Changing diagnosis coding routines may confound the results of longitudinal childhood pneumonia studies.

Aim: This Swedish study compared the discharge diagnosis codes used for children up to the age of five hospitalised for acute lower respiratory tract infections before and after the introduction of the pneumococcal conjugate vaccine in 2007.

Methods: The International Classification of Diseases-10th revision codes were used. We compared the discharge diagnosis codes at the Astrid Lindgren Children's Hospital from 1 July 2005 to 30 June 2007 (n=1,127) and 1 July 2011 to 30 June 2013 (n=1,240) in relation to the diagnostic methods used.

Results: There was a 54% reduction in the rate of all-cause pneumonia from the first to the second period in children aged 0-1 years, but some of this could have been due to the improved diagnosis of viral infections and us changing the code for respiratory syncytial virus infection from pneumonia to bronchiolitis. The overall rate of acute lower respiratory tract infections was unchanged.

Conclusion: We could not determine how much of the reduction in bacterial pneumonia in children under one was because of the introduction of the pneumococcal conjugate vaccine, based solely on discharge codes. Longitudinal register studies should take changes in diagnosis codes into account.