PEDIATRICS Vol. 105 No. 4 April 2000, pp. 753-759

Outcomes After Judicious Antibiotic Use for Respiratory Tract Infections Seen in a Private Pediatric Practice

Michael E. Pichichero, MD, John L. Green, MD, Anne B. Francis, MD, Steven M. Marsocci, MD, and Marie L. Murphy, MD

From the Elmwood Pediatric Group, Department of Microbiology/Immunology, University of Rochester Medical Center, Rochester, New York.

	ABSTRACT

Top Abstract Methods Results Discussion References

Background.  Most respiratory tract infections (RTIs) in children have a viral cause, they resolve on their own, and antibiotics need not be prescribed.

Objective.  We sought to provide evidence that judicious antibiotic use can be accomplished in private pediatric practice without observing an increase in return office visits or in the rate of bacterial infections that may follow.

Study Design.  This was a prospective 12-month study from July 1, 1996 through June 30, 1997. On the same 1 day each week, a representative convenience sample of acute respiratory tract illness patients was enrolled, and laboratory studies performed as appropriate, including viral cultures on all. Children were then followed for 30 days to ascertain the outcomes of not prescribing antibiotics except when specific bacterial infections were present at the initial visit.

Results.  Three hundred eighty-three children were enrolled; 293 (77%) did not receive antibiotics at the enrollment visit. Ninety children (23%) received antibiotics based on a diagnosis of acute otitis media (n = 53), acute streptococcal tonsillopharyngitis (n = 18), or other presumed or documented bacterial infections (n = 19). An unscheduled return visit related to the initial visit occurred for 86 (29%) of the 293 children not receiving antibiotics initially and in 40 (44%) of 90 children receiving antibiotics initially. Eighty-seven children (23%) had positive viral culture results. The most frequently isolated viruses were adenovirus, enterovirus, parainfluenzae virus, and influenza virus.

Conclusion.  Children with RTIs without a concomitant presumed or proven bacterial infection do not require antibiotics. In this busy office practice, >75% of the children presenting with an RTI did not have a presumed or proven bacterial infection. These children did not have a higher rate of return office visits or an increase in bacterial infections. This reinforces the judicious use of antibiotics in managing children with RTIs.outcomes, antibiotic, respiratory infections.

Most respiratory tract infections (RTIs) in children have a viral cause and resolve on their own.¹ Yet some physicians embrace the notion that presumptive antibiotic therapy will reduce return office visits and prevent subsequent bacterial infections. Others commonly prescribe antibiotics out of concern that parents will be dissatisfied and seek antibiotic treatment elsewhere if their expectations are not met.² Day care policies sometimes dictate that a child may not return to the care facility without an antibiotic if there is fever and/or green/yellow-colored rhinorrhea. Recovery from illness under the umbrella of antibiotic therapy leads both parent and physician to attribute improvement (or the absence of development of a secondary bacterial infection) to the antibiotic.³ This becomes a self-perpetuating cycle of expectation when the child experiences future similar infections.³

At the Elmwood Pediatric Group (EPG), we avoid prescribing antibiotics for RTIs with a presumed viral cause. In this article, we describe our office approach to the management with RTIs and a 1-year prospective collection of clinical information on 383 children from presentation with illness through follow-up for 30 days. We assess the consequences of our clinical decisions with respect to frequency of return office visits and occurrences of subsequent bacterial infections.

	METHODS

Top Abstract Methods Results Discussion References

Description of EPG Practice

EPG is a private pediatric practice located in the suburbs of Rochester, New York (greater metropolitan population of ~1 million). The group established by Burtis Breese and Frank Disney in 1938 has always maintained a rather strict policy of not prescribing antibiotics for RTIs presumed to be of viral cause. During the time frame of the current study, the group consisted of 9 board-certified pediatricians and 2 pediatric nurse practitioners, although only 4 to 5 providers worked on Tuesdays at our primary pediatric location (the day and site for this study). Our practice population is representative of the economic, racial, and ethnic diversity seen in suburban Rochester. Approximately 80% of our patient population participated in health maintenance organization (HMO) insurance plans, 12% in private insurance plans, and 8% in Medicaid programs. Urgent/emergency department care requires a referral for HMO patients and reports of the visits are routinely made to the primary care provider for all patients. We have always had a laboratory in our office practice and we did seek and receive Clinical Laboratories Improvement Act level III certification when that requirement became mandated by federal law. When seeing acutely ill children, each physician is scheduled to see 5 or 6 children per hour. Medical assistants and nurses perform necessary laboratory work, including cultures, white blood cell counts, and bedside cold agglutinins. Virtually all our patients have telephones and we have a long track record of good compliance for follow-up visits when needed. In addition to regular weekday hours, the office is open all Saturday, Sunday, and holiday mornings to see acutely ill children. At other times, the physicians of the practice share in after-hours and evening telephone call coverage of the practice. Antibiotics are not prescribed by telephone without first seeing the patient.

