Parachute approach to evidence based medicine

We read with interest the article ‘Parachute approach to evidence
based medicine.’ We feel this approach can also be appropriately applied
to interventions in the developed world, indeed this is an approach that
we naturally sometimes 'free fall' into. We would cite, as an example, the
adoption of CPAP (continuous positive airway pressure) in the treatment of
bronchiolitis. This treatment has been parachuted into this indication
with no trial evidence on the basis of extrapolation of extensive evidence
for CPAP in the safe treatment of respiratory failure in preterm infants.

We recently conducted a survey within the South West of England to
determine the feasibility of a randomized trial of CPAP for severe
bronchiolitis. We found that of the 190 infants treated with CPAP in the
South west 2000-05, 11.5% went on to require intubation (the principal end
point of a putative trial). We asked 30 experienced paediatricians to
estimate the effect of CPAP on intubation rates (mean estimate 30%
reduction). With this information we performed a sample size calculation.
This showed we would need to randomize 3026 infants with severe
bronchiolitis to a trial with 80% power, assuming no cross over (a design
which only 15% of surveyed paediatricians would accept). Given that only
approximately 40 infants reach the severity threshold in the south west
per year, which would multiply up to 500 p.a. in England and Wales. This
trial will never be done.

On this assessment UK paediatricians were completely justified to
parachute in behind evidence based lines.

Comment on BMJ. 2006 September 30; 333(7570): 701–703

Competing interests:
None declared