Initial psychometric properties of the provider-co-management index-RN to scale registered nurse-physician co-management: Implications for burnout, job satisfaction, and intention to leave current position

ABSTRACT

Team-based care has become a cornerstone of care delivery to meet the demands of high-quality patient care. Yet, there is a lack of valid and reliable instruments to measure the effectiveness of co-management between clinician dyads, particularly physicians and registered nurses (RNs). The purpose of this study was to adapt an existing instrument, Provider Co-Management Index (PCMI), previously used among primary care providers into a new version to scale RN-physician co-management (called PCMI-RN). We also aimed to explore preliminary associations between RN-physician co-management and burnout, job satisfaction, and intention to leave current job. Face, cognitive, and content validity testing, using mixed methods approaches, were preceded by initial pilot testing (n = 122 physicians and nurses) in an acute care facility. The internal consistency reliability (α=.83) was high. One-quarter of participants reported burnout, 27% were dissatisfied with their job, and 20% reported intention to leave their job. There was a weak significant correlation between co-management and burnout (p = .010), and co-management and job satisfaction (p = .009), but not intention to leave current position. Construct validity testing is recommended. Future research using PCMI-RN may help to isolate factors that support or inhibit effective physician-nurse co-management.

KEYWORDS:

• Comanagement
• instrument development; burnout
• physicians
• registered nurses

Introduction

Aging populations, increased numbers of patients living with chronic conditions and provider supply deficits have presented challenges to meeting the demand for high-quality care (Nicholson et al., 2019). Team-based and shared care among different professions has become a cornerstone of care delivery across the health continuum. Researchers have increasingly demonstrated that team-based models, such as nurse-physician dyads, help to alleviate missed care opportunities, increase continuity of care, and ultimately optimize patient outcomes (Bass et al., 2021; Funderburk et al., 2021; Hustoft et al., 2018; Reiss-Brennan et al., 2016). However, emerging evidence suggests that lack of interdisciplinary or interprofessional cohesion and poor communication between clinical team members may yield poorer clinician outcomes, such as burnout, job dissatisfaction, and workforce turnover (Smith et al., 2018; Tetzlaff et al., 2018). In an effort to validly measure interdisciplinary and/or interprofessional dyad cohesiveness and subsequently implement policies and practice changes that strengthen dyad interactions, valid and reliable instrumentation is needed (Salas et al., 2018).

Background

The construct of co-management first emerged in the literature as the effectiveness of shared patient care by more than one physician, such as a specialist and a primary care physician (Bowman et al., 2013; Weber et al., 2010). Additional evidence about co-management has focused on physician dyads co-managing patients in hospitals (Van Grootven et al., 2017) such as medicine attendings and a surgeon (Fierbinţeanu-Braticevici et al., 2019) or a physician and a pharmacist (Adnan & Pramaningtyas, 2021). Over the past decade, however, co-management has increasingly been investigated across professions, such as varying types of primary care providers working within the same team (Jones et al., 2017; Strobel, 2019).

Despite previous literature that includes the investigation of the impact of co-management across professions (e.g., pharmacists and physicians) and disciplines (e.g., hospitalists and surgeons), the theoretical dimensions of this latent construct were lacking. Therefore, we previously performed a conceptual analysis and qualitative study to determine the theoretical underpinnings, antecedents, and potential consequences when co-management occurs in the context of primary care (Norful, de Jacq, et al., 2018, Norful, Ye, Van der-Biezen, et al., 2018). Co-management is now defined as the process of sharing care delivery responsibilities for the same patient by more than one clinician, often across professions (Norful et al., 2019). Conceptually, co-management consists of three dimensions: effective communication, mutual respect and trust, and a shared philosophy of care (Norful, de Jacq, et al., 2018). Effective communication encompasses the timeliness of information exchange, a shared healthcare language, and effective and accessible modes to communicate. The second dimension involves mutual respect and trust by the co-managing clinicians, inclusive of a shared knowledge of scope of practice, trust with each other’s decision-making, and recognition of the contributions each clinician exhibits within the team. The third dimension, shared philosophy of care, refers to a clinical alignment with mutual goals for patient care as well as agreement on rational for care plan. This third dimension also encompasses conflict resolution and a similar worth ethic. The theoretical consequences of these three dimensions result in a cohesiveness and coordination between clinicians when sharing care responsibilities to meet a patient’s need. Subsequently, patient, provider, and organizational outcomes may improve (Norful, Ye, Van der-Biezen, et al., 2018). Co-management differs from similar constructs (e.g., collaboration and teamwork) because the focus is more on the attributes of a dyadic relationship needed during the process and coordination of shared care. The term collaboration is broad and in present nursing literature many times reflects restrictive physician oversight of a nurse’s scope of practice. Teamwork often refers to the collaborative effort of a group of multiple clinicians to achieve a common goal during patient care. Although the dimensions of collaboration or teamwork may overlap with co-management, they sometimes fail to isolate the factors that influence interprofessional relations, interactions, and coordination between two people during shared care (Norful, de Jacq, et al., 2018).

