Epidemiologists have tended to equate osteoporotic fractures with those whi
ch occur after only moderate trauma, or which affect elderly people. We set
out to critically examine this epidemiological approach, because even amon
g elderly people some fractures will be a result of severe trauma and these
cannot be assumed to reflect bone fragility. In a population based study o
f all Cardiff residents aged over 60 we identified everyone who presented w
ith a fracture during 1996. We considered the events that led to each fract
ure, to establish which fractures were a result of 'severe' trauma: trauma
greater than a fall from a standing height. We identified 1335 people who b
etween them sustained a total of 1372 fractures. 168 (12%) of the fractures
resulted from severe trauma, but in people aged over 80 only 6% of all fra
ctures, 4% of hip fractures, and 3% of wrist fractures followed severe trau
ma. In older subjects severe trauma makes only a very small contribution to
the causation of fracture, so that the age-sex distribution of fractures f
ollowing moderate trauma closely resembles that of fractures overall. For e
pidemiological purposes it does appear justifiable to equate the consequenc
es of bone fragility with the occurrence of osteoporosis associated fractur
es in elderly people. (C) 1999 Elsevier Science Ireland Ltd. All rights res
erved.