PEDIATRICS Vol. 117 No. 4 April 2006, pp. 1009-1017 (doi:10.1542/10.1542/peds.2005-2172)
Effectiveness of Centers for Disease Control and Prevention Recommendations for Outcomes of Acute Otitis Media
a Department of Pediatrics, Boston University School of Medicine, Boston, Massachusetts
b Departments of Epidemiology and Statistics and Data Coordinating Center, Boston University School of Public Health, Boston, Massachusetts
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVES. To determine whether we could increase adherence to the Centers for Disease Control and Prevention (CDC) recommendations with well-accepted approaches to improving quality of care and adherence to the CDC recommendations resulted in improved outcomes for acute otitis media (AOM).
METHODS. A cluster randomization study was conducted in 12 pediatric practices (6 intervention and 6 control sites). The main outcome measures were adherence to the CDC recommendations (modified to include 2 additional antimicrobial agents) and a subsequent antibiotic prescription for AOM within 30 days after diagnosis.
RESULTS. Of 3152 patients referred to research assistants, 2584 (82%) were eligible. Of those eligible, 1368 (99%) of 1382 at the intervention sites and 1138 (99%) of 1146 at the control sites consented to participate. Rates of adherence to the CDC recommendations were not significantly higher at the intervention sites than at the control sites, for initial enrollment episodes (78.2% vs 70.6%) or second episodes (62.6% vs 59.9%). After controlling for clustering according to site and covariates, children who were not treated in adherence to the CDC recommendations for both episodes had 1.60 times the odds of a subsequent prescription within 12 days, compared with those treated in adherence at both episodes.
CONCLUSIONS. Despite using evidence-based approaches that are known to influence physician behavior, we were unable to increase adherence to the CDC recommendations for treatment of AOM. However, we did establish that prescription of antimicrobial therapy consistent with the CDC recommendations for a second episode of AOM was associated with improved outcomes, measured as the need for subsequent antibiotic prescription. Because of the selection of resistant otopathogens, adherence to the CDC recommendations is likely more important in subsequent episodes of AOM than in the initial episode.
Key Words: acute otitis media • treatment • Centers for Disease Control and Prevention • recommendations
Abbreviations: AOM—acute otitis media • CDC—Centers for Disease Control and Prevention
An increase in antimicrobial agent resistance among otopathogens has made the treatment of acute otitis media (AOM) more challenging.1,2 Throughout the 1990s, Streptococcus pneumoniae became increasingly resistant to multiple antibiotics, including amoxicillin, cephalosporins, macrolides, clindamycin, and trimethoprim/sulfamethoxazole.3,4 Because practitioners rarely culture specimens from the middle ear and thus have no knowledge of antimicrobial agent-resistant pathogens for individual patients, the presence of risk factors for a resistant middle ear pathogen, including young age, day care attendance, and previous antibiotic exposure, should be incorporated into selection of an antimicrobial agent for the treatment of AOM.5–7 Certain antibiotics are more likely to result in improved clinical outcomes because they are more effective in the eradication of middle ear pathogens, on the basis of direct evidence from human bacteriologic efficacy trials or, when such evidence is lacking, pharmacokinetic and pharmacodynamic properties of the antibiotic.8,9
In 1999, the Centers for Disease Control and Prevention (CDC) released recommendations for the use of specific antimicrobial agents for the treatment of AOM.9 These recommendations were reached through consensus and were based on data available at the time. In general, these recommendations advocated use of amoxicillin as first-line therapy for children at low risk for resistant otopathogens and high-dose amoxicillin, high-dose amoxicillin/clavulanate, or cefuroxime for children at higher risk. Antibiotic use in the previous 30 days was the principal risk factor for antibiotic-resistant pathogens in the CDC recommendations. Day care attendance was cited as a risk factor but was not incorporated into the decision scheme.
