You are here

Article Text

Article menu

Download PDFPDF

Workplace
Original article
Effect of working hours and precarious employment on depressive symptoms in South Korean employees: a longitudinal study
  1. Woorim Kim1,2,
  2. Eun-Cheol Park2,3,
  3. Tae-Hoon Lee1,2,
  4. Tae Hyun Kim2,4
  1. 1Department of Public Health, Graduate School, Yonsei University, Seoul, Republic of Korea
  2. 2Institute of Health Services Research, Yonsei University, Seoul, Republic of Korea
  3. 3Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea
  4. 4Graduate School of Public Heath, Yonsei University, Seoul, Republic of Korea
  1. Correspondence to Professor Tae Hyun Kim, Graduate School of Public Health and Institute of Health Services Research, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea; thkim{at}yuhs.ac

Abstract

Objectives Long working hours and precarious employment are relatively common in South Korea. Since both can impact on mental health, this study examined their independent and combined effects on depressive symptoms of employees.

Methods Data were from the Korean Welfare Panel Study (KOWEPS), 2010–2013. A total of 2733 full-time employees without depressive symptoms were analysed. Hierarchical logistic regression models were used to investigate the effect of the number of working hours per week (eg, 35–40 hours, 41–52 hours, 53–68 hours, and >68 hours) and employment status (permanent vs precarious employment) on depressive symptoms, measured using the Center for Epidemiological Studies Depression (CES-D) 11 scale.

Results Compared with individuals working 35–40 hours/week, employees working above 68 hours (OR 1.57, 95% CI 1.05 to 2.34) had higher odds of depressive symptoms after full adjustment. Similarly, precarious employees (OR 1.34, 95% CI 1.02 to 1.75) showed worse mental health than permanent employees. In the combined effect model, employees in precarious employment who worked above 68 hours/week (OR, 2.03 95% CI 1.08 to 3.83) exhibited the highest odds of depressive symptoms compared with permanent employees working 35 to 40 hours/week.

Conclusions Long work hours and precarious employment status were associated with higher odds of depressive symptoms. The findings highlight the importance of monitoring and addressing the vulnerable groups of employees to reduce the mental health burden of economically active individuals.

  • working hours
  • precarious employment
  • depressive symptoms
  • East Asian societies

https://doi.org/10.1136/oemed-2016-103553

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    What this paper adds

    • Long working hours and precarious employment have been associated with depressive symptoms but most studies investigating this association have been cross-sectional and few have focused on East Asia.

    • This study investigated the independent and combined effect of work hours and permanent versus precarious employment status on depressive symptoms among full-time employees in South Korea, which ranks second among Organisation for Economic Co-operation and Development (OECD) countries in the number of working hours and first in the number of precarious employees.

    • Employees working above 68 hours/week showed higher odds of depressive symptoms compared with individuals working the standard 35–40 hours/week. Precarious workers also exhibited higher odds of depressive symptoms than permanent employees.

    • When investigating working hours and permanent versus precarious employment status concurrently, depressive symptoms were more common in permanent and precarious workers working above 68 hours/week. This finding is noteworthy as it demonstrates the adverse mental health effects of working above the legal maximum of 68 hours/week in permanent and precarious employees.

    Introduction

    The impact of long working hours on the mental health of workers has gained noticeable research interest as depression is known to contribute to work disability and reduced quality of life in working aged people.1 Previous studies have documented the adverse effects of long working hours on depressive state, anxiety, sleep condition and coronary heart disease among Japanese, European, Canadian and Australian participants.2 Studies investigating the effect of working hours on depressive symptoms in East Asian countries are relatively scarce and the majority of the researches present focuses on Japan. Yet the relationship between long work hours and depressive symptoms in South Korea requires significant addressing because South Korea ranks second among Organisation for Economic Co-operation and Development (OECD) countries in working hours, at an average of 45.9 hours/week. The International Labour Organization (ILO) reports that around 49.5% of South Korean workers work above 48 hours/week, which is noticeably higher than the world average of 22%.3 Moreover, mood disorders rank second among diseases compensated for by the South Korean Industrial Accident Compensation Insurance (IACI) and South Korea has the highest suicide rate among OECD countries, of which over 90% can be attributed to emotional or psychiatric status including depression.4

