Closing the Loop in Health Care Coordination

“Closing the loop” is a term that describes a well-functioning coordinated healthcare delivery system. It implies that clinical information flows easily during a medical referral – the referring clinician is able to let the recipient clinician know the pertinent information about the patient being referred, and the recipient can let the referring clinician know the opinions and recommendations that result from the referral. It is a two-way “loop.”

Achieving this has been a vexing problem for health care. Often, the referral information sent to a recipient is minimal, frequently filled out by office clerical personnel (rather than the clinician directly) – “headache” written on a prescription pad sent to a neurologist, or “chest pain” sent to a cardiologist. Specialists receiving such information often (rightly) complain of the inadequacy of the referral. What are the patient’s other diagnoses? What medications are being taken, and which ones have already been tried? What recent lab tests or other diagnostic steps have been taken? These may or may not be included in the referral sent to the recipient.

On the other part of the “loop,” feedback from the recipient back to the referral originator may be equally inadequate. Sometimes a good dictated note is created, including an initial assessment and recommended plan. Other times (depending on the specialist) the information is minimal, and the patient is scheduled for follow up and additional testing without ever letting the referring source know – each clinician taking care of the same patient function in separate silos, with minimal coordination. Primary care physicians, often the referral source, become quite frustrated when the patient comes back for follow up and is told “the cardiologist you sent me to change several of my medications and wants to see me back next month,” or “when I go to the lab, I need to get the blood tests you ordered, and the ones that Dr. A ordered, and also the ones that Dr. B ordered – can’t we just combine them?”

How to we improve on this?
Improving on this breakdown of information sharing is central to making health care delivery more coordinated, efficient and better. In coordinated, managed care-centered environments, such as with large groups and IPAs in California, “closing the loop” is an area of significant attention. Adequate referrals, and adequate responses to referrals, are part of performance-based assessment and performance-based compensation in these kinds of settings.

Faxing adequate information back and forth, perhaps prompted by referral forms within a managed-care network, has been the traditional way of sharing information between the two parties. This can be supplemented by satisfaction surveys on the adequacy of data-hand-off and coordination conducted periodically after-the-fact.

As we move into the era of Electronic Health Records (EHRs), the possibility of better data exchange becomes more palpable. In fact, of the 5 domains of healthcare priorities that underlie the federal EHR Incentive Program (Meaningful Use), “care coordination” is one of the main ones. Moderns EHRs are poised to be able to improve healthcare delivery in this area, but the promise is still potential, not actual.

Care coordination and health data exchange
The approach taken by the Meaningful Use program to date has stemmed from a hospital/institution-centered view. It envisions creating hubs that have “standard” data interfaces where such institutions can exchange data seamlessly. These hubs – Health Information Exchanges (HIEs) – are in various stages of being built, and rely (as a first step) upon uploading and downloading summary information from patient charts, using standardized data formats (CCDs and CCRs).

Stage 1 of Meaningful Use for ambulatory-practice EHRs contained the requirement to be able to create and import either CCRs or CCDs. However, actually transporting these documents between different settings of care was not required in the first stage; it is a Stage 2 concern (coming up in 2014). In order to successfully transport these documents back and forth, a functional and secure transport network needs to be in place – and building such an infrastructure is where much of the current attention resides.

Further, the data contained in CCRs and CCDs is mainly summary information – problem lists, medication lists, allergies, and the like. This is a great start, but falls short of what clinicians need for adequate referrals – often, copies of diagnostic tests (x-rays, CT scans, EKGs, stress tests, etc.) need to flow via fax in order to supplement the summaries. There is still a lot of work to do.

In addition, the business model for interfacing with these envisioned HIE hubs is still uncertain. Who will pay? At first, the use of such hubs will be focused around connecting community clinicians with their local hospital, since hospitals have approached data sharing with such an architecture in mind. In managed-care group settings, an intra-group (or intra-IPA) hub may be built for data exchange.

However, exchange of clinical data between clinicians in more general ambulatory settings may not be as clean – will there be multiple hubs? Which one is the recipient connected to? Is there a cost of integration with a hub? Will each integration take time – if so, is this a scalable solution? The “hub strategy” may struggle with low enrollment/subscriptions among ambulatory clinicians, unless subsidized by institutional sponsors – and even then, enrollment may lag.

Is there a better way?
Given that web-based EHRs have been particularly popular among smaller ambulatory practices (for all the obvious reasons), the possibility exists of sharing clinical charts among participants on the same platform.

In this envisioned model, the entire chart – not just summary information, but past chart notes, scanned documents, lab reports, etc. – becomes visible to both parties: the “host” practice originating the referral, and the “guest” practice receiving the referral. Further, the recipient clinician can enter information onto the shared chart upon seeing the patient, so that the response to referral can be seen by all (the loop is closed).

Of course, the underlying assumption with this “extending the platform” approach to data sharing is that both parties are on the same platform. However, when the platform is free, and a new account can be set up within minutes, the barriers to using this method go away. Regardless of whether the recipient has a different EHR, or no EHR, the information can be shared fully using this approach.

It is true, that such a method does not actually share data between disparate systems in the background – such a method is still a considerable ways off. But a view of the data can be shared fully, and can achieve the result much more quickly than otherwise might be had. What is more, this approach is scalable – the platform can be extended widely and quickly between any clinicians anywhere. This is the vision we are working on, and plan to launch, in the 2012 year.