Flash forward one borderline blood test result, several phone calls between myself, Lily, and the anesthesiologist, 14 emails, an office visit, and a completely normal stress test. Lily (and her heart) were in no better health, yet she was slightly less well off and more than slightly distressed by all the trouble she went through. I was disappointed that I’d failed her.
“If You Give a Mouse a Cookie,” Laura Numeroff’s classic children’s book, is a cautionary tale about the downstream consequences of a single, seemingly innocuous decision. You gave the mouse the cookie. Naturally, he wanted a glass of milk to go with it. Before long, the mouse was moving in and sharing your Netflix password.
This pattern is familiar to many of us, whether as doctors or as patients: A medical test spurs a “cascade” of phone calls, office visits, tests and treatments, each a logical, even inevitable, progression from the one before.
In a sense, this is how medical testing is supposed to work. But no test is perfect and every test has trade-offs — including false positives and incidental findings (results you weren’t looking for and are often better off not knowing). Some tests (like Lily’s) we even know to be low-value — meaning they have at best minimal benefit in a given situation.
Researchers and policymakers point to low-value medical services as a major source of U.S. health-care spending bloat — by a new estimate, up to $101.2 billion a year. Remarkably, this estimate does not even include the cost of cascades (my research team and others are working to fill that gap).
Our team started by looking at national Medicare bills for one of the most common procedures among older adults: cataract surgery. Studies clearly show that for most patients, preoperative tests such as EKGs don’t improve health or reduce complications from the low-risk, 20-minute procedure. Yet nearly a quarter of these cataract surgery patients still receive these low-value, preoperative EKGs each year.
Those extra cascade services cost 10 times that of the initial EKG — for a total of $38 million in one year across the United States. (Did some of these extra services help some of the patients who got them? No doubt. But others were most certainly harmed and on average, we know it was at best a wash.)
In another study using a cancer database linked to Medicare bills, Yale University researchers found that older men received the no-longer-routinely-advised prostate-specific antigen (PSA) tests for prostate cancer screening at a national cost of almost $450 million per year. Nearly three-fourths of this bill came from downstream biopsies and related complications.
One of my colleagues no longer remembers what prompted that first CT scan of his patient’s abdomen years ago, but he cannot forget what happened next. The scan showed an unexpected small, ill-defined mass abutting her kidney. A follow-up MRI and then a painful biopsy were just as ambiguous. So surgeons removed the mass — and the kidney with it — only to discover it was a harmless piece of fat. The devastating conclusion: Her remaining kidney failed soon after.
When our team surveyed doctors across the United States about cascades from incidental findings, we discovered how common such stories were.
Cascades are common because incidental findings are common. With widespread screening (including notoriously low-value, head-to-toe scans) and improvements in technology, laboratory and imaging tests are increasingly likely to pick up small abnormalities that are most often inconsequential. And once that result is there, it is hard to look away.
Doctors and patients alike are also swayed by our cognitive biases — like that one time in a thousand when that tiny speck turned out to be cancer and has haunted us ever since. Even though “nothing” cascades are far more common, we may brush these off as happy near-misses.
What’s becoming clear is that we also need better ways to navigate cascades once they begin — to maximize any upside and minimize the harm. For patients, that means understanding tests come with trade-offs. For doctors, it’s setting expectations about what tests might reveal, framing results for what they are, and working with patients to decide the next steps informed by evidence and patient preferences.
Ishani Ganguli is an assistant professor of medicine at Harvard Medical School and an internist at Brigham and Women’s Hospital.
Leave a Reply
You must be logged in to post a comment.