Development and Testing of a Conceptual Framework for Interprofessional Collaborative Practice

INTRODUCTION Validated conceptual frameworks are needed to guide interprofessional research in order to build a systematic body of knowledge of interprofessional collaborative practice (ICP). A conceptual framework derived from an extensive review of the interprofessional literature was developed. In the framework, constructs that include personal factors (i.e., beliefs in interprofessional collaboration, flexibility, trust, cooperation, and communication skills) and situational factors (i.e., leadership, empowerment, and support structures) are posited to influence effective ICP. ICP is conceptualized as understanding of roles, interdependence, knowledge exchange, and collective ownership of goals. Consequences of ICP include improved patient, organizational, and team and personal work behaviours and attitudes. METHODS A preliminary study was conducted to determine the relationships among the constructs in the conceptual framework with a sample of 117 interprofessional practitioners in Manitoba, Canada. Participants completed a survey derived from modified existing measures. RESULTS Exploratory factor analyses provided construct validity for the measures, and Cronbach alpha reliabilities were acceptable. CONCLUSION There is encouraging preliminary empirical support for the conceptual framework with trust, cooperation, communication skills, and support structures predictors of ICP, and ICP a predictor of all outcomes identified in the framework. Received: 12/13/2013 Accepted: 05/27/2014 Published: 09/23/2014 © 2014 Stutsky et al. This open access article is distributed under a Creative Commons Attribution License, which allows unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. H IP & Interprofessional Collaborative Practice ORIGINAL RESEARCH 2(2):eP1066 | 2 Introduction The interprofessional literature has been described as atheoretical (Reeves et al., 2011). Elements of interprofessional collaboration are poorly conceptualized, and a consistent theoretical framework to guide research and build a body of evidence to inform interprofessional education (IPE) and interprofessional collaborative practice (ICP) is missing (Reeves et al.). In this paper, we present a conceptual framework for ICP derived from an extensive review of the existing interprofessional literature. The results of a preliminary study designed to test the validity of this framework in a Canadian healthcare setting are described. Literature Review/Conceptual Framework Numerous databases including PubMed, Scopus, Embase, CINAHL, and PsycInfo were searched to examine theoretical and research papers relating to IPE and ICP. Main keywords included interprofessional collaboration, interprofessional relations, interdisciplinary collaboration, interprofessional education, teamwork, and patient care team. Of the 900 electronic abstracts reviewed, only research-based papers were selected for further review. Our proposed conceptual framework for ICP (see Figure 1, following page) was formulated by synthesizing concepts from 97 research papers and key national and international reports. We posit antecedents, that include personal and situational factors, influence ICP. ICP in turn results in a variety of consequences including improved work behaviors and attitudes, organizational outcomes, and patient outcomes. See Table 1 (page 3) for definitions/ descriptions of terms. Antecedents of ICP Researchers identified several antecedents to ICP. We separated the antecedents into personal factors that are controlled internally by an individual, and situational factors that professionals are exposed to within the workplace that either support or deter ICP. For ICP to be successful, interprofessional practitioners must first truly believe in the concept of ICP (Oandasan & Reeves, 2005; Parker Oliver, Wittenberg-Lyles, & Day, 2007) and have experience with being able to negotiate an interprofessional plan when disagreements occur (Bronstein, 2003; McGrail, Morse, Glessner, & Gardner, 2009). Relational skills are a precursor to ICP (McGrail et al., 2009), and interprofessional practitioners must have already developed strong cooperation (Gaboury, Lapierre, Boon, & Moher, 2011) and communication skills (Atwal & Caldwell, 2002; Havens, Vasey, Gittell, & Lin, 2010). Trust is critical, and according to D’Amour, Goulet, Labadie, San Martin-Rodriguez, and Pineault (2008), ICP is possible only when there is trust in each other’s competencies. D’Amour et al. claim that professionals place themselves in vulnerable positions all the time and take risks in trusting each other; however, when there is high uncertainty or low trust, professionals will avoid collaboration and hold onto their own responsibilities for patient care. Overall, it is important that individuals are comfortable with themselves and their own competencies before relying on others (Clark, 2011). Situational antecedents that either support or deter ICP include leadership (Canadian Interprofessional Implications for Interprofessional Practice • Healthcare leaders can use the proposed conceptual framework as a guide for facilitating interprofessional collaborative practice in organizations to enhance patient safety and quality. • Interprofessional