original research

Treatment of Acute Bronchitis in Adults A National Survey of Family Physicians Kevin C. Oeffinger, MD; Laura M. Snell, MPH; Barbara M. Foster, PhD; Kevin G. Panico; and Richard K. Archer Dallas and Temple, Texas

BACKGROUND. The purpose of this study was to determine how family physicians in the United States treat acute bronchitis in an otherwise healthy adult.

METHODS. A 33-item questionnaire on the diagnosis and treatment of acute bronchitis was mailed to a random sample of 500 physicians who are members of the American Board of Family Practice.

RESULTS. Thirty-two of the 500 sampled physicians could not be located by mail; 265 of those who received the questionnaire responded. The response rate was 57% (265/468). Sixty-three percent of responding physicians indi­ cated that antibiotics are their first choice of treatment for the otherwise healthy, nonsmoking adult with acute bron­ chitis. The decision to use antibiotics as the first choice of treatment did not vary by physician’s sex, age, years in practice, practice location, practice type, or percentage of HMO patients. Only 6% of responding physicians reported using p2 agonist bronchodilators as their first choice of treatment. Physicians in this study stated that they prescribe an antibiotic 75% of the time in treating nonsmoking patients with acute bronchitis (first choice or otherwise). If the patient is a smoker, physicians reported that they prescribe antibiotics 90% of the time (F=110.25; df= 1; P >.0001). Physicians reported that for patients who smoke it takes longer for coughs to totally resolve and longer for them to return to a normal activity level than for nonsmokers.

CONCLUSIONS. Family physicians report that antibiotics are their most common treatment for acute bronchitis in the otherwise healthy adult. Previous clinical trials have shown only marginal improvement in symptoms when patients with this condition are treated with an antibiotic. With antibiotic resistance emerging as a major global health problem, it is essential that other methods of treatment be evaluated.

KEY WORDS. Acute bronchitis; antibiotics; family practice; survey; physicians, family. (J Fam Pract 1998; 46:469-475)

cute bronchitis in the otherwise healthy been implicated in the etiology o f acute bronchitis, adult is one o f the most common med­ including rhinovirus, respiratory syncytial virus, ical problems encountered in primary parainfluenza virus, influenza virus, adenovirus, and care. It is a self-limiting, generally virai- enterovirus. Chlamydia pneumoniae (TWAR) and mediated clinical syndrome character­ Mycoplasma pneumoniae account for approxi­ Aized by cough and sputum production and accom­mately 10% to 15% of cases.2'5 Other bacterial path­ panied by upper respiratory tract and systemic man­ ogens are uncommonly isolated. ifestations o f infection. The duration o f this condi­ Tire treatment of acute bronchitis in the other­ tion is variable, with cough lasting more than 3 wise healthy adult remains controversial. Primary weeks in 50% o f patients and more than 4 weeks in care and respiratory medicine textbooks recom­ 25% of patients.1 mend providing symptomatic relief, only prescrib­ The prevalence of acute bronchitis peaks in the ing antibiotics in severe cases.'1-11 Clinical trials sug­ winter and is much less common in the summer. gest that antibiotics provide only marginal relief of Viral infection is generally considered to trigger the symptoms11218 and that P2 agonist bronchodilators majority o f episodes. A wide variety o f viruses have may be a more effective treatment.19 22 General prac­ titioners surveyed in Australia and the Netherlands Submitted, revised, January 19, 1998. report that they use an antibiotic more than half From the Department of Family Practice and Community the time in treating patients with acute bronchi­ Medicine, University of Texas Southwestern Medical Center at Dallas (K.C.O. L.M.S., K.G.P., R.K.A.) and DataSense, tis.23'24 Physicians in Kentucky treating Medicaid Temple, Texas (B.M.F.). Requests fo r reprints should be patients with acute bronchitis also use antibiotics addressed to K evin C. Oeffinger, MD, University o f Texas more than half of the time.25 Southwestern Medical Center, Department o f Fam ily Practice and Community Medicine, 5323 Harry Hines The purpose o f this study was to determine how Blvd, Dallas, TX 75235-9067. family physicians in the United States treat acute

