The principle of double effect

A child has metastatic cancer with no hope of cure. The metastases to the bone cause severe pain. Morphine is given with some relief but she continues to complain of pain.
More morphine can be given, but will likely result in cessation of breathing and death. Is it ethical to increase the morphine dose? Would it be ethical not to?

The offical, bioethicist-approved answer is to increase the morphine even if it hastens death. The reason this is okay is something called the principle of double effect. You can order more morphine only for the purpose of relieving pain. The adverse, or double, effect of bringing on death is tolerable only because the medicine is given for another, purer purpose.

I basically agree with this formulation, but it has some problems. First, what about patients in severe pain with better outlooks. If the child instead had a compound leg fracture, no one would ever think about giving enough morphine to kill her, because her long term prognosis is better. Second, theory and practice can be quite different. In the few cases like the original I've been involved in, things went more or less according to plan, although titration of the opiate dose was more to parent's assesment of difficulty breathing than any objective measure of pain. I have heard other's tell of experiences in adults where the opiate was regularly increased with the understanding on the team that the point was to stop the patients breathing, not to relieve pain. Residents were admonished for not turning up the infusion fast enough, even though there was no evidence the patient was in pain. Of course, the result is indistnguishable (terminal cancer patient gets lots of morphine and goes out in comfort) whatever the intent of the ordering doc.

I don't have a better way of looking at things, and generally feel that pain in under- not overtreated (though pain treatment in patients with terminal cancer is pretty good in my experience), so I'm not complaining, just commenting