Public Health
Conceptual Data Model
Premiere Edition
U.S. Department of Health and Human Services
Public Health Service
Centers for Disease Control and Prevention (CDC)
Atlanta, Georgia 30333
July 2000
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION JULY 2000
Table of Contents
INTRODUCTION............................................................................................................. 1
BACKGROUND .................................................................................................................. 2
GOALS AND OBJECTIVES................................................................................................... 4
PROJECT SCOPE................................................................................................................ 8
GUIDE TO UNDERSTANDING
THE PUBLIC HEALTH CONCEPTUAL DATA MODEL ........................................ 9
SUBJECT AREAS.............................................................................................................. 10
CLASSES AND RELATIONSHIPS........................................................................................ 11
ATTRIBUTES AND DATATYPES ........................................................................................ 15
KEY CONCEPTS............................................................................................................... 17
IMPLEMENTING THE PHCDM ......................................................................................... 19
PUBLIC HEALTH CONCEPTUAL DATA MODEL................................................ 25
HEALTH-RELATED ACTIVITIES SUBJECT AREA ............................................................... 29
LOCATIONS SUBJECT AREA ............................................................................................ 44
MATERIALS SUBJECT AREA ............................................................................................ 51
PARTIES SUBJECT AREA ................................................................................................. 57
APPENDICES ................................................................................................................. 64
DATATYPES .................................................................................................................... 65
MODEL SCENARIO .......................................................................................................... 77
FREQUENTLY ASKED QUESTIONS ................................................................................... 81
GLOSSARY...................................................................................................................... 87
BIBLIOGRAPHY............................................................................................................... 91
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 1 OF 91 JULY 2000
INTRODUCTION
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 2 OF 91 JULY 2000
Introduction
This document presents the Public Health Conceptual Data Model (PHCDM).
The document is comprised of four sections. Section one includes this
introduction, model background, a summary of the goals and objectives for the
model, and a project scope statement. Section two is a guide to understanding the
data model. Section three includes the graphical representation of the data model
and the supporting data dictionary. Section four consists of a series of appendices
including datatype definitions, a model scenario, frequently asked questions, a
glossary, and a bibliography.
The purpose of the Public Health Conceptual Data Model is to document the
information needs of public health so that the Centers for Disease Control and
Prevention (CDC) and its state and local partners in public health can:
• Establish data standards for public health, including data definitions,
component structures (such as for complex datatypes), code values, and data
use;
• Collaborate with national health informatics standards setting bodies to define
standards for the exchange of information among public health agencies, and
healthcare providers;
• Construct computerized information systems that conform to established data
and data interchange standards for use in the management of data relevant to
public health.
This model is a work in progress, and is being developed with the participation of
our public health partners. CDC is developing a process to receive comments
from its partners, address issues raised, and provide feedback on the disposition of
the comments. In the interim, contact the CDC Health Information and
Surveillance Systems Board (HISSB) Executive Secretariat with questions and
comments regarding the model: [email protected], or phone 770-488-8301 or 770-
488-8302.
Background
The PHCDM is one of many interrelated activities supporting CDC’s National
Electronic Disease Surveillance System (NEDSS) initiatives. The long-term
vision for NEDSS is a collection of complementary computerized information
systems that automate the process of gathering health data, facilitate the
monitoring of the health of communities, assist in the analysis of trends and
detection of emerging public health problems, and provide information for setting
public health policy.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 3 OF 91 JULY 2000
The focus of the NEDSS initiative is the development, testing, and
implementation of information management technology standards that will
support more complete and comprehensive integration of computerized health
information systems for use in public health. The NEDSS standards focus on five
important areas:
1. Data Architecture (data model, data definitions, and coding rules);
2. User Interface;
3. Information Systems Architecture (based on industry standards);
4. Tools for interpretation, analysis, and dissemination of data;
5. Secure data transfer.
The PHCDM is a major component of the NEDSS data architecture standards.
Together with Common Information for Public Health Electronic Reporting
(CIPHER) guidelines, the PHCDM provides a foundation for standardization of
public health data collection, management, transmission, analysis, and
dissemination.
The development of the PHCDM began in May 1999. The first step was the
construction of a high-level data model depicting the major subject areas to be
included in the PHCDM. The subject area data model was developed by
conducting an analysis of selected CDC disease surveillance systems, the Health
Level Seven (HL7) Reference Information Model (RIM), and other health-related
data models. Brainstorming sessions were held with CDC staff working on the
integration of public health surveillance systems to identify additional subject
areas that were overlooked in the analysis of existing data models. The subject
area model was used to define the project scope, estimate the work effort, and
develop the project plan.
In June 1999 an initial “class” diagram was created. A class is something about
which data are collected. The class diagram is a depiction of the major classes of
data within each subject area. It includes a description of the classes, as well as a
description of their inter-relationships. This class diagram was reviewed with
groups of CDC epidemiologists in July 1999 and revised based upon their
feedback. Attributes (i.e., information about the classes) were added to the class
model in August and reviewed within CDC and with state and local agencies in
September and early October 1999. Based upon feedback from these review
sessions the decision was reached to continue the enhancement of the data model
by:
• Developing a public health process model to provide context and clarify scope
for the data model;
• Adopting the HL7 Reference Information Model (RIM) representation of
health-related activities;
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 4 OF 91 JULY 2000
• Validating the data model by using it to develop a prototype database based
upon the information needed for a subset of CDC disease management and
surveillance systems such as the National Electronic Telecommunications
System for Surveillance (NETSS), the Sexually Transmitted Diseases
Management Information System (STD*MIS), and the Laboratory and
Epidemiological Public Health Information Tracking and Reporting System
(LITS+).
From October to December the model was revised to include data constructs from
the HL7 RIM and the Missouri Department of Public Health data model.
Meetings were held with the NETSS project team to discuss the objectives of
PHCDM and its potential application to the NETSS project. An approach to
developing the process model was devised in December.
The need for explicit description and publication of the rationale and objectives for
the model was acknowledged in January 2000. The PHCDM is critical to meeting
the data standardization objectives of the NEDSS initiative. The various model-
related activities have multiple objectives but their overall emphasis is on applying
the data model to data standardization issues facing CDC and its partners.
The model is expected to undergo continual refinement as it is used. It is a living
document that will need to be revised as public health information needs change,
as our understanding of those needs are improved, and as available technologies
increase the applicability of automation.
Goals and Objectives
The overarching purpose of the Public Health Conceptual Data Model is to
document the information needs of public health and facilitate the development of
data standards as part of the National Electronic Disease Surveillance System
initiative. One might ask, how does a data model facilitate the development of
data standards? That question is best answered by examining the goals and
objectives for the development of the PHCDM.
PHCDM Goals
1. Provide a framework for organizing data standards and guidelines
The initial CIPHER effort defined standards and guidelines for data
representation and code values. It included specifications for representing
concepts, such as dates, addresses, and person names as well as standard code
lists for coded elements, such as race, ethnicity, and sex. The CIPHER
standards can be linked directly to attributes in the data model. These
attributes represent characteristics of particular classes of information and will
eventually become fields in computerized information systems. The PHCDM
provides a context for these standards. By describing this context for the
CIPHER guidelines, CDC staff and public health partners working on data
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 5 OF 91 JULY 2000
standards are better able to envision the potential impact of their guidelines
and the implications of that impact. Also, by examining portions of the
PHCDM that do not map directly to CIPHER data standards, persons
continuing CDC’s work on data standards can readily determine additional
areas for which data standards and guidelines are required.
2. Reduce development effort for computerized information systems used
for public health
CDC and its state and local partners develop computerized information
systems for use in public health. Invariably these development projects
expend effort on requirements gathering, data requirement analysis, and
database design. The PHCDM will significantly reduce the effort expended
by these development projects by providing reusable data analysis and
database design and developing a common starting platform that can be used
or modified as necessary, resulting in reduced development time and cost.
By using the PHCDM at the beginning of analysis and design each individual
development project team can avoid rethinking data analysis and database
design issues. New findings or required revisions noted by a given
development team can be reflected in the PHCDM as part of its routine
maintenance so that they become available to other project teams for reuse.
3. Enhance data sharing through consistency
An additional advantage gained by using the PHCDM on multiple
development efforts is the reuse of PHCDM constructs in database design.
This reuse increases the consistency in data meaning and representation across
independently developed software systems. This increase in data consistency
will make it easier to share data, where appropriate, across the suite of
information systems used in public health. Use of the PHCDM will minimize
the need for complex data mapping and transformation processes prior to
sharing or reusing data.
Data consistency will permit data comparisons and linkages to be established
across multiple systems. Data consistency is an important system
characteristic that will facilitate the analysis of trends and detection of
emerging public health problems, and the use of information for setting public
health policy.
4. Represent public health data needs to national standards setting bodies
A critical aspect of this project is the ability to collaborate with national health
informatics standards setting bodies to define standards for the exchange of
information among public health agencies and healthcare providers. Health
Level Seven (HL7) is an essential part of the effort to establish standards for
information exchange in healthcare. The HL7 message development
methodology includes processes for constructing a reference information
model, defining datatypes, coordination of vocabulary domains, and
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 6 OF 91 JULY 2000
specification of Extensible Markup Language (XML) data interchange
standards.
HL7 is the undisputed leader in the establishment of standards for
interoperability among computerized information systems in healthcare. A
key aspect of the HL7 methodology is the HL7 Reference Information Model
(HL7 RIM). The HL7 RIM is the source for the data content of all HL7
version 3 standards. HL7 has defined an extensive process for ensuring that
the information needs of its constituencies are reflected in the RIM. It is
extremely important that the public health information needs be reflected in
the HL7 RIM.
The PHCDM will reuse as much of the RIM as is applicable to the needs of
public health and it will be the source of additions or modification to the RIM,
to represent unique public health data requirements. By including public
health needs in the HL7 RIM, we ensure that those needs are available to the
large body of information system vendors and provider organizations
participating in HL7, so that they can include them in the design of healthcare
information systems that serve as original sources of data relevant to public
health. CDC staff will also explore ways to coordinate public health
information needs with other accredited standards organizations.
5. Facilitate collaboration between CDC and its state and local partners in
public health
Collecting, analyzing, and reporting data related to public health is done at
local, state, and national levels. These data are used to monitor the health of
the public, identify public health problems and priorities, take immediate
public health action to prevent further illness, plan appropriate longer-term
interventions, and develop public health policy. Data on diseases and
conditions of importance to public health are reported on a regular basis to
Local Health Departments, which pass these data onto the states, which in turn
report voluntarily to CDC. CDC aggregates and reports these data back to
state and local authorities and to the public on a regular basis. The system of
surveillance, intervention, and planning requires collaboration between all of
the parties involved. Some states and localities rely upon and use CDC-
supplied software applications to electronically collect, report, and analyze
cases of notifiable conditions. Other states have the resources to develop such
applications on their own. Inconsistencies in data definitions, formats, and
code sets make the integration of data from the various sources and systems
difficult. The PHCDM will serve as a vehicle for collecting and reconciling
the information needs for public health at all levels. The data standards
defined by the PHCDM and CIPHER can be used by all parties involved in
public health. Through collaboration with state and local entities, the PHCDM
will lay a foundation for a unified view across the full spectrum of public
health activities.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 7 OF 91 JULY 2000
PHCDM Project Objectives
The goals of the PHCDM provide a basis of planning and evaluating the
assignment of resources for the PHCDM project. The goals of the PHCDM are
long term and will be achieved over the life of the PHCDM. The objectives of the
PHCDM project are more near-term and action-oriented. The PHCDM is to be
used for several years. The goals are expected to endure while projects and project
objectives are established on an annual or semi-annual basis.
The following objectives were achieved in 1999:
• Created a data model for public health based upon analysis of the database
structures of current CDC disease surveillance systems, CIPHER data
standards, the HL7 RIM, and other health-related information models, and
defined a maintenance process for the model;
• Reviewed the PHCDM with epidemiologists from within certain program
areas of CDC and from state and local public health agencies and updated the
PHCDM based upon their critique;
• Selected a computerized information system development project to make use
of the PHCDM and CIPHER data standards within CDC and develop a
workplan for applying the standards to the development effort.
Immediate next steps in 2000 are:
• Document and distribute copies of the PHCDM;
• Validate the PHCDM by using it to create a prototype database based upon the
information requirements of a selected set of CDC surveillance information
systems;
• Participate in the HL7 RIM harmonization process to introduce public health-
related information needs and to more closely align the PHCDM with the HL7
RIM;
• Participate in HL7 to use the HL7 RIM in the development of data exchange
messages specific to public health;
• Develop a high-level process model for public health to provide context for
the PHCDM;
• Continue to coordinate with state and local public health entities to ensure that
their information needs are represented in the PHCDM.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 8 OF 91 JULY 2000
Project Scope
The ultimate intention in developing the PHCDM is to represent the information
needs of all public health activities and entities. The immediate scope is to
document the information needs of public health surveillance including case
identification, reporting, investigation, intervention, and follow-up. For the
purpose of this project, public health surveillance will be viewed from the
perspective of CDC as a whole, not from any particular disease or
condition/program. To the extent possible the scope will include the perspectives
of State and Local Health Departments and other partners in public health
surveillance.
In addition to the data model, the scope of PHCDM project includes development
of a high-level process model for public health, harmonizing the model with the
HL7 RIM, and development of a prototype database based upon the PHCDM. In
addition the PHCDM project team will provide assistance to CDC system
development teams and state and local public health entities using the PHCDM in
their information management initiatives. These activities are some of many
anticipated sources of enhancements to the PHCDM.
Ongoing development of the PHCDM will follow an iterative approach. The
model will be continuously maintained and published on an annual basis. Each
year projects will be defined that specify the scope, process, and products to be
produced. Future project scopes might focus on areas such as public health
intervention programs, public health financing, or public health research. The
public health high-level process model will be used to help scope future projects.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 9 OF 91 JULY 2000
GUIDE TO UNDERSTANDING
THE PUBLIC HEALTH
CONCEPTUAL DATA MODEL
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 10 OF 91 JULY 2000
Guide to Understanding the Public Health Conceptual Data Model
The purpose of this section is to provide guidance to the reader to aid in
understanding the PHCDM. As a conceptual data model the PHCDM attempts to
present the information needs of Public Health in a way that lends itself to
validation by subject matter experts and has sufficient rigor and formality to be
used by experts in information technology in the development of database design
specifications.
To meet this objective the PHCDM avoids many of the details generally found in
logical and physical data models such as normalized data structures, primary and
foreign keys, and specification of field details such as length and decimal
positions. It makes extensive use of examples and explanatory text to describe
model classes and attributes. Its primary goal is to ensure that the concepts of
importance to public health are adequately depicted and documented.
The PHCDM uses a fairly high level of abstraction to document public health
concepts. This high level of abstraction extends the applicability of the model and
minimizes the need for maintenance. However, it can sometimes make it difficult
for subject matter experts to recognize specific details they might expect to find in
a public health data model. For example, “Where are items of interest to public
health, such as risk behaviors, infectious or environmental agents, drug-resistance,
case investigation, or populations-at-risk?”, might be questions a subject matter
expert might have after first browsing the data model. Rest assured that these
concepts are indeed included in the PHCDM. This guide to understanding the
PHCDM is intended to assist you, the reader, in finding answers to these questions
and others you may have as you review the model.
The first step in understanding the PHCDM is to become familiar with its
components, data model terminology, and the standards and conventions used.
The PHCDM uses the Unified Modeling Language (UML) modeling conventions.
UML is a widely used data modeling standard maintained by the Object
Management Group. References to information sources about UML can be found
in the bibliography included at the end of this document. The following
components of UML are used in this model:
• Subject Areas
• Classes and Relationships
• Attributes and Datatypes
Subject Areas
A subject area is a useful partitioning of a model into a cohesive collection of
classes. Subject areas are a way to subset a model into chunks that permit the
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 11 OF 91 JULY 2000
model to be more readily digested. There are four subject areas in the PHCDM:
Health-related Activities, Locations, Materials, and Parties.
The Health-related Activities subject area contains information about services,
conditions, and actions of interest to public health. A health-related activity might
be an observation, an intervention, a referral, or a notification. Typical examples of
health-related activities include observation of patient signs and symptoms,
clinical diagnoses, surgical operations, laboratory tests and results, as well as
public health notifications, case or contact investigations, population-oriented
health education campaigns, food item recalls. Cases, case reports (notifications),
and outbreaks are classes or health-related activities of particular importance to
public health surveillance.
