Chapter 4. Measure Mapping
Care Coordination Measures Atlas
Measure Mapping Table
To lay out information about the care coordination measurement landscape in two dimensions, a Measure Mapping Table was developed to show the intersection of care coordination domains and measurement perspectives. Measures were indexed, or "mapped," according to the care coordination domains included in the care coordination measurement framework (go to Figure 2), in order to indicate which aspects of care coordination a particular instrument measures. This measure indexing, or mapping, serves two main purposes:
- It provides an overview of the current care coordination measurement field, highlighting areas with many available measures and those with few measures.
- It allows Atlas users to quickly narrow the field of available care coordination measures, homing in on those that assess aspects of care coordination of particular interest to the user.
Measures relevant to care coordination that are included in this Atlas were mapped using the Measure Mapping Table (go to Table 4). The table is structured to simultaneously categorize measures by perspective—patient/family, health care professional(s), or system representative(s)—and by care coordination domain (specific care coordination activities and broad approaches). The perspective (seen in the columns of the table) reflects how care coordination is measured: who is providing the information (e.g., patients, primary care provider, chart review), what data are used (e.g., patient satisfaction survey scores, medical record information, administrative data), and how data are aggregated during analyses (e.g., by patient, by physician group, by payer, etc.). The domains reflect the specific components of care coordination that are addressed by each measure, or individual items within the measure. The Definitions of Care Coordination Domains were used to guide measure mapping.
A filled square (■) indicates that the measure contains 3 or more individual items that pertain to that domain. Composite measures or summary scores are also indicated with a filled square. An open square (□) indicates that the measure contains only 1 or 2 items that relate to a domain. This allows users to quickly assess the degree to which each measure focuses on a particular domain of care coordination, as well as the burden of data collection (i.e., number of items) related to the specific domain. Individual questions or items (measure components) within a measure may map to more than one domain. In addition, a single measure, or measure component, may address only one aspect of a particular domain. We mapped a measure to a domain if it addressed any aspect of the domain definition. For more detailed information on measure mapping, please refer to Appendix I: Measure Mapping Strategy.
Measure profiles follow each individual measure mapping table and contain more detailed information on the measure (go to Figure 3) for an explanation of what information is included). Decisions regarding the types of information to include were based on input from the advisory groups (go to Appendix III: Advisory Group Participants). Relevant information for each section of the profile was obtained and extracted from publications identified through a detailed literature search (go to Appendix II: Identifying Measures). The measure profiles also identify the specific measure items (i.e., survey questions or measure components) that map to each domain. Copies of the measure instruments will be collected in Appendix IV: Care Coordination Measures, currently under development. Appendix IV will be updated regularly.
Special Caution. Many measures included in the Atlas are survey instruments. Users are cautioned that even though individual items from surveys are mapped to particular domains, most instruments should be used in their entirety. Typically, measure testing is conducted on the entire measure; performance of measurement based on individual items is usually unknown. It may be possible to seek advice directly from a measure developer about any potential adaptations.
Table 4. Care Coordination Measure Mapping Table
|Care Coordination Activities|
|Establish accountability or negotiate responsibility|
|As coordination needs change|
|Assess needs and goals|
|Create a proactive plan of care|
|Monitor, follow up, and respond to change|
|Support self-management goals|
|Link to community resources|
|Align resources with patient and population needs|
|Broad Approaches Potentially Related to Care Coordination|
|Teamwork focused on coordination|
|Health care home|
|Health IT-enabled coordination|
Figure 3. Measure Profile Template
Title of Measure
Purpose. A short statement defining the main objective or goal of the measure.
Format/Data Source. A description of the type of instrument(s). If applicable, specific information is noted regarding the number of individual items and the domains, categories, or subtopics covered.
Date. The date the measure was published or released.
Perspective. The perspective—Patient/Family, Health Care Professional(s), or System Representative(s)—being measured.
Measure Item Mapping. A list of which measure items map to which domains. Measure items are typically survey questions or instrument components. For domain definitions, refer to the Definitions section. For the Communicate domain and its subdomains (Interpersonal Communication and Information Transfer), we also provide information on the participants involved in the communication (e.g., communication between health care professional(s) and patient/family; within teams of health care professionals; and across health care teams or settings).