The purpose of this study was to determine whether we could increase adherence to the CDC recommendations by using well-accepted approaches known to improve physician behavior and whether adherence to the CDC recommendations resulted in improved outcomes for AOM. Because the microbiologic features of recurrent AOM may be altered after treatment of an initial episode,10,11 we hypothesized that adherence to the CDC recommendations for treatment of a second episode of AOM would result in reduced rates of treatment failure, measured as the need for subsequent antibiotic treatment within 12 days after diagnosis of a recurrent episode.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study Design
The study was a prospective trial that used cluster randomization of 12 pediatric practices. The practices were identified from a list of all pediatric practices within 30 miles of Boston (Table 1). Letters and/or e-mail messages were sent to all practices whose volume of antimicrobial agent use in the previous year was above the median for the greater Boston area. The practices were asked to participate in a study involving the treatment of young children with AOM. Sites that expressed interest were visited by one of the investigators (H.B.), and a more-complete explanation of the study was provided. The first 12 sites that consented to participate were included in the study. The sites were unaware of their status as an intervention site or a control site. At no time during the study were the practices aware of the primary goals. After each site agreed to participate, the site was visited by one of the investigators (S.P. or C.D.M.), who delivered the educational intervention.
|
Intervention
The intervention was based on educational strategies that have been associated with changing physician behavior, including reverse academic detailing (use of local opinion leaders who meet with clinicians in small-group discussions), interactive educational sessions, feedback, and enabling tools (physician reminders).12,13 Each practice received 2 presentations. The first presentation was given within 3 months after study enrollment. The second presentation was given before the 2003 to 2004 respiratory illness season. Patients were enrolled between January 2002 and May 2004. At the intervention sites, the presentations were followed by discussion and included information about the CDC recommendations, the rationale for categorizing patients into high- and low-risk groups, and the use of specific antibiotics. At both intervention and control sites, the presentation emphasized the appropriate diagnosis of AOM. At the control sites, data on pneumococcal conjugate vaccine were presented but antibiotic therapy and the CDC recommendations were not discussed. In addition, at intervention sites, a reminder was placed in the chart of every child who was eligible for participation in the study (Fig 1), indicating the need to categorize the child as being at high or low risk and the recommended antibiotics, and a newsletter was distributed to all physicians 5 times during the course of the study. The newsletter included information about the distribution of antibiotic use among the intervention sites (not specific to any one site or physician) and recommended specific antibiotics, depending on the child's risk factors for resistant otopathogens.
|
Modified CDC Recommendations
At the beginning of the study, the 1999 CDC recommendations for the treatment of AOM among children were modified to incorporate day care attendance as a risk factor for antibiotic-resistant otopathogens and to allow the use of cephalosporins with antibacterial activity similar to that of cefuroxime, on the basis of information available at that time about the bacteriologic efficacy and pharmacokinetic and pharmacodynamic properties of available antibiotics (Table 2). 14
|
Eligibility
Children (and their families) were considered eligible if they were between 3 and 36 months of age, had a clinical diagnosis of AOM made by a health care provider, had no underlying major medical illness, had no history of premature birth, were not allergic to either amoxicillin or penicillin, and provided informed consent.
Enrollment and Follow-Up Monitoring
When a patient was diagnosed as having AOM and a provider thought that the patient was potentially eligible, the subject was referred to a research assistant. Research assistants were assigned to each practice for 35 hours per week. The research assistants reviewed the eligibility criteria and enrolled appropriate subjects. At that time, in addition to sociodemographic information, the research assistant gave the family a daily symptom diary to complete and arranged to contact the family 3 to 5 days and 9 to 11 days later. The family was instructed to complete the diary until the child was free of symptoms and to return it by mail. The research assistant contacted the family twice during the subsequent 2 weeks and obtained additional data about the child's well-being. This interview focused on resolution of symptoms and prescriptions for additional antibiotics. If a subject was diagnosed as having recurrent AOM, then the clinicians were to refer the patient back to the research assistant, so that a new diary could be initiated. Patients were again called 3 to 5 days and 9 to 11 days later.
