ISO/TR 22221:2006

TECHNICAL ISO/TR
REPORT 22221
First edition
2006-11-01

Health informatics — Good principles and
practices for a clinical data warehouse
Informatique de santé — Principes et indications d'exploitation d'un
entrepôt de données cliniques




Reference number
ISO/TR 22221:2006(E)
©
ISO 2006

---------------------- Page: 1 ----------------------
ISO/TR 22221:2006(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.


©  ISO 2006
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland

ii © ISO 2006 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/TR 22221:2006(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope .1
2 Terms and definitions .1
3 Data warehouse features for a health organization .3
3.1 General.3
3.2 Quality assurance and care delivery .4
3.3 Evaluation and innovation of health procedures and technologies .4
3.4 Disease surveillance, epidemiology, and public health .4
3.5 Planning and policy.5
3.6 Knowledge discovery.5
3.7 Education.5
4 Description in detail of each category.5
4.1 General.5
4.2 Quality assurance and care delivery .5
4.3 Services and technology evaluation and innovation.6
4.4 Disease surveillance, epidemiology and public health .7
4.5 Planning and policy.7
4.6 Knowledge discovery.8
4.7 Education.8
5 Governance and ethics considerations of clinical data .9
5.1 General.9
5.2 Governance requirements for data integrity and management.9
5.3 Perspectives of individual and social protection.13
5.4 Policies about people.18
5.5 Security review and audit .18
6 Architecture.19
6.1 Existing work on data warehousing .19
6.2 Characteristics of a clinical data warehouse.20
6.3 Methodology for clinical data warehouse development.25
6.4 Basic data models .26
6.5 Security and privacy.33
7 Metadata and education.34
7.1 Importance of metadata .34
7.2 Collection mechanisms.34
7.3 Ownership .34
7.4 Common definitions and standardization.35
7.5 Data quality.35
7.6 Change management .35
7.7 Education.35
8 Analytical and reporting tools.35
8.1 General.35
8.2 Deployment approaches .36
8.3 Enterprise business intelligence suites .36
9 Organizational approach.38
9.1 General.38
9.2 Multidisciplinary approach .39
Bibliography .40
© ISO 2006 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO/TR 22221:2006(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
In exceptional circumstances, when a technical committee has collected data of a different kind from that
which is normally published as an International Standard (“state of the art”, for example), it may decide by a
simple majority vote of its participating members to publish a Technical Report. A Technical Report is entirely
informative in nature and does not have to be reviewed until the data it provides are considered to be no
longer valid or useful.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 22221 was prepared by Technical Committee ISO/TC 215, Health informatics.
iv © ISO 2006 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/TR 22221:2006(E)
Introduction
0.1 General
A clinical data warehouse (CDW) is regarded as conceptually distinct from the clinical data repository of an
operational electronic health record. It is as yet a largely under-implemented and under-exploited resource
which, however, has many possible features with health care, education and research aspects. Such features
include:
⎯ quality assurance,
⎯ feedback to individuals and teams of caregivers,
⎯ infectious disease or medication surveillance, and
⎯ evaluation of organizational continuity as patients move between organizations.
Such data are also a crucial link between individual care, organizational and public health needs. The CDW
can provide a system view of different perspectives and levels of activity that cannot be provided easily and
properly by an operational system; these different levels and perspectives can require different characteristics
of the associated datasets.
This data access also has social, legal and ethical, epidemiological and informatics challenges, which may
variably impact the use dimensions of a CDW. This will be of particular importance as pedigree and genetic
data content of CDWs increases over time.
0.2 Purpose of this Technical Report
The data warehouse is not yet widely used by health organizations. There still is no common knowledge and
understanding about the creation and exploitation of data warehouse features by health organizations. The
purpose of this Technical Report is to enable the different CDW users to have a uniform understanding of a
CDW, including both general principles and particular characteristics of different major use perspectives.
0.3 Benefits of this Technical Report
The CDW is presently a largely under-exploited resource of invaluable information for supporting the service,
research and educative missions of the health system. It enables practice assessment as well as knowledge
discovery, but it also has the potential to support more efficient and effective innovation, as well as being an
essential tool for interdisciplinary collaboration. This Technical Report is intended to help orientate future
developments by creating the preliminary work for a technical specification of a clinical data warehouse and
leading to the development of standards for different use applications.
0.4 Target users
Target users include all stakeholders in the health system, public and private, including (but not limited to):
⎯ clinicians and para-clinical personnel,
⎯ administrators,
⎯ educators,
⎯ epidemiologists,
⎯ economists,
© ISO 2006 – All rights reserved v

