In reply: The “transfer to utility” TTU technique was devised to compare disparate interventions, using published clinical trial literature, where utility data are not reported. While collection of utility data in clinical trials would be preferable, until this occurs routinely a means to translate reported quality-of-life scores into utility scores is highly useful. Other groups are also grappling with this.1

In developing the TTU weights, various sophisticated statistical techniques were explored. However, added complexity did not improve the estimates. Contrary to the suggestion by Viney and colleagues, the TTU estimates are highly plausible and were vetted by our reference panel of clinical experts. Taking the example Viney and colleagues cite, a 0.304 increase in utility score for hip replacement indicates a patient would, on average, be willing to forgo 30.4% of remaining life-years to obtain the benefits of surgery — not 40% as incorrectly stated by Viney et al. This is consistent with the large increase in well-being observed following hip replacement (in the seminal trial, SF-36 mean scores increased from 26.9 to 66.6 for physical function, 14.6 to 58.7 for role physical, and 32.9 to 72.8 for bodily pain).

Whether the TTU introduces bias (a characteristic of other summative approaches to estimating health, such as the popular DALYs) is a matter for future research. Undoubtedly, all population-wide approaches suppress specific information, but the purpose of the priority-setting model supported by the TTU is not to provide information on individuals, but to compare management options to give clinicians and policymakers another way of understanding the comparative performance of disparate interventions.