    Although few studies have reported the relationship between long working hours and depressive symptoms in South Korea, previous studies have been cross-sectional in design and often only included participants engaging in specific sectors of the industry instead of the general labour force.5–7 In addition, most studies did not measure depressive symptoms using instruments that have been tested for validity and reliability but have depended on responses in a binary or a Likert scale format based on a single question, which is likely to be subjective.5–7 Additionally, the effect of long work hours and type of employment, referring to either permanent or precarious employment, on depression has not been investigated concurrently despite the fact that work hours and employment type may be related as precarious workers often report lower wages than permanent workers and are generally more prone to accepting overtime work on request.8 Whether long working hours exert similar patterns in permanent and precarious employees might hence be an important factor in investigating depressive symptoms. Therefore, the objective of this study was to examine the effect of long working hours on depressive symptoms of full-time workers using nationally representative longitudinal data from South Korea and, in addition, further analyse how the number of working hours and employment status together impacts the depressive symptoms of economically active individuals.

    Materials and methods

    Study population and data

    The Korean Welfare Panel Study (KOWEPS) is a panel survey conducted annually by the Korean Institute for Health and Social Affairs in conjunction with the Social Welfare Research Institute of Seoul National University on a nationally representative sample of South Korean households.9 Households are selected using a stratified multistage probability design. Data are collected through face-to-face interviews conducted by trained interviewers using structured questionnaires and cover topics including social service needs, healthcare usage patterns, economic and demographic background, sources of income, and subjective emotional and behavioural health status. Data are collected on all household members and interviews are conducted with individuals aged 15 or above. The KOWEPS includes multiple individuals from the same household and these individuals are repeatedly measured at each wave of data collection.

    This study was carried out using data from the 2010 to 2013 panel. Of the 3655 waged employees working 35 hours or above per week in 2010, individuals with depressive symptoms (Center for Epidemiologic Studies Depression Scale (CES-D) score cut-off value of 16) were excluded. Non full-time workers, referring to part-time individuals working <35 hours/week, were also not included in the analysis. After the exclusion of 2010 depressive symptom cases, 2733 waged workers remained to form the baseline population (figure 1).

    Figure 1
    Figure 1

    Flow of study participants.

    Measures

    Outcome variable

    The outcome variable of this study was depressive symptoms, measured using the Korean version of the CES-D 11 scale by the KOWEPS. The CES-D 11 includes the following 11 items based on the past week on a four-point scale (0: ≤1; 1: 2–3; 2: 4–; and 3: ≥6 days/week): (a) ‘No appetite’; (b) ‘I felt that I was doing generally well’; (c) ‘I felt quite depressed’; (d) ‘I felt difficulty in everything I did’; (e) ‘I could not sleep well’; (f) ‘I felt lonely’; (g) ‘I went on without much complaints’; (h) ‘I felt that people were treating me coldly’; (i) ‘My heart felt sad’; (j) ‘I felt that people disliked me’ and (i) ‘I was unable to have the courage to carry out something’. Scores are transformed, multiplied by (20/11) to be comparable to the standard CES-D 20 score.9 The CES-D 11 is akin to the CES-D 20-item standard version and a cut-off score of 16 has been widely used to indicate probable depression.10–12 The Korean version of the CES-D has been validated, with the α coefficient rating 0.9098 for the general population.13 Previous studies have measured depressive symptoms in the South Korean population using the CES-D scale based on the cut-off score of 16 and 25, in which a score of 16 denotes probable depression and a score of 25 definite depression.12 The KOWEPS recommends a score of 16 to indicate depression9 and this standard was also used in this study to identify the new onset of depressive symptoms.

