The impact of positive reinforcement on teamwork climate, resiliency, and burnout during the COVID-19 pandemic: The TEAM-ICU (Transforming Employee Attitudes via Messaging strengthens Interconnection, Communication, and Unity) pilot study

Abstract

Burnout is an internationally recognized occupational phenomenon that negatively impacts the healthcare workforce and its recipients. The aim of this pilot study was to test whether positive reinforcement and integrating a language of support among co-workers can enhance resiliency, facilitate psychological wellness, and encourage hope. This embedded mixed methods prospective, behavioral, interventional study evaluated the effects of positive feedback on wellness among intensive care unit clinicians during the COVID-19 pandemic in a single center, quaternary care medical center. The deliberate positive feedback paradigm has the potential to augment resiliency and improve attitudes toward a teamwork climate. The routine use of deliberate positivity may represent a scalable, low-cost initiative to enhance wellness in a healthcare organization.

Introduction

Burnout among physicians, nurses, and other health professionals continues to be an international crisis, affecting up to 50% of the front-line provider taskforce (Rotenstein et al., 2018; Zhang et al., 2018). A recent survey in the USA indicated that burnout is a serious problem that continues to intensify, with 70% of respondents (N = 733) believing that healthcare provider burnout will worsen in the next 2–3 years (Bees, 2021). Another survey of physicians 1 year after the beginning of the pandemic revealed that 61% of physicians often experience feelings of burnout, a significant increase from previous reports (The Physician’s Foundation, 2021). Though burnout is more prevalent in healthcare workers compared to the general population, its incidence is even more important in those caring for critically ill and dying patients. Moss et al. (2016) on behalf of the Critical Care Societies Collaborative (CCSC) proposed a call for action to combat burnout syndrome. Strategies to tackle burnout in critical care workers have included debriefing, simulation, peer support, and group sessions with variable success and response (El Khamali et al., 2018; Gazelle et al., 2015; Kashani et al., 2015). Without organizational strategies to combat burnout during crisis, healthcare workers are likely to suffer, and their personal resiliency too may waver. The current COVID-19 global pandemic has once again placed this vulnerable worker population directly in the eye of the storm, making it vital to immediately and optimally support their wellness in order to mitigate potentially devastating mental health consequences.

In the face of acute crisis, it is not uncommon for individuals to experience negative emotions such as stress, fear, frustration, anger, hopelessness, and helplessness. Studies of healthcare workers during the 2003 SARS pandemic provided insight into the psychological ramifications of a world-wide health crisis that were not only immediate but long-lasting with sequelae of post-traumatic stress disorder (PTSD) and depressive symptoms (Mak et al., 2009; Maunder et al., 2003). The psychological landscape amidst healthcare providers within the current COVID-19 global pandemic reveals clinically elevated psychological symptoms in alarming numbers: stress (86%; Mrklas et al., 2020); depression (20%–50%; Gautam et al., 2020; Mrklas et al., 2020; Pappa et al., 2020); anxiety (45%–47%; Gautam et al., 2020; Mrklas et al., 2020); post-traumatic stress (32%–44%; Li et al., 2021; Zandifar et al., 2020); sleep disturbance (25%–34%; Pappa et al., 2020); and substance use (25%; Cash et al., 2021). Combine reflexive human behaviors during a crisis with underlying burnout syndrome in the healthcare worker and the previously described mental health morbidity may be magnified.

In 2016, a US clinical trial addressing burnout entitled “Web-based Implementation for the Science of Enhancing Resilience Study (WISER)” was underway with interventions aimed at unique behavioral modifications, particularly sharing positive attitudes with colleagues (Profit et al., 2021). Gratitude and peer recognition have been shown to enhance resiliency and encourage hope in a variety of structured groups (Geng, 2018; McCanlies et al., 2018). Further, facilitating deliberate acts of kindness can also improve happiness (Rowland and Curry, 2019) and employee engagement and patient experiences (Landry et al., 2018). These straightforward interventions in turn can lead to happier individuals, cohesive teams, and unified culture and can have immediate effects with relatively low cost. Moreover, psychological constructs (i.e., acts of kindness; gratitude) and peer support play an important role in adaptive behavior (Chancellor et al., 2018; Geng, 2018; McCanlies et al., 2018).

