Abstract
Background:
Increased sick leave among educators can detrimentally impact studentsā productivity, and academic achievement. It remains unknown whether the number of sick days taken by educators in the preceding school year correlates with the prevalence or severity of psychological problems among educators in the subsequent school year.
Objective:
This study aimed to examine the number of self-reported sick days taken by educators in three Canadian provinces during the 2021/2022 academic year and its association with measures of stress, burnout, low resilience, depression, and anxiety during the 2022/2023 school year.
Methods:
Data was collected from educators in three Canadian provinces, Alberta, Nova Scotia, and Newfoundland and Labrador, from September 1, 2022, to August 30, 2023. The Maslach Burnout Inventory-Educator Survey (MBI-ES), the Brief Resilience Scale (BRS), and the Perceived Stress Scale were used to assess burnout, resilience, and stress, respectively. Likely Generalized Anxiety Disorder (GAD) and likely Major Depressive Disorder (MDD) were assessed using the Generalized Anxiety Disorder-7 and Patient Health Questionnaire-9 scales, respectively. Statistical analysis was conducted using SPSS version 28.
Results:
A total of 763 subscribers completed all the demographic, professional questions, and clinical scales, giving a response rate of 39.91%. Of these, there were 94 (12.3%) males and 669 (87.7%) females. Educators who reported taking 11 or more sick days in the previous academic year were at least three times more likely to exhibit high stress, emotional exhaustion, likely GAD, low resilience, and likely MDD than educators with no sick days during the preceding year. Similarly, educators with 11 or more sick days had significantly higher mean scores on the GAD-7 scale, the PHQ-9 scale, the PSS-10, the MBI Emotional Exhaustion subscale, and the MBI Depersonalization subscale than those with zero sick days.
Conclusion:
This study demonstrates a significant association between sick days and the prevalence and severity of high stress, low resilience, burnout, anxiety, and depression among educators. Short-term sick leave can escalate into long-term absences without adequate support for teachers. Governments and policymakers in the education sector must foster a supportive environment that enables teachers to thrive and effectively perform their professional role without taking prolonged sick days, which can undermine student learning and achievement.
1 Introduction
While teaching is inherently rewarding, it has unique challenges that can contribute to notable psychological health concerns. The responsibilities of educators are far-reaching, encompassing multifaceted roles that often extend beyond academics (). For example, as teachers perform their role, they engage with colleagues, parents, and caregivers of students, as well as students, both within and outside the classroom. Yet, despite their pivotal role in shaping future generations, teachers are often under-compensated compared to professionals in other fields, with significant disparities in salary distribution across different countries (). With a modest starting salary of $55,910, Canadian teachers are unsung heroes shaping the minds and character of the nationās future generations (). These comparatively low wages impact teachersā financial stability, job satisfaction, and morale. In 2021/2022, Canadaās average gross clinical payment per physician was $357,000, although most physicians are fee-for-service (independent contractors) (). With increasing work demands (such as larger class sizes, administrative responsibilities, expectations for standardized testing, etc.), this disparity between their dedicated efforts and rewards can exacerbate stress and burnout among teachers ().
Research indicates that teachers may experience higher rates of psychological disorders compared to other professions and are at higher risk of workplace stress, leading to burnout, depression, and other psychosomatic symptoms (). The stress experienced by teachers can be linked to three major overlapping problems: burnout, anxiety, and depression, all of which profoundly impact their health, well-being, effectiveness, or productivity (). The COVID pandemic exacerbated these challenges as educators transitioned to online instruction. Lack of adequate administrative support and resulting anxiety were significant predictors of teacher burnout during the COVID-19 pandemic (). A survey of mental health professionals reported that the majority characterized burnout as a state of exhaustion, which was viewed as a risk factor for developing a mental disorder (). Similarly, a study conducted during the second wave of the pandemic found high rates of depression, anxiety, and stress among teachers (67, 73, and 86%, respectively) ().
The demanding nature of teaching often takes a toll on educatorsā mental health, resulting in the need for sick leave to recuperate, which affects not only teachers but also carries significant social and economic costs for students, educational organizations, and society as a whole (). In the U.K., a survey by the Department of Education reported that 40,000 pre-retirement teachers resigned from state schools during the 2021/2022 academic year (). This survey also documented a notable increase in teacher absenteeism due to illness, with teachers accumulating over 3 million sick days during the 2021/2022 school year, marking a surge of more than 50% compared to pre-pandemic levels in 2018/2019 academic year (). Data from general practitioners also reported that 81% of mental health issues among the general workforce were associated with sickness absence (). Stress at work has also been linked to poor well-being and increased sickness leave due to depressive symptoms ().
Similarly, the increasing severity of depression has been associated with increased sick absences. The outcome of a study among workers in the U.S. reported that all levels of depression were linked to decreased work productivity, and absenteeism was significantly positively associated with the severity of depression (). While there was an uptick in sick days taken by civil servants in 2022 due to stress and other mental health issues, this increase was comparatively lower than that of teachers ().
Teachers reported an absence rate of 19%, with less experienced teachers exhibiting higher rates of absence (). Similarly, secondary school teachers have identified burnout (specifically emotional exhaustion) and job dissatisfaction as significant factors associated with psychiatric sick leave (). Further, elevated levels of emotional exhaustion were associated with long-term sickness absence, while low levels of professional efficacy (P.E.) did not show such a correlation (). In Canada, a survey conducted among members of the British Columbia Teachersā Federation in 2023 revealed heightened stress and workload among nearly two-thirds of the respondents, with approximately 40% reporting a decline in their physical or mental health compared to the previous year (). Another Canadian survey found that more than 25% of teacher respondents had taken a leave of absence, with stress being the primary reason cited (). This study also found that the stress teachers experience at school is frequently due to factors like increased workload and lack of support from the administration, which invariably adversely impact educatorsā home lives.
Teachers exhibited a 6% higher rate of sick leave compared to other professionals, such as physiotherapists and occupational therapists, resulting in a loss of 3.3% of available teacher time due to sickness absence (). While sickness absenteeism is commonly associated with physical illness (; ), its profound connection to mental well-being among teachers may often be overlooked. Again, limited Canadian research focuses on the association between educatorsā sick days and their psychological well-being. Further, what is driving the rise in sickness absences among teachers needs further exploration. Thus, the current study aims to evaluate the self-reported number of sick days taken by teachers during a school year and their association with the prevalence and severity of high stress, burnout, low resilience, anxiety, and depression during the subsequent school year.
This study aimed to achieve the following objectives:
Assess whether the number of sick days teachers take in Alberta, Nova Scotia, Newfoundland, and Labrador during the 2021/2022 school year predicts high stress levels among educators.
Examine whether the number of sick days taken by teachers in these provinces during the same period predicts the occurrence of burnout among educators.
Investigate whether the number of sick days teachers take in the three provinces during the same period predicts low resilience among educators.
Determine if the number of sick days teachers take in the three provinces during the same period predicts the likelihood of Generalized Anxiety Disorder (GAD) among educators.
Explore whether the number of sick days taken by teachers in the three provinces during the same period predicts the likelihood of Major Depressive Disorder (MDD) among educators.
1.1 Hypothesis
We hypothesize that the number of sick days taken by educators during the 2021/2022 academic year will be positively associated with the prevalence and severity of high stress, burnout, low resilience, and the likelihood of GAD and MDD in the subsequent school year.
2 Methodology
2.1 Study design
This is an exploratory study that utilized a cross-sectional survey design. Secondary analysis was performed on baseline survey data from the Wellness4Teachers program. Data were collected using web-based questionnaires distributed via a link to participantsā mobile phones. The questionnaires were administered using the University of Alberta REDCap platform, a secure web application designed to create and manage online surveys ().
2.2 Study settings
In 2024, Canadaās population reached 40,769,890 individuals (). Nova Scotia, Newfoundland and Labrador, and Alberta, where the study was conducted, are three of Canadaās 13 provinces and territories (). The Canadian Teachersā Federation is a national alliance of provincial and territorial member organizations representing over 232,000 teachers across the country ().
Nova Scotia, Newfoundland, and Labrador are both eastern provinces of Canada, with populations estimated at 1,066,416 and 540,418, respectively (). The Nova Scotia Teachers Union aims to improve the quality of public education and represents more than 10,000 public school teachers, community college faculty, professional support staff, and educators working in the Atlantic Provinces Special Education Authority (). In Newfoundland and Labrador, the English School District is the school board that oversees all English-language primary and secondary education, with approximately 63,000 students across 250 schools and more than 10,000 staff members (). There are 5,232 full-time equivalent teachers in Newfoundland and Labradorās 2023/2024 academic year (Government of Newfoundland and Labrador, 2024).
