In this chapter, we discuss care coordination measurement approaches that are still early in their development. We focus on three main areas of development: (1) care coordination measures utilizing data from electronic health records (EHR) or other health information technology (IT) systems, (2) public reporting of health IT-enabled care coordination, and (3) social network analysis as a novel approach to care coordination measurement. Because these areas of care coordination measurement are still evolving, we discuss them here with an emphasis on current level of development and growth potential, rather than including them in the review of individual measure instruments profiled in Chapter 6. These approaches were identified through the recent Atlas update measures search. Through this discussion, we aim to provide insight into future directions for measurement, and explore measurement potential, implementation challenges, and directions for further development.
EHR-based Care Coordination Measures
Much attention is being paid to the potential for using data from health IT systems, primarily EHRs, for quality measurement.a This interest has increased exponentially since passage in 2009 of the Health Information Technology for Economic and Clinical Health (HITECH) Act, part of the federal stimulus legislation. The HITECH Act allocated more than $25 billion towards building health information technology (IT) infrastructure and established two incentive programs (one each for Medicare and Medicaid) for adoption and 鈥渕eaningful use鈥 of certified EHR technology, including use for quality measurement.a Given the potential for EHRs and other health IT systems, such as health information exchanges, to facilitate information flow between providers, patients, and settings, health IT-based measures are of particular interest to the field of care coordination. Such cross-boundary measurement has traditionally been very difficult, yet it is crucial for understanding the process and effects of coordination across these care interfaces, whether or not they span separate organizations.
The Office of the National Coordinator for Health IT (ONC) within the Office of the Secretary for the U.S. Department of Health and Human Services recently reported that in 2012, 72% of all office-based physicians have adopted an EHR system, nearly double the rate just 3 years prior. Growth of EHR adoption has been even greater in recent years in the inpatient setting, where EHR adoption rates among non-Federal acute care hospitals more than tripled from 12% to 44% between 2009 and 2012. As of 2012, 85% of all non-Federal acute care hospitals use a certified HER.a These numbers are expected to grow in the coming years. This growth in EHR adoption and use offer much potential for major advances in the performance, and measurement, of care coordination over the next several years, even while many challenges remain.
In 2012, 大象视频published a report on the prospects for care coordination measurement using electronic data sources which evaluated opportunities for and barriers to measuring coordination processes using EHR data.b That report, based on interviews with 21 informants with expertise in health IT systems development and use, health information exchanges, EHRs, all-payer claims databases, insurance plans, health data standards, and quality measurement, highlighted three potential advantages of EHR data for use in measuring care coordination:
- Minimal data collection burden. Structured data within EHRs may be automatically extracted for quality measurement using computer programs or other advanced search techniques rather than through manual chart review.
- Rich clinical context. EHRs contain a trove of clinical data, including information on physician orders, laboratory and imaging results, medications prescribed, and progress notes. This information offers a view of processes of care and clinical outcomes not available within administrative claims data.
- Longitudinal patient data can be aggregated from multiple sources over time. EHRs aim (ideally) to aggregate information for each patient from multiple providers, settings, payers, and encounters into a single location.
While promising, both EHR technology and its implementation into clinical practice are evolving rapidly, and many barriers to EHR-based care coordination measurement have been highlighted in reports by 大象视频and others.a-e These barriers fall into three main categories: clinical workflow barriers, data limitations, and limited ability to share information across EHRs at different sites.
Just as the concept of care coordination is ambiguous in the health services research literature, there is as yet little agreement within the clinical sphere about what constitutes care coordination, who should do it, when, and how. This ambiguity limits clinicians鈥 efforts to coordinate care, and also limits documentation of coordination activities. As patterns of coordination-related clinical workflows emerge in the U.S. health care system, so too will the ability of EHRs to capture and facilitate those processes. Variability in care coordination documentation practices further limits development of standardized EHR-based measures of care coordination. Furthermore, heavy reliance on narrative documentation, rather than use of structured data fields, when entering clinical information into the medical record further limits use of information within EHRs for quality measurement.c (Structured data are contained within specific data fields that specify the type and format of recorded information, such as height recorded in meters. Unstructured data are generally recorded as free text, with no limitations in the format and often without clear specification of the type of information recorded in a particular location.)
Several aspects of EHR data present challenges for use in quality measurement, including measures of care coordination. Lack of data standardization complicates the process of specifying data elements to be used in EHR-based quality measures. Coding of lab results and medication information was of particular concern in prior reviews.a, c, d Proposed measures of care coordination that focus on the transfer of this kind of information across settings or providers will be limited by this lack of standardization. In addition, much work is needed to evaluate the reliability, accuracy, and completeness of information contained within EHRs when used for quality measurement..a, c Furthermore, many EHRs in use today require significant resources and technical support in order to extract data for the purpose of quality measurement.c
Poor EHR system interoperability presents a major obstacle to EHR-based measurement of care coordination processes. Without interoperability, EHRs cannot integrate into their record information about care received from other health delivery organizations or providers. This limits both coordination at the point of care and measurement of the coordination process using information contained within the EHRa, c, f Prior reports have noted that business models related to EHRs typically facilitate competition rather than cooperation, especially in ways that prevent a full picture of the steps taken to care for a patient across settings and time.c-e Much work is on-going to address EHR interoperability, but until increased information sharing becomes commonplace, one of the greatest potential advantages of EHR-based care coordination measures鈥攖he ability to capture processes of care that span providers and settings鈥攚ill remain largely unrealized.
Together, these reports underscore that EHR-based quality measurement is a nascent field, but one that is undergoing tremendous growth, spurred in particular by the HITECH Act.a
Specification of EHR-based Measures
The degree to which current measures can actually be calculated using EHR data depends upon the level of EHR-particular specifications available. By specification, we mean a set of definitions, instructions, codes, and/or software programs that allow any user to implement a measure in a precise, reliable, and replicable way. For example, while a measure definition describes what and who is measured, including a numerator and denominator description, the measure specification precisely specifies how the measure is to be calculated, including which fields within the data source are to be used and which values, such as particular diagnosis codes or ages, are included or excluded for a particular data field. For coordination, relative timing of events might be part of the specification (e.g., test result and interpretation communicated to patient within a particular time window relative to test performance).
Measure specifications designed to enable automatic extraction of clinical data from an EHR are necessary to realize one of the most promising benefits of EHR-based quality measurement: reducing resources needed for data collection while retaining rich clinical information, including timing and logistical steps related to care. Without such specifications, manual review of the electronic record would still be required, offering little additional benefit beyond traditional chart review of paper records. Accordingly, a new standard, the Health Level 7 (HL7) Health Quality Measure Format (HQMF), has been established to guide specification of EHR-based quality measures. eMeasure specifications are those that are fully specified in accordance with this standard, and that also include associated value sets for data elements used by the measure. Today, eMeasure specifications facilitate implementation of EHR-based quality measurement, although complete automation of EHR-based measurement has not yet been realized. Further automation of EHR-based care coordination measurement will advance as EHR technology and the HL7 HQMF standard continue to evolve in conjunction with changes in clinical workflow patterns that incorporate greater performance, and documentation, of care coordination processes.
Currently Available EHR-based Measures of Care Coordination
With this context, we now review currently available EHR-based measures of care coordination (Table 4 and Table 5). These measures were identified through the updated Atlas measure search (go to Appendix II for details), map to at least one of the Atlas framework care coordination domains, and were designed specifically for use with EHR data or have complete eMeasure specifications available. We omit from this discussion measures that included EHR or health IT system data as a potential data source within the measure documentation without any further specifications particular to EHR data. We do include in the discussion the Meaningful Use objectives that are being used in the CMS EHR incentive programs to document that participating eligible professionals and hospitals are using certified technology in accordance with program goals. While the purpose of these objectives differs somewhat from traditional health care quality measurement, we believe they represent an additional type of EHR-based measure that may shed light on processes of care coordination, and as such include them in this discussion. We reviewed all Stage 1 and Stage 2 Meaningful Use objectives and clinical quality measures (CQM) (collectively referred to in this discussion as Meaningful Use measures), and include here only those that evaluated a process of care that mapped to at least one of the Atlas framework domains.