A systematic review investigating co-management between physicians and nurse practitioners demonstrated evidence that effective co-management between these two types of primary care provider professions yield improved control of chronic conditions (Norful et al., 2019). Nurse Practitioners are registered nurses with a master’s or doctoral degree who have advanced clinical training and an expanded scope of practice. Scope of practice is regulated by local governing bodies and has increasingly included the ability to assess, diagnose, and prescribe a treatment plan similar to the role of generalists (Cooper et al., 2019). Evidence stemming from the systematic review demonstrated improved clinical control for hyperlipidemia and diabetes; increased provider adherence to recommended care guidelines; and improved patient and caregiver quality of life. Further, the review authors concluded that there were no valid and reliable instruments to measure co-management across professions (Norful et al., 2019). Subsequently, the Provider Co-Management Index (PCMI) was developed.

The original PCMI was initially developed and tested to validly measure interprofessional co-management of patients by more than one primary care provider (e.g., physicians, nurse practitioners, physician assistants; Norful, Ye, Shaffer, et al., 2018). PCMI is a 20-item instrument made up of three subscales that align with the conceptual dimensions and high internal consistency reliability (Cronbach’s alpha =.83). Previous researchers using the original PCMI have demonstrated that the more effective co-management is, the less primary position providers report burnout, job dissatisfaction, and intention to leave job (Norful, Ye, Van der-Biezen, et al., 2018, 2022a, 2022b; Strobel, 2019).

In the present landscape of peri-pandemic health-care system, RN and physician burnout, job dissatisfaction, and intention to leave one’s job has taken on a substantial level of urgency for health-care organizations and researchers (Murthy, 2022; Norful et al., 2021). One potential effort is to support RN-physician dyads; yet, there are no valid instruments to scale co-management between RNs and physicians in hospitals. Nurses deliver care to patients admitted to hospitals 24 hours per day, and effective co-management that yields cohesiveness, coordination, and communication, between nurses and physicians is essential to patient and organizational outcomes (Wang et al., 2018). Clinician well-being is a multi-dimensional trait, defined as the joy and satisfaction derived from one’s work and limitation of burnout and stress (Dyrbye et al., 2017). Most researchers have measured well-being within the workplace in attitudinal and behavioral contexts (e.g., job satisfaction, turnover intention) and psychological or emotional context (e.g., burnout). For example, researchers have demonstrated evidence that nurse-physician collaboration, (as opposed to measuring co-management) had an inverse association with turnover intention (Dyrbye et al., 2017). Other studies have measured similar constructs, such as nurse-physician collaboration, as a factor in job satisfaction. However, these studies were limited to a sample of only RNs (physicians were not included) thereby inhibiting evidence about co-management from the perspective of both professions (Dutra & Guirardello, 2021). As organizations seek to leverage co-management for improved operational efficiency and effectiveness outcomes, a valid measure is essential to empirically scale co-management in clinical and research settings. Further, the use of such instruments to measure co-management and determine outcomes associated with well-being is increasingly salient and may enable clinicians and organizational policymakers to improve co-management care delivery models aimed at optimizing patient and provider outcomes (Hustoft et al., 2018; Kang et al., 2020; Petit Dit Dariel & Cristofalo, 2018).

Thus, the purpose of this study was two-fold: (a) To adapt and validate an existing instrument, PCMI, among RNs and physicians practicing in an acute care setting into a new version called PCMI-RN; and (b) To determine preliminary associations between co-management and well-being outcomes among RN and physician dyads, including burnout, job dissatisfaction, and intention to leave one’s job.