EPG Standard Practice for Management of Children With RTIs

All children with RTIs have a history taken and a physical examination performed by 1 of the practitioners in the group. Laboratory testing is readily available and used with varying frequency according to the clinical situation and inclination of the practitioner. All children with suspected group A -hemolytic streptococcal (GABHS) tonsillopharyngitis have either a rapid antigen detection test or throat culture obtained. Empiric antibiotics are not prescribed for sore throat patients. In toxic appearing children and those with moderate to high fever, a manual white blood cell count is usually performed with results available within 1 to 3 minutes. Additionally, a blood smear for white blood cell differential is made in selected children (<5% of those who have a total white blood cell count). A white blood cell count in the normal range is a piece of clinical information shared with parents when the diagnosis of viral RTI is made and it provides some reassurance to the physician and the parent that at the time of the visit, a bacterial infection is unlikely to be present and antibiotics are not necessary (Casey et al, unpublished data). Bedside cold agglutinins are obtained on selected children with cough, with more frequent use in patients for whom cough is the chief complaint, a harsh or productive cough is a clinical concern, or when cough has persisted for >1 week. The bedside cold agglutinin test is primarily used to provide some assurance to both physician and parents that a Mycoplasma pneumoniae infection is unlikely and antibiotics, therefore, are not justified.

Description of the Study

This was a prospective study that began July 1, 1996 and ended June 30, 1997. Data were collected on a single, presumed representative, day of the week (Tuesdays). We chose Tuesdays because this is the day when the EPG secures viral cultures on a sample of acute respiratory illness patients seen. We do this for the National Institute of Allergy and Infectious Disease under a subcontract as a sentinel surveillance center. Our subcontract obligation and study design called for enrolling a representative (convenience) sample of 5 to 10 children on that single day of the week, who had RTI symptoms and signs typical of what we were seeing that week. This included patients with probable viral RTIs, viral RTIs concurrently associated with presumed bacterial infections (eg, acute otitis media), and viral RTIs with other concurrent nonrespiratory illnesses (eg, impetigo). Not all physicians participated fully and not all enrolled an equal number of children in the study. A minimum of 5 children were not enrolled in 2 weeks during summer months because respiratory viral illnesses were at a seasonal low. After enrollment, routine follow-up visits were not scheduled. However, we prospectively tracked whether the patients returned within 1 month for any acute office visits; if a patient visit occurred that the parent, patient, or physician considered possibly related to the initial visit, then data regarding that visit were collected, including diagnosis and antibiotic prescribing. If a scheduled well-child visit occurred within a month of enrollment and a concurrent acute illness was occurring, then that visit and its outcome were assessed for this study.

Information collected on each patient included age, gender, presence or absence of temperature elevation, and the recording of specific symptoms presentincluding the following: fever, rhinorrhea, sore throat, cough, otalgia, headache, hoarseness, nausea, vomiting, diarrhea, and rash. Signs recorded included conjunctival erythema/exudate; tympanic membrane erythema, fullness, mobility, and middle ear effusion; pharyngeal erythema and/or exudate; cervical adenopathy; and chest rhonchi, wheezes, and/or rales. The physician diagnosis was recorded for the acute illness visit, and children were then prospectively followed for 1 month to ascertain whether a return visit occurred.

Viral cultures, obtained by nasopharyngeal and oropharyngeal swab, were sent to the University of Rochester viral identification laboratory of Dr Caroline Breese Hall and processed in a standard manner on Hep-2, Rhesus monkey, MRC-5 and LLC-Mk2 (summers), or MDCK (winters) cell lines. Cytopathic effect was assessed 12 to 14 days after inoculation, which may not be sufficient for isolating of rhinoviruses. Results of these viral cultures were later entered into a database for analysis.

