Changes in nurses’ quality of work life prior to and during COVID: A repeated cross-sectional analysis Authors: Isaiah Hipel, Janice Keefe, Amber Duynisveld, Yuting Song, Yinfei Duan, Carole A Estabrooks, Peter G Norton, Priscilla Packiam, Janet Squires Published: 2026 ABSTRACT BackgroundThe COVID-19 pandemic had a profound impact on nurses in Long Term Care settings, who are responsible for resident care, clinical leadership, and staff training. These nurses face challenges such as supporting residents with complex health conditions, high workloads, and staffing shortages- challenges that intensified during the pandemic. Research suggests that these factors have affected nurses' well-being, and the quality of care provided to residents.ObjectiveThis study aimed to compare nurses’ (RNs/LPNs) quality of work life before and after the pandemic, focusing on changes in work-related stress, burnout, and well-being.MethodsWe conducted a repeated cross-sectional analysis of data from a stratified random sample of 39 urban Long Term Care homes in Anonymized and Anonymized that are part of the Translating Research into Elder Care Program. Data were collected pre pandemic (February 2020) and during the pandemic (December 2021), with only homes participating in both waves included. Two-level random-intercept regression models were used, adjusting for demographics and Long Term Care home characteristics. A total of 389 and 430 nurses completed surveys in 2020 and 2021, respectively.ResultsThe 2021 sample reported working significantly more hours (B = .04, p<.001), with increased emotional exhaustion (B = 0.17, p = .038, 95% CI: 0.09, 0.32), a key component of burnout, and a small but significant decline in mental health (B = -0.03, p = .007, 95% CI: -0.05, -0.01). Another component of burnout, cynicism, was significantly lower during the pandemic (B = -0.28, p<.001, 95% CI: -0.42, -0.14). No significant changes were observed in job satisfaction, professional efficacy, or physical health.ConclusionsDespite rising emotional exhaustion and declining mental health, cynicism did not follow the pattern expected under crisis conditions, suggesting that components of burnout may respond differently to pandemic stressors. Our findings have implications for supporting Long Term Care nurses through targeted interventions, particularly in managing workloads and mental health. What is already known •While emerging research has described pandemic-related strain across the LTC workforce, most studies have been cross-sectional, focused on care aides or mixed staff groups, leaving limited evidence on how nurses' quality of work life changed from before to during the pandemic.•As a result, little is known about how the pandemic specifically affected LTC nurses' emotional exhaustion, mental health, and job experiences over time. What this paper adds •This study provides a pre- and mid-pandemic comparison of LTC nurses in Canada, revealing significant increases in emotional exhaustion and declines in mental health, alongside an unexpected decrease in cynicism.•It also suggests that burnout dimensions may follow distinct trajectories under crisis conditions, a pattern consistent with theoretical accounts in which meaning- and value-based resources buffer against cynicism even when workload demands drive exhaustion.•The results identify specific areas, such as workload and mental health support, where targeted interventions are most urgently needed to sustain the LTC nursing workforce. What is already known •While emerging research has described pandemic-related strain across the LTC workforce, most studies have been cross-sectional, focused on care aides or mixed staff groups, leaving limited evidence on how nurses' quality of work life changed from before to during the pandemic.•As a result, little is known about how the pandemic specifically affected LTC nurses' emotional exhaustion, mental health, and job experiences over time. •While emerging research has described pandemic-related strain across the LTC workforce, most studies have been cross-sectional, focused on care aides or mixed staff groups, leaving limited evidence on how nurses' quality of work life changed from before to during the pandemic. While emerging research has described pandemic-related strain across the LTC workforce, most studies have been cross-sectional, focused on care aides or mixed staff groups, leaving limited evidence on how nurses' quality of work life changed from before to during the pandemic. •As a result, little is known about how the pandemic specifically affected LTC nurses' emotional exhaustion, mental health, and job experiences over time. As a result, little is known about how the pandemic specifically affected LTC nurses' emotional exhaustion, mental health, and job experiences over time. What this paper adds •This study provides a pre- and mid-pandemic comparison of LTC nurses in Canada, revealing significant increases in emotional exhaustion and declines in mental health, alongside an unexpected decrease in cynicism.•It also suggests that burnout dimensions may follow distinct trajectories under crisis conditions, a pattern consistent with theoretical accounts in which meaning- and value-based resources buffer against cynicism even when workload demands drive exhaustion.•The results identify specific areas, such as workload and mental health support, where targeted interventions are most urgently needed to sustain the LTC nursing workforce. •This study provides a pre- and mid-pandemic comparison of LTC nurses in Canada, revealing significant increases in emotional exhaustion and declines in mental health, alongside an unexpected decrease in cynicism. This study provides a pre- and mid-pandemic comparison of LTC nurses in Canada, revealing significant increases in emotional exhaustion and declines in mental health, alongside an unexpected decrease in cynicism. •It also suggests that burnout dimensions may follow distinct trajectories under crisis conditions, a pattern consistent with theoretical accounts in which meaning- and value-based resources buffer against cynicism even when workload demands drive exhaustion. It also suggests that burnout dimensions may follow distinct trajectories under crisis conditions, a pattern consistent with theoretical accounts in which meaning- and value-based resources buffer against cynicism even when workload demands drive exhaustion. •The results identify specific areas, such as workload