Antibiotic prescriptions in children

Journal of Antimicrobial Chemotherapy (2003) 52, 282 286 DOI: 10.1093/jac/dkg302 Advance Access publication 15 July 2003 Antibiotic prescriptions in children D. Resi*, M. Milandri, M. L. Moro and the Emilia Romagna Study Group on the Use of Antibiotics in Children Agenzia Sanitaria Regionale Emilia Romagna, Area Rischio Infettivo, Viale Aldo Moro 21, 40127 Bologna, Italy Received 20 December 2002; returned 27 February 2003; revised 17 April 2003; accepted 25 April 2003 Objectives: The aim of this study was to evaluate antibiotic prescription for children in Emilia Romagna, a Northern Italian region with 414 880 people aged 1 14 years. Methods: The regional Prescription Database of drugs reimbursed by the Italian National Health Service was used in this study. Antibiotic use was estimated as the proportion of children who received at least one prescription during the year 2000 (number of children treated per 100 inhabitants per year). To evaluate the frequency of exposure for each child, all the prescriptions given within a period shorter than 12 days were considered as a single treatment. Results: In the year surveyed, 511270 antibiotic prescriptions in 219 257 children were identified. In all, 52.9% of children received at least one antibiotic; this percentage decreased with age, ranging from 70.4% in children 1 2 years old to 35.8% in children >11 years old. Fifty-two per cent of inhabitants under the age of 15 years were treated with systemic antibiotics at least once in the year. Cephalosporins were mostly prescribed in the youngest children, while macrolides were most frequently used in children over 6 years old. In all 3.9% of children were treated with topical antibiotics. Conclusions: This study has shown that paediatric antibiotic prescription rates can be derived from analysis of regional drug and resident databases. High antibiotic usage is shown in the paediatric population of Emilia Romagna, similar to that observed in other regions of Northern Italy. Broad-spectrum antibiotics are predominantly prescribed. Comparison with prescription rates from other countries paediatric populations suggests there is extensive antibiotic overuse in Italy. This could be associated with selection for and dissemination of antibiotic resistance. Interventions are needed to reduce consumption. Keywords: pharmacoepidemiology, paediatrician s prescriptions, drug utilization Introduction In the USA almost three-quarters of all outpatient antibiotics are prescribed for acute respiratory infections. 1 A significant proportion of antibiotics prescribed every year in the community are given to children and 50% of prescriptions for children given by general practitioners are unnecessary. In the USA 44% of children with common colds were reported to be treated with antibiotics, 46% of those with upper respiratory infections, and 75% of those with bronchitis, conditions that typically do not benefit from antibiotic treatment. 2 In Canada, 74% of preschool children seeking care for respiratory infections received antibiotic prescriptions; in 85% of these cases such prescriptions were inappropriate. 3 Antimicrobial resistance among Streptococcus pneumoniae is rapidly increasing in several countries, 4 6 and the single most important factor in the emergence of antibiotic resistance among respiratory bacterial pathogens is selection pressure from antimicrobial agents. 7 Observational 8 and intervention 9 studies from several countries have supported a link between rates of antibiotic prescription and resistance in communities. Thus, reduction of inappropriate use of antibiotics in the community should be considered a major public health issue. Existing reports of population-based antibiotic use in children are relatively few. This paper describes the pattern of individual prescriptions of systemic and topical antibiotics used for eye infections in children and adolescents <15 years old in Emilia Romagna (Northern Italy) in the year 2000, reported to the Regional Health Service data system.... *Corresponding author. Tel: +39-051-6397134; Fax: +39-051-6397053; E-mail: dresi@asr.regione.emilia-romagna.it Members of the Emilia Romagna Study Group on the Use of Antibiotics in Children are listed in the Acknowledgements.... 282 Published by Oxford University Press

