Jacob Read
I haven't blogged about this yet, which is a bit surprising to me since I find myself describing it so often.
Let's start with an overview. We can view health information through a life cycle lens.
The promise of health information technology is to help us—and ideally the people we serve—achieve optimal health.
The HITECH Act began with the concept: “Adopt, Connect, Improve.”
These are the three pillars of meaningful use incentive programs.
adoption Technology allows us to connect system, therefore promote healthy.
Pretty simple, right?
Years later, one could argue use even connect Already (mostly) done.
But the bridge between measurement and health improve This is not a problem that we can easily solve with the tools currently available to us.
Why?
Many technological solutions, especially those that promote Dashboardthe most critical piece of the puzzle is missing. They got us close, but then they conceded the ball.
That's where this “simple” AAAA” model comes in.
For data and information to be truly valuable in healthcare, a full cycle needs to be completed.
Simply collecting and displaying is not enough. There are four basic steps:
1. get. This is where we collect raw data and information. EHR entries, device readings, patient-reported outcomes…all the information that flows into our system. Please note that I distinguished data (converted representations of the physical world: blood pressure, CBC, DICOM representation of MRI, actual medications taken) and information (Diagnosis, Thoughts, Symptoms, Problem List, Prescription Medications) Because the data is authentic and the information is reliable possible True, but possibly inaccurate. We need to weigh these two inputs correctly – because data is a better input than information. (I'll resist the temptation to make the case that data is also the preferred input for AI models… maybe that's another post.)
2. total. Once obtained, this data and information needs to be aggregated, normalized and cleaned. This is about getting disparate data sources to speak the same language, creating a unified repository so that we can ask questions about one data set rather than dozens or hundreds of data sets.
3. analyze. Now we can start to understand it. This is where clinical decision support (CDS) begins to take shape, how we identify trends, flag abnormalities, predict risks and highlight opportunities for intervention. Most current solutions end in the analysis phase. Dashboards, alerts, reports… they all throw advice (like a bowl of spaghetti) into the lap of humans to sort it all out and figure out what to do.
Sure…you can see patterns, understand the population, and identify areas for improvement…all of which are good things. The maturation of health information technology means that aggregated, standardized and sophisticated analytics are now more accessible and powerful than ever before. We no longer need a dozen specialized point solutions to handle every step; modern platforms can integrate it all. That's good – but not good enough
A dashboard or analytical report, no matter how elegant, is ultimately reactive. It shows you the truth but it's not the case Do Anything about it.
Behavior. This is where the rubber meets the road. It’s about translating insights into tangible interventions. What should happen (or not happen) next?
What is the benefit of knowing that a patient is at high risk for readmission but does not trigger specific follow-up protocols, social work consultation, or an adjusted discharge plan? What is the point of identifying prescribing patterns if the system cannot facilitate practice changes, provide immediate feedback to clinicians, or adjust order sets?
We rely on human intervention to bridge this gap. Clinicians may see trends on a report and initiate changes manually. We think it is necessary to screen and place an order…(一一).
So sad.
The real power of health IT, especially the advances we are seeing, is that close the loop. We should build systems that not only capture, aggregate, and analyze data; Facilitate the next best actionprioritizing what’s best for the people we serve, and (of course) WHO Should you be a recipient of this guide?
Imagine a system that not only flags potential problems but also:
* Automatically generate personalized patient education documents.
* Suggest an updated medication order (or set of orders) with just one click.
* Schedule a follow-up appointment with the appropriate specialist.
* Push notifications to care coordinators for intervention.
This is not about eliminating human judgment; It’s about empowering it. It's about doing the right thing, doing what's easiest.
The beauty of this loop is its iterative nature.
The actions we take then generate new data and information that feeds back into the acquisition phase, allowing us to continually refine our understanding and improve our interventions. The faster and more frequently we complete these four steps, the more responsive, efficient, and patient-centered our healthcare teams will become.
Next time you evaluate a new health IT solution, ask a key question: How can this system help us? Behavior?
Jacob Reider, MD, is a family physician who has served as deputy national coordinator for ASTP/ONC, CMIO for Allscripts and Albany Medical Center, CEO of Alliance for Better Health, and currently works in angel investing, consulting, and pickleballing. Find his occasional thoughts This is one of the few blogs older than THCB!
