Abstract
OBJECTIVE:
To determine whether a paediatric after-hours clinic uses evidence-based management in the treatment of acute otitis media, and compare this management with that provided in a paediatric emergency department and a general hospital emergency department.
METHODS:
A retrospective chart review of 573 patients (aged six months to five years) with a discharge diagnosis of acute otitis media was conducted in three after-hours settings: a paediatric after-hours clinic, a tertiary paediatric hospital emergency department and a secondary general hospital emergency department. The patients’ age, weight, sex and allergy to antibiotics were recorded as baseline characteristics. The physicians’ antibiotic choice, dose and duration, and the use of investigations were recorded as outcome variables.
RESULTS:
Amoxicillin was prescribed to 68% of patients at both the paediatric after-hours clinic and the paediatric hospital emergency department, compared with 53% of patients at the general hospital emergency department (P<0.01). The mean dose of amoxicillin prescribed at the paediatric after-hours clinic and the paediatric hospital emergency department were similar (43.4±9.7 mg/kg per day and 42.4±14.3 mg/kg per day, respectively) and higher than that prescribed at the general hospital emergency department (38.6±8.8 mg/kg per day, P<0.01). The paediatric after-hours clinic used investigations less often than did emergency departments (0.5% of cases compared with 9% and 20%, P<0.01).
CONCLUSION:
The paediatric after-hours clinic provided a high level of adherence to a clinical practice guideline and had a low utilization of resource intensive investigations.
Keywords: Acute otitis media, After-hours care, Evidence-based medicine
Abstract
OBJECTIF :
Déterminer si une clinique pédiatrique ouverte après les heures normales fait appel à un traitement probant de l’otite moyenne aiguë et comparer ce traitement à celui qui est administré à un service d’urgence pédiatrique et au service d’urgence d’un hôpital général.
MÉTHODOLOGIE :
Une étude rétrospective des dossiers de 573 patients (âgés de six mois à cinq ans) dont le diagnostic au congé était une otite moyenne aiguë a été menée dans trois établissements ouverts après les heures normales : une clinique pédiatrique, le service d’urgence d’un hôpital pédiatrique de soins tertiaires et le service d’urgence d’un hôpital général secondaire. L’âge, le poids, le sexe et les allergies aux antibiotiques des patients faisaient partie des caractéristiques de base. Le choix de l’antibiotique par le médecin, la dose et la durée du traitement ainsi que le recours à des explorations faisaient partie des variables d’issue.
RÉSULTATS :
De l’amoxicilline a été prescrite à 68 % des patients de la clinique pédiatrique ouverte après les heures normales et le service d’urgence de l’hôpital pédiatrique, par rapport à 53 % des patients du service d’urgence de l’hôpital général (P<0,01). Les doses moyennes d’amoxicilline prescrites à la clinique pédiatrique et au service d’urgence de l’hôpital pédiatrique étaient similaires (43,4±9,7 mg/kg par jour et 42,4±14,3 mg/kg par jour, respectivement) et plus élevées que celles prescrites au service d’urgence de l’hôpital général (38,6±8,8 mg/kg par jour, P<0,01). La clinique pédiatrique ouverte après les heures normales procédait à moins d’explorations que les services d’urgence (0,5 % des cas par rapport à 9 % et à 20 %, P<0,01).
CONCLUSION :
La clinique pédiatrique ouverte après les heures normales offrait un taux élevé de respect des directives de pratique clinique et utilisait peu les explorations exigeantes en termes de ressources.
Ideally, medical care for children should be accessible, continuous and comprehensive (1). However, it has been found that physicians’ after-hours availability varies markedly among four Canadian cities, from a low of 27% to a high of 92% (2). Thus, parents must frequently seek care for their acutely ill child in alternate settings. Paediatric after-hours clinics have developed in many urban centres to provide after-hours care to children with common acute illnesses. It is not known whether these clinics use evidence-based medicine or how they compare with other after-hours settings, such as emergency departments. This information would be important to health care practitioners, consumers and policy makers.
