Open Access Research Are children carrying the burden of broad-spectrum antibiotics in general practice? Prescription pattern for paediatric outpatients with respiratory tract infections in Norway Guro Haugen Fossum,1,2 Morten Lindbæk,1 Svein Gjelstad,1 Ingvild Dalen,1 Kari J Kværner3,4 Tocite:FossumGH, ABSTRACT ARTICLE SUMMARY LindbækM,GjelstadS,etal. Objectives:Toinvestigatetheantibioticprescription Arechildrencarryingthe patternandfactorsthatinfluencethephysicians’ Article focus burdenofbroad-spectrum antibioticsingeneral choiceofantibiotic. ▪ Scandinavia has a conservative prescription practice?Prescriptionpattern Design:Observationalstudy. pattern for antibiotics in respiratory tract forpaediatricoutpatientswith Setting:PrimaryhealthcareinNorway,December infections. respiratorytractinfectionsin 2004throughNovember2005. ▪ Norwegian guidelines recommend the use of Norway.BMJOpen2013;3: Participants:426 general practitioners, GPs, in phenoxymethylpenicillin in almost all upper e002285.doi:10.1136/ Norway, giving 24888 respiratory tract infection respiratory tract infection in general practice bmjopen-2012-002285 episodes with 19938 children aged 0–6years. (GP). Outcome measures: Assess antibiotic ▪ Specific challenges apply when prescribing ▸ Prepublicationhistoryand prescription details and patient and GP drugstochildren. additionalmaterialforthis characteristics associated with broad-spectrum and paperareavailableonline.To Key messages narrow-spectrum antibiotic use. viewthesefilespleasevisit ▪ The general prescription rate for antibiotics in Results: Of the 24888 episodes in the study, thejournalonline respiratory tract infections is relatively low in 26.2% (95% CI 25.7% to 26.8%) included an (http://dx.doi.org/10.1136/ Norway,buthigherinchildrenthanadults. bmjopen-2012-002285). antibiotic prescription. Penicillin V accounted for ▪ Norwegian GPs tend to overprescribe macrolide 42% and macrolide antibiotics for 30%. The antibioticstochildren. Received30October2012 prescription rate varied among the physicians, with ▪ The national guidelines for antibiotics in GP are Revised30October2012 a mean of 25.5% (95% CI 24.2% to 26.7%). not followed in children diagnosed with bron- Accepted3December2012 Acute tonsillitis gave the highest odds for a chitis/bronchiolitis. prescription, OR 33.6 (95% CI 25.7% to 43.9%), Thisfinalarticleisavailable compared to ‘acute respiratory tract infections and Strengths and limitations of this study foruseunderthetermsof symptoms’ as a reference group. GPs with a ▪ Large data set with representative GPs from dif- theCreativeCommons AttributionNon-Commercial prescription rate of 33.3% or higher had the larger ferentpartsofNorwayparticipating. 2.0Licence;see probability for broad-spectrum antibiotic ▪ Analyses performed at the episode level, limiting http://bmjopen.bmj.com prescriptions, OR 3.33 (95% CI 2.01% to 5.54%). theriskofincludingrevisitsforthesameillness. Antibiotic prescriptions increased with increasing ▪ Data mainly from 2005; however, recent data patient age. from the Norwegian Prescription Data base indi- Conclusions: We found a low antibiotic cate that the prescription pattern has undergone prescription rate for childhood respiratory tract fewchangessincethen. infections. However, our figures indicate an overuse of macrolide antibiotics and penicillins with extended spectrum, more so than in the INTRODUCTION corresponding study including the adult population. Almost 90% of antibiotics sold in Norway are Palatability of antibiotic suspensions and other prescribed in the primary healthcare sector.1 administrative challenges affect medication Themostcommonantibioticsubtypeisphenox- Fornumberedaffiliationssee compliance in children. To help combat antibiotic ymethylpenicillin,followedbyerythromycin,piv- endofarticle. resistance, guidelines need to be followed, in mecillinam and doxycycline.2 Although the particular for our youngest patients. Scandinavian prescription pattern is conserva- Correspondenceto Trial registration number (clinicaltrials.org): tive, the European trend of increased use of DrGuroHaugenFossum; NCT00272155. [email protected] broad-spectrum penicillins and macrolide FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 