and mental health support, where targeted interventions are most urgently needed to sustain the LTC nursing workforce. The results identify specific areas, such as workload and mental health support, where targeted interventions are most urgently needed to sustain the LTC nursing workforce. Background Residential Long Term Care and nursing homes were deeply affected by the COVID-19 pandemic which exacerbated existing challenges and introduced new stressors for healthcare professionals, particularly nurses (Cohen-Mansfield and Meschiany, 2022; Hanna et al., 2021; Reynolds et al., 2022; Robinson et al., 2021; White et al., 2021). Beyond direct care and care management of residents with complex health conditions, such as chronic diseases and cognitive impairments (Compton et al., 2023; González-Gil, 2016), nurses are responsible for overseeing and coordinating care plans, ensuring medication management, leading quality improvement initiatives, supervising the large unregulated workforce, and providing training to other staff members (Bolt et al., 2019; Montayre and Montayre, 2017). Further, nurses often serve in leadership roles, guiding teams through critical decisions and maintaining the standards of care within their facilities (Marquis and Huston, 2009). However, the pandemic intensified pre-existing issues such as staffing shortages and high workloads, as isolation requirements for infected staff further strained resources (Munkeby et al., 2023b; Reynolds et al., 2022; Shamon et al., 2023; White et al., 2021), compromising care quality and exacerbating nurse burnout (2). Additionally, the pandemic led nurses to experience increased emotional strain, role conflicts, moral distress, and ethical dilemmas, due to the inability to provide adequate care and the need to choose between their own safety and the care needs of residents, especially when protective equipment was in short supply (Blanco-Donoso et al., 2021; Munkeby et al., 2023a; Schulze et al., 2022; Shamon et al., 2023; White et al., 2021). Additionally, nurses often experienced elevated levels of stress, burnout, and moral distress, especially when faced with complex decisions in understaffed and under-resourced environments (Arble et al., 2021; Rushton et al., 2015). Job satisfaction among nurses is a critical factor influencing their well-being, retention, and the quality of patient care (Lu et al., 2012, 2019). In Long Term Care settings, job satisfaction is often influenced by factors such as staffing levels, work environment, leadership support, and opportunities for professional development (Aloisio et al., 2019; Ejaz et al., 2008). Low job satisfaction has been linked to higher levels of burnout, increased turnover intentions, and decreased quality of care (Al Sabei et al., 2020; Dall’Ora et al., 2020). During the pandemic, job satisfaction among nurses declined due to increased workloads, fear of infection, inadequate resources, and emotional exhaustion (Lavoie‐Tremblay et al., 2022; Said and El-Shafei, 2021). In Long Term Care homes, these factors have been particularly acute, given the high vulnerability of residents and the intense scrutiny these settings have faced (Lansbury et al., 2017; Peterson et al., 2021). Understanding changes in job satisfaction during the pandemic is essential for developing strategies to support nurses, improve their quality of work life, and ensure the delivery of high-quality care to residents. Burnout has long been recognized as a widespread phenomenon within the nursing profession (De Oliveira et al., 2021), particularly in Long Term Care homes, where the demands of care are high, and resources are often limited (Dall’Ora et al., 2020; Yeatts et al., 2018). Some of the key factors identified in the literature as leading to burnout include low job satisfaction, perceived inadequate staffing levels, poor care environments, and a lack of support from leadership (Chu et al., 2014; Compton et al., 2023; Costello et al., 2019; Dall’Ora et al., 2020; Shah et al., 2021). Organizational factors, like resource availability and training effectiveness, also influence burnout levels, while high workloads and role conflicts exacerbated emotional exhaustion (Dall’Ora et al., 2020; De Oliveira et al., 2021; Li and Yamamoto-Mitani, 2021; Yeatts et al., 2018). According to a recent survey by the Registered Nurses' Association of Ontario (Registered Nurses Association of Ontario (RNAO), 2022), burnout among nurses in Canada has reached alarming levels, with over 75% of nurses reporting feelings of burnout. This strain has led to a concerning trend where 69% of nurses indicated plans to leave their current positions within five years, and 42% are considering leaving the profession entirely (Registered Nurses Association of Ontario (RNAO), 2022). These statistics underscore the critical need to understand the well-being and quality of work life of nurses, particularly those working in Long Term Care, where challenges have been especially severe before, during, and after the pandemic. The cyclical relationship between quality of life and burnout among nurses is well-documented. For instance, a study by Babapour et al. (Babapour et al., 2022) found a significant negative relationship between job stress and quality of life among nurses, indicating that increased job stress can lead to lower quality of life, which in turn can exacerbate burnout. Current literature indicates that burnout in Long Term Care settings is driven by a complex interaction of multiple factors such as workload or staffing shortages, resource constraints, moral distress or ethical challenges, ageism, and compassion fatigue (Camero and Carrico, 2022; Costello et al., 2019; Iecovich and Avivi, 2017; Leskovic et al., 2020; Malagon-Aguilera et al., 2020; Yeatts et al., 2018). Even before COVID-19, nurses working in Long Term Care homes faced significant challenges that contributed to burnout and poor quality of life (QOL). A study found that mental health-related QOL among care staff in Long Term Care were much lower compared to age and sex-matched populations (Costello et al., 2019). These adverse conditions negatively impact both nurses' well-being and the quality of care provided to residents. Researchers have drawn some attention to how COVID-19 has affected Long Term Care staff, yet much of the focus has been on combined staff groups, leaving the unique experiences of nurses underexplored (Arble et al., 2021; Cho et al., 2023; Havaei