    Number of working hours

    The number of working hours were categorised into 35–40, 41–52, 53–68, and >68 hours based on self-responses provided by the study participants. Hence, the cut-off of working hours was at 40, 52 and 68 hours. This categorisation was based on the Korean Labor Standard Act, which defines 40 hours as the number of legal working hours per week and 52 hours as the maximum overtime hour limit. However, the Labor Standard Act also allows an additional 16 hours of holiday work, which refers to work conducted during legal holidays such as Sundays. Thus, employees can legally work up to 68 hours/week and the 40, 52 and 68 hour classification was applied accordingly.

    Precarious employment

    Employment was categorised into permanent and precarious (fixed-term contract) status. Permanent employment refers to full-time workers directly hired by their employers without a fixed-term contract and generally includes respondents with relatively high job security. All waged workers who were not permanent workers were classified as precarious workers. Precarious workers denote temporary workers under a fixed-term contract and can include subcontracted or dispatched workers. Fixed-term contracts can last up to a maximum of 2 years in South Korea and are subject to renewal afterwards. Since this study was limited to only permanent or precarious employees, self-employed individuals, defined as workers with their own business regardless of scale and type, were excluded from the analysis.

    Covariates

    Demographic, socioeconomic and health related covariates were included in this study. The covariates were age (20–29, 30–39, 40–49, or 50–59), gender (male or female), education level (middle school, high school or university graduate), equalised household income (quartiles), marital status (single or married), employment status (permanent or precarious), job satisfaction level (low, medium or high) and chronic disease status (none or 1 or above). Equalised household income was obtained by dividing household income by the square root of the number of household members, which is the standard method used in South Korea recommended by the OECD.14 Measuring equalised household income allows income comparability between households of different sizes. Job satisfaction level was measured based on the question ‘How satisfied are you with your occupation?’ The available answers were based on a five-point Likert scale ranging from (1) very dissatisfied, (2) somewhat dissatisfied, (3) mediocre, (4) somewhat satisfied to (5) very satisfied. In this study, the very dissatisfied and somewhat dissatisfied categories were classified into the low, the mediocre category into the medium, and the somewhat satisfied and very satisfied into the high job satisfaction level categories. Chronic disease included chronic hepatitis, diabetes, hypertension, chronic thyroid diseases, cardiovascular diseases, chronic bronchitis, glaucoma, chronic renal failure, asthma and hyperlipidaemia.

    Statistical analysis

    In order to examine the study participants' general characteristics, χ2 test was performed to compare differences between groups. Depressive symptoms were expressed as ORs and their 95% CIs. The KOWEPS data used in this study were hierarchically organised, in which households are selected using a stratified multistage probability design with multiple individuals from the same households being included. Furthermore, the KOWEPS data set is longitudinal in nature and can contain annual repeated measurements of the same individuals. Hierarchical logistic regression models were fitted using the generalized linear mixed models (GLIMMIX) procedure, which is adapted for hierarchical modelling of multiple exposures with a dichotomous or polytomous outcome.15 Multilevel models can properly account for the correlated structure of hierarchical data and controls for the non-independence of observations within groups in investigating individual level outcomes.15 Baseline cases of depressive symptoms were excluded from analysis. Analysis was conducted in three phases with model 1 adjusting for the exposure variable (working hours or permanent vs precarious employment) and sociodemographic factors (age, gender, education level, equalised household income and marital status), model 2 further adjusting for chronic disease, job satisfaction and year, and model 3 additionally adjusting for the remaining exposure variable (working hours or permanent vs precarious employment). The combined effect of the number of working hours and permanent versus precarious employment status was tested using an interaction term. The calculated p values in this study were all two-sided and considered significant at p<0.05. All analysis was carried out using the SAS software, V.9.4 (SAS Institute, Cary, North Carolina, USA).

    Results

    The general characteristics of the study observations are shown in table 1. Study participants were economically active at the baseline without depressive symptoms. The results show 338 (5.0%) observations with depressive symptoms. The largest proportion of observations were in the 35–40 hours/week group and the proportion of permanent workers exceeded that of precarious workers. Frequency of depressive symptoms generally increased as the number of working hours per week increased (35–40: 4.0%; 41–52: 5.1%; 53–68: 5.3%; and over 68 hours: 9.5%). Frequency of depressive symptoms was higher among precarious workers (8.0%) compared with permanent workers (3.4%).