© 1998 Appleton & Lange/ISSN 0094-3509 The Journal o f Family Practice, Vol. 46, No. 6 (June), 1998 469 TREATMENT OF ACUTE BRONCHITIS IN ADULTS

bronchitis in the otherwise healthy adult. Only the and ordinal variables. Years o f practice was a con­ results related to the treatment o f acute bronchitis tinuous variable and the standard measures of cen­ are included in this article; findings related to diag­ tral tendency and dispersion were computed. The nosis are reported elsewhere.26 demographic variables were most often used as grouping variables. Responses with either nominal METHODS or ordinal variables were compared with chi-square tests. Other tests used in this situation were the DIP A questionnaire concerning the diagnosis and treat­ and the Gradient in Proportions Test.30 In all cases ment of acute bronchitis was mailed to 500 family the P value was adjusted for multiple comparisons physicians in January 1996. The list o f physicians by the conservative Bonferroni adjustment. Some was generated by selecting a random sample from responses were continuous variables, but there was the 1995 Directory of the American Board of Family no statistical evidence that these variables were nor­ Practice.27 Before the first mailing, the survey instru­ mally distributed. Descriptive correlation coeffi­ ment was pilot-tested for clarity, usefulness, and face cients were obtained for certain pairs of variables. validity. A cover letter explaining the study and a pre­ stamped return envelope were included with the RESULTS questionnaire. Second and third mailings were sent to nonresponders. Thirty-two physicians could not be located by mail; The 33-question survey instrument consisted of 265 o f those who received the questionnaire six demographic questions, 13 questions concerning responded. The response rate was 57% (265/468). diagnosis, and 14 questions about treatment o f acute Data from physicians who were either retired or bronchitis in an otherwise healthy adult. This article practicing in another specialty (n=10) were not focuses on the findings related to the treatment ques­ included in the analysis. Responses o f the remaining tions. Physicians were asked seven questions regard­ 255 physicians were analyzed, for a net response rate ing their standard treatment of acute bronchitis in a of 56% (255/458). Not all physicians responded to nonpregnant, nonsmoking patient with no evidence each question; thus, the number (n) varies. o f sinusitis, systemic symptoms, or history o f asthma Responding physicians have been in practice an or chronic obstructive pulmonary disease. Three of average of 12.9 years (standard deviation = ±1.1). the questions were repeated for a similar patient Sixteen percent o f respondents are female, 68% who smoked. Physicians were also asked how long male, and 16% did not answer the question. Fifty-one it generally takes for cough to totally resolve and for percent of respondents reported that they practice in the resumption of normal activities in the average an urban setting, and 44% described their practice nonsmoking patient with acute bronchitis and no location as rural. Physicians’ age, practice type, and other complicating factors. The same question was the percentage of the practice composed of health asked with regard to the patient who smokes. maintenance organization (HMO) patients are dis­ For this study, a difference of .10 between two played in Table 1. proportions was considered to be clinically signifi­ Sixty-three percent of physicians stated that cant. This size difference translates into an effect antibiotics are their first choice of treatment for size o f h =.20 for the Difference in Proportions Test the nonsmoking patient. Only 6% chose P2 agonist (DIP).2829 A sample size of 350 would provide a bronchodilators as their first choice (Table 2). power o f .84. Since most variables were categorical, There were no statistically significant differences the chi-square test was expected to be the most fre­ in the selection o f antibiotics as the first choice of quently used statistical test. For chi-square tests o f 9 treatment by age or sex of the physician, years in df, a sample size of 180 provides a power of .82 to practice, practice type or location, or percentage detect a medium effect size, w = .30.29 On the basis of o f HMO patients. this information, a sample size o f 500 was selected. Surveyed physicians reported that they prescribe an antibiotic 75% of the time (first choice or other­ A nalysis wise) in treating nonsmoking patients with acute Descriptive statistics on the demographic variables bronchitis. If the patient smokes, physicians report­ consisted of counts and percentages for the nominal ed that they prescribe antibiotics 90% o f the time

