Coordination in healthcare is recognized as an important component of providing quality patient care. The Institute of Medicine and the Agency for Healthcare Research and Quality (AHRQ) have found that better quality outcomes are achieved by deliberately organizing patient care activities to facilitate healthcare delivery (i.e., care coordination) (McDonald et al., 2007) than in the absence of such organization. Effective teamwork among care providers requires coordination among team members, often consisting of work on interdependent tasks, team member communication, and task sequencing (Okhuysen & Bechky, 2009). Without proper team coordination, breakdowns can occur that negatively affect the quality of team performance (Hysong et al., 2011 ; Sui, Wang, Kirkman, & Li, 2015) and the job experience of team members (Helfrich et al., 2014 ; True, Solimeo, & Stewart, 2014). With this understanding, effective coordination of patient care as a central aspect of interprofessional care provision has emerged as an essential component in healthcare. The authors of the present article propose that, for healthcare providers to coordinate patient care successfully, effective interdependent and collective performance must take place among care team members. In other words, to coordinate care well, care providers must coordinate well among themselves. In turn, to achieve this goal, the elements of team coordination need to be measured accurately.
Although the need for and utility of coordinated care is strongly advocated (e.g., McDonald et al., 2007), the measures for evaluating team coordination and defining the critical components of coordinative activities have not been clearly identified. Considerable literature documents the importance of care coordination (Stille, Jerant, Bell, Meltzer, & Elmore, 2005 ; Forrest et al., 2000 ; Berendsen, de Jong, Meyboom-de Jong, Dekker, & Schuling, 2009 ; Bal, Mastboom, Spiers, & Rutten, 2007 ; McDonald et al., 2007). Forrest et al. (2000), for example, examined coordinative activities in the referral process and found that improved coordination was associated with increased physician satisfaction of care and successful referral. Similarly, Bal et al. (2007), in discussing a tool designed to increase communication between care providers, described the importance of team member communication in providing quality patient care. Research examining or recommending evidence-based best practices for the assessment of coordinative care, however, is scarce. Stille et al. (2005), who advocated for a team-based approach to coordinative care, call for more research on the development of clear definitions and measurement tools of team coordination.
This paucity of measurement recommendations arises from a lack of a conceptual understanding of what team coordination is and how best to achieve it in healthcare. For example, the AHRQ framework of coordinated care (McDonald et al., 2010) provides a starting point by enumerating context-specific coordination activities (e.g., facilitating transitions, assessing needs and goals, monitoring and follow-up) as well as broad approaches to care (e.g., health information technology-enabled coordination; a healthcare home or patient-centered medical home). This framework, however, stops short of identifying fundamental components (e.g., tasks, roles, responsibilities) of coordinated care regardless of the condition, patient, or context. Coordination in this sense requires the collective, effective provision of timely care by an interprofessional team with appropriate sequencing and transfer of responsibility within that team (Salas, Sims, & Burke, 2005 ; Okhuysen & Bechky, 2009). For successful provision of care, coordinating among care team members must occur. Thus, team coordination is the focus of this article.
In developing best practices for implementing successful team coordination, an appropriate, theory-based model of coordination must first be identified that is applicable to healthcare contexts. The conceptual definition of constructs in that model may then be used to identify and select appropriate measures as they relate to healthcare team coordination. This article employs Okhuysen and Bechky's (2009) model for coordination, referred to as the O&B framework, as a structure by which to systematically review the literature for appropriate metrics used in team coordination-related studies relevant to healthcare.
The O&B integrative framework (summarized in Table 1) explains the mechanisms of coordination and the integrating conditions necessary to achieve coordination. Because this model is context-free, it can easily be applied to a wide variety of healthcare coordination issues, such as breakdowns in coordination of patient referrals (Hysong et al., 2011). According to this framework, five basic mechanisms or structures (see the first column in Table 1) undergird team coordination: plans and rules, objects and representation, roles, routines, and proximity. These mechanisms enable teams to achieve three process outcomes, termed “integrating conditions” in the model (see columns 2-4 in Table 1): (a) accountability (clarity over who is responsible for what activity), (b) predictability (knowing what tasks are involved and when they should occur), and (c) common understanding (providing a shared perspective on the whole process and how individuals' work fits within the whole). These conditions allow individuals to collectively accomplish their interdependent tasks.
