Predictors of health-related absenteeism,presenteeism and sick leave among brazilian elementary school teachers: A cross-sectional study

Abstract

BACKGROUND:

The health risks that usually affect teachers are already known; however, the predictors of health related absenteeism, presenteeism, and sick leave have not yet been fully described.

OBJECTIVE:

To analyze the predictors of health-related absenteeism, presenteeism, and sick leave among elementary school teachers.

METHODS:

This study involved a probabilistic sample of 519 Brazilian elementary school teachers. The outcomes were days of health-related absenteeism, presenteeism, and sick leave in the previous 12 months. Work-place and individual factors were the independent variables analyzed.

RESULTS:

Inadequate infrastructure of schools, disability, and medical consultations were positively associated with all outcomes. Teachers who reported having a chronic disease, common mental disorders, and voice disorders presented higher absenteeism and presenteeism days. Musculoskeletal pain and low job support were associated with higher presenteeism and sick leave days. Teachers who performed strength and flexibility activities presented less presenteeism, those who reported physical violence at school were more frequently absent, and teachers with depersonalization presented a higher likelihood of sick leave.

CONCLUSIONS:

The prevention of health-related absenteeism, presenteeism, and sick leave among elementary teachers should ensure adequate work conditions and prevention and monitoring of health risks.

Keywords

Chronic disease health behavior mental health school health sick days

1 Introduction

It is well known that a worker’s occupational env-ironment can expose them to a variety of psychoso-cial stressors in which contribute to the development both chronic diseases and injuries, through physiological, psychological, and behavioral mechanisms [1 –3]. The presence of health related dis-orders among employees can lead to presenteeism, (i.e., employee works despite being sick), health-related absenteeism (occasional absence from work due to a medical problem), or sick leave (medically certificated absence from work due to illness) [4 –6]. It has been described that presenteeism is associ-ated with future absenteeism [6, 7], although pre-senteeism can also occur through early return to work after sick leave, when medical problems and symptoms are still present. Both absenteeism and presenteeism result in high economic costs due to productivity loss [6].

Work-place and health characteristics of teachers have received attention from researchers. It has been described that a high prevalence of Brazilian eleme-ntary teachers have a negative perception of their environment and work conditions, such as ergonomic suitability of furniture and equipment, noise and temperature, opportunities for growth and professional improvement, motivation and mood, remuneration, health benefits offered by the institution, violence, inadequate infrastructure, and safety of schools [8]. Since aspects related to work are associated with teacher health [8, 9], there is a large amount of information describing a high prevalence of behavioral risks, high stress, poor perceived physical and mental health, and musculoskeletal, mental, respiratory, and voice disorders among teachers [9 –12].

Teachers also present a higher rate of sick leave compared to many other professions [13]. Health-re-lated absenteeism among teachers has been positively associated with health risk behaviors, poor school environment, and voice, mental, and musculoskeletal disorders [5 , 14–19]. Although there is less research concerning presenteeism, the results appear to indicate similar findings as observed for absenteeism [20 –22]. These variables require investigation as they impact negatively on health, work performance, and social relationships, and can also negatively influence student academic performance [23].

Analyzing the literature on health related-abs-enteeism, presenteeism, and sick leave among teachers, two aspects should be considered. First, much of the information available comes from samples in developed countries, [24] where the differences between the educational systems prevent generalization of the results to developing regions. Although studies with South American teachers are available, they have only analyzed absenteeism related to voice disorders [18, 19]. There appears to be no studies focusing on the factors associated with presenteeism and sick leave. Second, the majority of studies performed focused on specific health characteristics and working conditions of teachers [5 , 19]. A multifactorial approach including work related factors, lifestyle risks, health disorders, disability, and health service use will bring knowledge on which characteristics presented by teachers are the main predictors of health-related absenteeism, presenteeism, and sick leave. The results will also provide epidemiological information regarding Brazilian teachers, guiding monitoring and intervention programs aimed at preventing these outcomes and promoting work well-being. Thus, the purpose of the present study was to analyze the predictors of health-related absenteeism, presenteeism, and sick leave among Brazilian elementary school teachers.

2 Methods

2.1 Population and study design

This is an epidemiological cross-sectional study conducted from July to December 2014 in the Municipal Education System of Londrina, Paraná, Brazil. An overview of Londrina city can be found elsewhere [25]. In 2014 the Municipal Education System of Londrina had 2500 teachers distributed in 74 schools in five regions of the city: north (32.8%), south (20.0%), east (20.3%), west (22.1%), and center (4.8%).

2.2 Sample

The sample was composed of elementary school teachers, stratified according to region of the city, who met the following inclusion criteria: a-) having been a teacher of municipal schools for at least 1 year and working in an elementary school; b-) not being retired or on medical leave during data collection; and c-) not having been work relocated (i.e., teacher working as a secretary or in administration). All schools (N = 74) were invited to participate in the study and 63 accepted. Sample size was estimated using the following parameters: N = 2.500, a 50% out-come prevalence, 5% sample error, confidence interval of 95%, design effect of 1.5, and sample loss of 15%, using the software OpenEpi 3.01. The minimum number of participants required was 500.