470 The Journal o f Family Practice, Vol. 46, No. 6 (June), 1998 TREATMENT OF ACUTE BRONCHITIS IN ADULTS

TABLE 1 ------TABLE 2 Demographics of Family Physicians Responding to a Physicians’ First Choice of Treatment for a Nonsmoking Questionnaire Regarding the Treatment of Acute Adult Patient with Acute Bronchitis Bronchitis Treatment Choice* Number (%) Characteristic No. (%) Antibiotic 161 (63) Age, y Prescription cough suppressant <35 28 (11) or expectorant 37 (15) 36 to 45 140(56) Over-the-counter symptomatic relief 46 to 55 59 (23) medications 29 (11) >56 26 (10) p2 agonist bronchodilators 15(6) Other 7(3) Practice Type No treatment 5(2) Solo 28 (11) 83 (33) Small single-specialty* 'Clinical scenario: First choice of treatment for a nonpregnant, non­ Large single-specialty! 51 (20) smoking adult patient with acute bronchitis and without sinusitis, sys­ Multispecialty 44 (17) temic symptoms or history of asthma/chronic obstructive pulmonary Academic 30 (12) disease. Other 17(7) Percentage of HMO Patients in Practice DISCUSSION <20 103 (41) 21 to 40 67 (27) Antibiotics are the primary treatment used by family 53 (21) 41 to 60 physicians in providing care to otherwise healthy 61 to 80 13(5) adults with acute bronchitis. If the patient smokes, >80 14(6) the likelihood of receiving a prescription for an 'Small single-specialty practices have 2 to 4 physicians. antibiotic significantly increases. Some antibiotics, (Large single-specialty practices have 5 or more physicians. such as amoxicillin, whose efficacy have not been studied in a controlled, randomized clinical trial, are (F=110.25; df= 1; P >.0001). Respondents reported commonly used for acute bronchitis. Agonist that they prescribe p2 agonist bronchodilators 25% bronchodilators are rarely used as the first choice of of the time for nonsmoking patients, and 41% o f the treatment. Most physicians prescribe treatment of time for patients who smoke. some kind as their first choice (84%); very few rec­ Amoxicillin and erythromycin are the two antibi­ ommend no treatment (2%) or the use of over-the- otics most commonly prescribed by responding counter medications (11%) as their first choice. physicians for the nonsmoking patient. If the patient These findings replicate previous studies. smokes, a newer macrolide or trimethoprim/sul- famethoxazole was reported to be used as often as _ TABLE 3 ______erythromycin and amoxicillin (Table 3). Physicians reported that it takes longer for cough Responding Physicians’ First Choice of Antibiotic, If an to totally resolve in patients who smoke than in Antibiotic is Prescribed patients who do not smoke. Smokers also take Nonsmoking Patient Who longer to return to a normal activity level. Fifty-nine Patient Smokes Antibiotic No. (%) No. (%) percent of physicians reported that cough generally lasts more than 7 days in the nonsmoking patient Amoxicillin 101 (40) 59 (23) with acute bronchitis. If the patient smokes, 88% of Erythromycin 89 (35) 56 (22) physicians reported that the cough lasts more than 7 Newer macrolides* 25 (10) 49 (19) days; 48% reported that it lasts more than 14 days. TMP/SMXt 20 (8) 46 (18) Tetracycline 17(7) Eighty-five percent o f physicians say that it general­ 11 (4) Cephalosporin 6(2) 25 (10) ly takes 7 days or less for a nonsmoker to return to Other 2(1) 2(1) normal activity. If the patient smokes, 41% of physi­ cians reported that it takes more than 7 days to * Azithromycin or clarithromycin. achieve a normal activity level (Table 4). tTMP/SMX denotes trimethoprim/sulfamethoxazole.

The Journal o f Family Practice, Vol. 46, No. 6 (June), 1998 471 TREATMENT OF ACUTE BRONCHITIS IN ADULTS

TABLE 4 Responding Physicians’ Perceptions About the Duration of Cough and Time to Return to Normal Activity for Otherwise Healthy Adults with Acute Bronchitis Cessation of Cough Return to Normal Activity Duration Nonsmokers, % Smokers, °/ cH P* ' Nonsmokers, % Smokers, % ' P‘'