In this study, our objective was to pilot test a messaging interface that delivered positive feedback and messages of gratitude to and from health professionals during a global pandemic and assess its effects on teamwork climate, resiliency, and burnout. We hypothesized that the participants engaging in deliberate acts of positivity and kindness would have improved ratings of wellness. We also explored qualitative feedback from participants.

Methods

Design

We conducted a nonrandomized prospective, behavioral, interventional embedded mixed methods study (Creswell and Plano, 2007) at Cedars-Sinai Medical Center (CSMC) over a 6-week period from June 2020 to July 2020. Adults greater than 18 years of age were eligible to volunteer for the study if they were front-line healthcare providers (physicians, nurses, and respiratory therapists) working in the Medical Intensive Care Unit (MICU). A convenience sample was used. Healthcare providers were excluded if they did not primarily identify with working in the MICU. Informed consent was obtained from all volunteers prior to study initiation. Participants were screened and recruited via general departmental announcements made by the principal investigator and co-investigators of the study (Supplemental Figure 1). There was no direct one-on-one participant recruitment, as stipulated by the rules and regulations of the site’s Institutional Review Board (IRB). Additionally, per CSMC Research Institute policy, registration on ClinicalTrials.gov was required as the research study prospectively assigned human participants or groups of humans to one or more health-related interventions to evaluate the effects on health outcomes (ClinicalTrials.gov Identifier NCT04441632; https://clinicaltrials.gov/ct2/show/NCT04441632).

Because this was a nonrandomized study, the principal investigator was aware of the identity of the participant volunteers. This investigator worked in a different department, was an unbiased third party to the participants involved, and had no connections to the MICU work environment. All participants were assigned a unique identification number (#ID-1 through #ID-24) to assess participant data and outcomes during the study period. One week prior to the intervention, participants were asked to complete a questionnaire (Supplemental Item 1) providing their ID number and pre-intervention baseline demographics, teamwork climate attitude, resiliency ratings, and burnout ratings.

Measures

The Safety Attitude Questionnaire (SAQ; Sexton et al., 2006): The SAQ is designed to assess six domains of relevance to patient safety, including the teamwork climate (TC) domain that we selected for use in this study. The SAQ domains and items were originally adapted from the aviation industry and tailored to the intensive care unit setting through incorporation of conceptual models from Vincent’s framework for analyzing risk and safety (Vincent et al., 1998) and Donabedian’s model for assessing quality (Donabedian, 1988). To streamline data collection in relation to the objectives, we used only the TC domain of the SAQ, which includes these six items: 1. Input is well received in this clinical area; 2. In this clinical area, it is difficult to speak up if I perceive a problem with patient care; 3. Disagreements in this clinical area are resolved appropriately; 4. I have the support I need from other personnel to care for patients; 5. It is easy for personnel here to ask questions when there is something that they do not understand; 6. The physicians and nurses here work together as a well-coordinated team. Each item uses a five-point Likert scale (1 = Disagree Strongly, 2 = Disagree Slightly, 3 = Neutral, 4 = Agree Slightly, 5 = Agree Strongly). Negatively worded items are reverse scored so that all items can be counted with 5 indicating the best teamwork climate for each item. The domain is usually scored using the mean of all six items; alternatively, the score can be converted using the percent of responses that scored positively (i.e., scores of 4 and 5 grouped together; Sexton et al., 2006). A score of 75% or higher therefore indicates a sense of positive teamwork climate.

The Brief Resilience Scale (BRS; Smith et al., 2008): The BRS is designed to gain insights into the extent an individual perceives themselves as resilient – having the ability to bounce back or recover from stress or adverse contexts. This ability may be particularly important for people who are already ill or are dealing with ongoing health-related stresses. The BRS is comprised of six items: 1. I tend to bounce back quickly after hard times. 2. I have a hard time making it through stressful events. 3. It does not take me long to recover from a stressful event. 4. It is hard for me to snap back when something bad happens. 5. I usually come through difficult times with little trouble. 6. I tend to take a long time to get over setbacks in my life. Participants rate the responses on a five-point Likert scale (1 = strongly disagree; 2 = disagree; 3 = neutral; 4 = agree; 5 = strongly agree). Negatively worded items are again reverse scored so that all items can be counted with 5 indicating the strongest resiliency. A higher score indicates higher sense of perceived resilience. Upon calculating the total scores, the BRS score interpretation ranges consist of: 1.00–2.99 = Low Resilience; 3.00–4.30 = Normal Resilience; 4.31–5.00 = High Resilience.

