More information about text formats
Gupta and colleagues  present the epidemiology of pneumothorax in
England using national data from the Hospital Episode Statistics for
emergency hospital admissions, which in turn is reliant on information
provided from local National Health Service (NHS) trusts. However, little
is known of the reliability of these locally collected data on the
incidence of pneumothorax.
We prospectively evaluated data from The Ipswich Hospital NHS Trust
over a four-year period (2001 – 2004). Data for episodes of pneumothorax
were obtained from hospital records identified through the appropriate
discharge codes. Chest X-ray reports which demonstrated the presence of
pneumothorax, acted as the ‘gold-standard’ of which the reliability of
data was assessed, as every pneumothorax episode would have required
The number of pneumothorax episodes as recorded by the hospital
database was 135 compared to 210 reported on chest radiographs, indicating
that the former correctly identified 64% of pneumothorax episodes compared
to the latter.
Formal data recording of pneumothorax episodes is unreliable in our
locality, and if our experience is mirrored in other local NHS trusts,
this may well influence the national epidemiology of pneumothorax as a
whole in England. A more reliable method of identifying pneumothorax
episodes is now urgently needed.
(1) Gupta D, Hansell A, Nichols T, et al. Epidemiology of pneumothorax
in England. Thorax 2000;55:666-71.