Occupational Medicine 2014;64:279–286 Advance Access publication 18 February 2014 doi:10.1093/occmed/kqu009
Shift work and burnout among health care workers A. Wisetborisut,1 C. Angkurawaranon,1 W. Jiraporncharoen,1 R. Uaphanthasath1 and P. Wiwatanadate2 Department of Family Medicine, Faculty of Medicine, Chiang Mai University, Muang Chiang Mai, Chiang Mai 50200, Thailand, Department of Community Medicine, Faculty of Medicine, Chiang Mai University, Muang Chiang Mai, Chiang Mai 50200, Thailand.
1 2
Correspondence to: A. Wisetborisut, Department of Family Medicine, Faculty of Medicine, Chiang Mai University, 110, Intawaroros Road, Sripoom, Muang Chiang Mai, Chiang Mai 50200, Thailand. Tel: +66 53 945 462; fax: +66 53 946 581; e-mail: [email protected]
Aims
To identify and describe the association between shift work and burnout among health care workers.
Methods
A cross-sectional study of health care workers in Chiang Mai University Hospital, Thailand. Data were collected via an online self-answered questionnaire and included details of shift work and burnout. Burnout was measured by the Maslach Burnout Inventory (MBI).
Results
Two thousand seven hundred and seventy two health care workers participated, a 52% response rate. Burnout was found more frequently among shift workers than those who did not work shifts (adjusted odds ratio [aOR] 1.4, 95% confidence interval [CI]: 1.0–1.9). Among shift workers, over 10 years of being a shift worker was associated with increasing burnout (aOR 1.7, 95% CI: 1.2–2.6) and having 6–8 sleeping hours per day was associated with having less burnout (aOR 0.7, 95% CI: 0.5–0.9). Nurses who had at least 8 days off per month had lower odds of burnout compared with those with fewer than 8 days off (aOR 0.6, 95% CI: 0.5–0.8).
Conclusions Shift work was associated with burnout in this sample. Increased years of work as a shift worker were associated with more frequent burnout. Adequate sleeping hours and days off were found to be possible protective factors. Policies on shift work should take into account the potential of such work for contributing towards increasing burnout. Key words
Burnout; health care workers; hospital; nurses; shift work; sleep.
Introduction When people experience prolonged exposure to stress or frustration at work, they may develop exhaustion of physical or emotional strength, which has been termed burnout [1]. Although there is consistent agreement neither on the definition of burnout nor on its existence in the commonly used standard classification systems, burnout appears to have medical significance and it is diagnosed in some clinical practices [2]. The most widely used instrument to measure burnout, the Maslach Burnout Inventory (MBI), was developed by Maslach and Jackson [3], who defined burnout as a syndrome characterized by emotional exhaustion (EE),
depersonalization (DP) and a perceived lack of personal accomplishment (PA) [4]. Health care workers seem to be at greater risk of burnout than others [5,6]. This may arise from experience of a wide variety of job stressors including emotionally demanding patient contacts, exposure to death and dying, time pressure and work overload [7,8]. Various adverse effects of burnout have been described in health care workers, including depression, alcohol and drug misuse, insomnia, suicidal ideas [9–11] and physical symptoms such as back and neck pain [12]. Furthermore burnout may affect the quality of patient care, trigger early retirement and even increase the risk of medical errors [13–15]. There are many risk factors for burnout
© The Author 2014. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://occmed.oxfordjournals.org/ at Apollo Group on January 17, 2015
Background Burnout, defined as a syndrome derived from prolonged exposure to stressors at work, is often seen in health care workers. Shift work is considered one of the occupational risks for burnout in health care workers.
