Antecedents and consequences of work–home interference among medical residents

https://doi.org/10.1016/S0277-9536(98)00425-0Get rights and content

Abstract

A cross-sectional field study is reported in which a comprehensive model of work–home interference (WHI) was developed and tested among 166 medical residents of an academic hospital in the Netherlands. It was hypothesized that WHI functions as a critical mediating pathway in the relationship between work and home characteristics on the one hand, and work-related and general psychological health indicators on the other. The results revealed that one home characteristic and three work characteristics put pressure on the interface between the work and home life, that is, (1) having a spouse who works overtime frequently, (2) an unfavorable worktime schedule, (3) a high quantitative workload and (4) a problematic dependency on the superior. The results further showed that WHI was positively associated with emotional exhaustion and depersonalization (i.e. work-related health indicators), as well as with psychosomatic health complaints and sleep deprivation (i.e. general health indicators). More importantly, the results strongly supported our basic hypothesis that WHI mediates the impact of some work and home characteristics on psychological health indicators. This seems to be particularly true for the general health indicators: none of the home and work characteristics just mentioned, had a direct impact on these general indicators, independent of WHI. With respect to the work-related health indicators, particularly depersonalization, the mediating role of WHI was also strong, though less consistent. The theoretical and practical implications of the findings are discussed.

Introduction

Interest in the relationship between employees’ work and home lives has substantially grown since the mid-eighties (cf. Neal et al., 1993). Traditionally, work and home have been considered separate domains. More recently, the interdependence between both domains is increasingly recognized and has captured the interest of many researchers. This is a result of several sociodemographic and economic trends in our society. Firstly, there has been an impressive increase in the number of women joining the work force. In 1991 already 39% of the European working population was female and this percentage has increased to 42% in 1996 (Paoli, 1997). Associated with this development is an increase in the prevalence of employed adults who are part of dual-career families (cf. Burke and Greenglass, 1987; Zedeck and Mosier, 1990; Frone et al., 1992). Secondly, the nature of work has changed, particularly demanding more mental and emotional effort (rather than physical effort). A recent large-scale European study conducted among a representative sample of over 15,000 employees from 15 European member states revealed that a growing number of employees, particularly in the Netherlands, is working under high speed and with tight deadlines (Paoli, 1997). Thirdly, due to, amongst others, the growing international competition (globalization of the economy) and the flexibilization of worktime schedules (e.g. 24-hours economy), a higher appeal is done to employees’ flexibility to work irregular hours and during `unsocial’ hours (i.e. evening-work, night-work, weekend-work and working overtime).

As a consequence, an increasing number of persons is confronted with high pressures in both their work and home life, and many of their daily hassles stem from job responsibilities that are incompatible with home or family responsibilities. This is often referred to as work–home interference (WHI). In the current study WHI is defined as the extent to which a person experiences pressures within the work domain that are incompatible with the pressures that arise within the family domain (cf. Kopelman et al., 1983). In a simpler form, WHI implies having to decide how to allocate one’s limited time and energy resources over both domains.

Interest in the interdependence between work and home has stimulated stress research in which WHI is generally assumed to play a mediating role in the relationship between potential stressors in both domains and psychological health indicators. A core assumption in Meijman’s effort-recovery model (Meijman, 1989) is that dealing with high work load and developing mental or emotional strains during the working period is not necessarily unhealthy, as long as one can recover sufficiently during the nonworking hours. However, when limited time and energy resources are further depleted due to high family demands, serious conflicts in fulfilling both work and family roles (i.e. WHI) are likely to occur. These conflicts may be time-based in a way that it is timely impossible to combine both work and family responsibilities (e.g. working overtime and picking up one’s child from a child care center simultaneously). They may also be effort-based (or strain-based) in a way that it is psychologically difficult to fulfil family demands (e.g. being an attentive parent or spouse), when strain-effects that were build up during the working period ‘spillover’ to the home situation (cf. Kopelman et al., 1983; Greenhaus and Beutell, 1985; Stephens and Sommer, 1996). The insufficient recovery from the incompatible pressures within both domains (i.e. WHI) is likely to result in psychological health complaints that in the long run become increasingly serious and chronic in nature.