The one-item burnout scale inventory (Dolan et al., 2015): This single-item burnout measure instructs respondents to define burnout for themselves: “Overall, based on your definition of burnout, how would you rate your level of burnout?” Responses are scored on a five-category ordinal scale, where 1 = “I enjoy my work. I have no symptoms of burnout”; 2 = “Occasionally I am under stress, and I don’t always have as much energy as I once did, but I don’t feel burned out”; 3 = “I am definitely burning out and have one or more symptoms of burnout, such as physical and emotional exhaustion”; 4 = “The symptoms of burnout that I’m experiencing won’t go away. I think about frustration at work a lot”; and 5 = “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.” This item often is dichotomized as ⩽2 (no symptoms of burnout) vs ⩾3 (one or more symptoms).

Procedures

All questionnaires were administered via a surveyplanet.com account, which was password protected. Over the next 4 weeks, participants were asked to submit positive feedback (a deliberate positive feedback paradigm) about co-workers within the MICU via a messaging interface. Participants were asked to send positive feedback about three unique individuals a minimum of twice per week over the duration of the study period. Positive feedback comments were submitted via surveyplanet.com, were only accessible to the principal investigator, and were checked routinely several times a day. Comments were allowed to be submitted anonymously. At the end of the workday (17:00, Monday through Sunday), the principal investigator then emailed the positive feedback to recipients. To optimize engagement and follow-through, weekly email reminders by the principal investigator were distributed to all study participants; co-investigators further made general departmental announcements to anonymously remind those involved to continue participation. One week after the behavioral intervention phase was completed, participants were asked to complete questionnaires by providing their ID number and responses to the questions pertaining to teamwork climate attitude, resiliency, and burnout (outcome measures) along with free-text responses providing feedback for the study. Questionnaire responses were submitted via the password-protected surveyplanet.com interface.

Data analysis

An embedded mixed methods assessment of teamwork climate, resiliency, and burnout was used pre- and post-intervention. Quantitative outcome assessments were made with validated questionnaires as the reference standards as aforementioned (SAQ, BRS, and one-item burnout scale inventory). Baseline demographic data included work role, years of experience, gender, and work shifts. The quantitative data were analyzed by calculating descriptive statistics for the cohort and interventional group, which were stratified by the primary outcomes. Categorical variables are reported as count (percentage). Continuous variables are reported as means. Qualitative data took the form of written notes in the form of feedback submitted via surveyplanet.com by participants who completed the study; submitters could opt in or out of making their feedback anonymous. These qualitative comments were then analyzed for recurring patterns presented in the form of supporting quotes. Furthermore, the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) checklist was used for transparency in reporting and is attached in the supplementary materials (Supplemental Item 2, Des Jarlais et al., 2004). The CSMC IRB reviewed and approved this behavioral study (#STUDY00000786).

Results

Twenty-four healthcare workers volunteered to participate in the study over a 6-week period from June 2020 to July 2020, of whom the majority were nurses (79%), women (67%), possessing greater than 5 years in clinical practice (83%), and primarily working on day shifts (67%) (Table 1). Across all participants (N = 24), baseline mean TC, BRS, and burnout scores were 83%, 3.68, and 2.75, respectively. When stratifying by demographic variables, nurses, women, those in clinical practice less than 5 years, and those primarily working night shifts were found to have the highest baseline burnout and the lowest baseline scores for resiliency and teamwork attitude (Table 2).

Table 1.

Study participant baseline demographics. Data are presented as number (percentage).

