Medpundit has a pretty good explanation of the recent study about herniated disk surgery. Basically it works, but if you wait two years, then the results are about the same as non-surgical treatment—roughly 70% of people get better, and there doesn’t appear to be any long-term harm from delaying surgery. As I have someone very close to me with a current case of extreme back and leg pain from a herniated disk, I’m very interested in the study, and actually more inclined to suggest surgery (especially arthroscopic) sooner rather than later. But in this case the patient, doctor and other advisors are more in favor of waiting it out.
So on a global level it’s more cost-effective not to do the surgery. But on an individual level it probably lessens the pain—and the pain is close to unbearable, and if you have to put up with it for several months, then surgery is probably an option the patient will want.
Note that this is only the case for herniated disks and not lots of the other back issues for which surgery is probably ineffective–but still done at a very high rate.
The good news is that ten years after AHCPR (the forerunner to AHRQ) was decimated by daring to discuss back surgery, we’re getting studies out about this type of issue. Even, as Medpundit points out, it’s not a great study and it’s very, very hard to do studies about this type of intractable medical problem.
CODA: One slightly disquieting anecdote. I asked a local back specialist (non-surgeon) what the best way of doing surgery was (open or athroscopic). He said that the choice depended mostly on the training of the surgeon! Er…shouldn’t the surgeon be trained in the most advanced manner? (I expect those who know to chime in here)