Clinical practice guidelines have been developed on the basis of evidence from randomized control led trials for a number of acute paediatric illnesses. The objective of clinical practice guidelines is to aid clinicians in practising evidence-based medicine with the ultimate goal of improving patient care (3). A relatively high degree of adherence to evidence-based guidelines is expected. For example, more than 80% of medical treatments for adult illness were evidence-based at a university-affiliated district hospital in the United Kingdom (4). We chose to study the after-hours treatment of acute otitis media (AOM) because considerable evidence from clinical trials exists regarding its treatment, a Canadian clinical practice guideline has been developed and the treatment of AOM plays a significant role in the development of antibiotic resistance.
AOM is the leading cause for antibiotic prescriptions in North America (5). It has been estimated that, on average, children spend over a month on antibiotic therapy for AOM in the first two years of life (6). Thus, the treatment of AOM has become a research priority to curtail the increasing prevalence of antibiotic-resistant bacteria while maintaining a cost effective treatment strategy (7).
The natural history of AOM is such that 80% of all cases resolve spontaneously in seven to 14 days without antibiotic treatment (5–8). The rate of clinical resolution rises to 95% with antimicrobial treatment and there appears to be no significant difference in the efficacy of treatment regardless of which of the antibiotics indicated for AOM are used (7–9). Most North American experts agree that antimicrobial therapy is the standard of care for AOM (9–11) because of the statistically significant improvement in outcome when antibiotics are used to treat AOM and because complications of AOM such as hearing loss, mastoiditis and brain abscesses have become rare since the dawn of the antibiotic era. Amoxicillin is generally considered the drug of first choice because of its good side effect profile, low cost and high palatability (12–15).
In July 1998, the Canadian Paediatric Society (CPS) released a position statement recommending standard dose amoxicillin (generally considered to be 40 mg/kg to 50 mg/kg per day) (16) for ten days as the first-line treatment for AOM (17). This statement was reaffirmed in February 2000 and, with respect to first-line therapy, is consistent with the most recent recommendations from the Center for Disease Control in Atlanta (18).
Using these guidelines as a benchmark for the appropriate care of children with AOM, we examined guideline adherence in a paediatric after-hours clinic in a large Canadian city and compared this with adherence in a paediatric hospital emergency department and a general hospital emergency department in the same city.
METHODS
Settings
Data from a paediatric after-hours clinic, a tertiary paediatric hospital emergency department and a secondary general hospital emergency department in the same Canadian city were abstracted from health records in this retrospective chart review. The paediatric after-hours clinic is not university-affiliated and is located within 1 km of the university-affiliated general hospital emergency department on the periphery of the city. These two settings serve the same population. The paediatric hospital emergency department is in a university-affiliated academic health sciences centre located in the downtown core of the city. The three settings differ in the number of paediatric patients treated each month, the types of physicians on staff and the utilization of medical students, residents and interns in the care of patients. At the paediatric after-hours clinic, paediatricians attend to 2000 paediatric patients per month. At the general hospital emergency department, approximately 1100 children are seen each month by emergency physicians, general physicians and family medicine residents. At the paediatric hospital emergency department, paediatric specialists (paediatric emergency physicians and paediatricians) and paediatric residents see approximately 4100 children per month.
Study population
Data were abstracted for all consecutive children attending each of the three settings that met the inclusion and exclusion criteria for the study. Children were included in this study if they were between six months and five years of age and had a discharge diagnosis of AOM. To avoid bias due to the time of day, children from all three settings were included only if they were seen during the regular working hours of the after-hours clinic (ie, between 18:00 and 22:00 on weekdays and 09:00 and 18:00 on weekends). Children who were on an antibiotic or who had a coexistent medical condition at the time of presentation were excluded from the study, as were patients who were admitted to hospital. From the paediatric after-hours clinic and the paediatric hospital emergency department, eligible patients were included if they were seen between December 1, 1998, and March 31, 1999. A longer but overlapping time frame (September 1, 1998, through June 31, 2000) was necessary for the general hospital emergency department because of considerably fewer cases meeting the inclusion criteria.
