Journal of the American
Association
Advance Access
January 20, 2015
Warner Medical
JL, et al. J Am Informatics
Med Inform Assoc 2015;0:1–10.
doi:10.1093/jamia/ocu015,
Research andpublished
Applications

Development, implementation, and initial
evaluation of a foundational open
interoperability standard for oncology
treatment planning and summarization

RECEIVED 10 July 2014
REVISED 16 September 2014
ACCEPTED 28 October 2014

Jeremy L. Warner1,2,*, Suzanne E. Maddux3, Kevin S. Hughes4, John C. Krauss5, Peter Paul Yu6, Lawrence N. Shulman7,
Deborah K. Mayer8, Mike Hogarth9, Mark Shafarman10, Allison Stover Fiscalini11, Laura Esserman11,12, Liora Alschuler13,
George Augustine Koromia13, Zabrina Gonzaga13, Edward P. Ambinder14

ABSTRACT
....................................................................................................................................................

....................................................................................................................................................
Key words: Medical Oncology; Breast Neoplasms; Health Information Management; Electronic Health Records;
Continuity of Patient Care; Information Science

BACKGROUND AND SIGNIFICANCE

physician (PCP) and an endocrinologist, both of whom encourage her to receive adjuvant chemotherapy at a network affiliate
50 km away and daily radiation therapy near her home. None
of her providers share an interoperable electronic health record
(EHR). The patient and her caregivers are faced with accessing
multiple portals, each with only a slice of the pertinent medical
information. As shown in Figure 1, the resultant paths of communication are often incomplete and susceptible to errors. This
problem will only grow worse as she enters the survivorship
phase of her treatment, which will extend for years to
decades.2

Cancer care is data-intensive, multidisciplinary, lifelong, and increasingly dependent on the seamless electronic transmission
of clinical data. As an example, consider a postmenopausal, diabetic woman who has just been diagnosed with early-stage
invasive breast cancer. This woman lives 150 km from a
National Accreditation Program for Breast Centers (NAPBC)
Center of Excellence.1 Results from surgery at the NAPBC center have determined that she will require adjuvant (postoperative) chemotherapy, radiation treatment, and hormonal therapy.
She has an established relationship with a local primary care

Correspondence to Jeremy L. Warner, Assistant Professor of Medicine and Biomedical Informatics at Vanderbilt University, 2220 Pierce Ave Preston Research
Building 777, Nashville, TN 37232, USA; jeremy.warner@vanderbilt.edu
C The Author 2015. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved.
V
For Permissions, please email: journals.permissions@oup.com
For numbered affiliations see end of article.

1

RESEARCH AND APPLICATIONS

Objective Develop and evaluate a foundational oncology-specific standard for the communication and coordination of
care throughout the cancer journey, with early-stage breast cancer as the use case.
Materials and Methods Owing to broad uptake of the Health Level Seven (HL7) Consolidated Clinical Document
Architecture (C-CDA) by health information exchanges and large provider organizations, we developed an implementation guide in congruence with C-CDA. The resultant product was balloted through the HL7 process and subsequently
implemented by two groups: the Health Story Project (Health Story) and the Athena Breast Health Network (Athena).
Results The HL7 Implementation Guide for CDA, Release 2: Clinical Oncology Treatment Plan and Summary, DSTU
Release 1 (eCOTPS) was successfully balloted and published as a Draft Standard for Trial Use (DSTU) in October 2013.
Health Story successfully implemented the eCOTPS the 2014 meeting of the Healthcare Information and Management
Systems Society (HIMSS) in a clinical vignette. During the evaluation and implementation of eCOPS, Athena identified
two practical concerns: (1) the need for additional CDA templates specific to their use case; (2) the many-to-many mapping of Athena-defined data elements to eCOTPS.
Discussion Early implementation of eCOTPS has demonstrated successful vendor-agnostic transmission of oncologyspecific data. The modularity enabled by the C-CDA framework ensures the relatively straightforward expansion of the
eCOTPS to include other cancer subtypes. Lessons learned during the process will strengthen future versions of the
standard.
Conclusion eCOTPS is the first oncology-specific CDA standard to achieve HL7 DSTU status. Oncology standards will
improve care throughout the cancer journey by allowing the efficient transmission of reliable, meaningful, and current
clinical data between the many involved stakeholders.

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

Figure 1. Illustration of some of the potential stakeholders in a routine cancer care scenario. Without standards (A),
communication pathways may be haphazard, incomplete, and nonsynchronous. Standards (B), illustrated by the HL7 logo,
enable reliable, complete, and replicable communication—with or without a central source of truth, such as a health
information exchange.