Alberta, located in Western Canada, has an estimated population of 4,756,408 and about 32,523 full and part-time teachers (; ). According to 2002/2023 data collected from the Canadian Teachersā Federation (CTF) survey, 72 percent of the members of teacher organizations in Canada are women, and 28 percent are men (). This shows some degree of gender imbalance in the teaching profession, with 63% of teachers being female in Alberta (). 84% of elementary school and kindergarten teachers in Nova Scotia are female (), and over two-thirds (68.7%) of teachers in 2007/08 were female in Newfoundland and Labrador ().
2.3 Data collection
Data collection commenced during the 2022/2023 academic year through the Wellness4Teachers program, which was initiated at the beginning of the academic year (). This program provides teachers with one-way (non-interactive) psychological intervention text messages based on cognitive behavioral therapy principles, delivered to their mobile phones upon signing up.
2.3.1 Participants
Teachers in Newfoundland and Labrador, Nova Scotia, and Alberta were invited to self-subscribe to the program (), facilitated by collaborative efforts with the Alberta Teachers Association, the Alberta Catholic School Boards, the Nova Scotia Teachers Union, and the Newfoundland and Labrador Teachers Union. Teachers subscribed by texting āTeachWellā to a designated phone number, which automatically registered them to receive supportive text messages (). Upon enrollment, participants were invited to complete a web-based survey via a link in the welcome text message. The baseline survey collected demographic information (age, sex, ethnicity, marital and housing status), professional details (the total number of sick days teachers took in the preceding year, years of teaching experience, class size, major role, area of specialization, and primary stressors), and clinical variables; stress, burnout, resilience, anxiety, and depression. Teachers responded to the question: āAs far as you can recall, in the 2021/2022 school year, how many sick days did you take off school?ā Typically, the survey took 10ā15āmin to complete. Completion of the baseline survey was entirely voluntary, irrespective of receiving daily supportive text messages.
2.4 Outcome measures
The primary outcome measures in this study are the measures of association between the number of sick days taken by the educators and the prevalence of high stress, burnout, low resilience, anxiety, and depression during the subsequent school year. The secondary outcome measures included measures of association between the number of sick days taken by the educators and the severity of high stress, burnout, low resilience, anxiety, and depression during the subsequent school year.
2.4.1 Instruments
The Maslach Burnout Inventory-Educator Survey (MBI-ES) (; ) was used to collect teacher burnout data. The MBI-ES is a 22-item instrument that measures emotional exhaustion, depersonalization, and personal accomplishment. For each of the three dimensions, emotional exhaustion scores ā„27 were deemed high, 17ā26 were deemed normal, and 0ā16 were deemed low. Similarly, for depersonalization scores, 13 and over was deemed high, 7ā12 was deemed normal, and 0ā6 was deemed low. Finally, 0ā31 was deemed high for personal accomplishment, 32ā38 was deemed normal, and 39 and over was deemed low. An analysis of the internal consistency of the MBI-ES questionnaire yielded a Cronbachās alpha of 0.785 for the full scale, which indicated scale reliability. For the emotional exhaustion subscale, the alpha coefficient was 0.93, suggesting excellent reliability, while the alpha coefficient for the depersonalization subscale was 0.618, suggesting questionable reliability, and for the personal accomplishment subscale, the alpha coefficient was 0.776, indicating satisfactory reliability ().
Resilience was measured using the Brief Resilience Scale (BRS; mean scores of 1.00ā2.99 suggest low resilience, 3.00ā4.30 suggest normal resilience, and 4.31ā5.00 suggest high resilience) (). The Perceived Stress Scale 10 (PSS-10; a score of ā„27 indicates likely high stress) was used to measure stress (). Again, a PSS-10 score of ā¤13 indicates low stress, scores of 14 to 26 indicate moderate stress, and scores of ā„27 indicate high stress. The BRS and the PSS-10 have Cronbachās α of 0.78 and 0.82, respectively, suggesting good internal reliability (; ).
Likely Generalized Anxiety Disorder (GAD) among participants was evaluated with the Generalized Anxiety Disorder-7 (GAD-7) scale (). This scale consists of seven self-reported items defining the symptoms of GAD. Participants with a GAD-7 scoreāā„ā10 were considered to have moderate to high anxiety or likely GAD symptoms, and those with GAD-7 scores <10 were considered to have low anxiety (GAD unlikely) (). Cronbachās alpha for the GAD-7 scale when used in primary care was 0.92, and the split-half reliability was 0.82, revealing a good level of reliability ().
Likely, Major Depressive Disorder (MDD) was measured with the Patient Health Questionnaire-9 (PHQ-9). PHQ-9 scoreāā„ā10 indicates moderate to severe depression or likely MDD (), whereas PHQ-9 score of <10 signifies at most mild depression (MDD unlikely). The Cronbachās α for PHQ-9 reported in one study was 0.74, demonstrating adequate internal consistency (). The MBI-ES, BRS and PSS-10, PHQ-9, and GAD-7 scales were studied as categorical variables for prevalence estimates and as means for severity estimates for the psychological conditions.
2.5 Statistical analysis
Data analysis was conducted using SPSS Version 28 (). The variable āSick daysā was categorized into five groups: āNo sick days,ā ā1ā5 sick days,ā ā6ā10 sick days,ā ā11ā15 sick days,ā and ā16 or more sick days.ā Seven separate multivariate binomial logistic regression analyses assessed the independent association between sick days and psychological problem variables (high stress, emotional exhaustion, depersonalization, lack of personal accomplishment, likely MDD, likely GAD, and low resilience).
Each regression model included demographic and work-related variables, including sick days, as predictors. Prior to regression analyses, correlational analysis was performed to eliminate strong intercorrelations (Spearmanās correlation coefficient of 0.7 to 1.0 or ā 0.7 to ā 1.0) among predictor variables. Odds ratios (OR) with two-tailed significance (p ā¤ā0.05) and confidence intervals were used to determine the effect of the āSick daysā variable on respondentsā self-reported psychological conditions while controlling for other variables in the model.
To compare the severity of the psychological conditions under study, a one-way analysis of variance (ANOVA) with two-tailed significance (p ā¤ā0.05) was used to evaluate differences in mean scores across the five categories of the āsick daysā variable. Post-hoc analysis was conducted using the HSD test. In cases where homogeneity of variance was violated, Welchās F-test and the Games-Howell post-hoc test were planned to be used.
3 Results
Table 1 summarizes the demographic, work-related, and professional characteristics of participants based on the number of sick days. Of the 1912 teachers who subscribed to the Wellness4Teachers program, 763 subscribers completed all the demographic, professional questions, and clinical scales, giving a response rate of 39.91%. Of those included in the study, a total of 535 (70.1%) live in Alberta, 127 (16.6%) live in Nova Scotia, and 101 (13.2%) live in Newfoundland and Labrador. Overall, there were 94 (12.3%) males and 669 (87.7%) females, and most were Caucasians (692, 90.7%). A total of 608 (79.7%) of participants taught in public schools and 140 (18.3%) in Catholic schools. Most participants, 290 (38.0%), had 20āyears or less but more than 10āyears of teaching experience, and 228 (29.9%) had more than 20āyears of teaching experience.