Antibiotic prescriptions in children Table 1. Paediatric population, treated children and number of treated children per 100 inhabitants by age groups and by sex Paediatric population Treated children Treated children/ 100 inhabitants/year F M total F M total F M total 1 2 31.029 33.168 64.197 21.511 23.710 45.221 69.3 71.5 70.4 3 6 60.219 63.893 124.112 38.445 41.570 80.015 63.8 65.1 64.5 7 10 60.256 63.814 124.070 27.886 29.431 57.317 46.3 46.1 46.2 11 14 49.785 52.716 102.501 17.473 19.231 36.704 35.1 36.5 35.8 Total 201.289 213.591 414.880 105.315 113.942 219.257 52.3 53.3 52.9 F, female; M, male. Materials and methods Setting Emilia Romagna is one of the 20 Italian regions and it is located in the north of the country. On 1 January 2000, the population of Emilia Romagna was nearly 4 million inhabitants, including 414 880 children 1 14 years old, accounting for 5.3% of the entire Italian paediatric population in this age group. Children <1 year of age were not included in this analysis because length of exposure to antibiotics in this age class varied according to actual date of birth, while in all other age classes we could observe exposure for a whole year. Inclusion of children <1 year of age would have required that analysis be carried out by person-years of exposure. The Italian health care system is organized uniformly; people are registered by their individual practitioner and receive free medical care. The National Health Service ensures free access to hospitals and reimburses 100% of the cost of oral antibiotics prescribed, independently of social status and income, and 100% of injectable antibiotics when considered appropriate. Pharmacies send prescription data to the regional database through a computerized accounting system. Analysis To identify antibiotic prescriptions, the regional database of drug prescriptions was used; the file includes all the prescriptions for drugs reimbursed by the Italian National Health Service. The file was linked to the regional database of the resident population to identify prescriptions of antibiotics provided to children 1 14 years old in Emilia Romagna, from 1 January to 31 December 2000. Information contained in the files and considered for the analysis were patients age and gender, dispensed product according to the Anatomical Therapeutic Chemical classification and date of prescription. All prescriptions for systemic antibiotics (J01) and topical antibiotics used in eye infections (S01AA, S01AX, S01CA) were identified. Childrens ages were calculated at a middle date of the year, 1 July 2000. The amount of antibiotic use was estimated using the following parameters: (i) the proportion of children who received at least one prescription during 2000 (number of treated children per 100 inhabitants per year); (ii) the proportion of children in the population who underwent 0, 1, 2 or 3 antibiotic treatments during 2000; all the prescriptions given to a single child within a time period shorter than or equal to 11 days (number of days from the first prescription to the following one) were considered as one treatment. 10 Distribution of prescriptions by antibiotic therapeutic category and by antibiotic molecule were analysed. For each category and type of antibiotic, the proportion of children who used at least once the specific antibiotic among children who received at least one prescription was calculated. Antimycotics were not included in this study. Data were analysed using SAS software. 