1 Are children carrying the burden of broad-spectrum antibiotics? antibiotics is also found in Norway.2–4 Recent figures from RTIs and prescription drugs for the elderly.22 As part of both Holland and the USA from 1998 to 2004 confirmed theRx-PADstudy,a1-yearsamplingofbaselinedatafrom thetendency.56 441 GPs were performed (December 2004 through The use of antibiotics in Norwegian children aged November 2005); the material of the present study. All 0–4years was approximately 30% in 2005/2006, com- RTIdiagnoseswereincluded, basedon theInternational pared to anaverageofabout 24% inthegeneralpopula- Classification of Primary Care (ICPC-2) coding system.23 tion.1 Users per 1000 aged 0–9years were 220 in 2005 Data were extracted from the doctors’ computer systems and 227 in 2011.7 Most prescriptions for children in usingsoftwaredevelopedbyMediataAS. primary healthcare are results of common airway infec- Of the 441 GPs in the project, 12 GPs with less than tions. A Norwegian study from 1989 found that 80% of 10 registered episodes, 72 episodes where patient general practice (GP) contacts for children aged 0– gender was missing and 3 GPs in rural practices with an 12years with the diagnoses tonsillitis, sinusitis, acute over-the-counter delivery of antibiotics were excluded. bronchitis,bronchiolitis andpneumoniawereprescribed The total number of episodes available for analysis were antibiotics, with phenoxymethylpenicillin as the drug of 24888, including 19938 patientsand 426GPs (figure 1). choice.8 The Rx-PAD database was merged with the Norwegian When prescription rates are considered, it is import- National Prescription Database (NorPD). Informations ant to keep in mind that several studies from GP show included in the database are the drug-specific anatom- that consultations for upper respiratory tract infections ical therapeutic chemical codes, name, strength, admin- (RTIs) are decreasing, while the proportion of antibiotic istration form, compound size and the date of prescriptions remains constant.9–12 Variations in phys- withdrawal by the patient.24 ician seeking behaviour and changing prescription The outcome of interest for the present study was anti- pattern among GPs are suggested explanations. biotic prescription details from the GPs. The following Adequate and rational use of antibiotics is one needed diagnoseswereincluded according totheICPC-2classifi- action to prevent and combat antibiotic resistance,13 14 cation system: acute RTIs and symptoms (R01–29, 74, hence several surveillance systems monitor the use of 80), acute bronchiolitis/bronchitis (R78), pneumonia antibiotics on national and European levels.15–17 The (R81), acute otitis media and ear pain (H01,71, 72, 74), proportion of pneumococcus airway isolates with acute tonsillitis (R72, 76) and other RTIs including reduced penicillin sensitivity was 3.3% in 2007 and sinusitis, laryngitis and asthma (R71, 75, 77, 79, 83, 95, shows an increasing tendency.16 Erythromycin resistance 96).23 in pneumococcus airway isolates increased from 2% in Consultations with preschool children, aged 0–6years, 2000 to 12.4% in 2006 but decreased to 9.4% in were extracted from the Rx-PAD baseline data to form a 2007.16 18 Resistance to macrolide antibiotics is of par- distinct database in SPSS. Consultations within 30days of ticular importance in children. Studies have shown each other were combined to form one episode, of resistant strains in their oral flora in 17% of patients which the first antibiotic prescription of the episode was 6weeks following treatment with erythromycin, and 85% selected for furtheranalysis. Analyses wereperformed by forazithromycin.19 episode; one patient may be included more than once. Norwegian guidelines present phenoxymethylpenicil- Less common antibiotics were grouped. Antibiotics lin asthe drug of choice in almost all upper RTIs if anti- common for urinary tract infections and rarely in use biotics are required.20 The level of resistance is low, and for respiratory infections (nitrofurantoin, trimethoprim, the only reasons for second-choice antibiotics are peni- pivmecillinam and methenamine) were coded as no cillin allergy or treatment failure. Compliance