The state of the science of interprofessional collaborative practice: A scoping review of the patient health-related outcomes based literature published between 2010 and 2018

Introduction If interprofessional collaborative practice is to be an important component of healthcare reform, then an evidentiary base connecting interprofessional education to interprofessional practice with significantly improved health and healthcare outcomes is an unconditional necessity. This study is a scoping review of the current peer reviewed literature linking interprofessional collaborative care and interprofessional collaborative practice to clearly identified healthcare and/or patient health-related outcomes. The research question for this review was: What does the evidence from the past decade reveal about the impact of Interprofessional collaborative practice on patient-related outcomes in the US healthcare system? Materials and methods A modified preferred reporting items for systematic reviews and meta-analyses (PRISMA) approach was followed. Results Of an initial 375 articles retrieved 20 met review criteria. The most common professions represented in the studies reviewed were physicians, pharmacists and nurses. Primary care was the most common care delivery setting and measures related to chronic disease the most commonly measured outcomes. No study identified negative impacts of interprofessional collaborative practice. Eight outcome categories emerged from a content analysis of the findings of the reviewed studies. Conclusions The results suggest a need for more research on the measurable impact of interprofessional collaborative practice and/or care on patient health-related outcomes to further document its benefits and to explore the models, systems and nature of collaborations that best improve population health, increase patient satisfaction, and reduce cost of care.


Introduction
If interprofessional collaborative practice is to be an important component of healthcare reform, then an evidentiary base connecting interprofessional education to interprofessional practice with significantly improved health and healthcare outcomes is an unconditional necessity. This study is a scoping review of the current peer reviewed literature linking interprofessional collaborative care and interprofessional collaborative practice to clearly identified healthcare and/or patient health-related outcomes. The research question for this review was: What does the evidence from the past decade reveal about the impact of Interprofessional collaborative practice on patient-related outcomes in the US healthcare system?

Materials and methods
A modified preferred reporting items for systematic reviews and meta-analyses (PRISMA) approach was followed.

Results
Of an initial 375 articles retrieved 20 met review criteria. The most common professions represented in the studies reviewed were physicians, pharmacists and nurses. Primary care was the most common care delivery setting and measures related to chronic disease the most commonly measured outcomes. No study identified negative impacts of interprofessional collaborative practice. Eight outcome categories emerged from a content analysis of the findings of the reviewed studies. PLOS

Introduction
Stimulated by a multitude of factors, including a heightened commitment to reforming healthcare delivery, attention to interprofessional practice and education has grown exponentially over the past decade. [1,2]. [2][3][4] In 2008 Berwick et al., published their seminal article on the triple aim, offering a roadmap for healthcare delivery reform. [5] Along with the Institute of Medicine's (IOM) 2001 report, Crossing the Quality Chasm, [6] and the World Health Organization's (WHO) 2013 report Transforming and scaling up health professionals' education and training, [7] these articles galvanized a renewed interest in interprofessional practice and education. Conceptually, these three works explicitly promoted interprofessional healthcare teams as a strategy to improve health services and outcomes. The IOM and WHO reports charged health professions' programs to incorporate interprofessional education (IPE) into their training with the conviction that these efforts would lead to enhanced communication and care coordination to advance population health, reduce healthcare costs, and improve patient health-related outcomes. Ultimately, if interprofessional practice is to be an important component of healthcare reform, then an evidentiary base connecting interprofessional collaborative practice (IPCP) with significantly improved health and healthcare outcomes is an unconditional necessity. [2,3,8] The 2013 WHO report gave a conditional recommendation of IPE, suggesting rigorous research be included in its implementation. Continued interprofessional education endeavors can only be justified if the return on investment from an IPCP model yields a positive impact on measurable patient health-related outcomes.
A 2014 scoping review [3] noted that despite its multi-decade history, interprofessional practice and education-related research focused mainly on short-term changes such as improved knowledge, skills and attitudes of learners or on intermediate policy changes in either education or clinical settings, but not on patient health and/or healthcare related outcomes. [3] As Brandt et.al. concluded ". . . little of the literature reviewed focused on population health or patient health outcomes, and none on the reduction in the cost of healthcare." [3] A 2016 editorial, [4] describing the key findings of the 2015 IOM's report on interprofessional practice and education, concluded that whether or not interprofessional practice and education improves clinical outcomes remains uncertain since few studies explicitly map interprofessional practice and education to health related outcome measures. [4] Since most health professions programs, including pharmacy, medicine, public health, nursing and dentistry, [9][10][11][12] mandate IPE it is understandable that most research focused primarily on educational methods and learner outcomes. However, this leaves a gap in the evidence linking interprofessional practice and/or care to patient health-related outcomes. As embedded as interprofessional education is in health professions training, for justify these efforts, research must extend to include health-related patient outcomes. Yet despite the mandate for IPE, as John Gilbert notes it remains, "a great truth awaiting scientific confirmation." An updated WHO report also supported the need for more evidence connecting IPE training to improved health outcomes.
This study is a scoping review [13] of the current peer-reviewed literature that connects interprofessional collaborative care and IPCP to clearly identified patient health-related outcomes. Scoping reviews are a process of mapping an existing literature or evidence base to answer a question. [13] For this study we explored the question: What does the evidence over the past decade reveal about the effectiveness of interprofessional collaborative care and/or interprofessional collaborative practice in the US healthcare system on patient related outcomes? In addition, the study also sought to identify gaps in the existing literature to help guide future research.

Materials and methods
For this review interprofessional collaborative care was defined as: the provision of comprehensive health services to patients by multiple caregivers from different professions (e.g., medicine, nursing, pharmacy, dentistry) who work collaboratively to deliver quality care within and across settings (e.g., hospital, primary care, dental clinics, hospice care). Interprofessional collaborative practice occurs when healthcare providers work with people from within their own profession, with people outside their profession and with patients and their families. When healthcare providers work collaboratively, they seek common goals and can analyze and address any problems that arise. Care is coordinated according to patients' needs and patient outcomes are explicitly tied to how care is provided (although without exception the influences on patient outcomes are multifactorial).
The review followed a modified preferred approach for reporting items for systematic reviews and meta-analyses (PRISMA). [14] The PRISMA approach is organized by five distinct elements or steps: beginning with a clearly formulated question, using the question to develop clear inclusion criteria to identify relevant studies, an approach to appraise the studies or a subset of the studies, a summary of the evidence using an explicit methodology, and interpreting the findings of the review.
The literature search was limited to peer-reviewed articles, from the time span of 2010-2018. The choice of the year 2010 as the floor was grounded in the historical fact that the US Affordable Care Act [15] became law during that year and 2018 chosen as the ceiling year because that was the last full year that had ended when the review commenced. Unpublished/ grey literature, opinion pieces/essays, letters to the editor, and review papers were excluded from this review. The literature search was further limited to papers written in English and based in the US. The rationale to focus solely on US based literature is that the US-based health care system of practice and reimbursement infrastructure are unique compared to other industrialized nations. Until the enactment of the Affordable Care Act, many low to moderateincome families didn't have access to medical home. The U.S. healthcare team is also more specialty based and focuses more primarily on clinician/patient model and not on team-based health care or universal insurance. This review aims to assess the impact of IPCP within the US healthcare system on patient related outcomes.
PubMed and Google Scholar were systematically searched to identify potentially relevant literature. Google Scholar was included because there is a growing literature [16,17] assessing the value of Google Scholar in relationship to other indexing databases such as PubMed. Current recommendations suggest including Google Scholar and PubMed for comprehensive health-related systematic review searches [18]. We decided to use both MeSH and broader terms in our search to capture all relevant studies. The initial search terms used for the review entailed the following: In keeping with the PRISMA approach the following five steps were adhered to.
Step 1: The initial research question for this review was: What does the outcome-derived evidence indicate about interprofessional collaborative care and/or interprofessional collaborative practice in the US healthcare system?
As the review of the literature was underway, two additional questions were formulated to guide the analyses. These were: 1. What patient health-related outcomes were measured in the literature reviewed?
2. What were the data-driven findings derived from the reviewed literature?
Step 2: The initial question directed the process for identifying the relevant work reviewed. The subsequent questions guided the analysis of the reviewed literature. The inclusion criteria the review proposed emerged directly from the question guiding the review.
Step 3: A systematic approach to appraise the studies was used. To assess the selected papers a comprehensive table that included the full paper citation and complete abstract along with the following five components comprising a checklist was developed and used: 1) Interprofessional Collaborative Care OR Interprofessional Collaborative Practice (yes/no), 2) United States (yes/no), 3) Practice Setting (e.g., acute, primary, hospice), 4) Peer-Reviewed (yes/no), 5) study design specified (yes/no), and 6) Data analyzed (identify specific outcomes) (yes/no). For a paper to be included in the review all of the criteria had to be met. Table 1 displays the definitions used for the inclusion criteria. Once the articles gleaned from the literature search were reviewed, the reasons for exclusion were determined by all of the study authors and recorded for depiction in a flowchart (see Fig 1) for discussion in the results section.
Step 4: The evidence presented in the articles reviewed was organized by study design (observational, randomized controlled trial, etc.), the type(s) of data analyzed, the study outcomes, and study results or findings. Organizing the information in this manner allowed for a comprehensive assessment of the reviewed articles. This information is displayed in Table 2 and is discussed in the results section. From this information (now data) categories were derived from the patient outcomes and findings of the included studies. Using a qualitative content analysis approach, study findings were also analyzed and organized into outcome categories by the group consensus of the authors. [18] The process/methodology used to derive categories is depicted in Fig 2 and the categories discussed in the results section.
Step 5: interpreting the findings. In the Discussion section of this paper, the findings of this review were interpreted in light of the contemporary manifestation of interprofessional collaborative practice and/or care in the US.

Results
Fig 1 illustrates the article selection process for this review. The search conducted in PubMed and Google Scholar yielded 375 papers-249 of which focused on interprofessional collaborative practice and 126 focused on interprofessional collaborative care. After an initial review by one of the study's authors (MNL), 129 articles tentatively met the study criteria and were then reviewed by the remaining authors. After full examination, a total of 355 articles failed to meet the inclusion criteria. Forty papers were excluded because they were review pieces, 113 papers were not US based, 92 papers did not have patient health-related outcomes, 21 papers did not address interprofessional collaborative practice or interprofessional collaborative care, 54 papers were opinion/commentary or program description papers, 31 papers were duplications, one paper was not peer-reviewed, and three papers were not available from any source. Ultimately, 20 papers [19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38] met the inclusion criteria for review. Table 2 displays the included papers summarized by study design, study setting, data analyzed and sample size, outcomes studied, and study findings. The selected studies utilized a variety of study designs including clinical trials, quality improvement and pre/post intervention studies. Ten studies were pre/post intervention studies, some of which were quality improvement studies. [20-25, 29, 30, 32, 35, 38] Two studies were clinical trials, [19,34] one of which was a randomized control trial. [34] Two studies were retrospective cohort studies [23,37] and one a prospective cohort study. [25] Two studies were cross-sectional, [28,35] one sequential mixed methods, [33] one a case study, [27] and one a case/control study. [31] In addition to the wide variety of study designs employed, sample sizes ranged widely from one paper reporting two case studies to a sample size [27] of 322,408 patients. [31] The case study project was difficult to assess since each case (two were included in the paper) entailed both a patient along with unspecified multiple healthcare providers as well as family members. [27]

Interprofessional Collaborative Care (IPC)
Occurs when healthcare is delivered by intentionally created, work groups that have a collective identity and shared responsibility for a patient or group of patients (e.g., rapid response team, palliative care team, primary care team, and operating room team).

Interprofessional Collaborative Practice (IPCP)
Occurs when multiple healthcare workers from different professional backgrounds work together with patients, families, caregivers, and communities to deliver the highest quality of care. When healthcare providers work collaboratively, they seek common goals and are able to analyze and address any problems that arise. Care is coordinated according to patients' needs.
Practice Settings hospital care: provide services to diagnose (laboratory, diagnostic imaging) and treat (surgery, medications, therapy) diseases for a short period of time; in addition, they usually provide emergency and obstetrical care specialty care: provide care for very specific types of diseases; for example, a psychiatric hospital nursing homes or long-term care facilities: provide long-term care for patients who need extra time to recover from an illness or injury before returning home, or for persons who can no longer care for themselves primary Care: provide services that do not require overnight hospitalization; the services range from simple surgeries to diagnostic testing or therapy home health care: provides nursing, therapy, personal care or housekeeping services in the patient's own home rehabilitation center: provides intensive physical and occupational therapy; includes inpatient and outpatient treatment.

hospice: provides supportive treatment to terminally ill patients and their families
Peer-Reviewed a process by which research for publication is evaluated by a group of experts in the appropriate field Data analyzed empirical referent that is analyzed in order to draw conclusions about the occurrence of a specific phenomena Sample Size number of entities (subjects, etc.) in a subset of a population selected for analysis Care or practice outcomes defined something that happens as a result of an activity or process, e.g., reduced blood pressure, reduced A1c, increased physical activity, reduced length of hospital stay, patient satisfaction, provider evaluation of provided care delivery https://doi.org/10.1371/journal.pone.0218578.t001   -the sum of asthma-related emergency department (ED) visits and hospitalizations at 9 months before, 9 months during, and 9 months after the intervention -Of 126 patients with asthma, the number of emergency department (ED) visits and/or hospitalizations decreased 30% during the intervention (p = 0.052) and then returned to pre-enrollment levels after the intervention was discontinued (p = 0.83) -The intervention reduced asthma-related ED visits and hospitalizations, however, the primary outcome was not statistically significant.   The specific outcomes measured in the studies reviewed varied greatly although some outcomes overlapped among studies. In aggregate, there were 22 distinct outcomes. Among the overlap outcomes displayed in Table 2 are: hospital length of stay, [20,29] hospital readmission rates, [20,29] direct cost of care, [20,21,31] A1c, [22,29,30,33,35,36] blood pressure, [19,23,30,33,36,37] and number of office visits. [28,31,32] While none of the studies reviewed detected any negative impacts of interprofessional collaborative care or interprofessional collaborative practice, not all findings favored a positive impact of interprofessional care or interprofessional practice.
While study analyses included quantitative, qualitative and mixed methods, descriptive statistical analysis was the most frequently used analytical technique. In all the reviewed studies, at least two professions were engaged with the most frequently engaged professions being physicians (medicine) included in 18 of the 20 studies and pharmacy in 15 of the 20 reviewed studies. Nursing was included in 6 of the reviewed studies. Other engaged professions (e.g., social work, occupational therapy, physical therapy) were represented in one or at most two studies.
Finally, an examination of the reviewed studies' outcomes and findings revealed eight distinct outcome categories. Fig 2 depicts the analysis process and Tables 3 and 4 describe the research outcomes and key study characteristics. The eight categories identified were: 1) Impact of interprofessional collaborative practice on chronic diseases with well-defined management measures [19, 22-24, 28-30, 33-35]; 2) Impact of interprofessional collaborative practice on specialty care outcomes with well-defined management measures [20,26,38]; 3) Impact of interprofessional collaborative practice on direct cost of care [20,29,31]; 4) Impact of interprofessional collaborative practice on prescribing practices and/or patient adherence [25,32]; 5) Impact of interprofessional collaborative practice on dental care [27]; 6) interprofessional collaborative practice impact on falls [21]; 7) impact of interprofessional collaborative practice on health services utilization [31,32]; 8) impact of interprofessional collaborative practice on patient satisfaction. [34]

Discussion
Interprofessional education is firmly entrenched as an essential component to prepare health professions for a complex and evolving healthcare environment. The Health Profession Accreditors Collaborative states that the need for health professions to work together is unprecedented. [39] However, the merit of interprofessional education is inextricably linked to the value of interprofessional collaboration, making evidence related to the value added proposition of collaborative care critically important. Surprisingly, this scoping review yielded only twenty studies that examined clinically relevant outcomes related to interprofessional collaborative care and interprofessional collaborative practice. Of an original yield of 375 studies published between 2010 and 2018, most studies focused on educational endeavors or provider impressions and not on outcomes documenting clinical impact. https://doi.org/10.1371/journal.pone.0218578.g002 Table 3. Analysis of studies findings from measured patient outcomes.
IPCP impact on patient outcomes (derived from an assessment of the study findings listed in Table 2)