---------------------- Page: 5 ----------------------
ISO/TR 22221:2006(E)
⎯ researchers,
⎯ system developers,
⎯ data and modelling specialists,
⎯ accreditation organizations,
⎯ citizen organizations, and
⎯ policy makers.

vi © ISO 2006 – All rights reserved

---------------------- Page: 6 ----------------------
TECHNICAL REPORT ISO/TR 22221:2006(E)

Health informatics — Good principles and practices for a
clinical data warehouse
1 Scope
The focus of this Technical Report is clinical databases or other computational services, hereafter referred to
as a clinical data warehouse (CDW), which maintain or access clinical data for secondary use purposes. The
goal is to define principles and practices in the creation, use, maintenance and protection of a CDW, including
meeting ethical and data protection requirements and recommendations for policies for information
governance and security. A distinction is made between a CDW and an operational data repository part of a
health information system: the latter may have some functionalities for secondary use of data, including
furnishing statistics for regular reporting, but without the overall analytical capacity of a CDW.
This Technical Report complements and references standards for electronic health records (EHR), such as
ISO/TS 18308, and contemporary security standards in development. This Technical Report addresses the
secondary use of EHR and other health-related and organizational data from analytical and population
perspectives, including quality assurance, epidemiology and data mining. Such data, in physical or logical
format, have increasing use for health services, public health and technology evaluation, knowledge discovery
and education.
This Technical Report describes the principles and practices for a CDW, in particular its creation and use,
security considerations, and methodological and technological aspects that are relevant to the effectiveness of
a clinical data warehouse. Security issues are extended with respect to the EHR in a population-based
application, affecting the care recipient, the caregiver, the responsible organizations and third parties who
have defined access. This Technical Report is not intended to be prescriptive either from a methodological or
a technological perspective, but rather to provide a coherent, inclusive description of principles and practices
that could facilitate the formulation of CDW policies and governance practices locally or nationally.
2 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
2.1
clinical data repository
CDR
operational data store that holds and manages clinical data collected from service encounters at point of
service locations
EXAMPLE Point of service locations include hospitals and clinics.
NOTE Data from a CDR can be fed to the EHR for that client, such that the CDR is recognized as a source system
for the EHR. The CDR can be used to trigger alerts in real time.
2.2
clinical data warehouse
CDW
grouping of data accessible by a single data management system, possibly of diverse sources, pertaining to a
health system or sub-system and enabling secondary data analysis for questions relevant to understanding
the functioning of that health system, and hence supporting proper maintenance and improvement of that
health system
NOTE A CDW tends not to be used in real time; however, depending on the rapidity of transfer of data to the data
warehouse, and data integrity, near real time applications are not excluded.
© ISO 2006 – All rights reserved 1

---------------------- Page: 7 ----------------------
ISO/TR 22221:2006(E)
2.3
dashboard
user interface based on predetermined data fields that facilitate domain-specific data queries, and suited to
regular use with minimal training
2.4
data dictionary
database used for data that refers to the use and structure of other data, i.e. a database for the storage of
metadata
[ISO/IEC 11179-1:2004]
2.5
data mart
subject area of interest within the data warehouse
EXAMPLE An inpatient data mart.
NOTE Data marts can also exist as a stand-alone database tuned for query and analysis, independent of a data
warehouse.
2.6
data warehouse
subject-oriented, integrated, time-variant and non-volatile collection of data
[1]
NOTE The term “data warehouse” is attributed to Inmon .
2.7
drill down
exploration of multidimensional data which makes it possible to move down from one level of detail to the next
depending on the granularity of data
EXAMPLE Number of patients by departments and/or by services.
2.8
episode of care
identifiable grouping of health care-related activities characterized by the entity relationship between the
subject of care and a health care provider, such grouping determined by the health care provider
[ISO/TS 18308:2004]
2.9
health indicator
single summary measure, most often expressed in quantitative terms, that represents a key dimension of
health status, the health care system, or related factors
NOTE A health indicator is to be informative and also sensitive to variations over time and across jurisdictions.
[ISO/TS 21667:2004]
2.10
metadata
information stored in the data dictionary that describes the content of a document
NOTE In a data warehouse context, metadata are data structure, constraints, types, formats, authorizations,
privileges, relationships, distinct values, value frequencies, keywords, and users of the database sources loaded in the
data warehouse and the data warehouse itself. Metadata help users, developers and administrators for information
management.
2 © ISO 2006 – All rights reserved

