Abstract
Background
Healthcare workers face an elevated risk of chronic stress and burnout, for which resiliency interventions are needed.
Methods
The Stress Management and Resiliency Training Program (SMART-3RP; 8 weekly 90-minute sessions) was offered to 254 hospital employees between 2/2021 and 1/2024. Participants were surveyed pre- and post-intervention for measures of resiliency, stress coping, positive affect, perceived stress, anxiety, and depression.
Results
The baseline sample was 84% female, 89.5% non-Hispanic, 71.3% White, and averaged 45 years of age (SD = 14.25). Baseline and follow-up survey completion rates were 71.3% (N = 181) and 35.8% (N = 91), respectively. Significant improvements were seen pre-post intervention for all measures (all ps <.001): resiliency (d = 0.57), stress coping (d = 1.1), positive affect (d = 0.83), perceived stress (d = −0.88), anxiety (d = −0.74), and depression (d = −0.43).
Conclusions
Alleviating employee stress is crucial for improving individual, clinical, and systems-level outcomes in hospitals. The SMART-3RP is a promising program that provides healthcare workers with resiliency and stress coping skills.
Introduction
Healthcare workers (HCWs) face an elevated risk of chronic stress due to the demands of their profession, 1 which, when left unmanaged, can lead to worsened mental and physical health. 2 As of 2022, HCWs in the U.S. reported alarming levels of burnout (46%), anxiety (57%), and depression (34%), 3 symptoms of which may include difficulty sleeping, 4 fatigue,5,6 irritability, 7 and lack of motivation that may bleed into other facets of life besides work. 8 Such chronic stress has been found to contribute to poorer perceived health, 9 increased risk for chronic medical conditions (such as cardiovascular disease 10 and substance use disorders 11 ), and even mortality. 12 Paradoxically, those dedicated to caring for others are themselves at high risk for compromised health due to workplace stress. Not only does this issue directly affect the lives of HCWs, but given their critical role, it can indirectly impact clinical care quality and hospital system efficiency: exhausted HCWs exhibit impaired teamwork and communication (a predictor of patient safety),13-15 reduced caring behavior toward patients, 16 decreased productivity and efficiency,17,18 and higher rates of turnover. 19 As a result, the material costs of burnout to the healthcare system are staggering—an estimated $4.6 billion is lost to burnout each year. 20 Thus, addressing stress among HCWs is valuable from individual, clinical, and systems-level perspectives.
As the COVID-19 pandemic brought a surge of added stress to hospitals through excessive workloads and unexpected changes in safety and policy, many studies reported heightened levels of emotional exhaustion, compassion fatigue, and depersonalization in HCWs. 21 Some found that repeated exposure to pandemic-related deaths, uncertainty, and overwork related to the pandemic appeared to be drivers of the spike in HCWs’ anxiety and stress, 22 as well as psychiatric symptoms and diagnoses (e.g., PTSD, anxiety, and depression),23,24 which some have reported as persisting up to 3 years after the start of the pandemic. 25 These findings underscore the urgent need for interventions supporting this population.
The Stress Management and Resiliency Training Program (SMART-3RP) is designed to help individuals manage stress effectively while fostering resiliency—the capacity to adapt to environmental changes and stressors. Developed by the Benson-Henry Institute for Mind Body Medicine, the SMART-3RP aims to mitigate the negative effects of stress by combining positive psychology, cognitive behavioral therapy, and mind-body techniques that elicit the relaxation response state. 26 The program is well-established and has been adapted for various populations, including cancer survivors, medical residents, older adults, and parents of children with learning and attentional disabilities.
More recently, the program has shown efficacy in improving stress and resiliency outcomes in frontline clinicians working in the Mass General Brigham (MGB) Hospital system during the COVID-19 pandemic. 27 Although findings showed efficacy of the program for improving stress and resiliency outcomes, effects were assessed within a limited timeframe (3/23/20 to 6/02/20) and only in frontline clinicians interacting with COVID-19 patients, leaving its potential benefits unexamined for the broader healthcare workforce. The present analysis aimed to explore changes in stress and resiliency outcomes among a broader sample of MGB employees who participated in the SMART-3RP clinical program during and following the COVID-19 pandemic.
Methods
Participants included any MGB employee who engaged in the SMART-3RP between February 2021 and January 2024. An electronic survey was emailed to participants before and after the 8-week virtual intervention. Surveys included validated self-report measures detailed below. All study methods were approved by the MGB Institutional Review Board (Protocol #2011P001081, approved 9/23/11).
Intervention
The SMART-3RP intervention comprised 8 weekly 90-minute sessions conducted via videoconferencing (i.e., Zoom) in groups of 10-20 individuals. Guided by principles of positive psychology, cognitive behavioral therapy, and mind-body science, sessions included topics of sleep, yoga, healthful eating, adaptive perspectives, humor, and eliciting the relaxation response through mind-body techniques like diaphragmatic breathing and mindfulness (including mindful awareness and mindful eating). Each session was led by a trained mental health clinician or nurse and included experiential mind-body practices, group activities, and discussion. Participants were encouraged to practice mind-body exercises between sessions (approximately 10-15 minutes/daily practice). More information on specific session content can be found elsewhere. 26 It was offered free of charge to encourage MGB employee attendance in the program.
