The Changing Face of Data Quality Within VHA

1
The Changing Face of Data Quality Within VHA

• Presented by
• Elizabeth Franchi, Director, HDI, Data Quality
  Program
• Sara Temlitz, Business Product Manager, HDI, Data
  Quality Program
• Jackie Houston, Identity Management Data Quality
  (IMDQ) Program Manager

2
Topics of Discussion

• Provide an overview of the Data Quality Program
• Review the existing and new programs within the
  Data Quality Program
• Discuss the overall structure and framework being
  developed to provide business benefits of a
  formal Data Quality Program throughout VHA
  enterprise

3

• Data Quality Program Overview

4
DQ Program Mission

• Mission
• Manage and implement a data quality framework to
  improve and ensure comprehensive and accurate VHA
  data for clinicians, administrators, veterans,
  sharing partners and business stakeholders.

5
DQ Program Overview

• The Veterans Health Administration (VHA) Data
  Quality Program
• Is part of the Office of Information (OI) Health
  Data and Informatics Office (HDI) and
• Provides, facilitates and supports strategic
  direction for data management
• Provides VHA with operational support to improve
  the quality of data required to provide and
  manage health care information
• Includes several key components to help formalize
  and establish accountability for management of
  VHA data

6
DQ Program Guiding Principles

• Technical and business strategy are connected
• Technology plays a supportive role in the data
  quality process
• Data governance and stewardship programs
  formalize accountability and management
  responsibilities for data
• Management of data is a method to achieve the
  data quality needed to support the delivery of
  health care.

• Information is an enterprise-wide business asset
  and resource
• Decision-making and business planning are based
  on information which is derived from data

7
Data Quality Roles Responsibilities

• VHAs data steward for Patient Identity Data
  ensuring the availability of the longitudinal
  health record across our enterprise and with
  other business partners, like DoD
• Support VHA EHR integrity within the MPI and
  other applications
• Subject matter expert for selected national VHA
  reference tables
• Serves as key member of domain action teams and
  other VHA committees

• Develop and manage programs for continuous data
  quality improvements
• Provide guidance and consultation to Lines of
  Business
• Manage data quality activities in support of
  VHAs business

8
OI HDI Data Quality Program Overview
9
DQ Program Structure
10
Data Quality Functional Areas
11
DQ Program Business Benefits

Reduction of inconsistent, inaccurate incomplete
and/or untimely data to provide higher quality
data that better meets business needs
Coordinated data quality efforts within VHA
collaboration and participation in enterprise
efforts
Authoritative and participative data governance
structure with business community involvement and
Common data quality principles and best practices
including resolution processes.
12
VHA Planning Strategy
Data Quality Goal
13
VHA Planning Strategy
Data Quality Goal
14
DQ Strategies Flow from Goals
DQ 1 Improve the comprehensiveness, timeliness,
consistency, and accuracy of VHA data for
clinicians, researchers, administrators,
veterans, sharing partners and lines of
business.
1.1.1.3 Work with management And decision bodies
to establish And formalize enterprise-wide DQ
responsibilities.
1.1 Support excellence in business practices
through a comprehensive VHA Data Quality Program.
1.1.1.4 Define governance communication and
prioritization process.
1.1.1 Establish and maintain DQ governance
structure including stakeholders,
roles, responsibilities, etc.
1.1.1.5 Establish a mechanism to collect
data quality analysis and improvement candidates
for each domain/system.
1.1.1.1 Implement and maintain VHA Data
Stewardship Program.
1.1.1.2 Identify and define roles and
responsibilities of DQ stakeholders Including
field reps.
15
Systematic Activities support a Proactive
Approach to Managing DQ
DQ 3 Maintain and ensure the integrity
of Patient identity for health care
delivery, management and patient safety (PS).
Facilitate identification/resolution of duplicate
patient records, catastrophic edits and
other identity-related anomalies.
Maintain and improve formal Identity Management
(IdM) practices
Participate as IdM Stewards in MPI, Person
Service, ADR and Toolkit UAT.
Provide IdM stewardship input to Identity
Management requirements And Use Cases.
Participate in Patient Safety incident reviews
related to IdM.
Participate in retrospective review of PS
incidents to identify systemic DQ issues..
16

• Data Quality Program Activities

From the mission - Providing comprehensive and
accurate VHA data for clinicians, administrators,
veterans, sharing partners and business
stakeholders.
17
VHAs Electronic Health Record

• VHA is recognized as a leader in the world of
  electronic health records
• VHAs integrated health information system,
  including its framework for using performance
  measures to improve quality, is considered one of
  the best in the nation.
• Institute of Medicine (IOM) Report,
  Leadership by Example Coordinating Government
  Roles in Improving Health Care Quality (2002)