Many of the Meaningful Use and other EHR-based measures included in this discussion assess additional aspects of quality of care beyond coordination processes. As with many of the measures profiled in Chapter 6, determining whether a particular measure evaluated care coordination or some other aspect of care was at times a difficult decision requiring subjective judgment and consideration of context. See the section on care vs. care coordination in Chapter 1 for further discussion of the challenges in distinguishing measures of care coordination from measures that assess other aspects of care, and how we addressed those challenges when considering measures for inclusion in the Atlas.
The original Atlas search completed in July 2010 found no EHR-based measures of care coordination. A brief discussion of the Meaningful Use Stage 1 objectives was included in the original Atlas, but the CMS EHR incentive programs were in a very early stage of initiation at the time of its publication, so a complete review of measures associated with those programs was not undertaken.
As Table 4 and Table 5 demonstrate, there has been much interest and development in this area since that time, with 26 new EHR-based measures identified in the Atlas search update, including 13 Meaningful Use measures (9 objectives and 4 CQMs).
Together, the 26 EHR-based measures shown in Table 4 and Table 5 evaluated nine Atlas domains (Figure 3). The Communicate domain was the most commonly measured, specifically the Information Transfer sub-domain (n=17), highlighting the predominant focus of early EHR-based care coordination measures on tracking the flow of information from one location to another as patients receive care (Figure 4). Measures that mapped to the Information Transfer sub-domain most often evaluated transfers of information occurring across health care teams or settings (n=10), with an additional seven measures evaluating transfers of information between a health care provider and the patient or family. As highlighted by the gaps in Figure 4, no measures evaluated communication among members of a health care team, such as providers and staff within a single clinic, and no measures evaluated the Interpersonal Communication sub-domain, an area less identifiable with currently collected electronic information.
Figure 3. Frequency of Care Coordination Domains Measured by EHR-Based Measures
[D] Select for Text Description.
Note: No measures mapped to some of the domains, as illustrated above. Many measures mapped to more than one domain. N = 26 measures total.
Twelve measures mapped to the Facilitate Transitions domain; all of these evaluated transitions occurring across health care settings (Figure 5). The transitions most frequently measured were those from primary care to outpatient specialty care (n=4), inpatient to primary care (n=2), inpatient to outpatient specialty (n=2), and inpatient to any other setting of care (n=2). One measure also assessed the transitions as coordination needs change, evaluating coordination as older adults who have experienced certain fractures (hip, spine or distal radius) transition from acute care to a period of rehabilitation.
Health IT-enabled coordination was also commonly measured among the set of EHR-based measures (n=9), not surprising given the focus of many of these measures on the use of EHR technology, particularly among the Meaningful Use objectives and CQMs that account for 13 of the 26 EHR-based measures identified (Figure 3). (Note that this domain reflects whether health IT system functionality was used to carry out care coordination activities, not whether health IT data were used in calculating the measure. Thus, not all EHR-based measures map to this domain). While previous evaluations of potential for EHR-based care coordination indicated interest in using EHR data to evaluate coordination facilitated by comprehensive care plans,c-e only one of the currently available EHR-based measures addresses this domain, in this case, evaluating provision of a home management plan of care to pediatric asthma patients. This likely reflects continued ambiguity around what constitutes a comprehensive plan of care and how to measure it. As highlighted in that report, such proactive, interactive, comprehensive and shared care planning is not widely used in current practice.c
Figure 4. Frequency of Communicate Domains and Sub-Domains Measured by EHR-Based Measures
[D] Select for Text Description.
Note: Measures mapped to the Communicate domain when the mode of communication was not specified as either Interpersonal Communication or Information Transfer. No measures mapped to the Interpersonal Communication sub-domain. No measures assessed communication within teams of health care professionals. N = 26 measures total.
Taken together, these EHR-based measures reflect the current health IT climate that is widely concerned with solving problems of interoperability and achieving greater information sharing across settings, providers, and other participants in patients鈥 care. They also reflect limitations in the ability of EHR technology to capture dynamic, interpersonal processes such as teamwork, care planning, and interpersonal communication. Advances in technology and its integration into clinical work flows may attenuate some of these limitations in the future, but some aspects of care coordination may never be well-captured in EHRs.c When resources allow, combining EHR-based measurement with other measurement approaches, such as surveys, can provide a more complete assessment of the many aspects of care coordination. Furthermore, EHRs represent just one view of care coordination processes (the system representative perspective). Measurement from the patient/family and health care professional perspectives is also important.
Figure 5. Types of Cross-Setting Transitions Evaluated by EHR-based Measures
[D] Select for Text Description.
Note: The sum of transitions listed above exceeds the total number of measures that evaluate any cross-setting transition (12) because some measures evaluated multiple transitions of care (i.e., transitions between Primary Care and Outpatient Specialty Care, as well as Primary Care and Inpatient).
Measures of EHR Use for Care Coordination - Meaningful Use
The Meaningful Use objectives and CQMs used in the CMS EHR incentive programs deserve particular attention, given the powerful impact those programs are having on health IT adoption. CMS reports that by mid-2013, more than half of all eligible professionals had received some incentive payment under the EHR incentive programs (Medicare and Medicaid combined). More than 309,000 unique eligible professionals and more than 4,000 unique eligible hospitals have received incentive payments. Payments as of June 2013 total more than $15.5 billion.p
Of the 26 EHR-based measures identified in the recent Atlas update search, 13 are used to evaluate Meaningful Use under the CMS EHR incentive programs (Table 4). These measures focus in particular on measuring the transfer of information (8 measures), either between providers and patients or their family (5 measures) or across health care providers or settings (3 measures). This reflects the focus of the Meaningful Use evaluation criteria to date (Stage 1 and Stage 2), which emphasizes data capture and sharing. It also reflects limitations in most EHR technology available today. One of the barriers to EHR-based care coordination measurement reported by 大象视频is that few options are available within current EHR technology to create, maintain, and share a longitudinal, comprehensive plan of care.c Similarly, much of the information needed for care coordination, such as documentation of needs assessments, patient preferences, responsibilities of the various participants in a patient鈥檚 care, and patient support networks, typically resides in unstructured text format (i.e., free text notes) or is simply not recorded anyplace, rather than in structured fields using standard terminology or code sets. To date, no EHR-based measures use unstructured data, and recent evaluations suggest this will be the case for the foreseeable future.c Enabling measurement of these aspects of care coordination will require a combination of advances in technology (building in structured data fields for this information), standardization (creating standards to encode this information), and clinical workflow (gathering information and documenting within structured fields using standards).c Public Reporting of Health IT-enabled Care Coordination
As increasing attention is focused on the adoption and use of EHRs and other health IT systems, some efforts are underway to publicly report health IT use. To the extent that these publicly reported measures specifically address care coordination, they also represent new opportunities for public reporting of coordination processes. Below, we summarize three such public reporting efforts identified as part of the Atlas measures search update. (Because the Atlas measures search was not designed specifically to identify public reporting initiatives, other examples may exist that report on some aspects of care coordination.)
- Rhode Island Health IT Adoption. As of August 2013, Rhode Island is the only state to mandate public reporting of health IT adoption and use by all licensed physicians. Beginning in 2013, advanced practice nurse practitioners and physician assistants must also participate, and will be individually identified in public reporting beginning in 2014. This public reporting is based on an annual survey that measures communication and information transfer across health care settings and use of EHRs to support patient monitoring and followup, as well as other aspects of EHR use not related to care coordination (go to Measure #75, profiled in this updated Atlas). Practitioner-level scores are reported for five composite measures of EHR use, of which two (scores for basic and advanced EHR functionality use) include most of the coordination-related survey items. Although these composite measures mask some of the specificity of the coordination items included within them, they represent one of the earliest attempts to publicly report the performance of care coordination for individual health care professionals. More information and physician-level measure scores are available from the State of Rhode Island Department of Health ().
- Minnesota Health Scores. This voluntary, state-wide public reporting initiative includes reporting the level of health IT-based care available from individual ambulatory care clinics within the state with respect to three functionalities: Adoption, Use, and Exchange. Most relevant to care coordination is the level of reported Exchange functionality, indicating whether an ambulatory care clinic sends or receives electronic data via an EHR with network hospitals (mid-level exchange functionality) and whether the clinic can also safely send or receive electronic information from its own EHR with hospitals outside its network (advanced level exchange functionality). Data are reported from a survey of most ambulatory care clinics in Minnesota; all clinics were invited to participate. In 2013, the first year of the program, 80% of clinics completed the survey. Clinic-level data are available online from Minnesota Health Scores ().
- State of California Office of the Patient Advocate (OPA) Quality Report Cards. Through this web site, consumers can view quality information about ten commercial health maintenance organizations (HMO), six preferred provider organizations (PPO), and more than 200 medical groups in California. The medical group ratings include information about use of health IT to facilitate communication and information transfer between health care providers and patients, such as whether patients can email their doctor, receive test results online, view their medical record online, or receive a visit summary with instructions after each visit. These ratings are generated by the Integrated Healthcare Association鈥檚 pay for performance initiative using the National Committee for Quality Assurance (NCQA) Healthcare Effectiveness Data and Information Set (HEDIS) measures as well as results from the Patient Assessment Survey. The HMO and PPO quality ratings include patient-reported experience of care based on the 大象视频Consumer Assessment of Healthcare Providers and Systems (CAHPS) health plan survey, including ratings of care coordination, but do not specifically address use of health IT to facilitate care coordination. More information and the HMO, PPO and medical group quality scores are available from the OPA Quality Report Cards web site at .
In addition to myriad research and public policy uses, these publicly available data on some care coordination processes may serve as a benchmark against which to gauge care coordination processes measured at a local level, such as local quality improvement efforts. In this sense, they are examples of tools to be used alongside the Care Coordination Measures Atlas, providing a reference point or sense of scale against which to interpret results from other coordination measures.
Social Network Analysis of Care Coordination
Social network analysis (SNA) is a method for mapping and analyzing relationships among actors within a network. A network consists of actors, such as individuals, organizations, programs, or other entities, who are connected with one another in some way.q SNA uses quantitative methods to evaluate the relationships and interactions of actors within a network, and can facilitate comparisons of one network to another, even when network structures differ.r In this section, we provide a brief overview of SNA methods, then discuss its application to the health care setting, in particular its use in the study of processes related to care coordination.
Method Overview
To use the SNA method, the first step is to identify actors within a particular network. When the boundaries around a network are clear, this is a straightforward task and all actors are identified before any data collection begins. When boundaries are not clear, data must be collected to identify actors, often through qualitative methods, such as interviews or snowball techniques. However actors are identified, once this has been done, relationships鈥攐r ties鈥攂etween each member of the network are mapped with each other member of the network, yielding a matrix of dyadic interactions. For example, if examining relationships among five nurses within a particular primary care clinic, the resulting network would take the form of a 5-by-5 matrix. The actors and ties among them may be depicted visually through a variety of network graphs, and quantitative methods can be used to evaluate various characteristics of a network. These include the degree of connectivity of particular actors within the network, the importance鈥攐r centrality鈥攐f particular actors based on their position between, proximity to, and directed connections with other actors, and the extent to which different actors within a network play similar roles (termed structural equivalence). While the metrics used to evaluate and describe networks vary depending upon the questions of interest, all SNA methods are similar in their focus on identifying actors within a network and quantifying the interactions or relationships among them.q
Social Network Analysis in the Health Care Setting
SNA has been used for the study of organizations since the 1950鈥檚, including some application in the health care setting. A recent systematic review of SNA methods published within the medical and health care literature identified 52 such studies published between 1950 and 2011.s All but one of the included studies used SNA to describe an existing social network in a health care delivery organization, typically gathering information about networks of physicians, nurses, other health professionals, administrators, and policy makers. These studies focused most often on (1) organizational management, such as physician-nurse interactions, staff relationships, team-functioning, and within-organization decision-making processes; (2) diffusion of innovations, including adoption of medical technology, prescribing practices, and evidence-based medicine; and (3) professional ties among providers from different organizations, settings, or health professions. Few examined connections across health care settings (just 9 of 52 studies) and none specifically examined care coordination.
