Health technology appraisal of interventional procedures: comparison of rapid and slow methods

To describe a method for the rapid appraisal of new interventional procedures and to compare its conclusions with those derived from a slower, more thorough method.

Explanation of an algorithm, pragmatically developed over a decade at the British United Provident Association (BUPA), to classify requests for funding for new interventional procedures as 'Fund routinely'; 'Fund as a one-off'; 'Fund in trial only'; 'Do not fund currently' within about 48 hours. Comparison of the resulting categorizations of 39 interventional procedures against the subsequent work of the English National Institute for Health and Clinical Excellence (NICE) Interventional Procedures team. The first two BUPA categories were equated with NICE's 'evidence adequate' and the second two with 'evidence inadequate'.

The algorithm is fit for purpose. It facilitated 114 requests for funding, received before June 2005, being successfully allocated: fund routinely, 33 (28.9%); fund as a one-off, 20 (17.5%); fund in trial only, 37 (32.5%); do not fund, 24 (21.1%). NICE subsequently categorized 18 being 'evidence adequate' and 21 'evidence inadequate'. There was concordance between BUPA and NICE on 35/39 (90%) of the topics. The four discrepancies are discussed.

Rapid appraisal of new interventional procedures using the BUPA algorithm is feasible and in most instances its output is similar to that obtained from a slower more thorough method.