Research

Antibiotics for lower respiratory tract infections

Still too frequently prescribed?

Warren J. McIsaac, MD, MSC, CCFP    Teresa To, PHD

This article has been peer reviewed. Cet article a fait l’objet d’une évaluation externe. Can Fam Physician 2004;50:569-575.

Dr McIsaac is a family physician in the Ray D. Wolfe Department of Family Medicine at Mt Sinai Hospital in Toronto, Ont, and is a researcher in the Department of Family and Community Medicine at the University of Toronto. Dr To is an epidemiologist in the Population Health Unit at the Hospital for Sick Children in Toronto.

Acute bronchitis is one of the most frequent reasons for prescribing antibiotics to adults.1,2 Acute bronchitis is distinct from acute exacerbation of chronic bronchitis in that patients who present with cough and symptoms of acute lower respiratory tract infections (LRTIs) usually do not have underlying chronic lung disease.3,4 Prescribing rates for adults with acute bronchitis have been reported to range from 66% to 81%2,5-9 even though experts recommend that antibiotics not be prescribed routinely for acute bronchitis (studies have found they have either limited or no benefit).10-13

Common lower respiratory tract pathogens have become more resistant to antibiotics.14 Penicillin-resistant and multidrug-resistant Streptococcus pneumoniae are increasingly common.15,16 Case reports have described treatment failures related to quinolone resistance.17,18 Other pathogens, such as Haemophilus influenzae and Moraxella catarrhalis are frequently resistant to common antibiotics such as ampicillin.19

Whether family physicians have reduced their use of antibiotics for acute bronchitis in light of concerns about resistance is unclear. A 1997 chart audit of family physicians in Newfoundland found 85% of patients with LRTIs received antibiotics.20 Other studies have relied on self-reported prescribing, chart audits, or data from case vignettes of hypothetical patients.21-23 Some studies have reported actual prescribing,24,25 but one study was limited to children and others studied only physicians in Saskatchewan and Ontario.23,24 The purpose of this study was to determine prospectively rates of antibiotic prescribing for adults with LRTIs presenting to a national sample of community-based family physicians.

METHODS

The study was conducted in collaboration with the College of Family Physicians of Canada (CFPC). The College maintains a national register, known as NaReS (National Research System), of community­based family physicians engaged in full-time clinical practice who have participated in research projects.26 In March 1999, English-speaking NaReS members were invited to participate in a study of adults with acute cough.

Physicians agreeing to participate were mailed a consent form and four clinical encounter sheets with accompanying patient consent forms and diaries. Any patients 16 years or older who presented with a new cough and the possibility of chest infection, such as acute bronchitis, were eligible and could be chosen by physicians for the study. Patients already taking antibiotics, immunocompromised, with chronic obstructive lung disease, or unable to complete the diary in English were ineligible. After the visit, patients were invited to read information about the diary portion of the study. Those expressing interest were given a package containing a diary and consent form that, when completed, were to be returned directly to the study centre.

A one-page form was used to record patients’ demographic characteristics, duration of illness, presenting symptoms and signs, diagnoses, tests ordered, and whether antibiotics were prescribed. Physicians were also asked how certain they felt that antibiotics were either indicated or not indicated, how concerned they were that patients might become sicker in the next week (using a scale from 0 to 100%), and whether they felt pressure to prescribe antibiotics (none, some, a lot, patient verbally requested antibiotics). The form was pretested on a convenience sample of family physicians.

There were approximately 700 English-speaking family physicians participating in NaReS in 1999. The study sought to enrol 300 physicians, based on a 50% response rate in a previous study.27 Cluster sampling was used: each physician was asked to complete four clinical assessments. Frequencies were used for categorical variables and means or medians for continuous variables; 95% confidence intervals were determined and adjusted for clustering of patients using estimation procedures for complex survey data.28 The study was approved by the University of Toronto’s Research Ethics Board.

RESULTS

Of 600 family physicians contacted, 270 agreed to participate, and 120 (44%) completed the study. Compared with all CFPC members,29 study physicians were younger (42.7 vs 46.6 years), more likely to be male (68.3% vs 65.0%), and had practised for fewer years (16.2 vs 18.5 years). Proportion of urban physicians was similar (85.0% vs 84.1%). Most family physicians were from Ontario (63.7%); followed by British Columbia (15.9%); Alberta (6.7%); Saskatchewan and Manitoba (7.0%); Quebec (3.0%); and Nova Scotia, Newfoundland, and Yukon Territories (3.7%). In all, 408 clinical encounters were completed, a median of two per physician (range one to four). One patient younger than 16 years was excluded. Little information on individual items on the form was missing (0.5% to 4.1%).