et al., 2022; Reynolds et al., 2022; Shamon et al., 2023; Song et al., 2024; White et al., 2021). Very few longitudinal studies have been conducted to compare the mental health and well-being of healthcare workers before and after the pandemic (Spányik et al., 2023). While some studies have focused solely on registered nurses, all were cross-sectional and qualitative studies (Leskovic et al., 2020; Munkeby et al., 2023a; Schulze et al., 2022). Only one longitudinal study on the healthcare workforce was conducted in Canada (Shamon et al., 2023), and to our knowledge, no studies have compared Long Term Care nurses in Canada pre- and post-COVID-19. Grasping the effects of the pandemic on the quality of work life for nurses is essential, as it sheds light on their unique challenges and stresses, directly affecting their well-being and job performance. It also underscores the need for targeted strategies to bolster nurse support and enhance patient care during the recent, as well as future public health crises. This study examines the impact of the COVID-19 pandemic on nurses' quality of work life in Long Term Care settings. Specifically, it explores changes in nurses’ well-being, mental health, burnout and factors contributing to these changes during a 2-year period. Our study defines February 2020 as the pre-pandemic period, as this was immediately before the onset of the COVID-19 pandemic in Canada. The first confirmed case of COVID-19 in Canada was reported in January 2020, and by March 2020, the situation escalated quickly with the virus spreading more widely. This led to all provinces and territories declaring states of emergency between late March and early April 2020 (Public Health Agency of Canada, 2020). These declarations marked a significant shift in public health responses, including the implementation of widespread lockdowns and public health measures aimed at controlling the spread of the virus. While previous research has documented adverse impacts of the pandemic on healthcare workers, most studies have assumed that all dimensions of burnout deteriorate in parallel during crises and few have looked at nurses in the long-term care sector. Our study tests this assumption with pre- and mid-pandemic data from the same long-term care homes, and by examining whether different components of burnout (emotional exhaustion, cynicism, and efficacy) may respond differently to crisis conditions. Factors such as solidarity and shared purpose may differentially buffer against psychological disengagement from work while failing to protect against energy depletion (Dall’Ora et al., 2020; Schulze et al., 2022; White et al., 2021). Methods Study design We completed repeated cross-sectional analyses of data from two waves of surveys completed in February 2020 (immediately pre-pandemic; Time 1) and again in December 2021 (22 months into the pandemic; Time 2). These data are part of the Translating Research into Elder Care program. Translating Research into Elder Care is a pan-Canadian health service research program that collected longitudinal data from regulated and unregulated healthcare workers in Long Term Care since 2008. The goal of Translating Research into Elder Care is to improve the quality of care for residents and healthcare professionals working in Long Term Care. For this study, we used survey data (Time 1 and 2) from nurses in Long Term Care. Setting We sampled nurses in urban Long Term Care homes in the Anonymized and Anonymized regions (Anonymized) and the Anonymized region (Anonymized). Provincial government licensed homes were sampled using proportional, stratified random sampling. The sampling strata were: (1) health region (Anonymized and Anonymized Zones and the Anonymized), (2) Long Term Care home size (small, <80 beds; medium, 80–120 beds; large, > 120 beds), and (3) owner-operator model (public non-profit, private for-profit, and voluntary non-profit). For each health region, we generated the list of eligible Long Term Care homes. A random number generator was used to create final lists of selected Long Term Care homes. The provincial lead investigators used these lists and followed a standardized procedure for recruitment, and if needed, replacement of Long Term Care homes. Participants Nurses were invited to participate if they had worked in a study Long Term Care home for longer than 3 months, could identify a unit where they worked for at least 50% of their time, and worked on that unit for 6 or more shifts in the past month. Nurses completed surveys online using Translating Research into Elder Care’s customized survey platform. We were unable to link participants from Time 1 to 2. Eligible nurses who agreed to participate completed the entire online survey anonymously. Measures The study's variables and measurements are compiled in Table 1. Briefly they were: Missed and rushed care tasks (Knopp‐Sihota et al., 2015; Song et al., 2020); burnout, the 9-item Maslach Burnout Inventory (Barnett et al., 1999; Beckstead, 2002; Wang et al., 2024); job satisfaction, an adapted version of the Michigan Organizational Assessment Questionnaire Job Satisfaction Sub-scale (Ginsburg et al., 2016; Schleicher et al., 2011); physical and mental health, the 8-item Short Form Survey (Ware et al., 2001). Demographic measures such as age, sex, years in current role, hours worked, education, and primary shift worked were also collected.Table 1Summary of measures.Table 1 dummy alt textVariableDescriptionNo. itemsScoringBurnout (MBI-GS-9)MBI emotional exhaustionRespondent feels emotionally exhausted, strained3Continuous: Mean of 7-point (0–6) Likert scale (never to daily)MBI cynicismRespondent feels cynical, that their work does not contribute to anything3Continuous: Mean of 7-point (0–6) Likert scale (never to daily)MBI efficacyRespondent feels their work is meaningful, has a sense of accomplishment3Continuous: Mean of 7-point (0–6) Likert scale (never to daily)Health status (SF-8))Physical healthRespondent’s perception of their own physical health in the most recent 4 weeks8Continuous: Scoring of items on 5- or 6point Likert scales based on scale developers’ algorithmMental healthRespondent’s perception of their own mental health in the most recent 4 weeksLong Term Care Home CharacteristicsOwner-Operator ModelOwnership model of the Long Term Care home1Categorical: Public not-for-profit, voluntary not-for-profit, private for-profitSizeNumber of beds in the Long Term Care home1Categorical: Small (≤79 beds), medium, (80–120 beds), large (>120 beds)ProvinceThe province where the Long Term Care home was located1Categorical: Anonymized, AnonymizedData Collection WaveTime when data collection occurred1Binary: February 2020, December 2021Nurse DemographicsAgeRespondent’s age, years1Categorical: <30, 30–39, 40–49, 50–59, ≥60Completed diplomaNurse completed diploma1Binary: Yes, NoCompleted bachelor's degreeNurse Completed bachelor's degree1Binary: Yes, NoShift worked most oftenShift the respondent works most often1Categorical: Day, evening, nightYears in current roleNumber of years respondent has worked as a nurse1ContinuousYears in current unitNumber of years respondent has worked on the current care unit1ContinuousNurse Quality of Work LifeWork Short StaffedRespondent feels working short-staffed1Binary: Every day to weekly, monthly to neverHours worked in two weeksNumber of hours that respondent worked in the past two weeks1ContinuousJob satisfaction (Michigan organizational assessment questionnaire job satisfaction sub-scale)Respondent satisfied with the current job3Continuous: Mean of 5-point (1–5) Likert scale (strongly disagree to strongly agree)Mental healthRespondent’s perception of their own mental health in the most recent 4 weeksWork Environment (Alberta Context Tool)LeadershipActions of formal leaders in an organization (unit) to influence changes and excellence in practice. Items generally reflect emotionally intelligent leadership6Mean of items on a 5-point Likert scale (strongly disagree to strongly agree)CultureThe way that “we do things” in our organizations and work units. Items generally reflect a supportive work culture6Mean of items on a 5-point Likert scale (strongly disagree to strongly agree)Social CapitalStructural elements of an organization (unit) that facilitate the ability to assess and use knowledge75-point Likert frequency scale (never to almost always). Recode 0 = no resource, 1 = resource. Take a count of recoded itemsEvaluationThe process of using data to assess group/team performance and to achieve outcomes in organizations or units6Mean of items on a 5-point Likert scale (strongly disagree to strongly agree) Table 1Summary of measures.Table 1 dummy alt textVariableDescriptionNo. itemsScoringBurnout (MBI-GS-9)MBI emotional exhaustionRespondent feels emotionally exhausted, strained3Continuous: Mean of 7-point (0–6) Likert scale (never to daily)MBI cynicismRespondent feels cynical, that their work does not contribute to anything3Continuous: Mean of 7-point (0–6) Likert scale (never to daily)MBI efficacyRespondent feels their work is meaningful, has a sense of accomplishment3Continuous: Mean of 7-point (0–6) Likert scale (never to daily)Health status (SF-8))Physical healthRespondent’s perception of their own physical health in the most recent 4 weeks8Continuous: Scoring of items on 5- or 6point Likert scales based on scale developers’ algorithmMental healthRespondent’s perception of their own mental health in the most recent 4 weeksLong Term Care Home CharacteristicsOwner-Operator ModelOwnership model of the Long Term Care home1Categorical: Public not-for-profit, voluntary not-for-profit, private for-profitSizeNumber of beds in the Long Term Care home1Categorical: Small (≤79 beds), medium, (80–120 beds), large (>120 beds)ProvinceThe province where the Long Term Care home was located1Categorical: Anonymized, AnonymizedData Collection WaveTime when data collection occurred1Binary: February 2020, December 2021Nurse DemographicsAgeRespondent’s age, years1Categorical: <30, 30–39, 40–49, 50–59, ≥60Completed diplomaNurse completed diploma1Binary: Yes, NoCompleted bachelor's degreeNurse Completed bachelor's degree1Binary: Yes, NoShift worked most oftenShift the respondent works most often1Categorical: Day, evening, nightYears in current roleNumber of years respondent has worked as a nurse1ContinuousYears in current unitNumber of years respondent has worked on the current care unit1ContinuousNurse Quality of Work LifeWork Short StaffedRespondent feels working short-staffed1Binary: Every day to weekly, monthly to neverHours worked in two weeksNumber of hours that respondent worked in the past two weeks1ContinuousJob satisfaction (Michigan organizational assessment questionnaire job satisfaction sub-scale)Respondent satisfied with the current job3Continuous: Mean of 5-point (1–5) Likert scale (strongly disagree to strongly agree)Mental healthRespondent’s perception of their own mental health in the most recent 4 weeksWork Environment (Alberta Context Tool)LeadershipActions of formal leaders in an organization (unit) to influence changes and excellence in practice. Items generally reflect emotionally intelligent leadership6Mean of items on a 5-point Likert scale (strongly disagree to strongly agree)CultureThe way that “we do things” in our organizations and work units. Items generally reflect a supportive work culture6Mean of items on a 5-point Likert scale (strongly disagree to strongly agree)Social CapitalStructural elements of an organization (unit) that facilitate the ability to assess and use knowledge75-point Likert frequency scale (never to almost always). Recode 0 = no resource, 1 = resource. Take a count of recoded itemsEvaluationThe process of using data to assess group/team performance and to achieve outcomes in organizations or units6Mean of items on a 5-point Likert scale (strongly disagree to strongly agree) Summary of measures. Statistical analysis We used descriptive statistics such as means, standard deviations, and ranges for continuous variables, and frequencies and percentages for categorical variables, to describe nurse demographics as well as unit and home characteristics. Two-level random-intercept regression models were used to compare outcomes between data collection waves, with nurses (Level 1) nested within long-term care homes (Level 2). Models included the time-point indicator (February 2020 = reference, December 2021 = comparison) as the primary predictor. Using these regression models, we controlled for the owner-operator model of the long-term care home, the province where the home was located, primary shift worked, age, and size of home to account for structural, organizational, and demographic factors known to influence nurses' workload, stress, and well-being. Additionally, we controlled for culture, evaluation, social capital, and leadership at the facility level. These Alberta Context Tool (ACT) variables were