is an antibioticprescription. important issue when antibiotic suspensions are sub- Descriptive statistics were estimated in IBM SPSS scribed, palatability being onedeterminant.21 Statistics V.19. Logistic regression models were used to The aim of the study was to investigate the prescrip- identify predictors of antibiotic prescription and predic- tion pattern of antibiotics for RTIs in Norwegian tors of broad-spectrum antibiotic prescription in Stata primary healthcare in preschool children; estimate the IC/11.2 for Windows. The results from logistic regres- prescription rate and assess factors affecting the GP’s sion analyses were adjusted for clustering by means of prescription pattern. multilevel modelling with random intercept at the group and doctor level, and random slope at diagnoses for doctors. MATERIAL ANDMETHODS The Prescription Peer Academic Detailing (Rx-PAD) Study was initiated in 2004 at the University of Oslo to RESULTS provide structured training in the use of prescription GP characteristics areshown in table 1. drugs for GPs. Eighty GP training groups including A total of 426 GPs were included in the study, 31.5% approximately 450 physicians from different parts of the and 68.5% females and males, respectively. A total of country were recruited with the purpose of raising 86% of the physicians were authorised GP specialists. awareness and improving GP’s prescription patterns in Most GPs worked ina group practice. 2 FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 Are children carrying the burden of broad-spectrum antibiotics? Figure1 Selectionprocessafter datafrom441general practitioners(GPs)and25093 respiratorytractinfection episodesfor1year(December 2004throughNovember2005) werecollected. The prescription pattern among GPs varied consider- Table1 Characteristicsof426participatinggeneral practitioners(GPs) ably,withameanof25.5(95%CI24.2to26.7)antibiotic prescriptions per 100 episodes. For broad-spectrum anti- Characteristic biotic proportion, the corresponding figure was 53.7 (n=426) Number %(95%CI) (95% CI 50.7 to 56.7). Among the latter, more than 50 Gender doctorsprescribed broad-spectrum antibiotics every time Female 135 31.7(27.3to36.1) (figure2). Male 291 68.3(63.9to72.7) Respiratory infections were most common among Grouppractice 2-year-olds to 3-year-olds, 41% experienced infections Yes 394 92.5(90.0to95.0) No 32 7.5(5.0to10.0) and fewer episodes were found in June, July and August. GPspecialist Of the 24888 episodes, 53% were boys and 47% were Yes 365 85.7(82.4to89.0) girls. The three most common diagnoses were acute No 61 14.3(11.0to17.6) upper RTI (R74), cough (R05) and acute otitis media Practicelocation (H71), 27%, 18% and 16%, respectively. A total of 20% City 233 54.7(50.0to59.4) of the episodesarerepeating visits. Rural 193 45.3(40.6to50.0) An antibiotic prescription was registered in 26.2% Mean Median(range) (6525) of all episodes. Penicillin V was prescribed most Age(years) 49 50(28to67) frequently, followed by macrolide/lincosamide antibio- Yearssince 19 20(0to41) tics. In the latter group, lincosamides accounted foronly authorisation 1.3%; the group is therefore referred to as macrolides.25 Numberoflisted 1328 1316(0to3385) patients Few registered prescriptions of tetracyclines and cepha- Consultationsperyear 3095 3029(309to11252) losporins for RTIs were found. Episodes with acute ton- sillitis resulted in a prescription in 77%, and more than FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 3 Are children carrying the burden of broad-spectrum antibiotics? childrenbelowage1(n=1552),whereonly10%received an antibiotic prescription. Predictors for antibiotic prescriptions are shown in table 3. Tonsillitis and pneumonia gave the highest OR for antibiotics among the diagnostic groups, with ORs of 33.6 (95% CI 25.7 to 43.9) and 16.5 (95% CI 12.7 to 21.4), respectively. Older children had an increased probability for antibiotics compared to the age group 0–1years (table 3). Gender or GP characteristics did not predict antibiotic prescriptions, except for GPs with a high work load (measured by annual consultations), which were more frequent prescribers. Bivariate analysis found increased ORs for antibiotic prescriptions among GPs in single practice and non-specialists, but these dif- ferences did not remain significant afteradjustment. Regression analysis on predictors for broad-spectrum antibiotics showed a similar pattern (table 4). The diag- Figure2 Antibioticprescriptionratesforrespiratorytract noses giving the highest ORs were the group of other infectionepisodesandproportionbroad-spectrumantibiotic respiratory infections and bronchitis, ORs of 3.04 (95% prescriptionsamongthe426participatinggeneral practitioners. CI 2.28 to 4.05) and 2.71 (95% CI 1.91 to 3.86), respect- ively gave the highest ORs compared to the reference group of acute RTIs. The probability for broad-spectrum 67% of the prescriptions were penicillin V. As shown in antibiotics was low for tonsillitis, and increasing age table 2, a high antibiotic rate for acute bronchitis was lowered the probability (table 4). Predictors of broad- found, 40%, with macrolide antibiotics as the choice of spectrum antibiotics with regard to the GP gave higher preference. When patients were diagnosed ‘other RTIs’, ORs for specialist GPs. The most significantly increased only 12% of the episodes resulted in antibiotic prescrip- OR was seen in high prescribers, OR 3.33 (95% CI 2.01 tions. In the latter, 52% of the episodes resulted in to 5.54), for GPs withprescription rates of33.3–77.8%. macrolide antibiotics and 22% in penicillins with extended spectrum. In the otitis media group where antibiotics were prescribed, penicillin V and penicillins DISCUSSION with extended spectrum were provided in 46% and Our results showed frequent use of broad-spectrum anti- 33%, respectively. A lower prescription rate was found in biotics for RTIs in children, but a low overall Table2 Antibioticprescriptionsof426GPsfor24888RTIepisodesfor1year(December2004throughNovember2005) accordingtoICPC-2diagnosesandtypeofantibiotic ICPC-2 Episodes Prescriptionrate Diagnoses codes N (%)(95%CI) PcVn(%) Pc-Extn(%) Macn(%) OtherJ01n(%) Acuteupper R01–29, 6806 13.5(12.7to14.3) 388(42) 245(27) 272(30) 17(2) respiratorytract 74,80 infectionsand symptoms Otherrespiratory R71,75, 8758 12.4(11.7to13.1) 271(25) 236(22) 562(52) 17(2) tractinfections 77,79, (including 83,95 sinusitisand laryngitis) Acute R78 1481 40.2(37.7to42.7) 141(24) 138(23) 309(52) 8(1) bronchiolitis/ bronchitis Pneumonia R81 606 67.0(63.3to70.7) 127(31) 97(24) 178(44) 4(1) Acuteotitisand H01,71, 5961 42.5(41.2to43.8) 1174(46) 832(33) 477(19) 53(2) earpain 72,74 Acutetonsillitis R72,76 1276 76.7(74.4to79.0) 654(67) 137(14) 176(18) 12(1) Total 24888 26.2(25.7to26.7) 2755(42) 1685(26) 1974(30) 111(2) GPs,generalpractitioners;ICPC-2,InternationalClassificationofPrimaryCare;Mac,macrolideandlincosamideantibiotics;OtherJ01, cephalosporins,co-trimoxazoleandtetracyclins;PcExt,penicillinswithextendedspectrum(non-penicillinV);PcV,penicillinV;RTI,respiratory tractinfection. 4 FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 Are children carrying the burden of broad-spectrum antibiotics? Table3 Multiplelogisticregressionanalysisshowingfactorsindependentlyassociatedwith426generalpractitioners’(GPs) antibioticprescriptionsfor24888respiratorytractinfectionepisodesduring1year(December2004throughNovember 2005) n(episodes) AdjustedOR(95%CI) Patientfactors Patient’sgender Male(ref) 13214 1 Female 11674 0.97(0.90to1.04) Groupofdiagnoses Acuteupperrespiratorytractinfectionsandsymptoms(ref) 6806 1 Otherrespiratorytractinfections 8758 0.76(0.67to0.87) Acutebronchitis 1481 4.35(3.53to5.35) Pneumonia 606 16.5(12.7to21.4) Acuteotitisandearpain 5961 5.16(4.55to5.85) Acutetonsillitis 1276 33.6(25.7to43.9) Patientage <2years(ref) 7909 1 3–4years 10118 1.34(1.24to1.46) 5–6years 6861 1.41(1.28to1.54) Doctorfactors Doctor’sgender Male(ref) 16281 1 Female 8607 1.06(0.88to1.28) Practicelocation City(ref) 14009 1 Rural 10879 1.00(0.84to1.18) Practicetype Group(ref) 23207 1 Single 1681 1.10(0.81to1.49) Educationallevel GPspecialist (ref) 21798 1 Notspecialist 3090 1.38(1.06to1.81) Doctor’sage 28–43(ref) 7096 1 44–49 7544 1.19(0.94to1.52) 50–54 5833 1.05(0.82to1.36) 55–67 4415 1.12(0.86to1.44) Patientcontactsperyear 309–2302(ref) 4025 1 2303–3028 5495 1.16(0.91to1.47) 3029–3711 7060 1.27(0.99to1.62) 3712–11252 8308 1.54(1.20to1.97) Significantvaluesinbold,p<0.05. prescription rate. Overprescription was found for bron- suspected atypical pneumonia or penicillin allergy. In chitis and in busy GPs. our study they have been used for other diagnoses as Although penicillin V was the most frequent preferred well. One explanation may, at least in part, be adminis- choice of antibiotic in 42% of childhood diagnoses, a trative convenience and a more preferential taste com- high percentage of broader spectrum antibiotics, espe- pared to the bitter-tasting penicillin V. It has been found cially macrolides, was found. An overall 30% macrolide that tolerability, taste and administrative frequency use is seen in the present study. A proportion of 52% for are of importance in paediatric populations.21 28 29 bronchitis differs from a bronchitis macrolide propor- When Hoppe and co-workers tested paediatric tion of 32% in the corresponding study whereadult data compliance in oral antibiotics, acompliance of 94% and also are included.26 The use of penicillin with extended 90% was found for clarithromycin and erythromycin. spectrum is recommended prior to the use of macro- Correspondingly, penicillin V suspensions had a compli- lides due to resistance issues.16 18 19 Macrolides, accord- ance rate of 62% and 56%.30 Hinnerskov et al31 have ing to the Norwegian guidelines for antibiotic use in highlighted a similar problem of macrolide overuse in primary care,20 27 are first choice of preference only in Denmark. FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 5 Are children carrying the burden of broad-spectrum antibiotics? Table4 Multiplelogisticregressionanalysisshowingfactorsindependentlyassociatedwith426generalpractitioners’(GPs) proportionofbroad-spectrumantibiotics(non-penicillinV)for6525respiratorytractinfectionepisodeswithprescribed antibioticfor1year(December2004throughNovember2005) n(episodes) AdjustedOR(95%CI) Patientfactors Patient’sgender Male(ref) 3490 1 Female 3035 1.04(0.91to1.19) Groupofdiagnoses Acuteupperrespiratorytractinfectionsandsymptoms(ref) 922 1 Otherrespiratorytractinfections 1086 3.04(2.28to4.05) Acutebronchitis 596 2.71(1.91to3.86) Pneumonia 406 2.10(1.46to3.02) Acuteotitisandearpain 2536 1.11(0.87to1.43) Acutetonsillitis 979 0.24(0.18to0.33) Patientage <2years(ref) 1533 1 3–4years 2888 1.13(0.94to1.35) 5–6years 2104 0.79(0.66to0.96) Doctorfactors Doctor’sgender Male(ref) 4310 1 Female 2215 0.83(0.56to1.24) Practicelocation City(ref) 3689 1 Rural 2836 1.07(0.72to1.59) Practicetype Group(ref) 6038 1 Single 487 1.21(0.61to2.34) Educationallevel GPspecialist (ref) 5658 1 Notspecialist 867 0.53(0.30to0.94) Doctor’sage 28–43(ref) 1878 1 44–49 1927 0.71(0.43to1.19) 50–54 1489 1.06(0.62to1.80) 55–67 1231 0.85(0.49to1.49) Patientcontactsperyear 309–2302(ref) 955 1 2303–3028 1367 1.48(0.90to2.44) 3029–3711 1889 1.23(0.74to2.06) 3712–11252 2314 1.44(0.85to2.46) Antibioticprescriptionrate(%) 0–15.60(ref) 580 1 15.61–23.64 1236 1.15(0.69to1.90) 23.65–33.33 1942 1.37(0.84to2.23) 33.34–77.78 2767 3.33(2.01to5.54) Significantvaluesinbold,p<0.05. The overall prescription rate of only 26% for RTIs in a prescription rate of only 26% shows a decreasing ten- children aged 0–6years in our study was rather low com- dency. In 1998, antibiotics were the rule rather than the pared to the corresponding data including the adult exceptionforchildrenvisitingGPsforotitismedia,tonsil- population.26 Although a low rate in preschool children litis,sinusitis,bronchitis/bronchiolitisandpneumonia.8 of 24% was found of the latter, a high proportion was Antibiotic prescriptions by diagnosis showed a similar broad-spectrum antibiotics.26 Blix et al1 reported popula- distribution pattern to that of the corresponding dataset tion prevalence for all antibiotic prescriptions of 24% including adults, but a higher rate for otitis media and a andforchildren0–4yearsabout30%in2005/2006.Our lower rate for symptom diagnoses and bronchitis.26 selection only included RTIs, and is, accordingly, some- SubstantialefforthasbeenperformedinNorwaythepast what lower. Compared to a Norwegian study from 1998,8 decade to decrease the prescription rate of antibiotics. 