Patient characteristics and clinical findings

Median age was 44.0 years (range 16 to 86), and 62.5% were women. Sixty-eight people (17.1%) were 65 years or older, and 114 (28.4%) had major comorbid conditions. Two thirds (63.8%) were seen within 7 days of onset of symptoms. Almost all reported frequent cough, two thirds had coloured sputum (Table 1), more than half reported wheezing or a tight chest, and 28.4% reported fever (>38°C). Most had concurrent upper respiratory tract symptoms, such as runny or stuffy nose and sore throat.

Physicians could record physical findings as present, absent, or not assessed; all were asked to record respiratory rate. Temperature was taken in 48.2% of cases, more often if fever was reported (55.8%, P = .05). Heart rate was recorded for 59.5% and respiratory rate for 79.6% of patients. Auscultation indicated that 57.1% of patients had clear chests. Most common findings were wheezes, followed by crackles and decreased breath sounds. Other clinical findings were rare.

Diagnoses, tests, and prescriptions for antibiotics

Diagnoses. Physicians used 35 separate diagnostic labels. These were grouped into common categories of variations of the same term (Table 2). The most common diagnosis was acute bronchitis or some variation (viral bronchitis, bacterial bronchitis, asthmatic bronchitis, sinubronchitis, tracheobronchitis). Pneumonia was diagnosed in 4.2% of patients. A nonspecific label (cough, chest cold) or viral etiology (viral LRTI, viral illness) was used in one in six cases.

Tests. Tests were rarely ordered; the most common test requested was chest x-ray examination. Antibiotics were prescribed in 86.5% of instances where chest x-ray was ordered and in 57.4% of instances where no x-ray was requested (P < .001).

Prescriptions. Prescribing data were available for 404 encounters (99.3%). Antibiotics were prescribed at 58.4% of visits (95% confidence interval [CI] 52.2 to 64.6); six of these (2.5%) were delayed prescriptions to be filled if patients did not improve. Median duration of prescriptions was 10 days (interquartile range 7 to 10 days). Viral conditions, asthmalike presentations, and cases with no diagnoses were combined into one category because prescribing rates were similar (Table 3).

Antibiotics were prescribed in 77.9% of encounters (95% CI 70.4% to 84.0%) where the diagnosis was acute bronchitis and in 28.9% where the diagnosis was viral, asthma related, or no diagnosis (95% CI 21.4% to 37.9%). Acute bronchitis accounted for 70.3% of antibiotic prescriptions (95% CI 62.8% to 76.9%).

Information about choice of antibiotic (Table 4) was available for 233 of 236 prescriptions (98.7%). Macrolides were most commonly prescribed (59.2%) (95% CI 51.3% to 66.7%); 64.5% of these prescriptions were for either clarithromycin or azithromycin. Aminopenicillins were the next most commonly prescribed antibiotics; amoxicillin accounted for 84.8% of these. Cephalosporins were prescribed in 9.0% of cases (95% CI 5.6% to 14.2%). Quinolones were prescribed less often (6.4%) (95% CI 3.5% to 11.7%).

Clinical uncertainty and patient pressure

Physicians expressed uncertainty about the need for antibiotics in 281 of 404 encounters (69.6%). Antibiotics were thought to be definitely indicated or definitely not indicated in 123 encounters (30.5%), probably indicated in 154 (38.1%), and probably not indicated in 127 (31.4%). Prescribing rate was 96.6% when antibiotics were definitely indicated, 98.0% when probably indicated, 24.0% when probably not indicated, and 0 when definitely not indicated. For 76.3% of their prescriptions (95% CI 68.7% to 82.5%), physicians expressed at least some uncertainty about the need for antibiotics.

Physicians reported feeling no pressure from patients to prescribe antibiotics in 272 of 405 encounters (67.2%). When they felt at least some pressure from patients, the prescribing rate was 82.7% (95% CI 75.3% to 88.3%) compared with 46.5% when no pressure was felt (95% CI 39.3% to 53.8%). In 145 of 402 encounters (36.1%), physicians were somewhat concerned that patients could become sicker in the next week. The prescribing rate was 82.6% in these encounters, but only 44.3% in visits where physicians thought deterioration unlikely.