aggregated to the facility level by computing the mean score for each Long Term Care home, and were grand mean centred prior to inclusion in the models. Adjusting for these variables helped to ensure that observed differences across time points reflected changes related to the pandemic rather than underlying contextual differences. All reported regression coefficients (B) are unstandardized and represent the adjusted mean difference between Time 2 and Time 1. To facilitate interpretation of effect magnitudes, we also calculated Cohen's d effect sizes for statistically significant outcomes by dividing the adjusted mean difference by the pooled standard deviation. <2% of data were missing for all variables, and we used listwise deletion for missing data. All analyses were conducted in SPSS 29. Ethics approval and consent This research involving human participants was performed in accordance with the Declaration of Helsinki and was reviewed and approved by the Research Ethics Board at Anonymized (Anonymized) and Anonymized (Anonymized). All participants indicated consent to participate by voluntarily completing the online survey. The recruitment information sheet about the study that each participant received indicated "By completing the survey online, you will be giving your consent to participate in the study." This approach was approved by the Research Ethics Board at Anonymized (Anonymized) and Anonymized (Anonymized). Results Comparison of sample characteristics at time 1 and time 2 As illustrated in Table 2, 382 nurses and 430 nurses across 37 Long Term Care homes completed the survey at Time 1 and Time 2, respectively. Age and education were the only two demographic variables that differed between time points such that the nurses in the December 2021 cohort were more educated and older than the February 2020 cohort.Table 2Demographics of nurses and characteristics of care units and Long Term Care homes.Table 2 dummy alt textNurse DemographicsFebruary 2020December 2021Mixed method RegressionNo. (%)No. (%)FSig.Age1.940.031<201 (0.26)0 (0)20–243 (0.78)9 (2.09)25–2938 (9.92)21 (4.88)30–3453 (13.84)43(10.00)35–3933 (8.62)62 (14.42)40–4458 (15.14)58 (13.49)45–4967 (17.49)66 (15.35)50–5450 (13.05)70 (16.28)55–5935 (9.14)47 (10.93)60–6433 (8.62)35 (8.14)65–709 (2.35)16 (3.72)>703 (0.78)3 (0.70)Diploma/certificate received9.490.002Yes280 (73.30)306 (71.16)No63 (16.49)34 (7.91)Missing*39 (10.21)90 (20.93)Bachelor's degree received21.62<.001Yes194 (50.79)235 (53.65)No130 (34.03)69 (16.05)Missing*58 (15.18)126 (29.30)Shift worked most of the time1.210.300Day Shift184 (48.17)223 (51.86)Evening Shift151 (39.53)146 (33.95)Night Shift46 (12.04)53 (12.33)Missing*1 (0.26)8 (1.86)Province0.2870.592Anonymized291 (76.18)320 (74.42)Anonymized91 (23.82)110 (25.58)February 2020December 2021Mixed method RegressionNurse DemographicsM (SD)M (SD)BSig.Years in current role10.36 (8.41)10.34 (8.91)0.070.143Years in current unit7.00 (6.18)6.60 (6.08)−0.060.179Number of hours worked in 2 weeks56.92(19.61)62.04 (20.03).00<.001February 2020December 2021Mixed method RegressionUnit CharacteristicsM (SD)M (SD)BSig.Leadership3.90 (0.84)3.94 (0.84)0.070.119Culture3.97 (0.66)3.94 (0.66)0.000.987Social Capital4.01 (0.55)3.96 (0.55)−0.020.523Evaluation3.69 (0.74)3.61 (0.74)−0.050.09February 2020December 2021Mixed method RegressionLong Term Care Home CharacteristicsNo. (%)No. (%)FSig.Owner Operator Model (Long Term Care home)0.0010.999Public not for profit8 (21.62)8 (21.62)Private for profit11 (29.73)11 (29.73)Voluntary not for profit18 (48.65)18 (48.65)Owner Operator Model (Participant)0.200.751Public not for profit82 (21.47)86 (20.00)Private for profit105 (27.49)112 (26.05)Voluntary not for profit195 (51.05)232 (53.95)Size of Long Term Care home0.000.999Small (< 80 beds)50 (13.09)56 (13.02)Medium (80–120 beds)60 (16.00)68 (16.00)Large (> 120 beds)272 (71.00)306 (71.00)Notes. All comparisons used mixed method regression models to control for nurses clustered in the same homes. In the mixed method regression, only one independent variable was included—time points. Categorical variables were analyzed using multinomial logistic regressions. Numerical data was analyzed using mixed linear regressions. * Denotes missing because the question was not asked. Table 2Demographics of nurses and characteristics of care units and Long Term Care homes.Table 2 dummy alt textNurse DemographicsFebruary 2020December 2021Mixed method RegressionNo. (%)No. (%)FSig.Age1.940.031<201 (0.26)0 (0)20–243 (0.78)9 (2.09)25–2938 (9.92)21 (4.88)30–3453 (13.84)43(10.00)35–3933 (8.62)62 (14.42)40–4458 (15.14)58 (13.49)45–4967 (17.49)66 (15.35)50–5450 (13.05)70 (16.28)55–5935 (9.14)47 (10.93)60–6433 (8.62)35 (8.14)65–709 (2.35)16 (3.72)>703 (0.78)3 (0.70)Diploma/certificate received9.490.002Yes280 (73.30)306 (71.16)No63 (16.49)34 (7.91)Missing*39 (10.21)90 (20.93)Bachelor's degree received21.62<.001Yes194 (50.79)235 (53.65)No130 (34.03)69 (16.05)Missing*58 (15.18)126 (29.30)Shift worked most of the time1.210.300Day Shift184 (48.17)223 (51.86)Evening Shift151 (39.53)146 (33.95)Night Shift46 (12.04)53 (12.33)Missing*1 (0.26)8 (1.86)Province0.2870.592Anonymized291 (76.18)320 (74.42)Anonymized91 (23.82)110 (25.58)February 2020December 2021Mixed method RegressionNurse DemographicsM (SD)M (SD)BSig.Years in current role10.36 (8.41)10.34 (8.91)0.070.143Years in current unit7.00 (6.18)6.60 (6.08)−0.060.179Number of hours worked in 2 weeks56.92(19.61)62.04 (20.03).00<.001February 2020December 2021Mixed method RegressionUnit CharacteristicsM (SD)M (SD)BSig.Leadership3.90 (0.84)3.94 (0.84)0.070.119Culture3.97 (0.66)3.94 (0.66)0.000.987Social Capital4.01 (0.55)3.96 (0.55)−0.020.523Evaluation3.69 (0.74)3.61 (0.74)−0.050.09February 2020December 2021Mixed method RegressionLong Term Care Home CharacteristicsNo. (%)No. (%)FSig.Owner Operator Model (Long Term Care home)0.0010.999Public not for profit8 (21.62)8 (21.62)Private for profit11 (29.73)11 (29.73)Voluntary not for profit18 (48.65)18 (48.65)Owner Operator Model (Participant)0.200.751Public not for profit82 (21.47)86 (20.00)Private for profit105 (27.49)112 (26.05)Voluntary not for profit195 (51.05)232 (53.95)Size of Long Term Care home0.000.999Small (< 80 beds)50 (13.09)56 (13.02)Medium (80–120 beds)60 (16.00)68 (16.00)Large (> 120 beds)272 (71.00)306 (71.00)Notes. All comparisons used mixed method regression models to control for nurses clustered in the same homes. In the mixed method regression, only one independent