Statistics

Treatment group demographics and outcomes were compared by Student's t test or ² analysis, as appropriate.

	RESULTS

Top Abstract Methods Results Discussion References

Of 1483 eligible children with RTIs seen on Tuesdays at the EPG, 383 were enrolled in this study; 293 (77%) did not receive antibiotics at their acute respiratory illness visit, whereas 90 (23%) did receive antibiotics because a concurrent presumed or documented bacterial infection was diagnosed (Fig 1). The mean (± standard deviation) age and gender mix of the 293 children not receiving antibiotics at the initial visit was 6.8 years (± 5.8 years; median age: 5 years) and 54.3% were males; for the 90 children receiving antibiotics because of concurrent bacterial RTIs the mean age was 4.3 years (± 4.0 years; median: 3 years) and 53.3% were males. Children receiving antibiotics for presumed or documented secondary infections were younger than the group not receiving antibiotic therapy (P = .001). The study children were generally representative of the overall patient population seen for acute illness visits during the time frame of the study; no significant differences in age of the study group and total practice population were observed (Fig 2). The distribution of diagnoses for the study population and the total practice population on Tuesdays for the study year were similar (Fig 3A and Fig 3B, respectively).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Outcomes for study population: proportion treated with antibiotics, subsequent acute illness visits within 1 month and antibiotic treatment decisions at that time. Eighty-six (29%) of 293 children who did not receive antibiotics at the initial acute respiratory illness visit returned within 1 month with another or persisting acute illness which was significantly less often than the 40 (44%) of 90 children who did receive antibiotics at the initial acute respiratory illness visit (P = .006).

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Month-to-month comparison of mean age of enrolled study patients and total population of children with acute illnesses seen at the Elmwood Pediatric Group, Rochester, NY, from July 1996 to June 1997.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Month-to-month distribution of respiratory illness diagnoses for the study population (A) and total population (B) of children seen at the Elmwood Pediatric Group, Rochester, NY, from July 1996 through June 1997.  = upper respiratory illness; = lower respiratory infection; = acute otitis media; = group A streptococcal pharyngitis; = upper or lower respiratory illness and nonrespiratory illness that may or may not have required antibiotic treatment.

The diagnoses rendered to the cohort of 90 children receiving antibiotics at the initial visit are shown in Table 1. Antibiotic selection was at the discretion of the prescribing practitioner; the dose was consistent with the manufacturers' recommendations of the individual products. Antibiotic treatment was generally recommended for 5 to 10 days, depending on the clinical condition.

A comparison of the frequency of subsequent visits within 1 month after diagnosis was made. A return visit within 1 month occurred for 126 (33%) of the 383 enrolled children. Not prescribing an antibiotic for a child with a viral RTI did not result in frequent return office visits related to that illness; 86 (29%) of 293 were seen for a return visit compared with 40 (44%) of 90 children in the group with a secondary bacterial infection receiving an antibiotic at the enrollment visit (P = .006). Two hundred seven (71%) of the 293 nonantibiotic-treated children recovered uneventfully as did 31 (62%) of the 50 antibiotic-treated children; 19 antibiotic-treated children had a return visit and required additional antibiotics for the same illness (within 1 month) attributable to an inadequate clinical response (Fig 1). Neither the occurrence or degree of fever nor any group of specific signs and symptoms at the initial visit identified which patients were destined to return for another visit and end up with an antibiotic.

The rate of bacterial infections occurring during a 1-month follow-up did not differ between the antibiotic-treated and nontreated groups. Seventy-three (57%) of 126 return visit children were prescribed an antibiotic as a consequence of varying diagnoses (Tables 2 and 3; Fig 1) and 53 (43%) were not prescribed an antibiotic at a return visit. Of the 73 children who received an antibiotic at a return visit because a presumed or documented bacterial infection occurred, 24 (33%) had received an antibiotic at the initial visit. In comparison, of the 53 children who did not receive an antibiotic at a return visit, 16 (31%) had received an antibiotic at the initial visit (P = not significant).

Eighty-seven (23%) of the 383 children enrolled in this study had positive viral culture results (Table 4). The most frequently isolated viruses were adenovirus, enterovirus, parainfluenzae serotype 3 virus, influenza B virus, and influenza A virus. Positive viral culture results were seen in 67 (23%) of 293 children in the nonantibiotic treatment and 20 (22%) of the 90 children in the antibiotic treatment group. Eighteen (21%) of 87 children with positive viral culture results were prescribed an antibiotic because of a presumed or documented concurrent bacterial infection (Table 5). Adenovirus, enterovirus, and influenza B virus accounted for two thirds of the viral isolations when a secondary presumed or documented bacterial infection was diagnosed.

