There are continuing advances in severity assessment and antibiotic therapy for community acquired pneumonia (CAP). The British Thoracic Society (BTS) updated its national guidelines on adult CAP management in 2004.1 2 A study was undertaken to examine local and national influences on guidelines used in UK hospitals for the management of adult CAP.

Methods

A questionnaire was sent to each of the 254 hospitals in the BTS directory in September 2006 (see fig 1 in online supplement). Data were analysed using Microsoft Excel. Differences in categorical variables were tested for statistical significance using the χ² test with the Fisher exact test.

Results

The response rate was 60% (n = 152); 92% of hospitals (n = 140) had locally written CAP guidelines (although in only 100 were policies being used in the emergency department as well as the medical department), 5% (n = 7) used the national BTS CAP guidelines and 3% (n = 5) had no guideline. Sixty-eight guidelines (49%) had been updated in the previous year and 88% (n = 123) had been updated since the BTS 2004 CAP guideline.

Self-reported concerns over healthcare acquired infections influenced local guidelines in 57 hospitals (Clostridium difficile (n = 57), meticillin-resistant Staphylococcus aureus (MRSA) (n = 22)). Other influences included the BTS 2004 guideline (n = 96), cost of antibiotics (n = 26) and local antibiotic profiles (n = 15).

Using severity assessment for planning management was recommended in 94% of guidelines (n = 131/140) including CURB65 tools3 in 76% (n = 106), CURB4 in 18, other tools in 9 and clinical judgement only in 9.

First-line antibiotic recommendations for non-severe and severe CAP as stated in local CAP guidelines are shown in table 1. For managing non-severe CAP, 61% of hospitals (n = 85) recommended amoxicillin plus macrolide and 24% (n = 34) recommended amoxicillin alone. For severe CAP, recommended first-line antibiotics were consistent with BTS recommendations in 87% of guidelines (113/130; no data from 10 hospitals) including a β-lactamase stable β-lactam plus a macrolide in 101 and the alternative BTS recommendation of a quinolone and β-lactam in 12. A simple β-lactam, such as amoxicillin or penicillin, plus a macrolide was recommended in 12 guidelines and other antibiotic choices in 5. In hospitals with C difficile concerns, cephalosporins were less commonly recommended as preferred treatment for CAP than in other hospitals (26% vs 47%, p = 0.01). As alternative therapy for severe CAP, 19% of hospitals (n = 27) recommended β-lactamase stable β-lactam plus macrolide combinations and 36% (n = 51) recommended quinolones (most commonly levofloxacin (n = 33), ciprofloxacin (n = 11) and moxifloxacin (n = 4)), mostly as combination therapy (n = 27). In 35 guidelines no alternative regime was stated, 6 recommended microbiology advice and 21 recommended other choices.

Discussion

This survey of 152 hospitals confirms that UK local guidelines for the management of adult CAP are widespread (but not always used in the emergency department), up-to-date, use severity assessment tools and are influenced by both national evidence-based guidelines and local factors, especially healthcare acquired infections, cost and local antibiotic profiles. Compared with 1999,5 the proportion of hospitals reporting C difficile infection as an influence on local CAP guidelines has increased significantly (from 19% (39/213) in 1999 to 41% (57/140) in 2006; χ² = 21, p<0.001). The response rate for this survey was only 60%, but there was no obvious difference between responders and non-responders.

This survey confirms the value of having national guidelines for common conditions, which can act as a framework to be adapted for local use.

Acknowledgements

The authors are grateful to all the BTS members who responded to the survey.