The Locations subject area contains information about addresses associated with
Parties, Health-related Activities, or Materials. Location information may be a
postal location, a telecommunication location, or a physical location. Typical
examples of locations include street addresses, post office boxes, telephone
numbers, e-mail or web-site addresses, geographic coordinates, and spatial
references such as three miles east of town on interstate 95.
The Materials subject area contains information about substances, equipment,
products such as food and medication, physical entities, and other tangible items
of interest to public health that are associated with health-related activities or
Parties. Typical examples of materials include food items, pesticides, blood
samples, specimens, medications, durable medical equipment, prosthetic devices,
and medical supplies. Physical entities of interest to public health may include
such items as a lake, pool, ship, or airplane that are potential sources of exposure
to health hazards.
The Parties subject area contains information about the participants of health-
related activities. A party may be an individual person or non-person living
organism, or a formal or informal organization. Typical examples of parties
include patients, physicians, public health nurses, epidemiologists, hospitals, and
laboratories, as well as organizations such as the Association of State and
Territorial Health Officials, the Council of State and Territorial Epidemiologists.
Groups of parties with common characteristics, such as smokers or children under
5 years of age, are also included.
Classes and Relationships
There are 29 classes in the PHCDM. A Class is anything about which information
can be collected. Classes can be persons, places, things, concepts, or events.
There are four core classes in the PHCDM. The four core classes correspond with
the four subject areas. The four core classes are Health-related Activity, Location,
Material, and Party. Classes are depicted in the data model diagram by a
rectangular box with a line dividing the box into two vertical sections. The name
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 12 OF 91 JULY 2000
of the class appears in the top section of the box. The following diagram
illustrates the four core classes of the PHCDM.
The 29 classes of information in the PHCDM are all interrelated. Direct
relationships between classes are depicted in the model diagram by lines
connecting the related classes. The UML modeling language defines many ways
in which classes can be related. The PHCDM uses three methods of relating
classes: Supertype/Subtype Relationship, Relationship Association, and
Participation Association.
Supertype/Subtype Relationship
The supertype/subtype relationship is used when generic concepts represented by
a class are further represented in one or more specialized classes depicting a subset
of the generalized concept. In the supertype/subtype relationship the more generic
class, referred to as the supertype, has one or more specialized subtype classes.
Each of the four PHCDM core classes is a generic supertype class with one or
more related subtype classes. The supertype/subtype relationship is depicted on
the data model diagram by a line drawn between the subtype and the supertype.
The line has an arrowhead on one end pointing to the supertype. The following
diagram depicts the supertype/subtype relationships to the PHCDM core classes.
Health-Related Activity Location PartyMaterial
Health-Related Activity Location PartyMaterial
Observation
Intervention
Referral
Notification
Postal Location
Physical Location
Telecommunication
Location
Specimen Individual
Organization
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 13 OF 91 JULY 2000
This hierarchical structuring of core model concepts makes it easier to understand
the model. Once you become familiar with the four core concepts the process of
becoming familiar with the specialized concepts in the class hierarchy is very
simple. For example, looking at the Health-related Activity hierarchy the classes
Observation, Intervention, Referral, and Notification are each special types of
Health-related Activity. An Observation is a Health-related Activity. All of the
data we capture about a Health-related Activity are data that are also collected
about an Observation. However, the Observation Class has additional attributes
that are captured only when the Health-related Activity is of type Observation.
Similarly, an Intervention is a Health-related Activity. It too inherits all of the
characteristics of Health-related Activity and supplements that information with
information unique to an intervention.
A subtype of one class may also be the supertype of one or more of its own
subtypes. This is illustrated in the following diagram depicting the full hierarchy
of the concepts modeled under Party.
The Individual and Organization classes are subtypes of the supertype class Party.
Individual is also the supertype for the Person and Non-Person Living Organism
subtypes, and Organization is the supertype for the Formal Organization and
Informal Organization subtypes.
Relationship Association
A relationship association is a special type of relationship used in the PHCDM to
reflect the relationship an instance of a core class or its subtypes has to another
Party
Individual Organization
Non-Person
Living Organism
Person
Informal Organization
Formal Organization
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 14 OF 91 JULY 2000
instance of the same core class or its subtype. These relationships are represented
in the PHCDM by four “relationship” classes, each associated with one of the four
core classes. The four relationship classes are: Activity Relationship, Location
Relationship, Material Relationship, and Party Relationship. The relationship
associations are depicted in the model diagrams by a rectangular box representing
the relationship class and a pair of association lines connecting the relationship
class to the core class that is linked by the relationship. The Activity Relationship
is illustrated in the following diagram.
The symbols “1” and “0..*” that appear on the association lines depict the
multiplicity of the association between the relationship class and the core class.
Multiplicity is an indication of the number of instances of a class that is capable of
being involved in any one association. In this case, the multiplicities indicate that
each instance of the activity relationship is associated with one and only one
health-related activity and that each health-related activity is associated with zero
or more activity relationships. Since there are two associations, each with the
same multiplicity, an instance of an activity relationship class is always associated
with two instances of a health-related activity class. A single health-related
activity may be associated with zero or more activity relationships relating it to
another health-related activity.
For example, an observation reflecting the presence of a vectorborne disease (with
mosquitoes as the vector) can be linked to interventions such as spraying
insecticide, using mosquito repellents, and issuing mosquito nets. Both
observation and intervention are subtypes of health-related activity.
Participation Association
The participation association is a special relationship used in the PHCDM to
depict the relationships that exist between the core classes. Each core class has a
many-to-many relationship to all of the other core classes. For example, an
instance of the party class may be related to many instances of the material,
location, and health-related activity classes. Instances of material, location, and
health-related activities classes may also be associated with many instances of the
party class. The participation association is depicted using a participation class.
There are five participation classes in the PHCDM: Actor Participation, Target
Participation, Party Location Participation, Material Responsibility, and Material
Location Participation. The following diagram depicts the Material Responsibility
participation association:
1
1
0..*
0..*
Health-Related Activity
Activity Relationship
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 15 OF 91 JULY 2000
The multiplicities on the associations indicate that a material responsibility is
always associated with one party class and one material class. Material and party
classes may be associated with zero or more material responsibility classes.
For example, if material is a specimen it may be important to reflect the party that
was the source for the specimen and the party that was responsible for obtaining
the specimen. This would be captured as two instances of the material
responsibility class, one for each party class, each associated with the same
specimen material class.
Attributes and Datatypes
Attributes are the specific items of data that can be collected for a class in the
PHCDM. Each attribute has a name, a description, and a datatype assignment.
An attribute name suggests the meaning of the attribute, while the description
defines it, provides examples, and includes relevant discussion. The datatype
assigned to an attribute extends the definition of the attribute. A datatype is a
specification of the allowed format for the values of an attribute.
Attributes and their datatype assignments are shown in the data model diagram by
listing them in the lower section of the rectangle representing the class. The
following diagram is an example of three classes and their attributes.
Party Material Responsibility Material
1 0..*
1
0..*
1
1
0..*
0..*
Relationship Date Time Range : IVL <TS>
Relationship Type Code : CV
Party Relationship
Birth Date : TS
Death Date : TS
Sex Code : CV
Individual
Party Identifier : Set<II>
Party
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 16 OF 91 JULY 2000
Attributes of a supertype are inherited by its subtypes. In this example, the
attribute Party Identifier in the supertype Party class is also an attribute of the
subtype class Individual (as well as all the other subtypes of Party).
Attribute names follow the form:
[Class Name] [{Qualifier Name}] Attribute-Type Name
The square brackets […] around class name indicates that the class name may be
omitted from the attribute name. The curly brackets {…} around qualifier name
combined with the square brackets indicate that there may be zero, one or more
qualifier names. Every attribute has an attribute-type name. The attribute-type
name provides an indication of the type of data the attribute conveys. The
attribute-type names used in the PHCDM are:
•
Amount
•
Code
•
Date
•
Description
•
Identifier
•
Name
•
Number
•
Quantity
•
Text
•
Time-Range
•
Value
The datatype assigned to an attribute is represented in the data model diagram by
the inclusion of the datatype name following the attribute name, separated by a
colon (“:”). The datatype assignments in the PHCDM appear in one of three
forms:
• Attribute Name : Datatype Name
• Attribute Name : SET<Datatype Name>
• Attribute Name : IVL<Datatype Name>
The collection of datatypes used in the PHCDM is drawn from the set of datatypes
defined for HL7. A complete list of the datatypes and their descriptions are
included in an appendix. The datatype name used in the model diagram is the
short name of the datatype.
Most attributes are expected to take on one value at a time. However, if the
datatype name is preceded by “SET” and enclosed in brackets, it is an indication
that the attribute may repeat. That is, there may be a set of one or more values for
the attribute. In the sample model the attribute Party Identifier is a set. That is an
indication that there may be multiple identifiers for a single instance of a Party
class.
If the datatype name is preceded by “IVL” it is an indication that the attribute
represents an interval of values from low to high. In the PHCDM the IVL prefix
to a datatype has been limited to intervals of time. In the sample diagram above
the attribute Relationship Date Time Range (for the class Party Relationship) is an
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 17 OF 91 JULY 2000
interval. The implication is that the relationship date time represents a start and an
end date time range.
Key Concepts
To fully appreciate the richness of the PHCDM it is necessary to understand a few
key concepts. Although the model contains 4 subject areas, 29 classes and nearly
100 attributes it is still fairly easy to digest. The key concepts included in this
section are also described in the model itself. They are described here because of
their importance to understanding the data model. The key concepts are:
Health-related Activity Mood Code
The activity mood code is an attribute of the health-related activity class. The
attribute is critical to determining the perspective of an instance of the health-
related activity class. The activity mood code captures the meaning or context of
the activity. Possible values of the activity mood code are:
• Fact
an activity that has occurred
• Command
an activity that has been ordered
• Master
a table entry of possible activities
• Definition
an algorithm for describing an activity
• Intent
a goal for an activity
Instances of the health-related activity class with an activity mood code of
“master” might be used to represent the list of health conditions under public
health surveillance. Other instances of the health-related activity class with an
activity mood code of “definition” might be used to represent the case definitions
established for the list of health conditions under public health surveillance. And
finally, instances of the health-related activity class with an activity mood code of
“fact” would be used to capture actual cases or occurrences of health conditions.
The activity mood code makes the health-related activity a very versatile concept
and minimizes the number of classes, attribute, and relationships need to represent
a wide range of concepts.
Actor / Target Participation
Actor participation is a class that captures the relationship between a party and a
health-related activity. Target participation is a class that captures the relationship
between a health-related activity and party, material, or location. The actor
participation is meant to capture information about parties that carry out the action
indicated in health-related activity. The activity type code indicates the nature of
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 18 OF 91 JULY 2000
the party’s participation. The type code identifies the various types of actors or
roles a party may assume. Examples of actor roles might include disease
intervention specialist, outbreak investigator, primary care physician, attending
physician. This structure permits the party class to assume many roles without
having to introduce specialized classes and attributes to the model. Target type
code in the target participation class performs a similar function for parties,
materials, and locations. The target is not an actor in the health-related activity but
is otherwise involved, either as the object of the activity (e.g., the patient) or
involved in a more passive mode (e.g., the medication in a medication
intervention). Targets may include persons with disease of interest or persons
exposed to disease.
Observation
The observation class represents both the act of observing and the results of the
observation. It represents objective, subjective, and derived observations. It
includes the concepts of test, assessment, vehicle condition, diagnosis, party
condition, and health status inquiry. Observations may be made about parties,
materials, or locations. Observations may be made about other health-related
activities, including other observations. This is a very rich concept. Essentially all
facts not included as attributes elsewhere could conceivably be represented by the
observation class. Because an observation may take many forms (e.g., numbers,
waveforms) depending on the target of the observation and the type of
phenomenon being observed, it has a datatype of “any”.
Individual Non-Person Living Organism
This subtype of the Party class is intended to be limited to those living organisms
that are identified individually. The class is necessary to permit the tracking of
health-related activities where the target of the activity is not a person. For
example, the target might be an animal such as a pet dog, or a circus elephant. It is
unlikely that a life form like a bacterium or a virus would ever be individually
identified and so it is not likely for them to be included as non-person living
organisms. Even insects, like mosquitoes, are unlikely to be included as non-
person living organisms. However, if a particular mosquito is isolated and
individually identified it might be valid to include it as a non-person living
organism. In most cases, however, non-person living organisms that are not
routinely individually identified would be represented either as specimens (a
subtype of material) or as members of an informal organization.
Informal Organization
An informal organization is intended to represent any defined population. The
members of the population do not have to be enumerated. The population is
treated as a unit with respect to its participation as a actor or target in a health-
related activity. Informal organizations could include a human family, a herd of
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 19 OF 91 JULY 2000
cows, people that smoke, dead crows in New England, or any other defined
population.
Implementing the PHCDM
Understanding the PHCDM requires a discussion of how it is to be implemented.
This section includes:
• a brief description of the hierarchy of data model types, identifying where a
conceptual data model fits within a family of interrelated data models;
• a discussion of the process of deriving multiple physical database models from
a common conceptual data model;
• a description of a harmonization process designed to facilitate ongoing
enhancements to the PHCDM.
Hierarchy of Data Model Types
A data model is documentation of data from a particular domain, for a specific
purpose, using a formal specification. A data model has both a graphical
expression and a supporting textual expression or data dictionary. Each
expression of a data model uses a pre-defined formalism of symbols, semantics,
and rules of construction.
The purpose of a data model is to aid in understanding data in a particular domain.
A data model communicates the modeler’s understanding of data and allows that
understanding to be assessed by others. A data model can be useful in reconciling
multiple perspectives of data because it reveals the underlying assumptions,
semantics, and constraints expressed in multiple models and requires their
harmonization into a single specification. A common use of a data model is to
document a database design (existing or planned) so that the design may be
evaluated.
There are multiple types of data models. Each type of data model has
characteristics that make it more useful than other types for a particular purpose.
There are models that are very useful for high-level planning and project
definition. These models tend to minimize technical details and focus instead on
delineating and defining subject areas and classes of information of interest to
executives, high-level decision-makers, and subject matter experts. These models
are not useful for evaluating or implementing a database design. For a data model
to be useful to a database design activity, it needs to include technical details such
as database key structures, datatypes, and the physical properties of tables and
columns. However, this latter type of model contains too many detail and
technical artifacts to make it useful for high-level planning and decision making.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 20 OF 91 JULY 2000
A data model should be constructed with a specific purpose in mind. The model
developer should choose from among the various data model types the type that is
most suitable for the intended audience and use. The following diagram identifies
a hierarchy of six data model types:
Each type of model is described below.
1. Subject Area Model (SAM):
A subject area model contains only subject areas and their connections, and
usually serves as a model for a large domain, such as the entire enterprise or a
major functional area. It is used for high-level planning and setting of project
scope.
2. Class Relationship Model (CRM):
A class relationship model contains only subject areas, classes, and
relationships, and generally depicts a limited domain, such as a single project
or enterprise business area. It is used for high-level analysis and estimation of
project size.
3. Conceptual Data Model (CDM):
A conceptual data model contains subject areas, classes, attributes, datatypes,
and relationships, and generally models a project-specific domain, such as
public health, finance, or material management. It results from a relatively
detailed level of analysis and is often a primary project deliverable.
SAM
CRM CDM
LDM DDM
PDM
Hierarchy of Data Model Types
•
Subject Area
Model
•
Class Relationship
Model
•
Conceptual Data
Model
• Logical Data
Model
•
Database Design
Model
•
Physical Database
Model
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 21 OF 91 JULY 2000
4. Logical Data Model (LDM):
A logical data model contains subject areas, normalized classes, atomic
attributes, relationships, and candidate/primary keys, and usually serves as a
model for an enterprise-specific implementation of a project-specific domain.
It signifies the completion of the most detailed level of data analysis and the
beginning of database design.
5. Database Design Model (DDM):
A database design model contains table spaces, tables, columns, datatypes, and
primary/foreign keys, and generally represents an existing or planned database
of a computerized information system. It indicates the completion of database
design and the beginning of database construction.