Patients were monitored for 60 days; the research assistants reviewed the medical records of each patient for 60 days after enrollment, to determine whether the child received any subsequent antibiotic treatment. The chart review was performed to ensure that we had complete follow-up data for all enrolled patients.
Statistical Analyses
Intervention and control practices were compared with respect to patient demographic features with 
2 tests. To account for our cluster-sample design, study groups were compared with respect to outcomes with generalized estimating equation logistic regression models, which account for clustering of patients within practices in estimations of both effects and SEs. Patient characteristics that were found to differ between study groups were controlled for as covariates in these analyses. For our first goal, we compared study groups with respect to treatment adherence to the CDC recommendations, for both the enrollment episode of AOM and second episodes of AOM during the 60-day study period. For our second goal, we compared those treated in adherence versus not in adherence to the CDC recommendations with respect to the need for a second antibiotic prescription for an episode of AOM. For enrollment episodes, we analyzed prescriptions by both day 12 and day 17. For second episodes, we categorized patients according to adherence to CDC guidelines during both the enrollment episode and the second episode and compared those treated in adherence to the guidelines during both episodes with those treated in other combinations. These analyses were also through generalized estimating equation logistic regression, controlling for patient characteristics found to differ between study groups. We also described the time to a second antibiotic prescription for second treatment episodes with Kaplan-Meier survival curves, although this analysis does not control for clustering according to practice.
The study was approved by the institutional review board of Boston University Medical Center. Oral consent was obtained from all participants.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The clinicians referred 3152 patients to the research assistants; 2584 (82%) were eligible and 568 (18%) were ineligible. Of those who were eligible, 1368 (99%) of 1382 at the intervention sites and 1138 (99%) of 1146 at the control sites consented to participate. Twenty-two additional patients were enrolled but were discovered subsequently to be ineligible because of penicillin allergy (n = 15), premature birth (n = 4), or other reasons (n = 3). A review by the research assistants of data for all patients seen in each practice during a 2-week period indicated that 40% of all eligible patients were enrolled. This approximates the proportion of office time, compared with the total number of office hours, that a research assistant was present at each practice. Patients who were not enrolled were similar to those who were enrolled with respect to prior antibiotic use and distribution of antibiotics prescribed for the current AOM episode (data not shown).
A majority of the children enrolled were male, were white, and had private insurance (Table 3). The mean age was 1.41 years for the intervention sites and 1.37 years for the control sites. Approximately one half of the children were in day care. The majority of primary caregivers were the mothers, were married, and had more than a high school education (Table 3). The following variables differed significantly between the control and intervention sites: race, Hispanic origin, child feeding history, number of smokers in the household, insurance status, race of primary caregiver, age of primary caregiver, marital status, educational status of primary caregiver, and family income. We controlled for these differences in our analyses.
|
The distribution and prevalence of antibiotic use in the enrollment episodes (n = 2528) and in second episodes (n = 919) occurring >12 days after the enrollment episode are presented in Table 4. Consistent in part with the CDC recommendations, there was a difference in antibiotic use for enrollment and recurrent episodes, with decreases in the use of low-dose amoxicillin (from 16.6% to 5.3%) and high-dose amoxicillin (from 49.1% to 18.2%) and increases in the use of amoxicillin/clavulanate (from 5.3% to 15.2%) and azithromycin (from 7.2% to 14.8%).
|
Adherence to the CDC recommendations was higher at the intervention sites than at the control sites for initial enrollment episodes (78.2% vs 70.6%), but this difference did not reach statistical significance. No difference in adherence to the recommendations was seen for second episodes (62.6% vs 59.9%). These results were adjusted for cluster randomization and a number of covariates (Table 5).