2.3 Data collection

The study was approved by the Ethics Committee for Research involving human beings of the State University of Londrina, process 118/2014. The guidelines of Resolution No 466/2012 of the Brazilian National Health Council were followed. After approval from the Municipal Department of Education of Londrina to conduct the study, all schools were visited randomly and the study was presented to the principals. The decision to visit the schools in a random order was because data collection was from July to December. Generally in Brazil, at the end of the semester, the teacher workload increases due to the exams and this can impact health. In schools where the principals agreed to participate; the study purpose and procedures were presented to teachers, an informed consent was signed, and data collection was scheduled. Data collection was conducted in each school where teachers were enrolled, by the coordinator of the project. Teachers completed an anonymous self-report questionnaire which was then stored in an opaque locked box. The schools were visited again to obtain data from teachers who were absent on the first scheduled day.

2.4 Variables

All variables of the present study were assessed through a self-report questionnaire. Self-report he-alth-related absenteeism and presenteeism present acceptable validity to be used in epidemiological studies, mainly when administrative data are unavailable [26 –28]. The outcomes of the study were:

Health-related absenteeism (occasional days of unplanned absence from work due to health problems during the previous 12 months);

Health-related presenteeism (days during the previous 12 months that teachers worked despite the presence of sickness and not being able to perform their activities as usual);

Medically certified sick leave (days of planned absence due to illness, attested by an institutional official medical board).

The legislation of the municipal system where the sample was recruited requires a medical evaluation performed by an institutional official medical board to provide sick leave for an employee. Sick leave is generally continuous and may last up to 24 months, a period which may be extended at the discretion of the institutional official medical board. Any occasional absence from work is characterized as absenteeism and if the employee provides a medical certificate at-testing their absence it is considered as justified absence, but not sick leave. However, if it is not attested by a doctor, the absence on this day will be deducted from the monthly payment and may lead to disadvantages in the institutional assessment. Previous studies conducted in other countries use different terms, as absenteeism, sick leave or sickness absence to describe the same phenomenon [4 –6]. Due to the specificity of the outcomes in Brazil, we decided to separate the analysis and both absenteeism and sick leave were analyzed. Open questions used to assess the outcomes were: Absenteeism “In the last year, how many days were you absent from work through your own decision because of your health”; Presenteeism “in the last year, while working, how many days did you not perform your activities as usual due to some health problem”; Sick leave “In the last year, how many days did you need to be on medical sick leave?”

The independent variables studied were grouped into work-place and individual factors [6]: Work-place factors: School infrastructure, physical viol-ence, work stress, and low job support; Individual factors: length of employment. overweight, common mental disorders, physical activity, strength and fle-xibility activities, musculoskeletal pain, alcohol consumption, tobacco use, problems related to dust, bur-nout, chronic disease, disability due to musculoskeletal disorders, and medical consultation. Additionally, the models were adjusted for the covariates age, income, and sex.

2.4.1 Work-place factors

School infrastructure was assessed through the question “Do you consider the infrastructure of your school to be appropriate?” (a list of items that should be considered was displayed: noise, temperature, lighting, cleaning, ventilation, size, and furniture) with answer options “yes”, “no”, and “partially”. Physical violence was reported by the question “Have you ever suffered physical violence at work?” with response options “Yes” and “No”. Work stress and Job support were assessed using the Brazilian version of the Job Stress Scale [29].

2.4.2 Individual factors

Length of employment was analyzed using an open question. Being overweight was assessed using body mass index and measures of height and weight were self-reported. Physical activity was estimated using the Brazilian version of the International Physical Activity Questionnaire - IPAQ [30], considered recreational, sport, exercise, and leisure time physical activity (cut-off adopted was 150 min moderate to vigorous physical activity). Strength and flexibility activities were estimated through the question “How many times a week do you perform muscle strengthening exercises?” and “How many times a week do you perform flexibility exercises?”. The cut-off used was equal to or more than two times a week in both activities. Musculoskeletal pain was estimated using the Brazilian version of the Standardized Nordic questionnaire [31]. Common mental disorders were assessed by the Brazilian version of a Self Report Questionnaire [32] and the cut-off adopted was 7/8. The Maslach Burnout Inventory was used to assess Burnout [33]. Some questions were asked to assess Alcohol consumption: “Usually, how many times in a normal month do you consume ≥5 doses of the drinks below (men) or ≥4 doses (women) in less than 2 hours? (the milliliters of beer, wine, and distilled drink doses were displayed); Tobacco use: “Have you smoked at least 100 cigarettes in your entire life?”, “yes” and “no”; “How often during a typical week do you smoke”, “none”, “1–2 days”, “3–4 days”, “5–6 days”, and “daily”; Voice disorders: “Do you have a frequent voice-related problem? “yes”, “no”; Problems related to chalk powder: “Do you have a frequent problem with dust or chalk powder?”, “yes”, “no” for each problem (nasal stuffiness, eye irritation, rhinitis, coryza, cough, and skin problems). Chronic disease was estimated using the following question: “Has a doctor or psychologist reported that you have any of the following chronic diseases?” with answer options “yes” and “no” for a list of categories of chronic diseases (cardiometabolic, psychological, orthopedic, respiratory, gastrointestinal, nervous system, and cancer). Disability due to musculoskeletal disorders in each body site was estimated using the Standardized Nordic questionnaire [31]. The number of medical consultations in the previous 12 months was estimated using the question “How many times in the last 12 months have you attended a medical consultation?”