Baseline characteristics and outcomes
Trained medical students abstracted data using a standardized data collection form. Each patient’s age, weight, sex and allergy to antibiotics were recorded as baseline characteristics. If the child’s weight was not noted on the chart, as was the case in the after-hours clinic, the child was assigned the 50th percentile weight for their age. The primary outcome was adherence to the CPS position statement on the choice of antibiotic (amoxicillin is considered first line) in the treatment of AOM. Secondary outcomes included the dose (mg/kg per day) and duration of amoxicillin therapy (if prescribed). In addition, use of laboratory and radiological investigations such as blood and urine cultures, blood counts and chest x-rays was recorded because this information was also thought to be important to health care practitioners, consumers and policy makers.
Data analysis
Previous investigators have identified that more than 80% of inpatient medical treatments are evidence-based (4). Some degree of second-line antibiotic use is expected for children at increased risk of a resistant infection such as those with amoxicillin-refractory persistent otitis media and those who attend group daycare (9). Therefore, it was hypothesized that first-line therapy would be used in 75% of AOM cases irrespective of setting. Based on this hypothesis, a sample size of 140 children was calculated to be sufficient to ensure, with 95% certainty, that the discrepancy between the study sample and the population of children diagnosed with AOM was within ±7%. To allow for ineligible patients and missing data in charts, it was decided that 150 to 200 patients would be selected at each of the three sites.
Statistical analysis was performed using Stat View (SAS Institute Inc, USA) software. Means for continuous variables (age, weight and amoxicillin dose) were compared using ANOVA, and proportions for categorical variables (sex, allergy, amoxicillin and investigations) were compared using χ2 tests. P values in Tables 1 and 2 are reported for the null hypothesis that there is no difference between the data sets for the three settings. Differences were considered statistically significant at P<0.05. If significant differences were found, pairwise comparisons were undertaken with differences considered statistically significant at P<0.01.
TABLE 1.
Baseline characteristics
| Paediatric after hours clinic (n=209) | Paediatric emergency (n=202) | General emergency (n=162) | Total (n=573) | P value | |
|---|---|---|---|---|---|
| Number of patients | 209 (36%) | 202 (35%) | 162 (28%) | 573 (100%) | |
| Number male | 110 (53%) | 109 (54%) | 88 (54%) | 307 (54%) | 0.9 |
| Antibiotic allergy | 12 (6%) | 9 (4%) | 13 (8%) | 34 (6%) | 0.4 |
| Mean age (years)* | 2.5±1.3 | 2.1±1.3 | 2.5±1.4 | 2.4±1.3 | <0.01 |
| Mean weight (kg)* | 13.2±2.9 | 12.9±3.8 | 13.5±3.4 | 13.2±3.4 | 0.2 |
Means reported with SD
TABLE 2.
Outcome variables
| Paediatric after hours clinic | Paediatric emergency | General emergency | Total | P value | |
|---|---|---|---|---|---|
| Amoxicillin use | 143/209 (68%) | 137/202 (68%) | 86/162 (53%) | 366/573 (64%) | <0.01 |
| Mean dose (mg/kg/day)*† | 43.4±9.7 (n=102) | 42.4±14.3 (n=100) | 38.6±8.8 (n=79) | 41.6±11.3 (n=281) | <0.01 |
| Duration less than 10 daysঠ| 10/66 (15%) | 15/115 (13%) | 11/59 (19%) | 36/240 (15%) | 0.6 |
| Investigations | 1/209 (0.50%) | 19/202 (9%) | 32/162 (20%) | 52/573 (9%) | <0.01 |
Means reported with SD;
Dosage of amoxicillin was documented for 281 of 366 cases (77%) prescribed amoxicillin; this rate of documentation varied between the three settings (71% at the paediatric after-hours clinic, 73% at the paediatric emergency department and 92% at the general hospital emergency department).
