Many Surgical Procedures No Better Than Placebo
As Acupuncture trials have to work triple time to counter the charges of placebo, surgical procedures have skated for years, with no concern for whether the surgery actually helped or not.
In my book, Radical Resilience, I contend that this is partly because of the “tyranny of the logical”, or how David Epstein calls it the “bio-plausibility.” What that means is, if something makes sense that it should work, like a spinal fusion, it’s hard to believe that it doesn’t.
The Economist has a recent article featuring recent studies showing that certain surgical procedures are no better than placebo. More than 300 million operations are performed worldwide each year, but growing evidence challenges the assumption that surgery necessarily delivers benefits beyond natural healing, placebo effects or nonsurgical treatment. Although some operations clearly save lives, many elective procedures lack rigorous evidence.
The article identifies several procedures whose benefits have disappointed in trials:
- Knee surgery: Arthroscopic flushing and removal of damaged tissue relieved pain no better than sham surgery. A separate Finnish trial found worse knee outcomes ten years after real surgery than after a sham procedure. (I talk about that in my book. The surgical patients go on to have more osteoarthritis than the control group.)
- Shoulder bone-spur removal: This operation reduced pain no better than sham surgery.
- Spinal fusion and rotator-cuff repair: Some trials found outcomes indistinguishable from placebo or nonsurgical care, including physiotherapy.
Other findings concern alternatives and harms rather than placebo comparisons. Antibiotics can match appendectomy for many patients.
In a prostate-cancer trial, surgery did not reduce mortality after fifteen years but increased sexual and urinary problems.
A 2022 analysis of 100 surgical trials found that, on average, approximately two-thirds of patients’ improvement reflected natural recovery or placebo effects; only one-third came from the procedure itself. Where researchers could separate these influences, natural improvement explained most of the benefit.
Changing practice remains difficult. Surgeons may overestimate success, resist evidence contradicting years of experience, or mistake age-related abnormalities on scans for the cause of pain. Patients understandably want a quick fix.
Fee-for-service payment creates a perverse incentive: surgeons earn money by operating, while deciding against surgery forfeits that revenue. Ineffective procedures decline faster where surgeons receive salaries. After negative trials, shoulder bone-spur operations fell sharply in Britain but remained popular in America. Fear of poor reviews and insurers’ concerns about losing customers also sustain demand.
The article’s central message: improvement after surgery does not prove surgery caused it. Stronger trials and willingness to question established practice are essential to preventing harm.
I discussed the perverse incentives that western medical systems face in my book Radical Resilience, in the chapter called Do No Harm.
“Remedies that seem logical according to this way of thinking do not necessarily work as intended, a pitfall New York Times bestselling author David Epstein, dubbed bio-plausibility. In a nutshell, it refers to treatments that make logical sense to the physician but may not work.”
The mental model behind it:
“And while the body is a sophisticated system of multiple feedback systems that can lead to homeostasis, it is often considered by modern medicine to resemble a machine whose individual parts need fixing.”
“In this man-as-machine model, the thinking goes that the machine needs something done to it; otherwise, it will not get better. But this can lead to incorrect diagnosis and treatment options.”
Why less invasive approaches get overlooked:
“Since the model of man-as-machine ignores the power of homeostatic feedback systems that can react to subtle inputs, it is difficult for many to believe that small changes, or minor treatments, can be enough.”
And the directly relevant line from “Follow the Money”:
“In this context, it is difficult to get a surgeon to stop doing a useless procedure when he makes his money doing it.”
Entire article follows.
Many surgical interventions are little better than placebo
A succession of clinical trials is upending the field
ACH YEAR, according to the World Health Organisation, over 300m operations are conducted worldwide. Patients in the rich world are more likely to go under the knife—about 60% of people in England are expected to undergo surgery at least once in their lifetime. Some of these procedures are clearly life-saving, such as organ transplants and emergency Caesareans. But as the number of other operations continues to grow, some in the field are starting to ask an awkward question: how many are really necessary?
Answers have long been hard to come by, in part because few randomised controlled trials (RCTs), the gold-standard type of research used to assess medications, were conducted on surgical procedures. That made it difficult to know if patients got better more often with surgery than they might have without. This is now changing. The number of surgical RCTs funded by Britain’s National Institute for Health and Care Research increased from 34 in 2011 to 188 in 2023, and national surgical-trials programmes are active across Europe as well as in Australia and Canada.
