An older patient with dementia is in the hospital and has trouble swallowing. A speech pathologist recommends thickening the liquids the patient drinks with starch or gum and specifies how viscous her tea, water or juice should be. Should it resemble honey? Or apricot nectar?

A doctor writes the order, and the discharged patient returns to her home or nursing facility. She may be drinking thickened liquids from then on.

The rationale is that this sludgy stuff prevents patients from drawing liquids into their lungs and from developing aspiration pneumonia. But does the practice work? Some geriatricians have doubted it for years.

Now, a large-scale study from the Feinstein Institutes for Medical Research in Manhasset, New York, has found that liquid thickening doesn’t actually help such patients.

This happens with some frequency: Medical practices so commonplace they rarely raise eyebrows turn out to have scant basis in fact.

“There are plenty of things we do in medicine that have no evidence,” said Dr. Matthieu Legrand, an anesthesiologist and critical care doctor at the University of California, San Francisco.

They continue, accepted but untested, “because we have always done them, so they just keep on happening.”

Legrand is the lead author of a new study examining another common policy: discontinuing certain blood pressure medications a few days before patients undergo major surgery.

Here’s a look at three common practices getting closer scrutiny.

Thickened liquids

About a decade ago, geriatricians at UCSF decided, as a one-day experiment, to try the same thickened liquids they order for patients.

“We got headaches. We got dehydrated,” recalled Dr. Eric Widera, among the participants and author of a recent skeptical editorial in JAMA Internal Medicine.

“We couldn’t do it for 12 hours. We were asking our patients with dementia to do it for the rest of their lives.” Yet, “there wasn’t any evidentiary basis” for the policy, Widera said.

Now, there is — and it doesn’t support the practice. The Feinstein researchers analyzed the medical records of nearly 9,000 older patients (average age: 86) hospitalized with dementia and swallowing difficulties. Their hospital diets consisted primarily of either normally thin or thickened liquids.

Matching the groups for key traits, researchers found no significant difference in length of hospital stay, readmissions or death rates. Those drinking thickened liquids were less likely to need ventilators but were actually more likely to develop pneumonia or other respiratory problems.

Moreover, drinking sludge “really takes a toll on their quality of life,” said Dr. Liron Sinvani, a hospitalist and geriatrician and the study’s senior author.

“We can’t say 100%, this is a wrong practice,” Sinvani said. “But we can question it. It’s not clear that what we’re doing is best for people.”

Interrupted blood pressure regimens

Between 25% and 50% of patients undergoing surgery are taking blood pressure medications called ACE inhibitors or angiotensin II receptor blockers, Legrand said.

For many kinds of surgery, patients are routinely advised to stop those drugs before the scheduled operation. Doctors fear that blood pressure will drop too low during the procedure, causing complications such as heart failure, stroke or kidney problems.

Without the medications, though, patients’ blood pressure could rise dangerously.

To determine what really happens, a study randomized 2,200 patients (average age: 68) undergoing a variety of noncardiac surgeries at 40 hospitals in France. Half of the patients continued using ACE inhibitors or ARBs until the day of surgery; half were told to stop taking ARBs 48 hours before the operations.

During the procedures, blood pressures were more apt to fall in the group continuing the drugs. “But the rate of complications was exactly the same,” about 22% in each group, said Legrand, lead author of the study, published in JAMA.

The groups had comparable rates of postsurgery heart attacks, strokes, sepsis, respiratory and kidney complications, intensive care admissions and deaths. A large international study and another in Britain recently reached similar conclusions.

Spine implant for back pain

The Food and Drug Administration reported in 2020 that about 50,000 spinal cord stimulators, devices intended to reduce chronic pain using electrical impulses, were being implanted annually — and that over four years the agency had received 108,000 reports of patient injuries, including 497 deaths, and malfunctioning stimulators.

Annual implantations are likely to have risen since, as doctors seek substitutes for prescription opioids. Do these stimulators work?

Here the dilemma is not lack of evidence but contradictory claims and conflicting findings, with researchers arguing about methodologies and results — a confusing landscape for patients desperately seeking relief.

Pain doctors consider a treatment effective if it reduces pain by half in 50% of patients. Potentially confounding such results, however, is the potent placebo effect.

For patients, “the more you have invested, the more likely you are to see an effect,” said Dr. Rita F. Redberg, a cardiologist at the University of California, San Francisco, and a co-author of a recent study in JAMA Neurology.

The new study analyzed insurance claims data for 7,500 patients (average age: 64) suffering chronic pain, most after failed back surgery. Over two years, the 1,260 patients using stimulators didn’t have lower use of opioids than did those pursuing conventional medical management.

“They wanted to feel better, but they didn’t,” Redberg said. Moreover, about 1 in 5 patients had the device removed or required a second surgery to repair or relocate it.

Two Cochrane reviews, meta-analyses by an independent network of researchers, have found “low to very low certainty evidence” that stimulation reduces pain intensity and “little to no sustained benefit” for low back pain.

Pain specialists and professional organizations were quick to criticize the new study’s methodology, however.

For now, disputes continue. “We all want to help patients with pain,” Redberg said. “This is not the way to do it.”