Variables	N = 24 participants
Primary role
Nurse/RN	19 (79%)
Physician/MD	3 (13%)
Respiratory therapist/RT	2 (8%)
Gender
Male	8 (33%)
Female	16 (67%)
Years in practice
<5	4 (17%)
5–10	7 (29%)
>10	13 (54%)
Work shift distribution
Days	16 (67%)
Nights	7 (29%)
Days and nights	1 (4%)

Table 2.

Baseline levels of teamwork climate (TC)^a, resiliency (BRS)^b, and burnout (BO)^c stratified by demographic variables (N = 24).

Variables	TC (%/100)	BRS (scale 1–5)	BO (scale 1–5)
Primary role
Nurse/RN	80	3.53	2.89
Physician/MD	90	4.11	2.00
Respiratory therapist/RT	96	4.50	2.50
Gender
Male	87	4.00	2.25
Female	81	3.52	3.00
Years in practice
<5	71	3.29	3.50
5–10′	79	3.43	3.00
>10	88	3.94	2.38
Work shift distribution
Days	83	3.84	2.56
Nights	80	3.21	3.29
Days and nights	88	4.33	2.00

^aNormative cut-off for poor TC <75%.

^bNormative cut-off for low resiliency <3.

^cNormative cut-off for symptomatic burnout ⩾3.

Baseline demographics and TC, BRS, and burnout scores comparing completers (N = 11) and non-completers (N = 13) are described in Tables 3 and 4. Overall, non-completers had modestly higher wellness scores at baseline compared to completers. Since our analysis strategy was intended to assess outcomes use a pre- and post-intervention design, non-completers were removed from post-intervention analysis. Eleven (46%) out of the original 24 participants therefore completed the study. Eight were nurses, two were physicians, and one was a respiratory therapist.

Table 3.

Baseline demographics between completers (N = 11) and non-completers (N = 13). Data are presented as number (percentage).

Variables	Completers (N = 11)	Non-completers (N = 13)
Primary role
Nurse/RN	8 (73%)	11 (84%)
Physician/MD	2 (18%)	1 (8%)
Respiratory therapist/RT	1 (9%)	1 (8%)
Gender
Male	2 (18%)	6 (46%)
Female	9 (82%)	7 (54%)
Years in practice
<5	2 (18%)	2 (15%)
5–10	3 (27%)	4 (31%)
>10	6 (55%)	7 (54%)
Work shift distribution
Days	7 (64%)	9 (69%)
Nights	3 (27%)	4 (31%)
Days and nights	1 (9%)	0 (0%)

Table 4.

Baseline scores of teamwork climate, resiliency, and burnout between completers (N = 11) and non-completers (N = 13).

Variables	Completers (N = 11)	Non-completers (N = 13)
Teamwork climatea (%/100)	82	84
Resiliencyb (scale 1–5)	3.61	3.74
Burnoutc (scale 1–5)	2.82	2.69

^aNormative cut-off for poor TC <75%.

^bNormative cut-off for low resiliency <3.

^cNormative cut-off for symptomatic burnout ⩾3.

After study completion, participants were found to have pre- to post-intervention trends toward greater teamwork attitude (7/11 participants; 82% vs 88%) and higher resiliency (6/11 participants; 3.61 vs 3.74). While teamwork climate and resiliency metrics were within normal levels, a positive trend in both increased metrics is contextually meaningful. Levels of burnout did not change, though two of the participants showed decreased burnout after the intervention. When analyzing components of the teamwork climate scale, participants reported augmented support from others to care for patients, a perception that it was easier to ask questions to colleagues or superiors for communication clarification, and an overall sense of improved team dynamics among all providers. When analyzing components of the resiliency scale, participants acknowledge experiencing difficulty during stressful events, though there is indication that they tended to bounce back faster after hard times or stressful events, a clinically and contextually meaningful finding (Table 5).

Table 5.

Pre- and post-intervention group teamwork climate (TC, %/100), resiliency (BRS, scale 1–5), and burnout (scale 1–5) scores divided into their subcomponents, N = 11.