DISCUSSION

More than half the adults who present to Canadian family physicians with cough and symptoms suggesting LRTI are prescribed antibiotics. The rate of prescribing is higher when acute bronchitis is diagnosed. The 78% rate for acute bronchitis in this study is similar to the 89% rate in a 1990 Australian study,5 the 75% rate in an American study of Medicaid claims from 1993 to 1994,6 and the 75% rate in a 1997 British study of more than 1000 adults.7 From 1980 to 1994 in the United States, the prescribing rate for acute bronchitis remained at 81% overall.9 Two other studies reported lower rates: 67% and 68%.2,8 This suggests that prescribing for acute bronchitis in Canada has remained as high as in other English-speaking countries during the last 20 years and has not changed appreciably.

Whether acute bronchitis should be considered a distinct diagnosis has been questioned.30 A consensus report on management of cough did not include acute bronchitis in the absence of underlying chronic lung disease as a cause of acute cough.31 Instead, common cold and postnasal drip syndromes were listed as the most common causes of cough and sputum of less than 3 weeks’ duration.31 Considerable variation in the criteria family physicians use in making this diagnosis has also been found.3

The contribution of diagnostic labels to the problem of antibiotic overuse has been described by Hutchison et al,20 who found the main difference between high and low prescribers of antibiotics was their use of diagnostic labels suggesting bacterial etiology, such as bronchitis or otitis media. An alternative term suggested is “chest cold,” which is consistent with the usually viral etiology of acute bronchitis. A randomized trial that encouraged physicians to use this term instead of acute bronchitis reduced antibiotic use from 74% to 48%.32

Patient pressure and clinical uncertainty played a role in high rates of antibiotic use.33 Patient pressure was a factor in only one third of visits; physicians’ uncertainty about the need for antibiotics was a factor in three quarters of visits. Use of patient information leaflets on the uncertain benefit of antibiotics for acute bronchitis together with delayed prescriptions reduced antibiotic prescriptions by 24% in one study.34 Canadian family physicians appear to use delayed prescriptions infrequently.

Some of the uncertainty arose from concern that patients might become sicker during the following week. This could reflect physicians’ preference to err on the side of caution to avoid missing early cases of pneumonia. Although no combination of findings is specific for pneumonia,35 physicians relied primarily on clinical examination and ordered chest x-ray examinations for fewer than one in 10 cases. Normal vital signs and a clear chest on auscultation reduce the likelihood of pneumonia, but a chest x-ray examination can allow greater diagnostic certainty when chest findings are present.35 About 40% of patients had chest findings, which would necessitate a substantial increase in the number of x-ray examinations over current practice, and might be impractical in many communities.

The most commonly prescribed antibiotics were macrolides. Between 1995 and 1998, S pneumoniae’s resistance to macrolides increased from 11% to 15% in the United States.16 Macrolide-resistant group A streptococci also appear more frequently,36 a problem that has been linked with the volume of macrolides used in outpatient settings.37 Quinolones are now being prescribed to manage cough in the community. Fluoroquinolone use increased throughout the 1990s in Canada while S pneumoniae susceptibility declined.17 The patterns of antibiotic use observed in this study suggest that multidrug-resistant S pneumoniae and macrolide-resistant group A streptococcus might continue to be a problem.

Limitations

Physicians in this study were volunteers and did not necessarily assess consecutive cases of LRTIs. This could have biased results toward more optimal prescribing. The observed prescribing rate of 78%, however, is similar to rates reported in other studies.5-7,9 Antibiotic use might have declined since this study was completed, but a recent American study found no decrease in visit-based prescribing for bronchitis among children and adolescents.38

Conclusion

Acute bronchitis remains a common diagnosis in adults presenting with symptoms of LRTIs and accounts for 70% of antibiotic prescriptions for managing cough. Specific strategies to help family physicians reduce antibiotic prescribing are needed and could include using the term “chest cold” instead of “acute bronchitis,” writing delayed prescriptions, and disseminating patient information leaflets. Addressing clinical uncertainty is also important if physicians are to have confidence in advising their patients that antibiotics are not needed.