However, several studies suggest how SNA methods might provide insight into coordination processes at the level of organizations, patients, or particular care transitions. Below, we highlight three studies identified from the prior review, the updated Atlas measure search (go to Appendix II for details), and other informal searches that demonstrate how SNA methods can be applied to the study and measurement of coordination-related processes.
In an example of an organization-level SNA approach to examining care coordination, Nageswaran and co-authors examined inter-agency collaboration in the care of children with complex chronic conditions in a single U.S. city.t The authors found that pediatric practices reported the greatest degree of collaboration with other agencies with respect to both referrals out to and in from other organizations. They also had strong connections with subspecialty practices, but weak ties with supportive services agencies. The latter had poor ties with many other agencies and the greatest gaps in collaboration. By asking network actors about desired as well as actual ties, the authors zeroed in on the Atlas domain of links to community resources, revealing potential gaps in the coordination of services for this patient population which may be ripe for establishing new connections among agencies that desire more collaboration.
Weenink and colleagues examined networks of providers caring for patients with type 2 diabetes and chronic heart failure (CHF) at three primary care clinics in the Netherlands, using information from patients, health care providers, and the medical record to construct patient-specific networks.u While small and of very limited generalizability, this study demonstrated feasibility of constructing patient-specific networks that arise during the provision of care. This patient-centric approach differs from other applications of SNA that have examined networks defined by organizational, professional, or disease boundaries. Thus SNA has the potential to provide measures from each of the three Atlas perspectives (patient/family, health care professional, and system representative), as well as linkages between the individuals representing each view.
Finally, Benham-Hutchins and co-authors examined the network of actors and communication patterns surrounding five patient hand-offs within a single hospital, such as admission to the hospital from the emergency department or transfers from one inpatient unit to another.v While these transitions occurred within a single hospital, the results illustrate that much care coordination, in particular communication, occurred during even intra-organization care transitions. Networks of providers included in the five hand-offs studied included between 11 and 20 providers. These networks were mapped by functional role, such as emergency department nurse or surgeon, rather than by individual name. Thus, the number of individuals involved in these hand-offs was likely greater than that reported from the analysis. The study found that none of the communication networks used in the five studied transfers had a centralized structure and that no single provider within any network coordinated information exchange. Gatekeepers were common among the networks, controlling the flow of information among various other actors. This study demonstrates that applying SNA techniques to examine care coordination processes is feasible, even at the very granular level of examining specific transitions for individual patients.
A key distinction between these three example studies is the level of analysis. Nageswaran and co-authors examined networks of organizations,t reflecting typical patterns of interaction that occur routinely over the course of providing care or services for many patients. Analyses conducted at this level can provide insight into patterns of information sharing, collaboration, and referrals that occur regularly across organizations, potentially suggesting structural gaps where stronger connections are needed, as well as links that bridge separate operational networks. Weenink and colleagues examined networks centered around patients, evaluating the degree to which certain aspects of care were centralized with a particular provider role or specialty group, or with the patients themselves.u This type of application might be useful for evaluating the effects of team-based or multidisciplinary care models or the effectiveness of improvement initiatives that employ care coordinators or technology to centralize coordination processes.
The study by Benham-Hutchins and co-authors examined coordination processes at an even more granular level, mapping networks of interactions that emerged on an ad hoc basis at the point of care as specific patient transitions occurred within a single institution.v This extreme micro-level view provided much more detailed insight into the roles and interactions of particular providers within the hospital of study, but results might not be reflective of typical patterns of interactions around other patient transitions within the same hospital, and are likely even less generalizable to other health care delivery organizations. However, the greater level of detail would likely be useful for quality improvement efforts that target team functioning.