|
Intervention and control sites did not differ with respect to the use of a second antibiotic prescription during the enrollment episode. For patients with second episodes of AOM during the study period, 9.8% of those treated in adherence to CDC recommendations during both study episodes received a subsequent antibiotic prescription within 12 days, compared with 13.5% of other patients (Table 6). After controlling for clustering according to site and covariates, those not treated in adherence to the guidelines during both episodes had 1.60 times the odds of a subsequent prescription within 12 days, compared with those treated in adherence during both episodes (Table 7). For the 919 patients with a subsequent episode of AOM, Fig 2 shows that those treated in adherence to the recommendations during both study episodes had a longer time to a subsequent antibiotic prescription than did those not treated according to the recommendations during the first episode, the second episode, or both (P = .046, log-rank test).
|
|
|
Because data that became available after this study was performed suggested that cefdinir might not provide the level of efficacy against penicillin-resistant S pneumoniae and Haemophilus influenzae thought previously,14 the analysis was repeated with cefdinir being coded as an antibiotic not consistent with the CDC recommendations. None of the results differed in the reanalysis (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study had 2 goals, that is, to increase adherence to the CDC recommendations for the treatment of AOM and to determine whether adherence to the CDC recommendations improved patient outcomes, measured as the need for additional antibiotics during the same episode of AOM. Despite the use of well-recognized and evidence-based approaches that change physician behavior, we were unable to increase adherence to the CDC recommendations; antibiotics inconsistent with the CDC recommendations (including modifications) were used in 11.5% of enrollment episodes and almost 31% of second episodes (macrolides, cefprozil, cefixime, trimethoprim/sulfamethoxazole, or low-dose amoxicillin). The increase in nonrecommended use represents specifically a doubling of azithromycin and cefprozil use; this suggests that some physicians are unaware of the lack of in vitro activity of these antimicrobial agents against multidrug-resistant S pneumoniae and most, if not all, H influenzae strains and are unaware that recurrent episodes are more likely to be associated with resistant otopathogens, 2 important premises of the CDC recommendations. This may become more critical in the era of universal immunization. Two publications reported that ß-lactamase-producing, nontypable H influenzae is the most common pathogen recovered among children who experience failure of initial therapy with amoxicillin.15,5
We did establish that prescription of antimicrobial therapy consistent with the modified CDC recommendations for a second episode was associated with improved outcomes, measured as the need for a subsequent antibiotic prescription. This finding, although expected from a microbiologic and pharmacologic perspective, is unusual from a methodologic standpoint. It is difficult to demonstrate the effects of antibiotic therapy in AOM by using clinical end points.16 AOM is a complex disease, often of combined bacterial and viral etiology. Improvement and persistence of symptoms do not have clear relationships with the ability of an antibiotic to eliminate pathogens from the middle ear.17,18 In addition, treatment for AOM needs to be understood in the context of local patterns of antibiotic resistance. A report from Massachusetts indicated that, in the post-heptavalent pneumococcal conjugate vaccine era, 33% of 190 nasopharyngeal isolates were not susceptible to penicillin.19 In our trial in a real-world practice setting, we used a repeat antibiotic prescription as a surrogate measure of a poor clinical outcome, and we made no attempt, except for academic detailing, to ensure an accurate diagnosis of AOM. These factors conspire against demonstrating an effect of less-optimal antibiotic therapy, but we demonstrated an effect by performing a large trial with >900 second episodes of AOM. There has been only 1 previous clinical trial of AOM with a sample size of similar magnitude.20
Adherence to the CDC recommendations for a recurrent episode of AOM reduces the need for subsequent antibiotics.21 Consistent with the hypothesis of the CDC working group that formulated the recommendations, we hypothesized that recurrences that occurred within 30 days after the enrollment episode would be more likely attributable to resistant otopathogens. Dagan et al21 demonstrated that such recurrent episodes are likely attributable to otopathogens that remain in the nasopharynx during treatment for the initial episode or are newly acquired shortly thereafter and have an increased likelihood of antimicrobial agent resistance. Consistent with our hypothesis, we found that using certain antibiotics, particularly after a second episode, reduced significantly the need for a third antibiotic within either a 12-day or 17-day window. Following the CDC recommendations for both the initial and subsequent episodes resulted in a 34% reduction in the need for an additional antibiotic. We evaluated outcomes at both 12 and 17 days after initiation of antimicrobial therapy. Twelve days has been a traditional time for assessing outcomes in clinical trials of AOM. We used a 17-day (10 days of traditional therapy and 7 additional days) outcome because Leibovitz et al22 demonstrated that recurrent episodes within 7 days after completion of antimicrobial therapy were more likely to be attributable to relapse (the same pathogen) than were recurrences after 7 days. We acknowledge that there is uncertainty about the appropriate time point to use for defining outcomes; therefore, we also presented the data as a survival plot. The difference remains significant for the 30-day period after a second AOM episode.