Acknowledgment

We acknowledge Dr Donald E. Low and Dr Richard Glazier for reviewing the manuscript and offering helpful suggestions, and Joanne Permaul for support and supervising data collection. The study was funded by the former Medical Research Council of Canada (grant MA 15474).

Contributors

Dr McIsaac and Dr To designed the study, analyzed and interpreted the data, and prepared the paper for publication.

Competing interests

None declared

Correspondence to: Dr Warren McIsaac, Ray D. Wolfe Family Medicine Centre, Mt Sinai Hospital, 600 University Ave, Toronto, ON M5G 1X5; telephone (416) 586-3190; fax (416) 586-3175; e-mail wmcisaac@mtsinai.on.ca. Reprints will not be available from the authors.

References

1. McCaig LF, Hughes JM. Trends in antimicrobial drug prescribing among office-based physicians in the United States. JAMA 1995;273:214-9.
2. Gonzales R, Steiner JF, Sande MA. Antibiotic prescribing for adults with colds, upper respiratory tract infections, and bronchitis by ambulatory care physicians. JAMA 1997;278:901-4.
3. Oeffinger KC, Snell LM, Foster BM, Panico KG, Archer RK. Diagnosis of acute bronchitis in adults: a national survey of family physicians. J Fam Pract 1997;45:402-9.
4. MacKay DN. Treatment of acute bronchitis in adults without underlying lung disease. J Gen Intern Med 1996;11:557-62.
5. Meza RA, Bridges-Webb C, Sayer GP, Miles DA, Traynor V, Neary S. The management of acute bronchitis in general practice: results from the Australian Morbidity and Treatment Survey, 1990-1991. Aust Fam Physician 1994;23:1550-3.
6. Mainous AG III, Zoorob RJ, Hueston WJ. Current management of acute bronchitis in ambulatory care. The use of antibiotics and bronchodilators. Arch Fam Med 1996;5:79-83.
7. Macfarlane J, Lewis SA, Macfarlane R, Holmes W. Contemporary use of antibiotics in 1089 adults presenting with acute lower respiratory tract illness in general practice in the U.K.: implications for developing management guidelines. Respir Med 1997;91:427-34.
8. Kawamoto R, Asai Y, Nago N, Masanobu O, Mise J, Igarashi M. A study of clinical features and treatment of acute bronchitis by Japanese primary care physicians. Fam Pract 1998;15:244-51.
9. Metlay JP, Stafford RS, Singer DE. National trends in the use of antibiotics by primary care physicians for adults with cough. Arch Intern Med 1998;158:1813-8.
10. Gonzales R, Sande M. What will it take to stop physicians from prescribing antibiotics in acute bronchitis? Lancet 1995;345:665-6.
11. Snow V, Mottur-Pilson C, Gonzales R. Principles of appropriate antibiotic use for treatment of acute bronchitis in adults. Ann Intern Med 2001;134:518-20.
12. Fahey T, Stocks N, Thomas T. Quantitative systematic review of randomised controlled trials comparing antibiotic with placebo for acute cough in adults. BMJ 1998;316:906-10.
13. Smucny JJ, Becker LA, Glazier RH, McIsaac W. Are antibiotics effective treatment for acute bronchitis? A meta-analysis. J Fam Pract 1998;47(6):453-60.
14. Health Canada and the Canadian Infectious Disease Society. Controlling antimicrobial resistance. An integrated action plan for Canadians. Can Commun Dis Rep 1997;23 (Suppl 7):i-iv,1-32.
15. Zhanel GG, Karlowsky JA, Palatnick L, Vercaigne L, Low DE, Hoban DJ. Prevalence of antimicrobial resistance in respiratory tract isolates of Streptococcus pneumoniae: results of a Canadian national surveillance study. The Canadian Respiratory Infection Study Group. Antimicrob Agents Chemother 1999;43:2504-9.
16. Whitney CG, Farley MM, Hadler JH, Harrison LH, Lexau C, Reingold A, et al. Increasing prevalence of multi-drug resistant Streptococcus pneumoniae in the United States. N Engl J Med 2000;343:1917-24.