RESEARCH AND APPLICATIONS

This vignette illustrates the norm for many cancer patients
today, including those obtaining most of their care in urban settings, because competing hospital systems often do not interoperate. This is one of the many reasons that the Institute of
Medicine (IOM) considers our cancer care system to be a “system in crisis” with disjointed, fractured, and often error-prone
care—a situation that has not changed appreciably between
their critiques in 1999 and 2013.3,4 In addition to the practical
implications for individual patients, the secondary use of clinical information for regional or national cancer analyses, including quality reporting, is impeded by a chronic lack of
interoperability.5–9 The IOM also criticized the cancer establishment for not adequately engaging patients regarding their
treatment values and concerns, not providing suitable educational materials, and not communicating information about
patients’ cancer for their own and their future clinicians’ use.
Accurate electronic health information encompassing a complete and interpretable record is critical for engaged patients
and coordination of care across all phases of the cancer journey, from diagnosis through end of life.10,11
National standards for the exchange of clinical data, including
narrative elements, have existed since 2000, when the Clinical
Document Architecture (CDA, currently in release 2) was first described.12–14 CDA-R2 is an Extensible Markup Language-based
documentation model that represents health concepts using the
Reference Information Model (RIM) distributed by Health Level
Seven International (HL7).15 CDA-R2 has been demonstrated to
be an effective medium for the exchange of structured clinical
data between both systems and providers.16,17
The Health Information Technology for Economic and
Clinical Health Act, which was part of the American Recovery
and Reinvestment Act of 2009, enacted the Meaningful Use
(MU) EHR Incentive Program. MU Stage One cites the
Continuity of Care Document (a constraint on CDA-R2) as the
format for clinical document exchange between EHRs and related systems.18 However, sharing data electronically across
multiple clinical practices remains difficult owing to lack of harmonization of data, inadequate use of structured data capture,
lack of standards for specialty care, the proprietary nature and

general incompatibility of current EHRs, and lack of consensus
on the role of unstructured or semi-structured narrative notes.
At the same time, recognition is growing that structured data
alone, although it may have importance for billing and compliance documentation, leaves much of the record unavailable.19
Unambiguous events such as services rendered and laboratory
tests are amenable to structure, but the nature of illness and
the cancer journey are not.20 This challenge was stated eloquently by Dr. Robert S. Foote:
“The medical record is not data. It contains data, as do
many forms of writing, but it is not data, nor is it simply
a repository into which data are poured. Although its raw
material is information—some of which, importantly,
can only be expressed with words and not with numbers—a finished medical record is information that has
been transformed by the knowledge, skill, and experience of the physician, motivated by the healing impulse,
into an understanding of human experience that makes
the care of the patient possible.”21
Although the Centers for Disease Control and Prevention has
previously developed the Implementation Guide for Ambulatory
Healthcare Provider Reporting to Central Cancer Registries,
August 2012,22 this profile was not balloted through an
American National Standards Institute (ANSI)-certified Standards
Development Organization (SDO), and is meant for cancer case
reporting rather than clinical care. Recently, an oncology-specific
implementation of the normative HL7 version 3 (v3) Care Record
message was described for the continuity of nursing care for oncology patients transitioning from inpatient to home settings.23
To our knowledge, an oncology-specific standard for the electronic transmission of data required for the overall coordination
of clinical care has not previously been described.

OBJECTIVE
In 2012, the Health Information Technology Work Group of
American Society of Clinical Oncology (ASCO), comprised of

2

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

volunteer cancer clinicians, created the Data Standards and
Interoperability Taskforce (DSIT) for the development of oncology interoperability standards. The primary objective was to
develop interoperable oncology-specific standards through an
ANSI-certified SDO, to enable a reliable source of truth that can
be used to untangle the complex web of interprovider and provider–patient communication (Figure 1). The DSIT selected HL7
for its focus on healthcare and broad market penetration; its
aspirational goal of attaining interoperability at the application
level of the Open Systems Interconnection model for health information exchange;24 and the rigor imposed by the consensus
methods used in balloting, reconciliation, and standard approval. A secondary objective was to evaluate the resultant
electronic Clinical Oncology Treatment Plan and Summary
(eCOTPS) product in artificial and clinical settings. This paper
reports the development of the first oncology-specific HL7
CDA-R2 standard and the insights gained from its early
implementation.

Table 1: Data elements in the ASCO TPS
templates
Diagnosis, site, and staging
Family history and major comorbidities
Eastern Cooperative Oncology Group performance status31
Surgical procedures and notable findings and complications
Biopsy results
Tumor markers and genomic data
Radiation and chemotherapy treatment data and potential
and actual side effects
Survivorship plan and follow-up monitoring
Contact information for all significant providers of cancer
care

Source material: ASCO chemotherapy treatment plan and
summary templates
In 2007, ASCO developed a suite of treatment plan and summary (TPS) templates for cancer care, motivated, in part, by
the shortcomings clarified by the aforementioned IOM reports,
the seminal IOM report “From Cancer Patient to Cancer
Survivor: Lost in Transition,”25 and the loss of continuity and
paper health records caused by the catastrophic Hurricane
Katrina in 2005.26,27 Oncologists and advanced nurse practitioners with clinical subject matter expertise participated in focused task forces to develop generic and histology-specific
TPS templates. The templates are brief by design, comprising
only the most critical data needed for basic coordination and
continuity of care. They are paper-based documents that can
be used during the cancer work-up and treatment planning
phase, during actual treatment, and as a summary after treatment is complete. The summary may be provided to the patient, caregivers, and PCPs (who may often be unaware of the
signs of recurrence or potential long-term side-effects of chemotherapy, radiation, and other methods used during cancer
care28–30). The templates vary, but generally contain several
common data elements (Table 1). Because the treatment of
curable breast cancer is a common but fairly complex scenario,
ASCO’s DSIT selected the Breast Cancer Adjuvant TPS (BCTPS)
as the source material for the foundational HL7 oncology standard (Figure 2).

