 0,0 SUB-TOTAL 1,2 1,3 1,2 0,9 52

0 - 19 years 0 - 2 years 3 - 9 years 10 - 19 years Drug class and name Prescriptions per % prescriptions Prescriptions per % prescriptions Prescriptions per % prescriptions Prescriptions per % prescriptions 1000 children (0-19) 1000 children (0-2) 1000 children (3-9) 1000 children (10-19) Antiinfectives for systemic use (J) Amoxicillin and enzyme inhibitor 167 6,0 174 4,8 193 6,6 147 5,1 Cefaclor 117 4,2 269 7,5 199 6,8 21 5,3 Clarithromycin 65 2,3 37 1,0 67 2,3 71 1,8 Benzathine phenoxymethylpenicillin 63 2,3 55 1,5 137 4,7 16 3,6 Azithromycin 60 2,2 54 1,5 62 2,1 61 1,6 SUB-TOTAL 16,9 16,3 22,4 17,5 Antineoplastic and immunomodulating agents (L) Other immunostimulants 12 0,4 9 0,3 18 0,6 9 0,5 Methotrexate 3 0,1 0 0,0 3 0,1 3 0,1 Azathioprine 3 0,1 0 0,0 0 0,0 5 0,0 Tacrolimus 3 0,1 0 0,0 2 0,1 4 0,0 Adalimumab 1 0,1 0 0,0 0 0,0 2 0,0 SUB-TOTAL 0,8 0,3 0,8 0,6 Musculoskeletal system (M) Mefenamic acid 42 1,5 38 1,1 46 1,6 40 1,2 Ibuprofen 20 0,7 21 0,6 30 1,0 14 0,8 Naproxen 13 0,5 4 0,1 10 0,3 17 0,3 Dexibuprofen 11 0,4 0 0,0 0 0,0 21 0,0 Diclofenac 10 0,4 0 0,0 0 0,0 19 0,0 SUB-TOTAL 3,4 1,7 2,9 2,3 Nervous system (N) Paracetamol 41 1,5 83 2,3 39 1,3 31 1,0 Methylphenidate 30 1,1 0 0,0 24 0,8 42 0,6 Valproic acid 20 0,7 8 0,2 24 0,8 21 0,6 Sertraline 13 0,5 0 0,0 1 0,0 25 0,0 Lamotrigine 11 0,4 0 0,0 4 0,2 18 0,1 SUB-TOTAL 4,2 2,5 3,1 2,5 Antiparasitic products, insecticides, and repellents (P) Permethrin 3 0,1 3 0,1 3 0,1 4 0,1 Metronidazole 2 0,1 0 0,0 0 0,0 4 0,0 Pyrantel 2 0,1 1 0,0 3 0,1 1 0,1 Mebendazole 1 0,0 0 0,0 1 0,0 1 0,0 SUB-TOTAL 0,3 0,1 0,3 0,2 Respiratory system ( R) Montelukast 50 1,8 43 1,2 75 2,6 35 2,0 Salbutamol 33 1,2 28 0,8 34 1,2 33 0,9 Fluticasone 31 1,1 54 1,5 48 1,6 14 1,3 Desloratadine 29 1,1 12 0,4 53 1,8 19 1,4 Sympathomimetics, plain 26 1,0 37 1,0 40 1,4 15 1,1 SUB-TOTAL 6,1 4,8 8,5 6,6 Sensory organs (S) Levocabastine 9 0,3 1 0,0 6 0,2 13 0,2 Gentamicin 5 0,2 14 0,4 7 0,2 2 0,2 Azelastine 5 0,2 0 0,0 6 0,2 6 0,2 Ketotifen 5 0,2 1 0,0 6 0,2 5 0,2 Ciprofloxacin 4 0,2 3 0,1 6 0,2 3 0,2 SUB-TOTAL 1,0 0,5 1,1 0,8 TOTAL 41,4 33,4 43,5 33,9

4.1.5. Prescription frequency of anti-bacterials for systemic use in primary care

In our data-base, 24% of drugs prescribed to children and adolescents in primary care were classed within anti-bacterials for systemic use (ATC code J01). This corresponds to 672 prescriptions per 1000 children, analysed within the entire cohort (0-19 years) (figure 10). Figure 11 and Table 9 show prescription frequencies of systemic anti-bacterials stratified by the three age cohorts. The number of prescriptions increased from 750 prescriptions per 1000 children for children aged 0-2 years to 818 for children pre-school and primary school age (3-9 years) and decreased to 555 prescriptions for older children (10-19 years). Overall, broad spectrum antibiotics (i.e. amoxicillin, ampicillin, pivampicillin, with or without enzyme inhibitor) were most often prescribed (32% of systemic antibacterials prescribed), followed by macrolides (erythromycin, clarithromycin, azithromycin) and 2nd generation cephalosporins, such as cefaclor and cefuroxim (22% and 18%, respectively). Stratified by age, we found the highest levels of broad spectrum antibiotics prescriptions for children aged 3-9 years (254 prescriptions per 1000 children versus 245 (0-2 years) and 186 (10-19 years) prescriptions, respectively). Still, in relation to the total number of systemic antibiotics prescribed in each age cohort, the distribution per age was rather balanced (33%, 31%, and 33% of systemic anti-bacterials prescriptions for children aged 0-2, 3-9, 10-19 years, respectively). Narrow spectrum penicillins, i.e. ß-lactamase sensitive penicillins (e.g. benzylpenicillin, phenoxy-methylpenicillin) and ß-lactamase resistant penicillins (e.g. flucloxacillin) were prescribed almost three times more often in children at 3-9 years of age than in each of the other two age cohorts (147 prescriptions versus 61 (0-2 years) and 55 (10-19 years) prescriptions per 1000 children, respectively). In children aged 0-2, 2nd generation cephalosporins were most frequently used (269 prescriptions per 1000 children, 36% of antibiotics prescribed in this age group). Macrolides were prescribed with a similar frequency in children aged 3-9 and 10-19 (154 and 152 prescriptions per 1000 children, respectively). However, comparing the patterns of systemic antibiotics prescription in both age groups, we found that proportion of macrolide prescription increased with age (19% versus 27% of all antibiotics prescribed in the 3-9 year old versus 10-19 year old patients, respectively). Generally, we observed an increase of diversity of systemic antibacterial prescriptions with age. Whereas more than 90% of prescriptions for children below 10 years of age were penicillins, 2nd generation cephalosporins, and macrolides, adolescents were additionally treated with lincosamides, sulphonamides/trimethoprim, tetracyclins and antibacterials that we defined as “others” (in total 17% of antibiotics prescriptions in patients aged 10-19).

Figure 10 Overall prescription frequencies of anti-bacterials for systemic use; described by prescriptions per 1000 insured children and by percentage of prescriptions of systemic antibiotics (J01)

Figure 11 Prescription frequencies of antibacterials for systemic use, stratified by age cohorts (0-2, 3-9, 10-19 years), described by prescriptions per 1000 insured children

Table 9 Prescription frequencies of anti-bacterials for systemic use, stratified by age cohorts; described by prescriptions per 1000 insured children and by percentage of prescriptions of systemic antibiotics (J01)

0-2 years 3-9 years 10-19 years Percent of Percent of Percent of Prescriptions per prescriptions of Prescriptions per prescriptions of Prescriptions per prescriptions of 1000 children systematic 1000 children systematic 1000 children systematic antibacterials antibacterials antibacterials Total % Total % Total % Broad spectrum penicillins 245 33 254 31 186 33 Narrow spectrum penicillins 61 8 147 18 55 10 Cephalosporins_1st_generation 17 2 11 1 14 3 Cephalosporins_2nd_generation 269 36 201 25 34 6 Cephalosporins_3rd_generation 47 6 44 5 24 4 Other ß-lactam antibiotics 0 0 0 0 0 0 Macrolides 107 14 154 19 152 27 Lincosamides 2 0 4 0 20 4 Sulphonamides / Trimethoprim 2 0 3 0 4 1 Tetracyclins 0 0 0 0 36 6 Others 1 0 2 0 31 6 OVERALL 749 100 816 100 524 94

By analysing the use of systemic anti-bacterials by region (Vienna versus rest of Austria) we could find only slight differences regarding broad and narrow spectrum antibiotics (more in Vienna than other federal states), 2nd generation cephalosporins and macrolides (less in Vienna than other federal states) (figure 12).

Figure 12 Prescription frequencies of anti-bacterials for systemic use, stratified by geographical region; described by prescriptions per 1000 children insured

4.2. Prescription frequency of medicines for children and adolescents in hospital care setting

4.2.1. Study population / data coverage

Hospital data included intermediate-care paediatric units, neonatology units, psychiatric and surgical wards caring for children, where applicable. Hospitals included in the survey covered about 50% of in-patient places for children and adolescents in Austria. Table 10 shows prescription data for hospital care data as average numbers of prescriptions per day and per hospital bed. In our dataset, AKH Vienna showed highest prescription frequencies, followed by LKH St. Pölten and the University Hospital of Innsbruck. Overall, we determined an average of 0,9 prescriptions per hospital bed and day. We defined University Hospital Vienna (AKH), University Hospital of Innsbruck (UK) and Paracelsus Medical University Salzburg (PMU) as tertiary care institutions; these three hospitals taken together amounted to 46% of all prescriptions in the data set, while 54% of prescriptions were in the remaining so-defined secondary care hospitals.

Table 10 Prescription data for hospital care setting

Average N of Average N of daily prescriptions N of paediatric Hospital prescriptions 95%CI per day, hospital beds per hospital abs. (rel.) bed OVERALL 900 (100%) 990 0,9 0,89 - 0,93 AKH Wien 202 (22,4%) 212 1,0 0,95 - 0,95 LKH St. Pölten 152 (16,9%) 67 2,3 1,92 - 2,74 UK Innsbruck 145 (16,1%) 117 1,2 1,14 - 1,36 Klinikum Wels 78 (8,6%) 43 1,8 1,52 - 2,27 PMU Salzburg 68 (7,5%) 125 0,5 0,46 - 0,63 Klinikum Klagenfurt 64 (7,1%) 104 0,6 0,52 - 0,70 Wilhelminenspital 55 (6,1%) 112 0,5 0,40 - 0,58 LKH Villach 44 (4,9%) 34 1,3 1,15 - 1,59 LKH Feldkirch 25 (2,8%) 30 0,8 0,66 - 0,93 LKH Leoben 22 (2,4%) 69 0,3 0,22 - 0,44 Klinikum Grieskirchen 19 (2,1%) 17 1,1 1,03 - 1,46 Barmherzige Brüder 16 (1,7%) 30 0,5 0,36 - 0,70 LKH Bregenz 12 (1,3%) 30 0,4 0,25 - 0,58

4.2.2. Prescription frequency on ATC level 1 (anatomical class)

Overall frequency: Figure 13 shows overall prescription frequencies of medicines in the hospital setting allocated to ATC level 1. Table 11 includes prescription frequencies in the different hospitals showing inter-hospital minimum and maximum ranges. Among the anatomical classes there was significant variation between hospitals (p=0.000, right side of table). Anti-infective drugs were most frequently prescribed for children and adolescents in Austrian hospitals (figure 13, table 11), with an average frequency of 24% (range between hospitals 10-47%), followed by preparations classified in anatomical class N (nervous system) and A (alimentary tract and metabolism) with 13% and 12%, respectively (range 7-25% and 8- 20%). Drugs of the blood and blood forming system (B), dermatologicals (D), as well musculoskeletal (M) and respiratory system (R) showed relative prescription frequencies between 5 and 10%, whereas the prescription ranges varied much for the different hospitals, especially for musculoskeletal (2-28%) and respiratory drugs (3-20%). 4% of drugs used were assigned to ATC classes C (cardiovascular), H (systemic hormones, without sex hormones and insulins) and S (sensory organs) each, with hospital ranges of 1-9%, 2-9% and 1-7%, respectively. Extemporaneous preparations and drugs without ATC code were prescribed with a frequency of 11%, however as stated above, we could not further investigate this group of drugs.

Figure 13 Overall prescription frequencies to children in Austrian hospitals;

A= Alimentary tract and metabolism, B= Blood and blood forming organs, C= Cardiovascular system, D= Dermatologicals, G= Genito urinary system and sex hormones, H= Systemic hormonal preparation, excl. sex hormones and insulins, J= Anti-infectives for systemic use, L= Antineoplastic and immunomodulating agents, M= Musculoskeletal system, N= Nervous system, P= Antiparasitic products, insecticides, and repellents, R= Respiratory system, S= Sensory organs, V= Various, Z= extemporaneous preparations and preparations without ATC code.

Table 11 Prescription frequencies, overall and per hospital (minimum and maximum values for the respective ATC group underlined)

Average BHB Klinikum Klinikum Klinikum LKH LKH LKH LKH LKH PMU UK Wilhelminen- ATC Code Level 1 prescription AKH Wien Eisenstadt Grieskirchen Klagenfurt Wels Bregenz Feldkirch Leoben St. Pölten Villach Salzburg Innsbruck spital frequency % % % % % % % % % % % % % % A (Alimentary tract and metabolism) 11,5 11,2 19,7 18,6 9,9 15,8 8,8 12,4 9,3 7,7 20,0 14,6 8,8 11,1 B (Blood and blood forming organs) 5,6 9,8 1,4 2,4 3,2 3,5 2,8 2,7 2,3 5,0 5,7 8,8 4,0 3,0 C (Cardiovascular system) 4,3 6,1 4,2 2,1 1,6 1,9 0,6 6,5 1,1 1,9 1,0 4,3 9,0 1,8 D (Dermatologicals) 5,2 7,0 8,3 4,9 4,6 1,0 4,6 8,1 6,2 2,1 10,6 6,6 1,6 14,4 G (Genito urinary system and sex hormones) 0,1 0,1 0 0 0 0,2 0 0 0,1 0,1 0,1 0,2 0 0 H (Systemic hormonal preparation, excl. sex 3,8 hormones and insulins) 4,6 6,2 8,7 2,3 4,9 2,8 2,1 3,5 4,1 3,1 3,9 1,7 4,1 J (Antiinfectives for systemic use) 23,8 28,0 17,8 10,1 47,2 28,3 32,7 26,0 41,5 14,3 18,5 22,5 16,3 20,3 L (Antineoplastic and immunomodulating 1,4 agents) 1,3 0,5 0,0 1,6 1,9 1,3 0,4 2,0 0,0 0,4 3,0 2,8 0,2 M (Musculoskeletal system) 9,2 2,7 6,1 17,3 5,9 14,0 7,9 5,7 4,0 28,2 6,2 6,0 1,5 5,6 N (Nervous system) 12,7 12,2 13,6 25,3 14,0 15,9 19,8 25,2 12,6 8,4 20,7 17,3 6,5 12,1 P (Antiparasitic products, insecticides, and 0,2 repellents) 0,1 0,1 0,3 0,1 0,3 0,2 0 0,1 0,5 0,1 0,1 0 0,1 R (Respiratory system) 7,3 7,7 17,7 5,2 6,1 4,2 16,0 8,8 12,8 4,4 11,2 7,6 3,0 19,8 S (Sensory organs) 3,9 7,3 4,2 0,5 3,2 4,2 2,3 2,1 2,3 2,3 2,4 3,9 1,4 5,8 V (Various) 0,3 0,3 0 0 0,3 0 0,3 0,1 0,5 0 0,1 0,6 0,7 0 Z (Extemp. preparations, without ATC) 10,9 1,7 0,2 4,6 0 3,7 0 0 1,8 21,0 0 0,6 37,7 1,6

Frequency stratified by hospital care type:

Figure 14Figure 14 shows prescription frequencies of medicines grouped on ATC level 1 and stratified by hospital care type. Stratification by hospital care type revealed significant differences in prescription patterns (p=0.000), mainly for drugs used for the muskuloskeletal (M) (15% versus 3% in in secondary and tertiary care type hospitals, respectively) and cardiovascular system (C) (3% versus 12%). Furthermore, higher prescription levels were seen in tertiary compared to secondary care type for blood products (B) (8% versus 4%), antineoplastic and immunomodulating agents (2% versus 0,7%) and sensory organ drugs (5% versus 3%), lower prescription frequencies were observed for drugs for the nervous (11% versus 14%) and the respiratory system (6% versus 8%). This reflects specialised care for children with cardiac and/or haematological diseases in tertiary care hospitals, while medicines for the muscuoskeletal and nervous system (e.g. pain medication) might be administered more frequently in secondary care hospitals. Anti-infective drugs, dermatologicals, hormonal preparations and medication for the alimentary system were prescribed with approximately equal frequency in both hospital care types.

Figure 14 Prescription frequencies stratified by care type; key as per figure 13

4.2.3. Prescription frequency on ATC level 2 (therapeutic class)

Overall 10 most frequent therapeutic classes: Figure 15 shows the 10 most frequently prescribed therapeutic classes, prescription frequencies per hospital are additionally displayed in table 12. On therapeutic class level, anti-bacterials for systemic use (J01) were most frequently prescribed (21% of overall prescriptions, inter-hospital range 10-45%), followed by anti- inflammatory / antirheumatic drugs (9%, inter-hospital range 1-28%) and analgesics (5%, inter-hospital range 2-18%). There was considerable variation in prescription frequencies between hospitals (p=0.000), ranging between prescription frequencies of 0-1% at minimum and 5-11% at maximum per hospital for ophthalmologicals (S01), nasal preparations (R01), corticosteroids for systemic use (H02), psychoanaleptics (N06), anaesthetics (N01), preparations for the treatment of wounds and ulcers (D03), and drugs for obstructive airways diseases (R03). Prescriptions of proprietary medicinal products or extemporaneous preparations without ATC code amounted to 11%. For lack of information on active substance, these were excluded from further analysis.