education can be strengthened with a validated framework for interprofessional collaborative practice. • With a validated framework, clinical professionals will become more aware of the importance of individual attitudes and behaviors and team interactions to improving patient safety and quality H IP & ISSN 2159-1253 Health & Interprofessional Practice | commons.pacificu.edu/hip 2(2):eP1066 | 3 Health Collaborative [CIHC] 2010; D’Amour et al., 2008; Oandasan & Reeves, 2005), empowerment (Tresolini & Pew-Fetzer Task Force, 1994), and support structures (Clark, 2011; McGail et al., 2009). Both central and local leadership is needed to promote collaboration, eliminate barriers (D’Amour et al., 2008), and promote an effective team culture (Clark, 2011). Leadership is also needed to create an empowering environment that includes having access to information, support, resources, and the opportunity for growth and mobility (Kanter, 1977; 1997). Support structures necessary for ICP include having adequate time for sharing knowledge and patient-related information (Atwal & Caldwell, 2002; Clark, 2011; Gaboury, Bujold, Boon, & Moher, 2009) and integrating daily collaborative behaviors into dayto-day functioning (Ottawa Hospital, n.d.). Support can also take the form of emotional support, helpful advice, or hands-on assistance from superiors, peers, or interprofessional practitioners (Kanter, 1977; 1997). Additional support structures include having formal procedures and mechanisms for facilitating dialogue (Parker Oliver et al., 2007) such as written policies and/or guidelines and various educational opportunities such as in-services and grand rounds. Interprofessional Collaborative Practice The World Health Organization (2010) defines collaborative practice as occurring “when multiple health workers from different professional backgrounds provide comprehensive services by working with patients, their families, carers, and communities to deliver the highest quality of care across settings” (p. 13). This definition is consistent with the model of Relationship-Centered Collaborative Care (RCC) that includes three key relationships necessary for effective patient care: (a) the patient-practitioner relationship, (b) the practitioner-practitioner relationship, and (c) the community-practitioner relationship (Tresolini & Pew-Fetzer Task Force, 1994). In the patient-practitioner dimension, the essential role of the patient as a partner in the interprofessional care process is emphasized, while in the practitioner-practitioner dimension, collaboration among healthcare providers is stressed. In the community-practitioner dimension, the need Figure 1. Conceptual framework for interprofessional collaborative practice H IP & Interprofessional Collaborative Practice ORIGINAL RESEARCH 2(2):eP1066 | 4 Table 1. Survey Subscales, Definition/Description, Sample Survey Items, Final Number of Survey Items, and Alpha Reliabilities Scales/Subscales Definition/Description Sample Survey Items Final # a Antecedents: Personal Beliefs in IPC2 Extent to which professionals identify strengths in interprofessional collaboration. At my workplace, interprofessional collaboration is a better answer than non-collaborative care to meet the patient’s/family’s biopsychosocial needs. 4 0.88 Flexibility1 “Deliberate...role-blurring...and includes reaching productive compromises in the face of disagreement” (Bronstein, 2003, p. 300-301). I am willing to take on tasks outside of my job description when that seems important. 2 0.69 Trust2 The confidence and reliance that interprofessionals have with one another. I never have to double-check information given to me by other team members. 5 0.83 Cooperation2 The manner in which interprofessionals work together for a common goal. I coordinate my efforts with professionals from other disciplines. 2 0.67 Communication3 The ease and effectiveness with which interprofessionals communicate with each other. I find it easy to ask the advice of others in my team. 3 0.68 Antecedents: Situational Leadership3 A team leader’s ability to foster ICP and “set and communicate clear goals and expectations and facilitate their implementation” (Temkin-Greener et al., 2004, p. 481). The team leader fosters professionals from different disciplines to work together. 3 0.76 Empowerment5 Having access to information, support, resources, and the opportunity for growth and mobility (Kanter, 1977; 1997). Overall, my current work environment empowers me to accomplish my work in an effective manner. 2 0.87 Support Structures1,4 Having the physical space, time, policies and procedures, and formal mechanisms to support ICP. Within my w


Introduction
The interprofessional literature has been described as atheoretical (Reeves et al., 2011).Elements of interprofessional collaboration are poorly conceptualized, and a consistent theoretical framework to guide research and build a body of evidence to inform interprofessional education (IPE) and interprofessional collaborative practice (ICP) is missing (Reeves et al.).In this paper, we present a conceptual framework for ICP derived from an extensive review of the existing interprofessional literature.The results of a preliminary study designed to test the validity of this framework in a Canadian healthcare setting are described.