2.4.3 Information quality

The questionnaire was composed mainly of instruments previously validated and culturally adapted to the Brazilian language [29 –33]. However, due to the absence of instruments in the Portuguese language, some questions (described in items above) were developed to achieve the aims of the present study. Steps were followed to evaluate content validity to ensure the quality of information [34]: a) The researchers of the project proposed the questions; b) The content validity of the questions was assessed by a panel of experts [35] in health science areas (from three different Universities, with Phd degrees, experience both in instrument validation and supervising research projects in post-graduate programs, and with relevant publications in their research areas); c) The questions were modified according to the suggestions of the reviewers; d) A Pilot study with 50 elementary teachers was conducted to assess comprehension and the reproducibility of the questions within seven days test-retest. Variables were included in this study if questions presented acceptable reproducibility (intraclass coefficient >0.5 for continuous/discrete variables and kappa index >0.40 for categorical variables).

2.5 Statistical analysis

Descriptive statistics are presented as mean, standard deviation, relative frequency, and rate. Bivariate and multivariate analyses were conducted using negative binomial regression to estimate the rate ratio (RR) and confidence intervals of 95% (CI95%). This procedure was adopted due to overdispersion presented by the outcomes. Variables that presented significant associations in the bivariate analysis were included in the hierarchical multivariate analysis. Work-place factors were inserted in the first level. Due to the interrelationship between the individual factors on the etiology of the outcomes studied, they were inserted in three other levels according to the conceptual hierarchical framework: health characteristics, chronic disease and disability, and health care need [36]. Upper-level variables were adjusted for those at the previous level, as well as for age, sex, and income covariates. Multivariate analysis was conducted considering strata, weight, and primary sample units using the package “survey” of the software STATA 13.0. In all cases, significance was set at P≤0.05.

3 Results

Data of 595 teachers were collected, however only 519 participants were included in the final sample due to missing data on the questionnaire. The sample was composed mainly of female teachers (95.1%). The distribution according to socioeconomic status was: low (16.3%), medium (43.1%), and high (40.6%). Participants reported lengths of employment of less than nine years (27.2%), ten to nineteen years (32.3%), and more than twenty years (40.5%) while a similar proportion of the sample were aged less than forty years (48.4%). Table 1 presents the description of the dependent variables. Means of absenteeism, presenteeism, and sick leave were 3.19, 4.86, and 4.78 days in one year, respectively. The prevalence of one or more days varied from 27.6% for sick leave to 64.4% for absenteeism.

Table 1
Description of dependent variables among the sample (n = 519)

Variables (days/year) Mean (SD) Prevalence≥1

Health-related absenteeism 3.19 (6.94) 64.4

Health-related presenteeism 4.86 (19.13) 42.8

Sick leave 4.78 (20.52) 27.6

Variables (days/year)	Mean (SD)	Prevalence≥1
Health-related absenteeism	3.19 (6.94)	64.4
Health-related presenteeism	4.86 (19.13)	42.8
Sick leave	4.78 (20.52)	27.6

SD: Standard deviation.

The bivariate association between independent variables studied and absenteeism, presenteeism, and sick leave is described in Table 2 and the multivariate analysis is presented in Table 3. The multivariate analysis revealed that inadequate infrastructure of schools (RR = 1.83 to 3.19), disability due to musculoskeletal disorders (RR = 1.37 to 2.70), and medical consultations (RR = 2.70 to 3.10) were positively associated with all outcomes. The variables common mental disorders (RR = 1.67 and 3.40), voice disor-ders (RR = 1.43 and 1.77), and chronic diseases (RR =2.21 and 1.98) resulted in a higher rate ratio of absenteeism and presenteeism days respectively. Teachers with low job support (RR = 1.73 and 2.27) and musculoskeletal pain in three or more regions of the body (RR = 2.43 and 2.93) presented a higher rate ratio of presenteeism and sick leave days respectively. Teachers who practiced strength and flexibility activities presented a lower rate ratio of presenteeism (RR = 0.44), those with depersonalization presented higher sick leave (RR = 2.89), and the experience of violence in school was associated with a higher rate of absenteeism (RR = 1.50).

Table 2

Bivariate analysis of the association between the independent variables analyzed and health-related absenteeism, presenteeism, and sick leave days among elementary teachers (n = 519)