11 Results In the year surveyed, 511 270 antibiotic prescriptions in 219 257 children were identified. In all 52.9% of children received at least one antibiotic prescription; this percentage decreased with age, from 70.4% in children 1 2 years old to 35.8% in children >11 years old (Table 1). The frequency of systemic and topical antibiotics prescriptions did not vary by gender. Systemic antibiotics The database included 491 906 prescriptions of systemic antibiotics in 215 696 children; the mean prescription rate was 119 prescriptions/ 100 children/year. Fifty-two per cent of inhabitants <15 years old were treated with systemic antibiotic at least once in the year. Table 2 shows that antibiotic use is greater in children <6 years old; two-thirds of children in this age group received one or more prescriptions. A high proportion of children received more than one antibiotic course during the study period. Figure 1 shows that the number of treatments varies by age group: the proportion of children who received two or more treatments decreased significantly after 2 years of age. The most common route of administration for systemic antibiotics was oral, while only a small proportion of children received a parenteral antibiotic (5% of treated children). This percentage decreased from 6.2% in the 1 2 years age group to <4% in the >6 years age group. The most frequently used antibiotics were cephalosporins (43.7% of treated children), followed by macrolides, a combination of penicillins plus β-lactamase inhibitors, and broad-spectrum penicillins (40.1%, 29.8% and 29.4%, respectively). Cephalosporins were frequently prescribed in the youngest children and macrolides were mostly prescribed in children >6 years of age (Figure 2). Table 3 shows that the combination of amoxicillin plus clavulanic acid and amoxicillin were the most frequently prescribed in the whole population. In the 1 2 years age group, amoxicillin was the most frequently prescribed antibiotic, while in older children the combination of amoxicillin and β-lactamase inhibitor was preferred. Cefaclor, a second-generation cephalosporin, was the third most fre- 283

D. Resi et al. Table 2. Number of children treated with systemic and topical antibiotics per 100 inhabitants by age group Systemic antibiotics Topical antibiotics treated children treated children/100 inhabitants/year treated children treated children/100 inhabitants/year 1 2 44.464 69.3 5.508 8.6 3 6 78.956 63.6 6.279 5.1 7 10 56.318 45.4 2.775 2.2 11 14 35.958 35.1 1.636 1.6 Total 215.696 52.0 16.198 3.9 Figure 1. Proportion of children with 1, 2 or 3 treatments with systemic antibiotics by age group (years). quently prescribed antibiotic in the 1 2 and 3 6 years age groups; in the other age groups macrolides such as clarithromycin were the third most frequently prescribed drugs. Topical antibiotics There were 19364 prescriptions for topical antibiotics for eye infections, in 16198 children corresponding to 3.9% of children treated with these antibiotics. The prescription rate was higher for 1 2 years age group (8.6%) and gradually decreased with age to 1.6% in the oldest age class. The mean prescription rate was 5 prescriptions/100 children/year. Overall, the most frequently prescribed groups (72.7%) were nonassociated antibiotics (S01AA and S01AX). S01CA (combination of antibiotics and corticosteroids) were the most used topical antibiotics in the 7 10 and 11 14 years groups (43.7% and 55%). Discussion Figure 2. Most frequently prescribed therapeutic groups of antibiotics by age (percentage of total treatments). Existing reports of antibiotics are usually based on wholesale statistics or studies of small populations, and few population-based reports on antibiotic prescriptions in paediatric populations have been published. As far as we know, the only paper reporting population-based data, while taking into account multiple antibiotic prescriptions to an individual child, is the paper by Thrane et al., 12 who report antibiotic data from a Danish county in 1997. In comparison to data reported by Thrane et al. 12 in the North Jutland county, the amount of antibiotic use in the paediatric population of Emilia Romagna is significantly higher, in terms of both the number of children receiving at least one prescription in a year and the number of antibiotic treatments per year. In Denmark, 429 prescriptions/1000 children/year were recorded, compared with 1190 prescriptions/1000 