Overall, more than half (53%) of all drugs prescribed were covered by the 10 most often used groups measured on therapeutic drug level (ATC level 2). Again, the range between hospitals was rather wide: whereas only 24% of all prescribed drugs were covered by the 10 most often used therapeutic drug groups at the University Hospital of Innsbruck, these covered around 75% of total drugs prescribed at the Klinikum Klagenfurt and LKH Bregenz.

Figure 15 Top 10 overall prescription frequencies per ATC code level 2;

J01= systemic antibiotics, M01= anti-inflammatory / antirheumatic agents, N02= analgesics, S01= ophthalmologicals, R01= nasal preparations, H02= corticosteroids for systemic use, N06= psychoanaleptics, N01= anaesthetics, D03= preparations for the treatment of wounds and ulcers, R03= drugs for obstructive airway diseases, Z00= extemporaneous preparations and preparations without ATC code.

Table 12 Prescription frequencies / ATC level 2, overall and per hospital (minimum and maximum values underlined)

Average BHB Klinikum Klinikum Klinikum LKH LKH LKH LKH LKH PMU UK Wilhelminen- prescription AKH Wien Eisenstadt Grieskirchen Klagenfurt Wels Bregenz Feldkirch Leoben St. Pölten Villach Salzburg Innsbruck spital frequency ATC Code Level 2

% % % % % % % % % % % % % %

Antibacterials for systemic use (J01) 20,9 23,7 16,1 10,0 45,4 25,7 31,7 21,6 39,8 13,7 16,5 17,3 12,9 17,3

Anti-inflammatory and anti-rheumatic agents (M01) 8,5 2,0 5,7 15,3 5,5 13,6 6,5 4,3 3,5 27,5 5,3 4,4 1,0 5,4

Analesics (N02) 4,7 1,9 4,4 7,5 7,2 6,4 14,6 18,3 6,3 3,3 12,0 4,5 2,5 3,0

Ophthalmologicals (S01) 3,8 7,3 2,0 0,2 3,2 4,2 2,3 2,0 2,2 2,3 2,4 3,8 1,3 5,6

Nasal preparations (R01) 3,2 3,9 9,9 3,3 3,3 0,9 11,4 2,8 7,3 0,7 7,1 1,5 0,4 11,4

Corticosteroids for systemic use (H02) 3,1 3,3 6,1 3,9 2,2 4,5 2,6 1,9 3,1 4,0 0,9 3,4 1,6 4,0

Psychoanaleptics (N06) 3,0 4,9 0,9 7,9 4,1 4,6 0,6 1,2 0,6 2,0 3,4 1,4 1,1 2,5

Anaesthetics (N01) 2,2 2,2 1,8 0,5 1,3 1,1 2,8 2,8 3,6 1,2 4,5 6,4 1,8 1,1

Preparations for the treatment of wounds and ulcers (D03) 2,1 3,0 0,5 0,1 2,2 0,2 1,2 2,6 0,8 1,4 6,9 3,5 0,4 4,2

Drugs for obstructive airway diseases (R03) 2,0 2,2 4,9 0,2 1,2 1,0 1,8 3,4 2,4 1,2 2,4 3,6 1,1 5,2

TOTAL 53,4 54,4 52,4 48,8 75,7 62,2 75,5 60,9 69,6 57,2 61,4 49,7 24,1 59,6

Prescription frequency of therapeutic classes (ATC level 2) stratified by hospital care type: Figure 16 shows sub-group comparison according to hospital care type (secondary versus tertiary care hospital). The biggest differences were seen in anti-inflammatory / antirheumatic (M01) (14% versus 2%) and analgesic medicine (N02) use (7% versus 3%) between secondary versus tertiary care hospitals, respectively (p=0.000). Systemic anti-bacterials (J01), nasal preparations (R01), and corticosterioids for systemic use (H02) were prescribed more frequently in secondary care hospitals whereas ophthalmologicals (S01) and anaesthetics (N01) were dispensed more frequently in tertiary care type hospitals. There were no significant differences between the two hospital care types regarding prescription of preparations for the treatment of wounds and ulcers (D03) and drugs for obstructive airway diseases (R03).

Figure 16 Top 10 prescription frequencies per ATC code level 2, stratified by hospital care type; key as described in figure 15

4.2.4. Prescription frequency on ATC level 5 (chemical substance)

Table 13 gives detailed information on the five most common preparations in each ATC anatomical class analysed overall for the Austrian hospital setting. Comparison between secondary and tertiary hospital settings is not shown as separate table but mentioned where relevant in the text. The five most frequently used medicines in ATC class J (anti-infectives for systemic use) amounted to 11% of overall prescriptions. Among these were broad spectrum penicillins (amoxicillin and enzyme inhibitor, ampicillin), cephalosporins (cefuroxim), as well as narrow spectrum antibiotics, like gentamicin and vancomycin. Regarding the medicines prescribed, there were prescription differences between secondary and tertiary care hospitals, for further information refer to chapter 4.2.5. The five most often used medicines for musculoskeletal system (M) and nervous system (N) represented 8% each of overall prescriptions of these ATC classes. Among medicines for the musculoskeletal system (M), mainly non-steroidal antiphlogistics (e.g. ibuprofen, mefenamic acid, diclofenac) were prescribed, in the nervous system group, analgesics like paracetamol, respiration stimulant (caffeine) and other psychostimulants, anaesthetics, hypnotics and sedatives (chloral hydrate, diazepam). In both ATC groups, variability of substances used was higher in tertiary care setting than in secondary care setting: the five most frequently used medicines of the musculoskeletal system amounted to 14% and 2% of overall prescriptions in this ATC class in secondary and tertiary care, respectively. Similarly, the five most frequently used medicines in ATC class N (“nervous system”) were 10% and 5% of overall prescriptions in this class in secondary and tertiary care setting, respectively (data not shown). The five most frequently used medicines in ATC class A (alimentary) and D (dermatologicals) constituted 4% of overall prescriptions in each of the corresponding ATC groups. Regarding the use of dermatologicals (D), there were no differences between hospital care types; tars, antibiotics and antimycotics for topical use were most often prescribed. Among alimentary medicines (A), pantoprazole was among the top 5, as well as ondansetrone, and various electrolytes. Among respiratory drugs (R), salbutamol and nasal preparations for topical use (oxymetazoline, xylometazoline) were applied. Among blood and blood forming medicines (B), phytomenadione, coagulation factors and antithrombotic drugs like heparins (enoxaparine, dalteparine) and erythropoietin were used. Epinephrine, norepinephrine, atenolol, dopamine, and clonidine were the five most often prescribed cardiovascular medicines (C). There were slight differences between care types (data not shown).

Among systemic hormonal preparations (H), corticosteroids and levothyroxine were the most frequently prescribed medicines. Among medicines for sensory organs (S), ophthalmologicals, like retinol, phenylephrine, gentamicin, and povidone iodine were most common. Summarising the top 5 drugs of each ATC group amounted to 51,5% of overall prescriptions.

Table 13 Five most frequently prescribed chemical substances (ATC level 5) per ATC group in Austrian hospitals, overall

Drug class and name % prescriptions Alimentary tract and metabolism (A) Pantoprazole 1,1 Ondansetron 0,8 Magnesium gluconate 0,6 Benzydamine 0,6 Silicones 0,5 SUB-TOTAL 3,6 Blood and blood forming organs (B) Phytomenadione 1,4 Albumin 1,1 Heparin 0,6 Enoxaparin 0,3 Erythropoietin 0,3 SUB-TOTAL 3,6 Cardiovascular system (C) Clonidine 0,9 Atenolol 0,5 Epinephrine 0,4 Dopamine 0,4 Norepinephrine 0,2 SUB-TOTAL 2,4 Dermatologicals (D) Dexpanthenol 2,0 Tars 0,6 Nystatin 0,5 Other antibiotics for topical use 0,4 Clotrimazole 0,3 SUB-TOTAL 3,8 Systemic hormonal preparation, excl. sex hormones and insulins (H) Prednisolone 1,9 Hydrocortisone 0,4 Dexamethasone 0,3 Methylprednisolone 0,3 Levothyroxine sodium 0,2 SUB-TOTAL 3,1 Antiinfectives for systemic use (J) Ampicillin 4,0 Cefuroxim 2,0 Vancomycin 1,9 Amoxicillin and enzyme inhibitor 1,9 Gentamicin 1,1 SUB-TOTAL 10,9

Drug class and name % prescriptions Antineoplastic and immunomodulating agents (L) Infliximab 0,2 Basiliximab 0,1 Methotrexate 0,1 Ifosfamide 0,1 Tacrolimus 0,1 SUB-TOTAL 0,6 Musculoskeletal system (M) Ibuprofen 5,2 Mefenamic acid 1,2 Dexibuprofen 0,8 Diclofenac 0,7 Ketoprofen 0,3 SUB-TOTAL 8,1 Nervous system (N) Paracetamol 3,9 Caffeine 2,6 Combinations (local anaesthetics, amides) 0,7 Chloral hydrate 0,6 Diazepam 0,4 SUB-TOTAL 8,2 Antiparasitic products, insecticides, and repellents (P) Permethrin 0,1 SUB-TOTAL 0,1 Respiratory system ( R) Oxymetazoline 1,2 Salbutamol 1,0 Xylometazoline 0,6 Natural phospholipids 0,4 Various (nasal preparations) 0,3 SUB-TOTAL 3,6 Sensory organs (S) Retinol 1,8 Gentamicin 0,4 Artificial tears and other indifferent preparations 0,4 Phenylephrine 0,3 Povidon-Iodine 0,3 SUB-TOTAL 3,1 Various (V) Calcium folinate 0,1 Mesna 0,1 SUB-TOTAL 0,2 TOTAL 51,5

4.2.5. Prescription frequency of anti-bacterials for systemic use in hospital care

Figure 17 shows overall prescription frequencies of anti-bacterials for systemic use in Austrian hospital care setting. Anti-bacterials for systemic use (J01) constituted 21% of total prescriptions in the hospitals studied. Thereof, 35% (inter-hospital range 17-60%) were broad spectrum penicillins, 18% (inter-hospital range 0-23%) cephalosporins of 1st, 2nd, and 3rd generation, 5% (inter-hospital range 2-16%) macrolides, and 28% (inter-hospital range 11-42%) were antibiotics like fluorochinolones (ciprofloxacin), aminoglycosides (gentamicin, tobramycin), fusidinic acid, linezolid, vancomycin, nitrofurantoin, etc. which we defined as “others” (figure 17).

Figure 17 Overall prescription frequencies of anti-bacterials for systemic use (J01)

Subgroup analysis on systemic antibiotics prescriptions by hospital care type is shown in figure 18. There were more prescriptions of antibiotics for systemic use in secondary than in tertiary care hospitals (58% versus 42%). Broad spectrum antibiotics and 2nd generation cephalosporins were more frequently prescribed in secondary care hospitals (37 versus 32% and 13% versus 7%, respectively), whereas other anti-bacterials were more frequently prescribed in tertiary hospitals (23% versus 34%). Similarly, trimethoprim and sulfonamides were more often used in tertiary than in secondary care (0,4% versus 4%).

Figure 18 Prescription frequencies of antibacterials for systemic use (J01), stratified by hospital care type

Table 14 shows that 78% of antibiotics dispensed in hospitals were for parenteral administration, 13% for oral administration, and in 9% both forms of administration were possible (p=0.000). This is due to the fact that information on medicines dispensing in hospitals was partly provided on ATC level 5 which only determines chemical substance but not mode of application or preparation name (also refer to methods section, chapter 3.4.2.). Only macrolides were more frequently administered via enteral than parenteral pathway.

Table 14 Mode of administration of systemic anti-bacterials in hospital

Percent of overall prescriptions J01 Type of antibacterial enteral parenteral application administration application not known % % % Broad spectrum penicillins 3,9 29,3 1,9 Narrow spectrum penicillins 1,2 2,2 0,1 Cephalosporins 1st generation 0,3 0,5 0,0 Cephalosporins 2nd generation 2,0 4,6 3,6 Cephalosporins 3rd generation 0,3 6,1 0,2 Other ß-lactam antibiotics 0,0 5,2 0,0 Macrolides 2,1 1,8 0,6 Lincosamides 0,6 2,8 0,1 Tetracyclins 0,5 0,3 0,0 Trimethoprim_Sulfonamides 0,7 1,4 0,0 Others 1,4 24,0 2,4 TOTAL 13,1 78,1 8,8

4.3. Prescription frequency of medicines in primary care versus hospital care setting

Prescription frequencies were compared between primary care and hospital settings. To overcome the differences in assessment of absolute quantities in the two settings, relative frequencies were calculated by percentage of prescriptions in relation to the overall prescriptions in the respective setting. Thereby, differences in distribution between drug groups were compared. Dispensings were used as proxy for prescription frequencies in hospitals.

4.3.1. Comparison on ATC level 1 (anatomical class)

Comparison of prescription frequencies in primary and hospital care setting is shown in figure 19. Medicines of most anatomical groups were relatively more frequently prescribed in hospital than in primary care setting (p=0.000). Particulary, the use of cardiovascular drugs was 10-fold higher in hospital than in primary care. Inversely, medicines for the respiratory (14% versus 7% of overall prescriptions per care type), genito-urinary system (0,9% versus 0,1%), and extemporaneous preparations or preparations without ATC code were more frequent in the primary care setting. Prescription frequencies of anti-infectives (25% versus 23%) and dermatologicals (7% versus 5%) were almost similar between the care settings.

Figure 19 Prescription frequencies in primary versus hospital care setting; A= Alimentary tract and metabolism, B= Blood and blood forming organs, C= Cardiovascular system, D= Dermatologicals, G= Genito urinary system and sex hormones, H= Systemic hormonal preparation, excl. sex hormones and insulins, J= Anti-infectives for systemic use, L= Antineoplastic and immunomodulating agents, M= Musculoskeletal system, N= Nervous system, P= Antiparasitic products, insecticides, and repellents, R= Respiratory system, S= Sensory organs, V= Various, Z= Extemporaneous preparations and preparations without ATC code.

4.3.2. Comparison on ATC level 2 (therapeutic class)

Table 15 and figure 20 show the top 10 therapeutic classes in each, primary care and hospital setting. Six of these groups belong to the 10 most frequently used ones in both settings, namely anti- bacterials for systemic use (J01), anti-inflammatory / antirheumatic agents (M01), analgesics (N02), psychoanaleptics (N06), drug for obstructive airway diseases (R03), and nasal preparations (R01). The remaining four most frequent therapeutic classes in hospital setting comprise ophthalmologicals (S01), corticosteroids for systemic use (H02), preparations for the treatment of wounds and ulcers (D03), and anaesthetics (N01). In primary care setting, anti- acne preparations (D10), anti-epileptic drugs (N03), cough and cold preparations (R05), and anti-histamins for systemic use (R06) comprise the remaining four top 10 therapeutic classes.

These top 10 therapeutic classes cover 54% (primary care) and 56% (hospital care) of overall prescriptions, respectively. Comparison of therapeutic groups most frequently prescribed in both, primary and hospital care setting, showed higher prescription rates of pain medication in hospitals (M01: 4% versus 9%, N02: 2% versus 5% in primary versus hospital care setting, respectively). Administration of systemic antibiotics (J01: 24% versus 21% in primary versus hospital care setting) and medicines for obstructive airway diseases (R03: 6% versus 2% in primary versus hospital care setting) was lower in hospital compared to primary care setting. Psychoanaleptics (N06: 3% each setting) and nasal preparations (R01: 3% each setting) showed no quantitative difference in use.

Table 15 Top 10 therapeutic classes (ATC level 2) per care setting; among top 10 in both care settings are marked in red, among top 10 in hospital care are marked in blue, among top 10 in primary care are marked in green.

Primary Hospital care care overall overall prescription ATC code level 2 prescription frequency frequency % % p-value Antibacterials for systemic use (J01) 24,3 20,9 Anti-inflammatory and anti-rheumatic agents (M01) 3,6 8,5 Drugs for obstructive airway diseases (R03) 5,9 2,0 Analgesics (N02) 1,8 4,7 Psychoanaleptics (N06) 2,6 3,0 Nasal preparations (R01) 3,1 3,2 SUB-TOTAL (top 10 both settings) 41,3 42,3 0,54 Ophthalmologicals (S01) 1,3 3,8 Corticosteroids for systemic use (H02) 0,7 3,1 Preparations for the treatment of wounds and ulcers 0,1 2,1 (D03) Anaesthetics (N01) 0,2 2,2 Anti-acne preparations (D10) 2,9 0,1 Antiepileptics (N03) 2,3 1,1 Cough and cold preparations (R05) 2,4 0,4 Antihistamins for systemic use (R06) 2,4 0,7 Extemporaneous preparations, without ATC (Z00) 27,4 10,9 TOTAL (without Z00) 53,6 55,8 0,17

Figure 20 Top 10 therapeutic classes (ATC level 2) per care system; J01= systemic antibiotics, M01= anti-inflammatory / antirheumatic agents, R03= drugs for obstructive airway diseases, N02= analgesics, N06= psychoanaleptics, R01= nasal preparations, S01= ophthalmologicals, D10= anti-acne medication, H02= corticosteroids for systemic use, D03= preparations for the treatment of wounds and ulcers, N01= anaesthetics, N03= anti-epileptic drugs, R05= cough and cold preparations, R06= antihistamins for systemic use, Z00= extemporaneous preparations and preparations without ATC code.