	Absenteeism		Presenteeism		Sick leave
Variables	Rate	RR (CI95%)	Rate	RR (CI95%)	Rate	RR (CI95%)
School infrastructure
Adequate	2.65	Reference	3.36	Reference	4.12	Reference
Not adequate	5.06	1.90 (1.40–2.58)	10.10	3.00 (1.80–5.00)	7.13	1.73 (0.82–3.63)
Physical violence
No	2.71	Reference	3.94	Reference	4.63	Reference
Yes	4.63	1.76 (1.31–2.37)	7.98	1.98 (1.18–3.31)	5.29	1.11 (0.53–2.34)
Low job support
No	2.53	Reference	3.44	Reference	2.54	Reference
Yes	4.06	1.60 (1.23–2.08)	6.77	1.96 (1.26–3.05)	7.79	3.05 (1.65–5.65)
Work stress
No	3.37	Reference	5.71	Reference	5.41	Reference
Yes	2.74	0.81 (0.60–1.09)	2.80	0.49 (0.30–0.79)	3.23	0.59 (0.30–1.18)
Length of employment
<9 years	3.00	Reference	6.26	Reference	4.62	Reference
10 to 19 years	2.68	0.89 (0.63–1.26)	4.52	0.72 (0.40–1.27)	4.54	0.98 (0.43–2.21)
>20 years	3.75	1.25 (0.90–1.74)	4.27	0.68 (0.39–1.18)	5.12	1.10 (0.50–2.42)
Overweight
No	3.32	Reference	3.98	Reference	5.51	Reference
Yes	3.00	0.90 (0.68–1.19)	5.63	1.41 (0.91–2.19)	3.67	0.66 (0.34–1.30)
Common mental disorders
No	2.45	Reference	2.47	Reference	3.66	Reference
Yes	4.87	1.98 (1.50–2.61)	10.36	4.18 (2.67–6.56)	7.36	2.01 (1.02–3.93)
Physical activity
No	3.02	Reference	5.42	Reference	5.45	Reference
Yes	3.78	1.25 (0.90–1.72)	2.78	0.51 (0.29–0.87)	2.28	0.41 (0.19–0.89)
Strength and flexibility
No	3,25	Reference	5.63	Reference	5.34	Reference
Yes	2.82	0.86 (0.63–1.17)	2.08	0.36 (0.22–0.61)	3.11	0.58 (0.28–1.20)
Musculoskeletal pain
No	2.12	Reference	1.21	Reference	1.81	Reference
One to two	2.42	1.50 (1.02–2.20)	4.17	3.43 (1.83–6.42)	4.92	2.71 (1.11–6.57)
≥three	3.66	1.72 (1.16–2.55)	7.34	6.03 (3.18–11.46)	5.93	3.26 (1.31–8.11)
Alcohol consumption
No	3.32	Reference	5.50	Reference	5.31	Reference
Yes	2.49	0.75 (0.56–1.05)	3.73	0.67 (0.41–1.10)	3.16	0.59 (0.30–1.17)
Tobacco use
Never	3.18	Reference	4.97	Reference	5.11	Reference
Not current smoker	3.57	1.12 (0.70–1.77)	2.46	0.49 (0.22–1.07)	3.31	0.60 (0.20–1.80)
Current smoker	2.45	0.77 (0.38–1.55)	8.00	1.60 (0.51–5.00)	1.25	0.24 (0.04–1.27)
Voice disorders
No	2.58	Reference	3.60	Reference	3.30	Reference
Yes	4.35	2.58 (2.19–3.03)	7.27	2.01 (1.27–3.18)	7.59	2.29 (1.20–4.38)
Problems related to
dust or chalk powder
No	2.51	Reference	2.77	Reference	4.09	Reference
One	2.82	1.10 (0.78–1.64)	3.78	1.33 (0.77–2.32)	5.41	1.39 (0.63–3.07)
≥Two	3.88	1.62 (1.17–2.25)	5.42	2.62 (1.53–4.48)	4.91	1.26 (0.58–2.73)
Burnout
Exhaustion
No	2.75	Reference	3.19	Reference	3.5	Reference
Yes	4.46	1.61 (1.20–2.18)	9.79	3.06 (1.88–4.98)	8.57	2.44 (1.20–4.96)
Depersonalization
No	2.95	Reference	3.31	Reference	3.13	Reference
Yes	3.75	1.27 (0.95–1.69)	8.62	2.60 (1.62–4.16)	8.79	2.80 (1.43–5.48)
Professional achievement
No	2.60	Reference	3.25	Reference	4.77	Reference
Yes	4.81	1.84 (1.38–2.47)	9.35	2.87 (1.77–4.66)	4.79	1.00 (0.49–2.03)
Disability due to
musculoskeletal disorders
No	2.35	Reference	2.37	Reference	3.45	Reference
Yes	4.71	2.00 (1.53–2.62)	9.39	3.95 (2.56–6.09)	7.19	2.08 (1.09–3.96)
Chronic disease
None	1.86	Reference	4.07	Reference	1.35	Reference
One to two	2.50	1.34 (0.95–1.90)	3.43	0.84 (0.47–1.49)	3.99	2.95 (1.31–6.62)
≥Three	5.02	2.70 (1.87–3.89)	7.44	1.82 (1.00–3.36)	8.12	6.01 (2.54–14.18)
Medical consultations
No	1.69	Reference	2.13	Reference	1.17	Reference
One to two	1.72	1.01 (0.69–1.49)	1.77	0.83 (0.44–1.55)	3.27	2.75 (1.14–6.84)
> =Three	4.61	2.73 (1.93–3.86)	7.71	3.62 (2.06–6.36)	7.03	6.01 (2.64–13.65)

RR (CI 95%) = Crude rate ratio (confidence interval of 95%). Bold denotes statistical significance at P < 0.05.