children/year in the present study (+177.4%); the proportion of children receiving at least one antibiotic prescription per year was 29% in Denmark compared with 52% in Emilia Romagna (+79.3%), with significant differences between age classes: in Emilia Romagna, the proportion of treated children was 39.4% higher among 1 2 year olds (49.7% in Denmark versus 69.3% in Emilia Romagna), but this proportion was 105.3% higher among 11 15 year olds (17.1% in Denmark versus 35.1% in Emilia Romagna). Moreover, in Emilia Romagna, three or more treatments were prescribed to 25.1% of the toddlers (1 2 years age group), to 7.1% of the 7 10 years group and to 3.7% of children >10 years old, compared with 12% of toddlers and 1% of children >7 years in Denmark. Independent of the overall use of antibiotics in Emilia Romagna and Denmark, antibiotic prescription in children was in both cases higher in the first years of life. 12 In our opinion, differences in population density (178 versus 80 inhabitants/km 2 for Emilia Romagna and North Jutland, respectively) between the two countries are not sufficient to explain variation in antibiotic consumption. 284

Antibiotic prescriptions in children Table 3. Most prescribed molecules of antibiotics by age classes (percentage of treated children) ATC Molecule 1 2 3 6 7 10 11 14 total J01CR02 amoxicillin and β-lactamase inhibitor 28.6 32.3 28.7 27.6 27.3 J01CA04 amoxicillin 39.8 28.3 25.0 23.8 26.7 J01FA09 clarithromycin 16.7 19.1 18.5 17.5 16.7 J01DA08 cefaclor 25.7 20.6 9.3 3.6 14.6 J01FA10 azithromycin 14.0 16.6 16.4 16.2 14.6 J01DA23 cefixime 16.8 16.0 10.0 6.7 11.9 J01DA39 ceftibuten 12.5 9.5 4.5 2.9 7.1 J01FA12 rokitamycin 6.8 8.2 6.7 4.6 6.3 J01DA41 cefprozil 7.6 7.1 3.5 2.7 5.1 J01DA06 cefuroxime 5.2 6.4 5.3 4.3 5.0 J01DA13 ceftriaxone 4.6 3.6 2.3 2.0 2.9 J01DA33 cefpodoxime 2.5 2.6 2.4 1.8 2.2 J01FA06 roxithromycin 0.9 1.3 2.5 4.4 1.9 J01EE01 trimethoprim/sulphonamide 1.5 1.5 1.9 2.5 1.6 J01FA01 erythromycin 1.1 1.2 1.6 1.6 1.2 ATC, Anatomical Therapeutic Chemical. The observed differences in antibiotic prescription behaviour between Denmark and Italy are confirmed by the results of a recent study that compared antibiotic prescription pattern in the Ravenna Local Health Authority, one of the 13 Emilia Romagna Health Authorities, and the county of Funen (Denmark). 13 These data showed significantly higher antibiotic prescribing in Ravenna, in all age groups, in terms of both prevalence of use and overall antibiotic prescription expressed as defined daily dosage (DDD) per inhabitant. Moreover, while in Funen the age group with the maximal antibiotic consumption was the elderly population (>80 years), with 20.1 DDD/ 1000 inhabitants day, in Ravenna the age group 0 9 years scored highest, with 23.6 DDD/1000 inhabitants day. The striking differences observed between Emilia Romagna and Denmark are due to the fact that Denmark, according to a recent survey, is one of the European countries consuming less outpatient antibiotics, while Italy ranked sixth after France, Spain, Portugal, Belgium and Luxembourg, the countries with the highest sales. 14 The antibiotic usage in the paediatric population reported in Emilia Romagna is similar to that observed in other Northern Italian regions: the prevalence of usage among children was 55.2% in 1999 in a study involving three northern regions 15 and 42% in 1997 in another study. 16 Antibiotic pressure on children is probably even higher in the south, given the large variations among Italian regions, according to Italian National Statistics on drug use published by the Italian Ministry of Health: 17 the overall consumption of antibiotics in Italy in year 2000 was 22.7 DDD/1000 inhabitants, ranging from 17.7/1000 in the north to 29.8/1000 in the south. A limitation of the present study is that it is based on reimbursement data, and this could have caused an underestimate of real exposure to antibiotics, due to antibiotics prescribed by private physicians or antibiotics obtained over the counter. A recent study involving 42 community paediatricians in Southern Italy showed that 22% of the expenditure for antibacterials was sustained directly by the family, reaching 60% for cheaper antibiotics such as amoxicillin. 