document types, such as “Discharge Summary” or “History
and Physical.” Each document type may contain a combination
of sections (e.g., problems, results) and entries (e.g., diagnosis
of cancer, result of genetic testing). As building blocks, these
templates, both sections and entries, may be reorganized into
different document types while maintaining the semantic accuracy of the clinical information. MU Stage 2 specified C-CDA
Release 1.1 as the standard for the exchange of clinical summaries and transfer documentation between EHR systems.34
Because of this citation of C-CDA and its fostering by large programs such as the Mid-South Clinical Data Research
Network35 and ONC’s Query Health,36 the DSIT intentionally developed the oncology HL7 standard to be congruent with
C-CDA principles.
Creation of the eCOTPS HL7 CDA-R2 IG
To begin translating ASCO’s BCTPS into a CDA-R2 standard,
the data were defined and disambiguated to map concepts to
standard terminologies, vocabularies, and nomenclatures or
modeled to the HL7 RIM. This process required extensive input
from medical and surgical oncologists in the DSIT, ASCO staff,
external oncology and interoperability stakeholders, and developers from the Lantana Consulting Group. The development included review and analysis of previously successfully balloted
clinical exchange standards and published IGs for existing
templates relevant to cancer treatment, such as the
Implementation Guide for Ambulatory Healthcare Provider
Reporting to Central Cancer Registries, Release 1.0,37 the HL7
Implementation Guide for CDA R2: Quality Reporting Document
Architecture—Category I (QRDA) DSTU Release 2 (US
Realm),38 and the HL7 Implementation Guide for CDA Release
2, IHE Health Story Consolidation, Release 1.1—US Realm.32
Each concept in the BCTPS was analyzed to determine
whether it comprised a distinct data element within C–CDA.

Consolidated CDA as a reference standard
Consolidated CDA (C-CDA) uses Extensible Markup Language
to transmit patient-specific medical data in structured and unstructured formats.32 It builds upon HL7’s CDA-R214 and the
HL7 v3 RIM,33 a consensus view of the way clinical information
can be abstractly represented. The CDA constrains the v3 RIM
by applying principles for the representation of information in
clinical documents. The C-CDA implementation guide (IG) provides building blocks, known as templates, to create specific

3

RESEARCH AND APPLICATIONS

These elements are rather generic across the multiple TPS templates, but are also site- and histology-specific.

MATERIALS AND METHODS

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

Figure 2. Page One of the ASCO Breast Cancer Adjuvant Treatment Plan and Summary template, available at http://
www.asco.org/quality-guidelines/breast-cancer-treatment-plan-and-summary-resources. This form is available as a modifiable Microsoft Word document, a Microsoft Word form, or a Microsoft Excel spreadsheet. In all cases, the product is intended to be a paper artifact that is printed for the patient, limiting the possibility of interoperability. Page Two has several
additional elements (data not shown).

RESEARCH AND APPLICATIONS
Because the family history C-CDA template is brief and inadequate for pedigree drawing, transmission of genetic test results, and other functions critical to an oncology patient, the
DSIT used the convention of CDA-R2 classes targeted at external information objects. Thus, a pointer to HL7 Version 3
Standard: Clinical Genomics Family Health History Pedigree
Model was included.39 This practice allows implementers to
use either the C–CDA-based family history template for the
transmission of minimal family history elements or the pedigree
model for risk evaluation, clinical decision support, genetic
testing results, and other advanced functions.

Concepts were then mapped to a template that modeled the
semantics of that particular concept. The IG consists of three
categories of templates: (1) existing; (2) modified; and (3) new.
Existing templates consist of constraints that are unchanged
from the original CDA-R2 template. Design aimed to reuse existing templates wherever possible. Modified templates further
define and restrain cardinality and vocabulary sets for coded elements to represent BCTPS-specific content. When no template
existed or could be further constrained, a new template was
designed to represent the relevant concept. Examples are listed
in Table 2.

4

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

Table 2: Examples of existing, modified, and new templates used in the eCOTPS
Category

CDA-R2 Template Name

eCOTPS Template Name

Explanation

Existing

Allergies Section

Allergies Section

Existing CDA-R2 allergies section was
reused in the eCOTPS document.

Modified

Medications Section

Medications Section BCTPS

Included entry:
Medication Activity

Included entries:
Anthracycline Lifetime Dose
Chemotherapeutic Drug
Therapy Discontinued
Chemotherapy Medication
Activity
Medication Activity

CDA-R2 medication section was modified to contain therapies administered
to the patient during cancer treatment.
Entry templates were added to this
section to represent BCTPS-specific
content such as chemotherapy activity
and lifetime dose of anthracyclines.

Not applicable

Anthracycline Lifetime Dose

New

•
•

•

RESULTS
Development of the draft standard
The initial draft of the IG was developed over the course of 6
months, during which weekly conference calls were held
between the developers and cancer clinicians. Several new
CDA-R2 templates were developed to represent BCTPS-specific
concepts, specifically data for family history of breast cancer,
the American Joint Committee on Cancer’s Tumor, Node, and
Metastasis Staging System breast cancer codes, breast cancer
chemotherapy and hormonal therapy, surgical findings, and potential adverse effects of breast cancer treatment. Several
Tumor, Node, and Metastasis codes were not represented by
extant Systematized Nomenclature of Medicine, Clinical Terms
codes; new codes were added in cooperation with the National
Library of Medicine.