6. Physical Database Model (PDM):
A physical database model contains the data definition language (DDL)
required to create tables and indexes, as well as data base management system
(DBMS)-enforced constraints. It is a machine-processable specification of an
existing or planned database of a computerized information system, and
corresponds to the final step of database design and construction.
In the diagram of the hierarchical taxonomy of models below, the six model types
are arranged from top to bottom by level of detail and target audience, and from
left to right by degree of precision and rigor of specification.
The top three models (subject area, class relationship, and conceptual) are
technology-independent and may be applicable to multiple organizations
performing the same functions that are supported by the data model. The lower
SAM
CRM
CDM
LDM
DDM
PDM
Detail
Detail
Precision
Precision
Executive / Decision Maker
Decision Maker / Subject Matter
Expert (SME)
SME / Information Technology
Expert
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 22 OF 91 JULY 2000
three models are more technology-specific and may be applicable to multiple
organizations that share the same business rules.
The PHCDM is a particular instance of a conceptual data model. A conceptual
model was chosen as the style for the PHCDM because of the desire to have a
model that is technology-independent and applicable to multiple organizations. A
conceptual data model also has sufficient detail to be useful as a definition of
information requirements in a specific domain (i.e., public health). The intended
audiences for this model are decision-makers and subject matter experts in public
health and information technology experts responsible for requirements analysis
and design of computerized information systems for use in public health.
Multiple Physical Data Models from one Conceptual Data Model
The PHCDM should be considered the base model for multiple physical data
models. The process of building a physical database model is expected to use the
PHCDM as input to creation of a logical data model (LDM). The LDM may be
derived from the entire PHCDM, or simply from a subset of it. Its design is
constrained by the business rules and project scope of the entity implementing the
model. These constraints will differ from implementation to implementation,
resulting in multiple LDMs that are semantically equivalent but that may vary
from each other on a technical level (i.e., choice of class identifiers, degree of
normalization, and relationship constraints). The database design models (DDM)
and the physical database models (PDM) will be derived from the LDMs as
depicted in the following diagram:
Multiple PDMs from one CDM
PDM
DDM
LDM
PDM
DDM
LDM
PDM
DDM
LDM
PDM
DDM
LDM
PDM
DDM
LDM
CDM
CRM
SAM
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 23 OF 91 JULY 2000
Each LDM-DDM-PDM triad is a separate implementation of the same CDM. In
the case of the PHCDM, these might represent separate or collaborative
implementations by Local and State Public Health Departments or by CDC
program areas. The physical database models might target different technologies
and enforce different business rules, however because they originated from a
common CDM, the semantics of the data content would be equivalent on a
conceptual level. This will greatly facilitate the sharing of information between
these independently developed databases.
PHCDM Harmonization Process
The PHCDM will be enhanced in an ongoing fashion, based upon input from sites
that use it in their information system development initiatives. As organizations or
programs implement the PHCDM, they will invariably identify omissions and
perhaps errors in the model. As errors and omissions are identified, they should be
brought to the attention of CDC in the form of proposed changes to the PHCDM.
PHCDM change proposals submitted from the multiple sites implementing the
model will be considered together and harmonized to ensure that conflicting
change requests are reconciled prior to being applied to the model. This
harmonization process is illustrated in the following diagram.
The current plan for the PHCDM is to re-issue the model on an annual basis,
incorporating as many of the change proposals as it is feasible to handle in one
year. The mapping of the PHCDM to CDC-developed information systems like
PDM
DDM
LDM
PDM
DDM
LDM
PDM
DDM
LDM
PDM
DDM
LDM
PDM
DDM
LDM
Public Health Conceptual Data Model
PHCDM Harmonization Process
PHCDM Harmonization Process
Site A Site B Site C Site D Site E
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 24 OF 91 JULY 2000
NETSS, STD*MIS, and LITS+ has already identified possible enhancements for
the 2001 version of the PHCDM. Additional input is expected from
harmonization of the PHCDM with the HL7 RIM and from use of the model in
additional activities related to the National Electronic Disease Surveillance System
initiative.
An attempt will be made to minimize the impact that enhancements to the
PHCDM have on the sites where the model has already been implemented.
Changes from release to release will be highlighted in each publication. An
assessment of the impact of the change and suggestions for forward migration will
be included in each release.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 25 OF 91 JULY 2000
PUBLIC HEALTH
CONCEPTUAL DATA MODEL
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 28 OF 91 JULY 2000
The model consists of 4 subject areas and contains 29 classes. These subject areas
are based on the most general categorization of the data relevant to public health
concerns. The four subject areas and the classes they contain are listed below.
HEALTH-RELATED ACTIVITIES
ACTIVITY RELATIONSHIP
ACTOR PARTICIPATION
CASE
HEALTH-RELATED ACTIVITY
INTERVENTION
NOTIFICATION
OBSERVATION
OUTBREAK
REFERRAL
TARGET PARTICIPATION
LOCATIONS
LOCATION
LOCATION RELATIONSHIP
MATERIAL LOCATION PARTICIPATION
PARTY LOCATION PARTICIPATION
PHYSICAL LOCATION
POSTAL LOCATION
TELECOMMUNICATION LOCATION
MATERIALS
MATERIAL
MATERIAL RELATIONSHIP
MATERIAL RESPONSIBILITY
SPECIMEN
PARTIES
FORMAL ORGANIZATION
INDIVIDUAL
INFORMAL ORGANIZATION
NON-PERSON LIVING ORGANISM
ORGANIZATION
PARTY
PARTY RELATIONSHIP
PERSON
Detailed descriptions of the classes and attributes are contained in the sections for
each subject area.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 29 OF 91 JULY 2000
Health-related Activities Subject Area
REFERRAL
Referral Reason Text : FTX
Referral Description : FTX
INTERVENTION
Intervention Reason Code
Intervention Form Code : CV
Intervention Route Code : CV
Intervention Quantity : PQ
Strength Quantity : PQ
Rate Quantity : PQ
OBSERVATION
Observation Value : ANY
Derivation Expression Text : ST
CASE
Confirmation Method Code : CV
Detection Method Code : CV
Transmission Mode Code : CV
Disease Imported Code : CV
Etiologic Status Code : CV
Classification Status Code : CV
OUTBREAK
Outbreak Jurisdictional Extent Code : CV
Outbreak Peak Date : TS
Outbreak Time Range : IVL<TS>
NOTIFICATION
Notification Reason Code : CV
MATERIAL
(from MATERIAL)
LOCATION
(from LOCATION)
ACTIVITY RELATIONSHIP
Activity Relationship Type Code : CV
Activity Relationship Date Time Range : IVL<TS>
HEALTH-RELATED ACTIVITY
Activity Identifier : SET<II>
Activity Mood Code : CV
Activity Type Code : CV
Activity Descriptive Text : FTX
Activity Status Code : CV
Activity Date Time : GTS
Activity Critical Date Time : GTS
Activity Method Code : CV
Subject Site Code : CV
Interpretation Code : CV
Confidentiality Code : CV
Maximum Repetition Number : INT
Priority Code : CV
0..*
1
0..*
1
0..*
1
0..*
1
ACTOR PARTICIPATION
Actor Type Code : CV
Actor Time Range : IVL<TS>
0..*
1
0..*
1
TARGET PARTICIPATION
Target Type Code : CV
Target Time Range : IVL<TS>
Target Awareness Code : CV
0..1
0..*
0..1
0..*
0..1
0..*
0..1
0..*
0..*
1
0..*
1
PARTY
(from PARTY)
0..*
1
0..*
1
0..*
0. .1
0..*
0..1
Figure 2. Health-related Activities Subject Area Diagram
The classes and attributes of the Health-related Activities subject area are
described below.
Class: ACTIVITY RELATIONSHIP
Associated with: HEALTH-RELATED ACTIVITY
Description of: ACTIVITY RELATIONSHIP
Activity relationship captures the relationship between a pair of health-related
activities. Generally, relationships between health-related activities fall into three
categories: an activity can be comprised of component activities; one activity can
cause another; one activity can be associated with another for any number of
reasons.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 30 OF 91 JULY 2000
Virtually any activity can be decomposed into its parts. In public health, an
outbreak of a particular disease can be composed of multiple individual cases of a
particular disease. To take a medical example, consider a surgical procedure, e.g.,
a laparoscopic cholecystectomy. This action consists of many smaller actions that
must occur in the right order and relation to each other. In the case of an invasive
surgery, preoperative preparation may be required as a precondition, while
anesthesia is conducted in parallel to the entire surgical procedure.
Causal associations are used to provide explanations for actions. For example, an
episode is defined as a case of a particular disease (event reportable to public
health) because of the results of a clinical evaluation combined with laboratory test
results. (Note that the definition of the case specifies these criteria.) Another
example is the instance of a test that was performed because of the results of two
earlier tests.
The notion of "associated with" is more general than “causal” and "comprised of"
associations. For example, in public health, a reportable case of disease is
commonly associated with multiple observations. These observations record such
items as specific behaviors that put the person at risk, the person's visits to
locations where they might have been exposed, or the test results that indicate the
person has a particular disease.
Associations for: ACTIVITY RELATIONSHIP
relates (1,1) :: HEALTH-RELATED ACTIVITY :: is_target_for (0,n)
Attributes of: ACTIVITY RELATIONSHIP
Activity Relationship Type Code : CV
The code that reflects the nature of the relationship that exists between two or
more associated health-related activities. The possible values include
“comprises”, “causes”, and “is associated with”. An example of a “comprises”
relationship is a case definition that is comprised of laboratory tests, symptoms,
and other qualifying criteria. An example of a “causes” relationship is a case
notification causes a case investigation. An example of an “is associated with”
relationship is an outbreak and the associated cases.
Activity Relationship Date Time Range : IVL<TS>
The period of time during which the relationship between the two activity
instances is effective.
Class: ACTOR PARTICIPATION
Associated with: HEALTH-RELATED ACTIVITY
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 31 OF 91 JULY 2000
PARTY
Description of: ACTOR PARTICIPATION
Actor participations include the active roles played by a party in the health-related
activity. Examples include an organization that provides physical therapy services,
a person who performs a surgical procedure, a public health worker who tracks
contacts of an infectious case, a person who conducts a test, a person who
conducts an interview.
Additional examples of actor participations are: a) the part played by an
epidemiologist or CDC program staff (party) in generating a public health case
definition; b) the part played by a provider, State or Local Health Department
(party) in the notification of a case.
Associations for: ACTOR PARTICIPATION
associates_to (1,1) :: HEALTH-RELATED ACTIVITY :: associates (0,n)
associates_to (1,1) :: PARTY :: associates (0,n)
Attributes of: ACTOR PARTICIPATION
Actor Time Range : IVL<TS>
The time range during which the associated party participated in the health-related
activity while taking on the role indicated by the specified actor type code value.
Actor Type Code : CV
Identifies the particular function or a set of functions that a party performs in the
health-related activity. Note that the actor type code designates the actual function
performed in a particular health-related activity in distinction to other roles or
occupation. Examples of actor type codes might include case investigator,
interviewer, and disease investigation specialist.
Class: CASE
Subtype of: OBSERVATION
Supertype of: OUTBREAK
Description of: CASE
A case is an observation that represents a condition or event that has a specific
significance for public health. The case can include a health-related event
concerning a single individual or it may refer to multiple health-related events that
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 32 OF 91 JULY 2000
are occurrences of the same disease or condition of interest to public health. An
outbreak involving multiple individuals is a type of case.
A case definition (a case whose mood code = "definition") includes the description
of the clinical, laboratory, and epidemiologic indicators associated with a disease
or condition of interest to public health. There are case definitions for conditions
that are reportable, as well as for those that are not. There are also case definitions
for outbreaks. A case definition is a construct used by public health for the purpose
of counting cases, and should not be used as clinical indications for treatment.
Examples include AIDS, toxic-shock syndrome, and salmonellosis and their
associated indicators that are used to define a case.
Attributes of: CASE
Classification Status Code : CV
Code for the classification status of the case. Possible values include confirmed,
probable, suspected, not a case, incomplete information. This status code differs
from the activity status code inherited from the health-related activity supertype to
case. The activity status code captures the lifecycle state of the case (active,
inactive, completed).
Confirmation Method Code : CV
Code for the mechanism by which the case was confirmed. This attribute is
intended to provide information about how the case classification status was
derived. Includes laboratory criteria met, clinical case inclusion criteria (alone)
met, epidemiologist- or other public health worker-assigned, epidemiologically
linked via investigation, and physician-reported.
Detection Method Code : CV
Code for the method by which the case was identified. Possible values include
provider report, patient self-referral, laboratory report, case or outbreak
investigation, contact investigation, active surveillance, routine physical, prenatal
testing, prenatal testing, prison entry screening, occupational disease surveillance,
and medical record review.
Disease Imported Code : CV
Code that indicates whether the disease was likely acquired outside the jurisdiction
of observation, and if so, the nature of the interjurisdictional relationship. Possible
values include not imported, imported from another country, imported from
another state, imported from another jurisdiction, and insufficient information to
determine. Note that if the specific jurisdiction is to be captured it is captured as a
target participation associated with a jurisdictional party.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 33 OF 91 JULY 2000
Etiologic Status Code : CV
Code for the strength of the causal relationship between the disease-causing agent
and the disease. This is particularly relevant for outbreaks where the cause is not
yet certain, or emerging/new diseases or conditions where the cause is not clear.
For example, in the case of an outbreak of gastroenteritis, blood in the stool may
indicate that the agent was most likely a Shiga toxin-producing E. coli (strong
suspicion), although other infectious or toxic agents may still be included in the
differential diagnosis, but to a lesser degree (weak or moderate suspicion).
Includes weak suspicion, moderate suspicion, confirmed, and unknown.
Transmission Mode Code : CV
Code for the mechanism by which disease was acquired by the party involved in
the case. Includes sexually transmitted, airborne, bloodborne, vectorborne,
foodborne, zoonotic, nosocomial, mechanical, dermal, indeterminate.
Class: HEALTH-RELATED ACTIVITY
Supertype of: INTERVENTION
NOTIFICATION
OBSERVATION
REFERRAL
Associated with: ACTIVITY RELATIONSHIP
ACTOR PARTICIPATION
TARGET PARTICIPATION
Description of: HEALTH-RELATED ACTIVITY
A health-related activity is an action performed for the purpose of documenting,
investigating, or improving the health condition of a party. It may also include
documenting the ability to affect the health status of a party. Examples of health-
related activities include all of the following:
• interventions such as surgical operations or vaccination;
• administration of a medication;
• referral to another provider;
• diagnostic observations about a patient's condition;
• diagnostic assessment that a condition meets the public health definition of a
case;
• a public health notification of a case of a reportable disease or condition;
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 34 OF 91 JULY 2000
• public health investigation of all persons exposed to a common source of
infection or toxin;
• food or consumer product recalls;
• an intervention targeted at a given population.
An instance of a health-related activity can be captured from several perspectives.
Possible perspectives for an instance of a health-related activity are:
• a fact about an activity that has occurred, such as the observation of
chickenpox in a child;
• a command, such as an order to vaccinate a child for chickenpox;
• a master table entry of possible activities, such as types of laboratory tests;
• a definition algorithmically describing an activity, such as a case definition for
chickenpox;
• an intent for an outcome of an activity, such as achievement of a 95%
immunization rate in children under age 2.
Associations for: HEALTH-RELATED ACTIVITY
is_source_for (0,n) :: ACTIVITY RELATIONSHIP :: relates (1,1)
is_target_for (0,n) :: ACTIVITY RELATIONSHIP :: relates (1,1)
associates (0,n) :: ACTOR PARTICIPATION :: associates_to (1,1)
associates (0,n) :: TARGET PARTICIPATION :: associates_to (1,1)
Attributes of: HEALTH-RELATED ACTIVITY
Activity Critical Date Time : GTS
The "biologically relevant" time for a health-related activity. The concept is best
understood with observations, where the time of the observation activity may
differ from the time of the observed feature. For instance, in history taking, when
the doctor records an episode of Hepatitis A suffered by the patient last year for
several weeks. The activity critical date time is the date/time when the patient
experienced the episode of Hepatitis A, and not the date and time when the doctor
records the history. That is to say, it is the time/dates that the patient actually had
hepatitis, and not when the patient tells the doctor, or when the doctor records it. In
another example, the provider may order a test, conducted on a blood sample
drawn today, for which results will not be available until next week. The activity
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 35 OF 91 JULY 2000
critical date time is the date and time of the taking of the specimen, not when the
results are available.
