A third of primary care consultations are due to an infectious disease and more than half are caused by a respiratory tract infection.1 Inappropriate use of antimicrobials for these infections is associated with increased risk of adverse drug events and higher overall health care costs.2,3 Data from 26 European countries demonstrated a correlation between the use of antibiotics and the level of antibiotic resistance.4 The danger of increasing antibiotic resistance has been recognised globally, resulting in extensive campaigns aimed at both prescribers and the public and in the development of treatment guidelines.5

Attempts to reduce unnecessary antibiotic use have had mixed results. A combination of patient and physician education has been shown to help reduce antibiotic overuse for a variety of acute respiratory tract infections,6 but levels of improvement have been limited, on average, to less than 20% absolute reduction across the study populations of physicians.7–10

The Happy Audit project has been a study financed by the European Commission, the main objective of which was to strengthen the surveillance of respiratory tract infections in primary healthcare through the development of intervention programmes targeting general practitioners (GP) and changing people's habits towards prudent use of antimicrobial agents.11 GPs from six countries have participated in this study (Denmark, Sweden, Lithuania, Russia, Spain and Argentina). However, Spain was the only country in which two types of interventions were undertaken. We present in this final paper the overall results of the study in Spain [with the participation of semFYC] aimed at evaluating the effect of two types of interventions on antibiotic prescribing in respiratory tract infections.

This is a before–after quality assurance study carried out in primary care with family physicians in nine Autonomous Communities in Spain. Detailed information about the study method and the intervention were previously published in the study protocol.11 Briefly, the data were registered according to the methodology of the Audit Project Odense described by Munck et al.,12 which follows a prospective self-registration methodology in which a simple reporting sheet was used. Approval was obtained from the Ethical Committee Board Fundació Jordi Gol i Gurina (Barcelona, Spain; registration number: 44154). All participants were instructed to fill out a template with all respiratory tract infections during a 3-week period in the winter months of 2008 (preintervention). On this sheet, the physician attending the patient noted different specific parameters of medical care, including the age and sex of the patient, the number of days of symptoms, presenting signs (fever, coughing, otorrhoea, odynophagia, tonsillar exudates, tender cervical glands, dyspnoea, increase in sputum volume, purulence of sputum), diagnosis (common cold, otitis media, pharyngitis, tonsillitis, influenza, acute bronchitis, acute exacerbation of chronic bronchitis or chronic obstructive pulmonary disease, pneumonia), aetiological suspicion (viral or bacterial), performance of chest radiograph and point-of-care tests – rapid antigen detection tests for detecting group A β-haemolytic streptococcus antigen and C-reactive protein rapid test in capillary blood, antibiotic treatment or not, allergy or not to penicillin, and whether the patient requested an antibiotic and referral to another healthcare setting (Fig. 1).

Figure 1.

Template used to register the respiratory tract infections (English version).

(0.58MB).

With the overall and individual results intervention activities were conducted including meetings with presentation and discussion of results and several training meetings on respiratory tract infections guidelines, information brochures for patients, workshops on point-of-care tests and provision of these rapid tests in the consultation. The point-of-care tests were specifically bought for this study and were provided free to the participants. All GPs participating in the study with the exception of Catalonia carried out the full intervention group; conversely, Catalan doctors underwent the same intervention excepted for the workshop on rapid tests and point-of-care tests were not provided (partial intervention group). The same registry was repeated in early 2009 (postintervention).

The data were analysed with the Stata v.11 statistical program. Bivariate analysis was carried out with the use of chi square tests for comparing categorical variables. For the calculation of the odds ratio of antibiotic prescribing after the intervention a multilevel logistic regression analysis was estimated with two levels: patients and GPs. Antibiotic prescription was considered as the dependent variable (yes/no). The model was also adjusted for covariables age, gender, signs and symptoms of patients, referral, demand of antibiotics and age, gender, and burden of GPs. Statistical significance was considered with p<0.05.

A total of 332 GPs were invited to participate in this study but 21 declined. Out of a total of 309 primary care physicians who performed the first registry in 2008, 281 completed the intervention and the second registry (90.9%), of which 210 were assigned to the full intervention group (89.4% of the GPs who were invited to take part) and 71 to the partial intervention group (73.2% of the GPs invited). Fig. 2 describes the flow of the GPs throughout the study. A total of 15,073 respiratory tract infections were registered in the first registry while 12,760 were registered after the intervention. The most common infection was common cold, with 11,190 infections (40.2% of all cases), followed by pharyngitis with 4189 cases (15.1%) and acute bronchitis with 3286 infections in total (11.8%) (Table 1). As shown in Table 2 the sign most frequently reported by GPs was cough, present in 20,996 contacts in both years (75.4%), followed by odynophagia (12,029 cases; 43.2%), fever (9030 cases; 32.4%) and increase of expectoration (5494 episodes; 19.7%).

Figure 2.

General scheme of the study.

(0.32MB).