Development and Testing. A summary of relevant information concerning the development of the measure, as well as reliability and validity testing. Measure developers were contacted to seek updated testing information.
Link to Outcomes or Health System Characteristics. A summary of results that link the measure to patient outcomes or health system characteristics.
Logic Model/Conceptual Framework. A brief description of a model, framework, or design if utilized in the development of the measure.
Past or Validated Applications
Source(s). List of relevant sources for the measure, and its development or testing.
Measure Selection Guide
This section of the chapter is intended to help users identify existing care coordination measures that can potentially be used to evaluate their care coordination interventions or demonstration projects.
Identify the measures relevant to your intervention.
Identifying the measures relevant to your intervention study involves several steps outlined below.
- Specify mechanisms of achieving care coordination.
- Find relevant domains on measure mapping table.
- Consider perspective(s) of interest.
- Identify relevant care coordination measures.
Review relevant measure profiles.
Once you have identified the relevant measures, go to the Measure Profiles section to examine the relevant measures in more detail and determine which may meet your evaluation needs.
This section augments the brief outline above with more detailed guidance on how to use the Atlas, including example material (in blue).
1. Identify the measures relevant to your intervention.
Step a. Specify the relevant mechanisms that your intervention will utilize to achieve its care coordination goals. Then identify the corresponding care coordination domain(s) (go to Domain Definitions).
A single intervention may employ multiple mechanisms so you will want to map each one individually to all applicable domains. Repeat this step for each mechanism, and highlight applicable rows on the measure mapping table. Keep in mind that a single mechanism may correspond to multiple domains.
Dr. Smith designed a program to improve post-discharge health outcomes for patients with congestive heart failure and to reduce readmissions related to CHF. The program aims to achieve this by actively facilitating the transition from the inpatient to outpatient setting, using a patient-centered case management approach to facilitate care during this transition. The study protocol includes activities such as: specially trained nurse case manager develops a care plan with the patient prior to discharge using a computerized protocol; a 30-minute patient education session with a nurse on the day of discharge to go over the patient care plan including 'red flags' (specific situations and the specific actions needed); faxing a complete medical record from the hospital, including test results, to the primary care provider within 48 hours of discharge; a followup phone call from a nurse to the patient within the first 7 days after discharge to assess the patient and trigger further followup as required. This program will be implemented at a single community hospital for 6 months. All patients admitted for CHF will be invited to participate.
Intervention mechanism: Facilitate transition from inpatient to outpatient setting → Domain: Facilitate Transitions Across Settings.
Intervention mechanism: The program uses a case management approach and a designated case manager → Domain: Care Management.
Intervention mechanism: Through red-flag discussion, help educate patient about how they can best react to changing symptoms → Domain: Interpersonal Communication; Support Self-Management Goals.
Intervention mechanism: Develop a care plan with the patient prior to discharge, using a computerized protocol → Domain: Create a Proactive Plan of Care.
Intervention mechanism: 30-minute patient education session with nurse on the day of discharge to go over patient care plan → Domain: Support Self-Management Goals; Create a Proactive Plan of Care.
Intervention mechanism: Faxing complete medical record from hospital stay, including test results, to primary care provider within 48 hours of discharge → Domain: Information Transfer; Monitor, Follow Up, and Respond to Change.
Intervention mechanism: Followup telephone call from nurse within the first 7 days after discharge → Domain: Monitor, Follow Up, and Respond to Change.
From the care coordination domains listed in the top rows on the left-hand side of the tables, find the domain(s) you selected and highlight across the row.
For the example listed above, find and highlight the rows for Facilitate Transitions (across settings); Care Management; Information Transfer; Interpersonal Communication; Monitor, Follow Up, and Respond to Change; Create a Proactive Plan of Care; Support Self-Management Goals.
Step c. Consider perspective(s) of interest.
Who is the intervention primarily targeted towards? Who will carry out the intervention? Which perspective are you most interested in assessing? Measurement from any of the three perspectives listed in the measure mapping tables may be relevant— Patient/Family, Health Care Professional(s), or System Representative(s). For example, an intervention that includes a patient education mechanism will certainly merit evaluation from a Patient/ Family perspective. But it may also be useful to assess it from a Health Care Professional(s), or System Representative(s) perspective, depending on the goals of the intervention. Thorough evaluation may require looking at your intervention from multiple perspectives. There are three Master Measure Mapping Tables, one for each measurement perspective (go to Tables 5, 6, and 7).