< 2 days 0 0 NS 21 6 <.001

3 to 7 days 41 12 <.001 64 53 <.005

8 to 14 days 49 40 NS 13 30 <.001

>14 days 10 48 <.001 2 11 <.001

* Differences in Proportion Test

Mainous and colleagues26 evaluated antibiotic use general and broad spectrum antibiotics in particular in otherwise healthy adults with acute bronchitis in the patient who smokes. Is there evidence to sup­ in a Medicaid population in Kentucky. They report­ port these treatment strategies? ed that 61% of patients were prescribed an antibi­ otic only, and 3% received a ffe agonist bron- Is There Evidence to Support the chodilator as the only medication. Fourteen per­ A moxicillin Choice? cent of patients received both, with a total of 75% Data from our study and previous reports23,25 indicate of patients receiving antibiotics. This is similar to that amoxicillin is one of the most commonly used our findings that physicians report using antibi­ antibiotics for otherwise healthy adults with acute otics in 75% o f nonsmoking patients and 90% of bronchitis. To date, placebo-controlled, randomized patients who smoke. Mainous et al25 also found that clinical trials have not evaluated the use of amoxi­ penicillins were the most commonly used antibi­ cillin for acute bronchitis. The theoretical coverage otics, being prescribed for 37% o f patients. The use for potential bacterial pathogens is minimal. o f p2 agonist bronchodilators was similar to the Amoxicillin is not effective for treatment o f atypical use reported in our study, with 3% o f the Kentucky bacteria such as M pneumoniae or C pneumoniae physicians prescribing a pa agonist only and 6% of or potential (3-lactamase-producing respiratory responding physicians in our study reporting a P2 pathogens such as Haemophilus influenzae or agonist as their first choice of treatment. A higher MoraxeUa catarrhalis. Amoxicillin is still effective use of antibiotics was seen in rural physicians in against pneumococcus, though the incidence of Kentucky. Rural physicians responding to our penicillin-resistant pneumococcus is increasing.3134 study did not report a higher use o f antibiotics than There is no evidence to support the continued use of physicians in an urban setting. amoxicillin in the treatment o f acute bronchitis in an Meza and colleagues23 reported that in Australia otherwise healthy adult. antibiotics were used in more than half o f patients with acute bronchitis, with amoxicillin being the Are A ntibiotics Proven Effective? most commonly used (28%). Similar trends are seen Potential bacterial respiratory pathogens are rarely in the Netherlands, where Verheij et al24 reported that cultured in patients with acute bronchitis. general practitioners prescribed an antibiotic in Approximately 40% o f patients with a history of a more than 65% o f encounters with patients with productive cough can produce a sputum sample, of acute bronchitis.24 which approximately one third have purulent spu­ Our findings, supported by other studies,23"26 sug­ tum by liberal criteria (> 5 WBC/HPF).14,16 Dere" gest that the traditional treatment of otherwise reported that o f 2218 patients with acute bronchitis healthy adults with acute bronchitis includes the use and microscopically purulent sputum (^25 o f an antibiotic, usually amoxicillin, in the majority WBC/HPF, <10 epithelial cells/HPF), only 21% had a o f patients, with an even greater use o f antibiotics in culture positive for Streptococcus pneumoniae, H

