Family Practice Vol. 17, No. 5, 386-388
© Oxford University Press 2000
Predictors of an antibiotic prescription by GPs for respiratory tract infections: a pilot
Centre for General Practice, University of Queensland Medical School, Herston, Queensland 4006, Australia.
Professor Chris Del Mar, Centre for General Practice, University of Queensland Medical School, Herston, Queensland 4006, Australia.
Murray S, Del Mar C and O'Rourke P. Predictors of an antibiotic prescription by GPs for respiratory tract infections: a pilot. Family Practice 2000; 17: 386–388.
Received 2 February 2000; Revised 11 May 2000; Accepted 16 May 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Antibiotics are over-prescribed for respiratory tract infections in Australia.
Objectives. The aim of this study was to describe the clinical predictors of GPs' prescribing of antibiotics.
Methods. We used Clinical Judgment Analysis to study the responses of GPs to hypothetical paper-based vignettes of a 20-year-old with a respiratory tract infection. The nature of four symptoms and signs (colour of nasal mucous discharge; soreness of the throat; presence of fever; and whether any cough was productive of sputum) was varied and their effect on prescribing measured using logistic regression.
Results. Twenty GPs participated. The nature of each symptom and sign significantly predicted prescribing of an antibiotic. Cough productive of yellow sputum; presence of sore throat; fever; and coloured nasal mucus increased the probability of an antibiotic being prescribed.
Conclusions. GPs are influenced by clinical signs and symptoms to use antibiotics for respiratory infections for which there is poor evidence of efficacy from the literature.
Keywords. Antibiotics, clinical signs, GPs, prescribing, respiratory tract infections.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Antibiotics are prescribed for respiratory tract infections more often than evidence of efficacy dictates.1 Unrealistic patient expectations that antibiotics will hasten, or are necessary for, recovery have been a major influence.2 We postulate that erroneous GP belief is also a major factor. To this end, we studied the signs and symptoms of respiratory tract infection used by GPs as clinical cues for prescribing antibiotics.
The symptoms of respiratory tract infections vary in the nature of their runny nose, sore throat, fever and cough. There is a spectrum of these clinical features across a range of diagnoses: thus in infections labelled ‘upper respiratory tract infections’, 11% of patients reported cough producing green phlegm, 25% producing white or yellow phlegm and 39% a dry cough.3 In acute bronchitis, whilst cough occurred in 92%, both runny nose and throat pain also occurred in 50% of cases.4 We decided to examine the effects of variation of these symptoms and signs on the likelihood of GPs prescribing antibiotics.
One difficulty is the strong influence of psycho-social factors on the likelihood of GPs to prescribe.5 The technique of Clinical Judgment Analysis6,7 allows vignettes to be developed for which such influences are controlled, and only the factors under scrutiny are varied. It has been used previously to study prescribing for sore throat,8 otitis media9 and sinusitis.10
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
A convenience sample of GPs was asked to study paper-based clinical vignettes of a 20-year-old with a respiratory tract infection, and indicate those for which they would prescribe an antibiotic. The introduction stated "you feel no pressure from the patient to prescribe an antibiotic", to control for patient expectations. Four symptoms and signs were varied to create 32 versions (2 x 2 x 2 x 4) of the vignette (Table 1
). Order effects were controlled by inserting three dummy vignettes, and randomizing the order of the questions. The significance that variations in the presence of each symptom or sign, including interactions, were associated with an antibiotic prescription was estimated by analysis of variance. We estimated the weight that each GP assigned to each symptom or sign when deciding whether or not to use an antibiotic using logistic regression. The best evidence from the literature is that antibiotics would have negligible influence on the clinical outcome for any of the 32 variations.11–13
|
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
All 20 questionnaires were returned with all 32 questions fully completed. The GPs indicated that they would prescribe an antibiotic for 212 of the 640 versions (20 x 32) of the 20-year-old with a respiratory tract infection. The effect on prescribing of the severity variables of the four symptoms and signs was highly significant (Table 1
). None of the interactions between the severity variables of the different symptoms and signs were significant.
The weight (standardized regression coefficient) that the GPs as a group assigned to each variable is shown in Figure 1. It is apparent that the significance of the ‘cough’ variable is due solely to the ‘cough productive of yellow sputum’ level. ‘Dry cough’ and ‘cough productive of clear sputum’ are not predictive of an antibiotic prescription.
|
The most important predictor was ‘cough productive of yellow sputum’: there was a 70% chance of receiving an antibiotic if this variable was present, compared with a 20% chance if there was either no cough, dry cough or cough productive of clear sputum.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
GPs were influenced by four clinical features to prescribe antibiotics for respiratory tract infections. To have prescribed antibiotics for any of the vignettes deviates from the best available evidence in the literature, which suggests that antibiotics would have little effect on the clinical outcomes.
