1

EHR/CR - Functional Profile Clinical Research Working Group – eClinical Forum, PhRMA EDC/eSource Taskforce

Electronic Health Records/Clinical Research

EHR/CR

Functional Profile

EHR/CR Functional Profile Working Group

Health Level 7

Co-Chairs

John D. Mestler

Suzanne Bishop

Version 0.2 (5 Jan 2007)

EHR/CR Functional Profile Working Group

Co-chairs
John D. Mestler / Procter & Gamble Pharmaceuticals
Suzanne Bishop. M.A. / Consultant,
eCF US Facilitator / SKB Consulting
EHR/CR Functional Profile Working Group
Mathias Poensgen / Altana Pharmaceuticals
Gary Drucker / Deputy Director, Clinical Data Management
NA Remote Data Operations
Global Data Acquisition and Management / Bayer Healthcare, Pharmaceutical Division
Seema Bhat / Bayer Healthcare, Pharmaceutical Division
Tricia Gregory / Sr. Assoc. Director, Biometrics and Data Management Dept. / Boehringer Ingelheim Pharmaceuticals, Inc.
Darlene Kalinowski, M.S. / Senior Manager, RCO / Bristol Myers Squib
Jane Griffin, RPh / Director, Pharmaceutical Research / Cerner Corporation (EHR Vender)
Marie MacDonald / Director Strategic Product Development / Clinphone Inc.
Richard Perkins / Consultant,
eCF EU Facilitator / ConCept Consulting
Ian Hamilton / Eli Lilly
Vincente Rodriguez / Eli Lilly
Christian Reich, M.D. / Millennium Pharmaceuticals, Inc.
Catherine Célingant / Director, Clinical Data Management
/ Millennium Pharmaceuticals, Inc.
Thomas Engel / Outcomes Research
Robert Case / Procter & Gamble Pharmaceuticals
Miguel Valenzuela / Sr Systems Validation Analyst, Pharma Development Quality - Systems Validation / Roche Products Limited
Paul Brandon / Product Manager, Direct Data Capture / Siemens Medical Solutions (EHR vender)

EHR/Clinical Research (EHR/CR) - Functional Profile

Clinical Research Working Group –VERSION 0.2 5 Jan 2007

Welcome to the EHR/CR Functional Profile (EHR/CR-FP) project of the Clinical Research Working Group. The project isaimed at developing an HL7 Informative Functional Profile that identifies capabilities that are critical for the conduct of clinical research utilizing EHR systems. Our efforts will establish conformance to the HL7 EHR Functional Model,under the auspices and direction of the EHR Technical Committee. The hope is to develop a functional profile that will provide EHR vendors conformance criteria that are specific to Clinical Research. This will encourage incorporation of the functions necessary to utilize the Electronic Health Records as the direct data source for clinical studies. Additionally, we intend to define additional functions that may be required and have not yet been included within existing functional models.

Background:

Nation-wide eHealth systems are in the planning and development stage in many countries, while the use of electronic health record systems in doctor’s offices that are also doing clinical trials is increasing. It seems appropriate to consider how clinical research can utilize these systems and avoid redundant collection of data while complying with government regulations.

The Pharmaceutical and Biotech industry have made significant progress in the execution of clinical studies. This has, in part, been the result of advances in the collection of patient data. The traditional paper based clinical studies (utilizing paper case report forms (CRFs)) have given way to electronic data capture (EDC). Advances in information technology with regard to regulatory compliant software applications, telecommunications and more importantly the internet has enabled clinical research data to be collected and reviewed in near “real-time”. It is estimated that 27% of clinical research is conducted using EDC systems. Currently however, data must be entered and transcribed into the EDC system from a variety of sources. These include paper based patient records and those maintained in Electronic Medical Records (EMR) / Electronic Health Records (EHR) systems that may be in use at the investigational site (i.e., hospitals, doctors’ offices, clinics, etc). This constitutes a duplication of effort for the study site personnel.

Neither EDC nor EHR systems have all that is required to serve the purposes of the other (i.e., health care or clinical research). In the case of clinical research, the sponsor must not have exclusive control of the investigator’s patient data. Regulatory agencies want to ensure that data cannot be compromised either accidentally or intentionally. Therefore, the investigator must hold the source (independent of the sponsor’s database and not under the sponsor’s control). Additionally, investigators are required to maintain accurate case histories for each of their patients. Increasingly, the investigational sites that are conducting clinical trials using sponsor-supplied EDC systems also have EMR / EHR systems that require input of the same patient data. The result is often the creation of a third record of information, on a paper copy that is printed from the EMR / EHR system. This paper copy may be placed in the patient’s file to satisfy a requlatory requirement for source data verification. With a sponsor provided EDC system, the trial data must either be printed out or provided electronically to the investigator to maintain once the clinical study has finished. This duplication of tasks and associated costs will grow with the increasing use of electronic data sources (i.e., through the progress of national initiatives to create an eHealth environment in which all patient records will be electronic).

Electronic Health Records are viewed by government agencies, payers, epidemiologist as well as the clinical research community as a quantum step forward in delivery of their respective needs. The current EHR environment can be described as on the rise and accelerating in its importance and adoption priority. The expansion and government-encouraged use of EHR systems in hospitals and physician offices means that patient data are increasingly being entered and maintained electronically. Recent reports suggest that 20-25% of US healthcare practices use electronic medical/health record systems. Within Europe, these figures vary greatly between countries from 3% to 90%. There has been significant media attention given to the national efforts of the US and the European Union (EU) to develop nationwide electronic health networks (eHealth).