Table 3

Multivariate analysis of the association between the independent variables analyzed and health-related absenteeism, presenteeism, and sick leave among elementary teachers

Variables	Absenteeism	Presenteeism	Sick leave
	Adjusted RR (CI95)	Adjusted RR (CI95)	Adjusted RR (CI95)
Work-place level
School infrastructure
Adequate	Reference	Reference	Reference
Not adequate	1.83 (1.32–2.55)	2.95 (1.76–4.93)	3.19 (1.16–8.77)
Physical violence
No	Reference	Reference	Reference
Yes	1.50 (1.02–2.26)	1.08 (0.58–2.02)	–
Low job support
No	Reference	Reference
Yes	1.24 (0.93–1.66)	1.73 (1.07–2.79)	2.27 (1.22–4.21)
Work stress
No	Reference	Reference	Reference
Yes	–	0.70 (0.49–1.35)	–
Individual level
Common mental disorders
No	Reference	Reference	Reference
Yes	1.67 (1.26–2.20)	3.40 (1.99–5.81)	1.24 (0.67–2.28)
Physical activity
No	Reference	Reference	Reference
Yes	–	0.85 (0.50–1.45)	–
Strength and flexibility
No	Reference	Reference	Reference
Yes	–	0.44 (0.23–0.83)	–
Musculoskeletal pain
No	Reference	Reference	Reference
One to two	1.11 (0.66–1.84)	1.43 (0.71–2.87)	1.58 (0.65–3.83)
≥Three	1.20 (0.76–1.92)	2.43 (1.12–5.30)	2.93 (1.19–7.20)
Voice disorders
No	Reference	Reference	Reference
Yes	1.43 (1.02–1.98)	1.77 (1.24–2.52)	1.58 (0.85–2.93)
Problems related to dust or chalk powder
No	Reference	Reference	Reference
One	0.78 (0.52–1.19)	0.69 (0.36–1.35)	–
≥Two	1.14 (0.71–1.81)	0.76 (0.41–1.39)	–
Work stress
No	Reference	Reference	Reference
Yes	–	0.96 (0.58–1.59)	–
Burnout
Exhaustion
No	Reference	Reference	Reference
Yes	0.91 (0.64–1.28)	1.54 (0.98–2.43)	1.79 (0.91–3.51)
Depersonalization
No	Reference	Reference	Reference
Yes	–	1.28 (0.85–1.91)	2.89 (1.39–6.02)
Professional achievement
No	Reference	Reference	Reference
Yes	1.13 (0.84–1.53)	0.94 (0.61–1.45)	–
Disability due to musculoskeletal disorders
No	Reference	Reference	Reference
Yes	1.37 (1.04–1.82)	2.29 (1.45–3.60)	2.70 (1.40–5.22)
Chronic disease
None	Reference	Reference	Reference
One to two	1.31 (0.82–2.09)	1.98 (1.11–3.52)	1.35 (0.51–3.55)
≥Three	2.21 (1.33–3.67)	1.37 (0.74–2.53)	2.09 (0.69–6.28)
Medical consultations
No	Reference	Reference	Reference
One to two	1.33 (0.79–2.23)	1.74 (1.08–2.82)	1.35 (0.64–2.83)
≥Three	2.70 (1.61–4.52)	3.10 (1.75–5.49)	3.05 (1.38–6.72)

RR (CI95%): Rate ratio (confidence interval of 95%). Final models were adjusted for variables that were significantly associated with the outcomes in the bivariate analysis, and for sex, age, and income. Bold denotes statistical significance at P < 0.05.

4 Discussion

The present study highlighted the predictors of health-related absenteeism, presenteeism, and sick leave among elementary teachers. Teachers presented 3.19, 4.86, and 4.78 mean days of health-related abs-enteeism, presenteeism, and sick leave, respectively, in the previous year. The prevalence of one or more days in the previous year was high (>27%), which indicates a relevant public health problem among teachers.

Disability due to musculoskeletal disorders was associated with all outcomes and musculoskeletal pain presented a positive association with presenteeism and sick leave. Both variables are interrelated since disability is a consequence of the severity of pain, which explains the agreement in associations and why they were inserted at different levels in the multivariate analysis. The results corroborate a previous study on teachers [14] and other workers [37] and are relevant since musculoskeletal disorders are common among teachers and are associated with a variety of work characteristics and health conditions [11]. Another variable related to physical health that was negatively associated with presenteeism was strength and flexibility activities. This type of exercise is recommended for all populations and is used as a non-pharmacological prevention and management method for musculoskeletal, cardiovascular, metabolic, and mental disorders [38, 39]. Teachers who report performing strength and flexibility activities are likely to have higher functional capacities and may be protected from work-related risks that result in presenteeism. However, although presenteeism was associated with absenteeism [6], these activities showed no protection for the other outcomes.

Teachers with voice disorders presented a higher rate ratio of health-related absenteeism and presenteeism. Voice disorders are a common health problem in teachers’ careers and the prevalence is higher compared to the general population [40]. The literature demonstrates the prevalence and variables associated with voice-related absenteeism (15, 18) and the present results corroborate these in part, since general health related absenteeism was analyzed. Greater attention has been given to health-related absenteeism compared to presenteeism when analyzing voice disorders, probably because absenteeism reflects higher clinical severity. However, the present study indicates that voice disorders should be monitored since they affect teacher performance both through working with a bad voice condition and being absent from work. Surprisingly voice disorders were not associated with sick leave in the sample studied. A previous study described that 34.6% of teachers are absent from work for only one day due to voice disorders [15]. The severity of voice disorders was not assessed in the present study, however it is probable the treatment and recovery generally do not require sick leave, or teachers are absent frequently on non-consecutive days during the year.