Table 1
| Variables | Zero sick days Nā=ā54 n (%) | 1ā5 sick days Nā=ā255 n (%) | 6ā10 sick days Nā=ā235 n (%) | 11ā15 sick days Nā=ā89 n (%) | 16 or more sick days Nā=ā130 n (%) | Chi2/Fisherās exact* test values | p-value | Total Nā=ā763 n (%) |
|---|---|---|---|---|---|---|---|---|
| Sociodemographic characteristics | ||||||||
| Age (years) | ||||||||
| 18ā25 | 4 (16.0%) | 10 (40.0%) | 8 (32.0%) | 2 (8.0%) | 1 (4.0%) | 11.09 | 0.52 | 25 (3.3%) |
| 26ā40 | 22 (7.6%) | 92 (31.9%) | 94 (32.6%) | 34 (11.8%) | 46 (16.0%) | 288 (27.7%) | ||
| 41ā60 | 27 (6.3%) | 144 (33.6%) | 125 (29.1%) | 52 (12.1%) | 81 (18.9%) | 429 (56.2%) | ||
| 61 and above | 1 (4.8%) | 9 (42.9%) | 8 (38.1%) | 1 (4.8%) | 2 (9.5%) | 21 (2.8%) | ||
| Sex at birth | ||||||||
| Male | 16 (17.0%) | 37 (39.4%) | 24 (25.5%) | 11 (11.7%) | 6 (6.4%) | 24.22 | <0.001 | 94 (12.3%) |
| Female | 38 (5.7%) | 218 (32.6%) | 211 (31.5%) | 78 (11.7%) | 124 (18.5%) | 669 (87.7%) | ||
| Relationship status | ||||||||
| Single | 15 (13.3%) | 42 (37.2%) | 31 (27.4%) | 12 (10.6%) | 13 (11.5%) | 23.6 | 0.10 | 113 (14.8%) |
| Married | 28 (5.7%) | 167 (33.9%) | 159 (32.3%) | 52 (10.6%) | 86 (17.5%) | 492 (64.5%) | ||
| Common- law or partnered | 7 (6.9%) | 33 (32.4%) | 31 (30.4%) | 15 (14.7%) | 16 (15.7%) | 102 (13.4%) | ||
| Separated or divorced | 3 (6.4%) | 12 (25.5%) | 13 (27.7%) | 7 (14.9%) | 12 (25.5%) | 47 (6.2%) | ||
| Other | 1 (11.1%) | 1 (11.1%) | 1 (11.1%) | 3 (33.3%) | 3 (33.3%) | 9 (1.2%) | ||
| Number of children | ||||||||
| No child | 23 (9.7%) | 85 (35.9%) | 74 (31.2%) | 25 (10.5%) | 30 (12.7%) | 12.47 | 0.71 | 237 (31.1%) |
| One child | 6 (5.3%) | 38 (33.6%) | 31 (27.4%) | 16 (14.2%) | 22 (19.5%) | 113 (14.8%) | ||
| Two children | 16 (5.4%) | 92 (31.2%) | 92 (31.2%) | 36 (12.2%) | 59 (20.0%) | 295 (38.7%) | ||
| Three children | 5 (6.0%) | 27 (32.5%) | 27 (32.5%) | 8 (9.6%) | 14 (16.9%) | 83 (10.9%) | ||
| Four or more children | 4 (11.4%) | 11 (31.4%) | 11 (31.4%) | 4 (11.4%) | 5 (14.3%) | 35 (4.6%) | ||
| Ethnicity | ||||||||
| Indigenous | 1 (5.9%) | 8 (47.1%) | 2 (11.8%) | 3 (17.6%) | 3 (17.6%) | 0.06 | 17 (2.2%) | |
| African descendants | 1 (16.7%) | 2 (33.3%) | 2 (33.3%) | 1 (16.7%) | 0 (0.0%) | 6 (0.8%) | ||
| East Asian | 3 (25%) | 4 (33.3%) | 2 (16.7%) | 0 (0.0%) | 3 (25.0%) | 40.23 | 12 (1.6%) | |
| Latino | 2 (22.2%) | 3 (33.3%) | 4 (44.4%) | 0 (0.0%) | 0 (0.0%) | 9 (1.2%) | ||
| Middle Eastern | 2 (40.0%) | 1 (20.0%) | 0 (0.0%) | 0 (0.0%) | 2 (40.0%) | 5 (0.7%) | ||
| South Asian | 0 (0.0%) | 6 (75.0%) | 1 (12.5%) | 0 (0.0%) | 1 (12.5%) | 8 (1.0%) | ||
| Caucasian (European descent) | 44 (6.4%) | 228 (32.9%) | 218 (31.5%) | 84 (12.1%) | 118 (17.1%) | 692 (90.7%) | ||
| Other ethnicities | 1 (7.1%) | 3 (31.4%) | 6 (42.9%) | 1 (7.1%) | 3 (21.4%) | 14 (1.8%) | ||
| Housing status | ||||||||
| Own home | 39 (6.1%) | 211 (33.2%) | 202 (31.8%) | 76 (11.9%) | 108 (17.0%) | 8.73 | 0.34 | 636 (83.4%) |
| Rented accommodation | 12 (11.3%) | 35 (33.0%) | 29 (27.4%) | 10 (9.4%) | 20 (18.9%) | 106 (13.9%) | ||
| Live with family or friends | 3 (14.3%) | 9 (42.9%) | 4 (19.0%) | 3 (14.3%) | 2 (9.5%) | 21 (2.8%) | ||
| Province of residence | ||||||||
| Alberta | 42 (7.9%) | 202 (37.8%) | 151 (28.2%) | 60 (11.2%) | 80 (15.0%) | 22.42 | 0.004 | 535 (70.1%) |
| Nova Scotia | 5 (3.9%) | 32 (25.2%) | 49 (38.6%) | 14 (11.0%) | 27 (21.3%) | 127 (16.6%) | ||
| Newfoundland and Labrador | 7 (6.9%) | 21 (20.8%) | 35 (34.7%) | 15 (14.9%) | 23 (22.8%) | 101 (13.2%) | ||
| Work-related variables | ||||||||
| Location of school | ||||||||
| Rural setting | 14 (4.7%) | 95 (32.2%) | 103 (34.9%) | 31 (10.5%) | 52 (17.6%) | 7.2% | 0.13 | 295 (38.7%) |
| Urban setting | 40 (8.5%) | 160 (34.2%) | 132 (28.2%) | 58 (12.4%) | 78 (16.7%) | 468 (61.3%) | ||
| Area of teaching specialization | ||||||||
| English | 9 (7.1%) | 45 (35.4%) | 37 (29.1%) | 15 (11.8%) | 21 (16.5%) | 0.000* | 0.00* | 127 (16.6%) |
| Mathematics | 9 (16.7%) | 18 (32.7%) | 18 (32.7%) | 3 (5.5%) | 7 (12.7%) | 55 (7.2%) | ||
| Sciences (physics, chemistry, biology) | 8 (15.4%) | 21 (40.4%) | 9 (17.3%) | 4 (7.7%) | 10 (19.2%) | 52 (6.8%) | ||
| Arts (history, geography, social studies, etc.) | 5 (7.1%) | 31 (44.3%) | 18 (25.7%) | 10 (14.3%) | 6 (8.6%) | 70 (9.2%) | ||
| Music | 0 (0.0%) | 5 (21.7%) | 9 (39.1%) | 3 (13.0%) | 6 (26.1%) | 23 (3.0%) | ||
| Physical education | 3 (15.0%) | 6 (30.0%) | 4 (20.0%) | 5 (25.0%) | 2 (10.0%) | 20 (2.6%) | ||
| Other | 20 (4.8%) | 129 (31.0%) | 140 (33.7%) | 49 (11.8%) | 78 (18.8%) | 416 (54.5%) | ||
| Teach only in your area of specialization | ||||||||
| No | 34 (7.8%) | 149 (34.0%) | 132 (30.1%) | 50 (11.4%) | 73 (16.7%) | 1.08 | 0.89 | 438 (57.4%) |
| Yes | 20 (6.2%) | 106 (32.6%) | 103 (31.7%) | 39 (12.0%) | 57 (17.5%) | 325 (42.6%) | ||
| Number of years teaching | ||||||||
| 5āyears or less | 15 (14.0%) | 41 (38.3%) | 34 (9.3%) | 10 (9.3%) | 7 (6.5%) | 30.54 | 0.002 | 107 (14.0%) |
| 10āyears or less but more than 5āyears | 10 (7.2%) | 38 (27.5%) | 48 (34.8%) | 23 (16.7%) | 19 (13.8%) | 138 (18.1%) | ||
| 20āyears or less but more than 10āyears | 11 (3.8%) | 103 (35.5%) | 89 (30.7%) | 31 (10.7%) | 56 (19.3%) | 290 (38.0%) | ||
| More than 20āyears. | 18 (7.9%) | 73 (32.0%) | 64 (28.1%) | 25 (11.0%) | 48 (21.1%) | 228 (29.9%) | ||
| Class size taught | ||||||||
| 20 or less | 11 (7.0%) | 52 (32.9) | 53 (33.5%) | 16 (10.1%) | 26 (16.5%) | 2.66 | 0.95 | 158 (20.7%) |
| 21ā27 | 28 (7.0%) | 129 (32.3%) | 124 (31.1%) | 46 (11.5%) | 72 (18.0%) | 399 (52.3%) | ||
| 28 or more | 15 (7.3%) | 74 (35.9%) | 58 (28.2%) | 27 (13.1%) | 32 (15.5%) | 206 (27.0%) | ||
| School institution (type) | ||||||||
| Catholic school | 16 (11.4%) | 51 (36.4%) | 37 (26.4%) | 16 (11.4%) | 20 (14.3%) | 14.32 | 0.07 | 140 (18.3%) |
| Public school | 35 (5.8%) | 199 (32.7%) | 192 (31.6%) | 73 (12.0%) | 109 (17.9%) | 608 (79.7%) | ||
| Other | 3 (20.0%) | 5 (33.3%) | 6 (40.0%) | 0 (0.0%) | 1 (6.7%) | 15 (2.0%) | ||
| Major role | ||||||||
| Elementary school teacher | 14 (4.1%) | 102 (30.2%) | 120 (35.5%) | 34 (10.1%) | 68 (20.1%) | 34.79 | 0.00 | 338 (44.3%) |
| Junior high school teacher | 13 (9.6%) | 42 (30.9%) | 36 (26.5%) | 18 (13.2%) | 27 (19.9%) | 136 (17.8%) | ||
| Senior high school teacher | 10 (8.8%) | 43 (38.1%) | 31 (27.4%) | 15 (13.3%) | 14 (12.4%) | 113 (14.8%) | ||
| Support staff | 1 (3.1%) | 11 (34.4%) | 8 (25.0%) | 7 (21.9%) | 5 (15.6%) | 32 (4.2%) | ||
| Administrator | 8 (11.9%) | 32 (47.8%) | 18 (26.9%) | 5 (7.5%) | 4 (6.0%) | 67 (8.8%) | ||
| Other | 8 (10.4%) | 25 (32.5%) | 22 (28.6%) | 10 (13.0%) | 12 (15.6%) | 77 (10.1%) | ||
| Source of stress | ||||||||
| Workload | 28 (6.6%) | 141 (33.4%) | 130 (30.8%) | 57 (13.5%) | 66 (15.6%) | 13.74 | 0.62 | 422 (55.3%) |
| Student behavior | 13 (8.1%) | 53 (33.1%) | 47 (29.4%) | 18 (11.3%) | 29 (18.1%) | 160 (21.0%) | ||
| Class size | 4 (9.5%) | 10 (23.8%) | 16 (38.1%) | 4 (9.5%) | 8 (19.0%) | 42 (5.5%) | ||
| Lack of support from the school administration | 4 (6.8%) | 17 (28.8%) | 18 (30.5%) | 8 (13.6%) | 12 (20.3%) | 59 (7.7%) | ||
| Other | 5 (6.3%) | 34 (42.5%) | 24 (30.0%) | 2 (2.5%) | 15 (18.8%) | 80 (10.5%) | ||
| Psychological variables | ||||||||
| High stress | ||||||||
| No | 46 (8.2%) | 200 (35.7%) | 174 (31.1%) | 56 (10.0%) | 80 (15.0%) | 17.34 | <0.001 | 560 (73.4%) |
| Yes | 200 (35.7%) | 55 (27.1%) | 61 (30.0%) | 33 (16.3%) | 46 (22.7%) | 203 (26.6%) | ||
| Low resilience | ||||||||
| No | 33 (7.3%) | 172 (38.2%) | 137 (30.4%) | 53 (11.8%) | 5 (12.2%) | 24.17 | <0.001 | 450 (59.9%) |
| Yes | 19 (6.3%) | 79 (26.2%) | 95 (31.6%) | 34 (11.3%) | 74 (24.6%) | 301 (40.1%) | ||
| Emotional exhaustion | ||||||||
| No | 20 (11.4%) | 74 (42.0%) | 45 (25.6%) | 14 (8.0%) | 23 (13.1%) | 17.88 | 0.001 | 176 (23.1%) |
| Yes | 34 (5.8%) | 181 (30.8%) | 190 (32.4%) | 75 (12.8%) | 107 (18.2%) | 587 (76.9%) | ||
| Depersonalization | ||||||||
| No | 43 (7.4%) | 206 (35.2%) | 181 (30.9%) | 61 (10.4%) | 94 (16.1%) | 7.37 | 0.12 | 585 (76.7%) |
| Yes | 11 (6.2%) | 47 (29.5%) | 54 (30.3%) | 28 (15.7%) | 36 (20.2%) | 178 (23.3%) | ||
| Lack of professional accomplishment | ||||||||
| No | 40 (7.6%) | 182 (34.5%) | 155 (29.4%) | 60 (11.4%) | 90 (17.1%) | 2.45 | 0.65 | 527 (69.1%) |
| Yes | 14 (5.9%) | 73 (30.9%) | 80 (33.9%) | 29 (12.3%) | 40 (16.9%) | 236 (30.9%) | ||
| Likely GAD | ||||||||
| No | 31 (7.6%) | 158 (38.5%) | 124 (30.2%) | 37 (9.0%) | 60 (14.6%) | 15.5 | 0.004 | 410 (54.0%) |
| Yes | 23 (6.6%) | 95 (27.2%) | 111 (31.8%) | 50 (14.3%) | 70 (20.1%) | 349 (46.0%) | ||
| Likely MDD | ||||||||
| No | 28 (8.3%) | 141 (41.7%) | 98 (29.0%) | 31 (9.2%) | 40 (11.8%) | 27.26 | <0.001 | 338 (44.3%) |
| Yes | 26 (6.1%) | 114 (26.8%) | 137 (32.2%) | 58 (13.6%) | 90 (21.2%) | 425 (55.7%) | ||
Frequency distribution of demographic, work-related, and professional variables based on participantsā number of sick days absences.
3.1 Multivariate binary logistic regression analyses
Since there were no strong correlations between the variables under study, all 15 sociodemographic and work-related variables in Table 1 and the āSick daysā variable were included in each of the seven multivariate binary regression models predicting the psychological conditions. Results are outlined in Table 2.
The model predicting the High Stress was statistically significant; Χ2(dfā=ā51; nā=ā763)ā=ā96.87, pā<ā0.00, explained between 11.9% (Cox and Snell R2) and 17.4% (Nagelkerke R2) of the variance; and correctly classified 74.7% of the cases.
The model predicting Emotional Exhaustion was statistically significant; Χ2(dfā=ā51; nā=ā763)ā=ā102.42, pā<ā0.00, explained between 12.6% (Cox and Snell R2) and 19.0% (Nagelkerke R2) of the variance; and correctly classified 79.0% of the cases.
The model predicting depersonalization was statistically significant; Χ2(dfā=ā51; nā=ā763)ā=ā129.53, pā<ā0.00, explained between 15.6% (Cox and Snell R2) and 23.6% (Nagelkerke R2) of the variance; and correctly classified 77.7% of the cases.
The model predicting lack of personal accomplishment was statistically significant; Χ2(dfā=ā51; nā=ā763)ā=ā60.35, pā<ā0.00, explained between 7.6% (Cox and Snell R2) and 10.7% (Nagelkerke R2) of the variance; and correctly classified 71.7% of the cases.
The model predicting low resilience was statistically significant; Χ2(dfā=ā51; nā=ā751)ā=ā119.03, pā<ā0.00, explained between 14.7% (Cox and Snell R2) and 19.8% (Nagelkerke R2) of the variance; and correctly classified 67.6% of the cases.
The model predicting the likely GAD was statistically significant; Χ2(dfā=ā51; nā=ā759)ā=ā117.00, pā<ā0.00, explained between 14.3% (Cox and Snell R2) and 19.1% (Nagelkerke R2) of the variance; and correctly classified 67.1% of the cases.
The model predicting the likely MDD was statistically significant; Χ2(dfā=ā51; nā=ā763)ā=ā97.40, pā<ā0.00, explained between 12.0% (Cox and Snell R2) and 16.0% (Nagelkerke R2) of the variance, and correctly classified 64.9% of the cases.
Table 2
| Characteristics | High stress | Emotional exhaustion | Depersonalization | Lack of personal accomplishment | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Sick days variable | ||||||||||||
| Zero sick days | 0.005 | 0.001 | 0.01 | 0.408 | ||||||||
| 1ā5 sick days | 1.54 | 0.64ā3.68 | 0.333 | 1.65 | 0.80ā3.37 | 0.173 | 1.25 | 0.54ā2.90 | 0.61 | 1.34 | 0.64ā2.82 | 0.438 |
| 6ā10 sick days | 1.88 | 0.78ā4.53 | 0.162 | 3.16* | 1.48ā6.75 | 0.003 | 2.18* | 0.92ā5.16 | 0.08 | 1.85 | 0.87ā3.93 | 0.110 |
| 11ā15 sick days | 3.47* | 1.34ā8.96 | 0.010 | 3.96* | 1.60ā9.84 | 0.003 | 3.00* | 1.17ā7.69 | 0.02 | 1.73 | 0.74ā4.03 | 0.204 |
| 16 or more sick days | 3.22* | 1.29ā8.03 | 0.012 | 3.44* | 1.47ā8.05 | 0.004 | 2.69* | 1.08ā6.69 | 0.03 | 1.63 | 0.73ā3.66 | 0.237 |
Summary from four multivariate regression models outputs showing the āsick daysā variable as a predictor of high stress, emotional exhaustion, depersonalization, and lack of personal accomplishment.
OR, Odds ratio; C.I., confidence interval. *Significance at pāā¤ā0.05.
Tables 2, 3 show the results of the regression analyses, highlighting the variable of interest, āSick days,ā across the seven regression models predicting high stress, emotional exhaustion, depersonalization, and lack of personal accomplishment, as well as likely MDD, likely GAD, and low resilience.
