Many different designs of total hip arthroplasty
(THA) with varying performance and cost are available. The identification
of those which are the most cost-effective could allow significant
cost-savings. We used an established Markov model to examine the
cost effectiveness of five frequently used categories of THA which differed
according to bearing surface and mode of fixation, using data from
the National Joint Registry for England and Wales. Kaplan–Meier
analyses of rates of revision for men and women were modelled with
parametric distributions. Costs of devices were provided by the
NHS Supply Chain and associated costs were taken from existing studies.
Lifetime costs, lifetime quality-adjusted-life-years (QALYs) and
the probability of a device being cost effective at a willingness
to pay £20 000/QALY were included in the models. The differences in QALYs between different categories of implant
were extremely small (<
0.0039 QALYs for men or women over the
patient’s lifetime) and differences in cost were also marginal (£2500
to £3000 in the same time period). As a result, the probability
of any particular device being the most cost effective was very
sensitive to small, plausible changes in quality of life estimates
and cost. Our results suggest that available evidence does not support
recommending a particular device on cost effectiveness grounds alone.
We would recommend that the choice of prosthesis should be determined
by the rate of revision, local costs and the preferences of the
surgeon and patient. Cite this article: