A systematic review critically appraising quantitative survey measures assessing power dynamics among multidisciplinary teams in acute care settings

ABSTRACT

By valuing the knowledge of each discipline holistic patient-centered care can be achieved as decisions arise from expertise rather than established hierarchies. While healthcare has historically operated as a hierarchical power structure (i.e., some voices have more influence), these dynamics are rarely discussed. This review addresses this issue by appraising extant quantitative measures that assess multidisciplinary team (MDT) power dynamics. By identifying psychometrically sound measures, change agents can uncover the collective thought processes informing power structures in practice and develop strategies to mitigate power disparities. Several databases were searched. English language articles were included if they reported on quantitative measures assessing power dynamics among MDTs in acute/hospital settings. Results were synthesized using a narrative approach. In total, 6,202 search records were obtained of which 62 met the eligibility criteria. The review reveals some promising measures to assess power dynamics (e.g., Interprofessional Collaboration Scale). However, the findings also confirm several gaps in the current evidence base: 1) need for further psychometric and pragmatic testing of measures; 2) inclusion of more representative MDT samples; 3) further evaluation of unmatured power dimensions. Addressing these gaps will support the development of future interventions aimed at mitigating power imbalances and ultimately improve collaborative working within MDTs.

KEYWORDS:

• Hierarchy
• hospitals
• multidisciplinary care team
• power
• quantitative
• survey

Introduction

Within healthcare, teams have become an integral feature of care provision emerging to manage the increasingly complex care needs and multimorbidity of a growing aging population, and to handle the rising prevalence of chronic disease (Hartgerink et al., 2014). Gawande (2011) outlines the evolution of health systems from care delivery by one “all-knowing” physician to current practice where patients are cared for by multidisciplinary teams (MDTs). MDTs include healthcare professionals (HCPs) from several disciplines, including physicians, nurses, health and social care professionals (i.e., physiotherapists, occupational therapists, dieticians, social workers, pharmacists), management, and support staff (i.e., healthcare assistants, multi-task assistants). By valuing the skills and knowledge of each discipline, holistic patient-centered care can be achieved (Engel et al., 2017). When MDTs appreciate the unique expertise of all professions and promote active participation from each member, care decisions are based on staff experience and knowledge rather than established role hierarchies. This represents a departure from compliance with doctor’s orders to care decisions that refocus on patient priorities (Fox & Reeves, 2015). By enabling voices across disciplines and seniority to contribute to care decisions a broader, more holistic perspective can be considered from a social (e.g., social workers) and environmental (e.g., occupational therapy) viewpoint in addition to the medical model.

However, interprofessional collaboration (IPC) is challenging. Each professional group has a unique identity that corresponds to their discipline-specific (typically siloed) training and clinical experience (Ferlie et al., 2005). These unique identities mean that despite sharing the same goal of improving patient outcomes, HCPs may have differing priorities, roles, and expectations about how care should be delivered (Braithwaite et al., 2016; Hall, 2005; Weller et al., 2014). These divergent interests can result in HCPs working in discipline-specific silos (nursing, medicine, allied health), where professions may leverage their discipline-specific knowledge to strengthen their voice within the MDT (Hall, 2005). However, traditional norms of organizations mean that some voices within MDTs are more valued and have more influence than others (Rogers, De Brún, Birken, et al., 2020).

Background

Healthcare has historically operated as a hierarchical power structure with physicians assuming dominant roles (Baker et al., 2011; Rogers, De Brún, Birken, et al., 2020), while other professions encounter challenges establishing their status in terms of patient care decisions (Hall, 2005). A recent review by Okpala (2020) identifies five domains that influence team power dynamics in healthcare: 1) team-related factors (unbalanced allocation of influence, respect for medical hierarchy); 2) role allocation (lack of recognition, lack of delineation of duties, lack of confidence in the skills and competencies of others); 3) communication (nature and tone of communication, receptivity and responsiveness); 4) trust and respect; and 5) individual-related traits (teamwork skills, team attitude). Engum and Jeffries (2012) suggest that imbalances of power between professions can influence communication, the coordination of care, and ultimately patient safety. Despite the reported impact of power dynamics, hierarchical structures in healthcare are rarely explicitly discussed (Gergerich et al., 2019) or researched (Baker et al., 2011; Paradis & Whitehead, 2015). The absence of this discourse suggests a hesitancy to acknowledge the realities of hierarchy in healthcare.

Without an understanding of MDT power dynamics, change agents are unable to address these fundamental patient safety issues (e.g., miscommunication) and develop appropriate strategies to mitigate power imbalances in practice. Many sub-cultures can exist within hospital settings that can create difficulties in conducting in-depth qualitative research across each unit/department. Therefore, employing a measure to assess MDT power dynamics will support change agents to uncover the collective thought processes informing local power structures, which can subsequently guide the development of context-specific strategies to address variations in power dynamics across teams.

The purpose of this review is to identify and appraise extant quantitative measures that assess MDT power dynamics or a dimension of power dynamics as outlined by Okpala (2020). By identifying and appraising these measures, change agents can choose psychometrically sound instruments to investigate MDT power dynamics. This enhanced understanding will help in the development and evaluation of future interventions aimed at exploring and mitigating power imbalances within healthcare teams. Enhanced consistency in the application of such measures will facilitate the conduct of high-quality research and promote the comparability of interventions aimed at improving care provision through improved interdisciplinary working and collaborative care approaches. To address this gap in the current evidence base, this research aims to answer the following research questions: 1) What quantitative survey measures exist to assess the dimensions of power within MDTs in acute care settings? and 2) How do the identified survey instruments compare in terms of their psychometric properties and how they were used?

Methods

This systematic review was conducted to explore the proposed research question. This study was informed by the Cochrane handbook’s (Higgins & Green, 2011) guidance for conducting systematic reviews and the Preferred Reporting Items of Systematic Reviews and Meta-Analyses (PRISMA) (Moher et al., 2009). The review protocol was published on the PROSPERO Database in November 2021 (CRD42021283355).

Search strategy

The five dimensions of power identified by Okpala (Okpala, 2020) informed the search strategy (Supplementary material 1). Using keywords in conjunction with truncation and Boolean operators, five electronic databases were searched: Medline, CINAHL, EMBASE PsychINFO, and ABI/Inform. Reference lists of included studies were also hand searched to identify potentially relevant studies that were not retrieved from the database searches. However, no additional relevant articles were retrieved.

Inclusion and exclusion criteria

The studies were restricted to peer-reviewed articles published in English before July 27, 2021. The eligibility criteria were broad to ensure a balance between a specific and sensitive search of the literature. Empirical studies were included if they reported on the development, validation, or empirical use of one or more quantitative measures assessing power dynamics or a dimension of power dynamics as outlined by Okpala (2020). Searches were limited to studies where healthcare staff working in acute/hospital settings used the quantitative survey measure(s) to self-assess power dynamics within their defined multidisciplinary team (i.e., intradisciplinary teams and studies using a random sample of hospital staff or students were excluded). Previous systematic, literature, narrative, and realist reviews were excluded in addition to descriptive studies, studies using only qualitative methods, and studies using external observers/examiners to assess power dynamics within MDTs. Studies that incorporated a random sample of hospital staff/students or used external observers to rate MDT power dynamics were excluded as contextual factors such as power relations are dynamic and require time to be appropriately understood (Rogerset al., 2020).

Study screening and data extraction

Covidence, an online data management system, was employed to manage the review process. Article screening and selection was performed independently by three reviewers (LR, SHS, PA) against the eligibility criteria. The reviewers met to discuss and resolve any conflicts or disagreements. To guide data extraction, the reviewers developed a standardized data extraction tool (Supplementary material 2). Adopting Powell’s et al. (2021) approach, articles were compiled into “measure packets,” which included (1) the measure itself; (2) the measure development article; and (3) all identified empirical uses of the measure retrieved from the database search. Researchers also extracted information relevant to nine psychometric rating criteria from the Psychometric and Pragmatic Evidence Rating Scale (Lewis et al., 2018; Powell et al., 2021): (1) internal consistency, (2) convergent validity, (3) discriminant validity, (4) known-groups validity, (5) predictive validity, (6) concurrent validity, (7) structural validity, (8) responsiveness, and (9) norms. When a full measure was relevant to assessing power dynamics, researchers reported the psychometric evidence for the full measure. However, if only subscales of a broader measure were relevant, researchers reported the psychometric evidence at the subscale level. Each criterion was rated using Lewis et al.’s (2018) scale: “poor” (−1), “none/absent” (0), “minimal/emerging” (1), “adequate” (2), “good” (3), or “excellent” (4). Ratings were summarized using a “rolled up median” approach to assign a single score for each criterion. If the measure had only one rating for a criterion, this value was the final rating. However, if a measure had multiple ratings for a criterion across several articles, the median score was calculated to generate the final rating. To obtain a conservative score, if the computed median resulted in a non-integer rating, the non-integer was rounded down. When a measure was used more than once, the range for each psychometric property was also provided.

Studies reporting less than two psychometric properties were excluded from phase two of the data extraction process (i.e., quality appraisal, and mapping survey power dimensions). This additional criterion ensured the most robust and reliable measures were appraised in further detail. This decision was determined following a review of the articles included in phase 1 of the data extraction process. While almost all articles reported one psychometric property (i.e., norms), only 21 reported three or more psychometric properties. Thus, to ensure a balance between an inclusive and meaningful appraisal, the authors applied the additional criterion of including studies that reported two or more psychometric properties for phase 2 of the data extraction process.

Quality appraisal and mapping of constructs

As appropriate for the study design, sections of the Mixed Methods Appraisal Tool (MMAT) were used to assess the quality of articles included in phase two of data extraction (Hong et al., 2018). LR appraised all articles included while SHS independently assessed 10% of included studies. Any disagreements over the quality or risk of bias of the included papers were resolved through discussion. To enhance the transparency in reporting the appraisal process, a summary of the quality assessment can be seen in Supplementary material 3. Subsequently, LR mapped all measures or their subscales to Okpala’s (2020) five dimensions of power (team-related factors; role allocation-related factors; communication; trust and respect; and individual-related traits).

Data synthesis

Due to the aim of this study and the heterogeneity of the articles included a narrative synthesis (Popay et al., 2006) and thematic analysis (Braun & Clarke, 2006) of the findings was the most appropriate approach to examine the review questions. Okpala’s (2020) five dimensions of power supported to structure the narrative synthesis. Results are reported in accordance with PRISMA guidelines (Moher et al., 2009).

Results

Overview

The search returned a total of 6,202 records. Of these, 1,502 were duplicates and were removed. In total, 4,138 articles were excluded following title and abstract screening and a further 465 were excluded during full-text review as they did not meet the inclusion criteria. Reasons for exclusion are evident in the PRISMA diagram (Figure 1). In total, 97 studies met the inclusion criteria and were reviewed. Sixty-two of these studies reported two or more psychometric properties and progressed to the second phase of data extraction (i.e., quality appraisal and mapping phase).

Figure 1. PRISMA diagram.

Phase 1 of data extraction (n = 97 studies)

Supplementary material 4 summaries all empirical uses of identified measures assessing MDT power dynamics (n = 86). Supplementary material 5 outlines study and measure characteristics for papers reporting the development (n = 6) or validation (n = 5) of surveys assessing power. Most studies were conducted in the USA (n = 52), Canada (n = 8), and the UK (n = 7). Studies completed in adult acute care settings were predominantly focused on surgical MDTs (n = 21), while pediatric and neonatal settings mostly conducted their studies among critical care services (n = 8) (Supplementary materials 4 and 5).

While two studies (Ginsburg & Bain, 2017; O’Donovan & McAuliffe, 2020) provided limited information about their study sample, the remaining ninety-five studies incorporated nursing staff in their sample (i.e., staff nurses, clinical nurse managers, clinical nurse specialists, and advanced nurse practitioners). Eighty-eight studies included physicians (junior and senior physicians), health and social care professionals participated in 42 studies (physiotherapists (n = 25), pharmacists (n = 16), social workers (n = 15) were the most frequently cited), while support staff (healthcare assistants most frequently cited (n = 15)) and administrative staff were incorporated in 17 and 13 studies, respectively. Many articles (n = 35) focused solely on power relations between nurses and physicians

Thirty-six of the included studies employed a cross-sectional approach and four conducted a mixed methods evaluation to explore staff perceptions of interdisciplinary teamworking (e.g., communication openness, MDT collaboration). The remaining studies (36 pre-post designs, 3 mixed methods papers, 6 cohort analytic studies, and 1 cluster randomized control trial) evaluated the impact of a new initiative or care model on team functioning (e.g., structured interdisciplinary rounds, interdisciplinary team training).

Phase 2 of data extraction (n = 62 studies)

Table 1 lists the 35 studies with limited psychometric information (i.e., less than two psychometric properties), which were excluded from phase two of data extraction. While Smits et al. (2003) reported limited psychometric properties for the Group Environment Scale used within their study, the authors provide adequate information for the additional bespoke measure employed, which resulted in the article's inclusion in the next phase of data extraction. Within the 62 remaining papers (shaded in gray in Supplementary materials 4 and 5), 43 measures were identified (Safety Attitudes Questionnaire (SAQ) full scale and relevant subscales referred to as one measure). Seventy-seven percent (n = 33) of these included measures were used once. The most frequently used surveys were the SAQ (as a full scale and relevant subscales) (n = 20), the Collaboration and Satisfaction about Care Decision-Making scale (n = 5), the Relational Coordination measure (n = 4), and the Hospital Survey on Patient Safety Culture questionnaire (n = 4). Table 2 presents the median and range psychometric properties available for the 43 identified measures. Table 2 reports the diverse psychometric values for the SAQ when researchers use the measure as a full scale or when subscales are adopted. Table 3 illustrates the dimensions of power dynamics associated with each included survey. While 81% of measures assessed communication processes within MDTs (n = 35), only 15 surveys evaluated role allocation. The Interprofessional Collaboration Scale (Kenaszchuk et al., 2010) and Team Development measure (Stock et al., 2013) were the only questionnaires that assessed all five power dimensions (Table 3). Figures 3–6 are categorized by Okpala’s (2020) five dimensions of power and present side-by-side comparisons of the psychometric information extracted for each measure.

Table 1. List of studies with limited psychometric properties.

Author	Measures used
Aaberg et al. 2021	TeamSTEPPS Teamwork Perceptions Questionnaire
	Collaboration and Satisfaction about Care Decisions in Teams
	Hospital Survey of Patient Safety Culture Questionnaire
Acal Jiménez et al. 2018	Safety Attitudes Questionnaire
Adams & Feudale 2018	Collaboration and Satisfaction about Care Decisions in Teams
Aeyels et al. 2019	Relational Coordination Scale
Austin et al. 2021	Interprofessional Collaborative Competency Attainment Scale*
Awad et al. 2005	Communication survey*
Bamford & Griffin 2008	Teamworking Questionnaire*
Bursiek et al. 2020	Professional Practice Environment Assessment Scale*
	The Mayo High Performance Teamwork Scale*
Catchpole et al. 2010	Safety Attitudes Questionnaire
De Lima Garcia et al. 2019	Hospital Survey of Patient Safety Culture Questionnaire
France et al. 2010	Safety Attitudes Questionnaire- ICU version
Gosling et al. 2003	Team Climate Inventory
Gupta et al. 2015	Safety Attitudes Questionnaire- subscales
Haller et al. 2008	Safety Attitudes Questionnaire- Labour and Delivery version
Halverson et al. 2009	Perception of Teamwork Survey
Hendricks et al. 2018	Safety Organising Scale
	Collaboration and Satisfaction about Care Decisions in Teams
Hill et al. 2015	Safety Attitudes Questionnaire- OR version
Hilt et al. 2020	Safety Attitudes Questionnaire
Kuy & Romero 2017	Safety Attitudes Questionnaire
Lau et al. 2020	Safety Attitudes Questionnaire- OR version
Lisbon et al. 2016	Hospital Survey of Patient Safety Culture Questionnaire
Masten et al. 2019	Self-developed measure*
Murphy et al. 2018	Hospital Survey of Patient Safety Culture Questionnaire
Notaro et al. 2019	Hospital Survey of Patient Safety Culture Questionnaire
O’Donovan & McAuliffe 2020	Psychological Safety measure*
O' Leary et al. 2015	Safety Attitudes Questionnaire- subscale
Opper et al. 2019	Collaborative Behaviour Scale*
Patterson et al. 2013	Safety Attitudes Questionnaire- subscales
Paull et al. 2013	Self-Efficacy of Teamwork Competencies Scale*
Piers et al. 2019	Interprofessional Practice and Education Quality Scales*
	Ethical Climate questionnaire*
Prenestini et al. 2013	Competing values framework questionnaire*
Prentice et al. 2016	Assessment of Interprofessional Team Collaboration Scale
Scotten et al. 2015	TeamSTEPPS Teamwork Attitude Questionnaire
	TeamSTEPPS Teamwork Perceptions Questionnaire
	Interprofessional Team Performance Scale*
Smits et al. 2003	Group Environment scale*
Strasser et al. 2005	Self-developed measure*
Zech et al. 2017	Hospital Survey of Patient Safety Culture Questionnaire

Note: * Measures not reported in detail in paper as limited psychometric properties available (<2).

Table 2. Summary of psychometric properties for included measures.

Measure name	Internal Consistency Median (Range)	Convergent Validity Median (Range)	Discriminant Validity Median (Range)	Known Groups Validity Median (Range)	Predictive Validity Median (Range)	Concurrent Validity Median (Range)	Structural Validity Median (Range)	Responsiveness Median (Range)	Norms Median (Range)	Total Score
Interprofessional Collaboration Scale	3	2	2	-	-	-	2	-	-	9
Assessment for Collaborative Environments	4	4	-	-	-	-	1	-	2	11
Communication and Sharing Information scale	3	-	-	1	-	-	2	-	4	10
Safety Attitudes Questionnaire (SAQ) (full scale) (n = 5)	3 (2,4)	-	-	1	-	-	3	-	4 (−1,4)	11
Hospital Survey on Patient Safety Culture (n = 4)	2	-	-	−1 (−1,1)	-	-	-	-	2 (1,3)	3
Relational Co-ordination (n = 4)	3 (1,4)	-	-	1	-	-	-	-	2 (−1,4)	6
SAQ subscales (including irrelevant subscales) (n = 2)	2 (2,3)	-	-	-	-	-	-	-	1 (1,2)	3
SAQ relevant subscales only (n = 13)	2 (−1,3)	-	-	1 (0,2)	-	-	-	-	2 (−1,4)	5
TeamSTEPPS Teamwork perception questionnaire (n = 2)	2 (1,3)	-	-	-	-	-	1	-	−1 (−1,2)	2
Collaboration and satisfaction about care decision-making (n = 5)	3	-	-	1 (−1,2)	-	-	-	-	−1 (−1, 2)	3
Cox conflict scale	3	-	-	−1	-	-	-	-	1	3
Healthcare Team Vitality Instrument	-	3	-	-	-	-	3	-	3	9
ICU Nurse- Physician Questionnaire (n = 3)	3	-	-	1 (1,2)	-	-	-	-	2 (1,4)	6
Professional Practice Environment Scale (n = 2)	3 (3,4)	-	-	−1	-	-	-	-	2 (1,4)	4
Teamwork Climate Scale	2	-	-	1	-	-	-	-	3	6
Self-developed communication and collaboration scale	3	-	-	-	-	-	-	-	2	5
Team Diagnostic Survey	2	-	-	2	-	-	-	-	2	6
Team psychological Safety (n = 2)	3	-	-	−1	-	-	-	-	1 (−1,3)	3
Work Collaborative Questionnaire	4	-	-	1	-	-	-	-	1	6
Job Diagnostic Survey	2	-	-	2	-	-	-	-	2	6
Reflexivity in Teams Scale	3	-	-	2	-	-	-	-	2	7
Satisfaction toward Coworkers	4	-	-	−1	-	-	-	-	1	4
Self-developed communication scale	3	-	-	−1	-	-	-	-	3	5
Team development measure	3	-	-	-	-	-	-	-	4	7
Self-developed- Teamwork competency questionnaire	2	-	-	-	-	-	-	-	2	4
Assessment of Interprofessional Team Collaboration Scale	−1	-	-	-	-	-	-	-	−1	−2
Collaborative Practice Scale	3	-	-	-	-	-	-	-	2	5
Ethical decision-making climate	-	-	-	2	-	-	-	-	4	6
Self-developed Interprofessional competency scale	4	-	-	-	-	-	-	-	−1	3
Team Assessment Questionnaire	4	-	-	-	-	-	-	-	3	7
Team Climate Inventory (n = 2)	4	-	-	1	-	-	-	-	3 (3,4)	8
Transactive Memory System Scale (n = 2)	2 (2,3)	-	-	-	-	-	-	-	−1 (−1,2)	1
Communication openness	4	-	-	-	-	-	-	-	2	6
Team Effectiveness Audit Tool	3	-	-	-	-	-	-	-	2	5
Job Content Questionnaire	2	-	-	-	-	-	-	-	2	4
TeamSTEPPS Teamwork Attitudes Questionnaire	2	-	-	-	-	-	-	-	−1	1
Operating Room Teamwork Assessment Scale	3	-	-	-	-	-	-	-	−1	2
Jefferson Scale of Attitudes Toward Physician Nurse Collaboration	2	-	-	-	-	-	-	-	2	4
Safety Organising Scale	3	-	-	-	-	-	-	-	2	5
Patient Safety Climate scale	3	-	-	-	-	-	-	-	4	7
Interprofessional socialization and valuing scale	-	-	-	−1	-	-	-	-	−1	−2
Intensity of Interprofessional Collaboration questionnaire	-	-	-	1	-	-	-	-	−1	0
TRUST questionnaire	-	-	-	-	-	-	-	-	2	2
Attitudes Towards Healthcare Team Scale	-	-	-	−1	-	-	-	-	−1	−2
Interpersonal Trust Scale	4	-	-	-	-	-	-	-	−1	3

Table 3. Summary of power dimensions assessed by included measures.

Scale	Team-related factors	Role Allocation	Communication	Trust/Respect	Individual-related factors	Total dimensions reported
Report 4 psychometric properties
Interprofessional Collaboration Scale	√	√	√	√	√	5
Assessment for Collaborative Environments	√	√	√	√		4
Communication and Sharing Information scale			√			1
Report 3 psychometric properties
Hospital Survey on Patient Safety Culture	√		√	√	√	4
Professional Practice Environment Scale	√		√	√	√	4
Relational Co-ordination		√	√	√	√	4
Safety Attitudes Questionnaire (full, subscale and versions)	√		√	√	√	4
TeamSTEPPS Teamwork perception questionnaire	√	√	√		√	4
Collaboration and satisfaction about care decision-making	√		√		√	3
Cox conflict scale			√	√	√	3
Healthcare Team Vitality Instrument	√		√	√		3
ICU Nurse- Physician Questionnaire	√		√		√	3
Teamwork Climate Scale	√		√	√		3
Self-developed communication and collaboration scale	√		√			2
Team Diagnostic Survey		√			√	2
Team psychological Safety			√	√		2
Work Collaborative Questionnaire		√	√			2
Job Diagnostic Survey	√					1
Reflexivity in Teams Scale			√			1
Satisfaction toward Coworkers					√	1
Self-developed communication scale			√			1
Report 2 psychometric properties
Team development measure	√	√	√	√	√	5
Self-developed- Teamwork competency questionnaire	√	√	√		√	4
Assessment of Interprofessional Team Collaboration Scale	√		√	√		3
Collaborative Practice Scale	√	√	√			3
Ethical decision-making climate	√		√	√		3
Self-developed Interprofessional competency scale	√	√	√			3
Team Assessment Questionnaire	√		√	√		3
Team Climate Inventory	√		√	√		3
Transactive Memory System Scale		√	√	√		3
Communication openness			√	√		2
Team Effectiveness Audit Tool		√	√			2
Job Content Questionnaire	√				√	2
TeamSTEPPS Teamwork Attitudes Questionnaire			√		√	2
Operating Room Teamwork Assessment Scale		√	√			2
Jefferson Scale of Attitudes Toward Physician Nurse Collaboration	√	√				2
Safety Organising Scale		√	√			2
Patient Safety Climate scale	√		√			2
Interprofessional socialization and valuing scale			√		√	2
Intensity of Interprofessional Collaboration questionnaire	√		√			2
TRUST questionnaire				√	√	2
Attitudes Towards Healthcare Team Scale	√					1
Interpersonal Trust Scale				√		1
Total studies	25 (58%)	15 (35%)	35 (81%)	19 (44%)	17 (40%)

Narrative synthesis

Given the heterogeneity of the included papers, studies and their relevant measures are described in the following narrative synthesis using Okpala’s (2020) dimensions of power. The Interprofessional Collaboration Scale assessed all five dimensions of power and provided the most comprehensive psychometric evaluation (recording four out of the nine criteria sought in the rating scale). Overall, measurement quality for the included scales was poor with only five measures (i.e., Assessment for Collaborative Environments, SAQ, Communication and Sharing Information scale, Interprofessional Collaboration Scale, and Healthcare Team Vitality Instrument) receiving an overall score of 9 or higher (out of a possible score of 36).

Team-related factors

Team-related factors refer to the unbalanced allocation of influence within MDTs. Twenty-five surveys identified in this review evaluated this power dimension and assessed whether team decision-making was collaborative (i.e., based on team members’ collective expertise) or embedded in traditional hierarchical assumptions (i.e., physician autonomy). Measures within this domain were primarily developed or used in the USA (76%) by nurses and physicians working in inpatient medical units (36%) (Bohmann et al., 2021; Deneckere et al., 2013; Drekonja et al., 2019; Hastings et al., 2016; Khoshab et al., 2019; San Martin-Rodriguez et al., 2008; Sim et al., 2020; Smits et al., 2003; Vazirani et al., 2005) or multiple team types (36%) (e.g., emergency department, critical care units) (Al-Araidah et al., 2018; Ballangrud et al., 2017; Deilkas & Hofoss, 2008; Ginsburg & Bain, 2017; Grymonpre et al., 2016; Lavoie-Tremblay et al., 2014; Ribeliene et al., 2019; Sexton et al., 2006; Zaheer et al., 2019). Evidence of norms was available for 23 measures, internal consistency for 21 measures, known-group validity for 9 measures, structural validity for 4 measures, convergent validity for 3 measures, and discriminant validity for 1 measure. No psychometric evidence was available for predictive validity, concurrent validity, or responsiveness. The median rating for internal consistency and convergent validity for the included measures assessing team-related factors was good (score = 3). An adequate median rating (score = 2) was identified for norms and discriminant validity (based on one measure, the Interprofessional Collaboration Scale (Kenaszchuk et al., 2010), while the rating for structural validity and known group validity was minimal (score = 1). The highest rated measures of team-related factors were the Assessment for Collaborative Environments (Tilden et al., 2016) and the SAQ (Sexton et al., 2006) which both scored 11 out of a possible score of 36. While the Assessment for Collaborative Environments was used by one study (Tilden et al., 2016) and reported excellent (score = 4) internal consistency and convergent validity, adequate norms (score = 2) and minimal structural validity (score = 1), the SAQ used by 20 papers (Abdi et al., 2015; Bognar et al., 2008; Bohmann et al., 2021; Carvalho et al., 2015; Chaboyer et al., 2013; Chu-Weininger et al., 2010; Davenport et al., 2007; Deilkas & Hofoss, 2008; Drekonja et al., 2019; Dwyer et al., 2018; Gillespie et al., 2018; Ginsburg & Bain, 2017; Müller et al., 2018; O’leary et al., 2010; Pannick et al., 2017; Profit et al., 2012, 2017; Schwendimann et al., 2013; Sexton et al., 2006; Zaheer et al., 2019) scored excellent norms (score = 4), good internal consistency and structural validity (score = 3), and minimal known group validity (score = 1).

Role allocation

Fifteen measures assessed role allocation that refers to staff understanding, recognition, and appreciation of team members’ roles. Sixty-seven percent of measures assessing role allocation were primarily developed and used in the USA and employed by nurses and doctors. Developmental and validation studies focused on broad team types (e.g., MDT working in an inpatient unit) (Kenaszchuk et al., 2010; Stock et al., 2013; Tilden et al., 2016; Yildirim et al., 2006), while empirical studies assessing this dimension most frequently investigated surgical MDTs (Burtscher et al., 2020; Gittell et al., 2000; Paige et al., 2009; Villemure et al., 2019). Evidence of internal consistency was available for all 15 measures, norms for 14 measures, known-group validity for 3 measures, structural validity for 3 measures, convergent validity for 2 measures, and discriminant validity for 1 measure. No psychometric evidence was available for predictive validity, concurrent validity, or responsiveness. The median rating for internal consistency and convergent validity for the included measures assessing role allocation was good (score = 3). A median rating of adequate (score = 2) was identified for norms and discriminant validity (based on one measure, the Interprofessional Collaboration Scale (Kenaszchuk et al., 2010)) and for known-group validity and structural validity the median rating was minimal (score = 1). The highest rated measure of role allocation was the Assessment for Collaborative Environments (Tilden et al., 2016) as outlined above. The next highest scoring measure was the Interprofessional Collaboration Scale (Kenaszchuk et al., 2010). This study reported good internal consistency (score = 3) and adequate convergent validity, discriminant validity, and structural validity (score = 2).

Communication

Thirty-five measures assessed MDT communication. Topics incorporated in this domain include information exchange (i.e., frequency, accuracy, and timeliness) and the inclusivity of communication. Measures evaluating communication were primarily developed and used in the USA (69%) among surgical MDTs (Bognar et al., 2008; Burtscher et al., 2020; Carvalho et al., 2015; Davenport et al., 2007; Gillespie et al., 2018; Gittell et al., 2000; Klipfel et al., 2011; Maxson et al., 2011; Müller et al., 2018; Paige et al., 2009; Villemure et al., 2019), where the sample predominantly incorporated nurses and physicians. Evidence of norms was available for 34 measures, internal consistency for 31 measures, known-group validity for 17 measures, structural validity for 5 measures, convergent validity for 3 measures, and discriminant validity for 1 measure. No psychometric evidence was available for predictive validity, concurrent validity, or responsiveness. The median rating for internal consistency and convergent validity for the included measures assessing communication was good (score = 3). For structural validity, norms and discriminant validity (based on one measure (Kenaszchuk et al., 2010)), the median rating was adequate (score = 2) and for known group validity the rating was minimal (score = 1). The highest rated measures of communication were the Assessment for Collaborative Environments (Tilden et al., 2016) and SAQ (scoring 11) (Abdi et al., 2015; Bognar et al., 2008; Bohmann et al., 2021; Carvalho et al., 2015; Chaboyer et al., 2013; Chu-Weininger et al., 2010; Davenport et al., 2007; Deilkas & Hofoss, 2008; Drekonja et al., 2019; Dwyer et al., 2018; Gillespie et al., 2018; Ginsburg & Bain, 2017; Müller et al., 2018; O’leary et al., 2010; Pannick et al., 2017; Profit et al., 2012, 2017; Schwendimann et al., 2013; Sexton et al., 2006; Zaheer et al., 2019) as outlined above.

Trust and respect

Eighteen measures assessed trust and respect within MDTs. These measures evaluated team psychological safety (i.e., staff comfort speaking up and taking interpersonal risks), and staff perceptions of feeling valued within their MDTs. Seventy-eight percent of measures were developed or employed in the USA, among nurses and physicians in surgical settings (Bognar et al., 2008; Burtscher et al., 2020; Carvalho et al., 2015; Davenport et al., 2007; Gillespie et al., 2018; Gittell et al., 2000; Müller et al., 2018; Sifaki-Pistolla et al., 2020). Evidence of norms was available for 17 measures, internal consistency for 16 measures, known-group validity for 9 measures, structural validity and convergent validity for 3 measures, and discriminant validity for 1 measure. No psychometric evidence was available for predictive validity, concurrent validity, or responsiveness. The median rating for internal consistency and convergent validity for the included measures assessing communication was good (score = 3). For structural validity, norms, and discriminant validity (based on one measure (Kenaszchuk et al., 2010)) the median rating was adequate (score = 2). The median rating for known group validity was absent (score = 0). The highest rated measures of communication was the Assessment for Collaborative Environments (Tilden et al., 2016) and SAQ (both scoring 11) (Abdi et al., 2015; Bognar et al., 2008; Bohmann et al., 2021; Carvalho et al., 2015; Chaboyer et al., 2013; Chu-Weininger et al., 2010; Davenport et al., 2007; Deilkas & Hofoss, 2008; Drekonja et al., 2019; Dwyer et al., 2018; Gillespie et al., 2018; Ginsburg & Bain, 2017; Müller et al., 2018; O’leary et al., 2010; Pannick et al., 2017; Profit et al., 2012, 2017; Schwendimann et al., 2013; Sexton et al., 2006; Zaheer et al., 2019) as outlined above.

Individual-related factors

Individual-related factors refer to the skills (e.g., conflict management) and attitudes of individual MDT members (friendliness, extra-role behaviors). Seventeen measures identified in this review evaluated this power dimension. Most were primarily developed or used in the USA (76%) among nurses and physicians in surgical teams or across multiple team types (Abu-Rish Blakeney et al., 2019; Al-Araidah et al., 2018; Ballangrud et al., 2017; Deilkas & Hofoss, 2008; Ginsburg & Bain, 2017; Grymonpre et al., 2016; Lavoie-Tremblay et al., 2014; Ribeliene et al., 2019; Sexton et al., 2006; Zaheer et al., 2019). Evidence of norms was available for 16 measures, internal consistency for 15 measures, known-group validity for 11 measures, structural validity for 2 measures, and convergent and discriminant validity for 1 measure. No psychometric evidence was available for predictive validity, concurrent validity, or responsiveness. The median rating for internal consistency for the included measures assessing individual-related factors was good (score = 3). For convergent validity, norms, and discriminant validity (based on one measure (Kenaszchuk et al., 2010)), the median rating was adequate (score = 2). While the median rating for structural validity was minimal (score = 1), a score for known group validity was absent (score = 0). The highest rated measures of individual-related factors was the SAQ (scoring 11) (Abdi et al., 2015; Bognar et al., 2008; Bohmann et al., 2021; Carvalho et al., 2015; Chaboyer et al., 2013; Chu-Weininger et al., 2010; Davenport et al., 2007; Deilkas & Hofoss, 2008; Drekonja et al., 2019; Dwyer et al., 2018; Gillespie et al., 2018; Ginsburg & Bain, 2017; Müller et al., 2018; O’leary et al., 2010; Pannick et al., 2017; Profit et al., 2012, 2017; Schwendimann et al., 2013; Sexton et al., 2006; Zaheer et al., 2019), followed by the Interprofessional Collaboration Scale (Kenaszchuk et al., 2010) (scoring 9) (details outlined above).

Discussion

The objective of this systematic review was to identify and appraise the existing quantitative measures assessing MDT power dynamics or a dimension of power dynamics as outlined by Okpala’s (2020) (i.e., team-related factors, role allocation, communication, trust and respect, and individual related factors). The extant literature emphasizes the importance of IPC in the provision of optimum patient care (Engel et al., 2017). However, to enhance shared responsibility and decision-making among MDTs, the influence of power should be named and addressed (Cohen Konrad et al., 2019). This recommendation links to an important finding of this review, which is that no study used their measure to explicitly investigate MDT power dynamics. Instead, these studies aimed to more broadly understand teamworking or evaluate the effectiveness of an intervention on team functioning (e.g., communication processes). Future research in the field of interprofessional care and teamworking needs to openly name and assess MDT power dynamics as understanding these team characteristics, support the development and integration of context-specific interventions to address power disparities in healthcare teams.

Wider macro-level determinants (e.g., governance of care) assist to reinforce these historical power structures in MDTs (i.e., medical dominance) (Baker et al., 2011). However, Rogers et al. (2020) emphasize that team-level contextual factors (e.g., team culture) have been almost entirely overlooked in the extant literature implementing change. Therefore, gaining a better understanding of MDT power dynamics and designing context-specific approaches to address these disparities ought to be a priority to support the delivery of collaborative, safe patient-centered care. However, this transparency may result in recruitment challenges and researchers must recognize the increased potential for self-selection bias toward HCPs with extreme experiences of MDT working (i.e., hierarchical vs. collaborative).

While the findings identified some promising measures to assess MDT power dynamics, many studies insufficiently reported on the psychometric properties of their chosen measures. The Interprofessional Collaboration Scale (Kenaszchuk et al., 2010) assessed all five power dimensions and provided the most comprehensive psychometric evaluation, recording four out of the nine criteria sought in the rating scale (Lewis et al., 2018). However, like most of the measures included (77% of measures used once), only one study used this scale. Most measures provided evidence of norms and internal consistency (95% and 86%, respectively). Although one measure received a poor rating for internal consistency (Assessment of Interprofessional Team Collaboration scale (Grymonpre et al., 2016)), over 25% of measures reporting norms scored poorly on the rating scale (see Figures 2–6). This finding likely reflects the limited empirical use of these measures that impacts scale generalizability. Additionally, evidence of other psychometric properties was sparse (i.e., known group validity (n = 20), structural validity (n = 4), convergent validity (n = 3), discriminant validity (n = 1)), insufficiently described or unavailable to extract (i.e., concurrent validity, predictive validity, responsiveness) (see Figures 2–6).

Figure 2. Side-by-side comparison: team related factors.

Figure 3. Side-by-side comparison: role allocation.

Figure 4. Side-by-side comparison: communication.

Figure 5. Side-by-side comparison: trust and respect.

Figure 6. Side-by-side comparison: individual-related factors.

Overall, measurement quality for the included scales was poor. With the exception of internal consistency and convergent validity (based on three studies), most median ratings ranged from absent (score = 0) to adequate (score = 2). Only five measures (i.e., Assessment for Collaborative Environments, SAQ, Communication and Sharing Information scale, Interprofessional Collaboration Scale, and Healthcare Team Vitality Instrument) received an overall score of 9 or higher (out of a possible score of 36). Therefore, future work needs to prioritize further psychometric testing of promising measures identified in this review. The pragmatic properties of these measures also require consideration, as the usability of these scales will likely impact their use in practice.

Additionally, the findings of this review emphasize a continued focus on the nurse–physician dyad in interprofessional care. While this result may relate to the acute care focus of this review, the extant literature suggests that these findings may also reflect the status of medicine, nurses making up the majority of the healthcare workforce, and the historical power relations that exist between these professions (i.e., the dominant–subordinate relationship) (Price et al., 2014; Reeves et al., 2008). Despite improving the range and quality of services offered to patients, power dynamics across other professional groups (e.g., health and social care professionals and support staff) appears to remain relatively unexplored. Boyce (2006) and Kessler et al. (2010) confirm that many of these disciplines feel marginalized and perceive their influence as invisible or undervalued by their MDTs. However, with health systems prioritizing the better integration of care across acute and community services, these undervalued professions will assume a critical role in this health reform. Therefore, future research exploring MDT power dynamics should expand their sample to accurately depict what represents an authentic/true team in modern healthcare.