	DISCUSSION

Top Abstract Methods Results Discussion References

Under the influence of misinformation and socioeconomic forces, presumptive antibiotic use for RTIs persists as a commonplace practice in pediatrics despite decades old evidence that no therapeutic benefit is derived for the patients and sometimes harmful consequences are produced.^4-12 In this study, we describe our practice characteristics, patient population served, and office policies for management. In a prospective study of a cohort of 383 children with RTIs, we did not prescribe an antibiotic for 77% of the children seen at the initial visit. We did prescribe an antibiotic for the remaining 23% of enrolled children who were diagnosed to have a viral RTI with a concurrent presumed or documented bacterial infection. Clearly, some of the children who received antibiotics may have been over-diagnosed but our avoidance of antibiotics in three fourths of children with RTIs is reasonably judicious although imperfect.

The EPG practice is typical in many ways of other practices in the United States. Physicians are of varying ages and experience, ancillary health care providers (nurse practitioners) provide care, a moderately diverse patient population is served, and patient encounters for acute illness are relatively brief. Aspects of the EPG practice that we find are essential to the judicious approach used in prescribing antibiotics include: 1) appropriate history and thorough physical examination, 2) availability and appropriate use of laboratory testing, 3) understandable and informative explanation of the diagnosis and its appropriate management, and 4) 365 days per year availability for follow-up reexamination and reconsideration of the diagnosis in the event that there is a change in the child's clinical condition.

The Rochester community has 2 large community-wide panels of physicians contracting with the 2 large HMOs and a switch between 1 of the 2 HMOs does not cause the need for a change in primary care physicians. Thus, there is very little insurance company switching and none to date that has caused patients a need to switch primary care physicians. This allows the traditional development of a trusting relationship between patient and physician; this trust is an essential component when diagnoses and management plans are offered to parents and patients. It has been an unfortunate consequence of Clinical Laboratories Improvement Act that so many physicians have reduced or entirely eliminated their office laboratory practices.¹³ We find the knowledge of a total white blood cell count, bedside cold agglutinin, or rapid streptococcal antigen detection test result is of considerable value to ourselves and our patients. Properly interpreted and communicated these tests can be reassuring to both physician and patient, particularly when no antibiotic is prescribed (J. Casey et al, unpublished data).

The study design did not allow us to answer the question of whether antibiotics prevented a progression from a presumed or documented viral RTI to a secondary presumed or documented bacterial complication and we were not able to study whether empiric antibiotics made the bacterial infections that did occur during the 1-month follow-up more difficult to manage attributable to a selection of resistant bacterial pathogens. However, the data suggest that avoidance of antibiotics during an initial visit for a presumed viral RTI does not result in an increased likelihood of a return visit for the same illness or more frequent visits for bacterial infections in the 4 weeks thereafter. In contrast, return visits occurred significantly more often in patients who received antibiotics at the initial visit compared with those who did not.

Physicians who use antibiotics for virtually all RTIs commonly offer several explanations for their actions.^1-3 There is, however, no scientific rationale for such antibiotic use.^4-1214-32 Increasing emergence of antibiotic-resistant bacteria has led to urgent pleas for judicious antibiotic prescribing.¹⁴ Knowledge about resistance does not translate into changed behavior. Schwartz et al³³ found that although 97% of pediatricians were concerned about antibiotic resistance, 53% would prescribe an antibiotic for an infant with scant purulent, green-colored nasal discharge of 1 day's duration with or without fever. To justify antibiotic use, physicians may make a different diagnosis, such as acute otitis media or sinusitis, when a patient has a viral respiratory infection.³⁴ Previous studies have shown that patients often have erroneous perceptions about antibiotics and indications for their use.^235-37 Some parents believe that acetaminophen is an antibiotic³⁷ or think that an antibiotic should be prescribed for all respiratory infections except a simple cold.³⁶ Others are increasingly concerned about excessive antibiotic usage and worry that their own child may receive too many antibiotics.² Parents seek antibiotics because they want recovery as quickly as possible for their ill child and they question whether a bacterial infection exists and whether an antibiotic will be helpful. The primary determinant of patient satisfaction is not the prescription of an antibiotic but rather a trusted physician's communication about the patient's illness.³⁸