As occurred for voice disorders, common mental disorders were associated both with health-related absenteeism and presenteeism and these results are similar to other studies [6 , 41]. Common mental disorders are prevalent among teachers and can be caused by a variety of work-related variables such as work satisfaction, job demand, social support, student behavior, student attendance, and employment conditions [12 , 42]. The present study adopted a hierarchical multivariate analysis including work and health characteristics that teachers usually present and, after statistical adjustments, common mental disorders was the variable that best predicted presenteeism. Furthermore, although with less magnitude, low job support was also significantly associated with presenteeism, in the same direction as described both in German teachers [20] and other professionals [6]. The results of the present study and those previously described show that independent of cultural, socioeconomic, and work condition differences across countries, mental health plays an important role in the work routine of teachers.

With regard to sick leave, mental health variables presented discrepant results. While teachers with low job support and depersonalization presented a significantly higher rate of sick leave days, common mental disorders was not associated in the multivariate analysis. Comparison of the results for sick leave with the available literature is difficult as studies use different terminology such as absenteeism, sick leave (short-term and long-term), medical leave, and sickness absence [5 , 43] which commonly represent the same phenomenon. In the present study, both health-related absenteeism and medically certified sick leave were considered due to the likelihood of a higher severity of illness in those with medical certified sick leave. Although caution should be adopted due to differences in outcomes across studies, the results are similar to those from two previous reviews describing that burnout and stress are associated with higher absenteeism among teachers [43] and other workers [6]. The same occurs for neurotic and stress-related disorders and long-term sickness absence among teachers [44]. Information provided in the present study reinforces that psychological health is a priority in health monitoring to prevent health-related absenteeism and sick leave and promote well being among teachers.

Inadequate infrastructure of schools seems to negatively influence work condition and consequently teachers’ health through the emergence of respiratory, musculoskeletal, and mental disorders [8 , 11]. It is probable this can explain the positive association between inadequate infrastructure of schools with the health-related absenteeism and presenteeism found in the present study. The results for absenteeism are in line with a review [43] that revealed higher absenteeism among schools with poor working conditions, school facilities, infrastructure, and classroom structure. These results are alarming since a large proportion of Brazilian teachers are dissatis-fied with ergonomic issues regarding furniture, equipment, noise, and temperature conditions in public schools [8]. Another variable related to work condition that was associated with absenteeism was physical violence. Violence is linked to fear among teachers and both variables impair physical and mental health, as well as teaching ability, and can result in absenteeism due to victimization of those exposed to violence [45]. The results of the present study corroborate a previous study that demonstrated higher absenteeism due to a voice problem among teachers exposed to violence [18] and suggested prevention of violence and improvement in school safety to decrease teacher absenteeism.

The analysis of the present study was conducted hierarchically according to relationships between variables. Chronic diseases were inserted in level three and were associated with health-related abs-enteeism and presenteeism. This result was expected due to clinical complications emerging from chronic diseases and suggests that work and behavioral variables associated with chronic diseases [1, 46] should be prevented. Similarly, medical consultation was associated with all outcomes. Medical care is associated with health disorders commonly presented by teachers [47] and for these reasons this variable was inserted in the final level. Teachers that required a medical consultation presented approximately three-fold health-related absenteeism, presenteeism, and sick leave compared to those who had not attended a medical consultation, from which two aspects should be noted. First, this indicates that health-related absenteeism and sick leave of teachers are due to see-king help from physicians to treat their health impairment and not to avoid work. Second, in addition to the outcomes investigated there is higher direct health expenditure due to the financing of medical consul-tations, tests, and medication. This information supports the need to monitor teacher health throughout their career and indicates that costs associated with health-related absenteeism, presenteeism, and sick leave should not be attributed only to productivity loss, but also to direct medical expenditure and the consequence on person-related financial difficulties [6].

Some limitations of the present study should be considered when interpreting the results. All variables studied were assessed by a self-reported questionnaire and could be influenced by recall. However, this limitation was reduced in the present study as the sample was composed of participants with a high level of schooling, added to which valid and reliable instruments were used. The healthy worker effect could also have influenced the results since only participants who were working during the period of data collection were included in the study. It is probable that if teachers on sick leave had been included in the sample, the mean days of absenteeism and the magnitude of associations could have been higher. Despite the limitations, the strengths of the study are the representative sample from the city analyzed, the use of multivariate analysis considering the main work characteristics and health disorders related to the teaching profession, and main outcomes regarding the relationship between work, productivity, and health.

5 Conclusion

The prevalence of teacher health-related absenteeism, presenteeism, and sick leave was high (>27%) in the sample studied. Although the associations varied according to the outcome, inadequate infrastructure of schools, medical consultation and disability due to musculoskeletal disorders were associated with higher health-related absenteeism, presenteeism, and sick leave. Furthermore, a higher rate ratio of two of the outcomes was described in teachers with common mental disorders, musculoskeletal pain, voice disorders, low job support, and chronic disease, while violence, strength and flexibility activities [negative association], and depersonalization predicted one outcome. Since the outcomes analyzed affect teachers worldwide, policies aimed to prevent and reduce health-related absenteeism, presenteeism, and sick leave should ensure adequate work conditions as well as health risk prevention and monitoring among elementary teachers.

Conflict of interest

None to report.

References

Schnall

, Dobson

, Landsbergis

. Globalization, work, and cardiovascular disease. Int J Health Serv. 2016;46(4):656–92.