Our data confirm that of other studies, which found that antibiotic prescribing is increased if nasal discharge or sputum is coloured rather than clear.3,14,15 Cough with coloured sputum was the strongest predictor.
This study was a small pilot, thus the results may not be generalizable to other GPs. Also, this particular questionnaire has not been compared with GPs' actual prescribing with real patients, though other studies have indicated the validity of this technique.7,9
If confirmed in a larger study, these data, juxtaposed with the best evidence from the literature, could act as a powerful tool for educational purposes.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Butler CC, Rollnick S, Kinnersley P, Jones A, Stott N. Reducing antibiotics for respiratory tract symptoms in primary care: consolidating ‘why’ and considering ‘how’. Br J Gen Pract 1998; 48: 1865–1870.[Web of Science][Medline]
2
Butler CC, Rollnick S, Pill R, Maggs-Rapport F, Stott N. Understanding the culture of prescribing: qualitative study of general practitioners' and patients' perceptions of antibiotics for sore throats. Br Med J 1998; 317: 637–642.
3 Gonzales R, Barrett PH, Steiner JF. The relation between purulent manifestations and antibiotic treatment of upper respiratory tract infections. J Gen Intern Med 1999; 14: 151–156.[Web of Science][Medline]
4 Gonzales R, Barrett PH, Crane LA, Steiner JF. Factors associated with antibiotic use for acute bronchitis. J Gen Intern Med 1998; 13: 541–548.[Web of Science][Medline]
5 Howie JGR. Clinical judgement and antibiotic use in general practice. Br Med J 1976; 2: 1061–1064.
6 Wigton RS. Use of linear models to analyse physicians' decisions. Med Decision Making 1988; 8: 241–252.
7 Kirwan JR, Chaput de Saintonge DM, Joyce CR. Clinical judgment analysis. Q J Med 1976; 281: 935–949.
8
Poses RM, Cebul RD, Wigton RS. You can lead a horse to water—improving physicians' knowledge of probabilities may not affect their decisions. Med Decision Making 1995; 15: 65–75.
9
Chaput de Saintonge DM, Hattersley LA. Antibiotics for otitis media: can we help doctors agree? Fam Pract 1985; 2: 205–212.
10 Little DR, Mann BL, Sherk DW. Factors influencing the clinical diagnosis of sinusitis. J Fam Pract 1998; 46: 147–152.[Web of Science][Medline]
11 Del Mar CB, Glasziou PP. Antibiotics for sore throat (Cochrane Review). In The Cochrane Library, Issue 3, Oxford: Update Software, 1999.
12
Fahey T, Stocks N, Thomas T. Systematic review of the treatment of upper respiratory tract infection. Arch Dis Child 1998; 79: 225–230.
13
Fahey T, Stocks N, Thomas T. Quantitative systematic review of randomised controlled trials comparing antibiotic with placebo for acute cough in adults. Br Med J 1998; 316: 906–910.
14 Mainous AG, Hueston WJ, Eberlein C. Colour of respiratory discharge and antibiotic use. Lancet 1997; 350: 1077.[Web of Science][Medline]
15
Kawamoto R, Asai Y, Nago N, Okayama M, Mise J. Igarashi M. A study of clinical features and treatment of acute bronchitis by Japanese primary care physicians. Fam Pract 1998; 15: 244–251.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Cadieux MSc, R. Tamblyn PhD, D. Dauphinee MD, and M. Libman MD Predictors of inappropriate antibiotic prescribing among primary care physicians Can. Med. Assoc. J., October 9, 2007; 177(8): 877 - 883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Akkerman, M. M. Kuyvenhoven, J. C. van der Wouden, and T. J. M. Verheij Determinants of antibiotic overprescribing in respiratory tract infections in general practice J. Antimicrob. Chemother., November 1, 2005; 56(5): 930 - 936. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Barley and P. W. Jones Analysis of the cues used by patients when making assessments of their asthma severity Eur. Respir. J., April 1, 2005; 25(4): 671 - 676. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nissen and C. Del Mar GPs should reduce antibiotic use with alternative treatments BMJ, May 4, 2002; 324(7345): 1099 - 1099. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||