The urgency of relieving severe physical pain reveals the serious limitations of the “QALYs gained” approach to measuring scale of impact. A person with terminal cancer treated with morphine for two months might remain highly disabled and in a very poor state of health, and gain only a fraction of a QALY, yet be spared two months of agony.
I’m not sure I follow this. QALYs allow negative values, so if morphine treatment increased health-related quality of life from, say, −0.5 to +0.1, it would gain 0.6 QALYs per year. Most/all currently-used value sets would give less weight than that to pain relief, but I don’t think that’s primarily because of their other health states. Extending that life would gain few QALYs, but that doesn’t seem to be your concern.
That said, it might depend on the method used to combine utility decrements for various health states. The most common approach to dealing with comorbidities is to multiply utility decrements, e.g. if the decrement for cancer is 0.4 and for pain is 0.6, you’d end up with (0.4*0.6) = 0.24 (assuming a baseline/counterfactual of full health). Maybe that’s what you were getting at?
I’m not sure I follow this. QALYs allow negative values, so if morphine treatment increased health-related quality of life from, say, −0.5 to +0.1, it would gain 0.6 QALYs per year. Most/all currently-used value sets would give less weight than that to pain relief, but I don’t think that’s primarily because of their other health states. Extending that life would gain few QALYs, but that doesn’t seem to be your concern.
That said, it might depend on the method used to combine utility decrements for various health states. The most common approach to dealing with comorbidities is to multiply utility decrements, e.g. if the decrement for cancer is 0.4 and for pain is 0.6, you’d end up with (0.4*0.6) = 0.24 (assuming a baseline/counterfactual of full health). Maybe that’s what you were getting at?