In tobacco screening, for example, accountability involves understanding the specific tasks that each member of the interprofessional team is responsible for (e.g., the nurse educates the patient about the risks of tobacco use, the physician orders appropriate treatment, the clerk schedules a follow-up appointment).
Predictability requires an agreement on care sequencing tasks. For example, first, the nurse must decide to proceed with tobacco use screening; second, assuming the patient agrees, the nurse and physician educate the patient about the risks of tobacco use; third, they encourage the patient to consider options for smoking cessation; and so on.
Finally, the team's state of common understanding relates to how each person's tasks fit in the overall goal of the care episode. This condition, in the tobacco screening example, involves an overarching understanding among staff that, by correctly deciding to proceed with screening about tobacco use (nurse's task), they are helping support the physician's assessment of treatment options and treatment initiation.
In this way, coordination is achieved by the combination of team members working through the aforementioned mechanisms (rows in Table 1) and across the integrating conditions (columns in Table 1) as measured by the specific elements detailed in the cells of Table 1. For example, plans and rules facilitate accountability and predictability by defining responsibilities for tasks. Proximity facilitates each integrating condition by promoting familiarity in different ways: accountability, by developing trust among team members (thereby increasing familiarity); predictability, by helping team members learn to anticipate and respond to others' needs; and familiarity, by building a common fund of knowledge (i.e., a transactive memory system).
Gaps in the Literature and Study Objective
Observational behavioral research has shown that the coordinating mechanisms and conditions described by the O&B framework are associated with improved team coordination and, subsequently, enhanced task performance (Marks & Panzer, 2004 ; Stout, Salas, & Carson, 1994). This conceptual framework of coordination provides a solid theoretical foundation for understanding the elements needed to help healthcare teams effectively coordinate care. Similar translations are seen from the literature discussing organizational capacity for change intervention, where researchers have successfully applied broad constructs from multiple areas in organizational research to healthcare settings in ways that offer tools for practitioners (e.g., Helfrich, Li, Sharp, & Sales, 2009 ; Stetler, Damschroder, Helfrich, & Hagedorn, 2011). The present aim is to offer a first step toward a similar translational goal in coordination.
Most recent research establishing relationships between team coordination and improved performance has primarily used observational methodologies and not survey items, limiting its potential generalizability. For team coordination to be accurately and comprehensively measured in healthcare settings for both research and organizational monitoring/feedback systems, the survey instrument used must be affordable and easy to administer. In the absence of such an instrument, a systematic review of existing survey metrics is warranted because surveys are easy to implement, replicable, and standardized.
The O&B framework is based on three decades of empirical research on coordination and related constructs. The authors of the present article viewed this literature as a likely source for reliable and valid survey instruments for measuring many, if not all, of the constructs posited in the O&B framework. However, as team coordination is a fundamental construct for organizations of all types, including healthcare, the literature is fragmented across numerous healthcare fields; currently, no centralized resource for survey measures of coordination is available. The authors' objectives for this study were to (a) systematically review and catalog the healthcare literature for validated survey instruments and items that capture the domains in the O&B framework and (b) identify target areas where scale modification or development may be necessary.
The authors conducted a systematic review of peer-reviewed journal articles using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement as a guide (Moher, Liberati, Tetzlaff, & Altman, 2009). This statement, consisting of a 27-item checklist, provides structured guidelines for conducting systematic reviews in the healthcare literature, as used for the procedures listed later in this section.
As the authors' interest in team coordination stemmed from the study of healthcare teams, the study began with an article search in the National Center for Biotechnology Information's PubMed database using the strategy and eligibility criteria detailed later. Upon yielding a very small number of articles (see Results section), the search was expanded to include databases from the industrial/organizational psychology and management domains, as teams are a basic component of organizations and are commonly discussed in these literatures. These additional databases included Clarivate Analytics' ISI Web of Science, the American Psychological Association's PsycINFO, and EBSCO Information Services' Academic Search Complete.
To be eligible for inclusion in the review, studies had to have been published in a peer-reviewed journal and contained the full text of survey items that measured one or more of the domains in the O&B framework.
The search strategy began with a bibliographic search for all articles cited in the O&B review, followed by a Web of Science search for all articles citing the O&B paper. Next, a keyword search was conducted of all the domains in the O&B model (see Table 1 for terms) as well as constructs related to coordination that could contribute items. The top eight management and psychology journals, as determined by impact factors and content coverage, included in the keyword search were as follows: Academy of Management Journal, Group & Organization Management, Journal of Applied Psychology, Journal of Management, Journal of Organizational Behavior, Organizational Behavior and Human Decision Processes, Organization Science, and Personnel Psychology. Journals with no coordination or team-related articles were not included in the review. Keywords used are provided in Table 2.
As the O&B review covered the coordination literature up to the time of its publication, the keyword search focused on articles published between 2010 and 2014 (since the publication of the O&B framework). However, as initial searches yielded too few articles, publication date restrictions were lifted. The PubMed keyword search was replicated to identify articles from the healthcare literature that could contribute instruments.
During the keyword search process, two review articles—a meta-analysis by Kleingeld, van Mierlo, and Arends (2011) and a seminal longitudinal study of project team performance that contained many constructs related to coordination (Keller, 2006)—were found that provided additional search opportunities. The authors conducted a bibliographic search of the articles included in Kleingeld et al. (2011) and a cited reference search of Keller (2006).
Two research assistants abstracted into a spreadsheet the articles identified by the search strategy. The following data elements were abstracted from each article: complete bibliographic citation, scale name (if mentioned), construct measured, construct definition (if provided), internal consistency reliability coefficient for scale (if reported), abstractor's initial assessment of the O&B domain being measured by the scale, scale response choice format, item text, and abstractor ID. Each record in the spreadsheet contained one survey item (i.e., multiple records per scale, and multiple records per article).
The research team responsible for coding was composed of five people: a master's-level industrial and organizational (I/O) psychologist, a bachelor's-level research assistant with a psychology background, two doctoral-level I/O psychologists, and a doctoral-level social psychologist. To obtain only high-quality potential items, the authors excluded scales with unacceptable or unreported reliability coefficients (α < .70, n = 513 items) and with response choices that did not conform to a Likert-style format (n = 376). The remaining scales were reviewed on an item-by-item basis by the research team, first individually and then as a group, to determine if the items could be categorized as a whole into one of the domains in the O&B model. To make this determination, the authors compared the text of the item to the definition of the domain as described in Okhuysen and Bechky (2009), shared individual decisions, and arrived at a consensus when discrepancies occurred.
Consistent with the authors' initial observations that most of the research on coordination relies on observational methodologies and not survey instruments, the review yielded zero full scales that could be classified as assessing an O&B domain of coordination. Therefore, the authors focused on individual items to determine which items mapped to a specific O&B domain. Items contributed by a single scale could be assigned to more than one domain (e.g., items 1-3 on a six-item scale originally intended to measure construct A could be assigned to one O&B domain, whereas items 4-6 might be assigned to a different O&B domain); however, any given item was assigned to only one domain. Team members individually and then collectively indicated which items belonged to which domain, and discrepancies were resolved by consensus. Any items that could not be categorized into one and only one O&B domain (either because no domain adequately fit or because the item could be assigned to more than one O&B domain) were discarded.
Due to the lack of full scales that assess the O&B domains, the authors created healthcare team coordination scales for each domain based on the items identified in this review and the construct definition as delineated by O&B. For each O&B domain, the research team followed this procedure to create scales: (1) review the list of items identified through the review and the definition of the domain, (2) adapt the existing survey items for consistent wording with each other (and the response scale) and apply to patient care teams in the healthcare industry, (3) identify measurement gaps for each construct based on the comparison of all pieces of the definition with the available adapted items, and (4) write new items to fill these gaps and reach the desired number of items. All scales have at least five items and use a Likert-style response format ranging from 1 (strongly disagree) to 5 (strongly agree).