Methods

Ethical approval was obtained from the Institutional Review Board in addition to the study site medical ethics committee.

Phase 1: face, cognitive, and content validity testing

The purpose of Phase 1 was to establish face, content, and cognitive validity of the PCMI-RN. Mixed methods were employed, with qualitative focus groups and a cross-sectional survey for psychometric testing. The current guideline for establishing face and cognitive validity include a minimum of six experts in the field to review each item and make suggestions for revision (Taherdoost, 2016). Therefore, we recruited two groups, made up of physicians (n = 6) and RNs (n = 6), with at least 5 years of clinical practice experience. All participants were recruited via e-mail, sent hospital-wide by nursing administration and a Department of Medicine list-serve. The e-mail explained the purpose of the study and contact information for the researchers. Interested participants were asked to contact the research team to organize a date, place, and time convenient for the group. The session took place remotely via Zoom® due to COVID-19 restrictions on in-person gatherings. All Zoom® session recordings and participant data were stored on a secure network drive, password-protected, and housed on a desktop computer in the primary researcher’s university office. Potential participants were eligible for inclusion if they (a) currently were registered nurses and/or physicians with clinical privileges (currently permitted by organizational policy to deliver direct patient care); (b) had greater than 5 years of experience; (c) were fluent in English language. Participants were not eligible to participate if they identified as (a) nursing or medical students; (b) medical or new graduate nurses; (c) physicians or nurses without hospital privilege; or (d) non-Englishspeaking clinicians.

First, we performed content validity testing using an electronic survey. Content validity establishes evidence of how much a tool is relevant and representative of the intended construct that is being assessed (Haynes et al., 1995). Each participant was asked to rate each of the items on a 4-point Likert-type scale of relevance (1=highly irrelevant to 4=highly relevant) and clarity (1=completely unclear, full revision needed to 4=clear, no revision warranted). Content validity scoring data for each item was exported to SPSS v.23 for analysis. A content validity index for both individual items (I-CVI) and each subscale (S-CVI) was computed. I-CVI and S-CVI greater than .8 (at least 80% agreement across respondents) were eligible for inclusion and further psychometric testing (Lynn, 1986).

Following content validity testing, we held focus group sessions to explore the construct of co-management between RNs and physicians, as well as establish face and cognitive validity. Focus groups were moderated by a PhD-prepared researcher with expertise in qualitative methods and instrument development. The group sessions were audio-recorded (not video recorded) to ensure descriptive validity and help with the interpretation of findings (Matthews et al., 2018). The principal investigators, study participants, and one study co-investigator were present during the focus groups. Electronic notes were taken by the co-investigator to help with the interpretation of recordings during data analysis. Open-ended questions were used to prompt discussion of RN-physician teams, the related dimensions of the latent variable, co-management, and clinician burnout (Table 1). For example, participants were asked to discuss a response to, “Describe your overall sense of cohesion (or lack of) between physicians and registered nurses in your day-to-day practice.” Probes such as “tell me more” or “can you give me an example” were used to allow the moderator the freedom to explore relevant issues and elicit rich responses (DeJonckheere & Vaughn, 2019). Next, the PCMI items and subscales underwent face and cognitive validity testing to determine that PCMI captures co-management and items are relevant to the content being measured (Haynes et al., 1995). The moderator read each item to the group and asked the participants to discuss their initial thoughts and interpretation of the item content. This approach helped to focus on participants’ interpretations of the survey items and how their own experiences may potentially inform their answers to questions.

Table 1. Interview guide (phase 1).

Part 1: Open-ended Qualitative Questions

(a) Tell me about your role, responsibilities, and environment where you work PROBE: What are you responsible for? What are some of your daily activities? (b) Explain who makes up your clinical team PROBE: What professions do you interact with on a daily basis? (c) Give me an example of a time when effective co-management (working with another profession to meet the demands for patient care) benefitted your role and/or your patients (d) Describe your overall sense of cohesion (or lack of) between physicians and registered nurses in your day-to-day practice (e) Explain how physicians and nurses within your team communicate (f) What is your perception about the influence of co-management on clinician wellbeing?

Part 2: Face & cognitive validity

(a) Please read each item aloud and explain your initial perception about what the item is intending to measure. (b) Describe how each item is relevant to your role. (c) Discuss the clarity of each item and whether you recommend any language-based revisions.