4.3.3. Comparison on ATC level 5 (chemical substance)

The top 80% of most frequently prescribed chemical substances in primary care setting are shown in appendix 8.1 (top 80% in age group 0-19), the top 80% of most frequently prescribed chemical substances in hospital care setting are described in appendix 8.2. Appendix 8.3 shows a combined list of the 80% most frequently prescribed chemical substances in each, primary care (n=66, age cohort 0-19 years) and hospital setting (n=114).

Of these, 30 chemical substances occur in both, primary and hospital care setting. Among these, amoxicillin and enzyme inhibitor, ibuprofen, and paracetamol are the substances most often prescribed in both settings, followed by cefaclor, clarithromycin, and mefenamic acid. Apart from the ones mentioned above, only phenoxymethylpenicillin, clindamycin, doxycylin, and ciprofloxacin occur among the most frequently used systemic anti-bacterials in both settings, the other antibiotics are predominant either in primary care setting (e.g. azithromycin, amoxicillin, etc.) or in hospital care setting (e.g. ampicillin, cefuroxime, vancomycin). In contrast, most pain medications like the ones mentioned above as well as dexibuprofen and diclofenac appear in both settings. Prednisolone is the only chemical substance of ATC anatomical class H (systemic hormonal preparations excl. sex hormones and insulins) that occurs among the top 80% medicines in both, primary and hospital care settings. Other class H substances like hydrocortisone or dexamethasone are among the top 80% in hospitals only. Regarding medicines for the respiratory system, salbutamol and fluticasone are the substances most frequently prescribed in both settings for obstructive airway diseases. Montelukast is among top 80% only in primary care setting whereas terbutaline only appears in the top 80% hospital care setting. Oxymetazoline and other plain sympathomimetics are the nasal preparations most frequently used in both care settings. Among systemic antihistamins, only levocetirizine is among the top 80% in both settings. In ATC anatomical class A (alimentary system), proton pump inhibitors (pantoprazole, esomeprazole) as well as macrogol, insulin aspart, colecalciferole and miconazole for local oral use appear among the top 80% medicines in both settings. Of medicines for nervous system disorders (N), the anti-epileptic drugs levetiracetam and valproic acid occur among the top 80% in both settings. Psychostimulants (caffeine), anaesthetics (fentanyl, lidocaine, propofol), and sedatives (diazepam, chloral hydrate) are frequent in hospitals, whereas centrally acting sympathomimetics (methylphenidate, atomoxetine), antidepressants (sertraline) and other anti-epileptics were frequent in the primary care setting.

Enoxaparin is the only substance of ATC group B (blood and blood forming organs) that is frequently prescribed in both care settings. Concerning dermatologicals, two substances (nystatin and methylprednisolone aceponate) are among the 30 chemical substances frequently used in both care settings.

4.3.4. Comparison of prescription frequency of anti-bacterials for systemic use

Figure 21 shows prescription frequencies of anti-bacterials for systemic use in primary compared to hospital care setting. Macrolides and 2nd generation cephalosporins, as well as narrow spectrum penicillins (ß- lactamase sensitive, ß-lactamase resistant penicillins) were more frequently prescribed in primary care setting, whereas ß-lactam antibiotics other than cephalosporins of 1st to 3rd generation (monobactam, carbapenem) and “other” antibiotics, such as fluorochinolones (ciprofloxacin), aminoglycosides (gentamicin, tobramycin), fusidinic acid, linezolid, vancomycin, nitrofurantoin, etc. were more frequently used in hospital.

Figure 21 Prescription frequencies of anti-bacterials for systemic use in primary vs hospital care

4.4. Evaluation of licence status of medicines used in children and adolescents

4.4.1. Off-label use in primary care setting

Around 100 different chemical substances (ATC level 5) were identified among 80% of most frequently used drugs in the five different age cohorts (0-2, 3-4, 5-6, 7-9, 10-19 years) in the primary care setting (see appendix 8.5). Licence status of these preparations was studied in the corresponding product information of Austria Codex and off-label use evaluated according to the criteria described in chapter 3.6.1. Among these chemical substances, we found a decreasing percentage of off-label use with increasing age in most anatomical groups, overall from 52% for 0-2 year old to 11% for 10-19 year old children (table 16).

Table 16 Percent off-label use in primary care overall and for anatomical classes by age groups

0-2 3-4 5-6 7-9 10-19 ATC Level 1*) years years years years years A (Alimentary tract and metabolism) 54% 31% 31% 31% 15% B (Blood and blood forming organs) 67% 67% 50% 67% 67% D (Dermatologicals) 64% 64% 64% 64% 0% G (Genito urinary system and sex hormones) 100% 100% 100% 100% 33% H (Systemic hormonal preparation, excl. sex 25% 25% 0% 0% 0% hormones and insulins) J (Antiinfectives for systemic use) 10% 14% 14% 9% 0% L (Antineoplastic and immunomodulating agents) 0% 0% 0% 0% 0% M (Musculoskeletal system) 57% 57% 43% 43% 29% N (Nervous system) 62% 60% 38% 25% 25% R (Respiratory system) 70% 35% 17% 13% 4% S (Sensory organs) 100% 33% 33% 0% 0% V (Various) 100% 0% 0% 0% 0% Overall 52% 40% 31% 26% 11%

*) ATC classes C and P not listed due to non-occurrence among 80% most frequently prescribed chemical substances

Amount of off-label use per ATC group in top 80% of medicines prescribed were as follows (also refer to appendix 8.5): Off-label use was lowest for anti-infectives (such as systemic anti-bacterials, fluconazole and valaciclovir). In ATC respiratory system group (R), 23 chemical substances (nasal preparations for topical use, throat preparations, adrenergics and steroids for inhalation and systemic drugs for obstructive airway disease, cough and cold preparations, and antihistamines) were prescribed in all age cohorts. Off-label use in children at 0-2 years was rather high (70%) but decreased to 4,3% in adolescents. Apart from analgesics (N02) and antiepileptics (N03), preparations of the nervous system (antipsychotics, antidepressants, and centrally acting sympathomimetics) were predominantly administered to children older than 4 years. Off-label use in ATC group “N” decreased from 62% for the 0-2 year old patients to 25% for adolescents. In the dermatological group (D), 8 out of 14 preparations were anti-acne drugs which are defined on-label only in the adolescents group (10-19 years). The remaining dermatological preparations were antifungals for topical use, corticosteroids and other dermatological preparations. In total, more than 60% of drug use was off-label for children below 10 years of age, whereas adolescents were treated completely on-label. Among alimentary system and metabolism group (A), 13 substances were classed among the top 80%. Seven of these (e.g. miconazole, proton pump inhibitors, antidiarrheals, metoclopramide, insulins, vitamin B and D) were defined as off-label which resulted in an off- label use ranging from 54% in children aged 0-2 years to 15% in adolescents. Rather high frequencies of off-label use were seen in genito-urinary (G) and blood forming group (B) with levels of 50% to 67% and 33% to 100%, respectively, however only few substance groups were listed among 80% of the most frequently used drugs. Ophthalmological drugs (ATC class S) were used off-label in children 0-2 years of age.

4.4.2. Off-label use in hospital care setting

Like in primary care setting, around 100 different chemical substance groups were among 80% of the most frequently prescribed medicines in hospital (also refer to appendix 8.6). Licence status of these preparations was studied in the corresponding product information of Austria Codex and off-label use evaluated according to the criteria described in chapter 3.6.1. Similar to the findings in primary care, the average frequency of off-label use was observed between 54% for newborns to 18% for adolescents (table 17). 82

Table 17 Percent of off-label use in hospital care by anatomical class and age

Off-label Off-label Off-label ATC Level 1*) 0 - 1 month°) 0 - 2 years°°) 0 - 19 years°°°) A (Alimentary tract and metabolism) 53% 33% 7% B (Blood and blood forming organs) 56% 44% 33% C (Cardiovascular) 83% 83% 83% D (Dermatologicals) 43% 14% 14% H (Systemic hormonal preparation, 20% 20% 0% excl. sex hormones and insulins) J (Anti-infectives for systemic use) 38% 14% 3% L (Antineoplastic and 100% 50% 0% immunomodulating agents) M (Musculoskeletal system) 83% 50% 33% N (Nervous system) 60% 40% 13% R (Respiratory system) 64% 43% 7% S (Sensory organs) 80% 80% 80% Overall 54% 35% 18% *) ATC classes G and P not listed due to non-occurrence among 80% most frequently prescribed chemical substances °) medicine is approved for children aged 1 month and above, i.e. use is off-label for children aged 0-1 month, °°) medicine is approved for children aged 2 years and above, i.e. use is off-label for children aged 0-2 years, °°°) medicine is approved for patients over 19 years of age, i.e. use is off-label for children aged 0-19 years

Again, anti-infectives, such as antibiotics, antimycotics (amphothericin B, fluconazole), antivirals for systemic use (aciclovir), and immunoglobulins were investigated for licence status. From our data we conclude that off-label use was quite common for newborns, but decreased to 3% for children older than 2 years. Within ATC group A (alimentary system and metabolism), stomatological preparations (miconazole, etc.), proton pump inhibitors (pantoprazole, esomeprazole), drugs for functional gastro-intestinal disorders, antiemetics (ondansetron), nystatin, multienzymes for digestive disorders, as well as vitamin D and mineral supplements were investigated. Off-label use ranged from 53% for newborns to 7% for children above 2 years. Nervous system drugs used in hospital differed quite much from those prescribed in primary care setting. Anaesthetics, analgesics, antiepileptic and sedative drugs, as well as caffeine

83 were prescribed most frequently, for these drugs we determined an off-label use of 60% to 13%. Antithrombotic agents (heparins), vitamin K and anticoagulation agents, erythropoietin, and plasmaproteins were investigated in ATC class B (blood and blood forming agents). Use of 56% of these drugs is not recommended for newborn children, 33% of them not for children below 19 years of age. Most drugs for the respiratory system (R) were nasal preparations for topical use but also antihistamins for systemic use; many of these drugs are indicated neither for newborns nor for toddlers (< 2 years); that´s why we determined an off-label use of 64% and 43% for newborns and children < 2 years of age in this anatomical class. In the cardiovascular drug class (C), adrenergic/dopaminergic drugs (e.g. dopamine, epinephrine) were administered, apart from these substances, antihypertensives (clonidine, spironolactone) and ß-blockers (atenolol). Of these drugs, only epinephrine was recommended for use in children. There were only 7 dermatological substances among 80% of most frequently used preparations (nystatin, clotrimazole, dexpanthenole, dimetindene, tars, antibiotics for topical use, methylprednisolone aceponate), 2 of them were definitely not recommended for the use below 1 month of age; despite otherwise used, we defined tars as off-label which led to an off-label use of 43% in newborns. Similar data were found for the non-sex hormones group (ATC anatomical class H), predominantly glucocorticoids were prescribed and we defined an off-label use rate of 20% for newborns and children below 2 years. Within class M (musculoskeletal system), predominantly non-steroidal anti-inflammatory drugs were used. None of these drugs is recommended for the use in newborns. Finally, within the group of sensory organ drugs (S) ophthalmological preparations were listed among the most often used drugs in hospital. Among these, only gentamicin is recommended for newborns, the use of all others (anti-infectives, mydriatics, others) was defined as off-label.

4.5. Qualitative assessment of medicine use in children and adolescents

4.5.1. Qualitative assessment of prescription habits by interview

Interviews with 10 general paediatricians in primary care: The general paediatricians were asked for their prescribing habits regarding approximately 45 different diseases covering seven therapeutic fields and treatment of unspecific symptoms like fever, acute and chronic pain and nausea/emesis (for detailed list of diseases refer to methods, table 3 in chapter 3.7.1). According to the doctors´ responses, the proposed panel of diseases covered the major part of their daily routine prescriptions. Apart from these, they normally continue treatments for (less common) diseases that have been initiated by paediatric specialists.

Table 18 shows a list of most frequently prescribed medicines and indications for their use. 122 different medicines were mentioned by the doctors (extemporaneous preparations and preparations without ATC code not included) for the different diseases. Ibuprofen showed to have the broadest spectrum of use, it is prescribed for 10 different diseases, followed by systemic antibiotics (penicillins, clarithromycin, cephalosporins), and sympathomimetic nasal preparation oxymetazoline. Furthermore, diphenhydramine are prescribed for 5 of the diseases mentioned in our interviews, lactic acid producing agents, cefalexine, paracetamol, plain sympathomimetics (i.e. single agent sympathomimetics), salbutamol, fluticasone and desloratadine are prescribed in 4 different diseases. The residual 106 substances that were mentioned are used in less indications (6 substances with 3, 18 substances with 2 different indications for use, and 82 with one indication for use each; data not shown).

Table 18 Medicines most often mentioned with indications for their use

Drug (N of different indications Indications for use for use) acute pain, chronic pain, fever, dysmenorrhoea, bacterial Ibuprofen (10) otitis media, secretory otitis media (non bacterial), laryngitis, pharyngitis, herpes simplex infection, varicella bacterial maxillodental infections, bacterial sinusitis, bacterial Amoxicillin and enzyme throat infection, borreliosis, erysipel, impetigo, pneumonia, inhibitor (9) urinary tract infection, vulvitis/balanitis bacterial bronchitis, bacterial throat infection, bacterial Amoxicillin (7) sinusitis, bacterial otitis media, pneumonia, borreliosis, urinary tract infection bacterial bronchitis, bacterial throat infection, bacterial eye Clarithromycin (7) infection, bacterial otitis media, pneumonia, borreliosis, impetigo allergic rhinitis, rhinitis acuta, pharyngitis, secretory otitis Oxymetazoline (7) media (non bacterial), bacterial otitis media, cough, laryngitis (Pseudo-Krupp) bacterial otitis media, bacterial sinusitis, bacterial throat Cefaclor (6) infection, pneumonia, borreliosis, urinary tract infection

bacterial otitis media, bacterial sinusitis, bacterial throat Cefpodoxime (6) infection, pneumonia, impetigo, urinary tract infection

Benzathine bacterial otitis media, bacterial sinusitis, bacterial throat phenoxymethylpenicillin (5) infection, borreliosis, vulvitis/balanitis

cough, laryngitis (Pseudo-Krupp), secretory otitis media (non Diphenhydramine (5) bacterial), nausea/emesis, urticaria

Lactic acid producing acute gastroenteritis, chronic diarrhea, nausea/emesis organisms (4)

Cefalexin (4) bacterial sinusitis, bacterial throat infection, erysipel, impetigo

Paracetamol (4) acute pain, dysmenorrhoea, fever, laryngitis (Pseudo-Krupp)

bacterial otitis media, cough, secretory otitis media (non Sympathomimetics, plain (4) bacterial), rhinitis acuta

Salbutamol (4) acute asthma, chronic asthma, cough, obstructive bronchitis

acute asthma, chronic asthma, laryngitis (Pseudo-Krupp), Fluticasone (4) obstructive bronchitis

Desloratadine (4) allergic conjunctivitis, allergic rhinitis, urticarial, varicella

For 20 of the 45 diseases and symptoms, at least one of the paediatricians responded that they would rather refer the patient to a specialist or to hospital”. Seven of 10 paediatricians would refer the child to a specialist or hospital for the treatment of chronic diarrhoea, chronic pain, or psoriasis; 5 of 10 paediatricians for initial treatment of hypertension in children (table 19).

Table 19 Diseases for which paediatricians refer to specialist or hospital rather than initiate medical treatment themselves

Number of paediatricians Diagnosis referring to specialist or hospital Chronic diarrhoea 7 Chronic pain 7 Psoriasis 7 Contraception 6 Erysipel 6 Hypertension 5 Acne vulgaris 4 Enuresis nocturna 4 Bacterial maxillodental infections 3 Chronic asthma 3 Dysmenorrhoea 3 Secretory otitis media 2 Acute asthma 1 Allergic Conjunctivitis 1 Bacterial eye infections 1 Borreliosis 1 Gastritis/Ulcus 1 Herpes simplex infect. 1 Mycotic infections / skin 1 Nausea / emesis 1

Extemporaneous preparations (ointments, suppositories, nosedrops, etc. manufactured in pharmacies) were prescribed for 10 diseases, preparations without ATC code (i.e. not mentioned as medical drug in Austria Codex) were prescribed for 17 diseases by at least one physician.

Interviews with 4 child psychiatrists: Table 20 shows medicines indicated in the interviews with child psychiatrists used in daily routine, including indication, age of use, and information regarding off-label use. Most frequently prescribed drugs were olanzapine, risperidone and quetiapine in schizophrenia, fluoxetine and sertraline in depression, methylphenidate and atomoxetine in attention deficit

87 hyperactivity syndrome (ADHS). Drugs used for schizophrenic disorders had the highest proportion of off-label use.

