Electronic Health Records: 12%
Hospitals are a center point for the Electronic Health Record (EHR) and a key to the National Health Record's sharing of health data among treatment teams. Less than 12% of them have effectively implemented software.
I was mortified to discover this. So shocked and disbelieving, in fact, that I needed corroboration of the original story run in Health Affairs, and found the Robert Wood Johnson Foundation reference that's linked above.
A hospital has huge Information Technology (IT) resources, and can share data among ancillary specialty clinics, like your psychiatrist, your cardiologist, or your pulmonary specialist. The doctors are associated with the hospitals, and when a patient is released to a doctor's care, the idea is that the doctor has availability to the electronic records in the hospital, as long as the patient signs a release of information and knows about it. This saves a ton of aggravation in filling out the same information many times, and delivers more accurate treatment data to these members of a consumer's treatment team.
The combined information is suddenly available in a secure environment, so when the patient moves, or needs to share that health data with somebody not in that hospital's network, it's available with a release of information via another component of the National Health Record, the Regional Health Information Organization (RHIO), or other similar health data organization. The RHIOs are all connected into the National Health Record, or will be soon, so our health information is secure behind firewalls and other technological wizardry. The idea is that nobody gets that information without a patient's say-so, and that when the patient says it's OK, the data is instantly in the doctor's hands.
Here's a scenario.
- Joe has a heart problem and a pulmonary problem
- Joe also has terrible anxiety that gives him breathing difficulty
- Joe goes to the emergency room, and the docs determine the current problem is a panic attack, and Joe's medicated and released, with instructions to follow up with the psychiatrist's Mental Health Clinic; the clinic is part of the hospital's network because the psychiatrist is affiliated with the hospital
- Joe goes to the Mental Health Clinic, and the notes, the medication order and the discharge summary from the hospital are available on line, so the psychiatrist knows the details
- Later, Joe goes to his cardiologist, who is also affiliated with the hospital and can consult the record of the emergency room visit in the EHR
- Ditto with the pulmonologist
Access to that record and the ability to treat the patient as a team, making sure all bases are covered so Joe not only doesn't die, and also has the highest quality life possible, isn't possible in 88% of the hospital service areas across the nation.
With only 12% of the hospitals effectively implemented, the National Health Record, which was mandated to be operational in a few years by then-president Bush, is essentially nowhere.
I've reported on the value of Rapid Cycle Implementation in getting an EHR up and running as quickly as possible. This method of solving the highest-priority problems that the software can address with target groups of professionals, then rolling out those features to all users, is nothing new…and it's effective. Soon enough, the organization has a functional EHR and is fine-tuning it, making those features that are already rolled out better. I wonder how many of the 88% of the hospitals without effective EHRs are incorporating Rapid Cycle Implementation into their software rollout.
There are many reasons software implementations fail. It could be that the CEO isn't promoting it with the staff effectively. It could be that the team doing the implementation is more interested in keeping good statistics on how much money's been spent on manpower than actually getting a feature rolled out. Resources (usually people) could be in such short supply that the project is sabotaged.
In these failings are the keys to successful implementations. Setbacks are unavoidable, but I'd call this failure.