variable was included—time points. Categorical variables were analyzed using multinomial logistic regressions. Numerical data was analyzed using mixed linear regressions. * Denotes missing because the question was not asked. Demographics of nurses and characteristics of care units and Long Term Care homes. Notes. All comparisons used mixed method regression models to control for nurses clustered in the same homes. In the mixed method regression, only one independent variable was included—time points. Categorical variables were analyzed using multinomial logistic regressions. Numerical data was analyzed using mixed linear regressions. * Denotes missing because the question was not asked. To examine whether the observed demographic differences between the two samples influenced our findings, we conducted sensitivity analyses in which education (bachelor's degree: yes/no) was added as an additional covariate in all regression models. The pattern and significance of results remained consistent with the primary analyses, suggesting that cohort differences in age and educational attainment do not account for the observed changes in nurses' quality of work life outcomes. Comparison of quality of work life at time 1 and time 2 In Table 3 we present the quality of work life comparisons for nurses at Time 1 and Time 2. On average, nurses reported higher levels of emotional exhaustion (B = 0.17, p = .038, 95% CI: 0.09, 0.32), and lower levels of cynicism (B = −0.28, p<.001, 95% CI: −0.42, −0.14), both key components of burnout, at Time 2 compared to Time 1. They also reported lower mental health at time 2 than pre-COVID-19 pandemic (B = −0.03, p = .007, 95% CI: −0.05, −0.01). No statistically significant changes were observed in job satisfaction, physical health, or the professional efficacy subscale of burnout. The number of responsive behaviours by residents approached significance (B = 0.06, p = .050). All changes, including statistically significant ones, were relatively small. Effect sizes for statistically significant outcomes were: emotional exhaustion (Cohen's d = 0.14, small effect), cynicism (d = 0.07, negligible to small effect), and mental health (d = 0.23, small effect). The unstandardized coefficient for mental health (B = −0.03) appears small because it reflects the effect of the binary time-point predictor after controlling for covariates; the raw mean difference of 2.29 points provides more intuitive context for the magnitude of change.Table 3Comparison of quality of work life for nurses at time 1 (February 2020) compared to Time 2 (December 2021).Table 3 dummy alt textQuality Of Work Life OutcomeTime 1February 2020Time 2December 202195% CIM (SD)M (SD)BLLULSigNumber of responsive behaviours by residents9.48 (3.73)10.28 (4.71)0.060.000.120.050MBI Exhaustion2.13 (1.42)2.34 (1.50)0.170.090.32.038MBI Cynicism2.11 (1.40)2.01 (1.38)−0.28−0.42−0.14<.001MBI Efficacy4.52 (1.11)4.56 (1.12)0.06−0.070.180.363Job Satisfaction4.14 (0.76)4.14 (0.70)−0.01−0.210.200.940SF-8 Physical Health49.68 (8.00)48.44 (8.16)−0.01−0.030.010.275SF-8 Mental Health50.42 (9.36)48.13 (10.17)−0.03−0.05−0.010.007Note: All B coefficients are unstandardized. Models controlled for owner-operator model, province, primary shift worked, age, size of home, and facility-level aggregates of ACT culture, evaluation, social capital, and leadership. Effect sizes (Cohen's d): Emotional exhaustion d = 0.14 (small); Cynicism d = 0.07 (negligible to small); Mental health d = 0.23 (small). Table 3Comparison of quality of work life for nurses at time 1 (February 2020) compared to Time 2 (December 2021).Table 3 dummy alt textQuality Of Work Life OutcomeTime 1February 2020Time 2December 202195% CIM (SD)M (SD)BLLULSigNumber of responsive behaviours by residents9.48 (3.73)10.28 (4.71)0.060.000.120.050MBI Exhaustion2.13 (1.42)2.34 (1.50)0.170.090.32.038MBI Cynicism2.11 (1.40)2.01 (1.38)−0.28−0.42−0.14<.001MBI Efficacy4.52 (1.11)4.56 (1.12)0.06−0.070.180.363Job Satisfaction4.14 (0.76)4.14 (0.70)−0.01−0.210.200.940SF-8 Physical Health49.68 (8.00)48.44 (8.16)−0.01−0.030.010.275SF-8 Mental Health50.42 (9.36)48.13 (10.17)−0.03−0.05−0.010.007Note: All B coefficients are unstandardized. Models controlled for owner-operator model, province, primary shift worked, age, size of home, and facility-level aggregates of ACT culture, evaluation, social capital, and leadership. Effect sizes (Cohen's d): Emotional exhaustion d = 0.14 (small); Cynicism d = 0.07 (negligible to small); Mental health d = 0.23 (small). Comparison of quality of work life for nurses at time 1 (February 2020) compared to Time 2 (December 2021). Note: All B coefficients are unstandardized. Models controlled for owner-operator model, province, primary shift worked, age, size of home, and facility-level aggregates of ACT culture, evaluation, social capital, and leadership. Effect sizes (Cohen's d): Emotional exhaustion d = 0.14 (small); Cynicism d = 0.07 (negligible to small); Mental health d = 0.23 (small). Discussion This study examined changes in nurses' quality of work life in Long Term Care settings before and during the COVID-19 pandemic. Our findings describe increased emotional exhaustion as well as decreased cynicism and mental health during the pandemic. Increased emotional exhaustion The significant increase in emotional exhaustion among nurses during the pandemic as noted in this study is consistent with previous literature linking COVID-19 stressors and increased emotional exhaustion among healthcare professionals (Ghahramani et al., 2021; Sexton et al., 2022). Previous research indicated that the pandemic intensified staffing shortages and increased workload demands, particularly in Long Term Care facilities (Estabrooks et al., 2020; Leskovic et al., 2020; Shamon et al., 2023). Similar trends have been observed in other studies conducted during the pandemic. A longitudinal study by Spányik et al. (2023) found that healthcare workers experienced significant increases in emotional exhaustion and secondary traumatic stress over time. Those working on the frontline were particularly affected (Spányik et al., 2023). Post-COVID-19 studies have further highlighted the challenges faced by healthcare workers, including concerns over protective equipment supply, evolving protocols, regulation challenges, and fears of contracting and spreading COVID-19 to family members (Estabrooks et al., 2020; White et al., 2021). These challenges may have collectively contributed to the increased