Our study has several potential limitations. Patients enrolled represented a convenience sample. In the context of our National Institutes of Health contract and according to our study design, selected patients were representative of patients seen that week in the practice. They were of similar age and gender and the distribution of diagnoses were generally representative of our total practice population when assessed on a week to week basis. However, we did not collect information on race, ethnicity, socioeconomic status, or pattern and frequency of previous office visits, so we cannot comment on these potential variables. Second, the EPG practice policies and approach to RTIs have several important elements (which may not be generally true for other practices) that may have influenced the behavior of the physicians and patients, and therefore, the study results. These elements include: 1) a long-standing policy of judicious antibiotic use that has trained patient expectations; 2) availability of an in-office laboratory; and 3) 365 days per year physician access in the office for acute illnesses. Third, our study design does not allow us to reach a conclusion that initial avoidance of antibiotics does not affect the likelihood of a return visit for the same illness. The only way to study that question would be to compare groups who are and are not treated for similar infections, as done by Townsend and Radebaugh,⁸ who showed no difference in outcome.

	FOOTNOTES

Received for publication Mar 23, 1999; accepted Oct 28, 1999.

Reprint requests to (M.E.P.) University of Rochester Medical Center, Department of Microbiology/Immunology, 601 Elmwood Ave, Box 672, Rochester, NY 14642. E-mail: mepo{at}uhura.cc.rochester.edu