Based on this theoretical perspective, the central question that is addressed in the current study is: to what extent does WHI play a mediating role in the relationship between work and home characteristics, on the one hand, and psychological health indicators on the other? Several studies have provided evidence for such a mediating role of WHI. For instance, the results of a recent study conducted by Stephens et al. (1997)among female employees with caregiving responsibilities revealed that job stress was related to depression, only through WHI. A similar intervening role of WHI was found by Parasuraman et al. (1996): male and female entrepreneurs who experienced workrole overload and high parental demands reported more WHI, which, in turn, was related to general life stress. A recent study conducted among employed men and women in Finland revealed that the number of children, having a fulltime job and the lack of a supportive relationship with one’s superior were important antecedents of WHI, which, in turn, was significantly associated with job anxiety, job depression, job exhaustion and psychosomatic complaints (Kinnunen and Mauno, 1998).

In order to fully examine this possible mediating role of WHI in the relationship between work and home characteristics and psychological health indicators, it can be argued that at least four criteria have to be met (cf. Frone et al., 1992): (1) WHI is preferably examined among both male and female employed adults who have arrived at a point in life during which both career opportunities and major family events are likely to occur simultaneously and, therefore, might interfere, (2) key work- and home-related antecedents of both WHI and psychological health should be assessed, (3) both work-related and general indicators of psychological health should be assessed and (4) the various relationships between antecedents and consequences of WHI should be tested simultaneously by employing covariance structure modeling.

In the current study, a comprehensive model of WHI is developed and tested that meets these four criteria. Firstly, our study is conducted among medical residents, that is, physicians who have successfully finished their study of medicine and now assist the staff of medical specialists in a hospital. WHI is likely to occur among this relatively young and highly educated occupational group, because residents often have to combine a demanding home situation (e.g. being part of a dual career couple, getting and raising children), with a highly stressful period in their career development (e.g. high patient loads, long and irregular working hours, high dependency on the superiors’ evaluation to start, continue and successfully finish one’s education) (see for a review: Butterfield, 1988). Secondly, our study assesses major antecedents of WHI in both the work and home domain, as well as potential consequences of WHI. With respect to the consequences, a distinction will be made between general indicators of psychological health (that is, indicators that are not specifically related to one domain, but might be attributed to and experienced in both domains) and work-related indicators of psychological health (that is, health complaints that are caused by the work situation and primarily affect people’s functioning at work). Finally, the hypothesized relationships between antecedents and consequences of WHI will be tested simultaneously by employing covariance structure modeling. The conceptual model of antecedents and consequences of WHI that guides the present research, illustrated in Fig. 1, will be discussed in more detail in the following sections. Because male and female residents may differ on one or more of the variables included in the model, gender is included as a control variable.

One of the most promising models addressing the impact of work characteristics on psychological health is the demand–control–support (DCS) model (Karasek, 1979; Karasek and Theorell, 1990; Johnson and Hall, 1994). The core assumption of this model is that psychological health complaints are most likely to occur in jobs where the psychological demands are high and one’s autonomy and superiors’ support are low. The high psychological demands in these so-called `isolated high strain jobs’ are assumed to produce a state of arousal that cannot be dealt with adequately due to the lack of autonomy and support. Although a considerable amount of research in the industrial sector has been based on this theoretical perspective, only a few studies have attempted to apply the DCS model to the work characteristics of health-care workers (cf. de Jonge et al., 1996). For instance, Landsbergis (1988)showed among nearly 300 American hospital employees that levels of burnout, psychosomatic symptoms, depression and job dissatisfaction were significantly higher in situations that were characterized by high job demands and low job control. In addition, Parkes et al. (1994)found in a heterogeneous sample of health-care workers that somatic symptoms were significantly predicted by an unfavorable combination of high job demands, low control and low support.

We argue that the job of medical resident is an isolated high strain job. Firstly, empirical studies show that both high time and effort demands are prevalent among residents: residents have long and irregular working hours (weeks consisting of 60–100 working hours are not exceptional), and they report to spend too many hours in the hospital, to see too many patients in too little time, and to have too little spare time to spend alone or with one’s family and friends, to get sufficient sleep, to study or to undertake physical activities (Gordon et al., 1987; Schwartz et al., 1987; Smith et al., 1988; Lemkau et al., 1988; Kirsling et al., 1989; Archer et al., 1991; Berkoff and Rusin, 1991; Jex et al., 1991; Phelan, 1992; Toews et al., 1993). In the current study, `worktime schedule’, indicating that too little time is provided to recover sufficiently from the effort one has put forth on the job, is included as a time demand. In addition, various studies have shown the stressful impact of both high quantitative and qualitative workload, such as beeper calls, call duty, heavy patient care demands and highly responsible tasks with often severe consequences of one’s mistakes (Lemkau et al., 1988; Berkoff and Rusin, 1991). Hence, two effort demands are included in the current study: `quantitative workload’ referring to too much work to do in too little time, and `mental workload’ (i.e. qualitative workload), referring to the high attention and accuracy required in the job.

It is also plausible to assume that residents have limited control over their tasks and conduct during the working day. In fact, they follow a training program in which their activities are highly supervised, evaluated and judged by their superiors. This, however, does not necessarily imply that residents experience their relationship with the superior as being supportive. In fact, high dependency on the superior who is continuously evaluating and judging one’s activities, is likely to be experienced as troublesome by the residents (cf. Firth-Cozens, 1989). Considering the theoretical framework provided by the DCS model, both `lack of job autonomy’ and high `dependency on superior’ are also included as potential stressful work characteristics in the current study. Accordingly, it is hypothesized that residents are more likely to experience WHI, the more they experience (1) an unfavorable worktime schedule, (2) a high quantitative workload, (3) a high mental workload, (4) lack of job autonomy and (5) a high dependency on the superior (see path 1 in Fig. 1).

In the literature on WHI, there is little evidence on the potential impact of home characteristics on WHI (e.g. Kopelman et al., 1983; Burke, 1988; Voydanoff, 1988). A study by Higgins and Duxbury (1992)revealed that males in dual career couples report more negative spillover from work to family than males in the traditional couples (that is, male breadwinner and fulltime housewife). In addition, a study of Parasuraman et al. (1992)among dual career couples showed that both men and women with preschool children had more trouble combining work and family roles than the `dinky’s’ (double income no kids). A study conducted by Holahan and Gilbert (1979)also revealed that the less spouse support was experienced by males and females in dual career couples, the more they experienced incompatible pressures in combining work and family roles.

In the present study some obvious home characteristics are included. Some of these can be regarded as time-based antecedents, because they require a person to spend large amounts of time in home activities, such as `dual career’ (i.e. whether or not one has a working spouse), `overtime partner’ (`whether or not one’s spouse has a time consuming job’) and `parental status’ (i.e. whether or not there are home-living children). In addition, two effort-based (or strain-based) antecedents are included, referring to home characteristics that may induce or reduce strain symptoms. These include `child care arrangements’ (i.e. whether or not one has flexible arrangements, referring to friends, relatives or professionals who take care of the children, particularly in the case of sudden emergencies) and `social support’ (i.e. the extent to which family members provide a supportive home situation). Accordingly, we expect that residents who are part of a dual career family (1), have partners who work overtime frequently (2), have children (3), have trouble to arrange flexible child care arrangements (4) and experience little support from their home situation (5), are more likely to experience WHI, than those to whom these characteristics do not apply (see path 2 in Fig. 1).