Data from EMR / EHR systems cannot be used directly for clinical research purposes. The number different systems and architectures make it difficult if not impossible to integrate with EDC systems on anything greater than a one-to-one integration, which is not economically feasible. Many do not have ways to either integrate or export data. Data are often unstructured (i.e., textual as opposed to data fields to collect individual, identifiable data items). Additionally, EHR systems are not covered under the control of clinical research regulations. As stated previously, the result is often the creation of a third record of information (either via paper or investigator-controlled, regulatory-compliant eSource system) to meet regulations..

Typically clinical research systems are designed to comply with FDA Guidances: The Guidance for Industry: Computerized Systems Used in Clinical trials (CSUCT) and 21 CFR Part 11. The current version of CSUCT was published in April 1999. A draft guidance was circulated for comment in September 2004. These documents can be accessed on

Both the FDA and ICH provide requirements for clinical trial records and the systems and processes that maintain them. The same responsibilities of the investigator towards the accuracy of source data exists whether that data is hand-written on paper or entered and stored electronically. If data are entered and stored into an EMR / EHR or EDC system as the sole source and used in clinical research, then that system must be compliant with these regulations (For example, EMR / EHR / EDC systems data that are used as eSource for clinical trials, under current regulation, require authority checks (as expected under 21 CFR Part 11) such as ensuring that only authorized persons can access the system and maintaining a clinical research-compliant audit trail).

Health care in general is under tremendous time and cost pressures. Patient records predominantly serve the purpose of the medical care of the patient and are optimized towards this purpose. It would be inefficient and costly for a hospital or physician’s office to carry out the meticulous documentation that is common in the bio-pharmaceutical industry unless the EHR system is designed to handle those tasks “behind the scene”.

The ideal environment provides non-redundant systems and processes that allow the use of patient electronic medical data for clinical research in a way that meets data protection, regulatory, and ethical research requirements and minimizes the challenges of clinical research for healthcare professionals. Additionally, collaboration on common data standards (including EHR narratives) and data transfer standards are needed to support both electronic national health records and clinical research (e.g., support for the CDISC/HL7 joint initiative). Data standards are essential for data collection, interpretation and exchange within the medical and research communities.

The EHR/CR-FP is a collaborative effort between the bio-pharmaceutical and healthcare industries and associated vendors to expand and adapt the structure of EHR and the associated systems, networks, and processes.

The project is planned in four phases:

  1. Organization–solicitation of participants, determination of scope and care setting of profile, and development of project plan and overview
  2. Formalization –step by step development of EHR/CRfunctions and conformance criteria (in process).
  3. Harmonization –comparison with, incorporation into, and alignment with the EHR FM (in process).
  4. Incorporate defined EHR/CRfunctions through alignment with EHR-FM.
  5. Accept or reject other functions from FM.
  6. Solicit input from EHR TC for possible new functions.
  7. Define priority timeframes (i.e. essential now, essential future) for functions.
  8. Incorporate and modify conformance criteria.
  9. Finalization– attention to detail, wording, language, conformance and preparation for final EHR – TC approval and external publication.

The development of the EHR/CR FP uses the HL7 approach. That is to say that everyone’s contribution or concern needs to be addressed. Your input is welcome.

Reading this Document

The EHR/CR FP like the EHR FM, is divided into three broad sections: Direct Care, Supportive Functions and Information Infrastructure. Each section defines a broad category of functions applicable to Direct Care, Supportive Functions, and Information Infrastructure.

Functional Priorities

Each function in the EHR/CR FP is assigned a single priority as follows:

Essential Now (EN) / Functions to ensure that patient data maintained in an EHR system will meet all clinical research regulatory requirements for data collection, data management, data extraction, and data security, and can be interpreted in a consistent manner. These functions must be present on “day 1” of using the EHR data for clinical research.
Essential Future (EF) / Functions to improve efficiency and performance of conducting clinical trial data capture through use of the EHR system. These functions will include those necessary to aggregate clinical data across EHRs for use in clinical research with additional functions added progressively to lead to an environment where all clinical research data capture and management will be through nation-wide health information networks and systems. The EF functions will be divided into levels along the evolution to this ultimate goal. None of the EF functions will be required to be present on “day 1” of using the EHR data for clinical research as the trial sponsors can provide work-arounds either through process or sponsor-supplied electronic systems.
Optional (O) / Functions in the HL7 Functional Profile that could facilitate clinical research, but will not prevent clinical research if they are absent.

Conformance Criteria

Conformance criteria have been developed in accordance with the standards set forth by the EHR technical committee. In order to ensure consistent, unambiguous understanding and application of the functional profile, the use of a consistent set of keywords (normative verbs) have been employed to describe conformance requirements.

  • SHALL – indicates a mandatory, required action. Synonymous with ‘is required’.
  • SHOULD – indicates an optional, recommended action that is particularly suitable, without mentioning or excluding other actions. Synonymous with ‘is permitted and recommended’.
  • MAY – indicates an optional, permissible action. Synonymous with ‘is permitted’

In addition, clarification is necessary to understand the standardized nomenclature used to describe the functions of a system. The following chart, adapted from the EHR FM, illustrates the hierarchy of nomenclature. For example, “capture” is used to describe a function that includes both direct entry “create” and indirect entry through another device “input”. Similarly, “maintain” is used to describe a function that entails reading, updating, or removal of data.

MANAGE
Capture / Maintain
Input Device (Ext.) / Create (Int.) / Read
(Present) / Update / Remove Access
View
Report
Display
Access / Edit
Correct
Amend
Augment / Obsolete
Inactivate
Destroy
Nullify
Purge