Literature Review/Conceptual Framework
Numerous databases including PubMed, Scopus, Embase, CINAHL, and PsycInfo were searched to examine theoretical and research papers relating to IPE and ICP.Main keywords included interprofessional collaboration, interprofessional relations, interdisciplinary collaboration, interprofessional education, teamwork, and patient care team.Of the 900 electronic abstracts reviewed, only research-based papers were selected for further review.Our proposed conceptual framework for ICP (see Figure 1, following page) was formulated by synthesizing concepts from 97 research papers and key national and international reports.We posit antecedents, that include personal and situational factors, influence ICP.ICP in turn results in a variety of consequences including improved work behaviors and attitudes, organizational outcomes, and patient outcomes.See Table 1 (page 3) for definitions/ descriptions of terms.

Antecedents of ICP
Researchers identified several antecedents to ICP.We separated the antecedents into personal factors that are controlled internally by an individual, and situational factors that professionals are exposed to within the workplace that either support or deter ICP.For ICP to be successful, interprofessional practitioners must first truly believe in the concept of ICP (Oandasan & Reeves, 2005;Parker Oliver, Wittenberg-Lyles, & Day, 2007) and have experience with being able to negotiate an interprofessional plan when disagreements occur (Bronstein, 2003;McGrail, Morse, Glessner, & Gardner, 2009).
Relational skills are a precursor to ICP (McGrail et al., 2009), and interprofessional practitioners must have already developed strong cooperation (Gaboury, Lapierre, Boon, & Moher, 2011) and communication skills (Atwal & Caldwell, 2002;Havens, Vasey, Gittell, & Lin, 2010).Trust is critical, and according to D' Amour, Goulet, Labadie, San Martin-Rodriguez, and Pineault (2008), ICP is possible only when there is trust in each other's competencies.D' Amour et al. claim that professionals place themselves in vulnerable positions all the time and take risks in trusting each other; however, when there is high uncertainty or low trust, professionals will avoid collaboration and hold onto their own responsibilities for patient care.Overall, it is important that individuals are comfortable with themselves and their own competencies before relying on others (Clark, 2011).
Situational antecedents that either support or deter ICP include leadership (Canadian Interprofessional

Implications for Interprofessional Practice
• Healthcare leaders can use the proposed conceptual framework as a guide for facilitating interprofessional collaborative practice in organizations to enhance patient safety and quality.
• Interprofessional education can be strengthened with a validated framework for interprofessional collaborative practice.
• With a validated framework, clinical professionals will become more aware of the importance of individual attitudes and behaviors and team interactions to improving patient safety and quality  (Tresolini & Pew-Fetzer Task Force, 1994), and support structures (Clark, 2011;McGail et al., 2009).Both central and local leadership is needed to promote collaboration, eliminate barriers (D' Amour et al., 2008), and promote an effective team culture (Clark, 2011).Leadership is also needed to create an empowering environment that includes having access to information, support, resources, and the opportunity for growth and mobility (Kanter, 1977;1997).Support structures necessary for ICP include having adequate time for sharing knowledge and patient-related information (Atwal & Caldwell, 2002;Clark, 2011;Gaboury, Bujold, Boon, & Moher, 2009) and integrating daily collaborative behaviors into dayto-day functioning (Ottawa Hospital, n.d.).Support can also take the form of emotional support, helpful advice, or hands-on assistance from superiors, peers, or interprofessional practitioners (Kanter, 1977;1997).Additional support structures include having formal procedures and mechanisms for facilitating dialogue (Parker Oliver et al., 2007) such as written policies and/or guidelines and various educational opportunities such as in-services and grand rounds.

