While the Population Health Blog continues to delight in the emerging science of "mHealth" as a newly minted start-up Chief Medical Officer, it ran across this interesting article on risk and patient safety. Authors Thomas Lewis and Jeremy Wyatt worry that "apps" can lead to patient harm.
They posit that the likelihood of harm is mainly a function of 1) the nature of the mistake itself (miscalculating a body mass index is far less problematic than miscalculating a drug dose) and 2) its severity (overdosing on a cupcake versus a narcotic). When you include other "inherent and external variables," including the display, the user interface, network issues, information storage, informational complexity and the number of patients using it, the risks can grow from a simple case of developer embarrassment to catastrophic patient loss of life.
In response, they propose that app developers think about this "two dimensional app space" that relies on a risk assessment coupled to a staggered regulation model. That regulation can range from simple clinical self assessment to a more complex and formal approval process.
What's clear to the PHB is that hidebound mainframe entities like the Food and Drug Administration are no match for the app "ecosystem". Rather than try to formulate a one-size-fits-all "not function as intended" model like this, maybe it should triage its oversight using the Lewis and Wyatt framework.
In addition, the PHB agrees with Lewis and Wyatt that safety is also a function of clinician input. Docs and nurses can assess possible mistakes, their downside severity and the impact of all those variables.
The PHB couldn't have put it better:
".... many app developers have little or no formal medical training and do not involve clinicians in the development process and may therefore be unaware of patient safety issues raised by inappropriate app content or functioning."
Without the insights of seasoned real-world doctors and nurses, apps could end up with the same safety issues that are plaguing electronic health records, many of which were also developed with little regard to physician or nurse input.
In other words, just because it's a "health" app doesn't mean its necessarily so.
Image from Wikipedia
