HHS Releases Landmark Report: Reforming America's Healthcare System

Noam H. Arzt, Ph.D.On December 3, 2018, the Department of Health and Human Services (HHS) released an extensive, 120-page report on the administration's proposals to reform the healthcare system. The report, titled Reforming America's Healthcare System Through Choice and Competition, is divided into four major sections. The report that government policy of the last few years has suppressed competition, increased prices for healthcare, and limited choices for consumers. Though rich in detail as it tries to prove each of these points, the more than fifty recommendations are often broad and aspirational rather than practical.

Since I am not a health economist, I will leave the market issues to others to discuss (many of the ideas in this report have been vetted and discussed by others previously). But there are two sections of the report which make direct mention of Health IT.

First, there is a subsection on Quality Improvement and the Measurement and Reporting of Quality (starting on page 88 of the report). After briefly reviewing the history of quality reporting and its impact on competition, the report then provides five recommendations focusing on the development of clearer quality measures, and specifically the exploration of the use of "machine learning techniques that can directly access data… from the provider Electronic Medical Records…." (See page 93)

But the real interest in Health IT comes at the very end of the report in a section titled, "Healthcare Information Technology and Non-Competitive Healthcare Markets. This subsection of the report starts on page 102. It begins with a discussion of how other industries have benefited from technology and automation. The focus in this section quickly shifts to interoperability and a discussion of barriers: medical complexity, lack of business drivers, lack of application programming interfaces (APIs), and lack of technical interoperability (what the report calls "lack of network exchange"). The report points out a disincentive to information sharing inherent in the fee-for-service model still dominant in the US, the proliferation of incompatible Health Information Exchanges (HIE) infrastructures, the closed nature of APIs from many vendors, that all prevent basic query for patient records across organizations.

The focus then shifts to the potential solution represented by the 21st Century Cures Act with its ban on information blocking, the Trusted Exchange Framework and Common Agreement (TEFCA), and open/free-market solutions such as open APIs like HL7's FHIR and CMS' Blue Button 2.0. The recommendations in this section focus, therefore, on implementation of the 21st Century Cures Act, reducing documentation burdens on providers (see the recent ONC Strategy Document on this topic), fostering interoperability, and continuing work to make standards "more comprehensive and robust."

Can't really argue with any of that, though the path forward seems far from clear. Both the information blocking rulemaking from ONC and the revised TEFCA proposal have been delayed into the new year. FHIR development continues, but its implementation is quite limited and vendor commitment is uncertain.

While major HIE initiatives like eHealth Exchange, Commonwell Health Alliance, Carequality, and the Strategic Health Information Exchange Collaborative (SHIEC) continue to develop some level of cooperation and collaboration there is no clear or consistent national strategy emerging. It seems pretty clear that our national investment in electronic health records (EHRs) has not quite achieved its goals. This is due to many factors, including the inherent complexity of the subject matter and the dominant fee-for-service payment model that does not incentivize data sharing and has EHRs overly focused on billing to the detriment of clinical functionality. The business case for EHRs and interoperability just needs to get stronger and clearer.

Open source and free-market solutions will likely play a key part in any solution, and some think it's the most important potential game changer (for example, see the article on The Politics of the EHR by Dr. Bruce Wilder). But it is much more than a software problem. Interoperability requires agreements on standards, governance, and patient consent at minimum to be effective. In many ways the industry is moving past intellectual property models as the primary driver of software strategy. Traditional hosting initially gave way to "lift and shift" cloud deployments (Software as a Service, or SaaS) but this is going to give way to more modular Platform as a Service (PaaS) deployments where applications are assembled from microservices developed internally or available on the open market.

Likely open source components will increasingly be leveraged in these offerings though their presence and source may become increasingly masked under a services umbrella. In the EHR space, this might look like a system that is more loosely coupled than today's EHRs, with core services (like patient matching, clinical decision support, and interoperability) assembled from best of breed cloud-based offerings and combined with proprietary components to create a system.

We'll just have to wait to see what 2019 brings, and to continue to push forward and hope for some convergence down the road.