Table 20 Use of drugs mentioned by child psychiatrists, including recommended indication and age of use and approval status

Off-label use Chemical Approval status in Therapeutic field Indication for use Age of use in daily substance regards to age routine? Aggression / behavioural disorder Schizophrenia Risperidone > 5 years > 5 years no Depressive episodes Aggression / behavioural disorder Not recommended in Schizophrenia Olanzapine > 15 years yes Depressive episodes children and adolescents Not recommended in Schizophrenia Amisulpride Productive + non-productive psychosis 10-19 years adolescents, yes contraindicated <14 years Bipolar disorders Not recommended in Schizophrenia Quetiapine Behavioural disorder 10-19 years yes children and adolescents Secondary prophylaxis Bipolar disorders Schizophrenia Aripiprazole > 7 years > 13 years yes Secondary prophylaxis Manic episodes Schizophrenia Ziprasidone 10-19 years > 10 years no Bipolar disorders Obsessive-compulsive disorder Behavioural disorder Depression Sertraline Anxiety disorder > 7 years > 6 years no Panic disorder Major depression Bulimia nervosa Panic disorder Depression Fluoxetine Social anxiety disorder > 8 years > 8 years no Posttraumatic stress disorder Sleep disorder Not recommended in Depression Mirtazapine Depression > 16 years yes children and aolescents Depression Escitalopram Social anxiety disorder 10-19 years > 18 years yes Social anxiety disorder Not recommended in Depression Trazodone > 13 years yes Sleep disorder children and aolescents Not recommended in Depression Venlafaxine Depression (resistent) 10-19 years yes children and aolescents ADHS Methylphenidate ADHS > 6 years > 6 years no ADHS Atomoxetine ADHS > 6 years > 6 years no Anxiety Other Hydroxizine > 6 years > 6 years no Sleep disorder

4.5.2. Evaluation of off-label use (based on interviews)

The frequency of off-label use prescriptions based on the interviews with the general paediatricians is shown in table 21. Analysing prescription habits obtained from interviews (general paediatricians only), we found that off-label use was less frequent than found in the entire primary and hospital care data-sets. Frequencies of off-label use were ranging from 16% for patients aged 0-2 to 9% for adolescents. Off-label use was highest for the treatment of skin and eye diseases, followed by ear-nose-throat and respiratory disorders, however, in any case, frequencies of off-label use were decreasing with increasing age.

By considering extemporaneous preparations of labelled medicines as ‘on-label’ use, frequencies of off-label use were even lower.

Table 21 Frequency of off-label use of drugs according to interviews with general paediatricians, described as percent off-label use per therapeutic field and as number of off- label drugs per total number of drugs

0-2 years 3-6 years 7-9 years 10-19 years Therapeutic field Comment % (abs.) % (abs.) % (abs.) % (abs.)

Eye only proprietary medicinal products°) 75 (3/4) 50 (4/8) 13 (1/8) 13 (1/8) Ear-nose-throat including extemp. preparations*) 42 (5/12) 28 (8/29) 10 (4/39) 10 (4/39) only proprietary medicinal products°) 17 (2/12) 14 (4/29) 0 (0/39) 0 (0/39) Respiratory including extemp. preparations*) 23 (6/26) 19 (7/36) 16 (4/25) 15 (4/26) only proprietary medicinal products°) 8 (2/26) 6 (2/36) 4 (1/25) 3 (1/26) Gastro-intestinal including extemp. preparations*) 25 (2/8) 17 (2/12) 13 (2/15) 6 (1/16) only proprietary medicinal products°) 13 (1/8) 8 (1/12) 7 (1/15) 0 (0/16) Uro-genital only proprietary medicinal products°) no prescriptions 0 (0/4) 0 (0/4) 0 (0/4) Skin including extemp. preparations*) 64 (9/14) 36 (5/14) 31 (4/13) 24 (5/21) only proprietary medicinal products°) 36 (5/14) 7 (1/14) 15 (2/13) 10 (2/21) Infections including extemp. preparations*) 6 (4/70) 5 (4/78) 3 (2/68) 3 (2/66) only proprietary medicinal products°) 4 (3/70) 4 (3/78) 1 (1/68) 2 (1/66) Unspecific symptoms only proprietary medicinal products°) 0 (0/10) 0 (0/13) 0 (0/14) 0 (0/13)

TOTAL including extemp. preparations*) 16 (23/144) 15 (30/194) 9 (17/186) 9 (17/198) only proprietary medicinal products°) 7 (10/144) 8 (15/194) 3 (6/186) 3 (5/198) *)Frequency of off-label use if extemporaneous preparations are classified as off-label used medicines °)Frequency of off-label use if extemporaneous preparations are classified as on-label used medicines

4.5.3. Comparison of qualitative data based on interviews to quantitative data obtained from reimbursement institutions

65 medicines at level of chemical substance (ATC level 5) represented 80% of the most frequently prescribed drugs reimbursed by health insurances in Vienna for the entire paediatric population (0-19 years). 42 of these were also mentioned by paediatricians during the interviews. In consequence, there is 65% concordance of the drugs reported by the physicians with the lists provided by HVB.

Eleven percent of drugs reimbursed by HVB in Vienna were extemporaneous preparations. According to the data retrieved by interviews, 12% (14 out of 122 different chemical substance groups) were reported to be prescribed as extemporaneous preparations.

However, the comparison of interview data with the global reimbursement data set needs to be interpreted with caution. The number of general paediatricians interviewed was small and, given they participated voluntarily, this must be considered a selected group of particularly interested and well informed paediatricians. Therefore, the interviewed paediatricians are not fully representative of general prescription habits for children and adolescents in Vienna or Austria.

4.5.4. Variability of prescription habits

Table 22 lists the number of different preparations on ATC level 2 (therapeutic class) and 5 (chemical substance) that were reported by the paediatricians per diagnosis. As drugs for contraception were not reported by any paediatrician, this indication is not shown.

The highest variability in prescription habits is seen for the treatment of cough and secretory, non-bacterial otitis media: in each of these diagnoses, the use of 13 different chemical substances was reported; for the treatment of cough, these substances belong to 5 different groups: nasal preparations (R01), drugs for obstructive airway diseases (R03), cough and cold preparations (R05), systemic antihistamins (R06), and extemporaneous preparations. Similarly, according to the interviewees, 12 different substances are prescribed for the treatment of allergic rhinitis, 8 of them nasal preparations (R01), 2 antihistamins for systemic use (R06), 1 otological preparation (S03) and 1 extemporaneous preparation. For laryngitis, 11 different drugs were reported, including 8 different therapeutic groups (extemporaneous preparations and preparations without ATC included). Bacterial infections, except bacterial eye infections, are predominantly treated with systemic antibiotics; however the type of antibiotic used varies considerably. For chronic asthma treatment, reported medicines were quite consistent with only 6 different chemical substances used, which all belong to one therapeutic drug class (R03: drugs for obstructive airway diseases). Treatment habits for hyperkeratosis (emollients and skin protectives), hypertension (ACE inhibitors), and influenza (direct acting antivirals) were consistent as well. It is worth noting that for this analysis we did not differentiate diseases that require a combination of medicines according to established treatment guidelines.

Table 22 Variability in prescription habits according to interviews with paediatricians

Number of different Number of different chemical substances therapeutic classes Diagnosis prescribed prescribed (ATC L5) (ATC L2) Cough 13 5 Secretory otitis media (non-bacterial) 13 6 Allergic rhinitis 12 4 Allergic conjunctivitis 11 4 Laryngitis (Pseudo-Krupp) 11 8 Rhinitis acuta 10 4 Bacterial otitis media 9 3 Bacterial eye infections 8 3 Bacterial throat infection 8 1 Mycotic infections / skin 8 4 Pharyngitis 8 6 Acute gastroenteritis 7 3 Atopic dermatitis 7 4 Herpes simplex infect. 7 5 Impetigo 7 2 Pneumonia 7 1 Urticaria 7 3 Varicella 7 5 Vulvitis / balanitis 7 6 Acne vulgaris 6 3 Bacterial sinusitis 6 1 Chronic asthma 6 1 Urinary tract infection 6 1 Acute asthma 5 3 Borreliosis 5 1 Chronic obstipation 5 3 Dysmenorrhoea 5 3 Nausea / emesis 5 4 Obstructive bronchitis 5 3 Erysipel 4 1 Gastritis / ulcus 4 2 Acute pain 3 2 Chronic pain 3 2 Fever 3 2 Mycotic infection / mouth-pharynx 3 2 Bacterial bronchitis 2 1 Bacterial maxillodental infections 2 1 Chronic diarrhoea 2 1 Enuresis nocturna 2 2 Psoriasis 2 2 Seborrhoic dermatitis 2 2

Number of different Number of different chemical substances therapeutic classes Diagnosis prescribed prescribed (ATC L5) (ATC L2) Hyperkeratosis 1 1 Hypertension 1 1 Influenza (other than symptomatic 1 1 treatment)

4.6. List of “Essential Medicines for Children and Adolescents in Austria”

A list of “essential medicines for children and adolescent in Austria” was created based on information from the two data-bases of primary and hospital care settings and adding further medicines as well as qualitative information (indication for use) from the interviews. The list comprises 196 medicines. The full list can be found in appendix 8.4. As for this project we have performed interviews only with general primary care paediatricians and child- psychiatrists, this list focuses on the primary care setting. For better overview, the essential list of medicines was structured by therapeutic areas of primary indications; due to their wide application range, systemic antihistamins were listed separately.

5. DISCUSSION

5.1. Patterns of medicine use

In this survey we studied medicine use in children and adolescents in Austria over a one year period. Compared to the findings of similar studies in literature, we could observe similarities but also differences that are worth to be discussed.

5.1.1. Primary care setting

In concordance with surveys performed in other countries [12, 13, 20], we observed anti- infectives, medicines for the respiratory system, and pain medication to be most often prescribed to children and adolescents in the primary care setting. This also reflects the findings in interviews with primary care paediatricians to be the focus of paediatric pharmacotherapy.

Sturkenboom et al. analysed data from three different countries (United Kingdom, Netherlands, Italy). They distinguished between three patterns of medicine use: medicines with a prescription rate of more than 10% of children per year, those with 1-10% of children per year, and preparations that are prescribed with a rate of less the 1% children per year. Prescriptions of medicines with high prevalences decrease with age but remain at a high level in all age cohorts. In contrast, the use of medicines that are less frequently prescribed to the youngest only moderately increase with age.

In some aspects our study shows similar results: for the most frequently used medicines (anti-infectives, respiratory drugs), prescription frequencies decrease with age but remain at high levels in all age groups, whereas those with lower prescription frequencies in younger children increase with age (blood and blood products, cardiovascular, genito-urinary, antineoplastic/immunomodulating drugs). In contrast to Sturkenboom et al., our data show an increase of prescriptions for alimentary and dermatological medicines for adolescents. In the former group, the frequently prescribed antacids, mainly pantoprazole, may be responsible for the high numbers in Austria. Among dermatologicals, anti-acne preparations were prescribed most frequently to adolescents in Austria, whereas Sturkenboom et al. found higher frequencies for fusidic acid and topical steroids, prescribed for all age groups. One reason for this difference may be the large amounts of extemporaneous preparations with topical steroids in Austria. Hence, these numbers are masked in the drug group 93

“extemporaneous preparations” which could not be analysed for active substances in our data-base. Additionally, in the reference study, frequency data for patients 15 to 18 years of age were not available from Italy.

For medicines for the nervous system, prescription increases with age in Austria whereas prescription of these drugs decreases with age in Italy, UK and Netherlands. Certainly, these data must be interpreted with caution: Sturkenboom found paracetamol and sedative preparations (e.g. diazepam) among the most frequently used substances in this drug group. In Austria, as the prices of paracetamol and diazepam are below the prescription charge, these substances may predominantly be obtained over-the-counter and therefore show up to a lesser account in our data-set. Hence, other preparations appear to be prescribed more frequently. Apart from analgesics and sedatives, psychostimulants (methylphenidate), antiepileptics (valproic acid, lamotrigine), and anti-depressants (sertraline) are among the most frequently prescribed medicines within this substance group, increasing with age in all countries.

Regarding drugs for the respiratory tract and anti-infectives, the results from our survey are comparable with the findings by Sturkenboom et al. Drugs for obstructive airway diseases were most frequently prescribed, among these salbutamol, fluticasone, and montelukast. Furthermore, nasal preparations, cough and cold preparations, and topical anti-allergics were frequently prescribed. According to Sturkenboom and others, however, there is only little evidence for the use of these drugs [12, 21, 22].

Among anti-infective drugs, systemic anti-bacterials were most frequently used in Austria, similar to Italy, UK and Netherlands, primarily broad-spectrum penicillins and macrolides. Another study comparing systemic antibiotics use between different European countries revealed vast heterogeneity in prescription patterns: Holstiege et al. analysed prescription data to paediatric outpatients in Denmark, Italy (Emiglia Romana region only), UK, Germany, and the Netherlands from 2005 to 2008 [19]. Frequencies of prescriptions of systemic antibiotics per year varied considerably between these countries: the highest prescription frequencies were registered in Italy which were almost twice as high as in Denmark and three times higher than in the Netherlands. Being aware of the limitations of direct comparison between those data and our survey, the frequency of prescriptions in the Austrian primary care setting is comparable to Germany and UK (672 versus 561 and 555 per 1000 person years), lower than in Italy (957 per 1000 person years) but higher than in Denmark and the Netherlands (481 and 294 per 1000 person years, respectively).

Among penicillins, broad spectrum penicillins were most commonly prescribed in Austria, like in Germany, Italy, UK and the Netherlands. On substance level, amoxicillin was most common in Germany, UK and the Netherlands, whereas amoxicillin with enzyme inhibitor is more frequently prescribed in Austria and Italy. In Denmark, narrow spectrum penicillins, all above penicillin V was predominant. In Austria, macrolides are the second most commonly used group of systemic antibiotics (146 prescriptions per 1000 children) with average prescription to every 5th Austrian child who is in need of systemic antibiotics. This pattern was similar to that in Italy and Germany (199 and 113 prescriptions per 1000 children) but differed quite strongly to the other countries (e.g. Netherlands with 47 prescriptions per 1000 children) [23]. Macrolides prescription increases with age in Austria, similar to the trends described in literature: due to good tolerability and a broad anti-bacterial spectrum, the use of intermediate- and long-acting macrolides, such as clarithromycin and azithromycin has increased in the last years [23]. However, there is evidence for anti-bacterial resistance derived from macrolides use [24, 25]. Holstiege et al. reported that cephalosporins were rarely prescribed to Danish and Dutch children whereas 2nd and 3rd generation cephalosporins were quite common in Germany and Italy, especially in children below 4 years of age. In Austria, 2nd generation cephalosporins ranked third after broad spectrum penicillins and macrolides. Prescription frequencies of this group of antibiotics decreased with age in Austria, similar to Germany and Italy.

Another study, dating from 1999 compared antibiotic prescribing patterns in general practice in Ravenna (Italy) and Funen (Denmark) [26]. Similar to the studies discussed above, prescription frequency in Italy was much higher than in Denmark, and the spectrum of 90% of most frequently used antibiotics varied between 8 different substances in Denmark and 17 in Italy.

Considering that all these data were collected in European countries where treatment guidelines and health care systems are aimed to be(come) harmonised, we find these results rather remarkable. Is it still appropriate in Europe of the year 2016 that it depends on the place of residence of a patient (and his/her care-giver) which or, if any, antibiotic treatment will be administered? In order to face upcoming problems like bacterial resistance, etc. we think that providing actual evidence-based medical information to health care providers should be a primary goal of all stakeholders engaged in the health care system.

5.1.2. Hospital care setting

To our knowledge, this is the first representative collection of epidemiological data regarding medicine use in children and adolescents in Austrian hospitals. In 2008, Prandstetter et al. [17] performed a study in a single neonatal unit in Linz, Austria, collecting prescription data during a 3 month period. Their results were consistent with the literature revealing the low availability of licensed drugs in the neonatal setting. Our survey covers medicine use in children of all ages and about half (50%) of Austrian hospitals with paediatric departments. As we do not know the proportion of medicines administered to neonates specifically, we cannot directly compare our results on substance level with the findings of Prandstetter et al.

Our survey revealed the highest prescription frequencies for anti-infectives (mainly systemic antibiotics), drugs for the nervous system (analgesics, anaesthetics) and alimentary tract (various groups, mainly antacids, anti-emetics). These data are comparable with results of surveys performed in Switzerland [27], Finland [28], and USA [29, 30]. Analysis by therapeutic class (ATC level 2) confirms that anti-bacterials for systemic use, anti- inflammatory and anti-rheumatic agents, analgesics and anaesthetics comprise more than one third of all drugs used for children and adolescents in hospitals.

Interestingly, the use of antipyretics/analgesics (NSAIDs and others) was more frequent in secondary care type hospitals than in university hospitals. However, frequencies of use of these substances varied considerably between the hospitals: for anti-inflammatory and antirheumatic drugs (mainly NSAIDs) from 1 to 28%, for analgesics (opioids, other antipyretics and analgesics like paracetamol) from 2 to 18% of all substances used in the respective hospital. In total, prescription frequencies of NSAIDs (such as ibuprofen amounting 5% of total prescriptions in the hospital setting) were almost twice as frequent as antipyretics/analgesics of ATC group N02 (mainly contributed to by paracetamol). We believe this fact is rather surprising as paracetamol is licensed for all ages and has a well-known safety and efficacy profile. Administered in the correct dose it may be advantageous over or at least as effective as ibuprofen [31, 32]. Psychoanaleptics (ATC group N06) ranged among the most frequently used drug groups in hospital care setting, however, this is due to caffeine preparations, used in neonatology, in contrast to primary care.

For systemic antibiotics, broad spectrum antibiotics (ampicillin, amoxicillin with enzyme inhibitor), 2nd generation cephalosporines (cefuroxime) and glycopeptide antibiotics

(vancomycin) were most frequently used in hospital setting. Third generation cephalosporines ranged at 4th place. Overall, we observed quite some variability in antibiotic use between hospitals (range 10 to 45%), however differences between secondary versus tertiary care hospitals were small (22 versus 19%). However, by considering the type of substance on ATC level 3 (pharmacological sub-group), we observed significant differences: broad spectrum antibiotics (amoxicillin, ampicillin, with or without enzyme inhibitor, sultamicillin), 2nd generation cephalosporines (cefuroxime, cefaclor) and macrolides (clarithromycin) were more frequent in secondary care hospitals, whereas trimethoprim/sulfonamides and other antibiotics (vancomycin, chinolones, aminoglycosides, linezolid), primarily needed as reserve antibiotics, were significantly more often used in university hospitals. These findings are similar to a survey recently performed in Italy by De Luca who identified high usage rates of chinolones and carbapenems in university hospitals [33]. They also report extensive use of 3rd generation cephalosporines. Regarding the administration form of antibiotics in hospitals, we observed parenteral use in 78%. For 9% of antibiotics both enteral and parenteral forms exist, thus we could not determine the mode of administration. Our findings are similar to the 82% described in the European Surveillance of Antibiotic Consumpton (ESAC) report by Amadeo et al. [34]. The authors called as therapeutic target to decrease the proportion of parenteral antibiotic administration. Parenteral administration is associated with a risk of vascular line infections, inconvenience to patients, longer hospital stays, and higher cost. On the other hand, parenteral administration may be justified by age and/or medical condition of the patient and the fact that antibiotics may not be available in both administration forms, especially those for in-hospital use.