280 Occupational Medicine
Methods This study was a cross-sectional analytical survey of health care workers at Chiang Mai University Hospital in Chiang Mai, Thailand. The data were collected during the registered period of the annual hospital health check-up (January to February 2013). The study was approved by the Ethics Committee of the Faculty of Medicine, Chiang Mai University. Contracted health care workers who voluntarily joined the annual health check-up programme registered via the university Faculty of Medicine website, using their identification number. After logging in, a consent form appeared on the web page. If informed consent was given a study identification number was assigned and the original employee identification number discarded from the database. Thus, no individual employee could be identified or linked to a hospital identification number. A computer-based, online self-answered questionnaire was then completed. This included questions on demographic data (age, sex, education, status, income and occupation), shift work and burnout. In this study, a shift worker was defined as a person who answered ‘yes’ to the question ‘For the last 12 months, did you perform shift work?’. The evening shift was defined as a working period from 16:00 to 0:00 and night shift from 0:00 to 8:00. Total shift work comprised the number of evening shifts and night shifts worked per month, years of shift working, number of days off per month
and number of sleeping hours per night. Burnout was measured using the MBI-Thai version [4,5], which has been tested for validity and reliability [22]. The MBI comprises 22 questions with three subscales to measure three domains of burnout: emotional exhaustion (EE), depersonalization (DP) and personal accomplishment (PA). There are 9, 5 and 8 items in the EE, DP and PA domains, respectively. Scores of at least 27 in EE, 13 in DP and 39 in PA were defined as high scores in that domain. Subjects were considered to have burnout if a high score was recorded in any domain. In each question, participants had the option of ‘do not want to answer’. If participants chose not to answer in any domain, information on that domain was excluded from the study. The sample size calculation was based on a crosssectional design and a prevalence of burnout of 30%, which was chosen in the light of the findings of a survey of health care workers in France [9]. We estimated that at least 318 participants were needed to achieve 90% power at a 5% significance level in detecting an association between shift work and burnout. The Stata v.12 software program (StataCorp., Texas, USA) was used for analyses. Descriptive statistics were used for demographic data. Comparisons between shift-working health care workers and non-shift workers, and between those with burnout and those without were made using chi-squared tests. Significant demographic characteristics, P married > divorced), higher
income, higher education and occupation were all associated with burnout (Table 2). Shift work was associated with burnout (OR 1.8, 95% CI: 1.4–2.3). The prevalence of burnout in shift workers was 25% (95% CI: 23–27) compared with 15% (95% CI: 12–18) in the non-shift work group (Table 3). After adjustment for sex, age and occupation, shift work was still associated with burnout (adjusted odds ratio [aOR] 1.4, 95% CI: 1.0–1.9). In each MBI subscale, shift work had a significant association with EE (aOR 1.7, 95% CI: 1.2–2.3) but lacked evidence for association with DP (aOR 1.6, 95% CI: 0.8–4.2) and PA (aOR 0.5, 95% CI: 0.2–1.2) (Table 3). Further analysis of the details of shift work among current shift workers was performed, adjusting for sex, age, and occupation (Table 4). The odds of burnout among those working shifts for 5–10 years were higher than in those who had worked shifts for less than 5 years (aOR 1.1, 95% CI: 0.8–1.6). The adjusted OR for burnout was increased to 1.7 (95% CI: 1.2–2.6) for those who worked for more than 10 years. Longer sleeping hours were associated with lower odds of burnout. The odds of burnout in shift workers who slept 6–8 h/day was significantly lower than in those who slept less than 6 h/ day (aOR 0.7, 95% CI: 0.5–0.9). An increased number of shifts per month (≥24 compared with ≤16 shifts per month) was not associated with burnout (aOR 1.3, 95% CI: 0.8–1.9) and the same result was found for the relationship between number of night shifts per month (≥16 compared with ≤8 night shifts per month) and burnout (aOR 0.7, 95% CI: 0.3–1.7). In the subgroup analysis among nurses only after adjusting for sex and age, the relationship between associated factors and burnout were similar to those seen when all shift workers were included in the analysis, except for number of days off per month (Table 5). Nurses who
282 Occupational Medicine
Table 1. Demographic characteristics of shift-working health care workers compared with non-shift-working health care workers Characteristics
Non-shift work (n = 602)
238 (76) 1207 (70)
76 (24) 526 (30)
37.1 (9.4)
48.5 (7.7)
389 (96) 500 (87) 389 (68) 163 (34) 4 (21)
15 (4) 74 (13) 184 (32) 314 (66) 15 (79)
722 (80) 629 (65) 94 (57)
186 (20) 346 (45) 70 (43)
501 (73) 944 (70)
190 (27) 412 (30)
142 (88) 734 (88) 338 (66) 231 (43)
19 (12) 98 (12) 176 (34) 309 (57)
14 (40) 30 (86) 724 (68) 324 (66) 19 (59) 334 (85)
21 (60) 5 (14) 339 (32) 164 (34) 13 (41) 60 (15)
P values NS
Shift work and burnout among health care workers A. Wisetborisut,1 C. Angkurawaranon,1 W. Jiraporncharoen,1 R. Uaphanthasath1 and P. Wiwatanadate2 Department of Family Medicine, Faculty of Medicine, Chiang Mai University, Muang Chiang Mai, Chiang Mai 50200, Thailand, Department of Community Medicine, Faculty of Medicine, Chiang Mai University, Muang Chiang Mai, Chiang Mai 50200, Thailand.
