[This post was written for the #HCLDR tweetchat Accessing Your Medical Record – Then What?]
A 3rd piece of pie! Now you are talking my language Though I think the sugar alarm would go off first 🙂 Great post!
— Colin Hung ()
If you have access to your medical record data, then …. what? The what is an action. I will argue that this question is best answered within the framework of what I call Pragmatic Interoperability.
Pragmatic : "compatibility btwn intended & actual effect of message exchange"
— Charles Webster MD ()
I’m sure you’ve heard of If/Then rules. On the left-hand side (in our left-to-right reading and writing culture) is some condition that must be met, in order to trigger some action described on the right-hand side. If “raining”, then “get umbrellas.” If the gram stain is negative, and if morphology is rod, and aerobicity is anaerobic, then consider Bacteriodes.
Compare the If/Then rule formulation to Pragmatic Interoperability (see my recent HIMSS16 10,000 word series), which “is the compatibility between the intended versus the actual effect of message exchange.” When an If/Then rule “fires” (that is, its left side conditions are met) its intended effect is the right-hand side action.
Pragmatic Interoperability occurs when the intended right-hand side action matches the actual effect of executing the rule, even if this intended action is in an other healthcare organization. For example, if the If/Then rule executes in the patient’s physician’s EHR, but the intended action occurs near, on, or even inside the patient, this is an example of pragmatic interoperability. This is also the case if an If/Then rule executes in the patient’s home monitoring systems, and the intended actual effect occurs inside the patient’s physician’s EHR.
1 example of higher levels of interoperability, most relevant mature tech 4 pragmatic interop is WF tech pic.twitter.com/VtYvrprAEK
— Charles Webster MD ()
So, if you have access to your medical record data, then …. what? Workflows, executed by workflow engines, triggered in a manner similar to If/Then rules. If I should be on a protocol but am not, trigger workflows to increase likelihood I will join the protocol (how? that’s part of the intelligence of the workflow). If something that should be measured isn’t being measured, then trigger workflows to increase likelihood of measurement. If some measured value is outside of normal limits, trigger workflows to increase likelihood of returning the value to within normal limits. The data in the medical record (broadly construed) represents the state of the patient. As the patient’s state changes, workflows are triggered. Today these workflows are mostly executed by human workflow engines. Tomorrow, increasingly, they will be executed by a combination of software-based and human-based workflow engines, hybrid cognitive systems, so to speak.
Reactive / dynamic re my 2010 ppr
— Charles Webster MD ()
Actions trigger workflows, which themselves are collection of actions. Sometimes actions are called tasks, but sometimes they are really experiences, if they happen to someone, such as a patient. In my series on Pragmatic Interoperability, I discuss three relevant areas from pragmatics, a subfield from linguistics. (Interoperability already leverages ideas from syntax and semantics, also subfields within linguistics.) One way to think about workflows, in this context, is as conversations among EHRs and health IT systems intended to serve patients and providers.
RE topic, patient-HCP communication, I'm working on a possibly relevant paper re conversation" workflow"
— Charles Webster MD ()
Another area of pragmatics is implicature and presupposition. Implicature is about being cooperative, relying on evidence, not saying too much or too little, and striving to communicate in a fashion most useful to the addressee. Implicature is obviously relevant to communicating with patients from EHR and a wide variety of health IT systems. Presumptions rely shared real world knowledge, we (patient and care team) share, that can be leveraged to communicate and cooperate most effectively and efficiently. I go on, at length, about these topics in part 4 of my series on pragmatic interoperability.
The “Then What” in the title refers to actionable data. The most relevant technology to make health data actionable is workflow technology. Indeed (finally!), we are seeing a new layer of workflow platforms running on top of an older layer of data platforms. These new process-aware workflow platforms rely on a variety of APIs, including FHIR, to access and change data. These care workflow platforms initiate and coordinate tasks and workflows for patient and healthcare personnel.
T1 What is the biggest hurdle to patient access to health data? Lack of incentive? Fear? Technology?
I don’t believe patients really want data. Patients really want useful, valuable actions driven by data. I need my MRI not to read it, but to make sure that someone who needs to read it get to do so, so they can make a diagnosis that will result in therapeutic actions moving me toward wellness. We’ve created a giant sea of databases. We need a giant sea of workflows.
T2 Which is more important to you, owning your health data or being able to freely access your health data?
Neither. Patients owning data is conceptually flawed (see Patient Data Ownership Cannot Resolve Data Access Problems: But Workflow Technology Might). Mere access to data is not sufficient. Patient data must cause action, action benefiting patients.
Patients “owning” data won’t get them any more legal leverage than what they already have under current law. Further more, this campaign distracts from the real issue, giving patients more control over the workflows creating the data, and then doing useful things with their data. See my previous HCLDR post on this topic, The Workflow Prescription: Patients Need Zapier, Workflow, and IFTTT-like Control Over Self-Care Workflow Automation At Home.
T3 What would you do first if you had full access to your health record?
I suppose I might skim it. (I have more than the average educational background necessary to understand the contents.) But what I really want is to push a button and set up a system of automatically occurring notifications, both into and out of the EHR, that will get me, and keep me, well. Everything from reminders nagging me to make relevant appointments to finger wagging about that third piece of pie I am about to consume (Internet of Things and wearables increasingly leverage workflow tech). When I get a reminder, it must be “actionable” in the sense I need to, with minimum effort (perhaps only pressing or saying “OK!”) instruct my collection of guardian angel workflows to go ahead and do what they suggest, such as consult my schedule and make that appointment, or strike that scrumptious key lime pie from my weekly auto-generated shopping list.
T4 What do you believe would be the biggest benefit TO YOU of full access to your health record?
At this stage, being relatively healthy, not much. However, if I were ill, possibly chronically, perhaps cancer-ridden and miserable, that last thing I want to be the workflow engine going through my medical records figuring out what needs to happen and in what order. Let’s build health IT systems that make patients’ lives easier, not harder. Let’s turn EHRs and related health IT systems into intelligent systems communicating and cooperating with each other (and the patient!) on the patient’s, on my, behalf. Communicating and cooperating intelligent systems is exactly the idea behind my recent 10,000 word, five-part, HIMSS16 post on Pragmatic Interoperability: Healthcare’s Missing Workflow Layer. I hope you’ll give it a read!