5.1.3. Comparison of medicine use between primary care and hospital settings

The comparison of results between between primary care and hospital settings needs to be interpreted with caution due to the different data-base structures. For example, whereas preparations without ATC code could not be analysed in any details in the primary care data- base, we had more information on active substances in the hospital data-base and could therefore add an adequate ATC code to some of the medicines (see also chapter 3.4.2). However, we assume that these methodological differences have no major impact on our overall findings.

The primary care setting showed a lower diversity of medicines used in. Treatment in primary care setting is dominated by use of anti-infectives, medicines for the respiratory system, and extemporaneous preparations whose active substance could not be identified in our analysis as they were prepared in pharmacies or are listed without ATC code in Austria Codex. Except for respiratory drugs and genito-urinary drugs, medicines of all other anatomical classes (ATC level 1) were more frequently prescribed in hospitals. In both the primary care and hospital setting, the top 10 therapeutic classes (ATC level 2) covered slightly more than half of all drugs prescribed (54% and 56% in primary and hospital care setting, respectively). Six therapeutic classes were equally among the top 10 classes in both settings: anti-bacterials for systemic use (J01), anti-inflammatory and anti-rheumatic drugs (M01), analgesics (N02), psychoanaleptics (N06), nasal preparations (R01), and drugs for obstructive airway diseases (R03). Among these classes, the following substances (ATC level 5) were most frequently used in both settings: amoxicillin with enzyme inhibitor, cefaclor, clarithromycin, ibuprofen, paracetamol, and mefenamic acid (for detailed listings refer to appendix 8.3.). For amoxicillin with and without clavulanic acid, clarithromycin, cefaclor and ibuprofen, the frequent use was confirmed by the interviews with the paediatricians in primary care setting. These drugs were reported for use in multiple indications (table 18, chapter 4.5.1).

Apart from broad spectrum penicillins which were used most frequently in both primary and hospital care settings, other systemic antibiotics showed large differences between the two settings. Whereas prevalence of 2nd generation cephalosporins, macrolides and ß-lactamase sensitive or resistant penicillins amounted to 18%, 22%, and 13% of total antibiotics use in primary care, these medicines were used to a much lower extent in hospitals (8, 5, and 4%). In contrast, “other” antibiotics, such as fluorochinolones, aminogylcosides, fusidinic acid and linezolid were used to a much higher extent in hospitals. This is certainly not surprising, however, has never been investigated for Austrian children so far [35].

Another remarkable finding is the high prevalence of pantoprazole prescriptions, both in primary care and in hospitals. Among medicines for alimentary tract and metabolism, pantoprazole was the most prevalent substance in both data-bases. We did not find similar results in any study from other countries assessing medicine use in children in the primary care setting [12, 13, 36]. Pantoprazole is recommended in adolescents aged 12 years and above for the symptomatic and long-term treatment as well as for relapse prevention of gastroesophageal reflux disease (GERD). For treatment of gastric or duodenal ulcera, pantoprazole is only indicated in adults. In our primary care data-base, pantoprazole use was predominantly prevalent in the cohort of 10-19 year old children. However, the reasons for 98 use of this substance might be an interesting subject of further investigations on medicine use in children in Austria.

5.2. Qualitative information on prescription patterns based on interviews

Qualitative information on individual prescription patterns could be performed with 10 primary care paediatricians in Vienna who agreed to participate in personal interviews. We are aware that volunteer bias, the relatively low number of interviews, and the restriction to Vienna limit the representativeness of our findings. Nevertheless, these data provide important complementary information, particularly regarding the specific indications for medicine use in various age groups. Moreover, to some extent, the data can serve for cross-validation of the results from the reimbursement database.

In our opinion, the following are the most important observations from the interview data: - Paediatricians in the “general” primary care setting, i.e. without further specialisation, routinely prescribe medicines for a panel of about 20 to 30 diseases. Beyond these, they either refer patients to specialised colleagues or clinics for consultation and treatment or continue prescriptions initiated or by specialists or hospitals. - The most frequent prescriptions occur for treatment of pain/fever, respiratory disorders, and infections (without consideration of homeopathics and other complementary medicines). - Primary care paediatricians reported to be aware of the licence status of medicines. In cases where a licensed medicine is available, it is preferably used over off-label or un-licenced alternatives. - Depending on the type of disease, prescription habits considerably vary among paediatricians (even in only 10 interviews). Prescriptions habits were most heterogeneous for unspecific symptoms like cough, non-bacterial otitis or rhinitis.

These data are comparable to the results of a poll conducted by R. Kerbl among 27 primary care paediatricians in Upper Styria in 2012 [18]. Similarly to our survey, they identified antipyretics (ibuprofen, paracetamol), medicines to treat respiratory disorders (salbutamol, montelukast) and anti-infectives (amoxicillin plus enzyme inhibitor, clarithromycin, phenoxymethylpencillin) to be most commonly used to treat children and adolescents in primary care. In total, 78 different chemical substances were reported by at least one paediatrician in this survey.

In contrast to our data, Kerbl (et al.) described that codein was frequently used to treat cough which was not mentioned by any of the paediatricians in our survey. However, 4 of the 10 doctors of our survey prescribe noscapine preparations for the treatment of cough which is not recommended for children and adolescents. This difference might be due to recent advice by EMA not to use codein in children below 12 years of age [37]. Another difference to our data can be seen in the treatment of laryngitis (Pseudo-Krupp): whereas in 2012 orally applied dexamethasone was hardly used, 7 of 10 paediatricians of our interviews prescribe this substance for laryngitis.

Interviews with child psychiatrists revealed that medicines used in schizophrenia (olanzapine, amisulpride, quetiapine, aripiprazole) and depression (mirtazapine, escitalopram, trazodone, venlafaxine) are widely used off-label. Again, the interviewees are aware of this issue and expressed their concerns. However, currently, there are no alternatives for lack of clinical studies and marketing authorizations for children. In a position statement of the British Association of Psychopharmacology (BAP), the authors call for long-term pharmaco-epidemiological, safety and efficacy data on off-label prescribing in children and adolescents [38]. Even for methylphenidate, a medication especially developed for paediatric patients, there is only retrospective information of effectiveness in long-term use [39]. However, the authors also confirm that prescribing physicians are aware of the limited evidence and need to weigh the available information independent of the licence status of these medications. This responsible use is even more important, as prescription rates for psychotropic medication in children and adolescents have increased in the last decade, among these mainly second generation antipsychotics (aripiprazole, risperidone), and serotonin reuptake inhibitors (fluoxetine, sertraline, fluvoxamine) [40]. Moreover, adverse effects of these drugs are common and especially long-term adverse effects, e.g. on cognition or fertility, have not yet been sufficiently investigated [41].

5.3. Available evidence for medicine use in children and adolescents

As extensively described in literature [7, 14, 28, 36, 42-44] and also shown in this survey, there is a high proportion of off-label use in the treatment of children and adolescents.

5.3.1. Definition of off-label use

Most of the surveys performed on medicine use in children and adolescents put much focus on the amount of off-label or unlicensed medicine use, as these variables can be seen as a 100 surrogate of unmet therapeutic needs [10]. However, definitions of off-label use vary considerably between the studies and, therefore, results of these studies cannot be objectively compared. A universally recognised definition of off-label and unlicensed use has not yet been established. For example, the Austrian Medicines Act does not define the terms “off-label” or “unlicensed”, neither do regulations on European level. On the homepage of the Austrian Federal Office for Safety in Health Care (www.basg.gv.at) we can find the following statement (only in German):

[“…Das BASG / AGES Medizinmarktaufsicht, die ÖGARI und die Patientenanwaltschaft Österreich haben sich zu nachfolgender Sprachregelung zum Thema Off-Label-Use geeinigt. Unter „Off-Label-Use“ versteht man die Anwendung eines Arzneimittels im Rahmen der medizinischen Heilbehandlung außerhalb der Informationen in der Fachinformation. Eine rechtlich verbindliche Definition ist dem österreichischen Recht, insbesondere dem Arzneimittelgesetz, nicht zu entnehmen. …“]

Hence, “off-label” use is, by definition, outside the term of the marketing authorisation and, therefore, difficult to regulate. In 2008, Neubert et al. performed a Delphi survey among experts (scientists, health care professionals, pharmaceutical companies, and regulatory agencies) to develop a definition of off-label and unlicensed use which was accepted by all stakeholders [45]. Besides off-label use (use of a marketed drug outside the indications, age subsets, strength (dose), pharmaceutical form and route of administration detailed in the SPC) and unlicensed use (use of a drug without marketing authorisation in Europe), they further distinguish paediatric off-label use by the following definition “… all paediatric uses of a marketed drug not detailed in the SPC with particular reference to therapeutic indication, therapeutic indication for use in subsets, appropriate strength (dosage by age), pharmaceutical form and route of administration …”. A drug with contraindication for children was defined as off-label, not unlicensed. However, the definition of extemporaneous formulations was not entirely resolved in this Delphi survey.

5.3.2. Off-label use in Austria

In our study, the databases lacked information on several criteria defining off-label or unlicensed use, such as indication for use (partly), dosage, pharmaceutical form, route of administration, contraindications, etc. Hence, we concentrated on off-label use based on the criteria age and indication, as available. Extemporaneous preparations were generally defined as off-label, as they increase the risk of toxicity [45].

101

Despite these methodological limitations, the observed frequencies of off-label use from our survey are comparable to the results from other publications [14, 27-30, 36, 46, 47], with average frequencies of 30-40% off-label use in the primary care setting. In Austria, off-label use decreases with age from 52% to 11% of all prescriptions in the primary care setting. Magalhaes et al. performed a systemic review investigating 34 clinical studies in regards to off-label use in hospitalised paediatric children. They found a range of off-label prescriptions from 71% (newborns) to 12% (adolescents). For our hospital analysis, as we did not have data on children´s age, we could not determine the exact frequency of off-label use per age (see also chapter 3.6.1.). However, we investigated age related licence status of medicines dispensed in hospitals. Taking this information as surrogate for off-label use, we can confirm a decrease of (assumed) off-label use by age, in average ranging from 54 to 18% in hospitals.

Among the 80% most frequently used medicines in primary care setting, the highest frequencies of off-label use occurs in systemic pain medication (dexibuprofen), antipsychotics (olanzapine, quetiapine), and antidepressants (escitalopram, trazodone). For hospital care, we identified cardiac stimulants (norepinephrine, dopamine), pain medication (ketoprofen, dexibuprofen), and some ophthalmologicals (retinol, phenylephrine) among the most frequently used medicines without specific label information for children and adolescents.

Of the most frequently used anti-infective medicines, off-label use primarily seems to be an issue in newborns. Almost one third (7/23) of anti-infectives (mainly antibiotics for systemic use) prescribed in hospitals are not licenced for children below one month of age. However, as we could not link prescriptions to patients age in our hospital data-base, i.e. could not determine to what proportion newborns were treated with these antibiotics, we can only estimate the degree of off-label use. Undoubtedly, among the most frequently used drugs in hospital is linezolid which is not recommended for children and adolescents. Against the background that linezolid targets to gram-positive bacteria that are resistant against other antibiotics, e.g. methicillin-resistant staphylococcus aureus (MRSA), its use in children and adolescents requires a thorough risk benefit evaluation.

Among drugs for respiratory diseases, only anti-asthmatics like inhalative sympathomimetics with corticosteroids (formoterol plus beclomethasone) and selective ß2-agonists (terbutaline) need to be used off-label for children. Again, babies and toddlers are more frequently treated with off-label medication, such as topical sympathomimetics, topical steroids (for nasal 102 application), or systemic antihistamines. Our observations are in line with findings described in literature [12] and with the latest inventories of paediatric therapeutic needs by the EMA [48].

5.3.3. Use of pain medication for children and adolescents

Use of anti-inflammatory / anti-rheumatic drugs and analgesics is highly common in both, primary care and hospital setting. Among these substances, we primarily found non-steroidal anti-inflammatory drugs (ibuprofen, mefenamic acid, ketoprofen, dexibuprofen, naproxen, diclofenac), paracetamol, and metamizole sodium. The age cut-offs for recommended use differ in the corresponding SPC´s, therefore the choice of the appropriate medicine needs to be made carefully. Especially in neonates and infants, overdosing is an issue, whereas older children might be at risk for underdosing [49]. Ibuprofen, licensed for children from 3 months of age, was found to be the substance most frequently used in hospital, followed by paracetamol (approved for children of all ages with a body weight of more than 3kg). In the primary care setting, data are similar; however we can assume that real “consumption” of these medicines is much higher, as many of these preparations are priced below prescription charge and, hence, were not captured in our data- base. In a retrospective cohort study performed by Neubert et al. [50], the authors compared prescription frequencies of non-steroidal anti-inflammatory drugs (NSAIDs) and opioids for the treatment of pain in children and adolescents in Italy, Netherlands, and the United Kingdom. NSAIDs, above all ibuprofen, turned out to be the most frequently used substance in these countries. Despite a relatively good safety profile, there are concerns regarding the tolerability of NSAIDs, particularly the increased risk of gastro-intestinal bleeding and impaired renal function [51]. Results of a retrospective multicentre study in Italy showed that 69% of children referred to emergency room for suspected gastrointestinal bleeding after NSAID intake had been treated with ibuprofen [52]. Further analysis revealed that 24 of these 51 children (47%) had been treated inappropriately in regards to the administered dose, duration of treatment and age. Knowing that actual ibuprofen use in children is even more widespread than in literature reports and our results, we must be wary of a rather high frequency of inappropriate use. Cardile et al. therefore recommend to better inform families and health care professionals regarding correct intake of NSAIDs [52].

103

5.4. Strengths and limitations of the survey

The present study reflects effective medicine use in children and adolescents in Austria during the year 2014 for both primary care setting and hospital settings. The survey provides insight and oversight of paediatric medicine use in Austria in general and examines specific indications, age groups, or care settings. It may also serve as the basis for future evaluations of time trends in pediatric medicine prescription patterns in Austria. Although the data were collected retrospectively, the potential for selection bias is small. Sources of potential selection bias will be discussed later in this chapter. As our data represent medicine prescriptions over a one year period, seasonal variation is not an issue.

There exists no nationwide data-base in Austria linking patient demographics, disease data, and therapies. Within the limitations of the systems, the data sources used in this survey were the best available to obtain representative and valid information on medicine use in children and adolescents in Austria. For the primary care setting, this information comprises prescriptions by paediatricians as well as general practitioners and specialists caring for children (ear-nose-throat, gynaecologists, etc). Thereby, the whole spectrum of medicines use in children and adolescents in Austria´s primary care setting is covered. However, for lack of information on individual prescribers, differences in their prescription patterns cannot be assessed. In primary care setting, restrictions are mainly owed to some missing data. For example, privately paid preparations are not contained in the insurance reimbursement database. This applies to medicines at a lower price than patient´s prescription charge and to all medicines that are generally not paid by health insurance. Therefore, data concerning over-the-counter medication, most vaccines, and oral contraceptives are not included in our data-base. Especially for oral contraceptives, careful choice of the right compound is important, as they have been poorly investigated in young patients and are frequently used off-label [10, 12]. Moreover, for extemporaneous preparations, the data base did not contain information on active ingredients. Consequently, some chemical substances might be underrepresented in the data-base, e.g. topical corticosteroids supplied as ointment mixtures. However, we know the overall frequency of extemporaneous preparations. Finally, the primary care data-base does not cover 100% of the Austrian population as around 5% of Austrians are not insured by the Main Association of Austrian Social Security Institutions (please see also chapter 3.3.1).

104

An important point of caution for the interpretation of data in the primary care setting is that prescription does not equal reimbursement which does not equal medicine use. Thus, our reimbursement data are only a proxy for medicine use in children and adolescents in Austria.

The hospital data in this survey cover about 50% of paediatric hospitals/departments in Austria and have a good range from small departments in provincial hospitals to large university hospitals. No central body for hospital administrations exists in Austria, such as HVB for the primary care setting. Therefore, we had to address every hospital pharmacy separately for delivery of medicines dispensing data. Taking into account that some hospital managements are quite restrictive regarding disclosure of information to external institutions, representativeness of the data can be judged as good. From the hospitals that provided data, we obtained fully representative data-sets for the year 2014 covering all paediatric wards in the respective institution. This allows to extrapolate results obtained from this analysis to all hospitals in Austria.

However, there are also some methodological limitations to be considered for the interpretation of data. First, we rely on the assumption that drugs which are delivered to a medical ward have been prescribed by a physician and in consequence are administered to the children admitted to this ward. Hence, we use dispensings from pharmacy as proxy for “prescription” and “‘drug use”. Second, we have no information whether a medicine dispensed to the ward is administered to a single or several patients. This issue particularly applies to medicines that were prepared extemporaneously in some hospitals but were supplied as proprietary medicinal product in others. To take a pragmatic approach to data analysis, we assumed that one pack of medication is administered to one patient. Obviously, such assumptions imply significant imprecision.