6 FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 Are children carrying the burden of broad-spectrum antibiotics? Our data indicate that these efforts are about to pay off. comprised a complete year, including all seasonal varia- For instance, we found a low antibiotic prescription rate tions. GPs from most parts of the Southern Norway inpatientswithotitismedia,42.5%.Thisisinaccordance participated; urban as well as rural areas. We believe it is withcurrentNorwegianguidelines.20Comparablefigures likely that our data represent the prescription pattern frompreviousstudiesonantibioticprescriptionsforRTIs for RTIs well and that our findings are generalisable to in children in several countries revealed an average pre- Norwegian GPs. We have chosen to analyse by infectious scription rate of 47%. Among the 14 study populations episodes, with a cut-off at 30days, well aware that a few included in the study, Italy and Canada have high levels new episodes may represent revisits due to treatment of42and57%,respectively.HollandandUKhavecorres- failure and follow-up visits by the same patient. Previous pondinglylowlevelsof14%and21%.32 studies havebeenperformedbasedonconsultationsand by infectious episodes.32 36 The consultations in the Predictors of antibiotic prescriptions present studyonly represent GPs’regularofficework, no The prescription rate of 40% for bronchitis/bronchio- out-of-hour emergency visits are included. In most of litis was worryingly high. This figure is lower than the Norway, the same office GPs also are on duty taking care former study from 1998,8 but not in correspondence of emergency cases during evenings and nights. A study with current Norwegian guidelines.20 Considerable byFagan39showedthatNorwegianGPs’prescriptionpat- improvement in prescription routines is needed. terns are identical when regular office consultations are Cochrane reviews have shown no evidence for treating compared to emergency visits forotitis media and tonsil- acute bronchiolitis in children with antibiotics, and only litis. Our data are mainly from 2005; however, figures a modest beneficial effect when treating patients from NorPD indicate similar patterns of antibiotic use 8–65.33 34 No obvious explanation to the high prescrip- nowin2011.7 tion rate for bronchitis is evident, but one reason may In conclusion, a low antibiotic prescription rate for be a perceived patient or parental demand for prescrip- childhood RTIs was found. However, the GPs tend tion, as reported by Little et al.35 We also suspect that to choose macrolide antibiotics and penicillins with some diagnoses may be registered based on symptom extended spectrum more often than the guidelines findings, although the GP suspected more serious illness recommend. We recognise the specific challenges that and subsequently prescribed antibiotics. C reactive are related to medication compliance in children. protein (CRP), CRP-testing, is usually performed follow- Administrative simplifications, such as the availability to ing clinical examination, and after the ICPC-2 coding. oral penicillin suspensions during meals, may help shift Elevated CRP may cause a prescription without a corre- the trend away from broad-spectrum antibiotics.40 sponding change in the initial ICPC-2 code. In the Palatability may also affect compliance, taking penicillin present study, 12–14% received antibiotics based on V’s bitter taste into account. Differences between GPs symptom diagnoses only. Jansen et al report similar can be targeted by more individual and specified means. results from Holland, with a symptom diagnose prescrip- We recommend that GPs replace narrow spectered peni- tion rate of 12–15% for preschool children.36 Such cillins with extended spectered penicillins rather than infections are likely to have viral pathology. A corres- macrolidesas second-choice antibiotic in RTIs. pondingly low prescription rate is expected. Children aged 5–6years had the highest probability Authoraffiliations for antibiotic prescriptions, children below age 2 the 1DepartmentofGeneralPractice,InstituteofHealthandSociety,Universityof Oslo,Oslo,Norway lowest. An opposite pattern was seen in the prescription 2DivisionofSurgeryandClinicalNeuroscience,Departmentfor pattern for broad-spectrum