Our inability to increase adherence to the CDC recommendations is surprising. We used a multifaceted approach that included reverse academic detailing with local opinion leaders, 2 of the authors visited the practices twice during the study, the intervention practices were given feedback about how their sites were performing with respect to the choice of specific antibiotics (although, admittedly, this did not include either site-specific or physician-specific information), and, finally, reminders were placed on the charts of every eligible patient and included in the newsletter. There are numerous potential explanations for our failure to have an impact on practices. First, it remains difficult to change physician behavior. This may be particularly true for specific antibiotics for treatment of AOM in the ambulatory care setting, because physicians may think that most of the available antibiotics are efficacious in AOM and may not recognize that some of the available antibiotics are not efficacious for some otopathogens. Second, we might have underestimated the extent to which parents influence the choice of specific antibiotics, as a result of their prior personal experience. Our study population was drawn largely from a population of middle or high socioeconomic status, and it is possible that parents in this group exert greater influence on physicians than do other parents. Interestingly, children of higher socioeconomic status are at greater risk for resistant S pneumoniae. Third, rates of adherence to the CDC recommendations were relatively high for the control practices, ie, 70% for the initial episode and 60% for the second episode. This made it more difficult to find a difference between the intervention and control sites. However, a doubling of the use of azithromycin and cefprozil for high-risk episodes is inconsistent with available evidence, because both are likely to be most effective when the risk for multidrug-resistant S pneumoniae or H influenzae is low. Multidrug-resistant S pneumoniae and H influenzae have been found to be increased in recurrent episodes.23
The challenge of conducting AOM research in clinical practices involves the validity of the diagnosis and an understanding of how patterns of antibiotic resistance affect the choice of antimicrobial agent. We did not use any objective test, such as tympanometry, that might have assisted practitioners in making more-accurate diagnoses of AOM; instead, we relied on their skills. This is a limitation of the study. However, the project was designed specifically to address the effectiveness of the CDC recommendations in the office practice setting, rather than in the context of carefully controlled clinical trials. Even with optimal diagnostic conditions, 20% of patients do not have bacterial infection of the middle ear. Under the diagnostic conditions of routine clinical care, many patients have otitis media with effusion rather than AOM or may not even have middle ear fluid and infection. Diagnostic error rates for AOM in clinical practice are not known but are thought to be substantial.24 However, the failure to make a proper diagnosis of AOM would only dilute any potential to observe differences (presuming that such failure is equivalent to physicians not following the recommendations) and would not increase artificially the benefit of prescribing medications in accordance with the recommendations. Therefore, for children with true bacterial AOM, the benefits of treatment in accordance with the recommendations are likely to be even greater than we observed.
