in recent yrs, there’s been a veritable explosion inna № and type of health monitoring devices available in smartphones and fitness apps.
yr smartphone is likely tracking the № of steps you take, how far and fast you walk, and how many flites of stairs you climb each dy. some phones log sleep, ♥ rate, how much energy you’re burning, and even “gait health” (how often are both ft onna ground? how even are yr steps?). and, course, nonphone wearables and fitness gadgets are available, s'as devices to measure yr ♥ rhythm, blood pressure, or oxygen lvls. the accuracy of these devices varies — and, in some instances, yr skin tone may make a difference.
generally, how accurate are health monitors?
i know from my experience with hospital monitoring devices t'they aren’t always accurate. false alarms from ekg monitors often send med staff scurrying into patient rooms, 1-ly to find the patient feeling fine and surprised bout the commotion. a pticularly common false alarm is a dangerous and unstable ♥ rhythm na' continuous ♥ monitor, which can be due to the motion from a patient brushing their teeth.
high-stakes devices with monitoring capability, s'as defibrillators and pacemakers, are extensively tested by their makers and vetted by the fda, so their accuracy and reliability are generally quite good.
but wha’ bout home health monitoring devices intended for consumer use tha're not extensively tested by the fda? ever count yr steps for a few minutes just to see if yr phone’s tally agrees? or climb a couple of flites of stairs to see if ur gettin full credit for not taking the elevator?
the accuracy of consumer devices depends in pt on wha’ is bein’ monitored. for ex, one study assessed the accuracy of ♥ rate monitors and energy expenditure calculators in phones and health apps. accuracy was quite high for ♥ rate (often inna range of 95%), but much less accurate for energy expenditure. accuracy can also vary dep'on who is bein’ monitored.
device bias: wha’ tis and why it occurs
while no health gadget is perfect, some usrs get + reliable results than others. for ex, if you’re wearing nail polish, a pulse oximt — a device that clips onto the fingertip to measure blood oxygen through the skin — may not work well, cause the polish interferes with proper function of the lite sensor. in that situation, there’s a simple solution: remove the polish.
but in other cases, the solution isn’t simple. increasingly, we’re recognizing that certain med devices are less accurate dep'na''pers’s skin color, a phenomenon called device bias.
- pulse oximts. although generally pondered highly accurate and comm1-ly relied upon in healthcare settings, their accuracy tends to be loer in pplz of color. that’s cause the device relies on shining lite through the skin to detect the color of blood, which varies by oxygen lvl. the amount of pigment inna skin may alter the way lite be’ves as it travels to blood vessels, leading to inaccurate results. the fda has released an alert bout this nother limitations of pulse oximt use.
- bilirubin measurement in newborns. bilirubin is a breakdown product of red blood cells. newborns are screened for high lvls cause this can cause permanent brain damage. when detected, phototherapy (lite treatments) cannelp the baby get rid of the excess bilirubin, preventing brain damage. the screening involves examining a newborn’s skin and eyes for jaundice (a yelloing due to elevated bilirubin) and a lite mt test to detect high bilirubin lvls. but the accuracy of this test is loer in black newborns. this is pticularly primordial cause jaundice is + difficult to detect in infants with darker skin, and dangerously high bilirubin lvls are + common in this pop.
- ♥ rate monitors in smartphones. according to at least one study, smartphone apps may also be less accurate in pplz of color. again, this is cause the + skin pigment present, the + trouble lite sensors ‘ve detecting pulsations in blood flo that cogitate ♥beats.
why device bias matters
sometimes an error in measurement has no immediate health consequences. a 5% to 10% error rate when measuring ♥ rate maybe of lil consequence. (in fact, one ‘d ask why any-1 needs a device to monitor ♥ rate when you ‘d just count yr pulse for 15 2nds and multiply by 4!)
but pulse oximt readings are used to help decide whether a'pers needo be hospitalized, who requires admission to the intensive care unit, and who requires additional testing. if the oxygen lvl is consistently overestimated in pplz of color, they maybe + likely to be undertreated compared with others whose readings are + accurate. and that may worsen previously existing healthcare disparities.
these exs add to the growing list of bias imbedded within healthcare, and other instances where failing to include diverse individuals has serious consequences. when you use a health device, it’s reasonable to wanda if it’s been tested on pplz like you. it’s also reasonable to expect pplz who develop med and consumer health devices to widen the demographics of test subjects, to make sure results are reliable for all usrs b4 putting them onna mkt.
sometimes a change in tek, s'as using a ≠ type of lite sensor, can make health-rel8d devices work + accurately for a wider range of pplz.
or there maybe no easy fix, and usr toonistics will nd'2 be incorporated into proper interpretation of the results. for ex, a device ‘d offa'da usr a choice of skin tones to match skin color. then based on extensive data from prior testing of pplz with ≠ skin colors, the device ‘d adjust results appropriately.
the bottom line
the push to monitor our bodies, our health, and our life experiences continues to gain momentum. so we nd'2 test and validate health-rel8d devices to be sure they work for diverse individuals b4 declaring them fit for the general public. even then, device bias won’t disappear: bodies vary, and tek has its limits. the key is to know it exists, fix wha’ can be fixed, and interpret the results accordingly.
follo me on twitter @robshmerling
original content at: www.health.harvard.edu…
authors: robert h. shmerling, md