These studies demonstrate just three of the ways that SNA methods can be applied to evaluate care coordination processes, but other applications exist and more will emerge as social network methods are more widely applied in this field. It is probable that additional applications of SNA to care coordination measurement have been published, but were not identified through the updated Atlas measure search. However, the identified evidence suggests that while promising and feasible, SNA has not yet been applied widely to questions of care coordination. Only one of 52 SNA applications from the health care setting identified by a recent systematic reviews related directly to care coordination, and another recent systematic review of boundary spanning roles within collaborative networks found only three examples from the health care setting, none of which addressed care coordination.w
SNA-based methods of examining care coordination processes hold promise because they consolidate great complexity into a few measures and are highly adaptable. However, data collection can be burdensome, particularly for networks without clear boundaries or with many actors, and analyses can be complex and often require special software programs. Future development of SNA-based care coordination measures must address these challenges, while refining methods particular to questions of coordination, care transitions, and collaborative care.
Landscape of Care Coordination Measures
These emerging trends will enhance the landscape of care coordination measurement options, supplementing the current predominance of survey-based measurement methods with additional data sources and approaches. For the most part, these newer approaches to measurement will not replace older methods, but rather complement them by providing additional lenses through which to view coordination-related processes of care. However, it is likely that one formerly common approach to care coordination measurement鈥攎anual chart review鈥攚ill be replaced in the future. As EHR technology and EHR-based measurement methodologies develop further, many measures that formerly relied on manual chart review will likely be supplanted by EHR-based measures for which data can be automatically extracted rather than requiring time-consuming manual review. In some cases this will involve revising measure specifications that were designed for chart review methods to instead adhere to the emerging standards for eMeasure specifications, as has been done for some of the currently available EHR-based measures. As the field of EHR-based measurement matures, additional measures will be developed that leverage the types of data most readily available from within EHRs.
Obtaining a comprehensive understanding of care coordination requires measurement from multiple perspectives, as is emphasized by the inclusion of three key perspectives in the Atlas framework: patient/family, health care professional, and system representative. While this chapter emphasizes development of novel measurement approaches, we do not wish to suggest that surveys鈥攖he predominant type of care coordination measure in use today鈥攁re outdated or inadequate. Indeed, we expect that surveys will continue to be the chief method of measuring care coordination for the patient/family and health care professional perspectives, and will continue to play an important role as one of several options for measuring the system representative perspective. Rather, as they are further developed and implemented, the emerging measurement approaches discussed in this chapter will provide additional options for measuring care coordination from each measurement perspective. EHR-based measures offer a new method for evaluating the system representative perspective, and in the future may provide an additional avenue for evaluating the patient/family perspective as opportunities increase for patients and their representatives to interact directly with EHRs. Social network analysis approaches can be adapted for measuring each of the perspectives, depending upon the level of analysis and source of information used to create network maps. Further development may also lead to combined or hybrid approaches, such as integrating questionnaires that collect data for social network analysis into existing care coordination-related surveys of patients or health care professionals, and then linking network characteristics to coordination processes evaluated through other means, such as EHR-based measures. While these possibilities are as yet unrealized, the rapid pace of care coordination measure development will ensure that many new measurement approaches continue to emerge and further enhance the measurement landscape.
As these and future measurement approaches emerge, the expanded landscape of care coordination measures will become broader, richer, and more diverse, but also potentially more difficult to navigate. It is our hope that this Atlas will serve as a valuable resource to guide measure selection, identify key measurement gaps, and build towards a common understanding of care coordination.
