Assessing clinical decision making: is the ideal system feasible?