Interprofessional Collaborative Practice
The World Health Organization (2010) defines collaborative practice as occurring "when multiple health workers from different professional backgrounds provide comprehensive services by working with patients, their families, carers, and communities to deliver the highest quality of care across settings" (p.13).This definition is consistent with the model of Relationship-Centered Collaborative Care (RCC) that includes three key relationships necessary for effective patient care: (a) the patient-practitioner relationship, (b) the practitioner-practitioner relationship, and (c) the community-practitioner relationship (Tresolini & Pew-Fetzer Task Force, 1994).In the patient-practitioner dimension, the essential role of the patient as a partner in the interprofessional care process is emphasized, while in the practitioner-practitioner dimension, collaboration among healthcare providers is stressed.
In the community-practitioner dimension, the need  Extent to which professionals identify strengths in interprofessional collaboration.
At my workplace, interprofessional collaboration is a better answer than non-collaborative care to meet the patient's/family's biopsychosocial needs.4 0.88 Flexibility 1 "Deliberate…role-blurring…and includes reaching productive compromises in the face of disagreement" (Bronstein, 2003, p. 300-301).
I am willing to take on tasks outside of my job description when that seems important.The confidence and reliance that interprofessionals have with one another.Having access to information, support, resources, and the opportunity for growth and mobility (Kanter, 1977;1997).As per below four subscales.13 0.78 Four subscales of ICP: Professionals' knowledge and understanding of their role and the roles of others within an interprofessional collaborative environment.
My colleagues from other professional disciplines do not treat me as an equal.

•
Interdependence 1 "The occurrence of and reliance on interactions among professionals whereby each is dependent on the other to accomplish his or her goals and tasks" (Bronstein, 2003, p. 299).
Working with professionals from other disciplines is not important in my ability to help patients/families."The perceived effectiveness of the team with respect to…the ability to meet patient (and family) care needs and outcomes (Temkin-Greener et al., 2004, p. 481).
Our team does a good job in meeting patient and family member needs.

Conflict 3
The degree to which disciplines disagree over the sharing of responsibilities and group decisions.to consider the patient's community, including one's family situation and available community resources to support health goals is highlighted.At a higher level, the role of the practitioner in enhancing community relationships and health is emphasized.Overall, the RCC model forms the underlying basis for ICP in our framework, because we see these relationships as foundational for comprehensive patient care in any setting.In examining other existing models and definitions of ICP, we found that effective relationships were either explicitly or implicitly identified as a fundamental component of ICP (Bronstein, 2003;CIHC, 2010;D' Amour et al., 2008;Orchard, Curran, & Kabene, 2005;Safran, Miller, & Beckman, 2006).Although the RCC model has been used primarily to guide education and practice, it has been used in rare cases for interprofessional research (Dix, Steggles, Baptiste, & Risdon, 2008;Gaboury et al., 2011).
Based on the literature, we conceptualized ICP as a four dimensional construct including understanding of roles, interdependence, knowledge exchange, and collective ownership of goals.Understanding of roles is key to the practitioner-practitioner dimension of RCC and the other two relationships where the patient's role must be understood (Tresolini & Pew-Fetzer Task Force, 1994).We found that interprofessional practitioners that work collaboratively are comfortable explaining their own role to other professionals, they put aside turf and role issues (Clark, 2011), they are aware of their own and other's limitations, and they have professional maturity and intellectual curiosity (Gaboury et al., 2009).
Regarding interdependence and knowledge exchange, interprofessional practitioners talk about the need for equality in terms of power relationships between professionals (Gaboury et al., 2009) and their willingness to share information (Atwal & Calwell, 2002).Respect for others and their knowledge is important (Clark, 2011), as well as an understanding of the value base of other professionals (Atwal & Caldwell, 2002).Interdependence and knowledge exchange among interprofessional practitioners and the patient is required for all three dimensions of RCC to be effective (Tresolini & Pew-Fetzer Task Force, 1994).
Collective ownership of goals is essential for effective ICP (Atwal & Caldwell, 2002).A vital component in developing and achieving healthcare goals is that patients and their families are active participants in the process (Bronstein, 2003).In our framework, the patient, which is a collective term referring to the patient, client, family, and/or community, is considered a key decision maker in terms of ownership of goals.Orchard et al. (2005) agree that the integral role of the patient in care planning and decision making is often overlooked in explanations of ICP.When looking at the relationship to RCC, we believe that collective ownership of goals transcends the patient-practitioner and community-practitioner dimensions (Tresolini & Pew-Fetzer Task Force, 1994).