Activity Date Time : GTS
The time when the action happened, is ordered or scheduled to happen, or when it
can possibly happen. The time specification could be a point in time, a time range
during which the activity occurred, or is supposed to occur.
Activity Descriptive Text : FTX
The description of an activity is a piece of free text or multimedia data that
describes the activity in all necessary detail. This attribute is a descriptive
supplement to an activity type code, not a replacement. There is no restriction on
length or content imposed on the description attribute. However, the content of the
description is not considered part of the functional information communicated
between systems. Descriptions are meant to be shown specifically to interested
individuals.
Activity Identifier : SET<II>
This is an instance identifier for a health-related activity. It uniquely identifies a
particular instance of a health-related activity class.
Activity Method Code : CV
The activity method code is a parameter of the health-related activity that specifies
one of the possible methods used to achieve a given end. The method is specified
for a given health-related activity, because there are different methods to achieve
results, and knowing the method is important for a more explicit interpretation.
For example, when carrying out an assessment of a person's risk-taking behavior,
possible methods include: written questionnaire, personal interview, third-party
interview (for children), and medical record review. When carrying out
interventions for public health education, possible methods include: mass media,
billboard, individually targeted automatic messages, and individual counseling.
Activity Mood Code : CV
The activity mood code determines the meaning or context for the activity. The
activity (corresponding to a verb in natural language) may be conceived as an
event that happened (fact), an ordered service (command), a possible service
(master), an algorithm for describing an event (definition), and a goal of health-
related activity (intent). Each of these is a different mood.
The activity mood code is critical to the design of this model. Without it, the
model described here would be at least three times as big, in order to distinguish
between the following:
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 36 OF 91 JULY 2000
a) The definition of the health-related activity (e.g., a case or test definition);
b) Health-related activities that are planned;
c) Scheduled health-related activities;
d) Health-related activities that have already occurred or been performed.
Activity Status Code : CV
A code for the state of the action (e.g., intended, ordered, in process, completed).
This attribute is not used to describe the classification status of a case; the case
classification status code should be used. (See the case attribute: case
classification status code.)
Activity Type Code : CV
A code for the kind of activity (e.g., physical examination, person interview,
serum potassium, public health notification, product sterilization or
pasteurization). The activity type code specifies the service conceptually by using
a code from a coding system. The activity type code or "name" is a handle on the
concept of the action, not on the individual action instance. Different coding
systems cover different kinds of activities, which is why there is not one single
coding system to be used for the activity type code.
When observations are recorded for outbreaks, the activity type code captures
information to indicate the category of the statistic, e.g., number ill, number
exposed, number hospitalized, number treated, number of fatalities, number
interviewed, incubation period days/hours, duration of illness (days/hours),
number not ill, % female, % male, % less than 18 years of age, ages of affected,
and information to indicate the type of statistic, e.g., minimum, maximum,
percentage, median, count.
Confidentiality Code : CV
Indicates limitations to disclosure and communication of information about a
health-related activity. Includes provider access only, limited to county or state
public health department access, disease program access only, or public use/
publicly available.
Interpretation Code : CV
The interpretation code allows for a very rough interpretation of the course or
outcome of an activity. These are sometimes called "abnormal flags", however the
judgment of normalcy is just one of the common rough interpretations, and is
often not relevant. For example, for the observation of a pathologic condition, it
doesn't make sense to state the normalcy, since pathologic conditions are not
considered "normal." In other words, context is required to make a final
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 37 OF 91 JULY 2000
determination, and this code may simply provide a judgment that these data are
worth investigating further. For example, this code may be used to indicate that an
antibody level is slightly elevated, which may be consistent with disease.
However, the interpretation of disease may require additional data, such as a
repeated antibody titer, to determine whether the value is rising or falling. This
attribute is also used to describe antibiotic susceptibility results as “susceptible”,
“intermediate”, and “resistant”.
Maximum Repetition Number : INT
The maximum number of repetitions of a health-related activity. Typical values
are 1, some other finite number, and infinity. This is relevant when the health-
related activity is a plan or a series of orders.
Priority Code : CV
Code for the priority of the activity. Possible values include routine, emergency,
and urgent.
Subject Site Code : CV
Most health care services focus on a particular part of the target on which the
health-related activity is performed. Typically, when the target party is a person,
this will be a feature related to the anatomic structure of the patient (the "target" of
the service). In the case of material entities other categorizations are used. For
example, when a sample is ordered from a restaurant to explain a case of food
poisoning, sites such as floor, meat grinder, refrigerator, or cutting board could be
used.
Class: INTERVENTION
Subtype of: HEALTH-RELATED ACTIVITY
Description of: INTERVENTION
An intervention is the administration of a substance or technique to provide care
for or to prevent a condition. This includes vaccinations and preventive therapy as
well as medication given directly for therapeutic purposes. An intervention need
not be administered solely to individuals, and may include population
interventions such as chlorinating or fluoridating the water supply, policies to
restrict tobacco sales, pasteurization of milk, and pesticide application in a specific
geographic area. Includes therapeutic and preventive treatments, counseling,
educational campaigns, needle exchange programs, media campaigns, food
recalls.
Attributes of: INTERVENTION
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 38 OF 91 JULY 2000
Intervention Form Code: CV
The physical form in which the intervention is delivered. For medications,
examples include tablet, capsule, suppository, and solution. For environmental
interventions, such as chlorination of the water supply, examples might include
chlorine in liquid or tablets. For food recalls, examples might include complete
meat packages or individual burgers. For media campaigns, examples might
include television commercials, radio ads, billboards, or pamphlets.
Intervention Quantity : PQ
The amount of the intervention associated with a single intervention instance. For
example, this might refer to the amount of pesticide to be sprayed during a single
application or the amount of gas or chemical to be used in a sterilization of a
medical device.
In the case of medication, the amount is the dose or amount of the therapeutic or
prophylactic agent given at one administration event. This attribute can be used all
by itself, or in combination with a strength.
Intervention Reason Code : CV
Code which describes the basis for the intervention. Includes treatment,
prophylaxis, post-exposure prophylaxis, high-risk individual or population.
Intervention Route Code : CV
The route by which the intervention is administered to the object of the
intervention. For medications, includes oral, intravenous, subcutaneous,
subdermal, and intramuscular. Medication route is similar to an anatomic body
site through which the therapeutic or prophylactic agent is incorporated or
otherwise applied to the body. Other kinds of intervention routes might include:
via public health nurse counseling, billboard campaign, newspaper advertisement,
helicopter spray (for pesticide treatment), injection of water supply (for
fluoridation).
Rate Quantity : PQ
The period of time over which a specified dose is delivered. This attribute only
applies to continuously divisible intervention forms such as fluids and gases. In
this case, the intervention rate indicates the amount of intervention within a
specified period of time. The rate quantity is a duration (physical quantity in time),
and it is the denominator of the intervention rate, while intervention quantity is the
numerator. For example, pesticide to be used for mosquito abatement may be
delivered at a rate of 20 liters per minute from a spray applicator.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 39 OF 91 JULY 2000
Strength Quantity : PQ
The strength of an intervention is the amount of the agent per each unit of
administration. This applies to pesticides, chlorination as well as medication. If the
intervention form is continuously divisible (e.g., fluid, gas), the strength is a
concentration.
When the strength attribute is used, the actual administered amount is the product
of intervention quantity and strength quantity.
Class: NOTIFICATION
Subtype of: HEALTH-RELATED ACTIVITY
Description of: NOTIFICATION
A notification is an interaction with a caseworker, person or party to report or
document a condition or health-related activity of importance to the health of the
public. Includes notification by a provider to a patient that they have a disease,
report by a provider or laboratory to public health of a case or positive isolate,
report of a gunshot wound to police, reminder of the need for immunization
against disease, notification of a possible adverse reaction to a drug.
Attributes of: NOTIFICATION
Notification Reason Code: CV
Code for the reason for the notification. Examples might include reportable
condition, positive laboratory test, positive screening results, self-motivated,
interview, referral, and positive gonorrhea test.
Class: OBSERVATION
Subtype of: HEALTH-RELATED ACTIVITY
Supertype of: CASE
Description of: OBSERVATION
Observations are actions performed in order to determine an answer or result
value. Observation result values are specific information about the observed
object. The type and constraints of result values depend on the kind of action
performed.
An observation, according to Webster's, is an "act of recognizing and noting a fact
[...] often involving measurement with instruments" and at the same time an
observation is also "a record or description so obtained" [i.e., obtained through
recognizing and noting]. Thus an observation is both the action and measurement
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 40 OF 91 JULY 2000
"procedure" and the resulting information that was obtained. The model
understands the result to be entirely dependent on the observation action, and thus
models the result as a component (attribute) of the Observation action rather than
as an independent entity.
The following concepts are included as observations:
• A test is a procedure followed to objectively measure or evaluate the presence
or status of a condition. It includes vital signs, physical exams, food tests,
animal tests, height, and weight;
• An assessment of causality is the relationship between a patient condition and
a source that may be causally related to that condition;
• A vehicle condition is the circumstances under which the vehicle became a
carrier for a disease-causing agent. An example of a vehicle condition includes
temperature abuse in storing or preparing food;
• A diagnosis is the conclusion drawn from analysis of the signs and symptoms
exhibited or described by an individual;
• A party condition is the state of health, contamination, or infection of a party;
• A health status inquiry is the account of a party's health-related background.
This could include an interview conducted anonymously as part of a risk
factor survey. It includes description of current symptoms; risk behaviors such
as alcohol, tobacco, or other drug use; exposures past and present; medical or
surgical history; current or previous medications, vaccinations, or
interventions (treatment or prophylactic); reproductive history; occupational
history or exposures; sexual habits; eating habits; travel history; educational
background; marital status; family history. For example, the patient's, parent's,
or guardian's report of drug use, life style, previous medical conditions, and
treatments.
In the public health context, case and outbreak information are captured as
observations. This includes information such as a count or percentage of cases
tracked for public health reporting. It also includes number ill, number exposed,
number hospitalized, number treated, number of fatalities, number interviewed,
incubation period, duration of illness, number not ill, % female, % male, and %
less than 18 years of age.
Attributes of: OBSERVATION
Derivation Expression Text : ST
The derivation expression text shows how an observation can be derived from
other observations. In this case, the activity relationship links the observations
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 41 OF 91 JULY 2000
through the value of the relationship code (activity relationship type code =
"derivation").
For example, to define a derived observation for a change in antibody titer, one
will associate the change in titer observation with the acute titer observation and
the convalescent titer observation. The derivation expression text would then be
“Change in Titer = Convalescent Titer / Acute Titer”. If this observation value is
abnormal, for example greater than 4, this would be indicated in the Interpretation
Code for the Change in Titer observation.
Observation Value : ANY
The result value of an observation activity. This value can be of any datatype. This
fact reflects the many different ways in which the value of an observation can be
captured. For outbreaks or reporting of aggregate numbers of cases, the number of
persons affected would be included as a value here.
It is worth noting that, as a result of the functionality introduced with the activity
mood code, reference values or ranges are captured as observation values. The fact
that an observation carries a reference value is indicated by the value of the mood
code.
Class: OUTBREAK
Subtype of: CASE
Description of: OUTBREAK
An outbreak or cluster is the occurrence in a community or region of cases of a
condition of public health importance in excess of those normally expected. The
designation of an outbreak implies that a public health assessment of causality or
at least of relatedness among cases has taken place. An outbreak is considered to
be a special type of case (where a case, in this instance, may include many affected
individuals), and may not simply be an aggregate of multiple cases although an
outbreak may also be designated as an aggregate of multiple individual cases.
Given that an outbreak is a subtype of observation, the number of parties (which
will generally equate to the number of cases) affected by the outbreak is captured
as the observation value.
Attributes of: OUTBREAK
Outbreak Jurisdictional Extent Code : CV
Code for the qualitative measure of the number of jurisdictions involved. Possible
values include single jurisdiction, multi-county, multi-state, and multi-national.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 42 OF 91 JULY 2000
Note that if the specific jurisdictions are to be captured they are captured as target
participations associated with a jurisdictional party.
Outbreak Peak Date : TS
Date of onset for the highest number of cases (mode) associated with the outbreak.
Outbreak Time Range : IVL<TS>
The period of time during which the outbreak takes place. The date on which an
outbreak starts is the earliest date of onset among the cases assigned to the
outbreak, and its ending date is the last date of onset among the cases assigned to
the outbreak.
Class: REFERRAL
Subtype of: HEALTH-RELATED ACTIVITY
Description of: REFERRAL
A referral is an introduction of an individual or individuals from one health care
organization to another, or from one part of an organization to another for the
purpose of diagnosis or treatment. It includes the referral of a case or the referral of
multiple exposed persons (or cases) by one State Health Department to another.
Attributes of: REFERRAL
Referral Description: FTX
Free form text describing the referral.
Referral Reason Text: FTX
Free form text providing the reason for the referral as well as the action that is
expected or requested upon receipt of the referral. Examples might include
partner, positive lab test, outside of referring jurisdiction and needs follow-up,
possible cancerous lesion for biopsy, and requires surgical intervention.
Class: TARGET PARTICIPATION
Associated with: HEALTH-RELATED ACTIVITY
LOCATION
MATERIAL
PARTY
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 43 OF 91 JULY 2000
Description of: TARGET PARTICIPATION
Target participations include the passive parts played by a party in the health-
related activity. The target of a health-related activity can be any party or material,
including humans, other non-person living organisms, and inanimate material.
For example, within a disease investigation, the person identified as an actual or
potential carrier is a target of the activity. If the "patient" is a child, and another
person, such as a parent, speaks for them (e.g., answering a questionnaire) that
representative is also an activity target.
Associations for: TARGET PARTICIPATION
associates_to (1,1) :: HEALTH-RELATED ACTIVITY :: associates (0,n)
associates_to (0,1) :: LOCATION :: associates (0,n)
associates_to (0,1) :: MATERIAL :: associates (0,n)
associates to (0,1) :: PARTY :: associates (0,n)
Attributes of: TARGET PARTICIPATION
Target Awareness Code : CV
Indicates whether the associated patient or family member is aware of the health-
related activity, and especially of the observation made. This is only relevant for
persons who are targets of a health-related activity. For example, a patient (or his
family members) may not be aware of a malignancy diagnosis, the patient and
family may be aware at different times, and some patients may go through a phase
of denial.
Target Time Range : IVL<TS>
The time range in which the associated party or material was a target of the
specified target type code in the associated activity.
Target Type Code : CV
Identifies the particular role in which the party appears as the target of the health-
related activity.
Examples of target type codes include: "State reporting case", "target of case",
"location imported from".
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 44 OF 91 JULY 2000
Locations Subject Area
Figure 3. Locations Subject Area Diagram
The classes and attributes for the Locations subject area are described below.
Class: LOCATION
Supertype of: PHYSICAL LOCATION
POSTAL LOCATION
TELECOMMUNICATION LOCATION
Associated with: LOCATION RELATIONSHIP
LOCATION RELATIONSHIP
MATERIAL LOCATION PARTICIPATION
PARTY LOCATION PARTICIPATION
TARGET PARTICIPATION
Description of: LOCATION
A location is a site of interest to public health. Examples of locations include
buildings, picnic grounds, regional areas, homes, test locations, specimen
locations, hospitals, day care centers, prisons, and other potential transmission
TELECOMMUNICATION LOCATION
Personal Identification Number : ST
Time Zone Text : ST
Electronic Address Text : TEL
POSTAL LOCATION
Street Address Text : AD
Address Directions Text : FTX
PHYSICAL LOCATION
Latitude Quantity : ST
Longitude Quantity : ST
Location Name : ST
Property Location Text : FTX
PARTY
(from PARTY)
MATERIAL
(from MATERIAL)
LOCATION RELATIONSHIP
Location Relationship Type Code : CV
Location Relationship Date Time Range : IVL<TS>
PARTY LOCATION PARTICIPATION
Participation Date Time Range : IVL<TS>
Participation Type Code : CV
Current Status Code : CV
Current Status Effective Date : TS
1
0..*
1
0..*
MATERIAL LOCATION PARTICIPATION
Participation Date Time Range : IVL<TS>
Participation Type Code : CV
1
0..*
1
0..*
LOCATION
Location Identifier : SET<II>
Location Setting Code : CV
Location Type Code : CV
Location Status Code : CV
Current Status Effective Date : TS
Location Narrative Text : FTX
0..*
1
0..*
1
0..*
1
0..*
1
1
0..*
1
0..*
0..*
1
0..*
1
TARGET PARTICIPATION
(from HEALTH-RELATED ACTIVITY)
0..1
0..*
0..1
0..*
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 45 OF 91 JULY 2000
locations. It also includes districts - that is to say one location may contain
another. The information for a location includes information such as an address
that makes it possible to find or to send messages to the location.