1 2
Correspondence to: A. Wisetborisut, Department of Family Medicine, Faculty of Medicine, Chiang Mai University, 110, Intawaroros Road, Sripoom, Muang Chiang Mai, Chiang Mai 50200, Thailand. Tel: +66 53 945 462; fax: +66 53 946 581; e-mail: [email protected]
Aims
To identify and describe the association between shift work and burnout among health care workers.
Methods
A cross-sectional study of health care workers in Chiang Mai University Hospital, Thailand. Data were collected via an online self-answered questionnaire and included details of shift work and burnout. Burnout was measured by the Maslach Burnout Inventory (MBI).
Results
Two thousand seven hundred and seventy two health care workers participated, a 52% response rate. Burnout was found more frequently among shift workers than those who did not work shifts (adjusted odds ratio [aOR] 1.4, 95% confidence interval [CI]: 1.0–1.9). Among shift workers, over 10 years of being a shift worker was associated with increasing burnout (aOR 1.7, 95% CI: 1.2–2.6) and having 6–8 sleeping hours per day was associated with having less burnout (aOR 0.7, 95% CI: 0.5–0.9). Nurses who had at least 8 days off per month had lower odds of burnout compared with those with fewer than 8 days off (aOR 0.6, 95% CI: 0.5–0.8).
Conclusions Shift work was associated with burnout in this sample. Increased years of work as a shift worker were associated with more frequent burnout. Adequate sleeping hours and days off were found to be possible protective factors. Policies on shift work should take into account the potential of such work for contributing towards increasing burnout. Key words
Burnout; health care workers; hospital; nurses; shift work; sleep.
Introduction When people experience prolonged exposure to stress or frustration at work, they may develop exhaustion of physical or emotional strength, which has been termed burnout [1]. Although there is consistent agreement neither on the definition of burnout nor on its existence in the commonly used standard classification systems, burnout appears to have medical significance and it is diagnosed in some clinical practices [2]. The most widely used instrument to measure burnout, the Maslach Burnout Inventory (MBI), was developed by Maslach and Jackson [3], who defined burnout as a syndrome characterized by emotional exhaustion (EE),
depersonalization (DP) and a perceived lack of personal accomplishment (PA) [4]. Health care workers seem to be at greater risk of burnout than others [5,6]. This may arise from experience of a wide variety of job stressors including emotionally demanding patient contacts, exposure to death and dying, time pressure and work overload [7,8]. Various adverse effects of burnout have been described in health care workers, including depression, alcohol and drug misuse, insomnia, suicidal ideas [9–11] and physical symptoms such as back and neck pain [12]. Furthermore burnout may affect the quality of patient care, trigger early retirement and even increase the risk of medical errors [13–15]. There are many risk factors for burnout
© The Author 2014. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://occmed.oxfordjournals.org/ at Apollo Group on January 17, 2015
Background Burnout, defined as a syndrome derived from prolonged exposure to stressors at work, is often seen in health care workers. Shift work is considered one of the occupational risks for burnout in health care workers.