Harvey

, Modini

, Joyce

, Milligan-Saville

, Tan

, Mykletun

, Bryant

, Christensen

, Mitchell

. B. Can work make you mentally ill? A systematic meta-review of work-related risk factors for common mental health problems. Occup Environ Med. 2017;74(4):301–10.

Scott-Marshall

, Tompa

. The health consequences of precarious employment experiences. Work. 2011;38(4):369–82.

Van Den Berg

, Burdorf

, Robroek

. Associations between common diseases and work ability and sick leave among health care workers. Int Arch Occup Environ Health. 2017;90(7):685–93.

Ervasti

, Kivimäki

, Kawachi

, Subramanian

, Pentti

, Oksanen

, Puusniekka

, Pohjonen

, Vahtera

, Virtanen

. School environment as predictor of teacher sick leave: data-linked prospective cohort study. BMC Public Health. 2012;12(1):770.

Lohaus

, Habermann

. Presenteeism: A review and research directions. Human Resource Management Review. 2019;29(1):43–58.

Skagen

, Collins

. The consequences of sickness presenteeism on health and wellbeing over time: a systematic review. Soc Sci Med. 2016;161:169–77.

Pereira

ÉF

, Teixeira

, Andrade

, Bleyer

, Lopes

ADS

. Associação entre o perfil de ambiente e condições de trabalho com a percepção de saúde e qualidade de vida em professores de educação básica. Cad Saude Colet. 2014;22(2):113–19.

Claudio

, Rivera

, Ramirez

. Association Between Markers of Classroom Environmental Conditions and Teachers’ Respiratory Health. J Sch Health. 2016;86(6):444–51.

10.

Bogaert

, De Martelaer

, Deforche

, Clarys

, Zinzen

. Associations between different types of physical activity and teachers’ perceived mental, physical, and work-related health. BMC Public Health. 2014;14(1):534.

11.

Coledam

DHC

, Pires-Júnior

, Ribeiro

EAG

, Oliveira

. Factors associated with musculoskeletal disorders and disability in elementary teachers: A cross-sectional study. J Bodyw Mov Ther. 2019;23(3):658–65.

12.

Borrelli

, Benevene

, Fiorilli

, D’amelio

, Pozzi

. Working conditions and mental health in teachers: a preliminary study. Occup Med. 2014;64(7):530–32.

13.

Silva

EBDF

, Tomé

LADO

, Costa

TDJGD

, Santana

MDCCP

. Transtornos mentais e comportamentais: perfil dos afastamentos de servidores públicos estaduais em Alagoas, 2009. Epidemiol Serv Saude. 2012;21(3):505–14.

14.

Darwish

, Al-Zuhair

. Musculoskeletal pain disorders among secondary school Saudi female teachers. Pain Res Treat. 2013;878570.

15.

Van Houtte

, Claeys

, Wuyts

, Van Lierde

. The impact of voice disorders among teachers: vocal complaints, treatment-seeking behavior, knowledge of vocal care, and voice-related absenteeism. J Voice. 2011;25(5):570–75.

16.

Negrini

, Perron

, Corbière

. The predictors of absenteeism due to psychological disability: a longitudinal study in the education sector. Work. 2014;48(2):175–84.

17.

Moriana

, Herruzo

. Variables related to psychiatric sick leave taken by Spanish secondary school teachers. Work Stress. 2006;20(3):259–71.

18.

Medeiros

, Assunção

AÁ

, Barreto

. Absenteeism due to voice disorders in female teachers: a public health problem. Int Arch Occup Environ Health. 2012;85(8):853–64.

19.

Cutiva

LCC

, Burdorf

. Medical costs and productivity costs related to voice symptoms in Colombian teachers. J Voice.776-e. 2015;29(6):15–22.

20.

Dudenhöffer

, Claus

, Schöne

, Letzel

, Rose

. Sickness presenteeism of German teachers: prevalence and influencing factors. Teachers and Teaching. 2017;23(2):141–52.

21.

Alker

, Wang

, Pbert

, Thorsen

, Lemon

. Impact of school staff health on work productivity in secondary schools in Massachusetts. J Sch Health. 2015;85(6):398–404.

22.

Rosow

, Szczupak

, Saint-Victor

, Gerhard

, DuPont

, Lo

. The economic impact of vocal attrition in public school teachers in Miami-Dade County. Laryngoscope. 2016;126(3):665–71.

23.

Miller

, Murnane

, Willett

. Do teacher absences impact student achievement? Longitudinal evidence from one urban school district. Educ Eval Policy Anal. 2008;30(2):181–200.

24.

Glewwe

, Kremer

. Schools, teachers, and education outcomes in developing countries. Handbook of the Economics of Education. 2006;2:945–1017.

25.

Instituto Brasileiro de Geografia e Estatística (IBGE). IBGE cidades, Londrina. Available in https://cidades.ibge.gov.br/brasil/pr/londrina/panorama accessed November 14th, 2018.

26.

Johns

, Miraglia

. The reliability, validity, and accuracy of self-reported absenteeism from work: A meta-analysis. J Occup Health Psychol. 2015;20(1):1–14.

27.

Short

, Goetzel

, Pei

, Tabrizi

, Ozminkowski

, Gibson

, Dejoy

, Wilson

. How accurate are self-reports? An analysis of self-reported healthcare utilization and absence when compared to administrative data. J Occup Environ Med. 2009;51(7):786–96.