18 Thus, antibiotic exposure can actually be significantly higher than that described by reimbursement data. Another limitation of our study is that reimbursement data do not provide information about the indications for the prescription of these drugs; as a consequence, no information is available regarding reasons for and appropriateness of prescribing. The strengths of the study are the large population size, and reproducible analysis and results. The selection of antibiotics by Emilia Romagna physicians in the paediatric population is significantly different from that reported in Denmark: 12 Italian children received significantly more cephalosporins, broad-spectrum penicillins, including combinations with β-lactamase inhibitors, and macrolides than Danish children. In Denmark there is no reimbursement for oral cephalosporins in the community and this alone may explain the difference. Other factors may contribute to this differences, such as industry pressure and existing policies on the use of antimicrobials in respiratory infections. In Emilia Romagna, penicillin V was even not included among the 15 most frequently prescribed drugs. This study raises considerable concern about the possible effects on antimicrobial resistance of the observed high selective antibiotic pressure in the paediatric population. 19 21 Resistance surveillance data available in Italy show that macrolide resistance in Streptococcus pyogenes and Streptococcus pneumoniae is >30%; penicillin non-susceptible S. pneumoniae strains in invasive diseases are still not as widely spread as in Spain or France, but their frequency is increasing. 22 25 Moreover, β-lactamase production is increasing in Haemophilus influenzae, and is nowadays present in >80% of the Moraxella spp. strains. 26 In conclusion, we have used linkage of reimbursement and population databases to estimate antibiotic prescription rates for children in an Italian region. Comparison of these rates with those for Denmark reveals a large difference, with Italian children receiving more courses of antibiotics and broader spectrum agents. We would encourage interventions aimed at reducing overuse and improving quality of care in the paediatric population, in order to prevent further selection for and dissemination of antibiotic resistant 285

D. Resi et al. strains. The effectiveness of any intervention could be monitored by repeating this survey. Acknowledgements Members of the Emilia Romagna Study Group on the Use of Antibiotics in Children: Asciano Mara, Dalla Casa Paola, Martelli Luisa, Morri Michela, Pascucci Maria Grazia, Polli Francesca, Reggiani Lamberto, Sapigni Ester, Valenti Elisabetta and Venturi Iole. References 1. McCaig, L. F. & Hughes, J. M. (1995). Trends in antimicrobial drug prescribing among office-based physicians in the United States. Journal of the American Medical Association 273, 214 9. 2. Nyquist, A. C., Gonzales, R., Steiner, J. F. et al. (1998). Antibiotic prescribing for children with colds, upper respiratory tract infections, and bronchitis. Journal of the American Medical Association 279, 875 7. 3. Wang, E. E., Einarson, T. R., Kellner, J. D. et al. (1999). Antibiotic prescribing for Canadian preschool children: evidence of overprescribing for viral respiratory infections. Clinical Infectious Diseases 29, 155 60. 4. Baquero, F. (1995). Pneumococcal resistance to beta-lactam antibiotics: a global geographic overview. Microbial Drug Resistance 1, 115 20. 5. Berezin, E. N.,Carvalho, E. S., Casagrande, S. et al. (1996). Streptococcus pneumoniae penicillin-nonsusceptible strains in invasive infections in Sao Paulo, Brazil. Pediatric Infectious Disease Journal 15, 1051 3. 6. Hortal, M., Algorta, G., Bianchi, I. et al. (1997). Capsular type distribution and susceptibility to antibiotics of Streptococcus pneumoniae clinical strains isolated from Uruguayan children with systemic infections. Pneumococcus Study Group. Microbial Drug Resistance 3, 159 63. 