Table 3
| Characteristics | Low resilience | Likely GAD | Likely MDD | ||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Sick days as a variable | |||||||||
| Zero sick days | <0.001 | 0.001 | <0.001 | ||||||
| 1ā5 sick days | 0.89 | 0.434ā1.81 | 0.737 | 0.891 | 0.45ā1.76 | 0.74 | 0.96 | 0.500ā1.86 | 0.912 |
| 6ā10 sick days | 1.34 | 0.647ā2.76 | 0.434 | 1.39 | 0.70ā2.79 | 0.35 | 1.79 | 0.913ā3.50 | 0.090 |
| 11ā15 sick days | 1.31 | 0.579ā2.95 | 0.519 | 2.36* | 1.08ā5.16 | 0.03 | 2.76* | 1.276ā5.98 | 0.010 |
| 16 or more sick days | 2.88* | 1.319ā6.27 | 0.008 | 2.01 | 0.95ā4.25 | 0.07 | 3.35* | 1.591ā7.05 | 0.001 |
Summary from three multivariate regression model outputs showing the āsick daysā variable as a predictor of low resilience, likely GAD, and likely MDD.
OR, Odds ratio; C.I, confidence interval; *Significance at pāā¤ā0.05.
Table 2 indicates that, after controlling for all demographic and work-related variables, educators who experienced 11ā15 sick days and those with 16 or more sick days were approximately three and a half times and three times more likely to present with high stress, respectively, compared to those with no sick days in the preceding academic year. Similarly, educators who had 11ā15 sick days and those with 16 or more sick days were four times and three and half times more likely to display emotional exhaustion, respectively, compared to those with no sick days in the preceding academic year. Also, educators with 6ā10 sick days, 11ā15 sick days, and 16 or more sick days were about two times, three times, and two and a half times more likely to present with depersonalization, respectively, compared with educators who had no sick days in the preceding year. No significant differences were observed between educators with no sick days and those who had sick days in the previous year regarding lack of personal accomplishment when controlling for all sociodemographic and work-related variables.
The results in Table 3 suggest that, even after controlling for all demographic and work-related variables, teachers with 16 or more sick days in the preceding academic year were roughly three times more likely to present with low resilience compared with teachers who had no sick days during the same period. Similarly, teachers who reported 11ā15 sick days in the preceding academic year were three times more likely to display symptoms indicative of likely GAD than those with no sick days. Additionally, teachers with 11ā15 sick days and those with 16 or more sick days in the previous academic year were about three times more likely to present with symptoms suggestive of likely MDD than teachers with no sick days during the same period.
3.2 Analysis of variance
As statistical analysis revealed, there was no violation of the homogeneity of variance for the mean scores across all the psychological conditions under study, so the F test and HSD test for post hoc comparisons were used in the ANOVA. The results of both the ANOVA and post hoc tests, which compare the mean differences in scores across the various scales for educators who had zero sick days and those with varying sick days, are summarized in Tables 4, 5, respectively.
Table 4
| Dependent variable | N | Mean | SD | F | p-value | |
|---|---|---|---|---|---|---|
| GAD-7 score | Zero sick days | 54 | 8.76 | 5.66 | 6.825 | <0.001 |
| 1ā5ādays | 253 | 8.53 | 5.51 | |||
| 6ā10ādays | 235 | 9.37 | 5.35 | |||
| 11ā15ādays | 87 | 11.17 | 6.00 | |||
| 16 or more days | 130 | 11.08 | 5.51 | |||
| Total | 759 | 9.54 | 5.61 | |||
| PHQāāā9 score | Zero sick days | 54 | 10.28 | 6.07 | 7.977 | <0.001 |
| 1ā5ādays | 255 | 9.71 | 5.64 | |||
| 6ā10ādays | 235 | 11.21 | 5.73 | |||
| 11ā15ādays | 89 | 12.52 | 6.12 | |||
| 16 or more days | 130 | 12.80 | 6.00 | |||
| Total | 763 | 11.07 | 5.93 | |||
| BRS score | Zero sick days | 52 | 3.16 | 0.83 | 9.295 | <0.001 |
| 1ā5ādays | 251 | 3.32 | 0.87 | |||
| 6ā10ādays | 232 | 3.15 | 0.91 | |||
| 11ā15ādays | 87 | 3.13 | 0.94 | |||
| 16 or more days | 129 | 2.74 | 0.85 | |||
| Total | 751 | 3.13 | 0.90 | |||
| MBI-ES emotional exhaustion subscale score | Zero sick days | 54 | 30.59 | 12.61 | 8.140 | <0.001 |
| 1ā5ādays | 255 | 32.29 | 10.52 | |||
| 6ā10ādays | 235 | 34.55 | 10.19 | |||
| 11ā15ādays | 89 | 37.14 | 10.58 | |||
| 16 or more days | 130 | 37.32 | 10.85 | |||
| Total | 763 | 34.29 | 10.84 | |||
| MBI-ES depersonalization subscale score | Zero sick days | 54 | 6.96 | 5.71 | 2.542 | 0.039 |
| 1ā5ādays | 255 | 7.62 | 6.167 | |||
| 6ā10ādays | 235 | 7.80 | 5.68 | |||
| 11ā15ādays | 89 | 9.38 | 6.57 | |||
| 16 or more days | 130 | 8.92 | 6.61 | |||
| Total | 763 | 8.06 | 6.14 | |||
| MBI-ES personal accomplishment subscale score | Zero sick days | 54 | 35.41 | 8.62 | 2.569 | 0.037 |
| 1ā5ādays | 255 | 35.52 | 6.94 | |||
| 6ā10ādays | 235 | 34.00 | 7.52 | |||
| 11ā15ādays | 89 | 33.55 | 7.08 | |||
| 16 or more days | 130 | 33.61 | 7.36 | |||
| Total | 763 | 34.49 | 7.37 | |||
| PSS-10 | Zero sick days | 54 | 21.04 | 5.90 | 6.66 | <0.001 |
| 1ā5ādays | 255 | 21.24 | 6.09 | |||
| 6ā10ādays | 235 | 22.50 | 6.055 | |||
| 11ā15ādays | 89 | 24.11 | 6.11 | |||
| 16 or more days | 130 | 23.78 | 5.55 | |||
| Total | 763 | 22.38 | 6.07 | |||
ANOVA output comparing the mean scores on standardized rating scales for psychological problems among educators against the number of sick days taken in the preceding school year.
MBI-ES, Maslach Burnout Inventory for Educators; BRS, Brief Resilience Scale; Std, Standard Deviation; *Significance at pāā¤ā0.05. PSS-10, Perceived Stress Scale; GADā7, Generalized Anxiety Disorder 7 Scale; PHQ-9, Patient Health Questionnaire 9 scale.
Table 5
| Dependent variable | (I) Sick days cat | (J) Sick days cat | Mean difference (I-J) | Sig. | 95% CI | |
|---|---|---|---|---|---|---|
| Lower bound | Upper bound | |||||
| GAD-7 scale scores | Zero sick days | 1ā5ādays | 0.23 | 0.78 | ā1.39 | 1.86 |
| 6ā10ādays | ā0.61 | 0.47 | ā2.24 | 1.03 | ||
| 11ā15ādays | ā2.40* | 0.01 | ā4.28 | ā0.52 | ||
| 16 or more days | ā2.32* | 0.01 | ā4.08 | ā0.56 | ||
| PHQ-9 scale scores | Zero sick days | 1ā5ādays | 0.56 | 0.52 | ā1.15 | 2.28 |
| 6ā10ādays | ā0.93 | 0.29 | ā2.66 | 0.79 | ||
| 11ā15ādays | ā2.24* | 0.03 | ā4.21 | ā0.27 | ||
| 16 or more days | ā2.52* | 0.008 | ā4.37 | ā0.67 | ||
| BRS scores | Zero sick days | 1ā5ādays | ā0.155 | 0.25 | ā0.42 | 0.11 |
| 6ā10ādays | 0.02 | 0.89 | ā0.25 | 0.28 | ||
| 11ā15ādays | 0.04 | 0.81 | ā0.27 | 0.34 | ||
| 16 or more days | 0.43* | 0.003 | 0.14 | 0.71 | ||
| MBI-ES emotional exhaustion subscale scores | Zero sick days | 1ā5ādays | ā1.69 | 0.29 | ā4.82 | 1.44 |
| 6ā10ādays | ā3.96* | 0.01 | ā7.11 | ā0.80 | ||
| 11ā15ādays | ā6.54* | <0.001 | ā10.15 | ā2.94 | ||
| 16 or more days | ā6.72* | <0.001 | ā10.11 | ā3.34 | ||
| MBI-ES depersonalization subscale scores | Zero sick days | 1ā5ādays | ā0.66 | 0.47 | ā2.46 | 1.14 |
| 6ā10ādays | ā0.84 | 0.36 | ā2.65 | 0.97 | ||
| 11ā15ādays | ā2.42* | 0.02 | ā4.49 | ā0.35 | ||
| 16 or more days | ā1.95* | 0.05 | ā3.90 | ā0.01 | ||
| MBI-ES personal accomplishment subscale scores | Zero sick days | 1ā5ādays | ā0.11 | 0.92 | ā2.27 | 2.04 |
| 6ā10ādays | 1.41 | 0.20 | ā0.77 | 3.58 | ||
| 11ā15ādays | 1.86 | 0.14 | ā0.63 | 4.34 | ||
| 16 or more days | 1.80 | 0.13 | ā0.5316 | 4.1310 | ||
| PSS-10 scores | Zero sick days | 1ā5ādays | ā0.20 | 0.82 | ā1.9613 | 1.5570 |
| 6ā10ādays | ā1.46 | 0.11 | ā3.2330 | 0.3113 | ||
| 11ā15ādays | ā3.08* | 0.003 | ā5.1010 | ā1.0497 | ||
| 16 or more days | ā2.74* | 0.005 | ā4.6411 | ā0.8387 | ||
LSD post hoc comparison of the mean difference in scores of various scales for educators who had zero sick days and those who had various sick days.