472 The Journal o f Family Practice, Vol. 46, No. 6 (June), 1998 TREATMENT OF ACUTE BRONCHITIS IN ADULTS

influenzae, M catarrhalis, or Haemophilus parain- not resolved in 30 days is the favored strategy rather fluenzae. Gaillat* reported a similar finding in 529 than empiric treatment o f all patients with acute patients with acute bronchitis and visibly purulent bronchitis or initial screening for M pneumoniae or sputum: 22% had one of the four organisms. C pneumoniae. Extrapolating from this data, a patient with acute Finally, the indiscriminate use o f antibiotics in the bronchitis and a history of a productive cough will treatment of acute bronchitis will adversely affect have a potential bacterial pathogen that can be cul­ the population by contributing to the growth of tured only 2% to 3% o f the time. Atypical bacteria, antibiotic resistance. The rapid emergence o f mul­ including M pneumoniae and C pneumoniae, are tidrug-resistant respiratory pathogens is a critical uncommon causes o f acute bronchitis, accounting global health issue.3942 The rate of community- for 10% to 15% o f the cases o f acute bronchitis.2'5 acquired, penicillin-resistant S pneumoniae contin­ King and Muncie37 suggest that the prevalence may ues to increase worldwide, including in the United be underestimated because o f difficulties in testing States, Canada, and Europe.31-34 During 1994 and for M pneumoniae. 1995, 23.6% o f outpatient isolates o f S pneumoniae The effectiveness o f antibiotics in the treatment collected in 30 different medical centers were resis­ of acute bronchitis is controversial. O f the eight ran­ tant to penicillin.34 Ten percent o f isolates were resis­ domized clinical trials we found that compared tant to erythromycin, azithromycin, or clar­ antibiotic with placebo, three show marginal symp­ ithromycin. Resistance of S pneumoniae to tom improvement with antibiotics.151718 In seven of cephalosporins, tetracyclines, and trimethoprim/sul- the studies, erythromycin or doxycycline was used; famethoxazole is also growing.3134 both are effective against both Mpneumoniae and C The link between the widespread use o f antibi­ pneumoniae. King et al18 reported that patients treat­ otics in the community and the growth o f antibiotic ed with erythromycin returned to work 1 day earlier resistance in respiratory pathogens has been estab­ than patients treated with placebo. Use of ery­ lished. In a recent Finnish study (1992), 16% o f thromycin did not affect the frequency o f cough, 10,000 group A streptococcus isolates from outpa­ sense of well-being, or chest congestion. In contrast, tients were erythromycin-resistant, increased from Williamson1 reported that patients treated with 5.4% in 1988.43 The proportion o f isolates resistant to placebo returned to work 1 day earlier than those erythromycin clearly increased with local ery­ treated with doxycycline. Notably, King reported thromycin use by outpatients. that patients with M pneumoniae received no greater benefit with the use o f erythromycin than the Study Limitations other patients. In evaluating the generalizability o f this study, three Patients treated with placebo are no more likely potential biases or limitations related to the survey to experience poor clinical outcomes than patients methodology need to be considered. First, a possi­ treated with antibiotics. Of the 756 total patients ble selection bias could result by having nonrespon­ treated in the eight clinical trials, five patients treat­ ders who are different from responders. Since the ed with an antibiotic and 6 patients treated with demographics o f the responders in this study are placebo were reported to have worsened.1,1248 Of similar to the demographics o f physicians in the those 11 patients, three from each group was subse­ American Board of Family Practice, however, and quently given a diagnosis of pneumonia; none the responders and nonresponders have a similar required hospitalization. geographic distribution, we are confident that our In all of the randomized clinical trials, subanalysis sample is representative o f the target population. demonstrates that smoking and nonsmoking Second, the response rate was lower than the sam­ patients do not respond differently to antibiotics.1,12'18 ple size calculation. A larger response rate would In addition to the lack o f evidence that antibiotics likely have improved the power of some of the sta­ significantly change the outcome o f otherwise tistical analyses. Thus, there may be some differ­ healthy patients with acute bronchitis, universal ences in the population o f physicians that were not treatment with erythromycin is not cost effective. In detectable in this sample. Third, the literature indi­ a cost-effective analysis, Hueston38 reported that cates that in physician surveys, there is a tendency treating only those patients with a cough that had toward overreporting the adherence to clinical and

The Journal o f Family Practice, Vol. 46, No. 6 (June), 1998 4 73 TREATMENT OF ACUTE BRONCHITIS IN ADULTS