    View this table:
    Table 1

    General characteristics of study observations, 2011–2013

    The results of the multilevel logistic regression models are depicted in tables 2 and 3, presenting the prospective association between working hours and permanent or precarious employment status on depressive symptoms. Results are presented using three models, with model 1 adjusting for the primary exposure variable and socioeconomic factors. Model 2 further adjusts from model 1 and incorporates job satisfaction level, chronic disease status and year. Finally, model 3 is a fully adjusted model which adjusts for the additional primary exposure variable, that is, permanent versus precarious employment status or the number of working hours per week. Table 2 shows the results with the number of working hours as the primary exposure variable. Compared with the 35–40 working hours per week group, individuals working above 68 hours/week had statistically significant higher odds of depressive symptoms (model 1: OR 1.92, 95% CI 1.30 to 2.85; model 2: OR 1.69, CI 1.14 to 2.51; model 3: OR 1.57, CI 1.05 to 2.34). Table 3 shows the results with permanent versus precarious employment status as the main exposure variable. When setting the permanent employment contract group as reference, the odds of expressing depressive symptoms increased among precarious workers in all three models (model 1: OR 1.66, CI 1.28 to 2.15; model 2: OR 1.39, CI 1.06 to 1.81; and model 3: OR 1.34, CI 1.02 to 1.75).

    View this table:
    Table 2

    Factors associated with depression in study participants

    View this table:
    Table 3

    Factors associated with depression in study participants

    The results of the multilevel logistic regression models on depressive symptoms based on the combined effect of the number of working hours and permanent versus precarious employment status are shown in figure 2. The p value for interaction showed marginal significance (p value 0.065). When placing the 35–40 hours permanent employment contract group as reference, the >68 hours permanent employment group (OR 1.83, CI 1.12 to 3.00) and >68 hours precarious employment group (OR 2.03, CI 1.08 to 3.83) showed higher odds of depressive symptoms after full adjustment of all covariates.

    Figure 2
    Figure 2

    Results analysing the combined effect of the number of working hours by permanent versus precarious employment.

    Discussion

    The result of this study exhibits an association between long work hours and increased odds of depressive symptoms. Study participants working above 68 hours/week showed statistically significant higher odds ratios of depressive symptoms in all three models. This finding is consistent with previous studies that have been conducted in the UK, Canada, the Netherlands, Japan and Korea, in which extended work hours were generally related with depressive states.2 A study by Virtanen et al1 revealed an around 17% increase in depressive symptoms with every 10 hour increase in work hours. As for studies performed based in Korea, cross-sectional studies confirmed an association between long work hours and depression using the WHO (five) well-being index,6 binary question-based response5 and the Beck Depression Inventory (BDI).7 This study is the first to prospectively recognise a relationship between long working hours and depressive symptoms using longitudinal data in South Korea, enhancing evidence on this subject. The results are particularly noteworthy because the negative mental health effects of working above the legal maximum of 68 hours/week are demonstrated. In fact, it is probable that a substantial number of employees work additional hours for no pay and the legal number of working hours may not be applied.16 This study denotes important implications by revealing that working above the legal limit can have negative mental health effects, inferring the prominence of addressing this subject. Furthermore, since this study was carried out using nationally representative data, the findings can be generalised to the South Korean population and conceivably to other East Asian populations sharing similar occupational characteristics.