Despite the cited importance of role clarity for successful interprofessional working (Ambrose-Miller & Ashcroft, 2016; Reeves et al., 2010) (i.e., prevent interdisciplinary tension by defining role expectations and avoiding infringement of professional boundaries), role allocation was the least frequently assessed power dimension by the included measures (n = 15). Instead, most (n = 35) investigated interpersonal communication, a factor central to much of the IPC literature (Wei et al., 2022). To gain a more holistic understanding of MDT power dynamics, future studies need to proportionately investigate all power dimensions. Similarly, the importance of context in interprofessional collaborative practice needs to be more consistently assessed across settings. While four power dimensions were primarily assessed among surgical staff, team-related factors were the only domain predominantly evaluated among medical MDTs. Literature suggests that the acuity and complexity of care mediates the level of IPC required, which may explain why many studies chose surgical settings as requiring greater investigation (DiazGranados et al., 2018; Nugus et al., 2010; Retchin, 2008). Additionally, the limited focus on team-related factors in surgical settings may reflect the assumption that physician authority is necessary due to the task-oriented nature of the specialty (Rogers et al., 2013; Teunissen et al., 2020). Emphasizing the importance of technical skills over interpersonal behaviors may explain why authority and influence are factors less frequently assessed among surgical MDTs. However, this assumption requires further investigation in the future research.

Limitations

Despite the final search yielding thousands of articles, the endeavor to strike a balance between a sensitive and specific search strategy increases the possibility that relevant articles may have been omitted from this review. The inclusion of purely empirical studies heightens the risk of publication bias as the gray literature was not appraised. However, we hope to have limited the impact of these challenges and present a comprehensive synthesis of the best available evidence by scanning reference lists of included articles to retrieve additional relevant studies.

Conclusion

This review set out to systematically identify and critically appraise extant quantitative measures that assess MDT power dynamics. The review reveals some promising measures to assess the five dimensions of power outlined by Okpala’s (2020). However, the findings also confirm several gaps in the current evidence base such as the need for 1) further psychometric and pragmatic testing of measures assessing power dynamics; 2) including more representative samples that depict true MDT membership; and 3) further evaluating unmatured dimensions of power (e.g., role allocation) in diverse healthcare contexts. Addressing these gaps will support the development and evaluation of future interventions aimed at mitigating power imbalances and ultimately enhancing care provision by improving collaborative working within MDTs.

Availability of data and material

Data used in this study is available through the journal articles cited herein.

Acknowledgements

The authors would like to thank Dr Emma Nicholson and Dr Robert Fox for providing their expertise during the planning phase of this systematic review.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/13561820.2023.2168632

Additional information

Funding

This research is part of the Collective Leadership and Safety Cultures (Co-Lead) research program which is being funded by the Health Research Board (RL-2015–1588). Funding sources were not involved in any stage of the review process in relation to data collection, data analysis, and the interpretation of the data.

Notes on contributors

Lisa Rogers

Dr. Lisa Rogers is an Assistant Professor at UCD Centre for Research, Education, and Innovation in Health Systems (UCD IRIS) in the School of Nursing, Midwifery, and Health Systems in University College Dublin, Ireland. She is a registered nurse, whose research interests include healthcare team dynamics and implementation science. Dr. Rogers holds a BSc. in General Nursing, MRes in Clinical Research and a PhD in Nursing, Midwifery, and Health Systems.

Shannon Hughes Spence

Shannon Hughes Spence was a research assistant at UCD IRIS in the School of Nursing, Midwifery, and Health Systems while writing this paper. Currently, she is a PhD student at the South East Technological University (SETU), Ireland. Her PhD focuses on young women’s experiences in the night time economy in Ireland, covering themes such as power, resistance, risk and safety. Hughes Spence holds a BA (Hons) in Community and Youth Development and a MSc in Sociology.

Praveenkumar Aivalli

Praveenkumar Aivalli is a PhD student at the UCD IRIS at the School of Nursing, Midwifery and Health Systems in University College Dublin, Ireland. His research interests include health policy and systems research, realist evaluation, implementation research, health system strengthening and quantitative epidemiological methods. He is an Ayurvedic Physician by profession and holds a post-graduate qualification in Public Health.

Aoife De Brún

Dr. Aoife De Brún is Assistant Professor at the UCD IRIS in the School of Nursing, Midwifery and Health Systems in University College Dublin, Ireland. She is a registered Chartered Psychologist with the British Psychological Society with experience of multi-disciplinary projects in health research. Her research interests include a range of topics in applied health and organisational psychology including team dynamics, collective leadership, and quality and safety in healthcare. Dr De Brún holds a BA(Hons) in Psychology and a PhD in Social Sciences.

Eilish McAuliffe

Prof Eilish McAuliffe is Professor of Health Systems at UCD and the Director of the UCD IRIS in the School of Nursing, Midwifery, and Health Systems. Her research activity is primarily focused on strengthening health systems. Utilising interdisciplinary approaches to identify problems in existing service provision, particularly in the areas of leadership, teamwork and organizational culture, she co-designs and evaluates new models and approaches to improve the quality and safety of healthcare. Prof McAuliffe holds a BSc. in psychology, an M.Sc. in Clinical Psychology, an MBA and a PhD in Health Strategy.