I am most interested in understanding the effects of this program on patients with CHF.
Perspective: Patients/Family → Specify population: CHF patients.
Step d. Identify relevant care coordination measures.
Using the Master Measure Mapping Tables (go to Tables 5, 6, and 7), look at the measures available that correspond to the intersections of interest (boxes in the grid) based on the previous steps (domains and perspectives). For example, if you wish to evaluate Information Transfer from the Patient/Family perspective, find the Patient/Family perspective column and scan down until you connect to the Information Transfer row. The box that connects these columns and rows lists the existing measures in the Atlas that evaluate information transfer from a patient or family perspective.
Note that interventions could have multiple mechanisms and perspectives and so you will need to do Steps a through d for each combination to identify all the relevant existing measures. Also, note that for some combinations, there may not be an existing care coordination measure included in the Atlas.
Go to the Master Measure Mapping Table for the Patient/Family perspective and look across the Care Management domain row. The numbers in this box correspond to related measures that may be of use in evaluating this intervention. For this example, the measures addressing care management from the patient/family perspective are: 11a, 14, 21, and 51.
Continue checking the table(s) for all domains and perspectives of interest in evaluating this intervention. This will provide the complete set of available measures contained in the Atlas for evaluating the mechanisms of the intervention. For this example, measure number 21 would be particularly important to review because it maps to the Patient/Family perspective of all 7 domains identified as relevant for this intervention.
2. Review relevant measure profiles.
Once you have identified each measure that maps to your intervention and evaluation mechanisms and perspectives, go to the Measure Profile section to find out more information about each of them. Each profile is preceded by an individualized measure mapping table that shows the care coordination domains and perspectives of the specific measure. The profile highlights the main features of the measure and key resources associated with it. These summaries also provide information on validity and testing, links to care coordination outcomes, application settings and populations, and format and data source. It also maps individual measure items (i.e., survey questions or questionnaire components) to each domain. This information should be used to guide the selection of specific measures for use in evaluating the intervention.
The individualized measure mapping tables provide information on the number of items related to each domain. A filled square (■) indicates that the measure has 3 or more items corresponding to that particular domain or that it is a composite measure related to that domain. An open square (□) indicates that a measure has only 1 or 2 items corresponding to that domain.
There are 37 different measures that map to the Patient/Family perspective of the 7 domains identified as relevant to the intervention in this example. To further narrow this list, you may begin by reviewing those measures that map to most of those 7 domains. For example:
Measure #4a. Consumer Assessment of Healthcare Plans and Systems—Adult Primary Care 1.0 (CAHPS) maps to 5 of the 7 relevant domains.
Measure #6. Client Perception of Coordination Questionnaire (CPCQ) maps to 5 of the 7 domains.
Measure #10. Patient Assessment of Care for Chronic Conditions (PACIC) maps to 5 of the 7 relevant domains.
Measure #11. Family Centered-Care Self-Assessment Tool - Family version maps to 6 of the 7 relevant domains.
Measure #13. Primary Care Assessment Survey (PCAS) maps to 5 of the 7 relevant domains.
Measure #16c. Medical Home Family Index and Survey (MHFIS) maps to 5 of the 7 relevant domains.
Measure #17a-b. Primary Care Assessment Tool (PCAT-CE) maps to 5 of the 7 relevant domains.
Measure #21. Resources and Support for Self-Management (RSSM) maps to all 7 relevant domains.
Measure #37. Patient Perceptions of Care (PPOC) maps to 6 of the 7 relevant domains.
Measure #40. Adapted Picker Institute Cancer Survey maps to 6 of the 7 relevant domains.
Measure #6 (CPCQ) has an open square in the box for Information Transfer from the Patient/Family perspective, indicating that the CPCQ has only 1 or 2 questions focusing on this domain and perspective. In contrast, Measure #4a (CAHPS) has a filled square (■) in the box corresponding to Information Transfer from the Patient/Family perspective, as it has 3 or more items addressing Information Transfer from this perspective. As a result, the CAHPS survey may, for example, be more appropriate for evaluating this component of the intervention. However, it also may require more resources to implement a lengthier measure.