C-reactive protein values in viral respiratory infections

We welcome the paper by Melbye et al on the course of C-reactive protein (CRP) response in upper respiratory tract infection.1 It provides valuable insight, broken down by virus type. However, the design of the study may possibly limit the generalizability of its results.

From the title and the abstract we are tempted to believe that all episodes that were treated with an antibiotic, were excluded. However, it is not clear to us what the indication for antibiotic prescription was. In the methods section, we read that, ‘subjects were excluded if a bacterial infection was suspected and antibiotics were prescribed’. The ambiguous word in this sentence seems to be ‘and’. When patients were prescribed antibiotics, did the investigators verify that it was a bacterial infection or did they assume that the GP had thought this?

Without doubt, the authors are aware of the fact that GPs frequently prescribe antibiotics for diseases of viral origin. The seriousness of the disease might well have played a role in the decision to prescribe antibiotics, indicating that those patients who were seriously ill were not included. Perceived patient preference is another reason. Further studies are needed that take a more comprehensive view on CRP in respiratory tract infections presented to general practice.

• © British Journal of General Practice, 2005.