We know at least some widely used surgeries are probably no better than a placebo. "Placebo" in surgery is when they get a surgical crew together, anesthetize the patient, make an the appropriate incision, and then dick around for the time it would take to perform a surgery, maybe passing some tools back in forth.
Basically, an elaborate ritual.
https://fivethirtyeight.com/features/surgery-is-one-hell-of-...
>A 2014 review of 53 trials that compared elective surgical procedures to placebos found that sham surgeries provided some benefit in 74 percent of the trials and worked as well as the real deal in about half.1 Consider the middle-aged guy going in for surgery to treat his knee pain. Arthroscopic knee surgery has been a common orthopedic procedure in the United States, with about 692,000 of them performed in 2010,2 but the procedure has proven no better than a sham when done to address degenerative wear and tear, particularly on the meniscus.3
>Meniscus repair is only one commonly performed orthopedic surgery that has failed to produce better results than a sham surgery. A back operation called vertebroplasty (done to treat compression fractures in the spine) and something called intradiscal electrothermal therapy, a “minimally invasive” treatment for herniated disks and low back pain, have also produced study results that suggest they may be no more effective than a sham at reducing pain in the long term.
There's even the question of is the pain even caused by the thing we are trying to fix?
>And then there’s what Thorlund calls “car repair” logic — something looks broken, so you try to fix it. A patient comes in with knee pain, and an X-ray or MRI exam shows a tear in the meniscus. The tendency is to assume that the torn meniscus is the cause of the pain and so should be fixed. However, studies show that MRIs can find all kinds of “abnormalities,” such as cartilage damage, even among people without knee pain. One such study looked at the MRI scans of more than 300 knees and found no direct link between meniscus damage and pain. “You can have a meniscal tear without having any problems,” Thorlund said.
>Back pain follows a similar pattern. Studies that examined MRIs of people’s backs show that things like slipped, bulging or herniated disks correlate very poorly with pain. Herniated disks and other supposed abnormalities are also common in people without back pain, and it’s telling that studies find that spinal fusion, another popular back surgery used to address disk problems, does not produce better results than nonsurgical interventions.