Articles
www.thelancet.com
Vol 388 October 1, 2016
1393
Non-parametric bootstrapping was used to generate
95% CIs
for the estimated diff erences in mean costs and
QALYs, and to ascertain the probability of the ICER for the
clear-lens extraction approach being cost-eff ective at
diff erent ceiling ratios. In accordance
with the guidance of
the National Institute for Health and Care Excellence,
28
we
report these probabilities at ceiling ratios of £20 000 and
£30 000 per QALY gained to represent decision makers’
maximum willingness to pay per QALY gained. All
analyses were also repeated with a multiple imputation
dataset (n=20) generated with the
use of chained equations
to deal with missing cost and utility data.
29
Analyses were
done with Stata version 12. This study is registered with
the ISRCTN registry, number ISRCTN44464607.
Chia sẻ với bạn bè của bạn: