Dr Tornado’s recent post on “How I met your EMR” takes a lighter look at the challenges of EMR use in healthcare and specifically makes the point that:
the goal of human-machine interaction engineering is to produce a user interface which makes it easy, efficient, and enjoyable to operate a machine in the way which produces the desired result. This generally means that the operator needs to provide minimal input to achieve the desired output, and also that the machine minimizes undesired outputs to the human.
But it is the visual that brings it home:
While the makes light of the problem the point is serious – there is a list of LE-GEN-DA-RY suggestions – don’t overload the screen or the user, make it quick, easy and *fun* to use.
There are better ways and attending the ONC hearing today Ensuring the Quality of Data Hearing today in Washington DC and while there was much testimony that pushed for structured data (for example Janice Nicholson from i2i) it was Dr Jospeh Heyman from the HIT Policy Certification Adoption Workgroup Committee who said:
“Just one extra click is not OK”
He qualified this by saying that if it offered some value to him and the patient then it could be acceptable(for instance making sure he carried out a clinical action that had been missed). In far too many cases these systems and processes just demand extra work for little benefit to the clinician. Focusing on capturing data for the wrong reasons is at the root of this problem – if we are trying to capture data for the purposes of billing, reporting on quality or other analytics tasks then that will be the focus. But clinicians want to focus on the patient and the care they deliver – and it was an interesting concept in this article The future state of clinical data capture and documentation: a report from AMIA’s 2011 Policy Meeting that covered AMIA’s 2011 Health policy meeting that included the shortcomings that are well documented:
They highlighted the shortcomings of current approaches that impede efficient data capture and presentation, fall short of accurately representing clinicians’ thinking, and fail to accommodate clinical workflow
But it was some of the potential solutions that were interesting:
Stetson et al noted that documents are created for many different purposes and their value and quality may be assessed using different metrics that may not be compatible. For example, a note might be written to inform a colleague about the clinical status of a patient without concern that it generates a ‘comprehensive bill’. Thus, it might be deemed of high value with respect to clinical communication but poorly compliant and not supportive of billing or utilization review or clinical quality measures.
There is a desire to codify data and that can be achieved using natural language processing. So perhaps offering tools that allow clinicians to capture the note in the way we were trained to do: to inform our colleagues on the clinical status and the medical decision making. This can be done through any means suited to the clinicians – per Doc Tornado
Article 5 – A bro EMR lets bros create, edit and sign off notes using any input method, any device and a combination of them.
And then use technology like Clinical Language Understanding to extract out the medical intelligence and offer feedback to help capture all the pieces necessary for the other purposes (clinical quality measures, billing, clinical management) with minimal interruption in the workflow but significant added value?