This study demonstrates that, for episodes of AOM at high risk for resistant pathogens, the selection of certain antimicrobial agents results in improved outcomes, measured as the need for subsequent antibiotics for treatment of AOM within 30 days. We are unaware of other practice-based data that support the importance of adherence to the CDC AOM recommendations.
|
ACKNOWLEDGMENTS |
|---|
Support for this investigator-initiated project was provided by Glaxo-Smith-Kline. The sponsor did not participate in study design, execution, or analysis or manuscript preparation.
The Boston-Based Pediatric Research Group includes Concord Hillside Medical Associates (Concord, MA), Framingham Pediatrics (Framingham, MA), Holliston Pediatric Group (Milford, MA), Hyde Park Pediatrics (Hyde Park, MA), Lahey Clinic-Pediatric and Adolescent Practice (Burlington, MA), Lexington Pediatrics (Lexington, MA), Main Street Pediatrics (Hopkinton, MA), Milford Holliston Pediatric Group (Milford, MA), Newton Center Pediatrics (Newton, MA), Patriot Pediatrics (Bedford, MA), Pediatric Associates of Medford (Medford, MA), and Pediatric Association of Northwoods (Taunton, MA).
We thank Paul Little, Eugene Shapiro, and Bruce Fireman, who were on our scientific oversight committee. They provided insight into the design and execution of the study and reviewed the manuscript before submission.
|
FOOTNOTES |
|---|
Accepted Oct 4, 2005.
Address correspondence to Howard Bauchner, MD, Boston Medical Center, 91 East Concord St, Boston, MA 02118. E-mail: howard.bauchner{at}bmc.org
Financial Disclosure: Dr Marchant is a consultant and serves on the speaker bureaus for Glaxo-Smith-Kline and Abbott; Dr Pelton serves on advisory boards for Glaxo-Smith-Kline and Sanofi Pasteur.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Pichichero ME. Acute otitis media, II: treatment in an era of increasing antibiotic resistance. Am Fam Physician. 2002;6 :2410 –2416
- Dowell SF. Acute otitis media caused by resistant pneumococci. Am Fam Physician. 2000;61 :318 –323[Web of Science][Medline]
- Paradise JL. A 15-month-old child with recurrent otitis media.
JAMA. 2002;288
:2589
–2598
[Free Full Text] - Jacobs MR, Dagan R, Appelbaum PC, Burch DJ. Prevalence of antimicrobial-resistant pathogens in middle ear fluid: multinational study of 917 children with acute otitis media.
Antimicrob Agents Chemother. 1998;42
:589
–595
[Abstract/Free Full Text] - Casey JR, Pichichero M. Changes in frequency and pathogens causing acute otitis media in 1995–2003. Pediatr Infect Dis J. 2004;23 :824 –828[Web of Science][Medline]
- Jacobs MR. Worldwide trends in antimicrobial resistance among common respiratory tract pathogens in children. Pediatr Infect Dis J. 2003;22(suppl) :S109 –S119[CrossRef][Web of Science][Medline]
- Uhari M, Mantysaari K, Niemela M. A meta-analytic review of the risk factors for acute otitis media. Clin Infect Dis. 1996;22 :1079 –1083[Web of Science][Medline]
- Froom J, Culpepper L, Green LA, et al. A cross-national study of acute otitis media: risk factors, severity, and treatment at initial visit: report from the International Primary Care Network (IPCN) and the Ambulatory Sentinel Practice Network (ASPN). J Am Board Fam Pract. 2001;14 :406 –417
- Dowell SF, Butler JC, Giebin GS, et al. Acute otitis media: management and surveillance in the era of pneumococcal resistance: a report from the Drug-Resistant Streptococcus pneumoniae Therapeutic Working Group. Pediatr Infect Dis J. 1999;18 :1 –9[CrossRef][Web of Science][Medline]
- Craig WA, Andes D. Pharmacokinetics and pharmacodynamics of antibiotics in otitis media. Pediatr Infect Dis J. 1996;15 :255 –259[CrossRef][Web of Science][Medline]
- Kaplan SL, Mason EO. Mechanisms of pneumococcal antibiotic resistance and treatment of pneumococcal infections in 2002. Pediatr Ann. 2002;31 :250 –260[Web of Science][Medline]
- Mazmanian PE, David DA. Continuing medical education and the physician as a learner: guide to the evidence.