Descriptive statistics were employed to calculate the number of items retained after review, the number of articles contributing items retained after review, the number of constructs represented by the items retained, and their respective mappings onto the domains in the O&B framework. Also compared were the aforementioned statistics by literature area (healthcare sources vs. nonhealthcare sources).
Figure 1 summarizes the article selection flow and results. The initial search for candidate items yielded 468 unique articles dating from 1978 to 2014, 37 of which came from healthcare journals. This initial set of articles yielded 1,401 candidate items, 274 of which came from healthcare sources. From this set, 279 items aligned with an O&B domain, 33 of which came from healthcare sources. The final set of items retained was composed of 279 items drawn from 37 articles, five articles of which came from healthcare sources (33 healthcare items). For the full list of articles, sorted by O&B domain, included in the review after applying the inclusion criteria, see the appendix to this article, published as Supplemental Digital Content at http://links.lww.com/JHM/A24.
Retained items were drawn from scales representing 59 constructs related to teamwork, roles, trust, coordination broadly, and ancillary constructs, as shown in Table 3 . Table 4 presents the mapping of article-espoused constructs to O&B domains, and the number of items available for each. Although the authors were able to identify items for all domains in the O&B framework, nine O&B domains had fewer than five items associated with them. Only two domains, physical proximity and plans and rules, were directly represented as such both in the O&B framework and as standalone constructs in the literature. The remaining constructs contributed items that indirectly tapped the domains of interest in the O&B framework.
Scales were created for the following O&B domains:
- Common understanding
- Routines (three dimensions: providing a template for task completion, handling handoffs, and bringing groups together)
- Familiarity (three dimensions: anticipating and responding, storing knowledge, and developing trust)
- Objects and representations (three dimensions: direct information sharing, scaffolding, and acknowledging and aligning work)
- Plans and rules (three dimensions: defining responsibilities for tasks, allocating resources, and developing agreement)
- Creating common perspective
Final scale construction included 99 items across 11 of the 28 constructs. Of these items, 11 were adapted from items identified in this review and 88 were generated based on O&B's construct definition. The most common reasons for needing to generate an item were that (a) not enough items were identified in the review to form a full scale or dimension, (b) a piece of the definition was not captured by the existing items, or (c) the item did not easily apply to healthcare or team settings.
Table 5 provides the items for the scales for the three integrating conditions: accountability, proximity, and common understanding. Although these scales were developed based on items from published scales and the definitions of the corresponding O&B construct, the functionality of these items as a set still needs to be verified.
To build an evidence base for defining and improving care coordination, validated instruments for understanding and measuring care team member coordination are needed. This study sought to systematically review the healthcare and management literatures for validated survey instruments and items that capture mechanisms, processes, and integrating conditions of coordination, as conceptualized in the O&B framework of coordination. The authors are not aware of other systematic reviews of coordination since publication of the Okhuysen and Bechky (2009) review, or of any systematic reviews of care team coordination measurement.
The authors found many articles, primarily from the management literature, that address numerous constructs ancillary or related to coordination; however, after applying the inclusion criteria, only 7% of articles contributed items that directly captured one or more coordination domains as posited by the O&B review; less than 1% of articles were contributed by the healthcare literature. In addition, the study was unable to identify any complete scales that directly capture coordination domains as delineated in the O&B framework. Of the items identified, 51 constructs (e.g., teamwork) were represented, two of which—physical proximity and plans and rules—fell within both the O&B framework and standalone constructs in the literature; the other 49 constructs only indirectly measured an O&B domain. Although general support is found among observers for the relationship between coordination and care outcomes, the clinical utility of these item pools as diagnostic tools is contingent on further validation work. If the healthcare field is to obtain a better assessment than currently offered by the literature on how current healthcare teams coordinate as a team, much more work in the measurement area of team coordination must be done.
The present findings highlight the inadequacy of the current state of healthcare team coordination measurement that is based on theoretical coordination frameworks. As research shows that the mechanisms within the O&B framework are associated with enhanced coordination and task performance (Okhuysen & Bechky, 2009), future research must help remedy this gap in healthcare team coordination metrics identified by this review.