Associations for: LOCATION
is_source_for (0,n) :: LOCATION RELATIONSHIP :: relates (1,1)
is_target_for (0,n) :: LOCATION RELATIONSHIP :: relates (1,1)
associates (0,n) :: MATERIAL LOCATION PARTICIPATION :: associates_to
(1,1)
associates_to (0,n) :: PARTY LOCATION PARTICIPATION :: associates (1,1)
associates (0,n) :: TARGET PARTICIPATION :: associates_to (0,1)
Attributes of: LOCATION
Current Status Date Time Range : IVL<TS>
The time range during which the current location status is or was active.
Location Identifier : SET<II>
An instance identifier that identifies the location. This could include, among other
things, identifiers assigned to a property within a registry office or other
organization tracking plots of land.
Location Narrative Text : FTX
A free text note that carries additional information related to the location. This
could include instructions for finding the location when postal information is
inadequate. It could also include information useful to people visiting the location
(e.g., “Beware of dog”).
Location Setting Code : CV
Code for the location environment. Examples might include public, private,
federal, and unknown.
Location Status Code : CV
An indication of the validity of the location. Examples might include verified,
unverified, and unable to verify.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 46 OF 91 JULY 2000
Location Type Code : CV
Code that indicates the type of location. Includes residence, office, restaurant,
hospital, daycare center, ship, prison, nursing home, or district such as census tract
or congressional district.
Class: LOCATION RELATIONSHIP
Associated with: LOCATION
Description of: LOCATION RELATIONSHIP
An association between two locations. This relationship is important in public
health reporting and investigations to describe how sites of public health
importance are associated, for instance: fourth floor of hospital "has as part" the
neonatal ICU. Here, the location relationship, "has as part", describes the
association between two locations, a particular ICU and the hospital floor.
Another example might be juice maker's apple orchard "is next to" farmer's cow
pasture. One can also link telecommunication locations or postal locations to
physical locations, for instance, 123 Main Street, Doraville, GA 30256 "is
geolocated by" +33 47.966, -84 19.508.
This structure is not needed to link multiple locations, e.g., home address, email
address, business address, to a single party. That requirement is supported through
linking location information to party with party location participation.
Associations for: LOCATION RELATIONSHIP
relates (1,1) :: LOCATION :: is_source_for (0,n)
Attributes of: LOCATION RELATIONSHIP
Location Relationship Date Time Range : IVL<TS>
The period in time during which the relationship between the two location
instances is effective. The time interval can be open at either end. That is, both the
start and stop dates for the participation could be indicated, or either start or stop
by themselves.
Location Relationship Type Code : CV
Indicates the type of relationship between the two locations. For example, "same
as", "adjacent to".
Class: MATERIAL LOCATION PARTICIPATION
Associated with: LOCATION
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 47 OF 91 JULY 2000
MATERIAL
Description of: MATERIAL LOCATION PARTICIPATION
Material location participation indicates the location where an item of material is
or was to be found.
Associations for: MATERIAL LOCATION PARTICIPATION
associates_to (1,1) :: LOCATION :: associates (0,n)
associates_to (1,1) :: MATERIAL :: associates (0,n)
Attributes of: MATERIAL LOCATION PARTICIPATION
Participation Date Time Range : IVL<TS>
Indicates the period in time during which the material item is or was to be found at
the location. For example, the date a specimen arrived at the location. The time
interval can be open at either end. That is, both the start and stop dates for the
participation could be indicated, or either start or stop dates by themselves.
Participation Type Code : CV
Code for the participation role of the material at the location. Examples might
include “resides at”, “originated at”, and “destined for”.
Class: PARTY LOCATION PARTICIPATION
Associated with: LOCATION
PARTY
Description of: PARTY LOCATION PARTICIPATION
Party location participation indicates the relationship between a party and a
location. The party may be an organization that owns several facilities or locations.
The participation role would be that of owner of the facility at this location.
Another role for a party would be a person who "works at" a location.
Associations for: PARTY LOCATION PARTICIPATION
associates (1,1) :: LOCATION :: associates_to (0,n)
associates_to (1,1) :: PARTY :: associates (0,n)
Attributes of: PARTY LOCATION PARTICIPATION
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 48 OF 91 JULY 2000
Current Status Code : CV
Code for the status of the participation between the party and the location.
Current Status Effective Date : TS
The effective date for the current party location role status.
Participation Date Time Range : IVL<TS>
Indicates the period in time during which the party is related to the location. The
time interval can be open at either end. That is, both the start and stop dates for the
participation could be indicated, or either start or stop by themselves.
Participation Type Code : CV
Code for the participation role of the party at the location. Examples might
include owner, occupant, visitor, worker, and client.
Class: PHYSICAL LOCATION
Subtype of: LOCATION
Description of: PHYSICAL LOCATION
Physical location information makes it possible to find the location on a map or by
examination of surveyor's documentation or by reference to a land or property
registry.
Attributes of: PHYSICAL LOCATION
Latitude Quantity : ST
Indicates the latitude of the location as measured in degrees north or south of the
equator.
Location Name : ST
The name of the location as it might be referred to on a map or in a registry.
Longitude Quantity : ST
Indicates the longitude of the location as measured in degrees west or east of the
prime meridian at Greenwich, England.
Property Location Text : FTX
A description of the property that is sufficiently precise to enable someone to
locate the property and to recognize its boundaries. The description can be
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 49 OF 91 JULY 2000
formulated as in terms of the property boundaries, or in terms of specific lots or
parcels that are located within a legal entity such as a township, county, or other
legally defined territorial entity. In some cases the description will be drawn from
the legal description of a property as recorded on a deed or other legal paper.
Class: POSTAL LOCATION
Subtype of: LOCATION
Description of: POSTAL LOCATION
Information used to direct mail to a particular location, or to find the location using
information to be found on a street map.
Attributes of: POSTAL LOCATION
Address Directions Text : FTX
Descriptive information to assist a party in finding a particular location. This
information is intended to supplement or replace street address information.
Street Address Text : AD
Text used for an address label. This could include street address information, or
postal directions using a box number to send mail to a post office box, a rural free
delivery box, or a military post office. It also includes lot or address number when
the address refers to an apartment building or housing complex.
Class: TELECOMMUNICATION LOCATION
Subtype of: LOCATION
Description of: TELECOMMUNICATION LOCATION
An electronic address for a party that provides the mechanism to contact the party,
to send messages, or to access information relevant to the party. Examples include
a telephone number, an email address, a World Wide Web URL. This is
distinguished from a postal address.
Attributes of: TELECOMMUNICATION LOCATION
Electronic Address Text : TEL
The number or other string that is entered to contact a particular telephone or other
electronic location.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 50 OF 91 JULY 2000
Personal Identification Number : ST
An identification number assigned to a person, and used to access a
communication device such as a beeper. Often referred to as a PIN.
Time Zone Text : ST
Text indicating the time zone in which a telephone is located.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 51 OF 91 JULY 2000
Materials Subject Area
Figure 4. Materials Subject Area Diagram
The classes and attributes for the Materials subject area are described below.
Class: MATERIAL
Supertype of: SPECIMEN
Associated with: MATERIAL LOCATION PARTICIPATION
MATERIAL RELATIONSHIP
MATERIAL RESPONSIBILITY
TARGET PARTICIPATION
Description of: MATERIAL
Material is defined according to Webster's: 1) the elements, constituents, or
substances of which something is composed or can be made; 2) matter that has
qualities which give it individuality and by which it may be categorized.
In public health, interest in materials commonly arises when a material is a vehicle
for a disease agent, or is suspected of being such a vehicle. For example, when a
case investigation considers the question of whether a bowl of potato salad is
contaminated with Salmonella organisms, the potato salad might be recorded as an
item of material. Note that this assumes that the identity of the potato salad needs
to be captured. In some cases it would be sufficient to record an observation that
SPECIMEN
Source Site Code : CV
PARTY
(from PARTY)
MATERIAL RELATIONSHIP
Material Relationship Type Code : CV
Material Relationship Date Time Range : IVL<TS>
MATERIAL RESPONSIBILITY
Responsibility Type Code : CV
Responsibility Date Time Range : IVL<TS>
Material Identifier : II
1
0..*
1
0..*
TARGET PARTICIPATION
(from HEALTH-RELATED ACTIVITY)
MATERIAL
Material Identifier : SET<II>
Material Type Code : CV
Material Description : FTX
Material Date Time Range : IVL<TS>
Handling Code : CV
Danger Code : CV
Material Quantity : PQ
Material Name : ST
1 0..*1 0..*
1
0..*
1
0..*
1
0..*
1
0..*
0..1
0..*
0..1
0..*
MATERIAL LOCATION PARTICIPATION
(from LOCATION)
1
0..*
1
0..*
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 52 OF 91 JULY 2000
the contaminated food was potato salad. It is also possible, when collecting
information about the bacteria, to capture it as an item of material.
Other materials or entities of interest to public health may include an independent,
separate, or self-contained substance or object, such as a lake, a pool, a waterpark,
resort, campsite, ship, airplane, or train that might serve as a source or vehicle of
exposure to a health hazard. For example, a public health investigation can center
around the question of bacterial or other contamination of a site such as a ship or
swimming pool. Specimens can be taken from such materials just as specimens
can be taken from parties, whether human or otherwise.
Associations for: MATERIAL
associates (0,n) :: MATERIAL LOCATION PARTICIPATION :: associates_to
(1,1)
relates (0,n) :: MATERIAL RELATIONSHIP :: is_source_for (1,1)
associates (0,n) :: MATERIAL RESPONSIBILITY :: associates_to (1,1)
associates (0,n) :: TARGET PARTICIPATION :: associates_to (0,1)
Attributes of: MATERIAL
Danger Code : CV
A code signaling whether there are certain dangers or hazards associated with this
material. For example, "Examine under hood", "Wear gloves".
Handling Code : CV
A code to describe how the material needs to be handled to avoid damage. For
example: "Do not expose to light", "Keep at certain temperature".
Material Date Time Range : IVL<TS>
An indication of the time interval during which the material is in existence.
Material Description : FTX
A free text description of the material. May contain multimedia, such as a drawing
or image depicting the material.
Material Identifier : SET<II>
The identifier assigned to an individual material item.
Ideally each entity will have only one identifier assigned to it. However, since
different systems will maintain different material databases, there may be different
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 53 OF 91 JULY 2000
instance identifiers assigned by different systems. Note that for serial numbers
assigned by specific manufacturers, catalog numbers of specific distributors, or for
inventory numbers issued by owners, the attribute Material Identifier in the
Material Responsibility class can also be used. This allows clearer expression of
the fact that a specific party associated with that material assigns such a code.
Material Name : ST
Name of the material. This is important in special cases such as the name of a lake,
an amusement park, or a cruise ship
Material Quantity : PQ
An indication of the amount of material. This could be a count or a quantity. For
example, 2 liters of water, 25 vials of blood.
Material Type Code : CV
This code describes the kind of material. No single terminology is expected to
provide all concepts that are types of material, since it is simply too broad a
domain. Instead of limiting the Material Type Code to a single domain, various
coding systems may be used.
For example, specimen types (e.g., whole blood, serum, and urine) can be used in
this attribute. For pharmacological substances the U.S. National Drug Code
(NDC) may be applicable. For other types of materials of interest to public health,
such as lakes, rivers, national parks, trains, planes, or ships, other coding systems
will be applicable.
Class: MATERIAL RELATIONSHIP
Associated with: MATERIAL
Description of: MATERIAL RELATIONSHIP
Material relationship captures the relationship between two items of material.
Material relates to other material largely in some kind of whole-part or
containment relationship. The special functioning of the material relationship
depends on the role of material, i.e., whether the material is a discrete thing, a
homogenous substance, or a container. Material can be all of those forms.
Associations for: MATERIAL RELATIONSHIP
is_source_for (1,1) :: MATERIAL :: relates (0,n)
is_target_for (1,1) :: MATERIAL :: relates (0,n)
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 54 OF 91 JULY 2000
Attributes of: MATERIAL RELATIONSHIP
Material Relationship Type Code : CV
Code for the type of material association. Every relationship type implies certain
roles for the material on either side of the relationship. For example, there is a
relationship between a blood specimen and the species of bacteria cultured from it,
between a dish of food and the ingredients used to make it, and a lake and the
sample collected from it. Thus examples of material relationship codes include: is
cultured from, is an ingredient of, and is a sample from.
Material Relationship Date Time Range : IVL<TS>
The period of time during which the relationship between the two materials is
valid.
Class: MATERIAL RESPONSIBILITY
Associated with: MATERIAL
PARTY
Description of: MATERIAL RESPONSIBILITY
Description of the type of relationship between a party and an item of material.
Material can have many kinds of relationships with parties. Relationships between
material and parties are included here since there are generally one or more parties
responsible for managing an item, or for performing particular functions with it.
For example, manufacturing is an activity in which a party or parties acts on
material. In some instances we may simply be interested in who made the
material. We may also be concerned with how the material item has been
processed or treated. For example, if the manufacturing of the material resulted in
contaminated or doctored medications, or food was not held at or cooked to proper
temperatures, there are significant implications for public health.
An important example of material responsibility is the role of a party as the
provider or receiver of a specimen. For example, a lake or a food item or a person
may be the source of a specimen, and a public health official may be the person
who obtains the specimen (the specimen may be a lake or food sample, a body
part, blood sample, sputum, or feces). Owner, distributor, and custodian/holder are
additional examples of relationship types between material and party.
Similarly, when a material item is implicated as a vehicle for a disease condition,
such as a food item that is contaminated with Salmonella organisms, the material
responsibility class provides a way to record party responsibility for the food item.
This could include recording the party who was responsible for its pasteurization,
the party who prepared the food, or the party responsible for storing it. For
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 55 OF 91 JULY 2000
medications or intravenous solutions, this might be the party responsible for
sterilization or for mixing the solution.
Associations for: MATERIAL RESPONSIBILITY
associates_to (1,1) :: MATERIAL :: associates (0,n)
associates (1,1) :: PARTY :: associated to (0,n)
Attributes of: MATERIAL RESPONSIBILITY
Material Identifier : II
An identifier assigned to a material item in the context of its relationship with a
responsible party. Different responsible parties may give the same piece of
material different identifiers. For example, a manufacturer may assign a
manufacturer ID and a distributor may assign a catalog number. All those
identifiers can in principle occur under the Material ID attribute, i.e., as a property
of the material itself. However, this attribute allows one to make the scope of the
ID more clear, i.e., it helps to easily distinguish a specific manufacturer's ID from
a distributor's ID much more clearly than can be done using the assigning
authority component of the instance identifier datatype.
Responsibility Date Time Range : IVL<TS>
Indicates the period of time during which the responsibility holds.
Responsibility Type Code : CV
Specification of the kind of responsibility that the party takes on with respect to
the material. Examples might include owner, responsible for preparation,
custodian.
Class: SPECIMEN
Subtype of: MATERIAL
Description of: SPECIMEN
A specimen is a part, fraction, aliquot, component, tissue sample, body fluid, food,
or other substance that is collected in a health-related activity to support the
assessment, diagnosis, or treatment of a party.
Attributes of: SPECIMEN
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 56 OF 91 JULY 2000
Source Site Code : CV
The source site code indicates from where, in relationship to the specimen source,
the specimen is taken. For persons and non-person living organisms, the valid
domain is a list of body sites. This is an attribute of the specimen, since it may be
relevant in some cases, e.g., if multiple liver needle biopsies are taken from
different lobes and locations of the liver. In the case of material items such as
restaurants or lakes, the site code indicates from where the specimen was taken. In
the case of a lake, this could be, "near intake", or "at swimming site". In the case
of a restaurant, this could indicate a typical site in the restaurant such as within the
meat grinder.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 57 OF 91 JULY 2000
Parties Subject Area
Figure 5. Parties Subject Area Diagram
The classes and attributes for the Parties subject area are described below.
