This paper produced by Duke University researchers and published in the Journal of the American Medical Informatics Association highlights how challenging mHealth is as a concept to Healthcare Informaticians.
mHealth Insights
Study size
Perhaps the journal is desperate to find articles to print but it’s hard to see how this piece got past a peer review process eg. it involved only 6 subjects (3 healthy participants and 3 Chronic Disease Patients) yet lists 11 authors listed (Ryan J Shaw, Dori M Steinberg, Jonathan Bonnet, Farhad Modarai, Aaron George, Traven Cunningham, Markedia Mason, Mohammad Shahsahebi, Steven C Grambow, Gary G Bennett and Hayden B Bosworth).
False consensus
I think it’s important to appreciate that it’s almost meaningless to find that 3 healthy people who were willing to participate in a medical research trial actually bothered to participate in that trial.
Similarly it’s pointless to be trying to compare the behaviour of these healthy subjects with the behaviour of Patients with something as vague as a chronic illness eg. if they’re obese/hypertensive for the last 20 years they’re more likely to have bigger concerns about using a weighscale/BPmonitor because of how it makes them feel rather than the fact that it’s now a device that’s got a SIM card inside it.
Devices/Apps and their lack of integration
I think a lot more consideration should’ve been paid to the mHealth devices that were chosen and the 3 different apps that were used with them in the trial. The 4 devices chosen for the trial were the (i) Fitbit Flex Physical Activity Tracker, (ii) iHealth Air Bluetooth Pulse Oximeter, (iii) iHealth HS5 Wirelesss Body Analysis Scale and (iv) iHealth Feel Blood Pressure Monitor. The apps used for the trial were the Fitbit app and iHealth app and a messaging system built using Twillo .
If I was in the trial as a healthy participant I think I would bore quickly of these devices because (i) I have never worn a physical activity tracker and don’t see the value of them as the tech in them has already converged to my smartphone (for 10 years senior citizens in Japan have had native pedometer apps running on their raku raku feature phones). (ii) My blood oxygenation doesn’t change over time so it would seem completely pointless launching the app, pairing the device, waiting for the data every time I’m reminded to test. (iii) My weight changes on a day to day basis in amounts that are so small that you’d require a calibrated clinical device before it could even begin to be meaningful. (iv) My blood pressure varies but it’s always within range so going through with the test procedure everytime the reminder alerts me got boring about 20 years ago when I first used a digital BP meter.
If I was in the trial as a participant with a chronic illness I can’t imagine being excited to record my lack of activity, the smoking habit that I’ve struggled to kick, the bulging waist line I’ve always meant to do something about and the hypertension that feels out of my control. I also might be discouraged from collecting data for a researcher because of concerns about how it might be used to influence clinical decisions that may be made about me by my Carers. Philippe Kahn makes some great points on this in his talks eg. where a major mobile operator in the USA with thousands of employees couldn’t get a single one of it’s employees to enrol even when they offered cold hard cash!
It feels ridiculous having to point this out because to most people I know personally it’s so obvious but trying to infer understandings on how Patients behave based on data being collected through this complex web of different sensors and apps is as daft as saying: “3 Patients weren’t adherent to the pogo sticks we gave them so we can conclude they’ll have no interest in bicycles either”
“Patients with chronic illnesses may be less likely to adopt and use these devices compared to healthy individuals”
I think it’s ridiculous to try and make this sweeping statement based on a study of 6 people. I’ve seen football teams introduce connected medical device monitoring programs and it’s rare to find a very well paid professional athlete (with millions to gain!) who has been able to be as adherent with BG monitoring in a short term trial as I know T1 diabetics have been for 10+ years!
Perhaps the researchers should’ve flipped the study and compared healthy participant group’s adherence when they are tasked to use medical devices that are ALREADY being used by the Patients in the Chronic Illness Patient group?
Related posts:
How would the Born Mobile generation redesign that Patient Care Experience?