*Statistically significant mean difference.
Table 4 indicates statistically significant differences in the mean scores across all scales among respondents based on the number of sick days taken in the preceding academic year. Table 5 summarizes the post hoc test outcomes, which compare mean scores on the various scales between educators who had no sick days in the previous academic year and those who had sick days.
Table 4 suggests statistically significant differences in the mean scores for all the scales for respondents based on the number of sick days they had taken in the preceding academic year. Table 5 summarizes the post hoc test results that compare the mean scores on various scales for educators who had zero sick days in the preceding academic year and those who had various sick days.
Table 5 indicates that educators who took 11 or more sick days had significantly higher mean scores on the GAD-7 scale, the PHQ-9 scale, the PSS-10, and the MBI emotional exhaustion and depersonalization subscales than those with zero sick days. In addition, those who had 6ā10 sick days showed higher mean scores on the MBI emotional exhaustion subscale than those with zero sick days. Conversely, educators who took 16 or more sick days in the preceding academic year had lower mean resilience scores, as measured by the BRS, than those with zero sick days.
4 Discussion
The current study assessed the number of sick days teachers took and their association with psychological issues. Our results suggest that participants who took 11 or more sick days in the preceding academic year were at least three times more likely to report high stress than those who took no sick days. The teaching profession has been associated with a higher risk of workplace stress, which can lead to burnout, anxiety, and depression (). Thus, this study demonstrates that teachers who took 11 or more sick days are more likely to present with high stress in the following school year. These findings complement existing research, which suggests that stress is a major factor associated with teachers taking a leave of absence from work (). A comparable study also reported a higher odds ratio for sickness absence associated with stress-related disorders ().
Similarly, educators who had 11 or more sick days in the preceding academic year were at least three and a half times more likely to present with emotional exhaustion than educators with no sick days. This finding is corroborated by a study among Spanish teachers, which identified burnout (emotional exhaustion) as a significant factor associated with psychiatric sick leave among teachers (). Further, a recent study among educators in Canada reported a relatively higher prevalence of emotional exhaustion (). Our study found no significant differences in the sense of personal accomplishment between educators with no sick days and those with sick days in the preceding year. A possible explanation is that teachers who persistently feel inadequate are more likely to leave the profession rather than return after a sickness absence. Lack of accomplishment is characterized by reduced productivity or capability, low morale, and an inability to cope (). Parallel findings were found among Brazilian teachers, where high levels of emotional exhaustion were linked to long-term sickness absence, but low professional efficacy levels were not (). Additionally, research among Danish human service workers found that long-term sickness absence was predicted by burnout and psychosocial factors at work ().
The current study also found that educators with six or more sick days in the preceding academic year were at least twice as likely to present with depersonalization compared to those without sick days. Teaching, as a human service occupation, relies heavily on interpersonal interactions. Hence, feelings of cynicism and detachment from the job can severely hinder teachersā ability to perform their duties, leading to more frequent sick absences (). Given the high tendency of burnout (depersonalization) rate among teachers who had six or more sick days in the previous school year, this study indicates that this cohort of teachers is at an increased risk of taking additional sick days in the current school year. Furthermore, teaching entails specific skills and training, making it challenging to replace experienced educators (). Therefore, it is vital for these teachers to be supported to reduce the tendency for increased sick leaves.
The current study found that teachers who had 11ā15 sick days in the preceding academic year were three times more likely to present with likely GAD compared to those who had no sick days. A report by general practitioners on cases of work-related ill-health found that 81% of psychological issues resulted in sickness absence (). Sick days may also serve as a means of escaping work-related anxiety for workers who struggle with participation in work, resulting in likely long-term sick leave (). A regression analysis in another study identified anxiety as the sole predictor for the length of sickness absence (). Additionally, research has shown that job-related anxiety significantly correlates with the duration of sick leave (). Thus, teachers who took 11ā15 sick days in the preceding school year may be at an increased risk of taking additional sick days in the current school year due to the higher prevalence of anxiety in this group.
Similarly, teachers with 11 or more sick days in the preceding academic year were about three times more likely to present with likely MDD than those without sick days. Consistent with this finding, a study reported moderate evidence linking more severe depressive disorders with a history of previous sick leave (). A systematic review and meta-analysis also found that depressive symptoms were associated with sick leave (). Additionally, a Dutch database showed that burnout, anxiety, and depression disorders are associated with longer sick leave durations, particularly among educators (). Similar findings were reported by a U.K. study, which showed an association between sickness absence and depressive symptoms (). Given the high prevalence of likely MDD among teachers with 11 or more sick days in the previous school year, this study indicates that these teachers may be at an increased risk of taking additional sick days in the current school year.
With respect to resilience, this study found that teachers who had 16 or more sick days in the preceding academic year were about three times more likely to present with low resilience compared to those with no sick days. Teachers with low resilience are more susceptible to other psychological problems, such as stress and burnout, which can lead to breakdowns and increased sick leave. Research has shown that a growing number of teachers are experiencing higher stress and workload, negatively impacting their physical and mental well-being (). Further, a recent study in Canada found that educators with low resilience were 3.26 times more likely to experience symptoms of emotional exhaustion (), which has also been implicated in mental health-related sickness absence among Spanish school teachers (). Thus, it can be inferred that teachers who had 16 or more sick days in the preceding school year may be at an increased risk of taking additional sick days in the current school year due to their low resilience levels.
The results of this study suggest that educators who had 11 or more sick days had significantly higher mean scores on the PSS-10, the MBI Emotional Exhaustion subscale, and the MBI Depersonalization subscale, the GAD-7 scale, and the PHQ-9 scale compared with those who had no sick days. In addition, those who took 6ā10 sick days had a higher mean score on the MBI Emotional Exhaustion subscale than those with no sick days. Emotional exhaustion is believed to develop first in response to high job demands and is predictive of stress-related health outcomes such as weariness, loss of energy, depletion, debilitation, and fatigue (). These results are expected, as teachers who take more sick days are likely to have increased vulnerability to developing or already having established psychological issues. Existing literature also indicates that anxiety, depression, emotional exhaustion, and depersonalization are risk factors for increased sick leave among educators (; ). Thus, the findings from this study support the proposition that there is an increased likelihood that educators who took six or more sick days in the previous school year are more likely to take additional sick days in the current school year due to ongoing mental health problems.
Resilience emerges from a dynamic interplay between individual risk and protective factors, enabling teachers to persevere and thrive in the face of challenges (). Consistent with this, our study shows that educators with 16 or more sick days in the preceding academic year had lower mean scores on the Brief Resilience Scale (BRS) than those with zero sick days. Another study found an inverse relationship between resilience and depressive symptoms, where higher resilience correlated with lower Beck Depression Inventory scores, indicating reduced burden from depressive symptoms (). Hence, our study findings also support the proposition that there may be an increased likelihood that educators who took more sick days in the preceding school year will take additional sick days in the current school year due to low resilience and vulnerability to ongoing mental health problems.
4.1 Study limitations
First, although standardized, the scales used in assessing educatorsā psychological issues are not meant to be diagnostic. Thus, the findings may not accurately reflect the prevalence of diagnosed psychological disorders among educators but provide an indication of psychological distress among the educators. Future research may explore incorporating structured clinical assessments conducted by qualified professionals to provide a more accurate and comprehensive understanding of the mental health challenges among educators and their association with sick days. Second, the study was undertaken in three provinces of Canada, and therefore, the findings may not be generalizable to all educators in Canada. The socio-economic contexts can vary significantly across provinces, and these differences could influence the psychological well-being of educators. Future studies should also aim to conduct similar research in other provinces of Canada, aside from the provincesā understudy, to determine if similar outcomes would be established. A significant proportion (over 80%) of the study participants were female. Although there is generally gender imbalance in the teaching profession, with females disproportionately represented, this percentage may not be reflective across the three provinces. Thus, the demographic variables may not reflect the demographics of teachers in the three provinces under study, and therefore, the findings may not be generalizable. Third, the survey questions about educatorsā sick days were not validated and relied on self-reported data, which may not accurately capture the actual sick days taken during the academic year. Notwithstanding these limitations, this study provides valuable insights into the relationship between sick days and psychological issues among educators, which would be of interest to policymakers and could inform policies in the field of education. This study also adds to the limited Canadian literature that explores the relationship between teacher sick days and mental health.