Normative—meets the highest threshold for consensus
under ANSI process, and is stable and implementationready.
Informative—provides detailed information regarding the interpretation or implementation of an HL7 specification.
Draft Standard for Trial Use (DSTU)—meets ANSI requirements for trial implementation; may not be forward-compatible with later, normative edition.
Comment—gathers input on the viability and clarity of a
proposed document; no votes are taken, but all comments
are considered.

Balloting and approval as a DSTU
The draft standard, IG, and artifacts were submitted to HL7’s
Structured Documents Work Group in the spring of 2013 for
open balloting. The standard met quorum for approval and all
comments were resolved per the usual HL7 process. In
November 2013, the revised standard was approved and published as HL7 Implementation Guide for CDA, Release 2: Clinical
Oncology Treatment Plan and Summary, DSTU Release 1.40

Votes, if applicable, may be submitted as affirmative, abstain, or negative. Negative votes must include comments that
document the reason for the negative vote. Comments as part
of an affirmative vote are encouraged and can improve the content or clarity of standards. DSTU standards that meet quorum
for approval must still address and reconcile all comments during the reconciliation process, with the intent of improving the
quality and clarity of the proposed draft standard. After all

Implementation in the Health Story Project
In 2014, the eCOTPS was successfully implemented in the
demonstration by Health Story Project (Health Story) at the
Healthcare Information and Management Systems Society

5

RESEARCH AND APPLICATIONS

comments have been resolved, the balloted document is
resubmitted for either publication or re-ballot.

Balloting of eCOTPS through HL7
HL7 operates under the consensus rules of an ANSI-certified
SDO. The HL7 ballot facilitates widespread collaboration in
the development of technical specifications. Participation
is free and open to HL7 members, and nonmembers must
register and pay an administrative fee. Interested parties
register prior to the release of the ballot packages. Once
per cycle, registered participants have at least 1 month to
review the draft standard and submit a vote and comments.
HL7 has four types of ballots to which proposed standards
can be submitted:
•

Because the use of anthracycline is
specific to oncology, a new template
was designed to represent the total
cumulative dose of drugs in the
anthracycline drug class that a patient
has received in his or her lifetime.

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

centers (UC Davis, UC Irvine, UC Los Angeles, UC San Diego,
and UC San Francisco), the Graduate School of Public Health at
UC Berkeley, and a number of public and private partners.41
Athena’s mission is to prototype new approaches to the
screening and treatment of breast cancer. Members of the
Athena team engaged in a comprehensive clinical workflow
analysis to identify opportunities for improving data capture at
the point of care. The analysis at four Athena sites involved 45
key informant interviews with clinicians, practice managers,
cancer registrars, and other stakeholders. Specifically, the project team focused on hand-offs and interfaces between clinical
workflows, as well as data capture, validation, and utilization
through existing health information exchange mechanisms.
From this emerged a subset of data that is critical for decision
making, clinical trials, and registry reporting. These data elements were reviewed by over 50 clinicians across the five UC
academic medical centers, with a primary focus on key clinical
and research data. The project team compared their data elements against existing relevant data standards, including the
eCOTPS, the College of American Pathologists electronic

(HIMSS) Interoperability Showcase. Health Story began in 2006
as a not-for-profit alliance of healthcare vendors, providers,
and associations. As part of its mission, Health Story coordinates an annual presentation at the HIMSS Interoperability
Showcase. A clinical vignette (Figure 3) similar to that described above was used to iteratively transmit accumulative
eCOTPS data through real-time interactions with various EHRs
and other electronic data applications. This realistic scenario
demonstrated modular use of the eCOTPS; coordination of care
for a cancer patient with multiple comorbidities; incorporation
of patient preferences into the clinician workflow; electronic
linkage of an interdisciplinary care plan manager and a Health
Information Exchange; multiple data capture methods directly
from devices; and patient-reported symptom and preference
information. The demonstration was well received by more
than 800 HIMSS attendees.

RESEARCH AND APPLICATIONS

Implementation and evaluation through Athena
The Athena Breast Health Network (Athena) is a collaboration
among the five University of California (UC) medical/cancer

Figure 3. The Health Story Project clinical vignette. The patient, “Ana,” is diagnosed with breast cancer and goes through
a series of health care interactions on her journey through treatment to survivorship. “Actors” in the vignette are labeled in
italics; vendors or organizations with a primary role in a given clinical interaction are shown adjacent to that interaction. All
information is passed by using established standards for structured and/or unstructured data, as shown. In particular, the
eCOTPS is passed to the Health Information Exchange during treatment planning and to Ana herself at the conclusion of
primary treatment. D/C: discharge; MIE: medical informatics engineering.

