The other night I participated in a very useful Google+ hangout with Adrian Gropper, Michael Mascia and Michael Chen. The discussion focused on a subject I think is incredibly important: the patient-centered health record. Unfortunately, this topic is hard to discuss without drowning in technical terms and acronyms. I consider myself fairly tech-savvy and still struggle.
A (55 minute) YouTube video is here: Click here
I think it is worth watching. before watching it, consider reviewing the following basic information to help set the stage, first without tech terms or acronyms, and then repeated with some of the key jargon.
The current EHR model is that each office or institution owns and manages an electronic record that contains information about the patients in that system. Despite the obvious need and lots of talk, there has been little actual progress towards making these separate and mostly proprietary systems ‘interoperable’ and therefore able to share information. The result is that clinicians routinely work with incomplete or outdated information, patients are locked into their home system, and it is extremely hard for patients to access their own information in any meaningful or useful way. Care is less safe and less reliable, patients are prevented from actively managing their care, and clinicians are frustrated.
The model discussed in this video takes a different approach, one that appeals to me for both philosophical and technical reasons.
In this model:
- The basic unit is a one-patient record unique to the individual patient.
- It is ‘open source’ meaning that the code is public, maximizing the ability to improve or modify it and create added pieces for new functions.
- This basic unit and its information are owned and controlled by the patient.
- This is where all the health and medical information about an individual patient is stored.
- It contains the most current, complete and up-to-date information.
- The patient has full access to their individual record.
- This patient-centered health record has a component that allows the patient to control access by others, essentially inviting their clinicians (or others) to see and use the information.
- Clinicians can access the system through their own software using one of two mechanisms:
- Using a straightforward process which gives them access to multiple separate patient records, and they work in the actual patient record.
- Using their proprietary clinician or institution owned system, which is linked to and synchronized with the patient’s individual record.
- In both cases, the ‘official’ version of the information is the patient-centered record, fully accessible to the patient.
- There are standardized connections between the clinician-based systems and the multiple patient-based records.
That was English. Now a version of the same information with some jargon:
- The patient has a record (EHR) to store his or her health information
- Ideally it is ‘open source’ such as NOSH (New Open Source Health Charting System) by Michael Chen. Read more about this at https://noshemr.wordpress.com/.
- It is owned and controlled by the patient.
- It is where all the health information about an individual patient is stored.
- It functions as the ‘Source of Truth.’
- The patient has full access.
- The patient-centered record (such as NOSH) has a component that allows the patient to control who can see, change or use the information.
- This is sometimes called UMA (for user management and authentication).
- It has other names such as HIE of One (health information exchange of one).
- This is what allows the patient to manage who has access.
- Clinicians access the system through their own software using one of two mechanisms:
- They use a parallel open-source software system that has a list of the patients using individual patient records they have been authenticated to use.
- They use (work in) their institution’s proprietary system which is linked to and synchronized with the official patient-owned record.
- In both cases, the patient’s record is the Source of Truth and fully accessible to the patient.
- There are interfaces that make connections work.
- FHIR (pronounced fire) is the one talked about here.
- Other ‘application program interfaces’ (APIs) can be developed or used, especially if the system is open source.
There are several important considerations in a system like this:
- Open source improves security, adaptability, flexibility
- It is based on the assumption that patients should have full control over their health information according to their own needs, not just be given limited access by clinicians or their systems.
- If patients have full control over access to their health information, it goes without saying that they can access their own health information without either delay or barriers.
- It can be adopted incrementally rather than requiring the entire US healthcare system to turn a switch.
- Patients and clinicians can help us move from where we are now to a patient-centered health record system by:
- Understanding the concept.
- Being at least somewhat familiar with the terminology.
- And MAKING NOISE about wanting a system like this in their local setting.
Peter Elias is a family physician in Maine. He blogs at http://petereliasmd.com.
Hmmmm. This harkens back to the EHR/PHR issue/debate. Providers are, of course, legally bound to keep medical records and they are liable for what’s in them–many a malpractice suit has turned on what’s in the “written” record. So it’s a non-starter to suggest that the entire EHR infrastructure can be easily turned into a consumer/patient owned or operated thing. That said, what us consumer/patient advocates have long wanted – stated simply – was an electronic record that was fully open and accessible to the patient and that the patient would have a “field”….or several….in the record to record and say whatever they want, and especially to provide what is called PRO…patient reported outcomes. We also strongly urge that patients’ goals, values and preferences be information that is collected and prominently displayed in a patient’s record. I realize that’s all general philosophy and overall approach stuff and the mechanics and systems are complex to design, build and implement. Adrian and others know much more about that than me.
Steven, I agree with the liability framing but I’m curious why you think the laws work against a patient-centered record. The law and common practice does not hold a physician responsible for reading 100% of a patient’s record whether it’s in an institutional EHR or in a patient-centered EHR. Physicians routinely access only that part of a patient’s history that they believe is relevant. I don’t know of any malpractice cases around a physician’s failure to consult the state health information exchange.
I believe that basing one’s decisions on a patient-controlled record – with appropriate forensic documentation in a separate system that the physician controls – reduces liability and improves the quality of care.
Wonderful and much appreciared effort. The video shows how much thought has gone into this. Lots more to do too. I’m glad open source is your modus.
One of a thousand questions:
I just wonder if the patient can ‘own’ his own record. Off the top, how can providers protect themselves if the patient changes his record to, say, wipe out contributory negligence on his part? Wouldn’t we have to time-stamp and authenticate every datum–as the record accumulates–in order to protect the providers AND patients?
Of course we would want to time-stamp and sign every entry. Also, each clinician or provider institution would likely run a “write-only” copy of the relevant parts of the record that they could use in case the patient removed their access for any reason.
The interesting thing about this is that it solves the problem we have today with “writes” into an institutional EHR from outside practices and patient-generated wearables. For example, projects such as SMART (as mentioned in my NEJM link in comment above) do not currently allow updates of the EHRs. This is not because FHIR and the technology don’t support EHR updates but simply because this opens up the EHR host to all sorts of liability that the software and the institution are not yet able to deal with. As a result, we see intermediaries like sandboxes, tethered PHRs, Apple HealthKit / ResearchKit, etc…
With a patient-centered health record, the liability issue of the institutional EHR is mostly eliminated because their copy need only manage things that are not clinically actionable (such as billing support) and that’s a much easier thing to do. This also shifts innovation to the open and peer-reviewable patient owned EHR/PHR.
Since the NOSH code was originally a provider-based EHR with a built in patient portal, the foundational design for a patient centered EHR was simply to “flip it around” so that the there was only 1 patient in the patient centered EHR and allowing multiple providers and practices to be registered.. Unlike other PHR designs, this allowed the expected security and auditing features (to protect providers as well as patents) to remain intact.
This week’s New England Journal of Medicine has a related article: Time for a Patient-Driven Health Information Economy? http://www.nejm.org/doi/full/10.1056/NEJMp1512142
Once Congress (HiTech) and CMS get into mandates and coercion for immature technology the natural evolution of technology is interrupted….and the rules and mandates stop such good ideas from attracting capital and the early adopters who are critical to developing innovative high value developments…….such as open source patient centered (and controlled) medical/health records…..but sometimes good ideas have a way of breaking through anyway….one can always hope.