Class: FORMAL ORGANIZATION
Subtype of: ORGANIZATION
Description of: FORMAL ORGANIZATION
A formal organization is an administrative and functional structure with common
objectives. Examples in public health might include state-based public health
membership organizations such as the Association of Public Health Laboratories
(APHL), Association of State and Territorial Health Officials (ASTHO), the
Council of State and Territorial Epidemiologists (CSTE), National Association of
County and City Health Officials (NACCHO), National Association for Public
Health Statistics and Information Systems (NAPHSIS), as well as individual
organizations such as California Department of Health Services, Dekalb County
Health Department, Blue Cross/Blue Shield Health Plans, Kaiser Permanente
Health Maintenance Organization, Quest Diagnostics, Environmental Protection
Agency.
Attributes of: FORMAL ORGANIZATION
ORGANIZATION
Organization Name : SET<ON>
INFORMAL ORGANIZATION
Group Type Code : CV
FORMAL ORGANIZATION
Industry Code : CV
INDIVIDUAL
Birth Date : TS
Death Date : TS
Sex Code : CV
PERSON
Ethnicity Code : CV
Race Code : SET<CV>
Occupation Code : CV
Person Name : SET<PN>
NON PERSON LIVING ORGANISM
Species Name : CV
Organism Name : ST
PARTY RELATIONSHIP
Party Relationship Date Time Range : IVL<TS>
Party Relationship Type Code : CV
ACTOR PARTICIPATION
(from HEALTH-RELATED ACTIVITY)
TARGET PARTICIPATION
(from HEALTH-RELATED ACTIVITY)
MATERIAL RESPONSIBILITY
(from MATERIAL)
PARTY
Party Identifier : SET<II>
1
0..*
1
0..*
10..* 10..*
0..*
1
0..*
1
0..*
0..1
0..*
0..1
1
0..*
1
0..*
PARTY LOCATION PARTICIPATION
(from LOCATION)
1
0..*
1
0..*
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 58 OF 91 JULY 2000
Industry Code : CV
Code for the type of activity or industry in which the organization is engaged.
Class: INDIVIDUAL
Subtype of: PARTY
Supertype of: NON-PERSON LIVING ORGANISM
PERSON
Description of: INDIVIDUAL
An individual is a human person or other single organism.
When non-person living organisms are under consideration, their identity should
only be recorded as a party when it is reasonable to do so, such as when they need
to be recorded in reference to a individual or series of health-related activities.
Note that parties can be identified in order to record an association to a material
item or to a location. This is not likely to occur for a non-human living organism
except in the non-trivial case of specimens. As a general rule, such non-
individually identified organisms as microorganisms and viruses will not be
recorded as parties. Information about them will be captured as observations.
Attributes of: INDIVIDUAL
Birth Date : TS
Date on which the individual was born.
Death Date : TS
Date on which the individual died.
Sex Code : CV
Code for the individual's sex at birth. Includes Male and Female.
Class: INFORMAL ORGANIZATION
Subtype of: ORGANIZATION
Description of: INFORMAL ORGANIZATION
An informal organization is a casual grouping or cluster of individuals with
common interests, characteristics or exposures, or relationships. An informal
organization can include individuals who do not recognize their relationship to the
rest of the group, and in fact, this class is particularly intended to represent
populations or groups of interest to public health, e.g., persons who are smokers,
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 59 OF 91 JULY 2000
persons of a certain age or race, persons exposed to the same chemical or agent,
and persons who are HIV-positive. The concept of informal organizations also
includes such clusters as families, neighborhoods, support groups, and groups of
migrant workers. The informal organization can group non-human parties as well.
Therefore it includes herds of cattle, canine litters, and prides of lions.
Attributes of: INFORMAL ORGANIZATION
Group Type Code: CV
Code for the type of informal organization. Examples include groups such as
families, Rotary Club members, girl scouts, retired persons, persons with heart
disease, alcoholics, persons vaccinated against measles, persons who are chronic
typhoid carriers, or patients on a given floor or ward of a hospital.
Class: NON-PERSON LIVING ORGANISM
Subtype of: INDIVIDUAL
Description of: NON-PERSON LIVING ORGANISM
A non-person living organism is an individual living thing other than a human
being that is sufficiently important in its own right to model as a party. For
example, this includes pets and working or farm animals whose condition is under
investigation.
Normally, other living clusters such as bacteria, parasites, viruses, prions, and
insects, are modeled as specimens. Information about them is captured as an
observation or observations. Such living clusters should only be recorded as
parties when it is necessary to capture multiple references to the same individual in
the course of a health-related activity.
Attributes of: NON-PERSON LIVING ORGANISM
Organism Name : ST
The name assigned to an animal or other organism. For example, the name
assigned to a pet or to a working animal such as a racehorse.
Species Name : CV
The name of the species, including both the genus and the species. This value is
drawn from a coded domain that contains the names of the known species.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 60 OF 91 JULY 2000
Class: ORGANIZATION
Subtype of: PARTY
Supertype of: FORMAL ORGANIZATION
INFORMAL ORGANIZATION
Description of: ORGANIZATION
Organizations provide a way to recognize the grouping and/or collective action of
individuals. An organization may be a group of functions operating as a unit.
Examples are managed care organizations, hospital systems, State Health
Departments, and regulatory agencies. Such an organization is modeled as a
formal organization. An organization may also be simply a group of interest that
has been assembled or defined in some informal manner. This type of
organization is modeled in the PHCDM as an informal organization. Examples of
such are social groups or units such as families, boy scouts, day care attendees,
and college students.
Attributes of: ORGANIZATION
Organization Name : SET<ON>
Name of the organization.
Class: PARTY
Supertype of: INDIVIDUAL
ORGANIZATION
Associated with: ACTOR PARTICIPATION
MATERIAL RESPONSIBILITY
PARTY LOCATION PARTICIPATION
PARTY RELATIONSHIP
TARGET PARTICIPATION
Description of: PARTY
A party is an individual or organization that is specifically of interest to public
health. This model includes the concept of "party", in order to clearly represent the
similar ways that the different kinds of party are related to health-related activities,
materials, and locations. These similarities are particularly relevant in the public
health context due to the broad range of concerns that come up.
Something is captured as a party when there is a specific interest in its associations
with health-related activities. That is to say, information is captured about a
particular individual or organization that makes it desirable to record its individual
existence. Usually this implies there will be a series of associations with that
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 61 OF 91 JULY 2000
individual that need to be linked. This distinction is important because information
can also be captured as an observation (i.e., a health-related activity). For example,
we expect that pets and specific farm animals such as horses and cows will be
captured as parties (non-person living organisms).
Concepts that are not considered parties include purely material entities, such as
lakes or parks. These are considered to be a type of material. Bacteria discovered
within a specimen will be captured as observations made on that specimen.
The best way to illustrate this point is through the use of examples. Public health
interventions are sometimes applied to specific persons. This includes the delivery
of treatment to prevent the development of tuberculosis, a vaccination given to a
patient exposed to rabies. It also includes the delivery of information, as when a
sexual partner of a patient with a sexually transmitted disease is provided with
counseling and clinical information about the disease (along with therapy to
prevent disease).
Public health interventions are sometimes applied to organizations. Note that this
model treats groups of people as informal organizations. Examples include
providing vaccinations and information to the members of a boarding school
where a case of meningitis was diagnosed, and the delivery of health warnings to
the general public when Shigella organisms are detected in a commercial food
product. Education campaigns related to such topics as AIDS prevention, the
dangers of tobacco use, and the importance of calcium in diets are regarded as
public health interventions and may be delivered to such "organizations" that
include the population of a city, state, or region, or to specific age cohorts or
otherwise identifiable groups.
Public health interventions are sometimes applied to non-person living organisms.
For example, dogs living as pets within a neighborhood might receive additional
rabies inoculations when several dead and infected raccoons were found in the
vicinity. Members of a herd of cattle might be treated when disease was
encountered in one of them. Note that within this model an informal organization
includes relevant groupings of individuals. These individuals could be persons or
non-person living organisms. Therefore, a herd of cattle is an informal
organization.
Associations for: PARTY
associates (0,n) :: ACTOR PARTICIPATION :: associates to (1,1)
associated to (0,n) :: MATERIAL RESPONSIBILITY :: associates (1,1)
associates (0,n) :: PARTY LOCATION PARTICIPATION :: associates_to (1,1)
relates to (0,n) :: PARTY RELATIONSHIP :: is source for (1,1)
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 62 OF 91 JULY 2000
associates (0,n) :: TARGET PARTICIPATION :: associates to (0,1)
Attributes of: PARTY
Party Identifier : SET<II>
A party identifier is a value that identifies a party.
Class: PARTY RELATIONSHIP
Associated with: PARTY
PARTY
Description of: PARTY RELATIONSHIP
A party relationship captures the relationship between two parties. Examples of
party relationships might include sexual partners, marital relationship, primary
caretaker and subject, and employment between parties. Further examples include
parent to child, health care provider to patient, health coverage organization to
patient. The relationship between a person and their foster parent, adoptive parent,
relative, emergency contact, or spouse is captured by this class. This association
generally refers to a relationship that exists outside of the particular event of
current interest, such as a specific health-related activity.
Associations for: PARTY RELATIONSHIP
is source for (1,1) :: PARTY :: relates to (0,n)
is target for (1,1) :: PARTY :: relates to (0,n)
Attributes of: PARTY RELATIONSHIP
Relationship Date Time Range : IVL<TS>
The period of time during which the relationship between the two parties is valid.
Relationship Type Code : CV
Code for the type of party relationship. Examples might include is an employee of,
is the sexual partner of, and is the parent/child of.
Class: PERSON
Subtype of: INDIVIDUAL
Description of: PERSON
A person is a human individual.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 63 OF 91 JULY 2000
Attributes of: PERSON
Ethnicity Code : CV
Code for the person's ethnic background (e.g., Hispanic, non-Hispanic).
Occupation Code : CV
Code for the occupation in which the person is employed.
Person Name : SET<PN>
A Person Name is a name assigned to a person.
Race Code : SET<CV>
Code for the person's race (e.g., American Indian/Alaskan Native, White, African
American, Asian, Hawaiian/Pacific Islander). The attribute repeats in order to
record the multiple racial categories to which a person can belong.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 64 OF 91 JULY 2000
APPENDICES
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 65 OF 91 JULY 2000
Appendices
Datatypes
This section contains the datatype definitions used within the Public Health
Conceptual Data Model. The datatypes are drawn from the HL7 Reference
Information Model, and represent a subset of the datatypes defined therein.
Figure 6. Datatypes Diagram
The classes and attributes that comprise the datatypes are described below.
Datatype: ADDRESS : AD
Is a Composite Datatype
Description of: AD
This Address datatype is used to communicate postal addresses and residential
addresses. The main use of such data is to allow printing mail labels (postal
address), or to allow a person to physically visit that address (residential address).
The difference between postal and residential address is whether or not there is
just a post office box. The residential address is not supposed to contain other
ANY: ANY
BINARY DATA: BIN
CHARACTER STRING: ST
CODE VALUE: CV
Code system : OID
Code System Version : ST
Print Name : ST
Replacement : ST
Value : ST
DISCRETE: DSCR
FREE TEXT: FTX
Charset : CV
Compression : CV
Data : BIN
Media Descriptor : CV
GENERAL TIME SPECIFICATION: GTS
INSTANCE IDENTIFIER: II
Assigning Authority Identifier : OID
Assigning Authority Name : ST
Identifier Type Code : CV
Identifier Valid Time Range : IVL<TS>
Value Text : ST
INTEGER: INT
INTERVAL: IVL
Low : T
Low Closed : BL
High : T
High Closed : BL
ISO OBJECT IDENTIFIER: OID
ORDERED: ORD
PHYSICAL QUANTITY: PQ
Unit : CV
Value : REAL
POINT IN TIME: TS
QUANTITY: QTY
REAL: REAL
Precision : N
Value : N
SET: SET
NUMBER: N
BOOLEAN: BL
ORGANIZATION NAME: ON
Type Code : CV
Value : ST
PERSON NAME: PN
Value : LIST<PNXP>
PERSON NAME PART: PNXP
Classifiers : SET<CV>
Value : ST
ADDRESS PART: ADXP
Role : CV = initval
Value : ST = initval
UNIVERSAL RESOURCE IDENTIFIER: URI
TELECOMMUNICATION ADDRESS: TEL
Address : URI
Use Code : SET<CV>
Valid Time : GTS
ADDRESS: AD
Bad Indicator : BL
Purpose Code : CV
Value : LIST<ADXP>
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 66 OF 91 JULY 2000
information that might be useful for finding geographic locations or doing
epidemiological studies. These addresses are thus not very well suited for
describing the locations of mobile visits or the "residency" of homeless people.
Components of: AD
Bad Indicator : BL
Indicates that this address is not working
Purpose Code : CV
A purpose code indicates the use for a given address. Examples might include
preferred residency (used primarily for visiting), temporary (visit or mailing, but
see History), preferred mailing address (used specifically for mailing), and some
more specific ones, such as "birth address" (to track addresses of small children).
An address without specific purpose code might be a default address useful for
any purpose, but an address with a specific purpose code would be preferred for
that respective purpose.
Value : LIST<ADXP>
This contains the actual address data as a list of address parts that may or may not
have semantic tags.
Datatype: ADDRESS PART : ADXP
Is a Composite Datatype
Description of: ADXP
This type is not used outside of the Address datatype. Addresses are regarded as a
token list. Tokens usually are character strings but may have a tag that signifies the
role of the token. Typical parts that exist in about every address are ZIP code, city,
country but other roles may be defined regionally, nationally, or on an enterprise
level (e.g., in military addresses). Addresses are usually broken up into lines that
are indicated by special line break tokens.
Components of: ADXP
Role : CV
The role of an address part (if any) indicate whether an address part is the ZIP
code, city, country, or post office box.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 67 OF 91 JULY 2000
Value : ST
The value of an address part includes the text for the specific component of the
address. It is what is printed on an address label.
Datatype: ANY : ANY
Is a Primitive Datatype
Description of: ANY
This is a generalized datatype that represents any other datatype within the model.
This concept is needed to support observation values, and to let those values take
on any datatype.
Datatype: BINARY DATA : BIN
Is a Primitive Datatype
Has Super Types: DSCR
Description of: BIN
Binary data is a sequence of uninterpreted raw bytes (8 bit sequences, or octets).
Datatype: BOOLEAN : BL
Is a Primitive Datatype
Has Super Types: DSCR
Description of: BL
The boolean type stands for the values of two-valued logic. A boolean value can
be either true or false.
Datatype: CHARACTER STRING : ST
Is a Primitive Datatype
Has Super Types: DSCR
Description of: ST
A string of characters where every character is represented by a uniquely
identifiable entity within the string.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 68 OF 91 JULY 2000
Datatype: CODE VALUE : CV
Is a Composite Datatype
Has Super Types: DSCR
Description of: CV
A code value is exactly one symbol in a coding system. The meaning of the
symbol is defined exclusively and completely by the coding system from which
the symbol originates.
Components of: CV
Code system : OID
An object identifier referring to the code system that defines the code value. The
OID supports unambiguous reference to standard coding systems - including HL7
codes, as well as to local codes.
Code System Version : ST
A version descriptor defined specifically for the given coding system.
Print Name : ST
A sensible name for the code as a courtesy to an interpreter of the message. The
name should not be considered as carrying the meaning of the code, it should
never be sent alone, and it does not modify the meaning of the code.
Replacement : ST
A name for the concept whose meaning is being conveyed. The replacement is
used if the concept cannot be captured by a code in the specified coding system. If
the value attribute is set, the replacement attribute MUST NOT be set. In no way
can a replacement string modify the meaning of the code value.
Value : ST
This is the plain symbol. E.g., "784.0"
Datatype: DISCRETE : DSCR
Is a Primitive Datatype
Has Sub Types: ST
REAL
OID
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 69 OF 91 JULY 2000
N
IVL
INT
II
CV
BL
BIN
Description of: DSCR
Abstract generalized type for any discrete type.