Importantly, both data-bases only contain summary data and no individual patient data. Therefore, apart from age (primary care setting only) and region, there is no information available on patients´ characteristics, such as demographics, disease information, indications for medicine use, or dosing, contra-indications and concomitant medication. In order to complement the information on medicine use in children and adolescents in Austria, we performed interviews with general paediatricians and child psychiatrists in primary practice. The interviews provided important qualitative information regarding the indications for use of about two thirds of the most frequently prescribed medicines.

In spite of the methodological limitations of our study, we think that the survey comprehensively reflects the current pharmacological paediatric treatment habits in Austria 105 and provides insight into strengths and weaknesses of established drug therapy in our country.

5.5. Essential medicines for Children and Adolescents in Austria

In our survey we assessed medicine use by children and adolescents in Austria in the year 2014. By combining data from primary and hospital care setting with the qualitative information obtained from the interviews with the paediatricians, we generated a list of essential medicines for children and adolescents in Austria (appendix 8.4). This list may be extended by further interviews with hospital paediatricians.

This list of essential medicines for children and adolescents in Austria can be the basis for the development of an (online) information platform providing structured, evidence-based information and recommendations on medicines for use in children and adolescents (www.kindermedika.at). For each of the medicines listed, relevant up-to-date information on child-appropriate formulations, dosing, efficacy, side effects, contraindications, interactions, and approval status shall be gathered and provided to practising health care professionals. Thus, they will have an information tool to be able to keep track of new developments and updated evidence in paediatric pharmacotherapy.

106

6. CONCLUSIONS

This is the first representative survey of medicine use in children and adolescents in Austria, performed over the period of one year. The results may serve as basis to investigate temporal trends in prescription patterns in the future, particularly to study the influence of measures aiming to improve the evidence base of paediatric medicine use in Austria. The survey allows to compare prescription patterns in Austria to those in other (European) countries, both on a scientific and economical level. Finally, the survey enables us to define an essential medicine list for children and adolescents in Austria, as basis for the development of an evidence-based information platform on medicines for use in children and adolescents for health care professionals.

107

7. REFERENCES

1. Wilson, J.T., An update on the therapeutic orphan. Pediatrics, 1999. 104(3 Pt 2): p. 585-90. 2. Johnson, T.N., Modelling approaches to dose estimation in children. Br J Clin Pharmacol, 2005. 59(6): p. 663-9. 3. Johnson, T.N., A. Rostami-Hodjegan, and G.T. Tucker, Prediction of the clearance of eleven drugs and associated variability in neonates, infants and children. Clin Pharmacokinet, 2006. 45(9): p. 931-56. 4. Kearns, G.L., et al., Developmental pharmacology--drug disposition, action, and therapy in infants and children. N Engl J Med, 2003. 349(12): p. 1157-67. 5. Sutherland, J.M., Fatal cardiovascular collapse of infants receiving large amounts of chloramphenicol. AMA J Dis Child, 1959. 97(6): p. 761-7. 6. Horen, B., J.L. Montastruc, and M. Lapeyre-Mestre, Adverse drug reactions and off- label drug use in paediatric outpatients. Br J Clin Pharmacol, 2002. 54(6): p. 665-70. 7. European Medicines Agency, Evidence of harm from off-label or unlicensed medicines in children. 2004. 8. Pandolfini, C. and M. Bonati, A literature review on off-label drug use in children. Eur J Pediatr, 2005. 164(9): p. 552-8. 9. European Medicines Agency, Regulation (EC) No 1901/2006 and 1902/2006 of the European Parliament and of the Council of 12 December 2006 on medicinal products for paediatric use, E.M. Agency, Editor. 2006. 10. European Medicines Agency, Survey on the survey of all paediatric uses of medicinal products in Europe. 2010. 11. European Medicines Agency, Needs for paediatric medicines http://www.ema.europa.eu/ema/index.jsp?curl=pages/regulation/document_listing/doc ument_listing_000096.jsp&mid=WC0b01ac0580925b1e. 12. Sturkenboom, M.C., et al., Drug use in children: cohort study in three European countries. BMJ, 2008. 337: p. a2245. 13. Olsson, J., et al., Paediatric drug use with focus on off-label prescriptions in Swedish outpatient care--a nationwide study. Acta Paediatr, 2011. 100(9): p. 1272-5. 14. Magalhaes, J., et al., Use of off-label and unlicenced drugs in hospitalised paediatric patients: a systematic review. Eur J Clin Pharmacol, 2015. 71(1): p. 1-13. 15. Saint-Raymond, A., et al., Usage of unpublished paediatric data. Arch Dis Child, 2016. 101(1): p. 81-4. 16. European Medicines Agency, Better medicines for children. Progress report of paediatric regulation. 2013. 17. Prandstetter, C., et al., [Medical prescriptions to premature and newborn infants in an Austrian neonatal intensive care unit]. Klin Padiatr, 2009. 221(5): p. 312-7. 18. Kerbl, R., Saalfelden 2012: Medikamentöse Therapie in der täglichen Praxis - Kindermedizin. ÖAZ, 2012. 05: p. 53-6. 19. Holstiege, J., et al., Systemic antibiotic prescribing to paediatric outpatients in 5 European countries: a population-based cohort study. BMC Pediatr, 2014. 14: p. 174.

108

20. Cazzato, T., et al., Drug prescribing in out-patient children in Southern Italy. Eur J Clin Pharmacol, 2001. 57(8): p. 611-6. 21. De Sutter, A.I., et al., Oral antihistamine-decongestant-analgesic combinations for the common cold. Cochrane Database Syst Rev, 2012(2): p. CD004976. 22. Isbister, G.K., F. Prior, and H.A. Kilham, Restricting cough and cold medicines in children. J Paediatr Child Health, 2012. 48(2): p. 91-8. 23. Holstiege, J., et al., Trends in paediatric macrolide use in five European countries-a population-based study. Eur J Clin Pharmacol, 2015. 71(8): p. 991-9. 24. Malhotra-Kumar, S., et al., Effect of azithromycin and clarithromycin therapy on pharyngeal carriage of macrolide-resistant streptococci in healthy volunteers: a randomised, double-blind, placebo-controlled study. Lancet, 2007. 369(9560): p. 482- 90. 25. Swedan, S.F., W.A. Hayajneh, and G.N. Bshara, Genotyping and serotyping of macrolide and multidrug resistant Streptococcus pneumoniae isolated from carrier children. Indian J Med Microbiol, 2016. 34(2): p. 159-65. 26. Vaccheri, A., et al., Antibiotic prescribing in general practice: striking differences between Italy (Ravenna) and Denmark (Funen). J Antimicrob Chemother, 2002. 50(6): p. 989-97. 27. Paolo, D., Unlicensed and off-label drug use in a Swiss paediatric university hospital. Swiss Med Wkly, 2006. 136: p. 218-222. 28. Lindell-Osuagwu, L., et al., Off-label and unlicensed drug prescribing in three paediatric wards in Finland and review of the international literature. J Clin Pharm Ther, 2009. 34(3): p. 277-87. 29. Hsu, B. and T. Brazelton, Off-label medication use in an academic hospital pediatric critical care unit. WMJ, 2009. 108(7): p. 343-8. 30. Lasky, T., et al., Estimating pediatric inpatient medication use in the United States. Pharmacoepidemiol Drug Saf, 2011. 20(1): p. 76-82. 31. De Martino, M. and A. Chiarugi, Recent Advances in Pediatric Use of Oral Paracetamol in Fever and Pain Management. Pain Ther, 2015. 4(2): p. 149-68. 32. Shepherd, M. and R. Aickin, Paracetamol versus ibuprofen: a randomized controlled trial of outpatient analgesia efficacy for paediatric acute limb fractures. Emerg Med Australas, 2009. 21(6): p. 484-90. 33. De Luca, M., et al., Antibiotic Prescriptions and Prophylaxis in Italian Children. Is It Time to Change? Data from the ARPEC Project. PLoS One, 2016. 11(5): p. e0154662. 34. Amadeo, B., et al., European Surveillance of Antibiotic Consumption (ESAC) point prevalence survey 2008: paediatric antimicrobial prescribing in 32 hospitals of 21 European countries. J Antimicrob Chemother, 2010. 65(10): p. 2247-52. 35. European Centre for Disease prevention and Control, Antibacterials for systemic use. 2014. 36. Palmaro, A., et al., Off-label prescribing in pediatric outpatients. Pediatrics, 2015. 135(1): p. 49-58. 37. European Medicines Agency, Codein not to be used for children below 12 years for cough and cold. 2015 (EMA/249413/2015). 38. Sharma, A.N., et al., BAP Position Statement: Off-label prescribing of psychotropic medication to children and adolescents. J Psychopharmacol, 2016. 30(5): p. 416-21.

109

39. Barbaresi, W.J., et al., Long-term stimulant medication treatment of attention- deficit/hyperactivity disorder: results from a population-based study. J Dev Behav Pediatr, 2014. 35(7): p. 448-57. 40. Persico, A.M., et al., Unmet needs in paediatric psychopharmacology: Present scenario and future perspectives. Eur Neuropsychopharmacol, 2015. 25(10): p. 1513- 31. 41. Arango, C., Present and future of developmental neuropsychopharmacology. Eur Neuropsychopharmacol, 2015. 25(5): p. 703-12. 42. Bonati, M. and C. Pandolfini, Off-label drug use in children should be rational. Arch Dis Child, 2011. 96(9): p. 870-1. 43. Choonara, I. and S. Conroy, Unlicensed and off-label drug use in children: implications for safety. Drug Saf, 2002. 25(1): p. 1-5. 44. Conroy, S., Unlicensed and off label drug use for paediatric patients. Optimal dosing schedules with gentamicin are needed for premature neonates. BMJ, 1998. 317(7152): p. 204-5. 45. Neubert, A., et al., Defining off-label and unlicensed use of medicines for children: results of a Delphi survey. Pharmacol Res, 2008. 58(5-6): p. 316-22. 46. Conroy, S., et al., Survey of unlicensed and off label drug use in paediatric wards in European countries. European Network for Drug Investigation in Children. BMJ, 2000. 320(7227): p. 79-82. 47. Lass, J., et al., Off label use of prescription medicines in children in outpatient setting in Estonia is common. Pharmacoepidemiol Drug Saf, 2011. 20(5): p. 474-81. 48. European Medicines Agency, Inventory of paediatric therapeutic needs: Respiratory. 2016. EMA/244726/2016. 49. Kazouini, A., et al., Paracetamol prescribing in primary care: too little and too much? Br J Clin Pharmacol, 2011. 72(3): p. 500-4. 50. Neubert, A., et al., The prescribing of analgesics and non-steroidal anti-inflammatory drugs in paediatric primary care in the UK, Italy and the Netherlands. Pharmacol Res, 2010. 62(3): p. 243-8. 51. Dills, R., L.A. Anderson, and C.A. Pierce, The role of nonsteroidal anti-inflammatory drugs in pediatric patients. Pharmacol Res, 2012. 65(1): p. 5-8. 52. Cardile, S., et al., Italian survey on non-steroidal anti-inflammatory drugs and gastrointestinal bleeding in children. World J Gastroenterol, 2016. 22(5): p. 1877-83.

110

8. APPENDICES

8.1. Eighty percent most frequently prescribed chemical substances (ATC level 5) in primary care, stratified by age group (n=66)

List of chemical substances is based on overall age cohort, i.e. 0-19 years. Extemporaneous preparations and preparations without ATC code are included within the 80% but are not listed below.

ATC % prescriptions per age group Code Name 0 - 19 0 - 2 3 - 9 10-19 level 5 years years years years A01AB09 Miconazole 0,4 1,8 0,1 0,1 A02BC02 Pantoprazole 0,6 0,0 0,0 1,4 A02BC05 Esomeprazole 0,3 0,1 0,1 0,5 A03FA03 Domperidone 0,3 0,4 0,5 0,2 A06AD65 Macrogole, combinations 0,4 0,4 0,6 0,2 A07CA Oral rehydration salt formulations 0,6 1,3 0,7 0,2 A07FA01 Lactic acid producing agents 0,5 0,5 0,5 0,5 A10AB05 Insulin aspart 0,6 0,0 0,2 1,1 A11CC05 Colecalciferole 0,3 0,9 0,2 0,2 B01AB05 Enoxaparin 0,6 0,0 0,0 1,2 D01AA01 Nystatin 0,3 1,6 0,1 0,0 D07AC14 Methylprednisolone aceponat 0,3 0,3 0,3 0,3 D10AD03 Adapalene 0,3 0,0 0,0 0,7 D10AF01 Clindamycin 0,4 0,0 0,0 0,9 D10AX03 Azelaic acid 0,4 0,0 0,0 0,8 D10BA01 Isotretinoin 1,3 0,0 0,0 2,8 D11AX Other dermatologicals 0,6 1,4 0,7 0,4 H02AB06 Prednisolone 0,5 1,2 0,5 0,2 J01AA02 Doxycyclin 0,3 0,0 0,0 0,6 J01AA08 Minocyclin 0,4 0,0 0,0 0,9 J01CA04 Amoxicillin 1,6 2,0 2,0 1,1 J01CE02 Phenoxymethylpenicillin 0,9 0,2 0,3 1,6 J01CE10 Phenoxymethylpenicillin-Benzathine 2,2 1,5 4,7 0,7 J01CR02 Amoxicillin and enzyme inhibitors 6,0 4,8 6,6 6,0 J01DB01 Cefalexin 0,5 0,5 0,4 0,6 J01DC04 Cefaclor 4,2 7,5 6,8 0,9 J01DD08 Cefixim 0,4 0,5 0,6 0,2 J01DD13 Cefpodoxim 0,8 0,8 0,9 0,8 J01FA07 Josamycin 0,6 0,4 0,8 0,6 J01FA09 Clarithromycin 2,3 1,0 2,3 2,9 J01FA10 Azithromycin 2,2 1,5 2,1 2,5 J01FF01 Clindamycin 0,4 0,0 0,1 0,8 J01MA02 Ciprofloxacin 0,3 0,0 0,0 0,6 L03AX Other immunostimulants 0,4 0,3 0,6 0,4 M01AB05 Diclofenac 0,4 0,0 0,0 0,8 M01AE01 Ibuprofen 0,7 0,6 1,0 0,6 M01AE02 Naproxen 0,5 0,1 0,3 0,7 111

ATC % prescriptions per age group Code Name 0 - 19 0 - 2 3 - 9 10-19 level 5 years years years years M01AE14 Dexibuprofen 0,4 0,0 0,0 0,8 M01AG01 Mefenamic acid 1,5 1,1 1,6 1,6 N02BE01 Paracetamol 1,5 2,3 1,3 1,3 N03AG01 Valproic acid 0,7 0,2 0,8 0,9 N03AX09 Lamotrigin 0,4 0,0 0,2 0,7 N03AX14 Levetiracetam 0,3 0,1 0,3 0,5 N05AX08 Risperidone 0,3 0,0 0,2 0,5 N06AB06 Sertraline 0,5 0,0 0,0 1,0 N06BA04 Methylphenidate 1,1 0,0 0,8 1,7 N06BA09 Atomoxetine 0,3 0,0 0,2 0,5 R01AA Sympathomimetics, plain 1,0 1,0 1,4 0,6 R01AA05 Oxymetazoline 0,3 0,9 0,3 0,1 R01AD09 Mometasone 0,6 0,1 0,5 0,9 R01AD12 Fluticason furoate 0,3 0,0 0,2 0,5 R02AB Antibiotics 0,3 0,0 0,1 0,5 R03AC02 Salbutamol 1,2 0,8 1,2 1,4 R03AK06 Salmeterol and fluticasone 0,5 0,0 0,3 0,8 R03BA02 Budesonide 0,4 0,6 0,5 0,3 R03BA05 Fluticasone 1,1 1,5 1,6 0,6 R03DC03 Montelukast 1,8 1,2 2,6 1,4 R05CB01 Acetylcysteine 0,3 0,0 0,2 0,5 R05CB06 Ambroxole 0,5 0,6 0,7 0,4 R05CB10 Combinations 0,3 0,3 0,6 0,1 R05DA04 Codein 0,8 0,4 1,5 0,5 R06AE07 Cetirizine 0,5 0,1 0,2 0,9 R06AE09 Levocetirizine 0,3 0,0 0,1 0,6 R06AX27 Desloratadine 1,1 0,3 1,8 0,8 S01GX02 Levocabastine 0,3 0,0 0,2 0,5 V01AA02 Grass pollen 0,3 0,0 0,3 0,5

112

8.2. Eighty percent most frequently prescribed chemical substances (ATC level 5) in hospital care setting (n=114)

ATC Code % of total Name level 5 prescriptions A01AB09 Miconazole 0,2 A01AD02 Benzydamine 0,6 A01AD11 Other stomatologicals for local oral treatment 0,3 A02BC02 Pantoprazole 1,1 A02BC05 Esomeprazole 0,3 A03AX13 Silicones 0,5 A03BA01 Atropine 0,2 A03BB01 Butylscopolamine 0,2 A04AA01 Ondansetrone 0,8 A06AD65 Macrogol, combinations 0,5 A07AA02 Nystatine 0,4 A09AA02 Multienzymes (lipases, protease etc.) 0,5 A10AB05 Insulin aspart 0,1 A11CC05 Colecalciferole 0,3 A12AA20 Calcium (different salts in combination) 0,5 A12CC03 Magnesium gluconate 0,6 B01AB01 Heparin 0,6 B01AB02 Antithrombin III 0,2 B01AB05 Enoxaparin 0,3 B01AC09 Epoprostenol 0,1 B02BA01 Phytomenadione 1,4 B02BD02 Coagulation factor VIII 0,2 B03XA01 Erythropoietin 0,3 B05AA01 Albumin 1,0 B05AA02 Other plasma protein fractions 0,2 C01CA03 Norepinephrine 0,2 C01CA04 Dopamin 0,4 C01CA24 Epinephrine 0,4 C02AC01 Clonidin 0,9 C03DA01 Spironolactone 0,2 C07AB03 Atenolol 0,5 D01AA01 Nystatin 0,4 D01AC01 Clotrimazole 0,3 D03AX03 Dexpanthenol 2,0 D04AA13 Dimetindene 0,2 D05AA Tars 0,6 D06AX Other antibiotics for topical use 0,4 D07AC14 Methylprednisolone aceponate 0,2 H02AB02 Dexamethasone 0,3