Chapter 4 References
a. Anderson KM, Marsh CA, Flemming AC, et al. Quality Measurement Enabled by Health IT: Overview, Possibilities, and Challenges (Prepared by Booz Allen Hamilton, under Contract No. HHSA290200900024I.). Rockville, MD: 大象视频; July 2012. 大象视频Publication No. 12-0061-EF.
b. Office of the National Coordinator for Health Information Technology. Update on the adoption of health information technology and related efforts to facilitate the electronic use and exchange of health information: A report to congress. Washington, DC: Office of the National Coordinator for Health Information Technology; June 2013.
c. McDonald K, Schultz E, Chapman T, et al. Prospects for Care Coordination Measurement Using Electronic Data Sources (Prepared by Stanford University under subcontract to Battelle on Contract No. 290-04-0020--大象视频SQI-II) 大象视频Publication No. 12-0014-EF. Rockville, MD: 大象视频; March 2012.
d. Samal L, Hasan O, Venkatesh AK, et al. Health Information Technology to Support Care Coordination and Care Transitions: Data Needs, Capabilities, Technical and Organizational Barriers, and Approaches to Improvement. Boston, MA: National Quality Forum; 2012.
e. Samal L, Dykes PC, Saranto K, et al. Patient-centered care across transitions: challenges and opportunities in clinical informatics. Studies in Health Technology and Informatics 2013;192:1238.
f. National Quality Forum. Critical Paths for Creating Data Platforms: Care Coordination. Draft Report for Public Comment. 2012.
g. Meaningful Use Stage 2. Centers for Medicare & Medicaid Services. Accessed: August 30 2013.
h. Meaningful Use Clinical Quality Measures. Centers for Medicare & Medicaid Services. Available at: . Accessed: August 30 2013.
i. American Medical Association Physician Consortium for Performance Improvement, National Committee for Quality Assurance, American Congress of Obstetricians and Gynecologists. Maternity Care Performance Measurement Set. Chicago, IL: American Medical Association; 2012.
j. Quality Net: Hospital Outpatient Quality Reporting Specifications Manual, v7.0. Centers for Medicare & Medicaid Services. Available at: . Accessed: August 30 2013.
k. American Medical Association Physician Consortium for Performance Improvement, American College of Cardiology Foundation, American Heart Association. Heart Failure Performance Measurement Set. Chicago, IL: American Medical Association; 2012.
l. Chan KS, Holzer J, Scholle SH, et al. The Development and Testing of EHR-based Care Coordination Performance Measures in Ambulatory Care. New York, NY: 2011.
m. Kern LM, Dhopeshwarkar R, Barron Y, et al. Measuring the effects of health information technology on quality of care: a novel set of proposed metrics for electronic quality reporting. Jt Comm J Qual Patient Saf 2009;35(7):359-69.
n. Centers for Medicare & Medicaid Services. 2013 Physician Quality Reporting System (PQRS) Measures List. Baltimore, MD: Centers for Medicare & Medicaid Services; 2012.
o. American Medical Association Physician Consortium for Performance Improvement, American Academy of Neurology, American Geriatrics Society, et al. Dementia Performance Measurement Set. Chicago, IL: American Medical Association; 2011.
p. Centers for Medicare & Medicaid Services. June 2013 EHR Incentive Program Report. Baltimore, MD: Centers for Medicare & Medicaid Services; 2013.
q. Luke DA, Harris JK. Network analysis in public health: history, methods, and applications. Annu Rev Public Health 2007;28:69-93.
r. Scott J, Tallia A, Crosson JC, et al. Social network analysis as an analytic tool for interaction patterns in primary care practices. Ann Fam Med 2005;3(5):443-8.
s. Chambers D, Wilson P, Thompson C, et al. Social network analysis in healthcare settings: a systematic scoping review. PLoS One 2012;7(8):e41911.
t. Nageswaran S, Ip EH, Golden SL, et al. Inter-agency collaboration in the care of children with complex chronic conditions. Acad Pediatr 2012;12(3):189-97.
u. Weenink JW, van Lieshout J, Jung HP, et al. Patient Care Teams in treatment of diabetes and chronic heart failure in primary care: an observational networks study. Implement Sci 2011;6:66.
v. Benham-Hutchins MM, Effken JA. Multi-professional patterns and methods of communication during patient handoffs. Int J Med Inform 2010;79(4):252-67.
w. Long JC, Cunningham FC, Braithwaite J. Bridges, brokers and boundary spanners in collaborative networks: a systematic review. BMC Health Serv Res 2013;13:158.
