This research provides critical insights into how and why emotional distress, measured approximately every six months, varied from March 2021 to December 2023. In our study of Canadian HCP, mean K10 scores decreased from 20.4 to 17.3 from study start to end. Despite the declared end of the global health emergency in May 2023 [16], close to half of all HCP were still reporting moderate or higher levels of distress by study end, indicating long-lasting emotional impacts. Of interest, K10 scores did not decline monotonically; rather they increased during periods of increased COVID-19 mitigation activities and during the winter months. Non-periodic factors were also associated with levels of distress. Those who were taking medication to treat anxiety, depression, or insomnia had higher K10 scores while lower K10 scores were associated with older age, being male, having more children in the household, having been previously infected with COVID-19, and being a non-physician regulated HCP versus a nurse.
Time-related changes in emotional distress have been noted by other researchers. López Steinmetz and colleagues, who used the Argentinian version of the K10 (cut off > 20), reported a significant increase in the percent of HCP with K10 scores indicative of distress between April/May (52.5%) and September 2020 (62.6%) [27]. Using the 6-item Kessler Psychological Distress Scale (K6) [6], Maunder et al. reported that Canadian HCP emotional distress increased between September 2020 and January 2021 [13]. Both studies examined differences prior to vaccines being widely available. Kameno et al. followed HCP from 19 Japanese hospitals from February 2021 to October 2022 [28] and, similar to the current study, found that distress scores, as measured by the K6, significantly decreased over time and that the rate of change varied over time [28].
The decline in mean distress scores was not uniform; periodic factors impacted distress level trajectories. Our results indicate that K10 scores, as well as anxiety and depression sub-scale scores, were higher during the winter months. Winthorst et al. [22] reported a small increase in depressive symptoms in healthy controls during the winters of 2004–2007. Similarly, Maunder et al. [13] reported an increase in K6 scores during the winter of 2021. Wang et al. reported that, among Chinese operating room nurses surveyed between December 2021 and January 2022, lower K10 scores were associated with reduced sunlight exposure hours [26].
K10 scores also increased by 5.6% during periods when the mitigation intensity score associated with the three domains (schools, work, and other) was higher (≥ 7). For example, this could be when schools were closed (school domain rank as 3), when working from home was strongly suggested or most businesses were closed except for specific sectors/worker categories (work domain ranked as 2), and when there were moderately strict public gathering restrictions, some inter-provincial travel restrictions, closure or significantly reduced capacity of most indoor activities, and closure of some outdoor activities (other domain ranked as 2). HCP are not just essential workers but people who live within a social context impacted by all mitigation strategies. For example, an American survey of HCP conducted in December 2020 found that 49% of respondents had emergency childcare needs that disrupted their work in the past year and that 41% anticipated having unmet childcare needs in the next year [29]. Contextual factors need to be considered when determining the overall impact of stressful events on the emotional states of HCP.
While unmet childcare needs led to increased distress in some studies, similar to two other studies of Canadian HCP [13, 30], we found that having more children living in the household was associated with lower K10 scores. Mehta et al. [30] reported that living with child(ren) was associated with lower anxiety subscale scores but not with depression scores in Canadian ICU staff surveyed in 2020. Styra et al. [31] reported that informal sources of support from family members and others mitigated Canadian HCP distress and that 77% of HCP relied on such informal supports early in the pandemic. Taken together, having familial social support appears to mitigate HCP emotional distress but being unable to meet the needs of these important family members appears to increase distress.
Non-periodic factors were also associated with emotional distress. HCP K10 scores were 10% higher, on average, for respondents who were taking antidepressants, anti-anxiety or anti-insomnia medications. These drugs may have been prescribed for a pre-existing emotional health issue and, if so, should have helped mitigate symptoms of COVID-19-related emotional distress if they had been taken for an appropriate period of time. While some studies have linked pre-existing mental health issues with increased distress during COVID-19 [32], we were unable to determine whether the use of these medications preceded or followed the stresses of working during the pandemic. Other researchers of coping strategies used by HCP during the COVID-19 pandemic reported that Turkish physicians and German HCP coped by taking antidepressants [33, 34] or using psychotropic drugs [35] while a Canadian study found that HCP used alcohol [36] to cope with the increased stress. A second Canadian study also found that HCP used alcohol to cope with stress during the pandemic, but in addition, and similar to other studies [37], during qualitative interviews they also identified physical exercise, yoga, meditation, and interacting with friends and family as frequently used coping strategies. Regardless of the method of stress relief, there is a clear need to provide HCP working during stressful events with evidence-based stress relief.
In a systematic review of mental issues among HCP working during COVID-19, Arias-Ulloa et al. also noted that females were at greater risk of emotional distress. These authors suggest that sex differences may be due to the fact that males find it more difficult to recognize psychological distress [11]. Zhang et al. postulate that gender differences may be due to gender roles that may vary with age [38] suggesting further examination of possible interactions between other exploratory factors and gender in future studies. Further, considering that in 2021, 91% of the Canadian nurses were female [39] while in 2022, 45% of Canadian physicians [40] and 70% of physiotherapists [41] were female, further research into the impact of gendered roles on jobs performed during the pandemic and subsequent risk of emotional distress is warranted.
In the current study, older participants tended to have lower K10 scores across the 2.5 years of follow-up; a ten-year increase in age was associated with K10 scores that were 8% lower than those a decade younger. The relationship between age and emotional distress has not been consistent. Umbetkulova and colleagues [9] found that younger HCP were at greater risk for developing mental health issues than older ones in their systematic review. Meanwhile, Arias-Ulloa et al. [11] postulated that inconsistencies may be due to how stress was measured, how age was grouped, what confounders were measured, and how they were used in the analysis. In a third review paper, Galanis and colleagues [10] suggested that younger nurses may be more susceptible to burnout as they may be less familiar with infection control measures and less able to handle extreme events, such as a pandemic.
These findings suggest that in the face of highly stressful situations, healthcare institutions should closely monitor the mental health of staff to provide needed psychological support and stress relieving strategies such as regular meetings to reflect on existing problems and open comprehensive evaluation of organizational risk [42]. Supportive institutional practices identified by more than half of Canadian ICU staff responding to one survey included the need for clear and unambiguous communication from their institution, expressions of gratitude from the hospital’s leadership, having leadership who were open to hearing staff concerns, free or subsidized parking, and scrubs so they could change before going home [30].
The results of our study may have limited generalizability among HCP as study participants were self-selected leaving room for possible selection bias where those more interested in SARS-CoV-2 enrolled in the parent study and those more interested in distress completed the K10 surveys. As well, there is some suggestion that our study participants may be younger, on average (41.4 years), than the Canadian HCP population (mean age; nurses: 43–44 years [39]; physicians: 49 years [40]). Selection bias may also be present as younger participants were less likely to continue for the full follow-up period, thus reducing the number of K10s completed later in the study. As well, all results are self-reported and may suffer from social desirability bias; i.e., respondents may have been reluctant to endorse symptoms associated with emotional distress. As no pre-pandemic K10 measures are available and data were not collected for the first year of the pandemic, this study cannot provide information to compare with those periods. However, these limitations are somewhat mitigated by the fact that this was a pan-Canadian study with a large sample size that collected data for 32 consecutive months using a validated distress scale.
link