Seamless interoperability between diverse healthcare IT systems has been pursued for many years. President Obama's re-election has all but secured continued focus on interoperability through government incentive programs. Achieving interoperability has been rather difficult, however, and doesn't seem to get easier despite the availability of mature standards. On top of that, the rise of Accountable Care Organizations (ACOs) has created interesting new dynamics that have the potential of slowing down interoperability efforts. But more about this a little later.
First, let's look at how interoperability is commonly defined. HIMSS defines it as the extent to which systems and devices can exchange data, and interpret that shared data. For two systems to be interoperable, they must be able to exchange data and subsequently present that data such that it can be understood by a user.
According to the Department of Health and Human Services (HHS), 20 percent of preventable medical errors are caused by the lack of immediate access to health information, and one of every seven primary care visits is impacted by missing medical information. Given these statistics, it's no wonder that interoperability is heavily pursued by private as well as government initiatives. Health Information Exchanges (HIEs) in particular have gotten a lot of attention over the last few years. According to a recent survey by the eHealth initiative, there are currently close to 300 HIE initiatives in the US, up from 255 a year earlier.
Given all of this, why are we still nowhere near widespread health interoperability? While there are technical challenges, they are not insurmountable and not at the heart of the problem. There now are robust standards available that facilitate not just the exchange of data, but also the semantic interpretation that enables the understanding of health data across diverse EHR systems. Examples are the Continuity of Care Document (CCD), particularly with the HITSP C32 constraints, IHE Patient Care Coordination and most recently, the Consolidated CDA standard that is mandated by the Meaningful Use Stage 2 rules.
Furthermore, what is often overlooked is that we already have a perfectly interoperable patient record in the form of narrative clinical notes. While you may not agree that these unstructured blobs of text qualify as ‘interoperable', this is only true if you ignore recent advances in Natural Language Understanding that make discrete clinical facts within those notes accessible and shareable electronically. Most of all, physicians get more clinical value out of a comprehensive narrative note compared to a discrete EHR record.
The Meaningful Use (MU) incentive program – which is providing funds to eligible physicians and hospitals to adopt certified EHR technology under the government's HITECH act – is particularly focused on achieving interoperability. But as with any government program, MU is politically charged and accused of being ineffective. Members of the House of Representatives recently sent this letter to the Secretary of HHS, expressing their concerns over the Meaningful Use Stage 2 rules and asserting that "we are no closer to interoperability in spite of the nearly $10 billion spent." However, President Obama's reelection suggests the MU program and funding will continue as planned.
The National Coordinator for Health Information Technology, Dr. Farzad Mostashari – who is in charge of the MU program – has recently made it very clear in this posting that MU stage 2 rules are focused on overcoming barriers to health information exchange:
we will pay close attention to whether the requirements in the rule are sufficient to make vendor-to-vendor exchange attainable for providers. If there is not sufficient progress or we continue to see barriers that create data silos or "walled gardens" we will revisit our meaningful use approach and consider other options to achieve our policy intent.
The challenge he is hinting at is the entrenched position of EHR vendors, who have little incentive for enabling the sharing of health information across vendor boundaries. In the current environment, it is in their best interest to promote sharing of information using only their own monolithic systems as opposed to supporting interoperability standards. Compare that to other industries like banking (where you can walk up to any ATM from any vendor to retrieve cash from any bank) or telephony (where you can place a call from your device of choice to any other device, cellular, landline or IP phone from any other vendor) with seamless interoperability.
Recent commentary in the New England Journal of Medicine by two renowned professors at Harvard Medical School makes the case for putting more pressure on the EHR vendor community to embrace innovation, support exchange standards and to break down data silos in support of true cross-vendor interoperability.
Interestingly, the push for hospitals, clinics and physician practices to collaborate more closely and form Accountable Care Organizations also has the potential of slowing down cross-vendor interoperability efforts. As hospitals have made multi-million dollar investments into their EHR system of choice, they tend to push their ancillary clinics and physician practices into adopting that EHR system as well to justify and recoup at least some of their investment.
As Jonathan Bush -- the CEO of practice EHR vendor athenaHealth -- stated in a recent earnings call:
So they [the hospitals] go out and sort of do some Bush Doctrine saying, in three years we're going to be live with this thing, it's going to slice and dice and bring world peace, and you're either going to be on it or not allowed in our hospitals.
Until we overcome these dynamics that preserve monolithic systems and associated data silos, meaningful and actionable exchange of health information across systems and vendors will remain around the next bend on the road to true interoperability.