emotional exhaustion observed among nurses during the pandemic. Decrease in cynicism Contrary to expectations, cynicism levels among nurses decreased during the pandemic. This is a key finding, as it challenges the common assumption that all components of burnout worsen uniformly during crises. While the Job Demands-Resources model (Bakker et al., 2023) typically predicts that increased demands lead to increased burnout across all dimensions, our findings suggest a more nuanced pattern. Drawing on previous research (Arble et al., 2021; Estabrooks et al., 2023) and theoretical frameworks such as Conservation of Resources theory (Hobfoll, 1989), one potential explanation for our findings is that factors such as resilience, teamwork, and a sense of collective mission may serve as buffers against negative emotions such as cynicism, particularly in high-stress environments like healthcare during crises (Murden et al., 2018; Rooney et al., 2024; Schulze et al., 2022; White et al., 2021). However, we acknowledge that these constructs were not directly measured in our study, and alternative explanations cannot be ruled out. For example, selective retention of more engaged workers or survivorship of resilient workers (see Limitations) may be alternative explanations. Most pre-pandemic research often linked increased workload and adverse job characteristics to higher burnout and cynicism among healthcare workers (Costello et al., 2019). However, the pandemic introduced conditions under which workers felt united by a common sense of risk, high workloads and uncertainty, but also a common purpose of protecting vulnerable residents (Chimed-Ochir et al., 2023). Meaning-based coping strategies, identified in the stress and coping literature (Folkman, 2008), may explain this divergence. When workers perceive their efforts as contributing to a meaningful collective goal, such as protecting vulnerable residents during a pandemic, they may maintain engagement with their work even while experiencing emotional depletion. Ambrose et al. (2022) reported that healthcare workers demonstrated enhanced teamwork and resilience in response to the pandemic. Additionally, Schulze et al. (2022) highlighted how the pandemic enhanced community cohesion and team spirit among nurses working in long-term care facilities. Despite experiencing significant psychosocial burdens from the pandemic, the nurses in our study reported high levels of efficacy in their work, which may explain the observed decrease in cynicism levels. However, they also faced increased emotional exhaustion, while their sense of professional efficacy remained unchanged. Notably, our finding contrasts with research on nursing home managers during the same period, who showed increased cynicism (Estabrooks et al., 2023). This divergence may reflect differences between frontline and leadership roles: managers face additional burdens of implementing rapidly changing policies and managing staffing crises, potentially without the same direct sense of purpose derived from hands-on resident care. Future research should examine whether role-specific factors moderate the relationship between crisis conditions and different burnout dimensions. Taken together, our findings suggest that burnout dimensions do not operate the same during a crisis. Future research should measure these resources directly and examine whether interventions that strengthen them can protect against burnout progression in Long Term Care settings. Decline in mental health The small but significant decline in mental health scores among nurses during the pandemic aligns with broader trends in the literature surrounding the healthcare workforce (Amit Aharon et al., 2021; Estabrooks et al., 2023; Varghese et al., 2021). Pre-pandemic literature already acknowledged the negative impact of workload and stress on nurses' mental health (Holland et al., 2019; Jenkins and Elliott, 2004; Tyler and Cushway, 1995). Many studies identified factors such as inadequate staffing, high patient-to-nurse ratios, and poor work environments as significant contributors to burnout and mental health issues among Long Term Care staff (Costello et al., 2019; Yeatts et al., 2018). However, the pandemic added unprecedented layers of stress, including ethical dilemmas, the fear of infection, and role conflicts, which likely intensified these pre-existing issues. Schulze et al. (2022) identified that nurses in Long Term Cares experienced heightened psychosocial burdens during the pandemic, including increased presenteeism, emotional exhaustion, and difficulty relaxing, all of which contributed to deteriorating mental health. Another study by White et al. (2021) found that nursing staff faced not only professional challenges but also personal concerns about their own health and that of their families during the pandemic. Nurses navigate disruptions in their everyday lives, concerns about family health, and limited social support due to lockdowns (Pfefferbaum and North, 2020). The dual burden of balancing the demands of patient care with the fear of bringing the virus home increased their mental health challenges. The decline in mental health scores from 50.42 at Time 1 to 48.13 at Time 2, a difference of 2.29 points, approaches or meets the minimally clinically important difference threshold typically cited for the SF-8 and related SF measures, which ranges from 2 to 3 points (Ware and Kosinski, 2001). Thus, while we described this change as "small but significant" in statistical terms, it likely represents a clinically meaningful deterioration in mental well-being for the nursing workforce. This magnitude of change, sustained across hundreds of nurses, has implications for both individual health and workforce stability. Overall, our study's finding of a small but significant decline in nurses' mental health scores during the pandemic aligns with existing research which documented how increased workloads, fear of infection, and social isolation cumulatively impact psychological well-being. Declining mental health among Long Term Care nurses can lead to anxiety, depression, and other psychological issues (Malagon-Aguilera et al., 2020), underscoring the need to address both professional and personal stressors to prevent long-term impacts. Strengths and limitations This study has several notable strengths. First, the use of data collected both before and during the pandemic allowed for direct comparisons of nurses' experiences