Audio recordings were reviewed by three members of the research team independently to achieve congruence and understand the context of the discussion. Once the data were coded, the researchers sorted the data by code and began discussion to identify categories and emergent themes. Audio recordings and electronic notes from the focus groups were iteratively discussed by the study team to determine the need for item revision (Miller et al., 2014). The original items of the PCMI were subsequently revised based on researcher consensus and informed by participant suggestions for removal and/or change of terms to better capture the interaction between RNs and physicians. At this stage, the newly revised instrument was named, “PCMI-RN” to indicate that this version was specific to RNs and physicians.

Phase 2: pilot testing

We used a cross-sectional survey design to (a) pilot test the PCMI-RN and (b) obtain preliminary data about the impact of co-management on burnout, intention to leave the job, and job satisfaction. We hypothesized that as effective co-management increased, burnout, intention to leave job, and job dissatisfaction would decrease. We distributed the survey via e-mail to conduct initial item analysis and preliminary assessment of internal consistency reliability of PCMI-RN. The full survey consisted of items measuring individual demographics (e.g., age, race), work experience characteristics (e.g., years of experience, years in the present position), and the PCMI-RN developed during Phase 1. PCMI-RN uses a 4-point Likert-type scale (1=strongly agree to 4=strongly disagree). Burnout was measured using a single validated item with five response options ranging from “I enjoy my work. I have no symptoms of burnout” to “I feel completely burned out and often wonder if I can go on. I am at the point where I may need some changes or may need to seek some sort of help soon” (Dolan et al., 2015). Responses were dichotomized to indicate “presence of burnout” and “no burnout.” The single item used to measure burnout has been shown to have 83.2% sensitivity and 87.4% specificity when compared to the Maslach Burnout Inventory (Mayzell & Normand, 2020). Intention to leave a job was measured with a single item, “Do you plan to leave your current position in the next year?” with a dichotomized response option (Yes or No). Job satisfaction was also measured using a single item, “On the whole, how satisfied are you with your present job?,” with a 4-point Likert-type response option very dissatisfied to very satisfied.

The suggested minimum sample for initial psychometric evaluation, is at least 30 participants per group type (Johanson & Brooks, 2010). Therefore, a convenient sample of RNs and physicians were recruited via hospital-wide list services provided by hospital leadership for voluntary participation. This study site was selected due to accessibility to the research team, similarity of demographics consistent with national workforce estimates and hospital-based statistics noting increased numbers of experienced clinicians (years) compared to other hospitals within the health-care system. The anticipated duration of survey completion was less than 10 minutes to reduce participant burden. The survey was not presented within subscales to allow for closer examination of each item’s psychometric properties and correlation to other items.

Data analysis

The survey data were exported into SPSS v26. Descriptive statistics were computed to describe participant characteristics, age, sex, years of education, and working experience. Next, we conducted an item analysis by computing descriptive statistics (e.g., means) and determining whether the items had limited range and high/low standard deviations. We evaluated the internal consistency reliability of the subscales by calculating the Cronbach’s alpha for each item to determine how well the items measure the latent construct. Next, we investigated the “alpha if item deleted” to estimate changes in reliability (Cronbach’s alpha) if a specific item was removed (Gliem & Gliem, 2003). Corrected item-total correlation (desired range between .30 and .70 was calculated for each item to indicate if each item was correlated to the whole scale and if the item was related to the other items in the scale (Fowler, 2018).

We calculated the percentages and frequencies of each response option for burnout, job satisfaction, and intention to leave items. We dichotomized the burnout item by combining the three response options that indicated some level of burnout and the two response options indicating no burnout. We dichotomized the job satisfaction item by combining very dissatisfied and a little dissatisfied to indicate job dissatisfaction and then combined moderately satisfied and very satisfied to indicate job satisfaction. Pearson’s correlations were used to determine the bivariate association between co-management and each workforce outcome while controlling for individual demographics and work characteristics.

Results

Phase 1: face, content, and construct validity testing

The final sample for face, content, and cognitive validity testing included 12 clinicians: 6 RNs and 6 physicians. The sample was predominantly White, female, and had greater than 10 years of experience (Table 2). Of the six RNs who participated, most had a bachelor’s degree or higher.

Table 2. Face, content, and cognitive validity panel demographics.