7. Baquero, F., Barrett, J. F., Courvalin, P. et al. (1998). Epidemiology and mechanisms of resistance among respiratory tract pathogens. Clinical Microbiology and Infection 4, Suppl. 2, S19 26. 8. Magee, J. T., Pritchard, E. L., Fitzgerald, K. A. et al. (1999). Antibiotic prescribing and antibiotic resistance in community practice: retrospective study, 1996 8. British Medical Journal 319, 1239 40. 9. Seppala, H., Klaukka, T., Vuopio-Varkila, J. et al. (1997). The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland. Finnish Study Group for Antimicrobial Resistance. New England Journal of Medicine 337, 441 6. 10. Resi, D., Castelvetri, C., Vaccheri, A. et al. (2001). The therapeutic course as a measure complementary to defined daily doses when studying exposure to antibacterial agents. European Journal of Clinical Pharmacology 57, 177 80. 11. SAS Institute, Inc. (1999). C.N.U.S. version 8. SAS Software. SAS Institute, Inc., Cary, NC, USA. 12. Thrane, N., Steffensen, F. H., Mortensen, J. T. et al. (1999). A population-based study of antibiotic prescriptions for Danish children. Pediatric Infectious Disease Journal 18, 333 7. 13. Vaccheri, A., Bjerrum, L., Resi, D. et al. (2002). Antibiotic prescribing in general practice: striking differences between Italy (Ravenna) and Denmark (Funen). Journal of Antimicrobial Chemotherapy 50, 989 97. 14. Cars, O., Molstad, S. & Melander, A. (2001). Variation in antibiotic use in the European Union. Lancet 357, 1851 3. 15. Rossi, E., De Rosa, M., Bonati, M. et al. (2001). Gruppo di lavoro Aziende USL La prescrizone farmaceutica pediatrica nell ambito delle cure primarie. Rapporto dalla banca dati ARNO. Giornale italiano di farmacia clinica 15, 26 9. 16. Troncon, M. G. (1999). Antibiotici per uso sistematico: la prescrizione negli anni 1993 97. [On-line.] http://www.medicoebambino. com/elettroniche/archivio/arch1999/ri/ri02990.htm (4 June 2003, date last accessed). 17. Ministero della Salute. (2001). L uso dei farmaci in Italia. Rapporto Nazionale anno 2000. [On-line.] http://www.ministerosalute.it/medicinali/ osmed/osmed.jsp (4 June 2003, date last accessed). 18. Campi, R., Garattini, L., Cazzato, L. et al. (2001). Analisi farmacoeconomica delle prescrizioni pediatriche extraospedaliere in un contesto italiano. Giornale italiano di farmacia clinica 15, 30 5. 19. Austin, D. J., Kristinsson, K. G. & Anderson, R. M. (1999). The relationship between the volume of antimicrobial consumption in human communities and the frequency of resistance. Proceedings of the National Academy of Sciences, USA 96, 1152 6. 20. Guillemot, D. (1999). Antibiotic use in humans and bacterial resistance. Current Opinion in Microbiology 2, 494 8. 21. Low, D. E. (2001). Antimicrobial drug use and resistance among respiratory pathogens in the community. Clinical Infectious Diseases 33, Suppl. 3, S206 13. 22. Marchese, A., Mannelli, S., Tonoli, E. et al. (2001). Prevalence of antimicrobial resistance in Streptococcus pneumoniae circulating in Italy: results of the Italian Epidemiological Observatory Survey (1997 1999). Microbial Drug Resistance 7, 277 87. 23. Pantosti, A., D Ambrosio, F., Tarasi, A. et al. (2000). Antibiotic susceptibility and serotype distribution of Streptococcus pneumoniae causing meningitis in Italy, 1997 1999. Clinical Infectious Diseases 31, 1373 9. 24. Bassetti, M., Manno, G., Collida, A. et al. (2000). Erythromycin resistance in Streptococcus pyogenes in Italy. Emerging Infectious Diseases 6, 180 3. 25. Varaldo, P. E., Debbia, E. A., Nicoletti, G. et al. (1999). Nationwide survey in Italy of treatment of Streptococcus pyogenes pharyngitis in children: influence of macrolide resistance on clinical and microbiological outcomes. Artemis-Italy Study Group. Clinical Infectious Diseases 29, 869 73. 26. Nicoletti, G., Blandino, G., Caccamo, F. et al. (2002). The Italian Epidemiological Survey 1997 1999. Antimicrobial susceptibility data of Haemophilus influenzae, Haemophilus parainfluenzae and Moraxella catarrhalis in Italy. International Journal of Antimicrobial Agents 20, 263 9. 286