laboratory guidelines when this information is com­ Philadelphia, PA: WB Saunders, 1994:982-3. pared with notes in the medical record or patient 10. Seaton A, Seaton D, Leitch AG, eds. Crofton and Douglas’ res­ piratory diseases. 4th ed. Boston, Mass: Blackwell Scientific self-reports. In our study, the cover letter specified Publications, 1989:276-7. that we were conducting a survey to examine com­ 11. Koster F. Respiratory infections. In: Barker LR, Burton JR mon practices in the medical management of acute Zieve PD, eds. Principles o f ambulatory medicine. Baltimore Md: Williams & Wilkins, 1986:357-8. bronchitis; that is, there were no right or wrong 12. Stott NCH, West R. Randomized controlled trial of antibiotics answers. There are also no clearly established guide­ in patients with a cough and purulent sputum. BMJ 1976- 2:556-9. lines for either diagnosing or treating this illness that 13. Franks P, Gleiner J A The treatment o f acute bronchitis with could have influenced what physicians reported. trimethoprim and sulfamethoxazole. J Fam Pract 1984- 19:185-90. 14. Brickfield FX, Carter WH, Johnson RE. Erythromycin in the CONCLUSIONS treatment o f acute bronchitis in a community practice. J Fam Pract 1986; 23:119-22. 15. Dunlay J, Reinhardt R, Roi LD. A placebo-controlled, double­ Bacteria are uncommon causes of acute bronchitis blind trial o f erythromycin in adults with acute bronchitis J in otherwise healthy adults. Treatment for bacterial Fam Pract 1987; 25:137-41. infection has not been conclusively shown to benefit 16. Scherl ER, Riegler SL, Cooper JK Doxycycline in acute bron­ chitis: a randomized double-blind trial. J Ky Med Assoc 1987; patient outcomes, and no treatment does not September:539-41. increase the likelihood of a poor outcome. In light of 17. Verhejj TJM, Hermans J, Mulder JD. Effects o f doxycycline in patients with acute cough and purulent sputum: a double­ the growing crisis o f antibiotic-resistant respiratory blind, placebo-controlled trial. Br J Gen Pract 1994; 44:4004. pathogens and the lack of evidence for significantly 18. King DE, Williams WC, Bishop L, Shechter A. Effectiveness of erythromycin in the treatment o f acute bronchitis. J Fam changed outcomes with the use o f antibiotics, antibi­ Pract 1996; 42:601-5. otics should not be used as the primary treatment for 19. Hueston WJ. A comparison o f albuterol and erythromycin for acute bronchitis. Other methods of treatment need the treatment of acute bronchitis. J Fam Pract 1991; 33:476-80. 20. Melbye H, Aasebo U, Straume B. Symptomatic effect of to be further evaluated. Early studies evaluating the inhaled fenoterol in acute bronchitis: a placebo-controlled use o f P2 agonist bronchodilators are promising,22 24 double-blind study. Fam Pract 1991; 8:216-22. but larger clinical trials are necessary to determine 21. Hueston WJ. Albuterol delivered by metered-dose inhaler to treat acute bronchitis. J Fam Pract 1994; 39:437410. their effectiveness. 22. Littenberg B, Wheeler M, Smith DS. A randomized controlled trial of oral albuterol in acute cough. J Fam Pract 1996; 42:49- ACKNOWLEDGMENTS 53. This project was supported by grant No. G9602RS from the 23. Meza RA, Bridges-Webb C, Sayer GP, et al. The management American Academy o f Family Physicians Foundation, the Texas o f acute bronchitis in general practice: results from the Academy o f Family Physicians Foundation, and predoctoral grant Australian Morbidity and Treatment Survey, 1990-1991. Aust No. 5 D15 PE86036 from the US Public Health Service. Fam Physician 1994; 23:1550-3. 24. Verheu TJM, Kaptein AA, Mulder JD. Acute bronchitis: aetiol­ REFERENCES ogy, symptoms, and treatment. Fam Pract 1989; 6:66-9. 1. Williamson HA. A randomized, controlled trial o f doxycycline 25. Mainous AG, Zoorob RJ, Hueston WJ. Current management of in the treatment o f acute bronchitis. J Fam Pract 1984; 19:481- acute bronchitis in ambulatory care: the use of antibiotics and bronchodilators. Arch Fam Med 1996; 5:79-83. 6. 2. Grayston JT. Infections caused by Chlamydia pneumoniae 26. Oeffinger KC, Snell LM, Foster BM, Panico KG, Archer RK. strain TWAR. Clin Infect Dis 1992; 15:757-63. Diagnosis o f acute bronchitis in adults: a national survey of family physicians. J Fam Pract 1997; 45:402-9. 3. Kuo CC, Jackson LA, Campbell LA, Grayston JT. Chlamydia pneumoniae (TWAR). Clin Microbiol Rev 1995; 8:451-61. 27. American Board o f Family Practice 1995 diplomate directory. Lexington, Ky: American Board of Family Practice, 1995. 4. Evans AS, Allen C, Sveltman S. Mycoplasma pneumoniae infections in University o f Wisconsin students. Am Rev Resp 28. Kirk R. Statistics: an introduction. 3rd ed. Ft. Worth, Tex: Holt Dis 1967; 96:237-44. Rinehart & Winston, Inc, 1990. 5. Mogabgab WJ. Mycoplasma pneumoniae and adenovirus res­ 29. Cohen J. Statistical power analysis for the behavioral sci­ piratory illnesses in military and university personnel. Am ences. 2nd ed. Hillsdale, NJ: Erlbaum Associates, 1988. Rev Resp Dis 1968; 97:345-58. 30. Fleis J. Statistical methods for rates and proportions. 2nd ed. 6. Rodnick JE, Gude JK Pulmonary infections. In: Taylor RB, ed. New York, NY: John Wiley & Sons, 1981. Family medicine: principles and practice. 4th ed. New York, 31. Simor AE, Louie M, Low DE. Canadian national survey of NY: Springer-Verlag, 1994:660-71. prevalence of antimicrobial resistance among clinical isolates 7. Prichard JG, Tierney LM. Acute bronchitis. In: Rakel RE, ed. o f Streptococcus pneumoniae. Am Soc Microbiology 1996; Textbook of family practice. 5th ed. Philadelphia, Pa: WB 40:2190-3. Saunders, 1995:420-1. 32. Johnson AP, Speller DC, George RC, Warner M, Domingue G, 8. Simon HB. Approach to the patient with acute bronchitis or Efstratiou A. Prevalence of antibiotic resistance and pneumonia in the ambulatory setting. In: Goroll AH, May LA, serotypes in pneumococci in England and Wales: results of Mulley AG, eds. Primary care medicine: office evaluation and observational surveys in 1990 and 1995. BMJ 1996; 312:1454-6. management of the adult patient. 3rd ed. Philadelphia, Pa: JB 33. Boswell TC, Frodsham D, Nye KJ, Smith EG. Antibiotic Lippincott, 1995:285-90. resistance and serotypes of Streptococcus pneumoniae 9. Hayden FG, Gwaltney JM. Acute bronchitis. In: Murray JF, at Birmingham public health laboratory, 1989-94. J Infect Nadel JA, eds. Textbook o f respiratory medicine. 2nd ed. 1996; 33:17-22.