	(n=12)
Occupation
RN (n)	6
Physician (n)	6
Age (mean) Years	55
	Percentages (%)
Race
White	82
Gender
Female	60
Education
BSN or higher (RNs only)	89.0
Years of experience
>10 Years	66

RN: Registered Nurse; BSN: Bachelor of Science in Nursing.

Content validity was calculated using content validity indices for each item and each subscale. I-CVI ranged from .833 to 1.0. No items were below the .8 threshold, and therefore all items were kept following content validity testing (Table 3). Following qualitative focus groups (face and cognitive validity testing) that aimed to collect expert input into item adaptation, several revisions to the instrument were made. Two items were removed due to lack of clarity, relevance, or redundancy as perceived by our expert panel. The first item removed was, “have knowledge of each other’s education and training background.” The second item removed was “recognize each other’s contributions to patient care.” Two items were revised: The item, “discuss patient care plans” was changed to “discuss patient’s plan of care” and a second item, “have a mutually agreed upon protocol to resolve conflict” was changed to “have a mutually agreed upon policy to resolve disagreement.” Following face, content, and cognitive validity testing, 18 items made up the PCMI-RN and were left for pilot testing to establish initial psychometric properties.

Table 3. PCMI-RN content validity testing.

Subscale (Item)	I-CVI
Effective Communication
(a) discuss patient care plans	.9
(b) review each other’s patient care documentation	.833
(c) communicate patient needs in a timely manner	1.0
(d) communicate changes in patient health status	.9
(e) have patient documentation that is available to one another	.9
(f) share a mutual medical language necessary to communicate	1.0
(g) notify each other about information that was relayed to the patient/family	.833
Mutual Respect & Trust
(a) have knowledge of each other’s education and training background	.9
(b) respect each other’s decisions about patient care	1.0
(c) trust each other’s decisions about patient care	1.0
(d) recognize each other’s contributions to patient care	.9
(e) have knowledge of each other’s scope of practice	.9
(f) identify each other’s strengths and limitations to deliver patient care	.9
Shared Philosophy of Care
(a) share the same values for patient care	1.0
(b) practice with the same work ethic (e.g., time, effort)	.9
(c) agree on rationale for diagnostic tests (e.g., labs, CT scan)	1.0
(d) agree on rationale for treatments (e.g., medications)	1.0
(e) have a mutually agreed upon protocol to resolve conflict	.833
(f) 6. share the same goals for patient care	1.0
(g) 7. share the same philosophy about how care should be delivered	1.0

I-CVI: Item-Content Validity Index.

Phase 2: pilot testing and initial psychometric analysis

The final sample for pilot testing consisted of 122 respondents (physicians and RNs). Our survey response rate was 38.8%. The majority of the sample were White, female, and had greater than 10 years of experience (46.2%). Table 4 displays sample demographics and the service lines within the hospital where the participants practice. Initial psychometric properties are found in Table 5. The internal consistency reliability of this adapted instrument, PCMI-RN, was high (Cronbach’s alpha =.833). The item means ranged from 1.48 to 2.69. The targeted range for corrected item-total correlation was .3–.7. All items that fell outside this range were reevaluated for inclusion. Six items fell outside the range, yet the Cronbach’s alpha if deleted would only slightly increase. The group consensus was that all items would be kept for future factor analyses in a subsequent study and with a new, larger sample.

Table 4. Pilot testing sample demographics.

	n=122
Age (Years)
Mean (SD)	48.3 (12.6)
Years Experience
Mean (SD)	24. 3 (15.1)
	Percentage (%)
Race
White	61.1
Gender
Female	78.2
Education
Associate’s Degree/Certificate in Nursing	8.6
Bachelor’s Degree (BSN)	50.8
Master’s Degree (MSN)	17.2
Doctor of Nursing Practice (DNP)	1.6
Medical Degree (MD/DO)	17.2
Other (e.g., PhD; MD/PhD)	4.7
Service Line
Emergency Department	14.1
Surgery/OR/PACU/Ambulatory	24.2
Obstetrics/Postpartum/Nursery/NICU	9.4
Med/Surg	16.4
Cardiology (e.g., Interventional, Telemetry)	3.9
Oncology	1.6
Intensive Care	13.3
Outpatient (e.g., Primary Care)	4.7
Other (e.g., Palliative Care; Infectious Disease)	12.5

SD: Standard deviation; BSN: Bachelor’s of Science in Nursing; MSN: Master’s of Science in Nursing; DNP: Doctorate of Nursing Practice; MD: Medical Doctor; DO: Doctor of Osteopathic Medicine; PhD: Doctorate of Philosophy; OR: operating room; PACU: Post-anesthesia care unit; NICU: Neonatal Intensive Care Unit.

