1282 Relationship Between Overtime and Mental Fatigue • Beckers et al Working Overtime Hours: Relations with Fatigue, Work Motivation, and the Quality of Work Debby G. J. Beckers, MA vertime work is a common phenom- Dimitri van der Linden, PhD enon all over the world. In Japan, workweeks that exceed 60 hours are Peter G.W. Smulders, PhD O no exception. It is well documented Michiel A. J. Kompier, PhD that this extreme type of overtime Marc J. P. M. van Veldhoven, PhD work can have severe health conse- quences and may eventually lead to Nico W. van Yperen, PhD death (“karoshi”).1,2 In Europe, de- tailed representative (inter)nationally Objectives: We sought to better understand the relationship between comparable data on the prevalence of overtime and mental fatigue by taking into account work motivation overtime work are scarce. A notable and the quality of overtime work and studying theoretically derived exception is a study by Merllie´ and subgroups. Methods: We conducted a survey-study among a represen- Paoli,3 which shows that 20% of the tative sample of the Dutch full-time workforce (n ϭ 1807). The employees work, on average, more prevalence of overtime work and the associations between overtime and than 44 hours a week. Occupational job demands, job variety, decision latitude, fatigue, and work motiva- health research demonstrates that tion was studied through descriptive statistics. We used MANCOVA overtime work may be associated (covariates: age, gender, salary level) to compare six overtime-fatigue with health problems such as high subgroups with respect to work motivation and job characteristics. blood pressure, increased risk of car- Results: A total of 67% of the respondents worked overtime (mean, 3.5 diovascular disease, and diabetes. hours). Overtime workers appeared to be nonfatigued, motivated workers Furthermore, overtime appears to be with favorable work characteristics. MANCOVA revealed no significant related to mental health problems such as depression and psychological overtime-fatigue interaction. Conclusions: Moderate overtime is common 4,5,6 among Dutch workers, who seem to be happy workers with attractive jobs distress. Various researchers4,7 have argued rather than fatigued employees. (J Occup Environ Med. 2004;46: that the relationship between over- 1282–1289) time and health problems may be understood in terms of a chronic imbalance between effort expendi- ture at work and the opportunities to recover from work. After all, over- time not only leads to more effort investment but also to less time for recovery after work.8 Accordingly, From the Department of Work and Organizational Psychology, Radboud University Nijmegen one would expect a relatively (Ms Beckers, Dr van der Linden, Dr Kompier, Dr Van Yperen); TNO Work & Employment, strong relationship between the Hoofddorp (Dr Smulders); University of Tilburg, Tilburg (Dr van Veldhoven); and University of number of overtime hours and a Groningen, Groningen, The Netherlands (Dr van Yperen). recovery-indicator such as mental Address correspondence to: Debby Beckers, Department of Work and Organizational Psychology, Radboud University Nijmegen, P.O. Box 9104, 6500 HE Nijmegen, The Netherlands; E-mail: fatigue. Ample studies, however, [email protected]. indicated that the relationship be- Copyright © by American College of Occupational and Environmental Medicine tween overtime and mental fatigue DOI: 10.1097/01.jom.0000147210.95602.50 is either significant but low or not JOEM • Volume 46, Number 12, December 2004 1283 significant at all (r ranging from geneous, and representative sample hours, only respondents who re- 0.00–0.12).5,9,10 of employees with enough contrast ported more than 32 contracted work We believe that at least five pos- with respect to overtime hours and hours a week were included in our sible explanations for the weak over- fatigue; 3) should take into account study. Consequently, the final sam- all association between overtime and the quality of overtime work; 4) ple consisted of 1807 persons (77% mental fatigue can be distinguished: should address potential positive out- males, 23% females). Respondents 1) Restriction of range in the amount comes such as work motivation as ranged in age from 16 to 63 (mean, of overtime: to demonstrate associa- well; and 5) should differentiate be- 41.7 years) and worked on average tions between overtime work and tween theoretically specified sub- 38.5 work hours on contract. The fatigue, enough variance in overtime groups. sample may be considered to be rep- is necessary; a study without contrast The current study meets these de- resentative for the total Dutch full- in amount of overtime work is un- sign-demands because it: 1 and 2) time working population in terms of likely to yield a correlation between uses a large and representative sam- age, salary level, and contracted overtime and any outcome variable; ple of the Dutch workforce; 3) in- work hours.16 2) Restriction of range in fatigue: cludes the three most important psy- employed