Antibiotics were prescribed in 4178 patients with respiratory tract infections in the first year (27.7%; 95% CI: 27–28.4%). The diagnoses in which antibiotics were most likely to be prescribed by the 281 GPs of the first registry were pneumonia (89.5% of antibiotic prescribing), acute otitis media (88.6%), acute tonsillitis (88.4%), acute sinusitis (87.4%), and exacerbations of chronic bronchitis or chronic obstructive pulmonary disease (81.9%). The percentages observed for acute bronchitis, other respiratory tract infections, acute pharyngitis, influenza and common cold were 58.1%, 20.6%, 18.5%, 2.9%, and 2.5%, respectively. The odds ratio of antibiotic prescribing after the intervention was 0.99 (95% CI: 0.89–1.10) among physicians who undertook the partial intervention, without statistical differences being observed. However, the odds ratio was significantly lower only in acute bronchitis (0.61; 95% CI: 0.42–0.88). The odds ratio of antibiotic prescribing after the intervention was statistically lower among GPs assigned to the full intervention, with an odds ratio of 0.50 (0.44–0.57; p<0.001). This reduction was more marked in flu infection (0.01; 95% CI: 0–0.07), common cold (0.03; 95% CI: 0.01–0.06), acute pharyngitis (0.15; 95% CI: 0.09–0.25), acute tonsillitis (0.18; 95% CI: 0.09–0.37), and acute bronchitis (0.31; 95% CI: 0.20–0.47) (Table 3).

The use of diagnostic tests after the intervention was higher among GPs assigned to the full intervention group, except for the request of an X-ray test. Before the intervention rapid antigen detection tests were performed in 76 cases and C-reactive protein testing only in 4 other cases. On the other hand, rapid antigen detection testing was carried out in 1509 cases (16.6% of all the patients) and CRP was performed in 1045 patients (11.5%) in the second registry (Fig. 3). As shown in Fig. 4, GPs assigned to the full intervention using both rapid antigen detection tests and C-reactive protein were less likely to prescribe antibiotics than their counterparts who did not use them. Antigen detection testing was used in 1499 patients with a result being positive in 345 cases (22.6%). Antibiotics were prescribed in 338 cases of positive test results (98%) but only in 69 individuals out of the 1164 negative test results (5.9%). Similarly, antibiotics were mainly withheld when the CRP value was lower than 10mg/L; however, they were prescribed in more than half of the cases with CRP values over 10mg/L.

Figure 3.

Use of rapid tests and request for X-ray by GPs in the different groups after the intervention.

(0.06MB).

Figure 4.

Percentage of antibiotic prescribing in the full intervention group depending on whether the GPs used the point-of-care tests.

(0.15MB).

A growing literature on strategies to reduce the overuse of antibiotics demonstrates that multifaceted interventions tend to be the most successful.10,13 The results of our study suggest that efforts to reduce antibiotic prescribing may be more effective if GPs intervene actively in the clinical encounter by means of carrying out the point-of-care provided.

This study has some limitations. This study was not a clinical trial and neither the groups were assigned randomly, which may be considered the main limitation of the study. Clinical outcomes of the patients recruited were not taken into account and thus, it is not known whether the percentage of clinical failure or incidence of complications differed between the groups. Nonetheless, the registry sheet included the referral of patients to secondary care. In addition, several papers have not observed any differences in clinical outcomes between patients with respiratory tract infections who were treated with antibiotics and those who were not.14–16 Another limitation that should be regarded in this type of study is that the mere fact of performing an audit may influence prescribing habits. However, the reliability of the Audit Project Odense methodology demonstrated in various projects carried out in other European countries is very high and is very well correlated with actual prescribing in medical offices.17 Nevertheless, the greatest strength of this study is the large number of physicians included, with the inclusion of 27,833 respiratory tract infections. The low drop-out rate observed in this study can be considered as another strength, since less than 10% of the professionals who carried out the first registry left the study. Another strength of this project is inherent in the reality of our country in which point-of-care tests were not incorporated into primary care by the time this study was carried out and, therefore, the effect of their use could be better established.

Many studies have been performed to determine the effectiveness of different types of intervention in reducing the prescription of antibiotics. Not all interventions have achieved positive results particularly when used alone. According to the last review of the Cochrane Library, published in 2005, only interventions taking combinations including result feedback, interactive educational sessions and strategies aimed at patients, achieve a reduction in the prescription of antibiotics in supposedly viral respiratory infections.10 These most successful have focused on combined communication skills training with point-of-care testing and the use of interactive booklets during consultations with patients.18–21 The results of the Happy Audit study confirm these results, since only GPs assigned to the full intervention group significantly reduced the prescription of antibiotics. Van den Velden et al. also observed that those that achieved more reductions of antibiotic overuse were the use of C-reactive protein and the enhancement of communication skills.13 In our study, GPs using rapid tests were more likely to prescribe fewer antibiotics. As far as C-reactive protein is concerned, the physicians felt more confident not to prescribe antibiotics where the results were less than 10mg/L. A recent meta-analysis confirms these results.22 Similarly, negative rapid antigen detections tests were associated with only a 6% of antibiotic prescribing. Interventions that did not include the former are associated with contradictory results or achieve only modest outcomes.23 A meta-ethnographic assessment of different interventions concluded that it is important to allow GPs to reflect on their own prescribing, and to educate GPs about appropriate prescribing and the benefit of implementing it in practice, in order to enhance the acceptability of the interventions.24

In conclusion, this study demonstrates that an intervention consisting of the presentation and discussion of overall and individual results, meetings on guidelines on respiratory tract and information brochures for patients achieved modest reductions on the antibiotic prescribing for respiratory tract infections. When all these interventions were combined with physician's active participation in workshops on point-of-care tests and these tests were provided in the consultation, the reduction was more marked.

This project is a Specific Targeted Research Project (STREP) funded by The European Commission: DG SANCO under the Frame Program 6 (SP5A-CT-2007-044154). Contract number 044154.

CL reports having a grant from the Fundació Jordi Gol i Gurina for a research stage at the University of Cardiff. He also reports receiving research grants from the European Commission (Sixth and Seventh Programme Frameworks), Catalan Society of Family Medicine, and Instituto de Salud Carlos III (Spanish Ministry of Health). The other authors have nothing to declare.