Healthcare Informatics
Geisinger's Chief Strategic Information Officer: OpenNotes is the Future
By: Mark Hagland
If anyone can testify to the power of the OpenNotes phenomenon, it is Alistair Erskine, M.D., the chief strategic information officer at the Danville, Pa.-based Geisinger Health System. Geisinger Health was one of the first hospital-based organizations in the U.S. to adopt the OpenNotes concept, as conceived by Thomas Delbanco, M.D. and Jan Walker, R.N., and implement it in their health systems (the other two being Harborview Medical Center in Seattle and Beth Israel Deaconess Medical Center in Boston). Erskine and his colleagues went live with OpenNotes in 2012.
It is in that context that HCI Editor-in-Chief Mark Hagland interviewed Dr. Erskine and other medical informaticists, healthcare leaders who are helping to change U.S. healthcare through the OpenNotes approach. OpenNotes was the subject of the July/August Healthcare Informatics cover story. To date, the OpenNotes organization estimates that 10 million patients across the U.S. have been given access to their physicians’ notes, a formidable accomplishment for a movement that is still relatively young, chronologically speaking.
Below are excerpts from Hagland’s interview earlier this summer with Dr. Erskine.
Let’s talk about Geisinger’s journey around OpenNotes so far. When did it begin, and when did you go live?
We were one of the first three health systems, along with Tom Delbanco’s group, to study this and to go live. 2012 was when we started. From the first study, three organizations. It was Harborview Medical Center [in Seattle], Beth Israel Deaconess [in Boston], and Geisinger, with primary care practices. We went live in 2012.
Did you coordinate with your colleagues at Beth Israel Deaconess and Harborview?
We absolutely coordinated with them, and that was really led by Tom Delbanco. He really initiated the whole thing. And each one of us wanted to find out what concerns there might be with the concept, before we went live, and what the concerns from patients and providers might be after we implemented it. And we wanted to find out how it might sustain itself over time. We went live in 2012.
What was your experience of it?
There’s a paper from the Annals of Internal Medicine on the study. It goes into the data from the three original sites. Our experience was that, before we went live, providers were more concerned than patients. Will it take longer? Will it require more messaging? Will doctors spend more time editing their notes, knowing the patient will read it? Will the doctor be more candid? How will the notes change around how we talk about cancer, or mental health, or obesity and weight loss? Will we be as efficient or more efficient? And will the notes improve patient education? Those were some of the questions we asked beforehand, and asked again after implementation. But the reality is that we’ve had virtually no complaints from either patients or providers.
And what did you find?
When it came to the number of secure e-mail communications that occurred after OpenNotes, those communications remained constant in volume; they didn’t go up. So the concern that patients would be calling clinics all the time, did not manifest, though that was a worry. And we found that the patients who initially participated, continued to participate. The idea that physicians would have to take more time addressing patients’ concerns after a visit, initially, 45 percent were worried, but afterwards, that went to 0. The concern about candid documentation went from 32 percent to 9 percent. That decreased significantly.
And the other thing that’s important about the way Geisinger approached it, differently from the other groups, is that we made it entirely voluntary. We said, if you as a group, want to share notes, there’s a mechanism to do that, and they can enroll; but if you don’t want to, we will not force you, we will not mandate that you do it. So of the 2 million visits we see at Geisinger, a portion of those—maybe 40 percent activation of our engaged patients who actually look at the record. There have been 600,000 encounters reviewed by patients out of the 2 million sets of notes released so far. So that means that the number of patients actually looking at their notes remains pretty high. In 2015, the total hits per month go from 450,000 to 50,000 per month. In other words, total hits of people looking at their notes, remains in the 50,000 range, for an annual total of 600,000. And the number of providers accounts for over 1,600 providers—physicians, PAs, residents, and fellows—the bulk are doing it. Funnily enough, the ENT doctors decided not to get involved; that was their particular culture. If you count the residents or advanced practitioners, we have just over 2,000 employed physicians. So yes, 75 percent are involved.
Have the physicians essentially said, ‘Yes, we’re glad’ you did it?