---------------------- Page: 8 ----------------------
ISO/TR 22221:2006(E)
2.11
online analytical processing
OLAP
set of applications developed for facilitating the collection, analysis and reporting of multidimensional data
[3]
NOTE The term “OLAP” is attributed to Codd .
2.12
organization
group of people that have their own structure rules and culture in order to work together to achieve goals
and/or to provide services through processes, equipments and technologies, etc.
2.13
performance indicator
measure that supports evaluation of an aspect of performance and its change over time
2.14
persistent data
data in a final form intended as a permanent record, such that any subsequent modification is recorded
together with the original data
2.15
roll up
method of regrouping and aggregating multidimensional data to move up the hierarchy into larger units
EXAMPLE Weekly count of patients aggregated by quarter or by year.
2.16
secondary data use
use of data for additional purposes other than the primary reason for their collection, adding value to this data
2.17
star schema
dimensional modelling concept that refers to a collection of fact and dimension tables
3 Data warehouse features for a health organization
3.1 General
The roles and capacities of each of the operational databases and informational databases or data
warehouses are complementary. An operational database is designed to perform transactions such as adding,
changing or deleting a patient. It has a limited capacity for data analysis supporting online care delivery.
Secondary data use refers to the exploitation of already existing persistent data. The concept of a clinical data
warehouse refers to a set of secondary data for analytic purposes relevant to a health organization. As health
care takes place in different organizations, including home care, family practice and care in institutions with
different missions, the notion of organization can apply to just one of these entities or to a group of entities,
e.g. a regional, provincial or national system of care. An organization uses different data sources, e.g. finance
data is usually separate from patient data. For certain purposes, it is appropriate to link finance and patient
data to analyse resource use. This clinical-administrative interface is one feature of a clinical data warehouse.
A data warehouse can accept data from several different databases, including from other human services
organizations such as social services or from technical devices, to facilitate different analyses pertinent for
one or more of the organizations. As described in more detail in Clause 7, and as is the case for all data
warehouses, there is a preliminary need to address different aspects of data quality prior to its transfer to the
data warehouse. This clause describes the use of a clinical data warehouse from different important
perspectives.
© ISO 2006 – All rights reserved 3

