Objectives This study aimed to identify decisions made by orthopaedic surgeons about whether patients on a waiting list should proceed to arthroscopy, and to describe surgeons' decisions.

Methods Surgeons were asked to think aloud as they made their decisions as the clinical management for patients from a previous randomised controlled trial (RCT) looking at the use of MRI for patients on a waiting list for knee arthroscopy. Audiotapes of the decision making were transcribed for analysis.

Results Surgeons agreed about proceeding with arthroscopy for five patients, although reasoning differed. In no cases did surgeons agree about not proceeding to arthroscopy. However, more súrgenos decided not to proceed to arthroscopy than to proceed. Patients with clinically diagnosed with osteoartritis were less likely to be cosen to have arhtroscopy than those with meniscal abnormalities for which the general consensus was to proceed with arthroscopy..

Conclusions There tended to be disagreement between súrgeons about proceeding with arthorscopy when osteoarthritis was diagnosed clinically. Surgeons decisions did not reflect the decision making as reflected in the original RCT. Surgeons' decisions were influenced by patient wishes. For some patients, the decision to proceed with arthroscopy was based solely on clinical diagnosis; MRI may not be advantageous in these instances. This study has implications for decision making in the current NHS patient choice environment. Patients may choose a treatment provider from a list of available providers at time of original clinical assessment and diagnosis. However, the treating surgeon does not necessarily re-examine the patient until the day of surgery. Given the variation between surgeons about the merits of proceeding with arthroscopy, surgeons may end up in the invidious position of providing surgery to patients whom they do not believe will benefit from arthroscopy.