Datatype: FREE TEXT : FTX
Is a Composite Datatype
Description of: FTX
This free text datatype can convey any data whose primary purpose is to be shown
to people for interpretation. Free text can be any kind of text, whether written
language (formatted or unformatted) or multi-media data.
Components of: FTX
Charset : CV
Definition of the character encoding if different from the default encoding.
Compression : CV
Indicates that the raw byte data is compressed, and which compression algorithm
is being used.
Data : BIN
Contains the free text data as raw bytes.
Media Descriptor : CV
Allows selection of the appropriate free text data. The default value is "text/plain".
Datatype: GENERAL TIME SPECIFICATION : GTS
Is a Primitive Datatype
Description of: GTS
This is a primitive datatype that is conceptually an arbitrary set of points in time. It
is any combination of 1) a point in time, and 2) an interval of time. This includes
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 70 OF 91 JULY 2000
uncertain points and intervals of time. The contents of a GTS instance contains
values that are defined in terms of a literal expression syntax that allows statement
of any needed frequency or time pattern.
For example, these are some ways in which the GTS datatype is used within the
PHCDM:
Dates are represented based on the precision needed and/or supplied. Y1999
represents the year 1999, while Y199909 indicates September, 1999, and
Y19990926 indicates September 26, 1999. If only month and day are available,
M0926 indicates September 26. The following are some of the period identifiers
available: Y for year, M (or MY) for month of the year, D (or DM) for day of the
month, H (or HD) for hour of the day.
The tag WY (w for week and y for year) indicates week of the year. Therefore
WY23 is the 23rd week of the year.
Date ranges are indicated by a pair of dates separated by a dash, "-". For example,
"Y20000110 - Y20000204" indicates a period beginning on January 10, 2000 and
ending on February 4, 2000. Open-ended periods can be indicated by a date either
preceded by a dash (period ending on the date indicated), or by a date followed by
a dash (period beginning on the date indicated).
Durations are indicated by square brackets, "[" and "]". The duration amount and
unit are within the brackets. For example "[10 min]" indicates 10 minutes, and "[3
day]" indicates 3 days.
These expressions can be concatenated together to express the union of multiple
time concepts. For example, "M09 D26" is an alternate way of indicating
September 26. Also, "Y20000224 [8 hour]" indicates a duration of 8 hours on
February 24, 2000.
As a general statement, the construction of GTS instances is based on the
following ideas:
• Singular time intervals as continuous sets of time points, specified through low
and high boundary or width (in case no boundary is known.)
• Periodic time intervals as discontinuous sets of time points, specified through a
period duration and a time offset (phase) interval.
• The set-operations intersection and union on such continuous and
discontinuous sets of time to form arbitrary sets of time.
• The reduction of any arbitrary set of time into an outer bound interval and a
sequence of occurrence intervals, no matter how complex the definition of this
arbitrary set is.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 71 OF 91 JULY 2000
• The use of probability distribution datatypes to account for the uncertainty in
scheduling and time orders, or, in other words, to allow "fuzzy" constraining
of time sets.
Time can be specified in terms of an absolute even flow of time, or events taking place in time
can be aligned to calendars.
Datatype: INSTANCE IDENTIFIER : II
Is a Composite Datatype
Has Super Types: DSCR
Description of: II
The datatype is used to uniquely identify an entity that exists within a computer
system or other well-controlled identification scheme.
Components of: II
Assigning Authority Identifier : OID
The ISO object identifier for the organization or identifier issuing scheme that is
responsible for the integrity and validity of the identifier. This field guarantees the
uniqueness of the identifier, and permits the origin of the identifier to be
determined. If the organization uses OIDs for internal object identifiers, this may
be the only field valued.
Assigning Authority Name : ST
The name of the organization or scheme responsible for the identifier.
Identifier Type Code : CV
A code representing the type of identifier. For example, the code might represent
the US national provider ID, US national payer ID, medical record number, and
social security number.
Identifier Valid Time Range : IVL<TS>
The time range during which the identifier is valid. It may be undefined on either
side since in some cases only the start date for ID validity will be known, while, in
others, only the end date is available.
Value Text : ST
The character string value of the identifier. For example the character string "123-
45-6789" for a medical record number.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 72 OF 91 JULY 2000
Datatype: INTEGER : INT
Is a Primitive Datatype
Has Super Types: DSCR
Description of: INT
Integer numbers are precise numbers that are results of counting and enumerating.
The set of integers is infinite but countable. Two special integer values are defined
for positive and negative infinity.
Datatype: INTERVAL : IVL
Is a Generic Datatype
Has Super Types: DSCR
Description of: IVL
Generic datatype that can express a range or interval of values. An interval is a set
of consecutive values of any totally ordered datatype. An interval is thus a
continuous subset of its base datatype.
Datatype: ISO OBJECT IDENTIFIER : OID
Is a Primitive Datatype
Has Super Types: DSCR
Description of: OID
The ISO Object Identifier is defined by ISO.
Datatype: NUMBER : N
Is a Primitive Datatype
Has Super Types: DSCR
Description of: N
The representation of a number. It is used as a generalized type for different
numeric representations.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 73 OF 91 JULY 2000
Datatype: ORDERED : ORD
Is a Primitive Datatype
Has Sub Types: TS
QTY
Description of: ORD
Abstract generalized type that at least contains naturally ordered subsets.
Datatype: ORGANIZATION NAME : ON
Is a Composite Datatype
Description of: ON
A name for an organization, such as "Centers for Disease Control and Prevention".
Components of: ON
Type Code : CV
A code identifying the use for an organization name. Possible values include: L -
Legal, A - Alias, D - Display, ST - Stock Exchange.
Value : ST
The actual name data as a simple character string.
Datatype: PERSON NAME : PN
Is a Composite Datatype
Description of: PN
A Person name is one full name of a person. A name such as "Jim Bob Walton,
Jr." is one instance of a Person name. The parts of this name "Jim", "Bob",
"Walton", and "Jr." are person name parts.
Components of: PN
Value : LIST<PNXP>
This contains the actual name data as a list of name parts that may or may not have
semantic tags.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 74 OF 91 JULY 2000
Datatype: PERSON NAME PART : PNXP
Is a Composite Datatype
Description of: PNXP
This type is not used outside of the Person Name datatype. Person Names are
regarded as token lists. Tokens usually are character strings but may have a tag
that signifies the role of the token. Typical name parts are given names and family
names; other part types may be defined culturally.
Components of: PNXP
Classifiers : SET<CV>
Classifications of a name part. One name part can fall into multiple categories,
such as given name vs. family name and name of public record vs. nickname.
Value : ST
The value of a name part.
Datatype: PHYSICAL QUANTITY : PQ
Is a Composite Datatype
Has Super Types: QTY
Description of: PQ
A physical quantity results from a measurement act. It consists of a value and a
unit.
Components of: PQ
Unit : CV
The unit of measure. Typically this is a unit, such as kilograms or miles per hour,
that is drawn from a table of units of measure. Note that "count" is also included as
a unit. This is used, for example, when collecting information about the number of
interventions of a particular type.
Value : REAL
The magnitude of the quantity measured in terms of the unit.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 75 OF 91 JULY 2000
Datatype: POINT IN TIME : TS
Is a Primitive Datatype
Has Super Types: ORD
Description of: TS
A point in time is a scalar defining a point on the axis of natural time.
Datatype: QUANTITY : QTY
Is a Primitive Datatype
Has Super Types: ORD
Has Sub Types: PQ
Description of: QTY
Abstract generalized type for any quantitative type.
Datatype: REAL : REAL
Is a Composite Datatype
Has Super Types: DSCR
Description of: REAL
A numerical amount. In order to facilitate computer representation, this is, by
assumption, a floating-point number.
Components of: REAL
Precision : N
The precision of the floating point number in terms of the number of significant
decimal digits.
Value : N
The value, expressed as an integer.
Datatype: SET : SET
Is a Generic Datatype
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 76 OF 91 JULY 2000
Description of: SET
SET is an unordered collection of unique items.
Datatype: TELECOMMUNICATION ADDRESS : TEL
Is a Composite Datatype
Description of: TEL
This is a token assigned as a mechanism for locating a telecommunication device
such as a telephone, website, or email address.
Components of: TEL
Address : URI
This is an arbitrary address string that uniquely identifies an address in a particular
domain.
Use Code : SET<CV>
The purpose of the "use code" is to advise in a system or user's selecting an
appropriate telecommunication address to reach a party for a given
telecommunication need. The following mandatory value domain is defined: PR -
primary residence (home) OR - other residence (other home) WP -
work/business/office communication address VR - vacation residence AS -
automated answering service EC - emergency contact BP - beeper/pager CL -
cellular/wireless phone
Valid Time : GTS
This is a General Time Specification (GTS) that identifies the periods of time
during which this telecommunication address can be used. For a telephone number
this can indicate the time of day in which the party can be reached on that
telephone. For a web address, it may specify a time range in which the web
content is promised to be available under the given address
Datatype: UNIVERSAL RESOURCE IDENTIFIER : URI
Is a Primitive Datatype
Description of: URI
The URI is used to refer to addresses of communicating entities used in order to
transmit any kind of information. This may be used for messaging addresses, and
for non-computer communication.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 77 OF 91 JULY 2000
Model Scenario
This section presents an example public health scenario to show how the classes within the
Public Health Conceptual Data Model correspond to data that are encountered over and
over in the course of public health work. The scenario is presented in the left-hand
column, and it is reinterpreted in the right-hand column in terms of concepts from the
PHCDM.
Meningitis Outbreak Scenario
Scenario Text Scenario Text Reinterpreted in
PHCDM Terms
On December 14 the Health Department
for County Z in the northern part of State F
was notified of a 2 year-old girl who had
presented to Hospital H a day earlier with
fever, nausea, vomiting, and a petechial
rash which was suspected to be
meningococcal sepsis.
A Notification, a subtype of Health-
related Activity, was recorded on
December 14.
Two Parties, a Person (the girl) and a
Formal Organization (the hospital), are
related to the Notification. The girl is
related to the Notification as Target
Participant. The hospital is related as
an Actor Participant.
Four Observations (fever, nausea,
vomiting, and petechial rash), which
are a subtype of Health-related
Activity, are linked to the Notification.
An additional Observation (suspected
meningococcal sepsis) is recorded and
linked to the Notification.
On the same day, a blood specimen was
drawn. The specimen that was cultured
from this girl grew Neisseria meningitidis,
confirming the suspected diagnosis.
A Specimen, a subtype of Material,
was drawn.
An Observation (presence of Neisseria
meningitidis) based upon testing the
Specimen leads to a further
Observation (confirmation of the
suspected diagnosis).
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 78 OF 91 JULY 2000
Scenario Text Scenario Text Reinterpreted in
PHCDM Terms
The Local Health Department consulted
with the State Health Department
regarding recommendations for
antimicrobial chemoprophylaxis of close
contacts of sporadic cases of
meningococcal disease.
The recommended Intervention
(antimicrobial chemoprophylaxis), a
subtype of Health-related Activity, is
based upon another Health-related
Activity (consultation) between two
Formal Organizations (the Local and
State Health Departments).
These Health-related Activities are
linked by an Activity Relationship.
Based on this consultation, rifampin or
ceftriaxone, in recommended dosages and
schedules, was administered to members
within the girl’s household and also to
other attendees and staff in the day care
center she attended.
An Intervention (rifampin or
ceftriaxone) is administered to several
Parties that are Informal Organizations
(members of the girl’s household and
other attendees and staff at the day care
center she attended).
The attendees and staff members as
well as the girl herself are related to a
Formal Organization (the day care
center). Each relationship is a Party
Relationship with the day care center.
On December 15, the Health Department
was notified of an 18 year-old female who
had been admitted to the hospital the day
before with fever, headache, and a stiff
neck.
Another Notification was recorded on
December 15.
A Person (18 year-old female) who
was the Target of several Observations
(the history or presence of fever,
headache, and stiff neck) is also the
Target of another Health-related
Activity (hospital admission).
The Observations have Activity
Relationships that link them to the
admission.
Cultures of cerebrospinal fluid grew N.
meningitidis.
An Observation (presence of N.
meningitidis) was recorded after
performing a Health-related Activity
(culturing) on a Specimen
(cerebrospinal fluid).
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 79 OF 91 JULY 2000
Scenario Text Scenario Text Reinterpreted in
PHCDM Terms
Over the next 2 weeks, five more cases
occurred with signs of meningitis.
Several Cases, subtypes of
Observations, were recorded.
In three of those cases, CSF cultures were
positive for N. meningitidis; in the other
two, latex agglutination tests were positive
though cultures were negative. These
cases occurred in persons 4 to 18 yearsold.
In each Case, Health-related Activities
(culturing and testing) on Specimens
(cerebrospinal fluid) were carried out.
Observations related to the tests were
recorded.
The occurrence of multiple cases prompted
discussions among the Local Health
Department, the State Health Department,
and CDC personnel.
Formal Organizations (Local and State
Health Department, CDC) were
involved in consultation and
discussion.
Available N. meningitidis isolates from the
cases with positive cultures were
forwarded to the State F Public Health
Laboratory where they were shown to be
serogroup C.
Specimens (N. meningitidis isolates)
were forwarded to a Formal
Organization (State F Public Health
Laboratory).
At the laboratory, Health-related
Activities (tests) were performed and
generated Observations (the isolates
are serogroup C).
Based on the conclusion that a cluster of
meningococcal diseases due to serogroup
C N. meningitidis was occurring in Towns
A and B, a decision was made to
vaccinate.
Public health authorities confirmed the
existence of an Outbreak (cluster of
meningococcal cases due to serogroup
C N. meningitidis) in a particular
Location (Towns A and B) and
decided on an Intervention
(vaccination).
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 80 OF 91 JULY 2000
Scenario Text Scenario Text Reinterpreted in
PHCDM Terms
Between December 29 and January 1, a
vaccination campaign was initiated
targeting residents of Towns A and B (total
population 33,000) between the ages of 2
and 22. Approximately 13,500 persons
were vaccinated with polyvalent
meningococcal polysaccharide vaccine.
A large scale Intervention (vaccination
campaign) was initiated. The Target
for the Intervention was an Informal
Organization (residents of Towns A
and B between the ages of 2 and 22).
Within this context, many individual
Interventions (vaccination with
polyvalent meningococcal
polysaccharide vaccine) were
performed.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 81 OF 91 JULY 2000
Frequently Asked Questions
1. What is data modeling?
Data modeling is the process of analyzing and representing the things (i.e.,
classes) that an organization must understand. A data model represents the
business facts (i.e., attributes) that the organization must know about the
classes, along with their associated relationships and business rules. Data
modeling requires that business personnel, facilitated by a data architect,
wrestle with and achieve consensus on the definition of what the specific
classes and their associated attributes are.
The purpose of data modeling is to develop an accurate model, or graphical
representation, of the client's information needs and business processes. A
well-developed data model is the architectural blueprint that enables stable and
flexible database and application development. The data model acts as a
framework for the development of new or enhanced applications.
2. What is a conceptual data model?
A conceptual data model is a high-level or abstract data model. It is based on
an analysis of client requirements and it describes the interests of an
organization or business. Conceptual data modeling is one of the most
powerful and effective analytical techniques ever developed for understanding
and organizing the information required to support any enterprise. This form
of model focuses on the big picture, and the really important strategic
objectives that will assure the health and prosperity of the enterprise. Data
subjects (i.e., classes) are shared across functional, process, and organizational
boundaries in the business. As a result, this model (and the applications and
databases which will be built from it) is the linchpin for removing waste and
unnecessary time and cost in the conduct of the enterprise processes by
increasing shared use and avoiding redundancy.
3. What is a logical data model?
A logical data model forms, defines, and standardizes data elements so that
they can be shared for all business purposes, instead of being tailored to fit
only one business unit or agency’s use or point of view. Logical models
ensure that the organization of the data is also optimal for all uses. A logical
data model really a blueprint of the data across an entire organization,
irrespective of platform, operating system, file structure, or database
technology. It defines your entire data landscape.
Logical data modeling is concerned with structuring the data to conform with
established database management principles. The logical data model
represents the actual data requirements that exist on forms and reports of an
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 82 OF 91 JULY 2000
information system. The logical data model for an enterprise is based on its
conceptual data model, but it will include an additional level of detail, and
more rigorous expression of data structures.