113

ATC Code % of total Name level 5 prescriptions H02AB04 Methylprednisolone 0,3 H02AB06 Prednisolone 1,8 H02AB09 Hydrocortisone 0,4 H03AA01 Levothyroxine sodium 0,2 J01AA02 Doxycycline 0,2 J01CA01 Ampicillin 3,9 J01CE01 Benzylpenicillin 0,4 J01CR01 Ampicillin and enzyme inhibitors 0,4 J01CE10 Benzathine phenoxymethylpenicillin 0,2 J01CR02 Amoxicillin and enzyme inhibitors 1,8 J01CR04 Sultamicillin 0,5 J01CR05 Piperacillin and enzyme inhibitors 0,3 J01DC02 Cefuroxim 1,9 J01DC04 Cefaclor 0,2 J01DD01 Cefotaxim 0,5 J01DD02 Ceftazidim 0,6 J01DD04 Ceftriaxone 0,2 J01DH02 Meropenem 0,9 J01EE03 Sulfametrole and trimethoprim 0,3 J01FA09 Clarithromycin 0,7 J01FF01 Clindamycin 0,7 J01GB01 Tobramycin 0,4 J01GB03 Gentamicin 1,1 J01MA02 Ciprofloxacin 0,2 J01XA01 Vancomycin 1,8 J01XD01 Metronidazole 0,6 J01XX01 Fosfomycin 0,9 J01XX08 Linezolid 0,2 J02AA01 Amphotericin B 0,3 J02AC01 Fluconazole 0,5 J05AB01 Aciclovir 0,5 J05AB06 Ganciclovir 0,2 J06BA02 Immunoglobulins, normal human, for intravasal use 0,4 L04AB02 Infliximab 0,2 L04AC02 Basiliximab 0,1 M01AB05 Diclofenac 0,7 M01AE01 Ibuprofen 5,1 M01AE03 Ketoprofen 0,3 M01AE14 Dexibuprofen 0,8 M01AG01 Mefenamic acid 1,2 M03AC09 Recuronium bromide 0,1 N01AH01 Fentanyl 0,4

114

ATC Code % of total Name level 5 prescriptions N01AX10 Propofol 0,2 N01AX14 Esketamine 0,2 N01BB02 Lidocain 0,2 N01BB20 Combinations 0,6 N02AF02 Nalbuphine 0,1 N02BB02 Metamizole sodium 0,3 N02BE01 Paracetamol 3,8 N03AA02 Phenobarbital 0,3 N03AG01 Valproic acid 0,2 N03AX14 Levetiracetam 0,3 N05BA01 Diazepam 0,4 N05BA06 Lorazepam 0,2 N05CC01 Chloral hydrate 0,5 N06BC01 Caffeine 2,5 R01AA Sympathomimetics, plain 0,2 R01AA05 Oxymetazoline 1,2 R01AA07 Xylometazoline 0,6 R01AB01 Phenylephrine 0,2 R01AX10 Other nasal preparations 0,3 R01AX30 Combinations (other nasal preparations) 0,3 R03AC02 Salbutamol 0,9 R03BA02 Budesonid 0,2 R03BA05 Fluticasone 0,2 R03CC03 Terbutaline 0,2 R06AA02 Diphenhydramine 0,2 R06AB03 Dimetindene 0,2 R06AE09 Levocetirizine 0,2 R07AA02 Natural phospholipids 0,4 S01AA11 Gentamicin 0,4 S01AX18 Povidone iodine 0,3 S01FB01 Phenylephrine 0,3 S01XA02 Retinol 1,7 S01XA20 Artificial tears and other indifferent preparations 0,4

115

8.3. Eighty percent most frequently prescribed chemical substances (ATC level 5) used in primary and hospital care each (n=150)

Medicines among top 80% in both systems are marked in red (n=30), among top 80% in primary care (age group: 0-19 years) are marked in green (n=36), among top 80% in hospital care are marked in blue (n=84). Extemporaneous preparations and preparations without ATC code are included within the 80% but are not listed below.

Primary care Hospital care overall overall ATC Code Name prescription prescription level 5 frequency frequency % %

J01CR02 Amoxicillin and enzyme inhibitor 6,0 1,8 M01AE01 Ibuprofen 0,7 5,1 N02BE01 Paracetamol 1,5 3,8 J01DC04 Cefaclor 4,2 0,2 J01CA01 Ampicillin *) 3,9 J01FA09 Clarithromycin 2,3 0,7 M01AG01 Mefenamic acid 1,5 1,2 N06BC01 Caffeine *) 2,5 J01CE10 Phenoxymethylpenicillin-Benzathine 2,2 0,2 H02AB06 Prednisolone 0,5 1,8 J01FA10 Azithromycin 2,2 *) R03AC02 Salbutamol 1,2 0,9 D03AX03 Dexpanthenol *) 2,0 J01DC02 Cefuroxim *) 1,9 J01XA01 Vancomycin *) 1,8 R03DC03 Montelukast 1,8 *) A02BC02 Pantoprazole 0,6 1,1 S01XA02 Retinol *) 1,7 J01CA04 Amoxicillin 1,6 *) R01AA05 Oxymetazoline 0,3 1,2 B02BA01 Phytomenadione *) 1,4 R03BA05 Fluticasone furoate 1,1 0,2 D10BA01 Isotretinoin 1,3 *) R01AA Sympathomimetics, plein 1,0 0,2 M01AE14 Dexibuprofen 0,4 0,8 J01FF01 Clindamycin 0,4 0,7 J01GB03 Gentamicin *) 1,1 N06BA04 Methylphenidat 1,1 *) R06AX27 Desloratadin 1,1 *) M01AB05 Diclofenac 0,4 0,7 B05AA01 Albumin *) 1,0 116

Primary care Hospital care overall overall ATC Code Name prescription prescription level 5 frequency frequency % % J01XX01 Fosfomycin *) 0,9 J01DH02 Meropenem *) 0,9 N03AG01 Valproic acid 0,7 0,2 C02AC01 Clonidine *) 0,9 J01CE02 Phenoxymethylpenicillin 0,9 *) A06AD65 Macrogol, combinations 0,4 0,5 R05DA04 Codeine 0,8 *) B01AB05 Enoxaparin 0,6 0,3 A04AA01 Ondansetron *) 0,8 J01DD13 Cefpodoxim 0,8 *) D01AA01 Nystatin 0,3 0,4 A10AB05 Insulin aspart 0,6 0,1 R03BA02 Budesonide 0,4 0,2 D11AX Other dermatologicals 0,6 *) N01BB20 Combinations *) 0,6 A02BC05 Esomeprazole 0,3 0,3 J01DD02 Ceftazidim *) 0,6 D05AA Tars *) 0,6 B01AB01 Heparin *) 0,6 R01AA07 Xylometazoline *) 0,6 R01AD09 Mometasone 0,6 *) A11CC05 Colecalciferole 0,3 0,3 J01FA07 Josamycin 0,6 *) N03AX14 Levetiracetam 0,3 0,3 A01AB09 Miconazole 0,4 0,2 A12CC03 Magnesium gluconate *) 0,6 J01XD01 Metronidazol *) 0,6 J01AA02 Doxycyclin 0,3 0,3 A01AD02 Benzydamine *) 0,6

A07CA Oral rehydration salt formulations 0,6 *) N05CC01 Chloral hydrate *) 0,5 A03AX13 Silicones *) 0,5 J01CR04 Sultamicillin *) 0,5 A12AA20 Calcium (different salts in combination) *) 0,5 R05CB06 Ambroxol 0,5 *) D07AC14 Methylprednisolone aceponat 0,3 0,2 R06AE07 Cetirizine 0,5 *) J01DB01 Cefalexin 0,5 *) A07FA01 Lactic acid producing agents 0,5 *)

117

Primary care Hospital care overall overall ATC Code Name prescription prescription level 5 frequency frequency % % N06AB06 Sertraline 0,5 *) J01MA02 Ciprofloxacin 0,3 0,2 A09AA02 Multienzymes *) 0,5 J01DD01 Cefotaxim *) 0,5 J02AC01 Fluconazole *) 0,5 R06AE09 Levocetirizine 0,3 0,2 R03AK06 Salmeterol and fluticasone 0,5 *) M01AE02 Naproxen 0,5 *) J05AB01 Aciclovir *) 0,5 C07AB03 Atenolol *) 0,5 L03AX Other immunostimulants 0,4 *) R07AA02 Natural phospholipids *) 0,4 C01CA24 Epinephrine *) 0,4 J01CR01 Ampicillin and enzyme inhibitor *) 0,4 C01CA04 Dopamine *) 0,4 S01AA11 Gentamicin *) 0,4 J01DD08 Cefixim 0,4 *) J01GB01 Tobramycin *) 0,4 D10AF01 Clindamycin 0,4 *) J01AA08 Minocyclin 0,4 *) D06AX Other antibiotics for topical use *) 0,4 A07AA02 Nystatin *) 0,4 N03AX09 Lamotrigin 0,4 *) Artifical tears and other indifferent S01XA20 *) 0,4 preparations H02AB09 Hydrocortisone *) 0,4 N05BA01 Diazepam *) 0,4 Immunglobulins, normal human, for J06BA02 *) 0,4 intravasalen adm. J01CE01 Benzylpenicillin *) 0,4 D10AX03 Azelaic acid 0,4 *) N01AH01 Fentanyl *) 0,4 D10AD03 Adapalene 0,3 *) A03FA03 Domperidone 0,3 *) D01AC01 Clotrimazol *) 0,3 R01AX10 Other nasal preparations *) 0,3 S01GX02 Levocabastine 0,3 *) H02AB02 Dexamethasone *) 0,3 N03AA02 Phenobarbital *) 0,3 R01AX30 Combinations (Other nasal *) 0,3

118

Primary care Hospital care overall overall ATC Code Name prescription prescription level 5 frequency frequency % % preparations) R05CB10 Combinations of mucolytics 0,3 *) V01AA02 Grass pollen 0,3 *) R05CB01 Acetylcysteine 0,3 *) H02AB04 Methylprednisolone *) 0,3 N06BA09 Atomoxetine 0,3 *) J01EE03 Sulfametrole and trimethoprim *) 0,3 R01AD12 Fluticasone furoate 0,3 *) S01FB01 Phenylephrine *) 0,3 R02AB Antibiotics (throat preparations) 0,3 *) M01AE03 Ketoprofen *) 0,3 N02BB02 Metamizole sodium *) 0,3 B03XA01 Erythropoietin *) 0,3 N05AX08 Risperidone 0,3 *) J01CR05 Piperacillin and enzyme inhibitor *) 0,3 S01AX18 Povidon-Iodine *) 0,3 J02AA01 Amphotericin B *) 0,3

A01AD11 Other agents for local oral treatment *) 0,3 N05BA06 Lorazepam *) 0,2 J01XX08 Linezolid *) 0,2 A03BB01 Butylscopolamine *) 0,2 H03AA01 Levothyroxine sodium *) 0,2 L04AB02 Infliximab *) 0,2 N01BB02 Lidocain *) 0,2 B02BD02 Coagulation factor VIII *) 0,2 N01AX14 Esketamin *) 0,2 B05AA02 Other plasma protein fractions *) 0,2 A03BA01 Atropine *) 0,2 C03DA01 Spironolactone *) 0,2 D04AA13 Dimetindene *) 0,2 J01DD04 Ceftriaxon *) 0,2 J05AB06 Ganciclovir *) 0,2 M03AC09 Recuronium bromide *) 0,2 N01AX10 Propofol *) 0,2 R01AB01 Phenylephrine *) 0,2 R03CC03 Terbutaline *) 0,2 C01CA03 Norepinephrine *) 0,2 B01AB02 Antithrombin III *) 0,2 R06AB03 Dimetindene *) 0,2

119

Primary care Hospital care overall overall ATC Code Name prescription prescription level 5 frequency frequency % % R06AA02 Diphenhydramine *) 0,2 B01AC09 Epoprostenol *) 0,1 L04AC02 Basiliximab *) 0,1 N02AF02 Nalbuphine *) 0,1

*) not among 80% of most frequently prescribed substances in the corresponding care setting (i.e. prescribed very rarely or not at all in this setting)

120

8.4. Essential list of medicines including qualitative information from interviews with selected paediatricians (n=196)

This list combines semi-quantitative information from primary and hospital care setting with qualitative information gathered from interviews with selected paediatricians. It therefore describes the most essential medicines prescribed to children and adolescents in Austria and may be basis for the development of an evidence-based information platform on paediatric medicines (focused on primary care).

ATC Code Therapeutic field Name level 5

Gastro-intestinal A01AB09 Miconazole disorders A02BC01 Omeprazole*) A02BC02 Pantoprazole A02BC05 Esomeprazole A02BX02 Sucralfat*) A03AX13 Silicones A03BA01 Atropine A03BB01 Butylscopolamine A03FA03 Domperidone A04AA01 Ondansetron A06AD11 Lactulose*) A06AD65 Macrogol, combinations A06AG11 Laurylsulfat, incl. combinations*) A07AA02 Nystatin A07AA07 Amphotericin B*)

A07CA Oral rehydration salt formulations A07FA01 Lactic acid producing agents A07FA02 Saccharomyces boulardii*) A09AA02 Multienzymes Various alimentary tract / metabolism A16AX products*) Nutrition and blood A11CC05 Colecalciferole A12AA20 Calcium (different salts in combination)

A12CC03 Magnesium gluconate

B01AB01 Heparin B01AB02 Antithrombin III B01AB05 Enoxaparin B01AC09 Epoprostenol B02BA01 Phytomenadione B02BD02 Coagulation factor VIII

B03XA01 Erythropoietin

121

ATC Code Therapeutic field Name level 5

Nutrition and blood B05AA01 Albumin B05AA02 Other plasma protein fractions Cardiovascular disorders C01CA03 Norepinephrine C01CA04 Dopamine C01CA24 Epinephrine C02AC01 Clonidine C03DA01 Spironolactone C07AB03 Atenolol Skin diseases D01AA01 Nystatin D01AC01 Clotrimazol D01AC02 Miconazole*) D01AC05 Isoconazole*) Imidazoles/triazoles in comb. w. D01AC20 corticosteroids*) D02AE51 Carbamide, combinations*) D03AX03 Dexpanthenol D04AA13 Dimetindene*) D04AX Other antipruritics*) D05AA Tars D06AX01 Fusidic acid*) D06AX07 Gentamicin*) D06AX Other antibiotics for topical use D06BB03 Aciclovir*) D06BB Antivirals*) D07AC13 Mometasone*) D07AC14 Methylprednisolone aceponat D10AD03 Adapalene D10AE01 Benzoyl peroxide*) D10AF01 Clindamycin D10AX01 Fusidic acid*) D10AX03 Azelaic acid D10BA01 Isotretinoin D11AH01 Tacrolimus*) D11AH02 Pimecrolimus*) D11AX Other dermatologicals Uro-genital disorders G04BD04 Oxybutynin*) H01BA02 Desmopressin*) Endocrine system A10AB05 Insulin aspart H02AB02 Dexamethasone

H02AB04 Methylprednisolone

H02AB06 Prednisolone

122

ATC Code Therapeutic field Name level 5

Endocrine system H02AB09 Hydrocortisone H03AA01 Levothyroxine sodium Infections A01AD11 Other agents for local oral treatment J01AA02 Doxycyclin

J01AA08 Minocyclin

J01CA01 Ampicillin J01CA04 Amoxicillin J01CE01 Benzylpenicillin J01CE02 Phenoxymethylpenicillin J01CE10 Phenoxymethylpenicillin-Benzathine J01CR01 Ampicillin and enzyme inhibitor

J01CR02 Amoxicillin and enzyme inhibitor

J01CR04 Sultamicillin J01CR05 Piperacillin and enzyme inhibitor J01DB01 Cefalexin J01DC02 Cefuroxim J01DC04 Cefaclor J01DD01 Cefotaxim

J01DD02 Ceftazidim

J01DD04 Ceftriaxon J01DD08 Cefixim J01DD13 Cefpodoxim J01DH02 Meropenem J01EE03 Sulfametrole and trimethoprim J01FA07 Josamycin

J01FA09 Clarithromycin

J01FA10 Azithromycin J01FF01 Clindamycin J01GB01 Tobramycin J01GB03 Gentamicin J01MA02 Ciprofloxacin J01XA01 Vancomycin

J01XD01 Metronidazol

J01XX01 Fosfomycin J01XX08 Linezolid J02AA01 Amphotericin B J02AC01 Fluconazole J05AB01 Aciclovir J05AB06 Ganciclovir

J05AH02 Oseltamivir*)