6

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

standard through an ANSI-accredited SDO; (2) to approach this
standard creation process in a manner enabling modularity;
and (3) to provide the resultant products for open consumption
by the healthcare market in accordance with ASCO’s not-forprofit approach.
Although healthcare-specific standards have been in development for decades, uptake has been frustratingly slow. The
many reasons for this include competing standards at the
national and international levels, loss of enthusiasm at early
implementation, lack of monetary reimbursement for implementation, perceived or actual lack of backward compatibility,
proprietariness, and conformance concerns.48 Fortunately, the
landscape is changing, largely driven by the widespread uptake
of EHRs necessitated by MU.49 MU Stage 2 cites many specific
standards (including C-CDA), and is hoped to accelerate the
dissemination and implementation of healthcare standards.50
Where the information needs of eCOTPS and MU certification
criteria overlap—as in medication and problem lists—adoption
is easier and reuse of data collected during clinical care is supported. Standards risk becoming obsolete if they are not used;
thus, it was a goal to implement the eCOTPS soon after DSTU
designation was achieved. ASCO was fortunate to identify two
enthusiastic early adopters: Health Story and Athena. Each of
these experiences was informative and will help guide and refine the eCOTPS so that it may eventually reach HL7 normative
status. Although the Health Story demonstration was constructed around a synthetic patient scenario, it is realistic in
that it brought together multiple for-profit and not-for-profit organizations who had to interoperate to succeed. Athena is a
very large network that cares for almost 45% of the breast cancer patients diagnosed in California. In their evaluation, Athena
found that using CDA-based data exchange for the eCOTPS is
likely to require additional CDA-R2 templates and vocabulary
bindings, which invokes nontrivial efforts. One possible solution
is the open source “greenCDA” implementation toolset (http://
www.openmapsw.com/index.htm), which was found to significantly reduce the time and resources needed for the implementation of the eCOTPS. After new templates were added, the
creation of the complete Athena CDA and the simplified
greenCDA with the tools was simplified for both designers and
implementers. Tools like this are an essential part of efficiently
and correctly implementing CDA-R2 specification.
There is an alternative oncology-specific nursing standard
for continuity of care, the HL7 v3 Care Record,23,51 and like
eCOTPS, it supports electronic information exchange among
cancer providers and patients. The use case for the HL7 v3
Care Record was to provide summary information to inform
nursing home care. This leaves an unmet need for a standard
to serve as a form of ongoing communication to augment the
overall coordination of care for an oncology patient, during and
after treatment. The C-CDA framework was chosen instead of
v3 messages because the eCOTPS requires a canonical human-readable format, which can be displayed on ubiquitous
tools, given a single style sheet. V3 messages (e.g., Care
Record) require a custom style sheet that is not reusable across
message or document types, increasing the level of effort to

DISCUSSION
In the mid-2000s, ASCO recognized that a standardized summary of cancer care was necessary. This recognition was fortuitously driven by an improved outlook for many cancer patients
that has increasingly created the need for summarization and
survivorship programs.42 With many cancer patients now outliving their disease by years or decades, major life events such
as geographic relocation or changes in employment status are
common.43,44 Likewise, with an expanding survivor population,
medical oncologists have increasingly relied on resumption of
care by PCPs after completion of primary therapy. Despite
ASCO’s successful creation of a suite of TPS templates, uptake
was unimpressive, partly because these templates remained
paper-based, tedious, and time-consuming to complete.45,46
According to one large recent survey, only one-third of cancer
survivors are currently receiving treatment summaries.47
The DSIT embarked on the process of translating and modifying the paper TPS templates to an interoperability standard
with three goals in mind: (1) to create a rigorously vetted

7

RESEARCH AND APPLICATIONS

Cancer Checklists, the California Cancer Registry, and the
National Cancer Institute Cancer Therapy Evaluation Program
Common Data Elements for clinical trials. Athena’s data elements were selected based on their perceived importance in
comprehensive care coordination, point-of-care data capture,
clinical registration, and quality improvement. Athena’s data
aligned well with ASCO’s BCTPS elements, making the eCOTPS
an ideal data exchange standard for the project. However, because the eCOTPS is, by design, a brief planning and summary
document, certain data required for certain aspects of care delivery were not represented; examples are measurements of
specific lesions and exact radiation treatment dosages (when
applicable).
A web-based Athena application with dynamic data entry
forms (“data entry checklists”) was mapped to the eCOTPS,
essentially ensuring semantic correctness when converting
question/answer pairs to HL7 observations. This was managed
by a team that included an HL7 expert, a breast cancer informatics analyst, and a software engineer. Although the Athena
checklists have data element groupings similar to the ASCO
TPS data elements (Table 1), they are more comprehensive
and certain concepts are more granular than those in the
eCOTPS. Additionally, mapping from the checklist format to the
eCOTPS document format was more complex because some of
Athena’s checklists contained data elements that mapped to
multiple eCOTPS document sections. It was then necessary to
perform many-to-many mapping of certain checklist data elements to eCOTPS document sections (i.e., document sections
containing data elements from more than one checklist). For
example, the Athena checklist Initial Diagnosis and Treatment
has detailed information about the individual lesions discovered
by one or more imaging techniques, including their identity, location, size, invasive grade, and whether molecular or genetic
testing was performed. Representing these data in the eCOTPS
required the creation of additional CDA-R2 templates including
new vocabulary bindings.