Design and Sample
A preliminary study, using a descriptive correlational design, was used to test the proposed relationships in the model excluding patient outcomes.The sample included 364 healthcare professionals currently employed in a regional health authority (RHA) in northern Manitoba, Canada.Sites included three hospitals, three long-term care facilities, and four primary healthcare centers.Participants were regulated direct care providers/supervisors who were involved in planning care and/or team decision making.

Procedure
After ethical approval from the University of Manitoba, survey packages were couriered to an onsite research manager who distributed the packages.Survey respondents received a letter of information and a $2.00 gift card was included.Anonymous surveys were returned directly to the research team and data were entered into a statistical analysis program.

Instrument
We constructed the Interprofessional Collaborative Practice Survey (ICPS) using selected items from existing instruments to measure all constructs in the framework including ICP and its antecedents and consequences (see Table 1).ICP is measured by four subscales and 13 items corresponding to the model constructs (i.e., understanding of roles, interdependence, knowledge exchange, and collective ownership of goals).Two items, based on Sicotte, D' Amour, and Moreault's (2002) work, measuring perceived degree of collaboration, were used to establish concurrent/construct validity of the newly developed measure of ICP.The total ICP score was significantly correlated to collaboration in both primary care and regional settings (r=0.60 and r=0.48 [p<.05] respectively).Personal antecedents are measured by five subscales (16 items), while situational antecedents are measured by three subscales (10 items).Six subscales (16 items) are used to measure ICP consequences.Due to length, redundant items were removed from original scales with further items deleted as a result of reliability testing and factor analysis.Minor wording changes were made for applicability to the population.The final tool consists of 55 items with nine demographic items.
High scores on the ICPS (including conflict) indicate a higher level of the construct with items rated on a 5-point scale of strongly disagree (1) to strongly agree (5).Scores are summed and averaged for each subscale.Twelve items are reverse coded.Perceptions of the state of patient safety, quality of care, and degree of collaboration were rated on a 5-point scale from low (1) to high (5).Patient safety (1 question) and quality of patient care (1 question) scores are averaged, while collaboration (2 questions) scores are summed and averaged.

Data Analysis
The Statistical Package for the Social Sciences (SPSS) was used to generate descriptive statistics and examine initial correlations between ICP and its antecedents and consequences using Pearson product-moment correlation coefficient.Armstrong (1981) supports the use of parametric statistics with ordinal level data.
Due to the number of changes to original subscales, the ICPS was essentially considered a new instrument; therefore, exploratory factor analyses (EFA) were used to examine the construct validity of the instrument and reliability was estimated using Cronbach's coefficient alpha.Path analysis techniques in AMOS (Arbuckle, 2005) were used to test the hypothesized paths in our conceptual framework.Path analysis within structural equation modeling analysis allowed us to test all hypothesized paths simultaneously and to take errors into account to get more precise estimates of the effects (Kline, 2011).

Sample Description
The response rate was 32%  factors (see Table 4).Four factors explained 74.71% of the cumulative variance for work behaviors and attitudes (see Table 5).Reliability of the ICPS was adequate with values ranging from 0.67 to 0.88 for antecedent and consequences scales, and 0.78 for the overall ICP scale (see Table 1).

Descriptive Statistics for Major Study Variables
As seen in In the path analysis, 3 of 5 paths for personal factors were significant predictors of ICP, with communication skills highest (β=0.33) and trust the lowest (β=0.17).Beliefs in interprofessional collaboration and flexibility were not significant predictors.Access to ICP support structures was the only significant situational predictor (β=0.33).Overall ICP was a significant predictor of all individual, team, and organizational outcomes (β=0.42-0.66)(see Figure 2, following page).The model had a fair fit for a newly developed framework (Comparative Fit Index [CFI] 0.81, Normed Fit Index [NFI]0 .77,and Root Mean Square Error of Approximation Index [RMSEA]0.14).

