Tests for trend were carried out for each frequency of use
variable by ordering the values from lowest to highest and including the resulting variable as
a one degree-of-freedom linear term in a Cox proportional hazards model. We also
examined combinations of the two agents, defined as follows: use of aspirin only, use of
non-aspirin NSAIDs only, use of both, and use of neither. We first assessed aspirin and
NSAID associations with overall breast cancer risk.
Bardia et al. Page 3
Breast