5 Conclusion, implications for policy, and future research
This study has demonstrated a clear link between the number of sick days taken by teachers and the prevalence and severity of stress, certain aspects of burnout, low resilience, anxiety, and depression. Existing research also indicates that these psychological issues are associated with increased sick days. Therefore, it is reasonable to suggest that teachers who reported sick days in the preceding school year are at an increased risk of taking additional sick days in the current school year due to the increased psychological burden among these educators. In order to enhance the reliability of self-reported data, future Canada wide studies should use mixed qualitative and quantitative methods, including structured clinical interviews to aid diagnostic accuracy and focus group discussions which could provide better insights into the association between psychological problems and sick days in educators.
Clearly, this inference needs to be validated through further research. Despite the need for additional studies, this research suggests that teachers who take sick days, regardless of the reason, should be offered interventions addressing psychological issues as a preventative measure to reduce the likelihood of further sick leave. Preserving teachersā psychological well-being is crucial for their effective performance and for supporting the overall educational experience of students, including their academic success and personal and social well-being. Short-term sickness absences can escalate to long-term absences if teachers are not adequately supported. Implementing workplace health promotion initiatives can also play a significant role in addressing mental health challenges among educators and enhancing their overall psychological wellness, which would, in turn, help to ensure that students attain their full potential. Additionally, interventions that focus on promoting self-compassion, with the possibility of increasing positive affect and reducing negative affect, could have a positive impact on both current and future resilience ().
Consequently, building on these findings, future research can help develop more effective strategies for supporting educatorsā well-being, ultimately improving outcomes for teachers and students. Governments and policymakers in the education sector should endeavor to create a supportive environment that enables teachers to thrive in their professional roles without resorting to increased sick leave, which can jeopardize student learning and achievement.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by the University of Alberta Ethics Review Board (Pro00117558) and the Dalhousie University Human Research Ethics Review Board (REB # 2023ā6,840). The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participantsā legal guardians/next of kin because implied consent was approved, wherein participantsā completion of the online survey indicated their consent.
Author contributions
BA: Writing ā review & editing, Writing ā original draft, Visualization, Validation, Software, Project administration, Methodology, Investigation, Formal analysis, Data curation. YW: Writing ā review & editing, Visualization, Validation, Supervision, Software, Investigation. RL: Writing ā review & editing, Visualization, Validation, Software, Investigation. AO: Writing ā review & editing, Visualization, Validation, Software, Investigation. PB-M: Writing ā review & editing, Visualization, Validation, Software, Investigation. VA: Writing ā review & editing, Visualization, Validation, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Conceptualization.
Funding
The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. RBC Foundation and Alberta Mental Health Foundation funded this research. The funder has no role in the design, data collection, analysis, preparation, or decision to publish the manuscript.
Acknowledgments
We thank the Nova Scotia Teachers Union, the Newfoundland and Labrador Teachers Union, the Edmonton Catholic School Board, and the Alberta Teachers Association for promoting the Wellness4Teachers program among their teachers. We are also grateful to all the teachers who completed the survey, despite their busy schedules.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisherās note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AdamsR. (2023). Record numbers of teachers in England quitting profession, figures show. London: The Guardian.
2
AgyapongB.da Luz DiasR.WeiY.AgyapongV. I. O. (2024). Burnout among elementary and high school teachers in three Canadian provinces: prevalence and predictors. Front. Public Health12:1396461. doi: 10.3389/fpubh.2024.1396461
3
AgyapongB.Obuobi-DonkorG.BurbackL.WeiY. (2022). Stress, burnout, anxiety and depression among teachers: a scoping review. Int. J. Environ. Res. Public Health19:10706. doi: 10.3390/ijerph191710706
4
AgyapongB.WeiY.da Luz DiasR.AgyapongV. I. O. (2022). Burnout and associated psychological problems among teachers and the impact of the Wellness4Teachers supportive text messaging program: protocol for a cross-sectional and program evaluation study. JMIR Res Protoc11:e37934. doi: 10.2196/37934
5
AhmadS.HussainS.ShahF. S.AkhtarF. (2017). Urdu translation and validation of GAD-7: a screening and rating tool for anxiety symptoms in primary health care. J. Pak. Med. Assoc.67, 1536ā1540
6
AmiriS.BehnezhadS. (2021). Depression symptoms and risk of sick leave: a systematic review and meta-analysis. Int. Arch. Occup. Environ. Health94, 1495ā1512. doi: 10.1007/s00420-021-01703-0
7
Antonio MorianaJ.HerruzoJ. (2006). Variables related to psychiatric sick leave taken by Spanish secondary school teachers. Work Stress20, 259ā271. doi: 10.1080/02678370600929354
8
BeltmanS.MansfieldC.PriceA. (2011). Thriving not just surviving: a review of research on teacher resilience. Educ. Res. Rev.6, 185ā207. doi: 10.1016/j.edurev.2011.09.001
9
BorritzM.ChristensenK. B.BültmannU.RuguliesR.LundT.AndersenI.et al. (2010). Impact of burnout and psychosocial work characteristics on future long-term sickness absence. Prospective results of the Danish PUMA study among human service workers. J. Occup. Environ. Med.52, 964ā970. doi: 10.1097/JOM.0b013e3181f12f95
10
BowersT.McIverM. (2000). Ill health retirement and absenteeism amongst teachers. Norwich NR3 1BQ: DFEE.
11
British Columbiaās Teachers' Federation (2023). āResults of the first annual British Columbiaās Teachersā Federation membership surveyā in British Columbiaās teachersā federation (Vancouver, BC).
12
CampbellD. (2024). Mental health sick days taken by civil servants rise 38% in one year. London: The Guardian.
13
Canadian Institute for Health Information (2022). A profile of physicians in Canada. Ottawa: Canadian Institute for Health Information (CIHI).
14
Canadian Teachers' Federation (2017). The Canadian Teachers' Federation honours exceptional educational leaders, educators, and a national literacy organization. Nova Scotia Teachers Union: Halifax.
15
CarlsenB. (2012). From absence to absenteeism? A qualitative cross case study of teachersā views on sickness absence. J. Socio-Econ.41, 129ā136. doi: 10.1016/j.socec.2011.12.007
16
CieslinskiR.SzumE. (2014). Burned out or just frustrated? Reasons why physical education teachers leave their profession. Phys Cult Sport63, 29ā35. doi: 10.2478/pcssr-2014-0020
17
CohenS.KamarckT.MermelsteinR. (1983). A global measure of perceived stress. J. Health Soc. Behav.24, 385ā396. doi: 10.2307/2136404
18
CoxJ. (2020). What is the role of a teacher?New York, NY: ThoughtCo.
19
EngmannB. (2013). Could resilience predict the outcome of psychiatric rehabilitation patients. Open J. Depr.2, 7ā10. doi: 10.4236/ojd.2013.22002
20
ErvastiJ.KivimakiM.PuusniekkaR.LuopaP.PenttiJ.SuominenS.et al. (2012). Studentsā school satisfaction as predictor of teachersā sickness absence: a prospective cohort study. Eur J Public Health22, 215ā219. doi: 10.1093/eurpub/ckr043
21
EryılmazA.Yıldırım-KurtuluÅH.YıldırımM. (2024). A cross-sectional and longitudinal study on the mediation effect of positive and negative affects in the relationship between self-compassion and resilience in early adolescents. Child Youth Serv. Rev.161:107669. doi: 10.1016/j.childyouth.2024.107669
22
FergusonK.JamesY.BourgeaultI. (2022). Teacher mental health and leaves of absences: a pilot study examining gender and care. Can. J. Educ.45, 315ā349. doi: 10.53967/cje-rce.v45i2.4485
23
FlachP. A.GroothoffJ. W.KrolB.BultmannU. (2012). Factors associated with first return to work and sick leave durations in workers with common mental disorders. Eur. J. Pub. Health22, 440ā445. doi: 10.1093/eurpub/ckr102
24
Government of Canada (2024). Data reference standard on Canadian provinces and territories. Government of Canada: Ottawa.
25
Government of Canada (2024). Job opportunities in Nova Scotia. Ottawa: Government of Canada.
26
Government of Newfoundland and Labrador (2008). The 2007/08 teaching workforce. St. Johnās, NL: Government of Newfoundland and Labrador.
27
Government of Newfoundland and Labrador (2024). Education statistics. St. Johnās, NL.
28
HamannJ.ParchmannA.MendelR.BühnerM.ReichhartT.KisslingW. (2013). Understanding the term burnout in psychiatry and psychotherapy. Nervenarzt84, 838ā843. doi: 10.1007/s00115-013-3804-x
29
HusseyL.TurnerS.ThorleyK.McNameeR.AgiusR. (2012). Work-related sickness absence as reported by UK general practitioners. Occup. Med.62, 105ā111. doi: 10.1093/occmed/kqr205
30
IBM (2022). Release notes: IBMĀ® SPSSĀ® Statistics 28. Armonk, NY: IBM.