The authors also found a disparity in the abundance of items and scales available to measure different components of the O&B model. The number of items ranged from one (acknowledging and aligning work) to 32 (trust) items per construct. Some constructs—trust (32 items), familiarity (25 items), coordination broadly (23 items), defining responsibility (22 items), and developing agreement (20 items)—may have sufficient items from which to create appropriate scales and metrics. However, many other constructs have large gaps in the availability of valid measures. For example, more than a third of the O&B domains (accountability, acknowledging and aligning work, proximity, visibility, roles, creating a common perspective, monitoring and updating, routines, and bringing groups together) had fewer than five items with strong face validity.
These low numbers suggest that some constructs are being underused in the study of coordination. Compared with the more prevalent constructs, these lack clear operational definitions. Creating a common perspective, for example, is conceptually defined as building a shared mental model of the team's work (Okhuysen & Bechky, 2009). However, identifying the appropriate way to measure coordinative elements is challenging because of the different operational definitions that can be derived from conceptual definitions. For example, common perspective could be operationalized as the availability of formal sharing tools (e.g., protocols), verbatim recall of the group's work process, or employee perceptions of understanding the group work process. In addition, unclear definitional boundaries of the constructs can lead to overlapping definitional features across O&B domains. Scaffolding, for example, is a component of both accountability and predictability. Fleshing out more concrete definitions and crisper definitional boundaries for these constructs in healthcare could promote measurement by providing guidance on the specific components that demarcate the construct.
In addition to issues surrounding the operationalization of constructs, low numbers of items could also be due to a lack of general popularity or interest in the constructs. Many of the constructs with the most items (e.g., trust) have recently received considerable attention in popular media and organizations; many of the low-item-count constructs (e.g., monitoring and updating) have received much less popular attention. These less popular constructs may be undervalued or deemed less important or, similar to a spotlight effect, these low-item-count constructs are being overlooked due to the popularity of other constructs. Perhaps the overemphasis on trust measurement, for example, leads to a lack of attention on the importance of other key aspects of coordination that may ultimately detract from patients' faith and trust in their providers downstream. Although perhaps less fashionable, these low-item-count constructs represent key pieces of the coordination process. To allow for the comprehensive coordination measurement, focus on the measurement of these low-item constructs needs to occur.
As with any study, this review is not without limitations. First, the authors restricted the review to include only top management and psychology journals; other journals outside these limits may have yielded suitable items, perhaps in some of the less represented constructs. Second, although the research identified 279 items that mapped to O&B coordination constructs, these items do not necessarily form complete scales and the reliability and validity information of these items is not clear. If future research intends to use the current scales for coordination, the utility, reliability, and validity of these items as scales must be examined.
Finally, many of the constructs proposed in the O&B framework lacked concrete, clearly bounded definitions, making categorization into a single construct difficult for some items. To address the problem of vague definitions, the research used methods from qualitative analysis: Each item was discussed and compared to the existing definition by the team, whereby clarifying boundaries and documenting decisions occurred in real time, with consultation of earlier decisions for consistency. Nevertheless, future development of thorough definitions of coordinative constructs is essential for progressing the study and measurement of coordination.
Care coordination is considered an essential component of high-quality care. However, care coordination is a vague term in healthcare, without clear definitions or metrics that are evidence based. Although few standardized and comprehensive measures exist, the current study supports this comprehensive conceptual model and offers some examples of validated measures to fit aspects of that model. Consistent with Okhuysen and Bechky's (2009) findings, which highlighted the fragmented nature of coordination, the present findings call for dedicated research that builds on this work to develop valid, reliable, orthogonal scales of the operating mechanisms and integrating conditions that make successful coordination possible. Additional research is needed to develop novel metrics for aspects of the model that lack valid measures. This study provides an important and critical step in developing an evidence base for care coordination based on valid conceptual and measurement models.
In healthcare in particular, future work should also concentrate on applying and adapting measurements to coordination-rich scenarios, such as outpatient referrals, patient navigation, and institution-to-institution handoffs.