	ABBREVIATIONS

RTIs, respiratory tract infections; EPG, Elmwood Pediatric Group; HMO, health maintenance organization; GABHS, group A -hemolytic streptococcal.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Monto AS, Ullman BM Acute respiratory illness in an American community: the Tecumseh study. JAMA 1974; 227:164-169 [CrossRef][Medline]
2. Bauchner H, Pelton SI, Klein JO Parents, physicians, and antibiotic use. Pediatrics 1999; 103:395-398 [Abstract/Free Full Text]
3. Mainous AG, Zoorob RJ, Oler MJ, Haynes DM Patient knowledge of upper respiratory infections: implications for antibiotic expectations and unnecessary utilization. J Fam Pract 1997; 45:75-83 [Medline]
4. Jones PN, Bigham RS Jr, Manning PR Use of antibiotics in nonbacterial respiratory infections. JAMA 1953; 153:262-264
5. Cronk GA, Naumann DE, McDermott K, Menter P, Swift MD A controlled study of the effect of oral penicillin G in the treatment of non-specific upper respiratory infections. Am J Med 1954; 16:804-809 [CrossRef][Medline]
6. Hardy LM, Traisman HS Antibiotics and chemotherapeutic agents in the treatment of uncomplicated respiratory infections in children. J Pediatr 1956; 48:146-156 [CrossRef][Medline]
7. Petersdorf RG, Browder AA, Feinstein AR IV Protection against infection in susceptible individuals. J Pediatr 1961; 58:174-194 [CrossRef][Medline]
8. Townsend EH, Radebaugh JF Prevention of complications of respiratory illnesses in pediatric practice. N Engl J Med 1962; 266:683-689
9. Fraser PK, Hatch LA, Hughes KEA A comparison between aspirin and antibiotics in the treatment of minor respiratory infections. Lancet 1962; 1:614-616 [Medline]
10. Lexomboon U, Duangmani C, Kusalasai V, Sunakorn P, Olson LC, Noyes HE Evaluation of orally administered antibiotics for treatment of upper respiratory infections in Thai children. J Pediatr 1971; 78:772-778 [CrossRef][Medline]
11. Soyka LF, Robinson DS, Lachant N, Monaco J The misuse of antibiotics for treatment of upper respiratory tract infections in children. Pediatrics 1975; 55:552-556 [Abstract/Free Full Text]
12. Chretien JH, McGarvey M, deStwolinski A, Esswein JG Abuse of antibiotics: a study of patients attending a university clinic. Arch Intern Med 1975; 135:1063-1065 [Abstract]
13. Schwartz B, Fries S, Fitzgibbon AM, Lipman H Pediatricians' diagnostic approach to pharyngitis and impact of CLIA 1988 on office diagnostic tests. JAMA 1994; 271:234-238 [Abstract]
14. 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; 101:163-165 [Abstract/Free Full Text]
15. Rosenstein N, Phillips WR, Gerber MA, Marcy SM, Schwartz B, Dowell SF The common coldprinciples of judicious use of antimicrobial agents. Pediatrics 1998; 101:181-184 [Abstract/Free Full Text]
16. Janeway CA The use of sulfonamides in the treatment of respiratory infections in children. N Engl J Med 1943; 229:201-203
17. Hodges RG The use of sulfadiazine as a prophylactic against respiratory disease. N Engl J Med 1944; 231:817-820
18. Mack JK The prevention of respiratory infections in children including sulfonamide prophylaxis. Illinois Med J 1950; 98:224-230
19. McLane RA Clinical evaluation of combined drug therapy in acute upper respiratory infections. J Med Soc N J 1952; 49:509-510 [Medline]
20. Cronk G, Naumann DE, McDermott K, Menter P, Swift MB A controlled study of the effect of oral penicillin G on the treatment of non-specific upper respiratory infections. Am J Med 1954; 16:804-809
21. Finland M, Wells EB, Collins HS, Gocke TM Aureomycin in the treatment of influenza and certain other acute respiratory infections with and without pneumonia. Am J Med 1950; 8:21-30
22. Hoagland RJ, Deitz EN, Myers PW, Cosand HL Aureomycin in the treatment of the common cold. N Engl J Med 1950; 243:773-775
23. Jones PM, Bigham RS, Manning PR Use of antibiotics in nonbacterial respiratory infections. JAMA 1953; 153:262-264
24. Finland M Antibiotic treatment of acute respiratory infections and influenza. N Engl J Med 1952; 247:557-567
25. Taylor B, Abbott GD, McKerr M, Fergusson DM Amoxycillin and co-trimoxazole in presume viral respiratory infections of childhood: placebo-controlled trial. Br Med J 1977; 2:552-554
26. Howie JGR, Hutchinson KR Antibiotics and respiratory illness in general practice: prescribing policy and work load. Br Med J 1978; 2:1342
27. Stott NCH Management and outcome of winter upper respiratory tract infections in children aged 0-9 years. Br Med J 1979; 1:29-31
28. Gonzales R, Steiner JF, Sande MA Antibiotic prescribing for adults with colds, upper respiratory tract infections, and bronchitis by ambulatory care physicians. JAMA 1997; 278:901-904 [Abstract]
29. Mainous AG III, Hueston WJ, Clark JR Antibiotics and upper respiratory infection: do some folks think there is a cure for the common cold? J Fam Pract 1996; 42:357-361 [Medline]
30. Arnold KE, Leggiadro RJ, Breiman RF, Risk factors for carriage of drug-resistant Streptococcus pneumoniae among children in Memphis, TN. J Pediatr 1996; 128:757-764 [CrossRef][Medline]
31. Cowan PF Patient satisfaction with an office visit for the common cold. J Fam Pract 1987; 24:412-413 [Medline]
32. Del Mar C Managing sore throat: a literature review. II. Do antibiotics confer benefit? Med J Aust 1992; 156:644-649 [Medline]
33. Schwartz RH, Freij BJ, Ziai M, Sheridan MJ Antimicrobial prescribing for acute purulent rhinitis in children: a survey of pediatricians and family practitioners. Pediatr Infect Dis J 1997; 16:185-190 [CrossRef][Medline]
34. Vinson DC, Lutz LJ The effect of parental expectations on treatment of children with a cough: a report from ASPN. J Fam Pract 1993; 37:23-27 [Medline]
35. Hamm RM, Hicks RJ, Bemben DA Antibiotics and respiratory infections: do antibiotic prescriptions improve outcomes? J Okla State Med Assoc 1996; 89:267-274 [Medline]
36. Branthwaite A, Pechere JC Pan-European survey of patients' attitudes to antibiotics and antibiotic use. J Int Med Res 1996; 24:229-238 [Medline]
37. Kandakai TL, Price JH, Telljohann SK, Holiday-Goodman M Knowledge, beliefs, and use of prescribed antibiotic medications among low-socioeconomic African-Americans. J Natl Med Assoc 1996; 88:289-294 [Medline]
38. Barden LS, Dowell SF, Schwartz B, Lackey C Current attitudes regarding use of antimicrobial agents: results from physicians' and parents' focus group discussions. Clin Pediatr 1998; 37:665-672 [Abstract/Free Full Text]