474 The Journal o f Family Practice, Vol. 46, No. 6 (June), 1998 TREATMENT OF ACUTE BRONCHITIS IN ADULTS

34. Doem GV, Braeggeman A, Holley HP, Rauch AM. effective. J Fam Pract 1997; 44:261-5. Antimicrobial resistance o f Streptococcus pneumoniae recov­ 39. Tenover FC, McGowan JE. Reasons for the emergence of ered from outpatients in the United States during the winter antibiotic resistance. Am J Med Sci 1996; 311:9-16. months of 1994 to 1995: results o f a 30-center national surveil­ lance study. Antimicrob Agents Chemother 1996; 40:1208-13. 40. Jones RN. The emergent needs for basic research, education, and surveillance of antimicrobial resistance. Diagn Microbiol 35. Dere WH. Acute bronchitis: results of US and European trials Infect Dis 1996; 25:153-61. of antibiotic therapy. Am J Med 1992; 92:53S-57S. 41. O’Brien TF. The global epidemic nature of antimicrobial resis­ 36. Gaillat J. A multicentre study comparing the safety and effi­ tance and the need to monitor and manage it locally. Clin cacy of dirithromycin with erythromycin in the treatment of Infect Dis 1997; 24:S2-S8. bronchitis. J Antimicrob Chemother 1993; 31(suppl C):139-51. 42. Norrby SR. Antibiotic resistance: a self-inflicted problem. J 37. King DE, Munciep HL Jr. High prevalence o f Mycoplasma Intern Med 1996; 239:373-5. pneumoniae in patients with respiratory tract symptoms: a rapid detection method. J Fam Pract 1991; 32:529-31. 43. Seppala H, Klaukka T, Lehtonen R, et al. Outpatient use o f ery­ thromycin: Link to increased erythromycin resistance in 38. Hueston WJ. Antibiotics: neither cost effective nor ‘cough’ group A streptococci. Clin Infect Dis 1995; 21:1378-85.

The Journal of Family Practice, Vol. 46, No. 6 (June), 1998 4 75