18
VHAs Electronic Health Record features

• More than 100 separate applications that support
  day-to-day activities of healthcare operations,
  including
• Registration/enrollment/eligibility
• Provider systems and
• Management and financial systems
• Through the Computerized Patient Record System
  (CPRS)
• Delivers an integrated electronic health record
  covering all aspects of patient care and
  treatment and
• Provides access to remote data through Remote
  Data Views, Inter-facility Consults, VistAWeb

19
Business Benefits of the electronic health record

• Improves quality and safety of healthcare to
  veterans
• Seamlessly provides health records and images
  when needed/authorized and
• Enhances interoperability among VA and other
  business stakeholders
• VHA implemented an electronic Master Patient
  Index (MPI) to facilitate the sharing of a
  patient centered longitudinal health record
  within VHA and with external business partners
  such as DoD

20
VHA Data Quality Efforts

• VHAs Master Patient Index
• Uniquely identifies patients
• Assigns a unique identifier (the Integration
  Control Number or ICN)
• Holds roughly 15 million unique patients
• Part of data quality efforts are to
• Ensure the availability of the electronic health
  record for patient care and related activities
• Manage the unique identifiers and patient
  identity traits
• Maintain record correlations across VHA systems
  like VistA, and
• Enable external linkages with DoD and other
  business partners

21
DQ Program Clinical Data Quality
Develops clinical data quality guidance and
operating policies for VHA. Establishes and
maintains mechanism to identify, resolve and
monitor clinical data quality opportunities.
22
Clinical Data Quality Program

• Emerging program within Data Quality
• Provides framework to identify and support
  activities to resolve clinical data quality
  issues within VHA
• Collaborate in the development of overall data
  quality metrics and goals for prioritized data
  quality opportunities
• Identify problem areas for audit and analysis for
  data verification and correction
• Monitor and resolve data quality integrity
  issues, conduct studies and other analyses

23
DQ Program Program Analysis
Review, monitor, and resolve data quality issues.
Performs data quality analysis on existing
administrative and healthcare data throughout VHA.
24
Program Analysis

• Program Objectives
• Provide support to program areas within DQ
• Assist in the development of overall data quality
  metrics and goals for prioritized data quality
  opportunities
• Provide audit and analysis for data verification
  and correction
• Monitor and resolve data quality integrity
  issues, conduct studies and other analysis

25
What is Data Quality Analytics?

• Applies techniques to
• Analyze the quality of data,
• Determine its ability to be used for its intended
  purpose, and
• Recommend corrective action when needed
• Example techniques include
• Data profiling
• Process assessments, e.g.,
• Healthcare Failure Mode and Effect Analysis
  (HFMEAs)
• Root Cause Analysis (RCAs)

26
What is Data Profiling?

• Assessment of data
• Content/values within databases
• The datas defined characteristics (or metadata),
  e.g., data type (a date field must include month,
  day and year)
• Profiling activity may be initiated
• To analyze a reported issue (assertion testing)
• To proactively discover data characteristics and
  anomalies

27
Data Profiling Benefits

• Enables a better understanding of the datas
  quality and ability to be used for its intended
  business purpose
• Identifies problematic areas and enables
  recommendations to be developed and implemented
  to improve inconsistent, inaccurate, and/or
  incomplete data
• Enables data descriptions and business rules to
  be verified and improved

28
Recent DQ Analytics Activity

• Profiling patient identity traits (e.g., Mothers
  Maiden Name)
• Comparing metadata (dictionaries, models, Primary
  View)
• Running queries against Corporate Data Warehouse
  (CDW) data from the VistA Patient file to
• Determine if the data meets the metadata
  description
• Identify other data anomalies
• Results will be used to improve Primary View
  business rules, e.g, Identification of additional
  blank equivalents for Mothers Maiden Name

29
Recent DQ Analytics Activity

• Initiating a HFMEA for Date of Death
• Erroneous Dates of Death cause cancellation of
  medications, future appointments, etc.
• Incorrect Dates of Death lead to data integrity
  errors, such as
• Vitals exist for after DOD
• Encounters exist after DOD
• Omission of Dates of Death cause business
  processes to continue, such as
• Appointment letters sent to deceased patients
• Medications delivered for deceased patients
• Appointment slots not available for waiting
  patients

30
DQ Program Identity Management
Responsible for integrity of patient identity
data within Master Patient Index (MPI). The
Identity Management Data Quality (IMDQ) team
supports VHA site personnel and is the primary
liaison with the MPI.
31
Identity Management Stewardship

• The Identity Management Stewardship Program is
  responsible for the business stewardship of
  identity data as part of the electronic health
  record within VHA. A major component of this
  program is the Identity Management Data Quality
  (IMDQ) team, which provides specific stewardship
  for the data required to link patient identities
  to their related information in VHA and other
  systems. The team is responsible for the data
  integrity and business analytics support of VHAs
  Master Patient Index (MPI), the unique identifier
  system for all patient records within VHA.

