Care coordination has been recognized as an important aspect of high quality, patient-centered care, and was identified by the National Priorities Partnership as a priority area for improving health care delivery in the U.S. Much work remains to be done to elucidate how best to achieve coordinated care and how care coordination relates to important outcomes such as hospitalization rates, readmissions, mortality, quality of life, and patient satisfaction. Yet even as this evidence base is developing, efforts are underway across the health care system to evaluate and improve care coordination. Robust measures of care coordination processes are essential tools for generating evidence about care coordination and its outcomes; evaluating current practices; designing, implementing and assessing improvement activities; and supporting payment initiatives that target care coordination.
The Agency for Healthcare Research and Quality (AHRQ) recognizes a particularly urgent need for measures that may be used in assessing or recognizing care coordination as it is carried out by primary care practices. Such measures may be used by health plans, insurers, or other payers to assess or recognize the degree and quality of coordination performed by primary care practices, as well as by practices themselves in assessing, and ultimately improving, their own performance.
As a critical step in providing measures to the field, AHRQ commissioned the development of the Care Coordination Measures Atlas, a compendium of existing measures of care coordination.1 It categorizes measures by their perspective (patient/family, health care professional, or system representative) and mechanisms used to coordinate care (activity domains). It includes measures that use a variety of data sources, although most rely on survey methods. Furthermore, it includes measures designed or used for three key purposes: quality improvement, research, and accountability.
This report presents measures selected systematically from the Atlas that are well-suited to primary care practice accountability and recognition purposes. It focuses on measures that are widely applicable and that reflect coordination as carried out by primary care practices rather than by other health care entities (e.g., hospitals, long-term care facilities, specialist providers). The report also includes measures that may be used to guide improvement efforts in response to the accountability measures. Measures were selected with four primary goals in mind:
- Ideally, measures included in the set should be comprehensive, covering all or most Atlas activity domains from the Atlas measurement framework. (For a list of these domain definitions, go to Appendix A).
- Measures should balance comprehensiveness with feasibility.
- Measures should be valid and reliable.
- Measures should be useful for accountability and recognition purposes, as the first priority, though measurement gaps can be addressed by measures that are useful for quality improvement purposes.
We selected measures separately for use in pediatric and adult populations.
In selecting measures, we considered the following criteria:
- Applicability to primary care practice evaluation.
- Focus on general population (not disease-specific).
- Broad coverage of activity domains from Care Coordination Measures Atlas framework (go to Appendix A for a list of domain definitions).
- Focus on care coordination (some measures in the Atlas embed care coordination items within a broader assessment of care).
- Evidence of reliability and validity.
We first narrowed the candidate measures (i.e., measures in the Atlas) to include only those that were applicable to a primary care setting or not setting specific. Second, we divided the measures into pediatric and adult indicator groups. Measures that were not age specific were included in both groups. Third, we further narrowed adult measures to those that focused on general chronic disease or that were not disease specific. We assigned the remaining measures to prioritization groups based on feasibility and degree of focus on care coordination. Following the goal to prioritize a comprehensive set, we identified high priority measures that mapped to the most activity domains from the Atlas measurement framework. We considered adult and pediatric measures separately. Finally, we assessed the validity and reliability of the most comprehensive measures.
We relied on published sources and information from measure developers in assessing these criteria. For complete details of selection methods, go to Appendix B. For detailed results of the measure selection process, including a detailed assessment of each selected measure, go to Appendix C.
About Measure Use
The measures evaluated in this report have been validated in their entirety. Although most measures contain items that are not focused on care coordination concepts, the validity reported here only applies to the care coordination related items in the context of the full instrument. In the case of survey-based measures, the respondent may answer differently based on other questions in the survey. For instance, one may assume a question does not include a concept contained in another question, but may not make that assumption if answering the same question in isolation. Therefore, further research is required to establish the validity of using only the care coordination portions of these measures.
An Emerging Field
This report is based on assessment of 64 measures included in the Care Coordination Measures Atlas.i We recognize that new measures of care coordination are being developed at a rapid pace, and existing measures are being further refined and tested. Thus, this report is necessarily limited to a sub-set of the full and dynamic universe of measures available for assessing care coordination. Furthermore, the level of reliability and validity testing that has become expected in other fields is largely lacking in the field of care coordination measurement.
Therefore, we emphasize that as this field matures and new measures are developed and existing measures are further tested and refined, the process of evaluating measures and making recommendations for primary care practice accountability evaluation will need to be revisited. We encourage measure users to share, and to the extent possible publish, their findings to help further development in this field. For further discussion of measure development and measurement gaps, see Measurement Development Gaps and Recommendations.
Organization of This Report
This report recommends two sets of measures:
Care Coordination Accountability Measures for Primary Care Practice contains measures recommended for evaluating care coordination performed by primary care practices in pediatric and adult populations for accountability or recognition purposes. Only measures from the patient/family perspective were considered because currently available measures from the health care professional and system representative perspectives rely on self-assessment, which is not appropriate for accountability purposes. In the future, new measures or audit procedures for current measures that reflect these additional perspectives would allow a more complete assessment of care coordination for accountability purposes.
The Companion Measure Set (not for accountability purposes) contains measures recommended as potential companions to the Accountability Measure Set for quality improvement purposes using a health care professional and system representative perspective. Selection criteria and processes are the same as those used for the Accountability Measure Set. The only reason these measures are not part of the Accountability Set is that all of the available measures are based on self-assessment, creating an inherent conflict of interest if used directly for accountability assessments
Recommendations are presented for pediatric and adult populations from each of these perspectives, and alternatives are provided based on expected differences in users' measurement priorities.
The final section, Measure Development Gaps and Recommendations, discusses measurement gaps and recommendations for research priorities. This section may be particularly of interest to measure developers and funders of quality measurement efforts.
Additional material is presented in a series of appendices:
Appendix A provides Atlas perspective and activity domain definitions.
Appendix B provides details of measure selection methods.
Appendix C provides detailed results of the measure selection process, including additional information about all measures included in the final measure sets.
Additional Key Sources
The Care Coordination Measures Atlas contains additional information about all measures considered for this measure set, and is a key source of information about the care coordination activity domains and perspectives referred to throughout this document.1 The Atlas is available for download at http://www.ahrq.gov/qual/careatlas/index.html.
An additional key resource is Appendix IV of the Atlas, which contains contact information for measure developers and copies of measure instruments for many of the Atlas measures. It is available for download as a separate document from: http://www.ahrq.gov/qual/careatlas/ (go to list of appendices).
i. Sixty-one measures are included in the original Atlas, available for download from http://www.ahrq.gov/qual/careatlas/index.html. An additional three measures were added to a Web-searchable version of the Atlas, which will be available soon from the AHRQ Web site.