Absolutely. Virtually no complaints from either providers or patients. It’s been crickets in terms of anyone saying this is a problem, I don’t like it, please turn me back off. No one who’s gone on has decided to go back off. And per the patients, we had 277 patients who were excluded by providers for one reason or another—we had given them that option. And that’s dramatically less than 1 percent.
What about the concerns that it would take more time or force physicians and providers to dramatically alter their documentation?
As far as taking more time, that didn’t happen at all. As far as certain sensitive topics, meaning cancer, substance abuse, mental health and obesity, with obesity, the concern went from 18 percent to 5 percent, around obesity.
After four years’ experience, what have been the biggest learnings for you and your colleagues around this? And what would you especially want for CMIOs to know?
The first thing for other CMIOs to do is to start with the vision that there is a national movement to be more transparent in healthcare, and to share what we’re doing in our ‘secret little medical records,’ with the patients. And share with the physicians the papers that document that these concerns don’t play out. And share that the vendors are making this easier to do. For example, Epic in their 2015 version, will make it that the default will be sharing with patients, and you have to check not sharing. So vendors are adding features and functions to support the OpenNotes movement.
Now, there is a laundry list of decisions around configuration. First, do your patients have access to a portal? Because if your patients don’t have access to a portal, they won’t be able to see the notes. So the CMIO needs to know the rate of patient portal adoption in their system. If for example they don’t have a portal or the rate of adoption is 5 percent, they could do OpenNotes all day long, but only 5 percent of patients would see the notes. Second, they need to decide whether this should be mandatory or voluntary.
Third, they need to decide on outpatient, emergency department, inpatient, notes. Fourth, are they going to allow residents to share notes, or only attendings? Fifth, are any notes going to be excluded? Does the psychiatrist or psychologist share notes? And do we turn off certain kinds of notes, so that psychotherapy notes are never shared, for example? And does a note get shared before it’s finalized, as a preliminary note? Do the patients with caregivers and family members, do they get access? Because oftentimes, they help with health literacy and support. So if you have automatic access to your mom’s record, do you get access to her notes? And lastly, teenagers—how do you negotiate what to do with pre-adults? So these are things to consider that are important configuration, and almost cultural, considerations. And based, on your organization and the willingness and openness of your organization, there’s a spectrum of choices there.
Would you agree that OpenNotes is one element in a bigger picture going forward?
I do agree that it’s one element in a bigger picture. In fact, OpenNotes 2.0 is being termed ‘OurNotes,’ in terms of its creating agenda-setting with the patient. Even before the patient comes to the clinic for a visit, they can log into their portal and do this on an iPad while in the waiting room, and can list for the doctor what they want to talk about during their visit. We’ve tried this in an open-ended way, for discussion between the doctor and patient during the visit. The physician then takes that information and uses it in the clinical note. And then when it’s all done, the patient gets a copy of that final note, for co-signature.” Though the initial pilot project has been small-scaled, Erskine believes that the “OurNotes” experiment points the way towards a new chapter in physician-patient relations going forward.
OpenNotes is part of a larger shift of the relationship between providers and patients, correct? In fact, consumers interacting with nearly all other industries are setting the terms for their interaction. Why should healthcare ultimately be any different?
It’s hard to argue with that; I think you’re right on target. And this notion that patients come to us on our terms, on our schedules—you know, when I was in medical school, if you said the word ‘consumer,’ that was anathema. Now, patients are in high-deductible health plans, and they are using Uber and Amazon and everything. When people go to Ritz-Carlton, they already know about them. And I think that what we’ll see, and what we’re pushing for at Geisinger, is health systems that don’t look at one patient at a time, they look at cohorts of patients, and then beyond that, the entire community of people in their area, those who are coming to them for care and those who aren’t. It’s typically the pre-sick who don’t engage, whom we could potentially engage before they become sick. So we all need to bring that into healthcare, figure out a way to interact with patients, how they want to be communicated with and treated, and put that into the system, the preferences, the same way as in other industries.
When healthcare ultimately gets there, that will be amazing, right?
Yes. And some health systems will get there, and some won’t. And it will be a market differentiator.