across time points, a design feature rare in pandemic research where most studies rely on retrospective accounts or post-pandemic-only data. Second, the focus on nurses in long-term care settings addresses an important gap in the literature, as most research has centred on acute care or general healthcare environments, or has combined nurses with other staff groups. Third, the study drew on a relatively large sample of nurses given the overall number of nurses working in Long Term Care, with 382 and 430 nurses completing surveys at Time 1 and Time 2, respectively. Fourth, the sample was carefully selected using proportional, stratified random sampling across health regions, Long Term Care home sizes, and owner-operator models, enhancing the representativeness of the findings within the sampled provinces. Fifth, the study used rigorously tested and validated measures, including the Maslach Burnout Inventory (MBI-GS-9), the SF-8 health survey, and the Michigan Organizational Assessment Questionnaire job satisfaction subscale. Finally, the use of two-level random-intercept regression models accounted for the nested structure of the data, with nurses clustered within Long Term Care homes, reducing the risk of biased standard errors and providing more accurate estimates of the relationships examined. Several limitations must be acknowledged. First, and most critically, we were not able to link the same nurses across Time 1 and Time 2, which means our study assumes and compares two independent cross-sectional samples rather than tracking individual-level change. This design introduces the possibility that observed differences reflect cohort composition changes rather than true within-person changes over time. For example, the Time 2 sample was significantly older and more educated than the Time 1 sample, which may indicate survivor bias. Second, regarding generalizability to other healthcare systems, our findings should be interpreted in the context of the Canadian long-term care system, which differs from other countries, among other things, in terms of funding models, staffing regulations, and nurse-to-resident ratios; as well pandemic response varied considerably across jurisdictions in terms of public health measures, vaccine availability, etc. Further, our sample was limited to urban long-term care homes in two Canadian provinces (Alberta and Manitoba). Our findings should not be generalized to rural contexts. Third, the current study relied on participants responding to survey questions. Surveys rely on self-report which can introduce issues such as recall bias and social desirability. Implications for Practice and Policy. The increase in emotional exhaustion and decline in mental health among nurses adds to a growing body of evidence that calls for interventions to support their well-being. Strategies could include implementing policies to manage workloads and providing mental health resources. The decrease in cynicism suggests that fostering teamwork and a shared sense of purpose may have protective effects on nurses' well-being. Organizations should consider promoting collaborative work environments and recognizing the collective efforts of staff during challenging times. Future research directions Further research is needed to explore the factors contributing to the decrease in cynicism and how this may be leveraged to improve nurses' work experience. Longitudinal studies could provide insights into the long-term effects of the pandemic on nurses' well-being. Qualitative research could also deepen our understanding of nurses' experiences and identify specific support mechanisms that are most effective, especially during times of crisis. Conclusion The COVID-19 pandemic had a significant impact on nurses in long-term care settings, increasing emotional exhaustion, and affecting mental health. The decline in mental health scores likely represents clinically meaningful deterioration. However, despite these challenges, nurses reported decreased levels of cynicism and reported no differences in job satisfaction, physical health, or professional efficacy, at least from baseline to 22 months into the pandemic. The divergent pattern across burnout components, with exhaustion increasing while cynicism decreased, may challenge assumptions that all dimensions of burnout worsen uniformly. The mechanisms underlying this divergence could not be identified with the present data and warrant direct investigation in future work. Addressing workload issues and providing mental health support are essential for sustaining this critical workforce and ensuring high-quality care for long-term care residents. Funding This work was supported by the Canadian Institutes of Health Research (CIHR #174890) and Alberta Health (AHW011810) and (AWHFP13-580). These sponsors were not involved in the design or implementation of the study. Availability of research data and materials The data supporting the conclusions of this article are housed in the secure and confidential Health Research Data Repository (HRDR) in the Faculty of Nursing at the University of Alberta (https://uofa.ualberta.ca/nursing/research/research-supports-and-services/hrdr), in accordance with the health privacy legislation of participating Translating Research into Elder Care jurisdictions. Data specific to this manuscript can be requested through the Translating Research into Elder Care Data Management Committee (htitley@ualberta.ca) on the condition that researchers meet and comply with the Translating Research into Elder Care and HRDR data confidentiality policies. CRediT authorship contribution statement Isaiah Hipel: Writing – review & editing, Writing – original draft, Formal analysis. Janice Keefe: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Methodology, Investigation, Formal analysis, Conceptualization. Amber Duynisveld: Writing – review & editing, Writing – original draft. Yuting Song: Writing – review & editing, Methodology, Investigation, Conceptualization. Yinfei Duan: Writing – review & editing, Methodology, Investigation, Conceptualization. Carole A Estabrooks: Writing – review & editing, Resources, Project administration, Methodology, Investigation, Funding acquisition, Conceptualization. Peter G Norton: Writing – review & editing, Methodology, Investigation, Conceptualization. Priscilla Packiam: Writing – review & editing, Writing – original draft. Janet Squires: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.