    Regarding the relationship between long work hours and depressive symptoms, stress has been proposed as a plausible explanation. First, since extended work hours often lead to time poverty and preclusion from family activities, work–family conflict can result, which in turn leads to higher stress and depression.17–19 Evidence has suggested that longer working hours are associated with work–family conflict, which contributes to less family interaction, conflict in marriage, less awareness about children's experiences and less participation in housework.20 In fact, when South Korea lowered the number of legal working hours from 44 to 40 hours in 2002, ‘enhanced family connectedness and quality of life’ has been reported as the foremost positive aspect.21 As the number of dual income families continues to increase in South Korea, more individuals will most likely experience time pressure as they become responsible for providing care to children and elderly dependants.22 Second, long work hours can lead to intermediary conditions, including fatigue and stress, which lead to occupational illnesses.23 This link is important because occupational illnesses include depression, with mood disorders ranking second among the number of compensations paid by the IACI of South Korea.24 ,25 Specifically, shift time workers may be particularly impacted by this aspect as the positive association between long working hours and occupational injuries was reported to be noticeably high for later shift working times.26 Therefore, considering that around 20% of South Korean companies implement shift work and extended work hours, stress can be an important mechanism for the decline in mental health resulting from long work hours.8

    Apart from working hours, precarious employment status also showed a statistically significant association with increased depressive symptoms. Depressive symptoms were more common in precarious workers than permanent workers. These results are in line with previous studies that have established a relationship between temporary work and depression.27 Specifically, a South Korean study concluded using self-reports that non-standard employees experience poorer mental health.28 The findings of this study also confirm that precarious employment is related to higher prospects of depressive symptoms. This is significant because evidence is added from South Korea, a country that currently ranks first among OECD countries in the number of precarious employees due to its flexible labour market.29 Such tendencies may result due to the structural conditions of the South Korean market where precarious workers often receive lower salaries and gain fewer opportunities for promotions.30 Precarious workers may also find it comparatively difficult to work under the maintenance of minimum labour standards as supervisory regimes are often less rigorously applied, resulting in increased instability allied with mental health declines.31

    This study also explored the combined effect of the number of working hours and type of employment contract. Working hours and permanent or precarious employment status exhibited borderline significance, with permanent and precarious workers working above 68 hours having higher odds of depressive symptoms compared with permanent workers working 35–40 hours/week. The comparatively higher odds of depressive symptoms among precarious workers than permanent workers in the over 68 hours group may be due to the fact that whereas long working hours and precarious employment are independently related to worse mental health outcomes, temporary workers are generally more prone to accepting overtime work on request as they receive lower wages and find difficulties in earning adequate wages.8 Nevertheless, results of this study show the independent effects of work hours and precarious employment on depressive symptoms, but their combined effect did not add up to the individual effects. This leads to the conclusion that exhaustive working hours can exert adverse mental health effects both in permanent and precarious employees, indicating a need to protect all workers from working extreme hours as working above 68 hours can increase the odds of depressive symptoms.

    This study has some limitations. First, the number of working hours was measured and classified based on self-reports, meaning that it may have been subjective. Second, there may have been a healthy survivor effect as individuals with severe depressive symptoms are comparatively unlikely to be able to work and have difficulties in maintaining full-time permanent occupations. However, individuals with baseline depressive symptoms based on the CES-D cut-off were excluded from the analysis to partially overcome this limitation. Third, although the CES-D 11 has been tested for validity and reliability, responses may still have been subjective. Fourth, part-time employees were not included as this study examined the effect of extended working hours compared with the standard 35–40 working hours per week. Finally, study participants were limited to employees and excluded employers or self-employed individuals. This was to enhance homogeneity of the study population because long work hours may infer different meanings for employers and the self-employed.

    Conclusion

    Long work hours were associated with higher odds of depressive symptoms in employees and precarious employment status also showed similar effects. In the combined effect model, working above 68 hours had negative mental health effects in permanent and precarious employees, revealing that employees working above the legal maximum are vulnerable regardless of employment type. Therefore, when considering the high average number of working hours and large proportion of precarious workers in South Korea, it is important to consider both factors when addressing the mental health of employees.