28.

Khanmohammadi

, Naghdi

, Ansari

, Hadian

, Kordi

, Hasson

. Single item presenteeism question: Reliability and validity of persian version in participants with low back pain. Work. 2018;61(3):339–46.

29.

Alves

MGDM

, Chor

, Faerstein

, Lopes

CDS

, Werneck

. Versão resumida da “job stress scale”: adaptação para o português. Rev Saude Publica. 2004;38(2):164–71.

30.

Matsudo

, Araújo

, Matsudo

, Andrade

, Oliveira

, Braggion

. Questionário internacional de atividade física (IPAQ): estudo de validade e reprodutibilidade no Brasil. Rev Bras Ativ Fís Saúde. 2001;6(2):5–18.

31.

Barros

ENC

, Alexandre

NMC

. Cross-cultural adaptation of the Nordic musculoskeletal questionnaire. Int Nurs Rev. 2003;50(2):101–8.

32.

Santos

KOB

, Araújo

, Oliveira

. Estrutura fatorial e consistência interna do Self-Reporting Questionnaire (SRQ-20) em população urbana. Cad Saude Publica. 2009;25(1):214–22.

33.

Carlotto

, Câmara

. Análise fatorial do Maslach Burnout Inventory (MBI) em uma amostra de professores de instituições particulares. Psicol Estud. 2004;9(3):499–505.

34.

Terwee

, Bot

, de Boer

, Van Der Windt

, Knol

, Dekker

, Boutter

, de Vet

. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol. 2007;60(1):34–42.

35.

Davis

. Instrument review: Getting the most from a panel of experts. Appl Nurs Res. 1992;5(4):194–97.

36.

Victora

, Huttly

, Fuchs

, Olinto

. The role of conceptual frameworks in epidemiological analysis: a hierarchical approach. Int J Epidemiol. 1997;26(1):224–7.

37.

Nagata

, Mori

, Ohtani

, Nagata

, Kajiki

, Fujino

, Matusa

, Loeppke

. Total Health-Related Costs Due to Absenteeism, Presenteeism, and Medical and Pharmaceutical Expenses in Japanese Employers. J Occup Environ Med. 2018;60(5):e273.

38.

Westcott

. Resistance training is medicine: effects of strength training on health. Curr Sports Med Rep. 2012;11(4):209–16.

39.

O’Connor

, Herring

, Caravalho

. Mental health benefits of strength training in adults. Am J Lifestyle Med. 2010;4(5):377–96.

40.

Behlau

, Zambon

, Guerrieri

, Roy

. Epidemiology of voice disorders in teachers and nonteachers in Brazil: prevalence and adverse effects. J Voice.665. e. 2012;26(5):9–18.

41.

Kidger

, Brockman

, Tilling

, Campbell

, Ford

, Araya

, King

, Gunnell

. Teachers’ wellbeing and depressive symptoms, and associated risk factors: A large cross sectional study in English secondary schools. J Affect Disord. 2016;192:76–82.

42.

Ferguson

, Frost

, Hall

. Predicting teacher anxiety, depression, and job satisfaction. J Teach and Learn. 2012;8(1):27–42.

43.

Lee

, Goodman

, Dandapani

, Kekahio

. Review of international research on factors underlying teacher absenteeism (REL 2015–087).Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory Pacific; 2015. Retrieved from http://ies.ed.gov/ncee/edlabs.

44.

Ervasti

, Kivimäki

, Puusniekka

, Luopa

, Pentti

, Suominen

, Ahola

, Vahtera

, Virtanen

. Students’ school satisfaction as predictor of teachers’ sickness absence: A prospective cohort study. Eur J Public Health. 2011;22(2):215–19.

45.

Wilson

, Douglas

, Lyon

. Violence against teachers: Prevalence and consequences. J Interpers Violence. 2011;26(12):2353–71.

46.

Hunter

, Reddy

. Noncommunicable diseases. N Engl J Med. 2013;369(14):1336–43.

47.

Wieser

, Riguzzi

, Pletscher

, Huber

, Telser

, Schwenkglenks

. How much does the treatment of each major disease cost? A decomposition of Swiss National Health Accounts. Eur J Health Econ. 2018;19(8):1149–61.

Predictors of health-related absenteeism,presenteeism and sick leave among brazilian elementary school teachers: A cross-sectional study

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1 Introduction

2 Methods

2.1 Population and study design

2.2 Sample

2.3 Data collection

2.4 Variables

2.4.1 Work-place factors

2.4.2 Individual factors

2.4.3 Information quality

2.5 Statistical analysis

3 Results

Table 1 Description of dependent variables among the sample (n = 519) Variables (days/year) Mean (SD) Prevalence≥1 Health-related absenteeism 3.19 (6.94) 64.4 Health-related presenteeism 4.86 (19.13) 42.8 Sick leave 4.78 (20.52) 27.6

5 Conclusion

Conflict of interest

References

Table 1
Description of dependent variables among the sample (n = 519)

Variables (days/year) Mean (SD) Prevalence≥1

Health-related absenteeism 3.19 (6.94) 64.4

Health-related presenteeism 4.86 (19.13) 42.8

Sick leave 4.78 (20.52) 27.6