workers mostly constitute chosocial job characteristics, that is, Measures a relatively healthy subset of the total job demands, job variety, and deci- Overtime Work. Overtime hours population. It may well be that over- sion latitude14 as indicators of the were assessed with the following time workers who persevere in work- quality of (overtime) work; 4) not item: “On average, how many hours ing overtime in turn constitute a only considers a possible negative a week do you work overtime?” healthy and nonfatigued subset of the health indicator (fatigue), but also a (paid AND unpaid overtime work; total population of (overtime) work- positive indicator (work motivation); include work you execute at home; ers. This might lead to an underesti- and 5) differentiates between sub- DONЈT include your commuting mation of the true relationship be- groups of employees, with different time). tween overtime and fatigue; 3) Not overtime-fatigue profiles. In this Quality of (overtime) work. Job paying attention to the quality of context, the aim of the current re- demands were measured with five overtime work4–6: whereas previous search was to answer two related items from the “Job demands” scale research mostly studied overtime questions: of the Job Content Questionnaire.17 from a quantitative perspective 1. What is the prevalence of over- A sample question is: “Do you have (number of hours worked overtime), time work, and to what extent is to work very fast?” Job variety was there are strong indications that it is the amount of overtime work re- measured with five items from the also the quality of overtime work lated to (a) fatigue, (b) work mo- scale “Skill discretion” of the JCQ, that counts. A recent study, for ex- tivation, and (c) the quality of for example: “Do you get to do a ample, suggests that overtime work (overtime) work (job demands, variety of different things on your might only be related to health prob- job variety, decision latitude)? job?” Decision latitude was mea- lems under adverse psychosocial sured using five items from the “De- 11 2. To what extent do subgroups of working conditions; 4) Overtime fatigued and nonfatigued (non) cision latitude” scale of the JCQ. An work may be fun: not only may overtime workers differ with re- exemplary question is: “Do you have high-quality overtime work be unre- spect to quality of work (job de- the freedom to decide how to do your lated to health problems (negative mands, job variety, decision lati- job?” The items of these three scales outcomes), it may also have positive tude) and work motivation? are scored on a four-point Likert outcomes, that is, contribute to psy- scale, ranging from 1, “never,” to 4, chological health, work motivation, Materials and Methods “always.” Higher scores on these and work satisfaction;12,13 5) Weak scales indicate higher (quantitative) and insignificant associations be- Sample and Procedures workload, more job variety and more tween overtime work and fatigue in The data were collected as part of decision latitude, respectively. Cron- the total study population may con- a questionnaire study on the work bach’s ␣ were 0.83, 0.74, and 0.83, ceal different patterns within mean- situation of Dutch employees.15 A respectively. ingful subgroups, for example, sub- random sample of 8000 persons was Fatigue. Fatigue was measured groups with different overtime- drawn from the total Dutch work- with five items from the Dutch ver- fatigue profiles. force, and 4009 workers (50%) actu- sion of the Maslach Burnout Inven- To assess the relationship between ally completed the questionnaire. To tory.18 An example item is: “I feel overtime work and fatigue, these five relate potential differences in fatigue used up at the end of the workday” issues need to be taken into account. and work motivation to differences (1 ϭ “never,” 7 ϭ “every day”). Preferably, such a study: 1 and 2) in overtime hours per se and not to Cronbach’s ␣ was 0.91. Norm-scores should be based on a large, hetero- the number of contractual work were available for this measure, which made it possible to qualify a respondents’ score as being “high” or “low” (see below). Work Motivation. JOEM • Volume 46, Number 12, December 2004 1285 job variety and decision latitude. Fi- overtime/nonfatigued; n ϭ 309; 3) (n ϭ 652) did not differ significantly nally, higher levels of job motivation high overtime/nonfatigued; n ϭ 103; with regard to overtime hours went together with lower levels of 4) no overtime/fatigued; n ϭ 171; 5) (F[1,1104] ϭ 0.47; P ϭ 0.49) nor did fatigue, and vice versa. low overtime/fatigued; n ϭ 293; 6) the overtime groups (no/low/high) high overtime/fatigued; n ϭ 88. differ significantly on fatigue Question 2: Differences Between The subgroups obviously differed (F[2,1663] ϭ 0.40; P ϭ 0.67), which Fatigued and Nonfatigued (Non) with respect to number of overtime is in accordance with the above re- Overtime Workers hours and fatigue in accordance ported not significant overall corre- lation between overtime and fatigue.