Duration of amoxicillin therapy was documented for 240 of 366 cases (66%) prescribed amoxicillin; this rate of documentation varied between the three settings (46% at the paediatric after-hours clinic, 81% at the paediatric emergency department and 68% at the general hospital emergency department)
The study was approved by the institutional review boards at the three clinical institutions.
RESULTS
Baseline characteristics
Of the 573 cases of AOM included in this study, 209 (36%) were from the paediatric after-hours clinic, 202 (35%) were from the paediatric hospital emergency department and 162 (28%) were from the general hospital emergency department. No statistically significant differences were observed in the sex, weight or proportion of patients with an antibiotic allergy (Table 1). However, there was a significant difference in the mean age among the sites (P<0.01, Table 1). The mean age of the children treated at the paediatric hospital emergency department was approximately 4.6 months younger than the mean age of those treated at both the paediatric after-hours clinic (P<0.01) and the general hospital emergency clinic (P<0.01).
Antibiotic choice
All the children in this study were prescribed an antibiotic. Amoxicillin was the most frequently prescribed antibiotic in all three settings. The frequency of amoxicillin prescription was the same at the paediatric after-hours clinic and at the paediatric hospital emergency department (68%, Table 2). At the general hospital emergency department, the frequency of amoxicillin prescription was 53%, which was significantly lower than either of the two other sites (P<0.01 for each pairwise comparison). The second most common antibiotic prescription at all three settings was cefaclor, which was prescribed to 12% of patients at the paediatric after-hours clinic, 7% of patients at the paediatric hospital emergency department and 17% of patients at the general hospital emergency department (P<0.01).
Dose of amoxicillin
The mean dose of amoxicillin was significantly different among the three settings (P<0.01, Table 2). The mean dose of amoxicillin prescribed at the paediatric after-hours clinic and at the paediatric hospital emergency department was 43.4±9.7 mg/kg per day and 42.4±14.3 mg/kg per day, respectively (P=not significant). The mean dose of amoxicillin prescribed at the general hospital emergency department was 38.6±8.8 mg/kg per day, which was significantly lower than at the paediatric after-hours clinic (P<0.01) but not at the paediatric hospital emergency department (P=not significant).
Duration of amoxicillin treatment
At all three sites the proportion of children prescribed amoxicillin for less than the recommended ten days was approximately 15% (P=0.5, Table 2). No child received amoxicillin therapy for greater than 14 days.
Use of laboratory and radiological investigations
Use of investigations was 0.5% (one case) at the paediatric after-hours clinic, 9% at the paediatric emergency department and 20% at the general hospital emergency department. These differences were significant (P<0.01, Table 2).
DISCUSSION
In this study, we examined adherence to a clinical practice guideline on the treatment of AOM at a paediatric after-hours clinic and compared this with a paediatric emergency department and a general hospital emergency department. Given the strong evidence and the broad consensus on management, we expected to find little difference among the settings. Considerable variation was identified in use of amoxicillin (recommended as the first line antibiotic), dosing of amoxicillin and performance of investigations.
There are some limitations of this study. The time frame used at the general hospital emergency department began earlier and extended later than the time frame used at the other sites. However, the earliest case reviewed at the general hospital emergency department was two months after the position statement from the CPS was released (17). In addition, during the time period after our chart review of the paediatric hospital emergency department and the paediatric after-hours clinic but before the completion of the review from the general hospital emergency department, the CPS reaffirmed its policy statement. Also during this time period, the Center for Disease Control published its own guideline, which supports amoxicillin as the first-line antimicrobial agent (18). Furthermore, when data before December 1, 1998 are excluded from the analysis, the results for the main outcome variables (antibiotic, dose and use of investigations) are the same even though the sample size for the general hospital emergency department is smaller. Thus, there is little reason to suspect that the longer time frame used for the general hospital site influenced the data in favour of second-line antibiotics.
Documentation was complete for choice of antibiotic, which was the primary outcome of interest. However, incomplete documentation led to missing values for 23% of the amoxicillin dosage data and 34% of the duration of amoxicillin data. Thus, caution should be used when drawing conclusions regarding the outcomes in the three settings for these variables.