antibiotics. Another factor Otorhinolaryngology,HeadandNeckSurgery,OsloUniversityHospital,Oslo, influencing antibiotic prescription was related to GP Norway experience, as GP non-specialists were more likely to 3Research,InnovationandEducationUnit,OsloUniversityHospital,Oslo, prescribe antibiotics. A high number of patient contacts Norway 4DepartmentofHealthManagementandHealthEconomics,Universityof per year increased prescription likelihood. High anti- Oslo,Oslo,Norway biotic prescription rates increased the probability for broad-spectrum antibiotics. Slightly surprising, GP spe- Contributors GHF,MLandSGconceivedtheideaofthestudyandwere cialiststend to use more broad-spectrum antibiotics than responsibleforthedesignofthestudy.GHF,SGandIDwereresponsiblefor those without the specialist authorisation. Similar results thedataanalysesandproducedthetablesandgraphs.IDdesignedthemodel arepresentedbyClavennaetal,suggestingthatphysician usedformultilevelanalysis.Allauthorsprovidedinputintothedataanalysis andtotheinterpretationoftheresults.Theinitialdraftofthemanuscriptwas attitude may play a role obtaining prudent antibiotic preparedbyGHF,MLandKJKandthencirculatedrepeatedlyamongall prescriptions37. Antibiotic overuse may also be affected authorsforcriticalrevision. by an existing gap between perceived and real determi- Funding TheRx-PADstudywascarriedoutwithgrantsfromtheNorwegian nantsof antibiotic prescription.38 MinistryofHealthandtheNorwegianMedicalAssociation.Thepresentstudy isfundedaspartofaPhD-projectbytheAntibioticCenterforPrimaryCare. Strengths and limitations Competinginterests ICMEJformsaresubmittedforallauthors.SGis The major strength of this study is the large number of workingparttimeandhasownershipinthecompanythatproducedsoftware infectious episodes in the data set. The study period fordataextractioninthisstudy.Otherauthorsnonetodeclare. FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285 7 Are children carrying the burden of broad-spectrum antibiotics? Ethicsapproval TheRx-PAD-study,mergedwiththedatafromNorPDwas 19. KastnerU,GuggenbichlerJP.Influenceofmacrolideantibioticson approvedbytheregionalethicscommittee,theNorwegiandatainspectorate promotionofresistanceintheoralfloraofchildren.Infection andtheNorwegianDirectorateofHealth. 2001;29:251–6. 20. LindbækM.Helsedirektoratet,Antibiotikasenteretforprimærmedisin. Provenanceandpeerreview Notcommissioned;externallypeerreviewed. Nasjonalefagligeretningslinjerforantibiotikabruki primærhelsetjenesten.Oslo:Helsedirektoratet:Antibiotikasenteretfor Datasharingstatement Therearenoadditionaldataavailable. primærmedin,2008. 21. SteeleRW,ThomasMP,BegueRE.Complianceissuesrelatedto theselectionofantibioticsuspensionsforchildren.PediatrInfectDis J2001;20:1–5. REFERENCES 22. GjelstadS,FetveitA,StraandJ,etal.Canantibioticprescriptionsin respiratorytractinfectionsbeimproved?Acluster-randomized 1. BlixHS,EngelandA,LitleskareI,etal.Age-andgender-specific educationalinterventioningeneralpractice—thePrescriptionPeer antibacterialprescribinginNorway.JAntimicrobChemother 2007;59:971–6. AcademicDetailing(Rx-PAD)Study(NCT00272155).BMCHealth ServRes2006;6:75. 2. LitleskareI,BlixHS,RonningM.AntibioticuseinNorway.Tidsskr NorLaegeforen2008;128:2324–9. 23. KITHAS,Sosial-oghelsedirektoratet,Norskselskapfor allmennmedisin,etal.ICPC-2:deninternasjonaleklassifikasjonen 3. GoossensH,FerechM,CoenenS,etal.Comparisonofoutpatient forprimærhelsetjenesten.Norskutg.,2.utg.ed.(Oslo):Sosial-og systemicantibacterialusein2004intheUnitedStatesand27 Europeancountries.ClinInfectDis2007;44:1091–5. helsedirektoratet;2004. 24. LindbaekM.Whatdoesinfluencetheprescriptionofantibioticsto 4. GoossensH,FerechM,VanderSticheleR,etal.Outpatient children?.TidsskrNorLaegeforen2004;124:2228. antibioticuseinEuropeandassociationwithresistance:a cross-nationaldatabasestudy.Lancet2005;365:579–87. 25. LeclercqR,CourvalinP.Bacterialresistancetomacrolide, lincosamide,andstreptograminantibioticsbytargetmodification. 5. CocoAS,HorstMA,GamblerAS.Trendsinbroad-spectrum AntimicrobAgentsChemother1991;35:1267–72. aSntatitbeiso,ti1c9p9r8e–sc2r0ib0i4n.gBfMorCchPileddreiantrw2i0th09a;c9u:4te1.otitismediaintheUnited 26. GjelstadS,StraandJ,DalenI,etal.Dogeneralpractitioners’ consultationratesinfluencetheirprescribingpatternsofantibiotics 6. deJongJ,vandenBergPB,deVriesTW,etal.Antibioticdruguse foracuterespiratorytractinfections?JAntimicrobChemother oPfhcahrmildarceonli2n0t0h8e;6N4e:9th1e3r–la1n9d.sfrom1999till2005.EurJClin 2011;66:2425–33. 