Discussion
Given that the participants had not been exposed to workplace-based IPE, it is encouraging that each of the four constructs making up ICP were at moderate to moderately high levels.Exposure to IPE may have provided them with a greater understanding of each other's roles and the need for interdependence and knowledge exchange to collectively generate patient goals that result in better patient outcomes.
The strongest personal predictors of ICP included trust, cooperation, and communication skills.The influence of personal factors/relational skills as critical antecedents to ICP is consistent with the literature (Atwal & Caldwell, 2002;CIHC, 2010;D' Amour et al., 2008;Gaboury et al., 2011;Havens et al., 2010;Safran et al., 2006;Temkin-Greener et al., 2004), andMcGrail et al. (2009) indicated that without strong relational skills, ICP will not be as effective.Beliefs in interprofessional collaboration was not found to be a significant predictor of ICP, which was surprising in light of previous research.Although interprofessional practitioners held strong beliefs in ICP, they were not predictive of the degree of ICP they experienced in their work setting.It is possible that the restricted range of this variable may have been a factor.Similarly, the non-significant influence of flexibility on ICP was not anticipated.Previous researchers identified the importance of the need for professionals to be flexible and willing to sacrifice a degree of autonomy to reach productive compromises when disagreement exists (Bronstein, 2003;McGail et al., 2009;Parker Oliver et al., 2007).facilitate ICP, then it is possible that negative work behaviors and attitudes may subside.

Limitations
Limitations included a small sample size and low response rate (32%).The small sample size limited the type of analysis that could be completed, and structural equation modeling would be used in the future with a large sample.Factors that may have decreased the response rate included distribution over the summer months, the length of the survey, the sensitive nature of the questions, and the concern of potentially being identified, given low numbers of allied health professionals, in particular.The results are potentially biased as those who completed the survey may have different feelings about ICP compared to those that elected not to complete the survey.The distribution of types of interprofessional practitioners in the sample is not equal, and only one health region was used to obtain the sample.Continual refinement and validation of the ICPS tool is needed including the addition of items measuring patient outcomes.Further validation of the framework will assist in broadening our understanding of ICP and its influence on quality of patient care.Finally, it is critical that patient outcomes be captured in future studies.

Conclusion
We have encouraging empirical support for our proposed ICP conceptual framework.Such a framework allows researchers to continue to build a sound body of evidence related to interprofessional practice that can be used by healthcare leaders, educators, and clinical professionals at all levels.
Healthcare leaders can use the framework as a guide for facilitating ICP in healthcare organizations to improve patient outcomes and enhance patient safety and quality of care.Specifically, leaders need to ensure that professionals are exposed to a working environment that enhances the development of personal factors/relational skills such as trust, cooperation, and communication skills.Healthcare leaders need to model effective leadership, provide the necessary support structures to enhance ICP, and create an empowering work environment.Within an empowering work environment, professionals will be better able to understand each other's roles, work interdependently, exchange knowledge, and collectively develop patient care goals.
The framework can be used by educators to strengthen IPE curricula by focusing education on specific skill development such as communication skills, or ICP concepts such as understanding of roles.During clinical placements, students need to be exposed to effective ICP so that they can learn how to work as an interdependent team that freely exchanges knowledge and works together to develop collaborative care plans.
Clinical professionals can ensure that those relational skills that are known to improve outcomes are selfassessed and enhanced if needed.Clinical professionals know about deficiencies in support structures in their work environments and can address those deficiencies by bringing them to the attention of their leaders.In addition, clinical professionals can make a concerted effort to focus on the four constructs of ICP in their everyday practices.
A primary goal for developing the conceptual framework for ICP was to be able to provide a guide that could be used to ultimately improve patient outcomes, patient safety, and quality of patient care.
With further research and validation, it is hoped that the proposed conceptual framework for ICP will be a valuable tool used by healthcare leaders, educators, and clinical professionals in meeting that goal.

Table 1 .
Survey Subscales, Definition/Description, Sample Survey Items, Final Number of Survey Items, and Alpha Reliabilities

Table 2 .
Rotated ractor loadings for interprofessional collaborative practice

Table 4 .
Rotated factor loadings for antecedents: Situational factors

Table 6 .
Means, Standard Deviations, Cronbach's Coefficient Alpha Reliabilities, Number of Subscale Items, and