Findings (aggregated from multiple studies)
1) Impact of interprofessional collaborative practice on chronic diseases with well-defined management measures [19, 22-24, 28-30, 33-36] -pharmacist intervention reduced blood pressure in racial minorities with socioeconomic disparities -intervention produced an average decrease in A1c of 2.3% -emergency department (ED) visits and/or hospitalizations decreased 30% during the intervention (p = 0.052) and then returned to pre-enrollment levels after the intervention was discontinued (p = 0.83) -BMI decreased from 33.0 to 32.3 kg/m2 -fewer diabetes-related PCP visits post intervention (5.09 vs 3.78 visits/year) -fewer non-insulin diabetes medications prescribed post intervention 2) Impact of interprofessional collaborative practice on specialty care outcomes with well-defined management measures [20,26,38] -97.5% and 91.7% of Hepatitis C patients achieved a sustained virological response (SVR) in the per-protocol analysis and the intention-to-treat analysis, respectively -length of stay (total hip arthroplasty [THA] reduced by 0.4 days and total knee arthroplasty [TKA] reduced by 0.6 days) reduced 3) Impact of interprofessional collaborative practice on direct cost of care [20,29,31] -direct costs (THA reduced by $1,020 per case and TKA reduced by $539 per case) significantly decreased -interprofessional treatment approach associated with estimated opportunity cost savings of $670,750.27.
4) Impact of interprofessional collaborative practice on prescribing practices and/or patient adherence [25,32] -increased communication between the primary care provider and the community pharmacy, coupled with targeted patient-specific interventions -significant reductions in nonadherence 5) Impact of interprofessional collaborative practice on dental care [27] -Collaborative care intervention with dental medicine improved patients health status including replaced dentures 6) interprofessional collaborative practice impact on falls [21] -new falls declined on average 31% across the facilities in the project 7) impact of interprofessional collaborative practice on health services utilization [31,32] -patients in intervention practices experienced a 7.4% increase in annual outpatient visits relative to baseline -intervention practices yielded statistically significant reductions in hospitalizations, emergency department visits, and ambulatory care-sensitive emergency department visits 8) impact of interprofessional collaborative practice on patient satisfaction [34] -no differences in patients' perceptions of shared decision-making, activation or satisfaction with care were found -also analysis found no difference in post-discharge patient satisfaction https://doi.org/10.1371/journal.pone.0218578.t003 In 2008 Berwick, et al. published their seminal triple aim article [5] which argued that healthcare reform should address patient satisfaction, population health improvement, and rising healthcare costs. Both Berwick's paper [31] and the US Affordable Care Act, [15] cite interprofessional collaborative care and interprofessional collaborative practice as important components of that reform. However, a 2014 interprofessional practice and education review [3] and 2015 IOM report [1] both noted that few studies examined the impact of interprofessional practice and education on triple aim outcomes-most notably patient health outcomes (or population health). Despite limited evidence, policy makers recommend practitioners develop the knowledge, attitudes, skills and behaviors to work collaboratively. [40] What is encouraging is that while the body of evidence is sparse, most studies examining interprofessional care found it positively impacts care.
The reviewed studies investigated outcomes across multiple settings, healthcare professions and study designs. The majority found interprofessional collaborative care and/or interprofessional collaborative practice improved health related outcomes such as A1c in patients with diabetes, polypharmacy for pain management in patients with lower back pain, and blood pressure outcomes for patients with hypertension. Healthcare cost savings were documented for chronically ill patients such as those with pancreatitis. The studies reviewed provide blueprints for others wishing to examine the impact of interprofessional collaborative practice or interprofessional collaborative care on appropriate health outcomes. Of note, while not all studies documented benefit, no study found a negative outcome related to interprofessional collaborative care and/or interprofessional collaborative practice, The majority of the studies included in this review researched outcomes in primary care settings [19, 23-25, 28, 30-33, 35-37] and addressed clearly defined chronic diseases with well-defined management measures. [19, 22-24, 28-30, 33-36] In these instances the challenge of defining, operationalizing, and measuring patient health-related outcomes is minimized.
Of the 20 studies, almost all included medicine and 11 included pharmacy. Surprisingly, no study reviewed included public health professionals, social workers, behavioral health providers or physician assistants and few studies engaged nurses. Only one study included dentists Table 4. Summary of the key study characteristics.

Characteristic
Group Frequency even though their accreditation standards advocate the importance of oral health to overall health and training that links oral health to colleagues in other professions. [41] Addressing these identified gaps in future research will strengthen the evidentiary base of interprofessional collaborative practice. Furthermore, clear articulations of the elements of interprofessional collaborative practice approaches, such as coordination, communication, cooperation, shared decision making and practice [42,43] need to be included in future research.

Limitations
A number of limitations to this review bear noting. First, the study used very specific definitions for our inclusion criteria. Because the field of interprofessional practice and education has yet to standardize the lexicon of its concepts, the definitions used may have been too restrictive and as a result missed some relevant research. For example, the Cochrane review identified studies with positive patient outcomes from integrated behavioral health care team but the terminology "Collaborative Care" used was differed from the IPCP definition used in this review. Therefore, this review paper was not identified in our initial literature search. While this might affect identification of public health and this work IPCP research studies, we do not believe this substantially changes our review. In addition, while the US healthcare system is unique, by focusing solely on the US healthcare environment, this review could have potentially missed studies from other developed countries that may have documented pertinent data-driven findings. Finally, the review only used the PubMed and Google Scholar search engines. There is always a possibility that an additional search engine might have yielded additional relevant articles.

Conclusions
The goal of interprofessional practice and education is to foster care collaboration that optimizes patient outcomes. Although advocates promote the benefits of removing silos among health professionals, there is surprisingly little evidence documenting the health-related outcome benefits of interprofessional collaborative practice and/or care. In addition, our review found inconsistency in use of terminology to describe health care team work and this might have resulted in difficulties identifying all relevant literature. [44] Continued effort to develop common and meaningful terminology and research on the measurable impact of interprofessional collaborative practice and/or care on patient health-related outcomes is needed to document its benefits and to explore the models, systems and nature of collaborations that best improve population health, increase patient satisfaction, and reduce cost of care.
Dedication: Dr. Nawal Lutfiyya unexpectedly passed away before completing the final revisions of this paper. Her co-authors would like to acknowledge her leadership in developing this review and to dedicate this paper in her memory.
Supporting information S1 File. Table 1