Wellness domains	Pre-intervention	Post-intervention
TCa, overall	82	88
My input is well received	85	85
It is difficult to speak up if I perceive a problem	87.5	85
Disagreements are appropriately resolved	80	80
I have support from others to care for patients	77.5	85
It is easy to ask questions when I do not understand	87.5	92.5
Healthcare providers work as a well-coordinated team	75	97.5
BRSb, overall	3.61	3.74
I tend to bounce back after hard times	3.7	4.2
I have a hard time making it through stressful events	3.7	4.2
It does not take me long to recover from events	3.5	3.8
It is hard me to snap back when something bad happens	3.3	3.6
I usually come through difficult times with little trouble	3.5	3.8
I tend to take a long time to get over set-backs in life	3.8	3.6
Burnoutc	2.82	2.82

^aNormative cut-off for poor TC <75%.

^bNormative cut-off for low resiliency <3.

^cNormative cut-off for symptomatic burnout ⩾3.

A total of 69 positive feedback responses were submitted between colleagues, all of which were sent by completers only; 29 in week 1, 16 in week 2, 11 in week 3, and 13 in week 4. The majority (42/69, 61%) of comments were directed to other nurses; the rest were distributed to physicians (12%), respiratory therapists (10%), and other members (care partners, pharmacists, social workers; 5%) of the healthcare team. Thirteen percent (9/69) were submitted anonymously. Figure 1 shows some of the most commonly generated positive words utilized in the feedback messages, including “thank,” “appreciate,” “amazing,” and “wonderful.” Three predominant themes borne out of the analysis of comments provided included: gratitude, inspiration, and [a sense of] belonging (contributing to and receiving from a community of co-workers). No adverse events of unintended effects of the intervention were noted.

Figure 1.

Word cloud generator for all comments submitted throughout the study.

Examples of positive feedback provided after the intervention included:

• “Although I honestly have not actively participated due to my work schedule and all, I strongly appreciate [it] and it really warmed my heart receiving positive words from my co-workers. It lifted my spirit.”

• “I really liked the feedback I got from one of the doctors. I just felt bad since I was fairly tired and only did a few responses.”

• “It’s a great study. Thank you for all you do.”

• “This is a great intervention during a difficult time. It is so great to express my appreciation to others. Thank you!”

• “I enjoyed giving feedback and receiving feedback from my co-workers.”

Discussion

This pilot study is the first in the published literature that utilizes a messaging interface to provide the crucial peer-to-peer support thought to be critical for moving the needle on healthcare provider wellness during the COVID-19 pandemic (Blake et al., 2020; Salas-Vallina et al., 2020). It demonstrates the feasibility and potential value in intervening with a strength-based, deliberate “positive” behavioral intervention, a psychological technique known to carry prosocial benefit in day-to-day interactions (Blake et al., 2020; Chancellor et al., 2018; Salas-Vallina et al., 2020). As hypothesized, the data for the participants who completed the study showed trends for greater teamwork attitude and higher resiliency. This finding is in-line with other research revealing that extremely stressful environments at work require team members to work together with strong “network ties” and “positive attitudes” (Salas-Vallina et al., 2020). Our results demonstrated that it is feasible to implement a low-cost behavioral intervention utilizing messages of positive feedback that may bolster a sense of unity, cohesion, and resiliency during times of peak stress and that participants enjoyed providing and receiving positive feedback.

The COVID-19 global pandemic has deeply impacted and uprooted the daily lives of front-line healthcare workers. Those providing care are not only facing the risks of this highly contagious virus in their personal lives, but also in the workplace. The review of the literature reveals consistent reports of heightened stress, anxiety, and depressive symptoms and coping deficits in healthcare providers as a result of COVID-19 (Shreffler et al., 2020). The pandemic has strengthened recent efforts around the world to better understand risk factors for burnout in healthcare workers. For example, in Italy greater resilience and ability to tolerate uncertainty have been found to mitigate burnout (Di Trani et al., 2021). In Serbia, individuals with augmented resilience were nearly three-times less likely to be burned out compared to their peers (Safiye et al., 2021). And in India, emergency room nurses with only mild levels of burnout were more resilient, underscoring interventions aimed at resiliency (Jose et al., 2020). An effective, straightforward, and simple support system to reduce such emotional distress and burnout is timely, urgent, and critical.