Measures
The Current Experiences Scale (CES) measured resiliency, or one’s ability to cope with daily stressors, across 6 domains: appreciation for life, adaptive perspectives, personal strength, spiritual connectedness, relating to others, and health behaviors. Subscales were calculated as composite scores of domain-specific items, and total scores were calculated as the composite of all 23 items. Total scores ranged from 0 to 115, with higher scores denoting greater levels of resiliency. The CES has previously shown good internal consistency, reliability, and construct validity. 26
The Measure of Current Status-A (MOCS-A) measured participants’ stress coping skills using a composite score of 13 items that assessed participants’ perceived ability to employ relaxation skills, adaptive perspectives, stress awareness, and other coping skills. Scored ranged from 0 to 52, with higher scores indicating a greater ability to utilize stress coping skills.
The Positive and Negative Affect Schedule (PANAS-PA) measured participants’ positive affect or positive emotions they experienced on average. Participants rated the degree to which ten positive emotions (e.g., “Interested,” “Strong,” “Inspired”) aligned with how they typically felt. Scores ranged from 10 to 50, with higher scores indicating greater levels of positive affect.
The Perceived Stress Scale (PSS-10) measured participants’ perceived stress levels using ten items about thoughts and feelings experienced in stressful situations over the past month. A composite score was created after 3 items were reverse-scored. Scores ranged from 0 to 40, with higher scores indicating greater perceived stress levels.
The Patient Health Questionnaire-4 (PHQ-4) measured symptoms of anxiety and depression using 4 items inquiring how often the participants experienced symptoms over the past 2 weeks. Two items pertained to anxious feelings (“Feeling nervous, anxious, or on edge” and “Not being able to stop or control worrying”) and 2 items pertained to symptoms of depression (“Feeling down, depressed, or hopeless” and “Little interest or pleasure in doing things”). Scores ranged from 0 to 6 for both anxiety and depression.
Data Collection
Data were collected and managed using Research Electronic Data Capture (REDCap) tools hosted at MGB. All data were stored in a secure REDCap data repository managed by the Benson-Henry Institute for Mind Body Medicine at Massachusetts General Hospital. All participants were informed that their surveys would be contributed to the data repository on the first page of the baseline survey; participants who consented proceeded to the next page of the survey, and any who did not consent exited the survey.
Statistical Analyses
Data was analyzed using the Statistical Package for Social Sciences (SPSS). Descriptive statistics and paired samples t-tests were calculated to explore participant characteristics and changes in outcomes pre-post intervention. Effect sizes were estimated according to Cohen’s d (0.2 = small, 0.5 = medium, 0.8 = large). List-wise deletion was used to handle missing data (i.e., t-tests were only run on participants with complete baseline and follow-up survey data). To investigate possible data biases, independent samples t-tests and chi-square tests were calculated to assess potential differences in the distribution of continuous and categorical variables (respectively) between completers and non-completers of the follow-up survey.
Results
Sociodemographic Data of MGB Employees Who Completed the Baseline Survey.
Means, Standard Deviations, and Paired Sample t-tests of Outcome Variables.
Abbreviations: CES, Current Experiences Scale; NP, new perspectives; PS, personal strength; SC, spiritual connectedness; RO, relating to others; HB, health behaviors; AL, appreciation for life; MOCS-A, Measure of Current Status-A; PANAS-PA, Positive and Negative Affect Schedule, positive items; PSS, Perceived Stress Scale; PHQ-4, Patient Health Questionnaire-4.
aCompleted both pre- and post-intervention surveys; 1 = higher scores are better functioning; 2 = lower scores are better functioning.
From the analysis of differences in characteristics of survey completers versus non-completers, it was found that follow-up survey non-completers had statistically higher scores of depression compared to survey completers (P = .025; 95% CI [0.083-1.21]). No other statistically significant differences were found between follow-up survey completers and non-completers regarding demographic characteristics or baseline scores of outcome variables.
Discussion
As quality interventions are critically needed to support HCWs in the wake of the COVID-19 pandemic, the present analysis sought to evaluate changes in stress and resiliency outcomes of HCWs before and following participation in a virtual, 8-session stress management intervention: the SMART-3RP. Improvements in all stress and resiliency outcomes were observed pre-post intervention. Large effect sizes were observed in stress coping (d = 1.1) and perceived stress level scores (d = −0.88), suggesting that the intervention may have had a particular impact on improving participants’ ability to cope with stressors effectively. A large effect size was also observed for changes in positive affect (d = 0.83). Notably, effect sizes for resiliency and stress coping were similar to previous findings observed in frontline clinicians who participated in the same program during the pandemic. 27 These findings suggest that resiliency interventions such as the SMART-3RP warrant further research in randomized trials to confirm the promising benefits for HCW stress and emotional well-being. Future work should explore mechanisms of potential moderation and mediation in stress and resiliency outcomes among HCWs (i.e., how differing baseline levels of positive affect, depression, and anxiety may moderate changes in resiliency outcomes or how learning coping skills may mediate decreases in perceived stress levels).