280 Occupational Medicine
Methods This study was a cross-sectional analytical survey of health care workers at Chiang Mai University Hospital in Chiang Mai, Thailand. The data were collected during the registered period of the annual hospital health check-up (January to February 2013). The study was approved by the Ethics Committee of the Faculty of Medicine, Chiang Mai University. Contracted health care workers who voluntarily joined the annual health check-up programme registered via the university Faculty of Medicine website, using their identification number. After logging in, a consent form appeared on the web page. If informed consent was given a study identification number was assigned and the original employee identification number discarded from the database. Thus, no individual employee could be identified or linked to a hospital identification number. A computer-based, online self-answered questionnaire was then completed. This included questions on demographic data (age, sex, education, status, income and occupation), shift work and burnout. In this study, a shift worker was defined as a person who answered ‘yes’ to the question ‘For the last 12 months, did you perform shift work?’. The evening shift was defined as a working period from 16:00 to 0:00 and night shift from 0:00 to 8:00. Total shift work comprised the number of evening shifts and night shifts worked per month, years of shift working, number of days off per month
and number of sleeping hours per night. Burnout was measured using the MBI-Thai version [4,5], which has been tested for validity and reliability [22]. The MBI comprises 22 questions with three subscales to measure three domains of burnout: emotional exhaustion (EE), depersonalization (DP) and personal accomplishment (PA). There are 9, 5 and 8 items in the EE, DP and PA domains, respectively. Scores of at least 27 in EE, 13 in DP and 39 in PA were defined as high scores in that domain. Subjects were considered to have burnout if a high score was recorded in any domain. In each question, participants had the option of ‘do not want to answer’. If participants chose not to answer in any domain, information on that domain was excluded from the study. The sample size calculation was based on a crosssectional design and a prevalence of burnout of 30%, which was chosen in the light of the findings of a survey of health care workers in France [9]. We estimated that at least 318 participants were needed to achieve 90% power at a 5% significance level in detecting an association between shift work and burnout. The Stata v.12 software program (StataCorp., Texas, USA) was used for analyses. Descriptive statistics were used for demographic data. Comparisons between shift-working health care workers and non-shift workers, and between those with burnout and those without were made using chi-squared tests. Significant demographic characteristics, P married > divorced), higher
income, higher education and occupation were all associated with burnout (Table 2). Shift work was associated with burnout (OR 1.8, 95% CI: 1.4–2.3). The prevalence of burnout in shift workers was 25% (95% CI: 23–27) compared with 15% (95% CI: 12–18) in the non-shift work group (Table 3). After adjustment for sex, age and occupation, shift work was still associated with burnout (adjusted odds ratio [aOR] 1.4, 95% CI: 1.0–1.9). In each MBI subscale, shift work had a significant association with EE (aOR 1.7, 95% CI: 1.2–2.3) but lacked evidence for association with DP (aOR 1.6, 95% CI: 0.8–4.2) and PA (aOR 0.5, 95% CI: 0.2–1.2) (Table 3). Further analysis of the details of shift work among current shift workers was performed, adjusting for sex, age, and occupation (Table 4). The odds of burnout among those working shifts for 5–10 years were higher than in those who had worked shifts for less than 5 years (aOR 1.1, 95% CI: 0.8–1.6). The adjusted OR for burnout was increased to 1.7 (95% CI: 1.2–2.6) for those who worked for more than 10 years. Longer sleeping hours were associated with lower odds of burnout. The odds of burnout in shift workers who slept 6–8 h/day was significantly lower than in those who slept less than 6 h/ day (aOR 0.7, 95% CI: 0.5–0.9). An increased number of shifts per month (≥24 compared with ≤16 shifts per month) was not associated with burnout (aOR 1.3, 95% CI: 0.8–1.9) and the same result was found for the relationship between number of night shifts per month (≥16 compared with ≤8 night shifts per month) and burnout (aOR 0.7, 95% CI: 0.3–1.7). In the subgroup analysis among nurses only after adjusting for sex and age, the relationship between associated factors and burnout were similar to those seen when all shift workers were included in the analysis, except for number of days off per month (Table 5). Nurses who
282 Occupational Medicine
Table 1. Demographic characteristics of shift-working health care workers compared with non-shift-working health care workers Characteristics
Non-shift work (n = 602)
238 (76) 1207 (70)
76 (24) 526 (30)
37.1 (9.4)
48.5 (7.7)
389 (96) 500 (87) 389 (68) 163 (34) 4 (21)
15 (4) 74 (13) 184 (32) 314 (66) 15 (79)
722 (80) 629 (65) 94 (57)
186 (20) 346 (45) 70 (43)
501 (73) 944 (70)
190 (27) 412 (30)
142 (88) 734 (88) 338 (66) 231 (43)
19 (12) 98 (12) 176 (34) 309 (57)
14 (40) 30 (86) 724 (68) 324 (66) 19 (59) 334 (85)
21 (60) 5 (14) 339 (32) 164 (34) 13 (41) 60 (15)
P values NS