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

RESEARCH AND APPLICATIONS

implement for clinical end users. Thus, successful implementation of the v3 Care Record message implies the existence of a
relatively sophisticated infrastructure. The eCOTPS use case
suggests that it can be deployed across a range of applications,
from sophisticated EHRs to any browser-enabled device.
Pragmatically, CDA is much more widely implemented than v3
messages in the United States and most countries with a national health information technology initiative. Wide implementation is one sign that the specification offers practical
advantages and that it will be easier to recruit vendors to adopt
the specification. In terms of modularity, the HL7 v3 RIM framework is ideal. Specifically, many elements are common across
cancer (which is not a distinct disease, but rather 100þ siteand histology-specific subtypes52). Other elements are quite
specific to subtypes of cancer, such as estrogen receptor status in breast cancer. A template-driven extensible standard
provides significant flexibility as more cancer subtype-specific
TPSs and other cancer-specific areas, such as survivorship,
are standardized in future work.
Whereas these initial implementations are mostly positive,
the ultimate success of the eCOTPS will be dependent on three
critical factors: (1) widespread uptake by EHR vendors with oncology-focused solutions; (2) auto-population of data elements
to eliminate redundant data entry; and (3) the willingness of
large practices and hospital systems to fully embrace seamless
interoperability. Although policy levers such as MU play their
part, a culture change toward sharing and transparency, while
simultaneously respecting patient privacy and autonomy, is still
needed. It remains to be seen whether disruptive standards
such as HL7’s Fast Healthcare Interoperability Resource
(FHIR)53 will change the general calculus of interoperability, as
well as the future of eCOTPS; fortunately, there are encouraging signs of cooperation between the FHIR and CDA communities (http://www.hl7.org/implement/standards/fhir/comparisoncda.html). Thus, if FHIR becomes a successful and widely used
standard, implementation of eCOTPS using FHIR will be
feasible.
As Athena goes live with the eCOTPS, transmitting breast
cancer data throughout the UC Healthcare System, ASCO’s
DSIT is already expanding the eCOTPS, adding data from the
Colon Cancer Adjuvant TPS. At the same time, the DSIT is undertaking improvements to the existing eCOTPS based on
Athena’s experiences. The DSIT plans to iteratively expand the
eCOTPS so that over the next few years, it will include diseasespecific data for the most prevalent cancers, critical survivorship information, and patient-reported data.

cancer care by allowing the efficient transmission of reliable,
meaningful, and up-to-date clinical data between all stakeholders involved in the cancer journey.

COMPETING INTERESTS
The authors have declared that no competing interests exist.

CONTRIBUTORSHIP
J.L.W., S.E.M., K.S.H., J.C.K., P.P.Y., L.N.S., D.K.M., L.A., Z.G.,
and E.P.A. participated in the design and creation of the
eCOTPS HL7 standard; J.L.W., L.A., G.A.K., Z.G., and E.P.A.
participated in the implementation of eCOTPS in the Health
Story Project; M.H., M.S., A.S.F., S.E.D., and L.E. participated
in the implementation of eCOTPS in the Athena Breast Health
Network; J.L.W. and S.E.M. wrote the initial draft manuscript;
all authors contributed to the manuscript revisions and approved the final manuscript.

FUNDING
This work was supported in part by ASCO, the University of
California Office of the President, the Safeway Foundation, and
Grant 90HT0029/01-02 (ONC). S.E.M, an ASCO employee, assisted in the preparation of the manuscript. ASCO had no role
in the study design, data collection and analysis, or decision to
publish. The other funders had no role in the study design, data
collection and analysis, decision to publish, or preparation of
the manuscript.

ACKNOWLEDGEMENTS
We would like acknowledge the contributions of Sarah E. Davis
MS (Department of Surgery, University of California, San
Francisco). We would like to extend special thanks to John
Paganini and the following vendors and organizations who participated in the Health Story Project’s demonstration at HIMSS
2014: Academy of Nutrition and Dietetics, Axesson Inc. (who
also provided portions of the graphic content for Figure 3),
Continua Health Alliance, Datuit LLC, Foothold Technology Inc.,
Healthwise Inc., InfraWare Inc., Inofile LLC, M*Modal IP LLC,
Medical Informatics Engineering, and Verizon Communications
Inc. We would also like to thank Jean Brook for help with editing, and the ASCO CancerLinQ Advisory Group for their generous support.

REFERENCES

CONCLUSION

1. Moran MS, Kaufman C, Burgin C, et al. What currently defines a breast center? Initial data from the national accreditation program for breast centers. J Oncol Pract. 2013;9(2):
e62–e70.
2. Ganz PA, Casillas J, Hahn EE. Ensuring quality care for cancer survivors: implementing the survivorship care plan.
Semin Oncol Nurs. 2008;24(3):208–217.
3. Hewitt M, Simone JV. Ensuring Quality Cancer Care.
Washington, DC: National Academies Press; 1999.

The eCOTPS is the first oncology-specific CDA-R2 standard to
achieve DSTU status through the HL7 balloting process. Early
implementers have demonstrated that the standard is functional and adaptable to different needs. Having the flexibility to
create additional CDA-R2 templates is essential for supporting
real-world patient care. Continuing experience gained through
trial implementation will inform the DSIT’s future work.
Oncology interoperability standards will improve the quality of

