• Bacterial infection
• pneumonia
• respiratory infection
• bronchiectasis
• bronchoscopy
• empyema
• lung cancer
• mesothelioma
• non-small-cell lung cancer
• pleural disease
• pneumonia
• clinical epidemiology
• tobacco and the lung
• viral infection
• asthma
• cytokine biology
• lymphocyte biology
• paediatric asthma

The H1N1 pneumonia cohort studied was a subset of a much larger cohort of adults hospitalised in the UK with confirmed pandemic influenza A/H1N1 2009 infection (FLU-CIN cohort, n=1046); part of that cohort has been described elsewhere.1 The depth and breadth of bacteriological testing of patients recruited into FLU-CIN was at the discretion of attending physicians; only 3 of 1046 patients had evidence of bacterial co-infection recorded, probably an underestimate of the true burden of bacterial co-infection. Patients with identified bacterial co-infection were similar in age to patients without co-infection (mean age 27.0 (SD 13.1) years vs 39.5 (SD 16.4) years). The inclusion of patients with bacterial co-infection in the cohort of patients with H1N1 pneumonia would be expected to reduce any differences between the two study cohorts if patients with co-infection are indeed clinically distinguishable.

We are not aware of any publications arising from the 2009 pandemic demonstrating that patients with H1N1 pneumonia and bacterial co-infection can be reliably differentiated from patients without co-infection on clinical or demographic grounds alone. Some groups have examined the role of procalcitonin in this regard,2 and ongoing investigations may provide additional information. Conclusions from studies of viruses other than H1N1/09 should be interpreted cautiously with regard to H1N1/09 disease patterns and vice versa.