It seems unlikely that the difference in the use of second-line antibiotics was due to differences in antibiotic allergy rates across the settings because this was not statistically significant (P=0.4, Table 1). Furthermore, when cases with antibiotic allergy are excluded from the analysis, rates of amoxicillin use change only slightly to 71% at the paediatric after-hours clinic, 70% at the paediatric hospital emergency department and 58% at the general hospital emergency department (P<0.01).
Some level of second-line antibiotic use is expected for children at increased risk of a resistant infection such those who have persistent otitis media despite a recent course of amoxicillin or those who attend group daycare (9). Due to the retrospective nature of the study, we were unable to identify reasons for second-line antibiotic use. Although it is likely that the sickest children or children with complicated AOM sought care in an emergency department, it was not expected that the rates of second-line antibiotics would vary between two emergency departments.
Nonadherence to guidelines may be due to lack of awareness of the guideline, lack of familiarity and of agreement with the guideline, or the inertia of previous practice (19). In addition, as discussion with colleagues has been found to be a popular source of information for physicians (20), familiarity with a guideline may be influenced by access to experts. We speculate that these factors may have influenced practice differentially in the three settings that we studied.
In the present study, we found variation in the use of laboratory and radiological investigations. Only patients with the discharge diagnosis of AOM were included in the present study. Therefore, it is unlikely that the results of the investigations contributed to the patients’ ultimate management. Factors related to use of investigations may include access to the infrastructure required to perform investigations; the degree of patients’ ill appearance raising physician diagnostic uncertainty, which investigations may alleviate; and employment of clinically inexperienced medical trainees.
AOM in children is a common problem that frequently requires care in an after-hours setting. In the present study, we found variability in the management of AOM among three after-hours settings. The paediatric after-hours clinic provided a high level of adherence to an evidence-based clinical practice guideline and a low level of resource intensive investigations. We believe that the paediatric after-hours clinic may be an excellent venue for the after-hours treatment of paediatric AOM.
Acknowledgments
The Pediatric Outcomes Research Team (PORT) is supported by a grant from the Hospital for Sick Children Foundation, Toronto. Jonathon L Maguire was supported by a scholarship from the Jane and Howard O Jones Bursary, Faculty of Medicine, University of Toronto; Jane Healey was supported by a scholarship from the Robert and Annie MacDonell Bursary, Faculty of Medicine, University of Toronto. We gratefully acknowledge the summer medical students who assisted with data collection: Robin Shaefer and Alex Gubanov (supported by a summer student research scholarship from Bristol-Myers Squibb Canada Inc) and Diana Tamir (supported by a scholarship from the Robert and Annie MacDonell Bursary, Faculty of Medicine, University of Toronto).
