It’s Not About The Box Improving Care at Group Health with People, Process and Technology

March 4, 2009

Why does this patient have high blood pressure?

Filed under: Documentation,Epic — Tags: , , — Matt Handley @ 1:46 pm

I love it when our patients help remind us of how we can better serve them.  I had the pleasure of talking with one of our patients on the phone yesterday about his story and suggestions for improvement.

40 plus years ago, at age 11, the patient had a repair of a coarctation of the aorta.  He and his family were told “He’s fixed!”.  No thoughts about follow up.  So earlier this year, when at age 60 he started to have higher blood pressures, it was assumed that it was from essential HTN.  His own investigation led to a referral to Alan Golston, cardiologist extraordinaire in Tacoma, and he started down a path that has led to a successful surgical repair of a paper thin graft – one that was at significant risk of rupture.

Many patients would have stopped there, but this patient committed to helping us develop safer systems for managing patients with a history of surgical repair of congenital heart disease.  The gaps we discovered in our conversation included awareness of the needs for follow up, and in how we enter important data into Epic in a way that helps us take great care of these patients.

It will be a team sport.  The clinical content will be developed with a cardiologist from UW with an interest in congenital heart disease and Dr Golston.  Gwen O’Keefe and her  team will help clarify and reinforce how we enter discrete data into Epic, and how we use Epic to track chronic illness needs.

I think this is a great story to share.   Engaged patient, great doctor, great care, great outcome, and we are going to be learning from it and improving the care of other patients with similar care needs.   A story that highlights the advantages of our model from patient engagement to great multispecialty group practice, to great tools and technology to support both.

I have promised the patient to close the loop as we take the steps we talked over.  He is eager to share that story through the non-profit that supports patients with congenital heart disease.

September 23, 2008

Problems with Diabetes, decision support and problem list management

Filed under: Epic,Improvement News,Quality / Affordability — Tags: , — Matt Handley @ 4:11 pm

A great challenge has presented itself in how we use a problem list entry to manage alerts and reminders for diabetes.  Years ago we created a “dummy” diagnostic code for diabetes (GHC CIS CARE COORDINATION: DIABETES [GHC.04]) so that Health Maintenance and Best Practice Alerts can be turned off for the rare patient with diabetes for whom the alerts are not clinically appropriate (e.g., hospice/palliative care).  The alerts fire based on this dummy code being in the problem list, rather than the presence of the specific diagnostic codes we use for documenting and coding diabetes care.  Because the diagnosis is assigned by an algorithm it isn’t perfect – it is wrong about 3% of the time, implying that a patient has diabetes when they do not.  It is possible to remove the dummy code from the problem list for the two circumstances where it is clinically appropriate (when the algorithm got the diagnosis wrong or when it is clinically inappropriate to continue to monitor their diabetes) in two ways:

1. Delete GHC CIS CARE COORDINATION: DIABETES [GHC.04] from the problem list – removes it, but the algorithm will add it back next month, or…
2. Resolve GHC CIS CARE COORDINATION: DIABETES [GHC.04].  This suppresses Epic from adding the dummy diagnosis back to the problem list forever (or until the problem is “un-resolved”)

Some clinicians may delete or resolve the dummy code inappropriately in their efforts to clean up the problem list, not understanding that the dummy diagnostic code is the tool that we use for managing alerts.  Good intent (love that attention to the problem list), but most of the time that this happens it is not for clinically appropriate reasons.   This has happened about 500 times, and most of these have been in patients who very clearly have diabetes.

We all need to work together to manage problem lists, but need to exercise care in how we do that.  We are working on a way to both “un-resolve” the problem list entry for patients whose diabetes is being actively managed, and set up a regular system to make sure that when the GHC.04 code is resolved it is on purpose.  We are looking to the Diabetes Expert Team as a stakeholder to assist in this work, under Dave McCulloch’s direction.

Powered by WordPress