Table 5. Phase 2: initial psychometric testing.

Item	Mean	SD	Corrected item-total correlation	Cronbach’s alpha if item deleted
(a) discuss patients’ plan of care	1.66	.645	.652	.825
(b) review each other’s patient care documentation	2.47	.927	.158	.840
(c) communicate patient needs in a timely manner	2.29	1.004	.113	.843
(d) communicate changes in patient health status	1.64	.904	.667	.816
(e) have patient documentation that is available to one another	1.52	.664	.569	.824
(f) share a mutual medical language necessary to communicate	1.48	.581	.576	.825
(g) notify each other about information that was relayed to the patient/family	1.97	1.335	.610	.816
(h) respect each other’s decisions about patient care	1.62	.598	.725	.820
(i) trust each other’s decisions about patient care	1.59	.589	.574	.825
(j) have knowledge of each other’s scope of practice	1.89	1.296	.608	.816
(k) identify each other’s strengths and limitations to deliver patient care	2.69	1.893	.453	.836
(l) share the same values for patient care	1.78	1.122	.581	.818
(m) practice with the same work ethic (e.g., time, effort)	1.70	.673	.573	.824
(n) agree on rationale for diagnostic tests (e.g., labs, CT scan)	1.89	1.086	.582	.819
(o) agree on rationale for treatments (e.g., medications)	1.68	.554	.554	.826
(p) have a mutually agreed upon policy to resolve disagreement	2.45	.926	.200	.838
(q) share the same goals for patient care	2.29	1.022	.725	.820
(r) share the same philosophy about how care should be delivered	2.41	.942	.173	.839

A posthoc power analysis was performed and determined that at least a sample size of 98 participants would demonstrate statistical significance, thus our sample size was appropriate to explore correlations between variables (Quach et al., 2022). One quarter of participants in our sample reported burnout, more than a quarter were dissatisfied with their job, and 20% reported their intention to leave their job within one year. There was a weak significant correlation between co-management and burnout, and co-management, and job satisfaction, but not intention to leave one’s job (Table 6).

Table 6. The frequency and association between co-management and burnout outcomes.

		PCMI-RN Pearson Correlation
Outcome	Percentage (n=122)	$r$	$p$
Burnout	25%	−.241	.010*
Job Dissatisfaction	27%	−.001	.009*
Intention to Leave Current Job in next year	20%	−.182	.055

Discussion

This study developed an adapted version of the original PCMI now focusing on RN-physician co-management and called the PCMI-RN. We conducted its initial psychometric testing through expert assessment of face and content validity, and pilot tested the survey to explore initial evidence of reliability. Items were based on the established conceptual dimensions and empirical evidence about co-management (Norful, de Jacq, et al., 2018): Effective Communication; Mutual Respect and Trust; Shared Philosophy of Care. Face, content, and cognitive validity were supported by the subjective input of 12 expert clinicians. The final PCMI-RN, following pilot testing, consisted of 18 items, and had strong internal consistency reliability (Cronbach’s alpha =.83). PCMI-RN is now ready for further psychometric testing including exploratory and confirmatory factor analyses to assess its dimensionality. We will use the data in this present study to calculate an a priori sample size for factor analysis, as recommended, to ensure instrument precision (Kyriazos, 2018).