REFERENCES
- 1.American Academy of Pediatrics Ad Hoc Task Force on Definition of the Medical Home:The medical home Pediatrics. 1992;90:774. [PubMed] [Google Scholar]
- 2.Patel H, Macarthur C, Feldman W. Where do children go? Comparing the after-hours availability of family physicians and primary care pediatricians in four Canadian cities. Can Fam Physician. 1997;43:1235–9. [PMC free article] [PubMed] [Google Scholar]
- 3.Merritt TA, Palmer D, Bergman DA, Shiono PH. Clinical practice guidelines in pediatric and newborn medicine: Implications for their use in practice. Pediatrics. 1997;99:100–14. doi: 10.1542/peds.99.1.100. [DOI] [PubMed] [Google Scholar]
- 4.Ellis J, Mulligan I, Rowe J, et al. Inpatient general medicine is evidence based. Lancet. 1995;346:407–10. [PubMed] [Google Scholar]
- 5.McCaig LF, Hughes JM. Trends in antimicrobial drug prescribing among office-based physicians in the United States. JAMA. 1995;273:214–9. [PubMed] [Google Scholar]
- 6.Paradise JL, Rockette HE, Colborn DK, et al. Otitis media in 2253 Pittsburgh-area infants: Prevalence and risk factors during the first two years of life. Pediatrics. 1997;99:318–33. doi: 10.1542/peds.99.3.318. [DOI] [PubMed] [Google Scholar]
- 7.Dagan R, Leibovitz E, Leiberman A, Yagupsky P. Clinical significance of antibiotic resistance in acute otitis media and implication of antibiotic treatment on carriage and spread of resistant organisms. Pediatr Infect Dis J. 2000;19(Suppl 5):S57–65. doi: 10.1097/00006454-200005001-00009. [DOI] [PubMed] [Google Scholar]
- 8.Rosenfeld RM, Vertrees JE, Carr J, et al. Clinical efficacy of antimicrobial drugs for acute otitis media: Metaanalysis of 5400 children from thirty-three randomized trials. J Pediatr. 1994;124:355–67. doi: 10.1016/s0022-3476(94)70356-6. [DOI] [PubMed] [Google Scholar]
- 9.Blumer JL. Fundamental basis for rational therapeutics in acute otitis media. Pediatr Infect Dis J. 1999;18:1130–40. doi: 10.1097/00006454-199912000-00028. [DOI] [PubMed] [Google Scholar]
- 10.Dowell SF, Marcy SM, Phillips WR, Gerber MA, Schwartz B. Principles of judicious use of antimicrobial agents for pediatric upper respiratory tract infections. Pediatrics. 1998;101:163–5. [Google Scholar]
- 11.Rosenfeld RM. An evidence-based approach to treating otitis media. Pediatr Clin North Am. 1996;43:1165–81. doi: 10.1016/s0031-3955(05)70512-5. [DOI] [PubMed] [Google Scholar]
- 12.McCracken GH. Considerations in selecting an antibiotic for treatment of acute otitis media. Pediatr Infect Dis J. 1994;13:1054–7. doi: 10.1097/00006454-199411000-00036. [DOI] [PubMed] [Google Scholar]
- 13.Barnett ED, Klein JO. The problem of resistant bacteria for the management of acute otitis media. Pediatr Clin North Am. 1995;42:509–17. doi: 10.1016/s0031-3955(16)38976-3. [DOI] [PubMed] [Google Scholar]
- 14.Klein JO. Review of consensus reports on management of acute otitis media. Pediatr Infect Dis J. 1999;18:1152–5. doi: 10.1097/00006454-199912000-00031. [DOI] [PubMed] [Google Scholar]
- 15.Bluestone CD. Clinical course, complications and sequelae of acute otitis media. Pediatr Infect Dis J. 2000;19(Suppl 5):S37–46. doi: 10.1097/00006454-200005001-00007. [DOI] [PubMed] [Google Scholar]
- 16.Glasziou PP, Del Mar CB, Sanders SL, et al. The Cochrane Library. Vol. 1. Oxford: Update Software; 2003. 2003. Antibiotics for acute otitis media in children (Cochrane Review) [DOI] [PubMed] [Google Scholar]
- 17.Canadian Paediatric Society Infectious Diseases and Immunization Committee. Antibiotic management of acute otitis media. Paediatr Child Health. 1998;3:265–7. doi: 10.1093/pch/3.4.265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Dowell SF, Butler JC, Giebink GS, et al. Acute otitis media: Management and surveillance in an era of pneumococcal resistance – a report from the Drug-resistant Streptococcus pneumoniae Therapeutic Working Group. Pediatr Infect Dis J. 1999;18:1–9. [PubMed] [Google Scholar]
- 19.Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA. 1999;282:1458–65. doi: 10.1001/jama.282.15.1458. [DOI] [PubMed] [Google Scholar]
- 20.Hayward RS, Guyatt GH, Moore KA, McKibbon KA, Carter AO. Canadian physicians’ attitudes about and preferences regarding clinical practice guidelines. CMAJ. 1997;156:1715–23. [PMC free article] [PubMed] [Google Scholar]