Riddle: What test is easy, fast, reproducible, cheap, can be done in any physician’s office and is a good predictor of mortality?
Answer: Gait speed.
As the authors of a short essay in JAMA argue, measuring how quickly a person walks 4 meters at his or her usual walking speed is just such a test. Since decisions from whether to screen for breast cancer to how tightly to control diabetes depend on prognosis, a simple way to estimate life-expectancy is very useful.
Detecting low gait speed should alert the clinician to think about possible causes: undiagnosed Parkinson’s disease, unrecognized heart failure, or other potentially treatable conditions. And monitoring gait speed may prove to be a good way to objectively assess the response to treatment.
But here's what's so surprising. Nobody measures gait speed. Open the medical record of any 85 year old patient and you won’t find gait speed recorded anywhere, though 52% of women over 85 have a gait speed less than the 0.6 meters/second cutoff that defines “dysmotility” as do 31% of men over 85.
There’s been a lot of interest recently in defining “low value medical care,” tests and treatments for which the ratio between cost and benefit is high. Here’s a high value test, one where the cost is measured in pennies and the benefit is at least moderately great. While it seems prudent for doctors to perform less low value care, maybe it would also be wise to provide more high value care.