    References

    1. ↵
      1. Virtanen M,
      2. Ferrie JE,
      3. Singh-Manoux A, et al
      . Long working hours and symptoms of anxiety and depression: a 5-year follow-up of the Whitehall II study. Psychol Med 2011;41:2485–94. doi:10.1017/S0033291711000171
      OpenUrlCrossRefPubMed
    2. ↵
      1. Bannai A,
      2. Tamakoshi A
      . The association between long working hours and health: a systematic review of epidemiological evidence. Scand J Work Environ Health 2014;40:5. doi:10.5271/sjweh.3388
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Lee S,
      2. McCann D,
      3. Messenger JC
      . Working time around the world. Geneva: ILO, 2007.
    4. ↵
      1. Kim SW,
      2. Yoon JS
      . Suicide, an urgent health issue in Korea. J Korean Med Sci 2013;28:345–7. doi:10.3346/jkms.2013.28.3.345
      OpenUrlCrossRefPubMed
    5. ↵
      1. Kim I,
      2. Kim H,
      3. Lim S, et al
      . Working hours and depressive symptomatology among full-time employees: results from the fourth Korean National Health and Nutrition Examination Survey (2007–2009). Scand J Work Environ Health 2013;39:515–20. doi:10.5271/sjweh.3356
      OpenUrlCrossRefPubMed
    6. ↵
      1. Lee KH,
      2. Kim JE,
      3. Kim YK, et al
      . Long working hours and emotional well-being in Korean manufacturing industry employees. Ann Occup Environ Med 2013;25:38. doi:10.1186/2052-4374-25-38
      OpenUrlCrossRefPubMed
    7. ↵
      1. Heo HT,
      2. Kim DW,
      3. Lee JS, et al
      . An association between working schedules and depression in public sector employees. Korean J Occup Environ Med 2012;24:347–55.
      OpenUrl
    8. ↵
      1. Son M,
      2. Kong JO,
      3. Koh SB, et al
      . Effects of long working hours and the night shift on severe sleepiness among workers with 12-hour shift systems for 5 to 7 consecutive days in the automobile factories of Korea. J Sleep Res 2008;17:385–94. doi:10.1111/j.1365-2869.2008.00675.x
      OpenUrlCrossRefPubMed
    9. ↵
      The Korean Institute of Social and Health Affairs. User's guide the 8th wave of the Korean Welfare Panel Study. Seoul, South Korea: The Korean Institute of Social and Health Affairs, 2013.
    10. ↵
      1. Gellis ZD
      . Assessment of a brief CES-D measure for depression in homebound medically ill older adults. J Gerontol Soc Work 2010;53:289–303. doi:10.1080/01634371003741417
      OpenUrlCrossRefPubMed
    11. ↵
      1. Kohout FJ,
      2. Berkman LF,
      3. Evans DA, et al
      . Two shorter forms of the CES-D (Center for Epidemiological Studies Depression) depression symptoms index. J Aging Health 1993;5:179–93. doi:10.1177/089826439300500202
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Park JH,
      2. Kim KW
      . A review of the epidemiology of depression in Korea. J Korean Med Assoc 2011;54:362–9. doi:10.5124/jkma.2011.54.4.362
      OpenUrlCrossRef
    13. ↵
      1. Cho MJ,
      2. Kim KH
      . Diagnostic validity of the CES-D (Korean version) in the assessment of DSM-III-R major depression. J Korean Neuropsychiatr Assoc 1993;32:381–99.
      OpenUrl
    14. ↵
      1. Lee K
      . Income movement in South Korea: current status and characteristics. Korea Dev Econ Assoc 2009;15:55–77.
      OpenUrl
    15. ↵
      1. Witte JS,
      2. Greenland S,
      3. Kim LL, et al
      . Multilevel modeling in epidemiology with GLIMMIX. Epidemiology 2000;11:684–8. doi:10.1097/00001648-200011000-00012
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Bell DN,
      2. Hart RA
      . Unpaid work. Economica 1999;66:271–90. doi:10.1111/1468-0335.00169
      OpenUrlCrossRef
    17. ↵
      1. Cha S
      . Associations between paid work hour and time pressure: focused on the role of rest and leisure. Popul Assoc Korea 2014;37:25–52.
      OpenUrl
    18. ↵
      1. Harrington JM