Shapiro and McDonald (2020) emphasized the importance of peer support as a vehicle that could potentially change organizational culture during this pandemic. Indeed, a heightened sense of team identification has already been shown to support healthcare provider wellness during these challenging times (Sangal et al., 2020). Both team coordination and resiliency are key in helping providers sustain positive mental health in stressful environments, which can have generalized health benefits and ultimately helps to provide better care for the patients they serve (Landry et al., 2018; Nelson-Coffey et al., 2017; Rowland and Curry, 2019; Sieg, 2020; Wardman, 2020). This study also found that burnout is not a resilience deficit disorder. In other words, despite risks and experiences to feelings of burnout, individuals reported perceived resilience. Notably, resilience is considered a good preventive measure against burnout, and the negative correlation between burnout and resilience among medical staff has been confirmed in research conducted worldwide (Safiye et al., 2021; West et al., 2020). In a recent study conducted in the USA, participants who did not show the presence of burnout had higher resilience, but 29% of doctors who had the highest resilience had burnout, which indicates that preventive measures should still be taken to preserve their mental health (West et al., 2020). Taking time to provide positive feedback is a feasible intervention that may be advantageous to team climate and resiliency. In line with other research findings pertaining to prosocial behaviors in the workplace (Chancellor et al., 2018), positive feedback, akin to deliberate acts of kindness, has the potential to be of benefit to both the “receiver” and the “giver” of feedback, an impact that may indeed have a ripple effect on psycho-socio-cultural elements in the work setting. Moreover, in the high-stress work environments where burnout is commonplace, prosocial expression of positive feedback may have a favorable impact on psychological health and wellness individually and in the broader sense of team culture. Investigating the contributing factors to burnout and addressing targeted interventions to reduce feelings of burnout and to promote psychological health and wellness remain indicated.

Our pilot study had several limitations. First, it was completed in a single-center hospital setting and had a small sample size. Although the intervention had a positive trend on measures of teamwork climate and resiliency, a larger group with more participants would be necessary to perform inferential statistical testing. The findings of this study provide important preliminary information for calculating required sample size for such studies. Second, due to the volunteer nature of the study, there was also a considerable drop-out rate (54%). Several volunteers who were unable to participate in the trial to completion commented on a desire to partake in the study, but reported a lack of time to do so. Perceived “limited time” to engage in a wellness intervention is in itself a variable contributing to the “symptoms” at that moment in time. It has been shown to be a commonly reported variable in having a direct and indirect impact on mental health and wellness (Pollock et al., 2020), furthering the need to facilitate a culture of brief, accessible behavioral health-related interventions for staff. Qualitative feedback about the positive impact of even knowing about the intervention, however, supports the notion that the known existence of the active intervention toward a wellness initiative may be in and of itself an intervention for some individuals. Modifying the intervention in future studies to augment engagement will be important to bolster study completion rates. Third, there was a self-selection bias in those who eventually completed the study and both questionnaires; these individuals seemed to be at higher risk with worse baseline wellness scores compared to those who did not complete the intervention. Fourth, though we were able to compare wellness scores pre-and post-intervention, matching the interventional arm with a control arm would be helpful in powering future studies utilizing this intervention. We also posit that burnout perception did not change pre- and post-intervention due to the inherent subjectivity of the one-item burnout scale. Finally, the duration of the study was 6 weeks (1 week pre-intervention, 4 week intervention, and 1 week post-intervention), which is relatively short. Though positivity had a short-term effect on wellness in the group of participants, the longitudinal impact was not measured; longitudinal follow-up in this cohort would provide clarity in better understanding the effects of positivity on long-term wellness. However, even short-term gains have implications for intermittent application of the feasible intervention so as to contribute to a culture of wellness within a healthcare setting.

In conclusion, we hypothesized that the participants engaging in deliberate acts of positivity and kindness would have improved ratings of wellness, and our results demonstrated a trend in and potential for improved teamwork climate and resiliency. A messaging interface to exchange mutual positivity can be a simple, low-cost intervention to provide an effective peer-to-peer support system. We hope to utilize and scale-up this intervention with modifications to enhance participation, diversity, and follow-through in other units and teams that desperately require wellness interventions during this pandemic.