This study expands on the understanding of RN-physician co-management of patients. Nurses manage patient care within the context and scope of nursing practice, while their physician colleagues manage the diagnostic and treatment needs of the patient. These two professions collectively contribute their expertise to patient care and may be potentially viewed as a concept different from interprofessional collaboration. Collaboration may or may not be present during co-management but can be a consequence if co-management is effective. The ability to scale RN-physician co-management enables the measurement of a distinct and important aspect of interprofessional care delivery, particularly the cohesiveness between a nurse and a physician, and the need to optimize clinical team processes. As more and more organizations implement team-based care, the guarantee that clinicians will effectively and cohesively work together is unknown. This instrument illuminates the professional interaction between two clinicians at a granular level beyond the whole clinical team. Further, although the instrument is aligned with the theoretical dimensions of co-management (effective communication; mutual respect and trust; shared philosophy of care), organizations may use the instrument to identify gaps in organizational infrastructure that could be unknowingly inhibiting effective team processes. For example, if electronic health record systems are structured to silo nursing and physician documentation, communication to meet the coordination and demands for patient care may be hindered. In addition, if a physician is not well versed in understanding the scope of nursing practice, the efficiency of clinical care may be impacted or respect for a nurse’s skillset could be reduced, and vice versa. We recommend that the instrument be used at the organizational level to evaluate team processes as well as used at the individual dyad level to inform appropriate workforce allocation.

This study also addressed an identified research gap in early studies measuring burnout as a manifestation of workplace well-being, being that most studies to date have operationalized well-being as job satisfaction or turnover intent. The findings of this study indicate that co-management perceptions share an inverse relationship with burnout, providing evidence toward the importance of team-based care, mutual respect, and valuation of professional practice as an organizational determinant of the trait. Our study also measured burnout outcomes within a combined sample of both nurses and physicians, aiming to address yet another identified research gap. Our findings contribute to the understanding of how burnout can have a substantial impact on workforce turnover and other psychological manifestations in clinicians. Negative implications of clinician burnout prior to the current COVID-19 pandemic include increased risk of chronic physiological conditions (e.g., cardiovascular disease; obesity; Bianchi et al., 2015; Toker et al., 2012), psychiatric conditions (e.g., depression, post-traumatic stress disorder [PTSD], and suicidal thoughts and behavior; Bianchi et al., 2015; Giesinger et al., 2020; Yates, 2020), and adverse organizational outcomes (e.g., low workforce retention rates; poorer quality of care; Helfrich et al., 2017; Van Bogaert et al., 2017).

Our findings were consistent with evidence of burnout rates in clinicians prior to the COVID-19 pandemic, although our data collection was performed 6 months following the initial pandemic wave in New York. Evidence has emerged that burnout rates have more than doubled since that time (Fessell & Cherniss, 2020; Rodriguez et al., 2020), making continued studies of factors that mitigate or exacerbate burnout an important focus. Understanding the effects of environmental factors, such as interdisciplinary or interprofessional relations, can help organizations to further measure team dynamics and in turn optimize clinician relations in all states of the healthcare response to public health emergencies. As the COVID-19 pandemic enters the recovery and resilience-building stages, ongoing surveillance of clinician well-being, and in particular burnout and poor retention, will be necessary. The measurement of co-management with PCMI-RN may play an important measure to understand the multiple organizational factors affecting clinician health.

Implications of the findings and recommendations to the organization

This study offers evidence toward innovative staffing models in organizations. Measuring co-management between interdisciplinary and interprofessional dyads may allow clinicians, policymakers, and administrators to strategically place specific clinicians together to subsequently improve quality of care, alleviate clinician burden or burnout, and improve the clinical outcomes of patients. In addition, PCMI-RN can be used to alert organizational practice managers to suboptimal co-management models so that interventions can be implemented to improve the quality of patient care delivery, promote patient safety, and optimize clinician and organizational outcomes, such as workforce retention. Organizations may also consider measuring nurse-physician co-management using the PCMI-RN to help determine which dyads have the highest cohesion. Dyadic cohesion is a theoretical consequence of effective co-management (Norful, de Jacq, et al., 2018). This knowledge may be considered for RN staffing decisions, to test how co-management, and in turn cohesion, manifests over time and determine interventions to strengthen RN–physician relations (e.g., daily huddles, patient rounds, joint Grand Rounds). Future research may include a granular investigation of daily physician and nurse tasks or responsibilities to increase the evidence about which clinical team processes are most impacted by poor co-management. Further, future studies should aggregate samples by services line or clinical settings to establish new evidence about co-management that is specific to a setting, such as a comparison between acute and primary care.

