Low-Value Care

The most-cited accounting of US health spending puts total waste at roughly a quarter of the system — hundreds of billions of dollars annually — with overtreatment and low-value services alone estimated in the tens of billions, and the OECD reaches similar proportions across rich countries: around one-fifth of health spending contributes little or nothing to outcomes. Low-value care is the clinical core of that waste: tests, drugs and procedures delivered to patients who cannot benefit from them — imaging for uncomplicated back pain, antibiotics for viral infections, pre-operative test batteries before cataract surgery, annual ECGs in the asymptomatic, vitamin D screening of the general population. Each item looks small; the aggregate is one of the largest line items in any health economy.

The strongest evidence that entrenched care can be worthless comes from the trials nobody wanted to run: arthroscopic surgery for knee osteoarthritis — once among the commonest orthopaedic operations — performed no better than sham surgery in 2002, a result reinforced by later trials of meniscectomy; vertebroplasty for painful spinal fractures matched sham injection in two simultaneous 2009 trials; percutaneous stenting for stable angina failed to beat a placebo procedure on exercise capacity in 2017, after a 2007 trial had already shown no reduction in death or infarction over medical therapy. Each result was contested by the specialty concerned, each procedure declined slowly and partially, and each remains performed today at meaningful volume — the natural history of low-value care in three case studies.

Low-value care is overdetermined: fee-for-service pays for doing, not for declining; defensive medicine converts liability fear into test orders; patient expectation equates action with caring — and satisfaction scores punish the physician who explains why not; training instils intervention bias; cascade effects convert one unnecessary scan into a chain of incidental findings, each demanding follow-up; and every service line, once established, is somebody's revenue, staffing and identity. Choosing Wisely — the 2012 campaign in which dozens of specialty societies each named their own top low-value practices, eventually producing hundreds of recommendations across many countries — proved the profession can say the words; a decade of evaluations finding modest practice change proved that naming is not stopping.

Removing established care is now a research field of its own, and its central finding is asymmetry: adding a new practice requires evidence and enthusiasm, while removing one requires overcoming loss aversion, income effects, patient attachment and institutional inertia simultaneously — with the evidential bar set higher for stopping than it ever was for starting. What measurably works is structural, not educational: payment reform that stops rewarding volume, decision-support defaults that make the low-value order the effortful one, audit-and-feedback showing clinicians their outlier status, and public reporting. Campaigns of awareness alone consistently underperform; the systems that changed defaults changed practice.

The deepest misreading of this topic is that unnecessary care is merely expensive. Every low-value test carries direct harm potential — radiation, contrast, false positives — and initiates cascades: the incidental adrenal nodule, the borderline PSA, the whole-body scan's dozen findings, each generating biopsies, anxiety and occasionally injury in people who were well; unnecessary antibiotics drive resistance at population scale; and every unit of clinical capacity spent on the well is unavailable to the sick — the opportunity cost that ties this hub to underuse. Globally the picture is a double burden: the same world contains systems overproviding scans to the worried well and systems where basic surgery, imaging and essential medicines never arrive — overuse and underuse as twin failures of the same misallocation.

For patients, the internationally promoted questions do real work: Do I really need this test or procedure? What are the risks? Are there simpler, safer options? What happens if I do nothing? A clinician comfortable with those questions is practising well; one affronted by them is telling you something. For clinicians: the evidence-based defaults are public — imaging rules for back pain and head injury, antibiotic guidance, pre-operative testing standards — and the professional skill of this decade is the confident, kind no. For systems: pay for value, default to restraint, and measure what stopping saves. Less medicine, correctly targeted, is more health — that sentence is the entire field.