Worse, some types of surgery may do more harm than good. A trial in Britain found that prostate-cancer surgery had no effect on patient mortality 15 years on, but worsened sexual and urological problems. A Finnish trial of a common knee procedure, meanwhile, concluded that patients who had real surgery had more problems in the affected knee ten years on than those who had had sham surgery (anaesthesia followed by a superficial incision). David Ring, an orthopaedic surgeon at the University of Texas at Austin, reckons that most operations in his field may be unnecessary. Surgery itself, it would seem, could use a major intervention.
If so many operations are unnecessary, how to explain the fact that patients often leave the operating table feeling better? In a study published in 2022 in JAMA, a journal, researchers pooled data from 100 surgical trials covering 32 interventions. They found that two-thirds of the improvement patients felt after surgery was due, on average, either to the body healing on its own or else to the placebo effects of feeling cared for and prepped for surgery. Only one-third of the benefit, in other words, came from the procedure itself.
Natural healing is also powerful, if slow-acting. In the 2022 study in JAMA the researchers compared real and sham surgeries with standard non-surgical care for a subset of procedures on which data were available. They found that the placebo effect could explain only a small part of the improvement in the sham-surgery group. Most of the benefit came instead from natural improvement.
Achieving changes in clinical practice has proved difficult. Surgeons continue to perform some operations that have been shown to be no better than placebo surgery at the same rate as before. They find it hard to believe that something which they have been doing for years and seems to help their patients is, in fact, useless, says Ian Harris, a surgeon at the University of New South Wales, in Sydney. That occasionally leads to what he describes as outlandish criticisms. In 2014, for example, a group of surgeons wrote a commentary on the use of sham surgery in trials in Arthroscopy, a medical journal, in which they worried that “Patients who may not be of entirely sound mind are selected as research subjects, and research performed on such individuals would not be generalisable to mentally healthy patients.”
Operating theatre
“You have to remember surgeons have no uncertainty,” says Jane Blazeby, a surgeon at the University of Bristol who has conducted many such trials. “They always know what’s best,” she adds, archly. This certainty is not always justified: research has shown that, for one common type of knee surgery, surgeons’ predictions about whether a patient will improve is as good as flipping a coin. Surgeons who have been doing a procedure for years also tend to overestimate its benefits, says Stefan Lohmander from Lund University, in Sweden, because the patients who return for check-ups are mostly those who got better. Moreover, many of these returning patients do not want to disappoint their doctors, says Dr Lohmander—which may, in part, be why surgeons often feel procedures are more successful than their patients do.
For Dr Blazeby, convincing her colleagues to enroll their patients on randomised controlled trials required teaching them to “confidently be unconfident”, she says. Such trials required them to tell patients that an operation they had been doing for years might not be the right choice for them. Encouragingly, younger surgeons are more open to change. Studies in America and Australia, for example, have found that the probability that a surgeon stops performing a procedure a trial has shown ineffective decreases with time. And when trial results reinforce each other, scepticism becomes less tenable.
But it is not just surgeons who need to trust the data: a growing number of patients insist on surgery on the basis of inconclusive tests. All too often, says Dr Ring, scans reveal abnormalities that are little more than tissue tears; as natural a by-product of ageing as wrinkles and grey hair. What’s more, he says, these things are unlikely to be what is causing patients’ problems. A growing number of studies, in fact, show that the abnormalities in knees, backs and shoulders that orthopaedic surgeons try to fix are very common in people who have no symptoms at all.
But the promise of a quick fix is hard to resist. “It is becoming more and more difficult to say ‘no’ to the patient,” says Dr Lohmander. In America some surgeons fear their patients will leave them a bad review online, and insurance companies still cover ineffective procedures for fear of losing customers to their competitors.
Even more trials should help doctors hold their ground more firmly. For now, most non-essential surgical techniques remain untested against either sham surgery or non-surgical interventions. A review in 2020 found that just 1% of the RCTs on common surgical procedures conducted for chronic musculo-skeletal pain compared doing the surgery to not performing the procedure at all. To fulfil their Hippocratic promise of doing no harm, surgeons must be prepared to ask how much good they are actually doing. ■