4. What is a physical data model?
The physical data model captures the physical constraints imposed by the
business (performance of certain transactions, data volume limitations,
physical distribution of the data at different locations, and security protection
of selected data), and creates a best-balanced physical design of the computer
database which meets those (sometimes conflicting) constraints without
sacrificing the natural subject organization of the data. Each physical model is
based on the logical data model, but takes account of performance
requirements and the special characteristics of the chosen implementation
software.
5. How do the three models fit together?
Once user requirements are gathered, a conceptual model is designed. This
model provides an overview of the subject areas (classes and their
relationships). The next step is to design a logical data model. The logical
model focuses on the data required by the system, independent of how the data
are stored, the database software, and how the data are processed. When the
logical model is complete, a detailed physical model(s) can be developed. The
physical model is the logical model implemented according to the constraints
of the chosen database management system (DBMS) software.
6. Does a conceptual data model tell me how to set up a computerized
system?
No, the conceptual data model expresses the data requirements of the
enterprise independently of enterprise functional requirements (such as for
volume of transactions, response time, and data security), and independently
of the characteristic features of the chosen application and the database
software. However, when the enterprise follows a disciplined application
development strategy, the logical and then the physical data models will be
developed using the conceptual data model as the basis and starting point.
7. If the PHCDM is not a physical data model, then how do I use the
PHCDM in developing my system?
The conceptual model provides a framework for the development of new or
enhanced applications. It is a blueprint that database and software developers
use to design and implement new or enhanced software information systems.
First, the logical data model is developed through a process of refinement from
the conceptual data model, and later, the physical data model is constructed on
the basis of the logical data model.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 83 OF 91 JULY 2000
8. How does the PHCDM relate to data standards?
The PHCDM provides a framework from which we can identify and organize
public health concepts and data standards. Below are some examples of how
the PHCDM can be used to further data standardization efforts:
• Categorize areas to identify where standard code sets are needed, e.g.,
codes for demographic characteristics, test names, or risk factors. These
code sets may be borrowed from existing national standards or, when not
available, may be developed through collaboration between CDC and
public health partners.
• Mappings from existing logical data models (CDC, state, local and other
relevant systems) to the PHCDM can help determine whether the PHCDM
truly reflects and represents the broad spectrum of public health. The
mapping process may result in the identification of common concepts
between data models (as well as the identification of concepts requiring
standardization that are not represented in the PHCDM.) Common
concepts may include general demographic, case management, and
behavioral risk factor information (e.g., smoking, IV drug use, and
unprotected sex).
• The PHCDM can serve as a springboard for discussion with standards
development organizations (SDOs). Presenting the PHCDM to SDOs is
one way of ensuring public health needs are considered in future revisions
of national standards. If the SDO has a data model (e.g., HL7 Reference
Information Model), comparisons of relevant areas of the PHCDM to the
SDO model may provide an opportunity for further harmonization of data
concepts.
A CDC standards body will be involved in such data standards activities. The
PHCDM will help facilitate the iterative process of identifying subject areas
and data concepts requiring standardization and disseminating data standards
as they are adopted. Collaboration with CDC staff, public health partners, and
standards development organizations is necessary to resolve differences in the
way concepts are defined and represented from each perspective. Once these
differences are resolved, and a common concept(s) can be agreed upon, use of
the newly standardized concept(s) in subsequent iterations of the model(s) and
existing/future system design will allow for the exchange of this information
in a more uniform, consistent way.
9. How does this model facilitate data exchange?
The conceptual model, when it is shared across application and organizational
boundaries, expresses the data that are common, and that will be shared. In
effect, it provides a standard for data definitions and representation.
Applications and databases built on basis of the conceptual model need to
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 84 OF 91 JULY 2000
store the data in a form consistent with that model or support a mapping to that
format, in order to support data exchange.
10. Who should be the users of this model?
Parties within public health who need to use or manage public health data
should utilize the model. This includes subject matter experts, end-users that
analyze data and create reports, and application and database developers.
11. What is the role of the data model in the National Electronic Disease
Surveillance System (NEDSS)?
The National Electronic Disease Surveillance System (NEDSS) is a set of
related activities designed to electronically integrate and link together the
myriad information systems currently used for public health surveillance. The
goal of NEDSS is to facilitate more efficient, accurate, and timely collection,
interpretation, and use of data by communities, Local and State Health
Departments, and CDC to improve the health of the public. Other NEDSS
activities include an information systems architecture based on industry
standards, including a standard user interface for CDC surveillance
information systems; shareable tools for interpretation, analysis, and
dissemination of data; a network for secure internet-based data transfer that is
also based on industry standards. Because data are vital to doing our jobs in
public health, the PHCDM is a critical element of NEDSS; it documents our
information needs in public health, providing a framework for organizing data
standards and guidelines and facilitating data comparability and exchange with
other systems. For example, we are using the PHCDM as a tool to
communicate public health data needs to national health informatics standards
setting bodies, to enable development of standards for the exchange of
information among public health and healthcare providers.
12. What is the scope of this model? Does it include all public health
activities?
At this point the primary focus of this model is to represent data needs for the
key “Essential Public Health Functions” (as defined by the United States
Public Health Functions Steering Committee in 1994, available at
http://www.health.gov/phfunctions/project.htm) of “Monitor health status to
identify community health problems” (including public health surveillance)
and “Diagnose and investigate health problems and health hazards in the
community”. We are developing a process model to help explain the scope of
the model and to relate the concepts included here directly with specific public
health activities.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 85 OF 91 JULY 2000
13. What is a process model, and how does a data model differ from a
process model?
A process model is a model of the activities, functions, and processes of an
organization. Processes in a process model are often defined in terms of their
inputs and outputs. Because of this, a data model that provides definition for
the data consumed and produced by a process often accompanies a process
model. A data model does not reflect any action or flow of information. Only
a static view of data is presented in a data model.
14. Does this represent ALL of public health, i.e., local, state, and federal
issues?
It is hoped that this model will be broadly representative of public health
issues at all levels. The data model used in Missouri was used as input to the
development of this model, and we have held a meeting and requested
feedback from our partners in State and Local Health Departments. However,
the purpose of publishing the model is to ensure its use and critique by our
partners in community organizations, and local, state, and federal agencies.
We are requesting your feedback, to ensure the continued growth and
improvement of this model, and to ensure that it truly represents all of public
health’s data needs. For example, due to a relative lack of focus at the CDC
level on direct provision of healthcare, contact follow-up, or issues with
patient insurance coverage, this model may incompletely represent data needs
associated with financial management of patients, or follow-up of contacts of a
case of communicable disease. Please provide specific suggestions for
possible inclusion in the model.
15. Does this model include subject areas and classes related to both
infectious and non-infectious disease?
This model is generally intended to represent data needs in public health. It
should be noted that concepts included in the model, such as individuals,
locations, risk factors, and interventions, are generic and important to many
kinds of public health activities. We have tried to be inclusive in our thinking
as we developed this model, although participants in model development have
included mostly colleagues with backgrounds in infectious diseases.
However, there has been participation from our colleagues in other areas, such
as injury prevention and environmental health, and we hope through
publication of the model to obtain specific suggestions on how to facilitate its
representation of data needs for the broad spectrum of public health issues.
16. How should State or Local Health Departments use this model?
State or Local Health Departments that already have data models should
compare their models to the PHCDM in order to determine what changes may
be needed to ensure that the PHCDM truly represents public health at all
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 86 OF 91 JULY 2000
levels. If a State or Local Health Department does not yet have a data model
underlying its information system design and development, it should consider
evaluating and adopting the PHCDM.
17. How will the PHCDM be maintained?
A team at CDC consisting of subject matter experts, data modelers, and
technical specialists will maintain the PHCDM. As the model is applied to
public health initiatives such as system development, database design,
interface specifications, and other data standardization initiatives areas where
the model is weak will be reviewed and attempts will be made to enhance the
model to make it useful.
CDC is developing a process to receive comments from its partners
concerning the data model. These comments will be used to identify areas of
the model that need refinement or extension. For example, in the early
versions of the PHCDM, there was a class entitled, "patient coverage", which
referred to the health insurance that would (if it existed) pay for the patient's
clinical care. However, it was determined that there were not enough systems
at CDC that required this information; thus, we had no data to use to flesh out
this class (with its attributes, for example). If our public health partners feel
that the PHCDM should include this class, we would welcome their
suggestions on its attributes and associations with other classes.
The model will be continuously maintained, but published no more frequently
than once per year.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 87 OF 91 JULY 2000
Glossary
American National Standards Institute (ANSI): A voluntary standards
organization that serves as the coordinator for national standards in the United
States and the U.S. member body to the International Organization for
Standards. ANSI accredits standards committees and provides an open forum
for interested parties to identify, plan, and agree on standards; it does not itself
develop standards. Standards are developed by Standards Development
Organizations (SDOs).
URL: www.ansi.org
Association: An association is a structural relationship that specifies that instances
of one thing are connected to instances of another.
Attribute: The specific items of data that can be collected for a class.
Centers for Disease Control and Prevention (CDC): An agency of the
Department of Health and Human Services that promotes health and quality of
life by preventing and controlling disease, injury, and disability.
URL: www.cdc.gov
Common Information for Public Health Electronic Reporting (CIPHER): A
set of standards and guidelines for data representation and code values which
includes specifications for representing concepts as well as standard code lists
for coded elements. The CIPHER standards can be linked directly to
attributes in the PHCDM.
URL: http://www.cdc.gov/od/hissb/docs/cipher.htm
Class: A description of a set of objects that share the same attributes, relationships,
and semantics.
Data Model: A framework for the development of a new or enhanced application.
The purpose of data modeling is to develop an accurate model, or graphical
representation, of the client's information needs and business processes.
Datatype: A specification of the allowed format for the values of an attribute.
Examples include string, number, code, and text.
Electronic Data Interchange (EDI): A standard format for exchanging business
data. An EDI message contains a string of data elements, each of which
represents a singular fact, such as a price, product model number, and so forth,
separated by delimiters (a character that identifies the beginning and end of a
character string). The entire string is called a data segment. EDI is one form of
e-commerce, which also includes e-mail and fax.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 88 OF 91 JULY 2000
HIV/AIDS Reporting System (HARS): A computerized management
information system developed to assist State and Local Health Departments in
managing and analyzing HIV/AIDS surveillance data.
URL: http://www.cdc.gov/hiv/software/hars.htm
Health Information and Surveillance Systems Board (HISSB): A board
established by the authority of the Director of CDC/ATSDR. The mission of
HISSB is the formulation and enactment of policy concerning the planning,
development, maintenance, and use of public health information and
surveillance systems.
URL: http://www.cdc.gov/od/hissb/
Health Level 7 (HL7): A standards development organization formed in 1987 to
produce a standard for hospital information systems. HL7 received ANSI
accreditation as an Accredited Standards Developing Organization in 1994.
The HL7 standard is an American National Standard for electronic data
exchange in health care that enables disparate computer applications to
exchange key sets of clinical and administrative information. HL7 is primarily
concerned with movement within institutions of orders; clinical observations
and data, including test results; admission, transfer and discharge records; and
charge and billing information (coordinating here with ASC X12). HL7 is the
selected standard for the interfacing of clinical data for most health care
institutions.
URL: www.hl7.org
HL7 Reference Information Model (HL7 RIM): A conceptual model that
defines all the information from which the data content of HL7 messages is
drawn.
URL: www.hl7.org
International Organization for Standardization (ISO): A worldwide
federation of national standards bodies from some 100 countries, one from
each country. Among the standards it fosters is Open Systems Interconnection
(OSI), a universal reference model for communication protocols. Many
countries have national standards organizations, such as the U.S. American
National Standards Institute (ANSI), that participate in and contribute to ISO
standards development.
URL: www.iso.org
Laboratory and Epidemiological Public Health Information Tracking and
Reporting System (LITS+): A computerized information management
system that provides seamless integration of laboratory and epidemiologic
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 89 OF 91 JULY 2000
data. The system was developed by CDC and is in use in many public health
and reference laboratories.
URL: http://www.cdc.gov/ncidod/dbmd/litsplus/newsletters/fall99.pdf
National Electronic Disease Surveillance System (NEDSS): A set of related
activities designed to electronically integrate and link together the myriad
information systems currently used for public health surveillance. The goal of
NEDSS is to facilitate more efficient, accurate, and timely collection,
interpretation, and use of data by State and Local Health Departments and
CDC to improve the health of the public. When complete, NEDSS will
include data standards; tools for interpretation, analysis, and dissemination of
data; a network for secure internet-based data transfer that uses industry
standards; and policy-level agreements on data access, sharing, burden
reduction, and protection of confidentiality.
URL: http://www.cdc.gov/od/hissb/
National Electronic Telecommunications System for Surveillance (NETSS):
A computerized public health surveillance information system that uses a
standard ASCII record format to provide CDC with weekly data regarding
cases of nationally notifiable diseases.
URL: http://www.cdc.gov/epo/dphsi/netss.htm
Public Health Conceptual Data Model (PHCDM): A high level conceptual
model, developed as part of the CDC NEDSS initiative, which provides the
foundation for standardization of public health data collection, management,
transmission, analysis and dissemination.
Sexually Transmitted Diseases Management Information System
(STD*MIS): A management information system designed to assist state and
local STD control programs in managing and operating their programs. The
system allows programs to maintain surveillance data on both laboratory and
case reports, allows tracking and monitoring of disease intervention activities
such as patient interview and follow up, and allows capture of data regarding
patient clinic visits and outcomes.
URL: http://www.cdc.gov/nchstp/dstd/STD-MIS.htm
Subject Area: A way of organizing classes into groups within a model, where
classes grouped together into higher-level units. Within the UML, a subject
area is referred to as a package.
Subtype: A specialization of another class, which inherits the attributes of its
parent class.
Supertype: A generalized class that is related to subtypes that inherit its attributes.
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 90 OF 91 JULY 2000
Tuberculosis Information Management System (TIMS): A software program
that automates the administration of tuberculosis prevention, surveillance, and
control programs. TIMS facilitates the management of TB cases and the
tracking and reporting of TB control program activities. Data collection and
transcription on all TB clients can be performed daily at the service delivery
level at one or more facilities.
URL: http://www.cdc.gov/nchstp/tb/tims/tims.htm
Unified Modeling Language (UML): A graphical language for visualizing,
specifying, constructing, and documenting the artifacts of a software-intensive
system.
URL:http://www.omg.org/uml/
World Wide Web Consortium (W3C): An industry consortium that seeks to
promote standards for the evolution of the Web and interoperability between
WWW products by producing specifications and reference software.
URL:http://www.w3.org/
X12: A standards development organization that develops uniform standards for
inter-industry electronic interchange of business transactions-- electronic data
interchange (EDI). X12N, a subcommittee of X12, develops standards for
healthcare insurance and claims processing.
URL: http://www.disa.org/
eXtensible Markup Language (XML): A specification developed by the World
Wide Web Consortium. XML is designed especially for Web documents. It
allows designers to create their own customized tags, enabling the definition,
transmission, validation, and interpretation of data between applications and
between organizations. XML provides a file format for representing data, a
schema for describing data structure, and a mechanism for extending and
annotating HTML with semantic information.
URL: http://www.w3.org/
PUBLIC HEALTH CONCEPTUAL DATA MODEL
PREMIERE EDITION PA GE 91 OF 91 JULY 2000
Bibliography
Booch G, Rumbaugh J, Jacobson I. The Unified Modeling Language User Guide.
Reading, MA: Addison-Wesley, 1999.
Institute of Medicine. The Future of Public Health. Washington, DC: National
Academy Press, 1988.
Fowler M.. UML Distilled. Reading, MA: Addison-Wesley, 1997.
Health Level Seven, Modeling & Methodology Technical Committee. HL7
Message Development Framework. 1999.
Health Level Seven, Modeling & Methodology Technical Committee. HL7
Reference Information Model. Version 0.96. 2000.
Centers for Disease Control and Prevention. Integrating Public Health
Information and Surveillance Systems: A Report and Recommendations from
the CDC/ATSDR Steering Committee on Public Health Information and
Surveillance System Development. 1995.
URL: http://www.cdc.gov/od/hissb/docs/katz.htm