Introduction

Clinical documents are an electronic representation of clinical information prepared for two main purposes:

  1. To support interoperability between systems by providing a standard, extensible machine-readable format for representing clinical data
  2. To provide a human-readable format for clinical data that is consistent with the machine-readable format within the same document

Why use clinical documents?

  • They reduce cost in reproducing and transporting paper records.
  • They can improve the speed and accuracy of data absorption into other systems.
  • They can provide patients with an accurate, readable record of some or all of their medical record.
  • May be easier to generate than other forms of e-data.
  • At the point of care, they reduce hassle to patient in completing new provider registration materials (that awful clipboard), allow physicians to receive critical health data at transfer of care, and improve the quality of care through complete and timely information.

How do clinical documents compare to other data formats?

For years health IT professionals have been enabling interoperability between systems, both within organizations and between organizations. There are two approaches to preparing and processing information that originates in one system and is destined for another. The first approach is data-centered:

DataFmtFig1
In this approach data is extracted from a participating system, transformed into a data file (e.g., fixed length or delimited file) and sent to the recipient who processes the data into a database. This has been a common mode of data interoperability for many years. Data-centered files can contain information about just one patient or about many patients depending on the use cases and systems involved. Public health usually receives data-centered files for public health reporting and most public health agencies are skilled at processing and absorbing these types of files.

The second approach is document-centered:

DataFmtFig2
In this case, data extracted from a participating system is formed into a clinical document and sent to the recipient. But rather than process the data into a database, the clinical document is usually stored intact in a document repository (like an electronic filing cabinet). An additional “meta” database is maintained which has entries about each document to facilitate their retrieval (like information about the patient whose clinical data is in the document, but not the clinical data itself). Clinical documents typically contain data about just one patient at a time.

Data stored in a document-centered file can be extracted and processed since the underlying electronic data is structured and machine-readable. The purpose of an electronic document, however, is to maintain the integrity of the data represented in it at the moment in time it was produced.

What does this mean to public health?

Clinical documents have become a key strategy in promoting and using electronic information to support patient care. They provide clinicians with a critical tool for quickly and accurately moving data between parties to support a wide variety of clinical needs and information-sharing requirements. Clinical summaries – a form of clinical document – are prominent in the Centers for Medicare and Medicaid Services (CMS) EHR Incentive Programs to encourage eligible professionals and hospitals to implement health information technology, in particular electronic health record (EHR) systems and their “meaningful use” (MU).

As public health agencies increase their reliance on data received from EHR systems, support for clinical documents will become increasingly important. But there are some important issues:

  • The EHR vendor community has placed increasing emphasis on clinical documents as the most desired format for interoperability between systems. More energy and focus will likely be placed there in the future than in data-centered approaches.
  • The public health measures related to meaningful use explicitly require HL7 v2 messages and not clinical documents. This produces a disincentive for public health agencies to add clinical document support to their capabilities.
  • Data-centered approaches continue to dominate the public health interoperability landscape, and supporting multiple approaches can place a large burden on public health agencies’ limit technical capabilities.
  • Different systems may use different coding and semantic standards, making exchange with retention of clinical meaning more challenging. A lot of work has been done within HL7 v2 messaging standards to ensure semantic interoperability. Similar care needs to be taken within the clinical document community.