Several European countries advise doctors in primary care to do this, but the evidence is conflicting, with some studies suggesting benefit and others suggesting harm. Two papers in this week's BMJ add to this uncertainty.

One shows that children who are given antibiotics before admission to hospital are more likely to die on reaching hospital. The other a review of all the current evidence cannot conclude whether or not pre-hospital antibiotics improve survival.

In the first study, UK researchers analysed 158 children diagnosed with suspected meningococcal disease by a general practitioner before admission to hospital. Two thirds were given parenteral (injected) penicillin, in accordance with national guidelines.

The children who were given penicillin were more likely to die than those who were not given penicillin.

However, the children who received penicillin also had more severe disease on reaching hospital. So, although a harmful effect of penicillin cannot be excluded, a more likely explanation for the higher mortality is that there is a strong bias towards giving penicillin to the most severely ill children, conclude the authors.

In the second paper, an international group of researchers reviewed evidence from 14 studies and found that oral antibiotics given before hospital admission were associated with reduced mortality. Results for parenteral antibiotics were inconsistent, though the data suggest that they might have a beneficial effect when a substantial proportion of patients is treated.

Once again, it is suggested that bias linked to illness severity may explain these results. For example, doctors are likely to prescribe oral antibiotics only in patients with milder disease.

"We cannot conclude from this review whether or not antibiotics given before admission have an effect on case fatality, thoug
'"/>

Contact: Emma Dickinson
edickinson@bmj.com
44-207-383-6529
BMJ-British Medical Journal
1-Jun-2006