---------------------- Page: 9 ----------------------
ISO/TR 22221:2006(E)
3.2 Quality assurance and care delivery
The predominant paradigm for quality assurance is a cycle consisting of problem definition, data collection,
data analysis, and planning for problem resolution. The step of data collection often depends on searching for
this data in a paper record, which is both time-consuming and possibly frustrating, depending on the quality of
the record’s maintenance. Although the paper record will not completely disappear, at least for some time,
with the advent of the EHR and increasing use of electronic data collection, the CDW should dramatically
reduce the time for data access and analysis. It should enable quality control teams to return from abstracted
data analysis to the original data, to explore and ask related questions to obtain additional data to strengthen
the evidence on the nature of the problem. The CDW is also a source of prospective data for monitoring
improvement. It can be used to establish trends, identify changes and provide alerts. Knowing in advance the
data categories that could be followed over time enables the creation of tailored interfaces, sometimes known
as a dashboard, which enable checking of updated data as well as drill down to detailed data for a particular
sub-question.
3.3 Evaluation and innovation of health procedures and technologies
An extension of the concept of quality assurance is the assessment of the impact with the introduction of a
new technology or a change in procedure. The paradigm for new technology development is a series of steps
that start in the research context and move progressively through
⎯ development,
⎯ performance, robustness and safety testing,
⎯ controlled clinical trials, and
⎯ market release and market surveillance.
The CDW has two roles: one at the beginning and one at the end of this process. The CDW is increasingly a
source of information on existing patterns of care, and especially the relative importance of particular
investigations and treatments. Indeed, it is this process which is under continual examination as part of quality
assurance. Companies and research groups can use this information to direct their development choices,
selecting areas of testing and treatment where significant improvement might be obtained. At the end of the
process, following the introduction of a new technology, the CDW becomes a source of data for surveillance of
optimal use and also for evaluation of the impact of its use, as well as unexpected findings. The importance of
post-market surveillance for ensuring appropriate uptake and early awareness of unexpected benefit or risk is
already well known for new pharmaceuticals.
3.4 Disease surveillance, epidemiology, and public health
The CDW is a rich source of information that can profile communities and assess the health status to assist in
planning, expose changes in patterns of care, or trends in use of procedures, or disease profiles including
infections. The need for a CDW has been particularly promoted by epidemiologists and health services
researchers, who need to understand a population profile of health and disease, aiming for disease prevention
and risk minimization, as well as evaluation of variation in population outcomes and their causes. A major
impediment is always the access to quality data, and the need to rely on imperfect data from a mix of sources
with heterogeneous data organization. It is still common to come across a population data set which provides
a clue of disease variation, but where the next step of getting more detailed data that might explain this
variation is practically impossible. The CDW should be able to link to data sets or use indicators from other
human services organizations, such as justice, education, social services, etc., for public health to analyze
population health and related community needs. Depending on access, networking and permission, the CDW
represents a new opportunity to delve finely into causes of variation and to link data between intervention and
outcome, e.g. to better assess whether a preventative procedure results in improved outcomes. Furthermore,
the CDW could be a source of information to understand probability distributions for different health care
activities. The patterns could be used to develop simulation models for macro or micro system components to
explore different options.
4 © ISO 2006 – All rights reserved

---------------------- Page: 10 ----------------------
ISO/TR 22221:2006(E)
3.5 Planning and policy
Administrative and policy decision making depends on access to objective data, usually in abstracted form. In
common with clinical decision makers, there can be a need to explore the data and to move from abstracted
to particular data. Abstracted analysed data from the CDW may become a main way that data is shared
between decision makers with different roles, such as between clinicians and administrators, and form a basis
of negotiation: hence data should as far as possible be clearly presented and interpretable for a given
purpose. Health and performance indicators are increasingly used for quality and economic reasons, as
metadata can be, describing the way in which the indicators are derived. Their effectiveness depends on
efficient data access and continual examination of validity, which can be supported by analysis of related data
from the CDW, including comparison across systems of care. Certain abstractions are subject to coding,
increasingly using semi-automated methodologies dependent on the quality of primary data. These codes can
be available in the CDW.
3.6 Knowledge discovery
As well as providing evidence for quality assurance and to support technology assessment, the CDW using
different analytical methodologies could be a source of unexpected new knowledge about disease evolution
and treatment response, similar to that previously discussed concerning post-market surveillance. This should
most probably be in a sub-population where manifestations are uncommon and the CDW provides the
opportunity to analyse these cases in detail in comparison to the population to which the sub-population
belongs, a task which was previously very difficult because of variable data quality and access difficulties.
3.7 Education
The CDW is a window on actual health care practice. It is an opportunity to study disease and practice
variation, and hence a repository of teaching material of clinical cases and case management that can be
correlated to the teaching of best practice. The teaching of quality assurance is variably practiced at present.
Being a key resource for quality assurance, including query tools such as the dashboard, the CDW should
provide an enhanced quality assurance education environment.
4 Description in detail of each category
4.1 General
The more detailed descriptions in this clause provide an appreciation of the different processes and roles
related to the perspectives of CDW use. Security and privacy issues, as well as different analytical tools to
support the CDW in these different perspectives, are considered in subsequent clauses.
4.2 Quality assurance and care delivery
4.2.1 Description of business processes involved
Data about patient care are collected as a function of an area of concern identified to or by a quality
assurance professional or team. Detailed analysis may lead to a requirement for additional evidence before a
correction plan is proposed and adopted. Regular data collection can check subsequently
...