The work reported herein was funded by the U.S. Department of Veterans Affairs Health Services Research & Development Service, grants # IIR 12-383 and CIN 13-413. The salaries of all authors were funded at least in part by this funding.
Bal R., Mastboom F., Spiers H. P., Rutten H. (2007). The product and process of referral: Optimizing general practitioner-medical specialist interaction through information technology. International Journal of Medical Informatics
, 76(Suppl. 1), S28–S34.
Berendsen A. J., de Jong G. M., Meyboom-de Jong B., Dekker J. H., Schuling J. (2009). Transition of care: Experiences and preferences of patients across the primary/secondary interface—A qualitative study. BMC Health Services Research
, 9, 62.
Forrest C. B., Glade G. B., Baker A. E., Bocian A., von Schrader S., Starfield B. (2000). Coordination of specialty referrals and physician satisfaction with referral care. Archives of Pediatrics & Adolescent Medicine
, 154, 499–506.
Helfrich, C. D., Dolan, E. D., Simonetti, J., Reid, R. J., Joos, S., Wakefield, B. J., & Nelson, K. (2014). Elements of team-based care in a patient-centered medical home are associated with lower burnout among VA primary care employees. Journal of General Internal Medicine
, 29, 659-666.
Helfrich C. D., Li Y. F., Sharp N. D., Sales A. E. (2009). Organizational readiness to change assessment (ORCA): Development of an instrument based on the Promoting Action on Research in Health Services (PARIHS) framework. Implementation Science
, 4(1), 1.
Hysong S. J., Esquivel A., Sittig D. F., Paul L. A., Espadas D., Singh S., Singh H. (2011). Towards successful coordination of electronic health record based-referrals: A qualitative analysis. Implementation Science
, 6, 84.
Keller R. T. (2006). Transformational leadership, initiating structure, and substitutes for leadership: A longitudinal study of research and development project team performance. Journal of Applied Psychology
, 91(1), 202.
Kleingeld A., van Mierlo H., Arends L. (2011). The effect of goal setting on group performance: A meta-analysis. Journal of Applied Psychology
, 96(6), 1289.
Marks M. A., Panzer F. J. (2004). The influence of team monitoring on team processes and performance. Human Performance
, 17(1), 25–41.
McDonald, K., Schultz, E., Albin, L., Pineda, N., Lonhart, J., Sundaram, V., & Malcolm, E. (2010). Care coordination measures atlas. Agency for Healthcare Research and Quality. Retrieved from https://archive.ahrq.gov/professionals/systems/long-term-care/resources/coordination/atlas/index.html
McDonald, K., Sundaram, V., Bravata, D., Lewis, R., Lin, N., Kraft, S., & Owens, D. K. (2007). Closing the quality gap: A critical analysis of quality improvement strategies. Vol. 7—Care coordination
(Report No. 04(07)-0051-7). Rockville, MD: Agency for Healthcare Research and Quality.
Moher D., Liberati A., Tetzlaff J., Altman D. G. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine
, 151, 264–269.
Okhuysen G. A., Bechky B. A. (2009). Coordination in organizations: An integrative perspective. Academy of Management Annals
, 3, 463–502.
Salas E., Sims D. E., Burke C. S. (2005). Is there a “big five” in teamwork? Small Group Research
, 36, 555–599.
Stetler C. B., Damschroder L. J., Helfrich C. D., Hagedorn H. J. (2011). A guide for applying a revised version of the PARIHS framework for implementation. Implementation Science
, 6(1), 1.
Stille C. J., Jerant A., Bell D., Meltzer D., Elmore J. G. (2005). Coordinating care across diseases, settings, and clinicians: A key role for the generalist in practice. Annals of Internal Medicine
, 142, 700–708.
Stout R. J., Salas E., Carson R. (1994). Individual task proficiency and team process behavior: What's important for team functioning? Military Psychology
, 6, 177–192.
Sui Y., Wang H., Kirkman B. L., Li N. (2015). Understanding the curvilinear relationships between LMX differentiation and team coordination and performance. Personnel Psychology
, 69(3), 559–597.
Supplemental Digital Content
© 2018 Foundation of the American College of Healthcare Executives