8

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

9

RESEARCH AND APPLICATIONS

and flexible documentation. J Am Med Inform Assoc. 2011;
18(2):181–186.
20. Inman AW. The cancer journey. Biofeedback. 2010;38(1):
24–27.
21. Foote RS. The challenge to the medical record. JAMA Int
Med. 2013;173(13):1171–1172.
22. Meaningful Use of Electronic Health Records [cited 2014
July 8th]. http://www.cdc.gov/cancer/npcr/meaningful_
use.htm
23. Duim M, Boterenbrood F, Goossen WT. Continuity of care
with HL7 v3 care record for oncology nursing. Stud Health
Technol Inform. 2014;201:476–482.
24. Shortliffe EH, Cimino JJ. Biomedical Informatics. New York,
NY: Springer; 2006.
25. Hewitt M, Greenfield S, Stovall E. From Cancer Patient to
Cancer Survivor: Lost in Transition. Washington, DC:
National Academies Press; 2005.
26. Berggren RE, Curiel TJ. After the storm—health care infrastructure in post-Katrina New Orleans. N Engl J Med. 2006;
354(15):1549–1552.
27. Joseph DA, Wingo PA, King JB, et al. Use of state cancer
surveillance data to estimate the cancer burden in disasteraffected areas—Hurricane Katrina, 2005. Prehospital
Disaster Med. 2007;22(4):282–290.
28. Nissen MJ, Beran MS, Lee MW, et al. Views of primary care
providers on follow-up care of cancer patients. Family Med.
2007;39(7):477–482.
29. Bober SL, Recklitis CJ, Campbell EG, et al. Caring for cancer
survivors: a survey of primary care physicians. Cancer.
2009;115(18 Suppl):4409–4418.
30. Mao JJ, Bowman MA, Stricker CT, et al. Delivery of survivorship care by primary care physicians: the perspective of
breast cancer patients. J Clin Oncol. 2009;27(6):933–938.
31. Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group.
Am J Clin Oncol. 1982;5(6):649–655.
R
32. HL7 Implementation Guide for CDAV
Release 2, IHE Health
Story Consolidation, Release 1.1—US Realm 2012 [cited
September 5, 2014]. http://www.hl7.org/implement/
standards/product_brief.cfm?product_id¼258.
33. HL7 Version 3: Reference Information Model (RIM) [cited
September 5, 2014]. http://www.hl7.org/implement/
standards/product_brief.cfm?product_id¼77.
34. Centers for Medicare and Medicaid Services (US). Medicare
and Medicaid Electronic Health Record Incentive
Program—Stage 2 (CMS-0044-P); the Federal Register,
Washington, DC; 2011.
35. Rosenbloom ST, Harris P, Pulley J, et al. The mid-South
clinical data research network. J Am Med Inform Assoc.
2014;21(4):627–632.
36. Klann JG, Buck MD, Brown J, et al. Query Health: standards-based, cross-platform population health surveillance.
J Am Med Inform Assoc. 2014;21(4):650–656.
37. Implementation Guide for Ambulatory Healthcare Provider
Reporting to Central Cancer Registries, Release 1.0 2012
[cited September 5, 2014]. http://www.cdc.gov/phin/

4. Levit L, Balogh E, Nass S, Ganz PA. Delivering High-quality
Cancer Care: Charting a New Course for a System in Crisis.
Washington, DC: Institute of Medicine; 2013.
5. Abernethy AP, Etheredge LM, Ganz PA, et al. Rapid-learning
system for cancer care. J Clin Oncol. 2010;28(27):
4268–4274.
6. Friedman CP, Wong AK, Blumenthal D. Achieving a nationwide learning health system. Sci Transl Med. 2010;2(57):
57cm29.
7. Sledge GW, Hudis CA, Swain SM, et al. ASCO’s approach to
a learning health care system in oncology. J Oncol Pract.
2013;9(3):145–148.
8. Sledge GW Jr, Miller RS, Hauser R. CancerLinQ and the future of cancer care. American Society of Clinical Oncology
Educational Book. American Society of Clinical Oncology
Meeting. 2013;33:430–434.
9. Yu P, Artz D, Warner J. Electronic health records (EHRs):
Supporting ASCO’s vision of cancer care. American Society
of Clinical Oncology Educational Book. American Society of
Clinical Oncology Meeting. 2014;34:225–231.
10. Yu PP. The evolution of oncology electronic health records.
Cancer J. 2011;17(4):197–202.
11. Warner J, Hochberg E. Where is the EHR in oncology? J Natl
Compr Canc Netw. 2012;10(5):584–588.
12. Dolin RH, Alschuler L, Beebe C, et al. The HL7 clinical document architecture. J Am Med Inform Assoc. 2001;8(6):
552–569.
13. Alschuler L, Dolin R, Boyer S, Beebe C. HL7 Clinical
Document Architecture Framework, Release 1.0. ANSIApproved HL7 Standard, Nov. 2000. Ann Arbor, MI: Health
Level Seven, Inc; 2000.
14. Dolin RH, Alschuler L, Boyer S, et al. HL7 clinical document architecture, Release 2. J Am Med Inform Assoc.
2006;13(1):30–39.
15. Dolin RH, Alschuler L, Behlen F, et al. HL7 document patient
record architecture: an XML document architecture based
on a shared information model. Proc AMIA Symp. 1999;
52–56.
16. Paterson GI, Shepherd M, Wang X, et al. Using the XMLbased clinical document architecture for exchange of structured discharge summaries. In: Proceedings of the 35th
Annual Hawaii International Conference on System
Sciences (HICSS) June, 2006; Salt Lake City, Utah. 2002;
1200–1209.
17. Bilykh I, Jahnke JH, McCallum G, et al. Using the clinical
document architecture as open data exchange format for
interfacing EMRs with clinical decision support systems. In:
19th IEEE International Symposium on Computer-Based
Medical Systems (CBMS) January 2002, Hilton Waikoloa
Village, Island of Hawaii. 2006:855–860.
18. Centers for Medicare & Medicaid Services. 2014 Definition
Stage 1 of Meaningful Use [cited 2014 June 15th]. http://
www.cms.gov/Regulations-and-Guidance/Legislation/
EHRIncentivePrograms/Meaningful_Use.html.
19. Rosenbloom ST, Denny JC, Xu H, et al. Data from clinical
notes: a perspective on the tension between structure

 Warner JL, et al. J Am Med Inform Assoc 2015;0:1–10. doi:10.1093/jamia/ocu015, Research and Applications

38.