17. Chen DK, McGeer A, De Azavedo JC, Low DE. Decreased susceptibility of Streptococcus pneumoniae to fluoroquinolones in Canada. N Engl J Med 1999;341:233-9.
18. Davidson R, Cavalcanti R, Brunton JL, Bast DJ, De Azavedo JCS, Kibsey P, et al. Resistance to levofloxacin and failure of treatment of pneumococcal pneumonia. N Engl J Med 2002;346:747-50.
19. Zhanel GG, Karlowsky JA, Low DE, Hoban DJ. Antibiotic resistance in respiratory tract isolates of Haemophilus influenzae and Moraxella catarrhalis collected from across Canada in 1997-1998. J Antimicrob Chemother 2000;45:655-62.
20. Hutchinson JM, Jelinski S, Hefferton D, Desaulniers G, Parfrey PS. Role of diagnostic labeling in antibiotic prescription. Can Fam Physician 2001;47:1217-24.
21. McIsaac WJ, Goel V. Sore throat management practices of Canadian family physicians. Fam Pract 1997;14:34-9.
22. Arnold SR, Allen UD, Al-Zahrani M, Tan DH, Wang EE. Antibiotic prescribing by pediatricians for respiratory tract infection in children. Clin Infect Dis 1999;29:312-7.
23. Paluck E, Katzenstein D, Frankish CJ, Herbert CP, Milner R, Speert D, et al. Prescribing practices and attitudes toward giving children antibiotics. Can Fam Physician 2001;47:521-7.
24. Wang EE, Einarson TR, Kellner JD, Conly JM. Antibiotic prescribing for Canadian preschool children: evidence of overprescribing for viral respiratory infections. Clin Infect Dis 1999;29:155-60.
25. Pennie RA. Prospective study of antibiotic prescribing for children. Can Fam Physician 1998;44:1850-6.
26. Lewis J. NaReS: your national research system. Can Fam Physician 1989;35:837-9.
27. McIsaac WJ, Goel V, To T, Permaul JA, Low DE. Effect on antibiotic prescribing of repeated clinical prompts to use a sore throat score: lessons from a failed community intervention study. J Fam Pract 2002;51:339-44.
28. Stata Corporation. Stata statistical software: release 0.6. College Station, Tex: Stata Corporation; 1999.
29. College of Family Physicians of Canada. 2001 CFPC National Family Physician Workforce Survey. Part of the Janus Project: Family physicians meeting the needs of tomorrow’s society. Mississauga, Ont: College of Family Physicians of Canada; 2001.
30. Hueston WJ, Mainous AG, Dacus EN, Hopper JE. Does acute bronchitis really exist? A reconceptualization of acute viral respiratory infections. J Fam Pract 2000;49:401-6.
31. Irwin RS, Boulet LP, Cloutier MM, Fuller R, Gold PM, Hoffstein V, et al. Managing cough as a defense mechanism and as a symptom. A consensus panel report of the American College of Chest Physicians. Chest 1998;114(Suppl):133-81S.
32. Gonzales R, Steiner JF, Lum A, Barrett PH. Decreasing antibiotic use in ambulatory practice. Impact of a multidimensional intervention on the treatment of uncomplicated acute bronchitis in adults. JAMA 1999;281:1512-9.
33. Miller E, Mackeigan LD, Rosser W, Marshman J. Effects of perceived patient demand on prescribing anti-infective drugs. CMAJ 1999;161:139-42.
34. Macfarlane J, Holmes W, Gard P, Thornhill D, Macfarlane R, Hubbard R. Reducing antibiotic use for acute bronchitis: blinded, randomized controlled trial of patient information leaflet. BMJ 2002;324:91-4.
35. Metlay JP, Kapoor WN, Fine MJ. The rational clinical examination. Does this patient have community-acquired pneumonia? Diagnosing pneumonia by history and physical examination. JAMA 1997;278:1440-5.
36. Martin JM, Green M, Barbadora KA, Wald ER. Erythromycin-resistant group A streptococci in schoolchildren in Pittsburgh. N Engl J Med 2002;346:1200-6.
37. Seppala H, Klaukka T, Lehtonen R, Nenonen E, Huovinen P. Outpatient use of erythromycin: link to increased erythromycin resistance in group A streptococci. Clin Infect Dis 1995;21:1378-85.
38. McCaig LF, Besser RE, Hughes JM. Trends in antimicrobial prescribing rates for children and adolescents. JAMA 2002;287:3096-102.