Monday, August 12, 2019

Low value care

Doctors and policy wonks hear a lot these days about “low value care.” What exactly is that, and should you care? The term refers to tests, medications and procedures that add little to a patient’s health or well-being but which cost them or the system more money than any benefit warranted. Some of the earliest research that looked at this was done at Dartmouth, where they found enormous variations in the cost of treating various conditions in different parts of the country with no correlation with outcomes. In my own practice, I remember seeing a long-time patient who had become a snow bird. When he returned north and came to see me, he told me that he decided to hook up with a local doctor should he get sick while in Florida. The doctor did not have any available appointments, but it was suggested he come in and get an ECG and blood tests before the visit (inappropriate for an unknown patient!), which he did. The next day he got a call: his ECG was abnormal, showing he had had a heart attack, and so additional tests were scheduled. An echocardiogram and stress test were done, followed by a catheterization – all of which were normal. Fortunately, he suffered no adverse consequences beyond some anxiety, but this whole sequence, starting with a “routine” ECG, epitomized low value care.
Numerous examples abound. Complex imaging such as MRIs and CT scans for non-specific back pain rarely lead to any change in therapy and often find red herrings. Most of us over 60 have some abnormality on a back image, even those who have never had a backache in their life, and these may lead to totally inappropriate surgical procedures. Tests for various rheumatologic disorders, such as ANA (anti-nuclear antibody, a screening test for lupus) and Rheumatoid factor, elevated in rheumatoid arthritis, are often positive in healthy elderly people. When these tests are ordered for people who have osteoarthritis, the meaningless positive test may lead to unhelpful and even dangerous treatments for a disease they do not have. PSA testing in men well over 75 may discover low grade prostate cancers that would not have bothered them if never discovered but which lead to treatments with serious and life-long side effects, lowering their quality of life. Other tests, such as imaging for plantar fasciitis or CT scanning for uncomplicated sinusitis do little harm but are costly and add nothing to change in therapy.
The difference between settings and individual doctors are remarkably large, with the use of studied low value services varying as much as 8-fold. Researchers have been unable to find any obvious characteristics that differ between those who order a huge number of such tests and procedures and those who do not: neither age, gender nor medical school seem to affect this behavior. The solution lies in both continuing education of the doctors and perhaps in harnessing the electronic record to request that certain orders be justified before proceeding. Patients, too, should play a role, and not request tests that the doctor feels are not needed after some discussion about the usefulness of the test. There have been various attempts to estimate how much of our bloated health care expenses represent waste, but a growing consensus is that 20-25% of what we spend goes for low value care. To get a handle on costs, this would seem to be the “low hanging fruit.”

Prescription for Bankruptcy. Buy the book on Amazon


  1. Several easy studies:
    Is there a relationship between productivity-based care and test ordering? (I.e., do payment system reinforce this type of test ordering?)
    I work in a system that uses the automatic "education" screens. How much time is spent looking at/reading the screen? About as much as we spend reading the EULA on internet sites?

  2. Easier to order a test (or lots of tests) than to think! And it takes less time, which is a huge consideration with 15 minute appointment slots.