27. BerildD,LindbaekM.Aretoomanyantibioticsprescribedto 7. Nasjonaltreseptbasertlegemiddelregister(theNorwegian Norwegianchildren?TidsskrNorLaegeforen2002;122:2690. PrescriptionDatabase).www.reseptregisteret.no(accesseddate 28. BlockSL,SchmierJK,NotarioGF,etal.Efficacy,tolerability,and Oct2012). parentreportedoutcomesforcefdinirvs.high-doseamoxicillin/ 8. StraandJ,RokstadK,HeggedalU.Drugprescribingforchildrenin clavulanateoralsuspensionforacuteotitismediainyoungchildren. gSetundeyr.alApcrtaacPticaee.dAiatrre1p9o9rt8f;r8o7m:2t1h8e–M24o.re&RomsdalPrescription CurrMedResOpin2006;22:1839–47. 29. RamgoolamA,SteeleR.Formulationsofantibioticsforchildrenin 9. AshworthM,CharltonJ,BallardK,etal.Variationsinantibiotic primarycare:effectsoncomplianceandefficacy.PaediatrDrugs pUrKesgcerinbeinraglapnradcctiocnessu1lt9a9ti5o–n2r0a0te0s.BforrJacGuetenrPersapcitra2t0o0ry5;i5n5fe:6ct0io3n–8in. 2002;4:323–33. 30. HoppeJE,BlumenstockG,GrotzW,etal.ComplianceofGerman 10. KuyvenhovenM,vanEssenG,SchellevisF,etal.Managementof pediatricpatientswithoralantibiotictherapy:resultsofanationwide upperrespiratorytractinfectionsinDutchgeneralpractice;antibiotic survey.PediatrInfectDisJ1999;18:1085–91. prescribingratesandincidencesin1987and2001.FamPract 2006;23:175–9. 31. HinnerskovM,TherkildsenJM,CordobaG,etal.Macrolideoveruse fortreatmentofrespiratorytractinfectionsingeneralpractice.Dan 11. McCaigLF,BesserRE,HughesJM.Trendsinantimicrobial MedBull2011;58:A4356. prescribingratesforchildrenandadolescents.JAMA 2002;287:3096–102. 32. RossignoliA,ClavennaA,BonatiM.Antibioticprescriptionand prevalencerateintheoutpatientpaediatricpopulation:analysisof 12. NeumarkT,BrudinL,EngstromS,etal.Trendsinnumberof surveyspublishedduring2000–2005.EurJClinPharmacol consultationsandantibioticprescriptionsforrespiratorytractinfections 2007;63:1099–106. between1999and2005inprimaryhealthcareinKalmarCounty, SouthernSweden.ScandJPrimHealthCare2009;27:18–24. 33. SmucnyJ,FaheyT,BeckerL,etal.Antibioticsforacutebronchitis. CochraneDatabaseSystRev2004;18:CD000245. 13. Hoiby EA, Vestrheim DF, Caugant DA, et al. Bacterial 34. SpurlingGK,FonsekaK,DoustJ,etal.Antibioticsforbronchiolitisin resistance against antibiotics. Tidsskr Nor Laegeforen 2008;128:2452–6. children.CochraneDatabaseSystRev2007;24:CD005189. 35. LittleP,DorwardM,WarnerG,etal.Importanceofpatientpressure 14. CostelloeC,MetcalfeC,LoveringA,etal.Effectofantibiotic andperceivedpressureandperceivedmedicalneedfor prescribinginprimarycareonantimicrobialresistanceinindividual investigations,referral,andprescribinginprimarycare:nested patients:systematicreviewandmeta-analysis.BMJ2010;340: observationalstudy.BMJ2004;328:444. c2096. 36. JansenAG,SandersEA,SchilderAG,etal.Primarycare 15. CoenenS,FerechM,Malhotra-KumarS,etal.European managementofrespiratorytractinfectionsinDutchpreschool SurveillanceofAntimicrobialConsumption(ESAC):outpatient children.ScandJPrimHealthCare2006;24:231–6. macrolide,lincosamideandstreptogramin(MLS)useinEurope. JAntimicrobChemother2006;58:418–22. 37. CpalaevdeiantnriacAo,utBpoantiaetnitMs..ADricffherDeinscCeshiilnda2n0t1ib1i;o9t6ic:5p9r0e–s5cr.ibingin 16. SimonsenGS.NORM/NORM-VET2007.UsageofAntimicrobial 38. MoroM,MarchiM,GagliottiC,etal.Whydopaediatricians AgentsandOccurenceofAntimicrobialResistanceinNorway. prescribeantibiotics?ResultsofanItalianregionalproject.BMC Tromsø/Oslo2008. Pediatrics2009;9:69. 17. Statensinstituttforfolkehelse.Seksjonforforebyggende 39. FaganM.Isotitisandtonsillitishandledinthesamewaywithin infeksjonsmedisin.MSIS-rapport/Meldesystemetforsmittsomme normalworkinghoursandout-of-hours?.TidsskrNorLaegeforen sykdommer.Oslo:Folkehelsa. 2008;128:2340–2. 18. Simonsen GS. Surveillance and prevalence of antimicrobial 40. WestinAA,WalstadRA,SpigsetO.Shouldpenicillinbegivenonan r2e0s0is9t;a1n2c9e:6i2n3–N7o.rway. Tidsskr Nor Laegeforen emptystomach?.TidsskrNorLaegeforen2009;129:2610–12. 8 FossumGH,LindbækM,GjelstadS,etal.BMJOpen2013;3:e002285.doi:10.1136/bmjopen-2012-002285