32
IMDQ Team Responsibilities

• Defines business rules and processes governing
  identity management to ensure accuracy and
  integrity of VHAs longitudinal health record
• Monitors and resolves data integrity issues and
  conflicts on the MPI and local VistA systems
• Resolution of duplicates
• Mismatches and catastrophic edits
• Ensure the quality of data
• Improve process and method of entering patient
  record data
• Facilitate information dissemination and training
  
• Improve the understanding of the MPI and identity
  management and its direct relation to the
  patient record

33
Identity Management Activities

• Management of Catastrophic Edits and other data
  quality issues
• Support of Master Patient Index (MPI) development
• Enhancements to existing software, including
  Primary View Implementation
• Re-hosting to Administrative Data Repository
  (ADR)
• Assistance for implementation of Person Service
  Identity Management (PSIM)
• IMDQ Toolkit development
• Non-person enumeration
• Enrollment System Redesign implementation
• Other HeV application development support

34
Catastrophic Edits

• Common Causes of a Catastrophic Edit
• Erroneous Patient Selections
• Patient Records Used as Templates
• Erroneous or Catastrophic Duplicate Record Merges
• Recycling of old patient records

Catastrophic Edit
35
Catastrophic Edits to Identity

• Effects of Catastrophic Edits
• Incorrect association of patient medical record
  data to a different patients record
• Intermingling of two different patients records
• Actions taken to reduce Catastrophic Edits
• Changes to VistA software to warn of potential
  catastrophic edits
• Distribution of notifications of occurrences to
  local management
• Tracking of resolution to ensure medical records
  are restored
• National staff within the Identity Management
  Data Quality group monitor alerts about potential
  problems daily
• Patient safety risk is high
• Treatment may be administered to wrong patient
• May not have complete or accurate information
  about patients, including allergies, current
  medications
• Electronic health record may become fragmented or
  have conflicting data

36
DQ Program Data Stewardship
Establishes and formalizes accountability for the
characteristics and management of VHA data and
ensures business stakeholders fully participate
in decision-making regarding data essential to
them.
37
Data Stewardship Objectives

• Develop an effective data stewardship program for
  the organization, leveraging existing activities
• Coordinate/support business and technical
  stewardship efforts for VHA business community
• Work with Line-of-Business (LOB) Data Stewards
  and subject matter experts and ensure that they
  have necessary knowledge, training, and support
  to fulfill stewardship responsibilities
• Lead/coordinate stewardship activities for the
  domains defined in enterprise information models
  and by LOB managers and
• Collaborate with DoD, Indian Health Service,
  business partners, VA technical teams and
  external data architecture and standards
  organizations.
• Emerging program not yet fully staffed

38
DQ Program Business Products
Ensures business stakeholder DQ requirements are
identified and communicated Liaison for data
quality issues between business groups, technical
communities and the DQ program.
39
Business Product Management

• The Business Product Management Program ensures
  that business stakeholder data quality
  requirements are identified and communicated
  through appropriate processes, and monitors
  progress to ensure business needs are met. This
  Program serves as a liaison between business
  stakeholder groups, technical communities and the
  Data Quality Program in issues related to data
  quality, including those involving identity
  management.

40
Business Product Management Program

• Provides expertise to business stewards on
• Business requirements development,
• Enterprise requirements management process, and
• IT development process
• Works with stewards and other stakeholders to
• Define detailed business requirements
  specifications for both legacy VistA and
  HealtheVet applications
• Develop and obtain approval of overarching data
  quality requirements
• Collaborate with ESMs, architects, developers and
  other technical staff to develop formal business
  models and technical specifications
• Participate in design validation and user
  acceptance testing

41
Enterprise DQ Requirements

• Develop Requirements for Enterprise Repository
• Integrate DQ requirements into the development
  lifecycle
• Inception
• Elaboration
• Construction
• Transition
• Provide guidance and support to development and
  program staff

42
Summary

• The VHA Data Quality Program continues to develop
  and expand its role in improving data for all
  aspect of healthcare within VHA and with other
  business partners
• High quality data is essential for the provision
  of quality healthcare for our patients and for
  other business needs
• Collaboration with others within our organization
  is critical in ensuring data quality

43
Resources

• Data Quality
• http//vaww.vhaco.va.gov/dataquality
• Identity Management Data Quality
• http//vista.med.va.gov/mpi_dqmt/
• Administrative Data Quality Council
• http//vaww.vhaco.va.gov/dataquality/adq.htm
• Data Consortium
• http//datacon.vssc.med.va.gov/default.aspx

44
Contacts

• Beth Franchi, Director, VHA Data Quality
  Program(414) 389-4191Elizabeth.Franchi2_at_va.gov
  Sara Temlitz, Business Product Manager(414)
  389-4192Sara.Temlitz_at_va.govJackie Houston,
  Identity Management Data Quality Program
  Manager(205) 554-3449Jackie.Houston_at_va.gov