J06BA02 Immunglobulins, normal human

123

ATC Code Therapeutic field Name level 5

Infections L03AX Other immunostimulants Immunosuppressive L04AC02 Basiliximab disorders L04AB02 Infliximab Musculo-skeletal M01AB05 Diclofenac diseases M01AE01 Ibuprofen M01AE02 Naproxen M01AE03 Ketoprofen M01AE14 Dexibuprofen M01AG01 Mefenamic acid Anaesthesia M03AC09 Recuronium bromide N01AH01 Fentanyl N01AX10 Propofol N01AX14 Esketamin N01BB02 Lidocain N01BB20 Combinations Nervous system N02AE01 Buprenorphine disorders N02AF02 Nalbuphine N02BB02 Metamizole sodium N02BE01 Paracetamol N03AA02 Phenobarbital N03AG01 Valproic acid N03AX09 Lamotrigin N03AX14 Levetiracetam N05AX08 Risperidone N05AH03 Olanzapine*) N05AH04 Quetiapine*) N05BA01 Diazepam N05BA06 Lorazepam N05CC01 Chloral hydrate N06AB03 Fluoxetine*) N06AB06 Sertraline N06BA04 Methylphenidate N06BA09 Atomoxetine*) N06BC01 Caffeine Ear-nose-oropharynx A01AD02 Benzydamine disorders R01AA Sympathomimetics, plain R01AA05 Oxymetazoline R01AA07 Xylometazoline

R01AA08 Naphazoline*)

R01AB01 Phenylephrine R01AB05 Ephedrine*) 124

ATC Code Therapeutic field Name level 5

Ear-nose-oropharynx R01AC01 Cromoglicic acid*) disorders R01AC03 Azelastine*) R01AD08 Fluticasone R01AD09 Mometasone R01AD12 Fluticasone furoate R01AX10 Other nasal preparations R01AX30 Combinations (Other nasal preparations) R02AA02 Dequalinium R02AA20 Various antiseptics*) R02AB Antibiotics Respiratory diseases R03AC02 Salbutamol R03AC12 Salmeterol*) R05CA12 Hederae helicis folium*) R05CA Expectorants*) R05CB01 Acetylcysteine R05CB06 Ambroxol R05CB10 Combinations R05DA04 Codeine R05X Other cold preparations*) R07AA02 Natural phospholipids R03AK06 Salmeterol and fluticasone R03BA02 Budesonide R03BA05 Fluticasone furoate R03CC03 Terbutaline R03DC03 Montelukast Various disorders (anti- R06AA02 Diphenhydramine histamins) R06AB03 Dimetindene R06AE07 Cetirizine R06AE09 Levocetirizine R06AX17 Ketotifen*) R06AX27 Desloratadin Eye disorders S01AA11 Gentamicin S01AA13 Fusidic acid

S01AA26 Azithromycin

S01AE01 Ofloxacin S01AX18 Povidon-Iodine S01CA01 Dexamethasone and anti-infectives S01GA05 Phenylephrine S01GA51 Naphazoline, combinations S01GX01 Cromoglicic acid

S01GX02 Levocabastine

125

ATC Code Therapeutic field Name level 5

Eye disorders S01GX08 Ketotifen S01XA02 Retinol S01XA20 Artifical tears and other indifferent preparations *) based on qualitative information from interviews with primary care paediatricians

126

8.5. Licence status of eighty percent of most frequently prescribed medicines in primary care

Medicines listed below are among top 80% in at least one age group. Licence status is defined as follows: *) 0= on-label, 1= off-label, 2=medicine not used in this age group. Licence status per age group*) ATC Code Name 0 - 2 3 - 4 5 - 6 7 - 9 10 - 19 level 5 years years years years years A01AB09 Miconazole 0 0 0 0 0 A02BC02 Pantoprazole 1 1 1 1 0 A02BC05 Esomeprazole 0 0 0 0 0 A02BX02 Sucralfate 1 1 1 1 0 A03FA01 Metoclopramide 0 0 0 0 0 A06AD65 Macrogole, combinations 1 0 0 0 0 A07CA Oral rehydration salt formulations 0 0 0 0 0 A07EC02 Mesalazine 1 1 1 1 1 A07FA01 Lactic acid producing agents 0 0 0 0 0 A10AB05 Insulin aspart 1 0 0 0 0 A10AE04 Insulin glargine 1 0 0 0 0 A11CC05 Colecalciferole 0 0 0 0 0 Vitamin B1 in combination with 1 1 1 1 1 A11DB vitamin-B6 and/or vitamin B12 B01AB04 Dalteparin 1 1 2 1 1 B01AB05 Enoxaparin 1 1 1 1 1 B02BD02 Coagulation factor VIII 0 0 0 0 0 D01AA01 Nystatin 0 0 0 0 0 D01AC08 Ketoconazole 1 1 1 1 0 D07AC14 Methylprednisolone aceponat 0 0 0 0 0 D07CC01 Betamethasone and antibiotics 0 0 0 0 0 D10AD03 Adapalene 1 1 1 1 0 D10AD04 Isotretinoin 1 1 1 1 0 D10AD54 Isotretinoin,combinations 2 1 1 1 0 D10AE01 Benzoyl peroxide 2 1 1 1 0 D10AE51 Benzoyl peroxide, combinations 2 1 1 1 0 D10AF01 Clindamycin 1 1 1 1 0 D10AX03 Azelaic acid 1 1 1 1 0 D10BA01 Isotretinoin 1 1 1 1 0 D11AH01 Tacrolimus 1 0 0 0 0 D11AX Other dermatologicals 0 0 0 0 0 G01AA10 Clindamycin 1 1 1 1 0 G03DB01 Dydrogesterone 2 1 1 2 1 G04BD09 Trospium 1 1 1 1 0 H01AC01 Somatropin 0 0 0 0 0 H01BA02 Desmopressin 0 0 0 0 0 H02AB06 Prednisolone 1 1 0 0 0 H03AA01 Levothyroxine sodium 0 0 0 0 0 J01AA02 Doxycyclin 1 1 1 0 0 J01AA08 Minocyclin 2 0 0 0 0 J01CA04 Amoxicillin 0 0 0 0 0

127

Licence status per age group*) ATC Code Name 0 - 2 3 - 4 5 - 6 7 - 9 10 - 19 level 5 years years years years years J01CA08 Pivmecillinam 0 0 0 0 0 J01CE02 Phenoxymethylpenicillin 0 0 0 0 0 Phenoxymethylpenicillin- 0 0 0 0 0 J01CE10 Benzathine J01CR02 Amoxicillin and enzyme inhibitors 0 0 0 0 0 J01CR04 Sultamicillin 2 1 1 1 0 J01DB01 Cefalexin 0 0 0 0 0 J01DC02 Cefuroxim 0 0 0 0 0 J01DC04 Cefaclor 0 0 0 0 0 J01DD08 Cefixim 0 0 0 0 0 J01DD13 Cefpodoxim 0 0 0 0 0 J01FA06 Roxithromycin 0 0 0 0 0 J01FA07 Josamycin 0 0 0 0 0 J01FA09 Clarithromycin 0 0 0 0 0 J01FA10 Azithromycin 0 0 0 0 0 J01FF01 Clindamycin 0 0 0 0 0 J01MA02 Ciprofloxacin 0 0 0 0 0 J01XX01 Fosfomycin 0 0 0 0 0 J02AC01 Fluconazole 0 0 0 0 0 J05AB11 Valaciclovir 1 1 1 1 0 L03AX Other immunostimulants 0 0 0 0 0 L04AX01 Azathioprine 0 0 0 0 0 M01AB05 Diclofenac 1 1 0 0 0 M01AE01 Ibuprofen 0 0 0 0 0 M01AE02 Naproxen 0 0 0 0 0 M01AE14 Dexibuprofen 1 1 1 1 1 M01AG01 Mefenamic acid 0 0 0 0 0 M02AA15 Diclofenac 1 1 1 1 0 M03BX02 Tizanidine 1 1 1 1 1 N02BB02 Metamizole sodium 0 0 0 0 0 N02BE01 Paracetamol 0 0 0 0 0 N03AG01 Valproic acid 0 0 0 0 0 N03AX03 Sultiame 0 0 0 0 0 N03AX09 Lamotrigin 1 0 0 0 0 N03AX14 Levetiracetam 0 0 0 0 0 N05AH03 Olanzapine 1 1 1 1 1 N05AH04 Quetiapine 2 1 1 1 1 N05AX08 Risperidone 1 1 0 0 0 N05AX12 Aripiprazole 1 1 1 0 0 N06AB03 Fluoxetine 1 1 1 0 0 N06AB06 Sertraline 1 1 0 0 0 N06AB10 Escitalopram 1 1 1 1 1 N06AX05 Trazodone 2 1 1 1 1 N06BA04 Methylphenidate 1 1 0 0 0 N06BA09 Atomoxetine 2 1 0 0 0 R01AA Sympathomimetics, plain 0 0 0 0 0 R01AC02 Levocabastine 1 1 1 0 0 R01AD09 Mometasone 1 0 0 0 0 R01AD12 Fluticason furoate 1 1 0 0 0 R01AD58 Fluticasone, combinations 1 1 1 1 0 128

Licence status per age group*) ATC Code Name 0 - 2 3 - 4 5 - 6 7 - 9 10 - 19 level 5 years years years years years R02AA02 Dequalinium 1 0 0 0 0 R02AB Antibiotics 1 0 0 0 0 R03AC02 Salbutamol 0 0 0 0 0 R03AK06 Salmeterol and fluticasone 1 0 0 0 0 R03AK07 Formoterol and budesonide 1 1 0 0 0 R03AK08 Formoterol und Beclometason 1 1 1 1 1 R03AL01 Fenoterol and ipratropium bromide 0 0 0 0 0 R03BA02 Budesonide 1 1 0 0 0 R03BA05 Fluticasone 0 0 0 0 0 R03DC03 Montelukast 0 0 0 0 0 R05CB01 Acetylcysteine 1 0 0 0 0 R05CB06 Ambroxole 0 0 0 0 0 R05CB10 Combinations 1 0 0 0 0 R05DA04 Codein 1 1 1 1 0 R06AE07 Cetirizine 1 0 0 0 0 R06AE09 Levocetirizine 1 1 0 0 0 R06AX13 Loratadine 1 0 0 0 0 R06AX27 Desloratadine 0 0 0 0 0 S01GX02 Levocabastine 1 1 1 0 0 S01GX07 Azelastine 1 0 0 0 0 S01GX08 Ketotifen 1 0 0 0 0 V01AA02 Grass pollen 1 0 0 0 0

129

8.6. Licence status of eighty percent of most frequently prescribed medicines in hospital care

Medicines listed below are among top 80% in at least one age group. Licence status is defined as follows: *) 0= on-label, 1= off-label <1 month of age, 2= off-label <2 years of age, 3= off-label <19 years of age

ATC Code Licence Name level 5 status*) A01AB09 Miconazole 1 A01AD02 Benzydamine 1 A01AD11 Other stomatologicals for local oral treatment 0 A02BC02 Pantoprazole 2 A02BC05 Esomeprazole 2 A03AX13 Silicones 0 A03BA01 Atropine 0 A03BB01 Butylscopolamine 2 A04AA01 Ondansetrone 1 A06AD65 Macrogol, combinations 2 A07AA02 Nystatine 0 A09AA02 Multienzymes (lipases, protease etc.) 0 A10AB05 Insulin aspart 0 A11CC05 Colecalciferole 0 A12AA20 Calcium (different salts in combination) 0 A12CC03 Magnesium gluconate 3 B01AB01 Heparin 1 B01AB02 Antithrombin III 0 B01AB05 Enoxaparin 3 B01AC09 Epoprostenol 3 B02BA01 Phytomenadione 0 B02BD02 Coagulation factor VIII 0 B03XA01 Erythropoietin 2 B05AA01 Albumin 0 B05AA02 Other plasma protein fractions 3 C01CA03 Norepinephrine 3 C01CA04 Dopamin 3 C01CA24 Epinephrine 0 C02AC01 Clonidin 3 C03DA01 Spironolactone 3 C07AB03 Atenolol 3 D01AA01 Nystatin 0

130

ATC Code Licence Name level 5 status*) D01AC01 Clotrimazole 1 D03AX03 Dexpanthenol 0 D04AA13 Dimetindene 0 D05AA Tars 3 D06AX Other antibiotics for topical use 0 D07AC14 Methylprednisolone aceponate 1 H02AB02 Dexamethasone 0 H02AB04 Methylprednisolone 0 H02AB06 Prednisolone 2 H02AB09 Hydrocortisone 0 H03AA01 Levothyroxine sodium 0 J01AA02 Doxycycline 2 J01CA01 Ampicillin 1 J01CE01 Benzylpenicillin 0 J01CR01 Ampicillin and enzyme inhibitors 0 J01CE10 Benzathine phenoxymethylpenicillin 0 J01CR02 Amoxicillin and enzyme inhibitors 0 J01CR04 Sultamicillin 1 J01CR05 Piperacillin and enzyme inhibitors 2 J01DC02 Cefuroxim 0 J01DC04 Cefaclor 0 J01DD01 Cefotaxim 0 J01DD02 Ceftazidim 0 J01DD04 Ceftriaxone 0 J01DH02 Meropenem 1 J01EE03 Sulfametrole and trimethoprim 1 J01FA09 Clarithromycin 0 J01FF01 Clindamycin 1 J01GB01 Tobramycin 0 J01GB03 Gentamicin 0 J01MA02 Ciprofloxacin 0 J01XA01 Vancomycin 1 J01XD01 Metronidazole 0 J01XX01 Fosfomycin 0 J01XX08 Linezolid 3 J02AA01 Amphotericin B 1 J02AC01 Fluconazole 0 J05AB01 Aciclovir 0 J05AB06 Ganciclovir 2 J06BA02 Immunoglobulins, normal human, for intravasal use 0 L04AB02 Infliximab 2 L04AC02 Basiliximab 1

131

ATC Code Licence Name level 5 status*) M01AB05 Diclofenac 2 M01AE01 Ibuprofen 1 M01AE03 Ketoprofen 3 M01AE14 Dexibuprofen 3 M01AG01 Mefenamic acid 1 M03AC09 Recuronium bromide 0 N01AH01 Fentanyl 2 N01AX10 Propofol 1 N01AX14 Esketamine 0 N01BB02 Lidocain 1 N01BB20 Combinations 0 N02AF02 Nalbuphine 0 N02BB02 Metamizole sodium 1 N02BE01 Paracetamol 0 N03AA02 Phenobarbital 3 N03AG01 Valproic acid 0 N03AX14 Levetiracetam 2 N05BA01 Diazepam 2 N05BA06 Lorazepam 2 N05CC01 Chloral hydrate 3 N06BC01 Caffeine 0 R01AA Sympathomimetics, plain 0 R01AA05 Oxymetazoline 0 R01AA07 Xylometazoline 2 R01AB01 Phenylephrine 2 R01AX10 Other nasal preparations 2 R01AX30 Combinations (other nasal preparations) 2 R03AC02 Salbutamole 1 R03BA02 Budesonid 0 R03BA05 Fluticasone 0 R03CC03 Terbutaline 3 R06AA02 Diphenhydramine 1 R06AB03 Dimetindene 1 R06AE09 Levocetirizine 2 R07AA02 Natural phospholipids 0 S01AA11 Gentamicin 0 S01AX18 Povidone iodine 3 S01FB01 Phenylephrine 3 S01XA02 Retinol 3 S01XA20 Artificial tears and other indifferent preparations 3

132

CURRICULUM VITAE

Mag. Elisabeth Rauch

 Date of birth: February 08, 1972  Citizenship: Austrian

Professional Experience / Education:

Since May 2014 Freelancer for the project “kindermedika.at” at the Medical University Vienna, Dept. of Paediatrics and Adolescent Medicine, as part of the PhD program “Applied Medical Sciences” at the Medical University of Vienna, part-time

Since Oct 2014 Clinical Project Manager “AGMT Arbeitsgemeinschaft Medikamentöse Tumortherapie gemeinnützige GmbH“ (project management / investigator initiated studies in oncology), part-time

2008 – 2014 Clinical Trial Manager “Österreichisches Forum gegen Krebs” (project management, data analysis / investigator initiated clinical studies in oncology)

2006 - 2008 Freelancer for the pharmaceutical industry providing services in clinical research (project management), medical information, content management/marketing

2004 – 2005 Hospital pharmacist - General hospital of Vienna

2002 Freelance Project Manager Clinical Research in Munich, Germany (Preparation, Performance and Reporting of Clinical Trials, Medical Writing)

1999 – 2001 Medical Monitor Essex Pharma München GmbH, Munich, Germany

1997 - 1998 Clinical Research Assistant Synthelabo Arzneimittel GmbH, Munich, Germany

1996 - 1997 Pharmacy internship in Vienna (Completion „summa cum laude“).

1990 - 1996 Pharmacy studies University of Vienna, Austria

133

1982 - 1990 Secondary School – Neusprachliches Gymnasium der Kreuzschwestern in Linz, Austria

Professional Education: - PhD studies at Medical University “Applied Medical Sciences” (since March 2014) - Post graduate diploma “Clinical Research” at the Medical University of Vienna (MUW) and Vienna School of Clinical Research (VSCR) (2006-2009) - Several courses Clinical Research (GCP, ICH-Guidelines), Medical Writing and Clinical Pharmacy

Publications: Bendamustine-Bortezomib-Dexamethasone is an active and well tolerated regimen in patients with relapsed or refractory multiple myeloma. Ludwig H, Kasparu H, Leitgeb C, Rauch E, Linkesch W, Zojer N, Greil R, Seebacher A, Pour L, Weißmann A, Adam Z. Blood. 2014 Feb 13;123(7):985-91

Lenalidomide and dexamethasone for acute light chain-induced renal failure: a phase II study. Ludwig H, Rauch E, Kuehr T, Adam Z, Weissmann A, Kasparu H, Autzinger EM, Heintel D, Greil R, Poenisch W, Müldür E, Zojer N. Haematologica 2014 Nov 14; pii: haematol.2014.115204

Languages: English, French (both fluently)

134