JAMA. 2002;288
:1057
–1060
[Free Full Text] - Davis DA, Thomson MA, Oxman AD, Haynes RB. Changing physician performance: a systematic review of the effect of continuing medical education strategies.
JAMA. 1995;274
:700
–705
[Abstract/Free Full Text] - Adler M, McDonald PJ, Trostmann U, Keyserling C, Tack K. Cefdinir vs. amoxicillin/clavulanic acid in the treatment of suppurative acute otitis media in children. Pediatr Infect Dis J. 2000;19(suppl) :S166 –S170[CrossRef][Web of Science][Medline]
- Block ST, Hedrick JA, Tyler RD, et al. Microbiology of acute otitis media recently treated with aminopenicillins. Pediatr Infect Dis J. 2001;20 :1017 –1021[CrossRef][Web of Science][Medline]
- Marchant CD, Carline SA, Johnson CE, Shurin PA. Measuring the comparative efficacy of antibacterial agents for acute otitis media: "the Pollyanna Phenomenon. " J Pediatr. 1992;120 :72 –77[CrossRef][Web of Science][Medline]
- Carlin SA, Marchant CD, Shurin PA, et al. Host factors and early therapeutic response in acute otitis media: does symptomatic response correlate with bacterial outcome? J Pediatr. 1991;118 :178 –183[CrossRef][Web of Science][Medline]
- Dagan R, Leibovitz E, Greenberg D, et al. Early eradication of pathogens from middle ear fluid during antibiotic treatment of acute otitis media is associated with improved clinical outcome. Pediatr Infect Dis J. 1998;17 :776 –782[CrossRef][Web of Science][Medline]
- Finkelstein JA, Huang SS, Daniel J, et al. Antibiotic-resistant Streptococcus pneumoniae in the heptavalent pneumococcal conjugate vaccine era: predictors of carriage in a multicommunity sample.
Pediatrics. 2003;112
:862
–869
[Abstract/Free Full Text] - Kaleida PH, Casselbrant ML, Rockette HE, et al. Amoxicillin or myringotomy or both for acute otitis media: results of a randomized clinical trial.
Pediatrics. 1991;87
:466
–474
[Abstract/Free Full Text] - Dagan R, Leibovitz E, Cheletz G, et al. Antibiotic treatment in acute otitis media promotes superinfection with resistant Streptococcus pneumoniae carried before initiation of treatment. J Infect Dis. 2001;183 :880 –886[CrossRef][Web of Science][Medline]
- Leibovitz E, Greenber D, Piglansky L, et al. Recurrent acute otitis media occurring within one month from completion of antibiotic therapy: relationship to the original pathogen. Pediatr Infect Dis J. 2003;22 :209 –215[CrossRef][Web of Science][Medline]
- Quach C, Collet JP, Lelorier J. Effectiveness of amoxicillin, azithromycin, cefprozil and clarithromycin in the treatment of acute otitis media in children: a population-based study. Pharmacoepidemiol Drug Saf. 2005;14 :163 –170[CrossRef][Web of Science][Medline]
- Rothman R, Owens T, Simel DL. Does this child have acute otitis media?
JAMA. 2003;290
:1633
–1640
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Ostini, D. Hegney, C. Jackson, M. Williamson, J. M Mackson, K. Gurman, W. Hall, and S. E Tett Systematic Review of Interventions to Improve Prescribing Ann. Pharmacother., March 1, 2009; 43(3): 502 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Berman Otitis Media, Shared Decision Making, and Enhancing Value in Pediatric Practice Arch Pediatr Adolesc Med, February 1, 2008; 162(2): 186 - 188. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