      . Health effects of shift work and extended hours of work. Occup Environ Med 2001;58:68–72. doi:10.1136/oem.58.1.68
      OpenUrlFREE Full Text
    19. ↵
      1. Grant-Vallone EJ,
      2. Donaldson SI
      . Consequences of work-family conflict on employee well-being over time. Work Stress 2001;15:214–26. doi:10.1080/02678370110066544
      OpenUrlCrossRefWeb of Science
    20. ↵
      1. Hill EJ,
      2. Hawkins AJ,
      3. Ferris M, et al
      . Finding an extra day a week: the positive influence of perceived job flexibility on work and family life balance*. Fam Relations 2001;50:49–58. doi:10.1111/j.1741-3729.2001.00049.x
      OpenUrlCrossRefWeb of Science
    21. ↵
      1. Park Y,
      2. Kim C-W,
      3. Lee J, et al
      . A research on the actual condition of leisure activities stimulated by the introduction of the 40-hour workweek system. J Leisure Stud 2004;2:43–68.
      OpenUrl
    22. ↵
      1. Jacobs JA,
      2. Gornick JC
      . Hours of paid work in dual-earner couples: the United States in cross-national perspective. Sociological Focus 2002;35:169–87. doi:10.1080/00380237.2002.10570696
      OpenUrlCrossRef
    23. ↵
      1. Schuster M,
      2. Rhodes S
      . The impact of overtime work on industrial accident rates. Ind Relations J Econ Soc 1985;24:234–46. doi:10.1111/j.1468-232X.1985.tb00992.x
      OpenUrl
    24. ↵
      1. Choi KS,
      2. Kang SK
      . Occupational psychiatric disorders in Korea. J Korean Med Sci 2010;25:S87–93. doi:10.3346/jkms.2010.25.S.S87
      OpenUrlCrossRefPubMed
    25. ↵
      1. Dembe AE,
      2. Erickson JB,
      3. Delbos RG, et al
      . The impact of overtime and long work hours on occupational injuries and illnesses: new evidence from the United States. Occup Environ Med 2005;62:588–97. doi:10.1136/oem.2004.016667
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Hänecke K,
      2. Tiedemann S,
      3. Nachreiner F, et al
      . Accident risk as a function of hour at work and time of day as determined from accident data and exposure models for the German working population. Scand J Work Environ Health 1998;24(Suppl 3):43–8.
      OpenUrlPubMedWeb of Science
    27. ↵
      1. Virtanen M,
      2. Kivimäki M,
      3. Joensuu M, et al
      . Temporary employment and health: a review. Int J Epidemiol 2005;34:610–22. doi:10.1093/ije/dyi024
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Kim IH,
      2. Muntaner C,
      3. Khang YH, et al
      . The relationship between nonstandard working and mental health in a representative sample of the South Korean population. Soc Sci Med 2006;63:566–74. doi:10.1016/j.socscimed.2006.02.004
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Shin K-Y
      . Globalisation and the working class in South Korea: contestation, fragmentation and renewal. J Contemporary Asia 2010;40:211–29. doi:10.1080/00472331003597554
      OpenUrlCrossRef
    30. ↵
      1. Saloniemi A,
      2. Virtanen P,
      3. Vahtera J
      . The work environment in fixed-term jobs: are poor psychosocial conditions inevitable? Work, Employment and Society, 2004.
    31. ↵
      1. Quinlan M,
      2. Bohle P
      . Overstretched and unreciprocated commitment: reviewing research on the occupational health and safety effects of downsizing and job insecurity. Int J Health Serv 2009;39:1–44. doi:10.2190/HS.39.1.a
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Contributors WK and THK designed the study, collected the data, performed the statistical analysis, and wrote the manuscript. WK, E-CP, T-HL and THK contributed to the discussion and reviewed and edited the manuscript. THK is the guarantor of this work and, as such, had full access to all of the data. THK assumes responsibility for the integrity of the data and the accuracy of the data analysis.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.