39.

40.

41.

RESEARCH AND APPLICATIONS

42.
43.

44.

45. Salz T, Oeffinger KC, McCabe MS, et al. Survivorship
care plans in research and practice. CA Cancer J Clin.
2012;62(2):101–117.
46. Salz T, McCabe MS, Onstad EE, et al. Survivorship care
plans: is there buy-in from community oncology providers?
Cancer. 2014;120(5):722–730.
47. Rechis R, Beckjord EB, Nutt S. Potential benefits of treatment summaries for survivors’ health and information
needs: results from a LIVESTRONG survey. J Oncol Pract.
2014;10(1):75–78.
48. Hammond WE, Jaffe C, Cimino JJ, et al. Standards in biomedical informatics. Biomedical Informatics. SpringerVerlag, London, UK; 2014:211–253.
49. Kuperman GJ. Health-information exchange: why are we
doing it, and what are we doing? J Am Med Inform Assoc.
2011;18(5):678–682.
50. Burke JG, Lich KH, Neal JW, et al. Enhancing dissemination
and implementation research using systems science methods. Int J Behav Med. 2014:1–9.
51. Goossen W, Langford LH. Exchanging care records using
HL7 V3 care provision messages. J Am Med Inform Assoc.
2014;21(e2):e363–e368.
52. Kleihues P, Sobin LH. World Health Organization classification of tumors. Cancer. 2000;88(12):2887.
53. Bender D, Sartipi K. HL7 FHIR: An Agile and RESTful approach to healthcare information exchange. In: 26th
International Symposium on Computer-Based Medical
Systems (CBMS). University of Porto, Porto, Portugal, June,
2013;326–331.

library/guides/Implementation_Guide_for_Ambulatory_
Healthcare_Provider_Reporting_to_Central_Cancer_
Registries_August_2012.pdf.
R
HL7 Implementation Guide for CDAV
R2: Quality Reporting
Document Architecture—Category I (QRDA) DSTU Release
2 (US Realm) 2012 [cited September 5, 2014]. http://
www.hl7.org/implement/standards/
product_brief.cfm?product_id¼35.
HL7 Version 3 Implementation Guide: Family History/
Pedigree Interoperability, Release 1 [cited July 8, 2014].
http://www.hl7.org/implement/standards/
product_brief.cfm?product_id¼301.
R
HL7 Implementation Guide for CDA, Release 2V
: Clinical
Oncology Treatment Plan and Summary, DSTU Release 1
2013 [cited September 5, 2014]. http://www.hl7.org/imple
ment/standards/product_brief.cfm?product_id¼327.
Elson SL, Hiatt RA, Anton-Culver H, et al. The Athena Breast
Health Network: developing a rapid learning system in
breast cancer prevention, screening, treatment, and care.
Breast Canc Res Treat. 2013;140(2):417–425.
Centers for Disease Control and Prevention (CDC). Cancer
survivorship—United States, 1971–2001. MMWR Morbidity
and Mortality Weekly Report. 2004;53(24):526–529.
Mols F, Thong MS, Vreugdenhil G, et al. Long-term cancer
survivors experience work changes after diagnosis: results of a population-based study. Psycho-oncology. 2009;
18(12):1252–1260.
Short PF, Vasey JJ, Tunceli K. Employment pathways in a
large cohort of adult cancer survivors. Cancer. 2005;103(6):
1292–1301.

AUTHOR AFFILIATIONS
....................................................................................................................................................
1

Department of Medicine, Division of Hematology & Oncology,
Vanderbilt University, Nashville, TN, USA

8

2

Department of Biomedical Informatics, Vanderbilt University,
Nashville, TN, USA

9

3

10

UNC Lineberger Comprehensive Cancer Center, Chapel Hill,
NC, USA
Department of Pathology and Laboratory Medicine, University
of California, Davis, Sacramento, CA, USA

Quality and Guidelines Division, American Society of Clinical
Oncology, Alexandria, VA, USA

Shafarman Consulting, Oakland, CA, USA

11

Department of Surgery, University of California, San
Francisco, CA, USA

4

Massachusetts General Hospital, Harvard Medical School,
Boston, MA, USA

12

Department of Radiology, University of California, San
Francisco, CA, USA

5

Department of Internal Medicine, Division of Hematology/
Oncology, University of Michigan, Ann Arbor, MI, USA

13

Lantana Consulting Group, East Thetford, VT, USA

6

Palo Alto Medical Foundation, Sunnyvale, CA, USA

14

Tisch Cancer Institute, Icahn School of Medicine at Mount
Sinai, New York, NY, USA

7

Dana-Farber Cancer Institute, Harvard Medical School,
Boston, MA, USA

10