It is noteworthy that all outcomes showed significance, given that data were collected over 3 years from many different groups and across multiple interventionists delivering the SMART-3RP. However, the sample was largely homogenous regarding demographic characteristics (84% female, 89.5% non-Hispanic, 71.3% White); although time points and interventionists varied, the sample’s demographics did not. The lack of racial, ethnic, and gender diversity in the sample is an evident limitation to generalizability.
Selection bias likely also impacted the generalizability of our findings, as follow-up survey non-completers had statistically higher levels of depression compared to the completers (P = .025; 95% CI [0.083-1.21]). These results should be acknowledged when considering the positive results found, as the follow-up data for all outcomes were taken from a sample of individuals who were experiencing fewer symptoms of depression at baseline and who may have been more likely to engage with and draw benefits from the program. These findings also suggest that specific outreach, treatment, and retention efforts may be needed for HCWs with elevated depression symptoms; future studies may want to make additional efforts to identify, engage, and retain these individuals, as they are a population in need of greater support. 28 No other recorded baseline variables were statistically different between completers and non-completers, as both groups were similar in age, race, ethnicity, sex, marital status, education level, and baseline scores of anxiety, stress levels, stress coping, positive affect, and resiliency.
As a limitation to survey response accuracy, we acknowledge the social desirability bias that comes with the nature of the items asked, particularly because the surveys were provided through the participants’ workplace. For instance, some individuals may have felt pressured to report greater resiliency or more positive affect (e.g., reporting greater feelings of being excited, strong, enthusiastic, interested, etc.), or they may not have felt comfortable reporting symptoms of mental illness (e.g., reporting if they recently felt depressed or anxious). It may be important in the future to reiterate to participants that their survey responses are not accessible by their employers, encouraging safety for candidly reporting outcomes.
Beyond the aforementioned psychological barriers to program participation, there may exist notable structural barriers, such as time constraints, low digital literacy, and limited access to technology with reliable internet. Although individuals working night shifts or long hours could have particularly benefitted from the intervention, 29 they may have been unable to attend sessions since the program was offered after a typical 9-to-5 workday. Additionally, those who have difficulty operating the Zoom platform may have been deterred from participating. Instructors of the program provided an introductory pamphlet for using Zoom prior to the first session, however it may be important to offer more hands-on support in navigating the digital platform for those who indicate greater need. Finally, although the digital nature of the intervention makes it so that transportation is eliminated, it requires participants to have access to a computer or smartphone with reliable internet access, which may have limited some participation.
As we look to further understand and build resiliency in HCWs, researchers may benefit from utilizing qualitative and/or mixed methods work to study how individuals conceptualize and respond to stress, and relate to this construct of “resiliency” that has gained popularity in recent years, both colloquially and within research. For instance, it may be valuable to explore how one’s “objective” scores of resiliency may map onto one’s subjective reports of stress coping and resiliency. Further, qualitative work may uncover other aspects of intervention impact that were not directly measured. For instance, individuals may report in open-ended responses that the program improved their communication in their relationships or self-compassion, constructs that may be important for understanding mechanisms of resiliency.
Alleviating employee stress is crucial for systems dedicated to providing quality healthcare. A hospital’s highly stressful work environment places significant demands on employees, often resulting in burnout,3,30 decreased productivity and efficiency, 17 impaired teamwork and communication,13-15 and compromised patient care. 16 Therefore, interventions that address stress among HCWs are highly beneficial from individual, clinical, and systems-level perspectives. As a result of the present analysis, the SMART-3RP intervention offers a promising approach to equipping employees with resiliency and stress coping skills to manage their work and life demands. Moreover, the COVID-19 pandemic demonstrated the need for systems to be prepared and equipped to adapt during critical periods of heightened urgency. Providing stress management and resiliency skills to HCWs now may lay a strong foundation for individuals to be more prepared for unprecedented times in the future.
Lastly, although individual-level interventions are necessary to address burnout, more is needed. These solutions place the responsibility of resiliency on individual workers rather than addressing key system-level root causes of chronic stress in the healthcare workplace (e.g., limited autonomy, expectations of perfectionism, excessive workloads, staffing shortages, administrative burdens such as inefficient electronic health records, and misalignment of organizational and physician values driven by performance metrics). 31 Future work must address underlying structural issues that erode employee health and well-being. The pressure put on individual HCWs to always be resilient is unrealistic, and through shifting cultural mindsets of what it means to be resilient (e.g., by giving physicians permission to be vulnerable and self-compassionate rather than perfect and unwaveringly persistent), we can foster positive professional culture. These changes will take time, and while they are unfolding, HCWs must have access to evidence-based resiliency skills that can support them in improving their health and quality of life.
Footnotes
Acknowledgements
The authors would like to extend their sincere gratitude to the healthcare workers at Mass General Brigham for their dedication to caring for so many patients and families.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Mass General Brigham Behavioral and Mental Health initiative.