31
JainG.RoyA.HarikrishnanV.YuS.DabbousO.LawrenceC. (2013). Patient-reported depression severity measured by the PHQ-9 and impact on work productivity: results from a survey of full-time employees in the United States. J. Occup. Environ. Med.55, 252ā258. doi: 10.1097/JOM.0b013e31828349c9
32
JanssonC.AlexandersonK. (2013). Sickness absence due to musculoskeletal diagnoses and risk of diagnosis-specific disability pension: a nationwide Swedish prospective cohort study. Pain154, 933ā941. doi: 10.1016/j.pain.2013.03.001
33
JohnsonS.UlvenesP. G.ĆktedalenT.HoffartA. (2019). Psychometric properties of the general anxiety disorder 7-item (GAD-7) scale in a heterogeneous psychiatric sample. Front. Psychol.10:1713. doi: 10.3389/fpsyg.2019.01713
34
KidgerJ.BrockmanR.TillingK.CampbellR.FordT.ArayaR.et al. (2016). Teachers' wellbeing and depressive symptoms, and associated risk factors: a large cross sectional study in English secondary schools. J. Affect. Disord.192, 76ā82. doi: 10.1016/j.jad.2015.11.054
35
KroenkeK.SpitzerR. L.WilliamsJ. B. (2001). The PHQ-9: validity of a brief depression severity measure. J. Gen. Intern. Med.16, 606ā613. doi: 10.1046/j.1525-1497.2001.016009606.x
36
LagerveldS. E.BültmannU.FrancheR. L.van DijkF. J. H.VlasveldM. C.van der Feltz-CornelisC. M.et al. (2010). Factors associated with work participation and work functioning in depressed workers: a systematic review. J. Occup. Rehabil.20, 275ā292. doi: 10.1007/s10926-009-9224-x
37
LizanaP. A.LeraL. (2022). Depression, anxiety, and stress among teachers during the second COVID-19 wave. Int. J. Environ. Res. Public Health19:5968. doi: 10.3390/ijerph19105968
38
MaslachC.JacksonS. E.LeiterM. P. (1997). Maslach burnout inventory. Lanham Maryland, USA: Scarecrow Education.
39
MaslachC.JacksonS. E.SchwabR. (2022). MBI: educators survey. California: Mind Garden Inc.
40
MaslachC.LeiterM. P. (2016). Understanding the burnout experience: recent research and its implications for psychiatry. World Psychiatry15, 103ā111. doi: 10.1002/wps.20311
41
MizalaA.ĆopoH. (2016). Measuring the relative pay of school teachers in Latin America 1997ā2007. Int. J. Educ. Dev.47, 20ā32. doi: 10.1016/j.ijedudev.2015.11.014
42
MuschallaB. (2016). āWork-related anxieties and work-place phobia: a topical phenomenon at the interface of clinical and organizational research and practiceā in Healthy at work: interdisciplinary perspectives. eds. WienckeM.CacaceM.FischerS.Springer, Cham. 53ā68.
43
MuschallaB.LindenM.OlbrichD. (2010). The relationship between job-anxiety and trait-anxietyāa differential diagnostic investigation with the job-anxiety-scale and the state-trait-anxiety-inventory. J. Anxiety Disord.24, 366ā371. doi: 10.1016/j.janxdis.2010.02.001
44
Newfoundland and a.L.E.S. District, District Overview (2022). Newfoundland and Labrador English School District. Canada: St. John Newfoundland and Labrador.
45
Nova Scotia Teachers Union (2015). Nova Scotia teachers recognized for making a difference. Nova Scotia Teachers Union: Halifax.
46
OāHaireN. (2024). āGender and leadershipā in CTF surveys gender composition of Canadaās teaching population. (Edmonton, Alberta: The Alberta Teachersā Association).
47
PressleyT. (2021). Factors contributing to teacher burnout during COVID-19. Educ. Res.50, 325ā327. doi: 10.3102/0013189X211004138
48
REDCap (2022). University of Alberta. Edmonton: Faculty of Dentistry & Dentistry.
49
RemorE. (2006). Psychometric properties of a European Spanish version of the perceived stress scale (PSS). Span. J. Psychol.9, 86ā93. doi: 10.1017/S1138741600006004
50
SalvagioniD. A. J.MesasA. E.MelandaF. N.GonzĆ”lezA. D.de AndradeS. M. (2022). Burnout and long-term sickness absence from the teaching function: a cohort study. Saf. Health Work13, 201ā206. doi: 10.1016/j.shaw.2022.01.006
51
SchneiderA.HilbertS.HamannJ.SkadsemS.GlaserJ.LƶweB.et al. (2017). The implications of psychological symptoms for length of sick leave. Dtsch. Arztebl. Int.114, 291ā297. doi: 10.3238/arztebl.2017.0291
52
SchonfeldI. S.BianchiR.Luehring-JonesP. (2017). āConsequences of job stress for the mental health of teachersā in Educator stress: an occupational health perspective, eds. McIntyreT.McIntyreS.FrancisD.Springer, Cham. 55ā75.
53
SchwabM. (2019). Alberta (Canada) who are todayās principals and teachers and the students in their classrooms?Paris: OECD Teaching and Learning International Survey (TALIS).
54
SmithB. W.DalenJ.WigginsK.TooleyE.ChristopherP.BernardJ. (2008). The brief resilience scale: assessing the ability to bounce back. Int. J. Behav. Med.15, 194ā200. doi: 10.1080/10705500802222972
55
SoerR.Six DijkstraM. W. M. C.BielemanH. J.StewartR. E.RenemanM. F.OosterveldF. G. J.et al. (2019). Measurement properties and implications of the brief resilience scale in healthy workers. J. Occup. Health61, 242ā250. doi: 10.1002/1348-9585.12041
56
SpitzerR. L.KroenkeK.WilliamsJ. B. W.LƶweB. (2006). A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch. Intern. Med.166, 1092ā1097. doi: 10.1001/archinte.166.10.1092
57
Statistic Canada (2024). Canada's population estimates: Strong population growth in 2023. Ottawa: Government of Canada.
58
Statistics Canada (2023). Annual statutory teachers' salaries in public institutions, by level of education taught and teaching experience, Canadian dollars: Ottawa, Canada.
59
Statistics Canada (2023). Population estimates, quarterly, statistics, editor. Ottawa: Government of Canada.
60
The Alberta Teachers' Association (2023). āTeacher count: province only 38 teachers aheadā in The Alberta Teacher's association (Edmonton: ATA).
61
TitovN.DearB. F.McMillanD.AndersonT.ZouJ.SunderlandM. (2011). Psychometric comparison of the PHQ-9 and BDI-II for measuring response during treatment of depression. Cogn. Behav. Ther.40, 126ā136. doi: 10.1080/16506073.2010.550059
62
Toppinen-TannerS.OjajƤrviA.VƤƤnƤnenA.KalimoR.JƤppinenP. (2005). Burnout as a predictor of medically certified sick-leave absences and their diagnosed causes. Behav. Med.31, 18ā32. doi: 10.3200/BMED.31.1.18-32
63
UsmanS.AkhmadiSuryadarmaD. (2007). Patterns of teacher absence in public primary schools in Indonesia. Asia Pac. J. Educ.27, 207ā219. doi: 10.1080/02188790701378826
64
VukmirovicM.RajovicN.PavlovicV.MasicS.MirkovicM.TasicR.et al. (2020). The burnout syndrome in medical academia: psychometric properties of the Serbian version of the Maslach burnout inventory-educators survey. Int. J. Environ. Res. Public Health17:5658. doi: 10.3390/ijerph17165658
65
WinklerD. R. (1980). The effects of sick-leave policy on teacher absenteeism. ILR Rev.33, 232ā240. doi: 10.1177/001979398003300207
Summary
Keywords
burnout, educators, generalized anxiety disorder, moderate depressive disorder, resilience, sick days, stress, teachers
Citation
Agyapong B, Wei Y, da Luz Dias R, Orimalade A, Brett-MacLean P and Agyapong VIO (2024) Psychological problems among elementary and high school educators in Canada: association with sick days in the prior school year. Front. Psychol. 15:1442871. doi: 10.3389/fpsyg.2024.1442871
Received
10 June 2024
Accepted
24 September 2024
Published
07 October 2024
Volume
15 - 2024
Edited by
Murat Yildirim, AÄrı İbrahim ĆeƧen University, Türkiye
Reviewed by
Dragan Mijakoski, Institute of Occupational Health of RNM, North Macedonia
Juan Pedro MartĆnez-Ramón, University of Murcia, Spain
Updates
Copyright
Ā© 2024 Agyapong, Wei, da Luz Dias, Orimalade, Brett-MacLean and Agyapong.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Belinda Agyapong, bagyapon@ualberta.ca
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.