Based on our findings, organizations may also target burnout-mitigation initiatives that are not a “one-size-fits-all” strategy, but rather focused on the individual teams and clinician dyads, inclusive of their workflow, relations, expertise, and resources. The co-management theory itself focuses on provider–provider dyads and promotes a shared clinical alignment with effective communication strategies and trust (Norful, de Jacq, et al., 2018). Attention to individual clinician dyads may be more effective at mitigating stress rather than focusing on the whole team and provide more precision to well-being interventions

Limitations

There are limitations to this study. The initial psychometric testing was achieved by a purposive convenience sample. Although the sample sizes were adequate for psychometric testing, the responses of other clinicians may differ. Also, this study did not establish the factorial structure of this adapted version of PCMI and was informed strictly by classical test theory. The findings of this study should be interpreted with the understanding that this new adaptation of the PCMI has yet to undergo construct or discriminant validity testing. The results of this present study explore potential associations between co-management and RN/physician wellbeing outcomes but additional studies and further psychometric testing, including factor analyses is recommended. This present study also did not compare differences in PCMI-RN responses by profession. In future studies, researchers should explore differences between physicians and RNs about the perceived importance and consequence of co-management and investigate whether associations between co-management, burnout, job satisfaction, and intention to leave one’s job vary by profession. This knowledge may help to understand profession-specific factors that prevent or inhibit burnout.

Conclusions

Care delivery models, including team-based approaches, are a fundamental and critically important modifiable structure in clinical settings. Despite substantial evidence about team-based processes and effectiveness, this novel instrument enables the measurement of co-management, between a nurse and physician, at the individual clinical dyad-level, a sub-tier of the overall clinical team. Future researchers may use the instrument to evaluate suboptimal dyad-related processes, such as communication infrastructure or clinical alignment involving patient care goals. At the dyad level, organizations may use the instrument to inform effective staff allocation by pairing groups of nurses and physicians together that demonstrate effective co-management, and subsequently improving team-based outcomes. Future research should include large-scale field testing to establish construct validity, and the exploration of RN-physician co-management across clinical settings.

Disclosure statement

The PCMI is an invention licensed by Columbia University Tech Ventures. More information to license its use can be found at http://innovation.columbia.edu/technologies/CU21271

Additional information

Funding

The author(s) reported that there is no funding associated with the work featured in this article.

Notes on contributors

Allison A. Norful

Allison A. Norful, PhD, RN, ANP-BC, FAAN, is a board-certified adult nurse practitioner and health services researcher with a focus on interdisciplinary care delivery models and the impact of work environment factors on physiologic stress that precipitates adverse psychological outcomes in nurses including depression, anxiety, and suicidality. She is jointly appointed as an Assistant Professor at Columbia University School of Nursing and as a nurse scientist in the New York Presbyterian Hospital enterprise. Dr. Norful has been recognized as a leading international researcher and expert consultant in latent construct measurement, care delivery model analysis, nurse work environments and psychometric testing. Email address: [email protected]

Katherine C. Brewer

Katie Brewer, PhD, RN, is an Assistant Professor at Towson University, where she teaches, conducts scholarly activities, and serves as a member of the university, nursing profession, and Baltimore-area community. Her areas of scholarship are organizational determinants of nurse and patient well-being, with a current focus on institutional betrayal and trust. Dr. Brewer’s expertise includes population and health systems research, population-level analysis and interventions, organizational and social determinants of health, and health policy. She has several publications on these topics. Email address: [email protected].

Margaret Adler

Margaret Adler, MSN, RN, NEA-BC, is the Program Coordinator for the Magnet Program at NewYork-Presbyterian Hudson Valley Hospital. Ms. Adler is responsible for facilitating the nursing research program at the hospital and ensuring the organization meets the standards for research set by the Magnet Program. She has acted as a principal investigator on a number of research studies. Ms. Adler’s recent research focus has been on the stress of registered nurses who served during the pandemic. She is a co-author of a peer-reviewed article on managing the principles of Magnet during a pandemic. Email address: [email protected]

Andrew Dierkes

Andrew M. Dierkes, PhD, RN, is an Assistant Professor at University of Pittsburgh School of Nursing. Dr. Dierkes’ research aims to leverage the nation’s largest healthcare workforce—nurses—to improve outcomes for patients and providers while lowering costs of care. Within this framework, he has examined a range of topics, including postoperative sepsis, Pay for Performance programs, Medicaid Expansion under the Affordable Care Act, and California’s nurse staffing mandate. He is also interested in innovative work to translate research findings into practice. Email address: [email protected]