Burnout, referring to the draining of energy resources caused by chronic job stress and primarily affecting human service professionals, is considered a work-related indicator of psychological health in the current study. Originally burnout was defined by Maslach and Jackson (1982)as a three-dimensional construct, characterized by emotional exhaustion (i.e. a feeling of emptiness, of having no emotional resources left and little to offer psychologically to clients/patients and colleagues), depersonalization (i.e. negative and cynical attitudes and feelings towards people one has to deal with professionally, like patients) and reduced feelings of personal accomplishment (i.e. the tendency to evaluate oneself negatively, particularly with regard to one’s work with clients/patients). A study by Schaufeli and van Dierendonck (1993), however, provided strong evidence for two, rather than three dimensions underlying the burnout construct. They conclude that emotional exhaustion is the core dimension of burnout, which is accompanied by negative attitudes towards one’s clients/patients or colleagues (i.e depersonalization). Therefore, the current study includes `emotional exhaustion’ and `depersonalization’ as the two core dimensions of burnout.

Evidence for a relationship between WHI and burnout comes from Jackson and Maslach (1982)who compared husbands’ job-related affective well-being with their wives’ descriptions of at-home behavior and experience. They observed that the level of husbands’ job-related emotional exhaustion was significantly associated with wives reporting (1) that their husband came home tense, unhappy, tired and upset, (2) that he had difficulty sleeping at night and (3) a low quality of family life. Similar evidence is provided by a study from Piotrkowski (1978)among lower and middle class American families. This study shows that husbands’ negative work-related feelings were brought into the family system, reducing husbands’ emotional and interpersonal availability for family interactions, or even requiring family members to help the husband manage his feelings of strain. Also a study by Burke (1988), conducted among more than 800 police officers, revealed that WHI was significantly related to burnout and work alienation. In the current study it is, therefore, hypothesized that WHI is positively associated with (1) emotional exhaustion and (2) depersonalization (see path 3 in Fig. 1).

In the current study `psychosomatic health complaints’ and `sleep deprivation’ are included as general psychological health indicators, because they may be attributed to and experienced in both the work and family domain. A recent study conducted by Stephens et al. (1997)among female employees with caregiving responsibilities revealed that the more their job interfered negatively with their caregiving duties, the more they reported general psychological health complaints (e.g. feelings of depression). A study by Burke (1994)conducted among more than 800 police officers showed that WHI was significantly related to psychosomatic symptoms (e.g. headaches and pain in the heart). In addition, those who reported to have too little spare time for self-development, experienced more negative feelings, such as sleep deprivation and depression. Accordingly, in the current study it is hypothesized that WHI is positively associated with (1) `psychosomatic health complaints’ and (2) `sleep deprivation’ (see path 4 in Fig. 1).

Section snippets

Study sample and procedure

A cross-sectional study was conducted among the whole population of medical residents of an academic hospital in the Netherlands (n=293). All residents were informed about the purpose and the interests of the study by a publication from the association of medical residents, as well as by the hospital’s department of social affairs. Two weeks after the residents had received a survey at their home address, a reminder was sent out in which the importance of the participation of each individual

Results

Five stepwise regression analyses were executed with, respectively, WHI and the four psychological health indicators as criterium variables; gender was included as a control variable and the work and home characteristics as predictor variables. Both gender (males=`0’ and females=`1’) and the home characteristics (except `social support’) were entered as dummy variables (`not true’=`0’ and `true’=`1’).

Regression model A, addressing the relative impact of the work and home characteristics on WHI,

Discussion

The aim of the present study was to develop and test a comprehensive model of antecedents and consequences of work–home interference (WHI) among residents. The core question was to what extent WHI would function as a critical mediating pathway in the relationship between work and home characteristics, on the one hand, and psychological health indicators on the other. The results provided evidence for three work-related antecedents of WHI, that is, an unfavorable worktime schedule, a high

References (67)

  • R Kopelman et al.

    A model of work, family and interrole conflict: a construct validation study

    Organizational Behavior and Human Performance

    (1983)
  • S Parasuraman et al.

    Work and family variables, entrepreneurial career success and psychological wellbeing

    Journal of Vocational Behavior

    (1996)
  • K.R Parkes et al.

    Social support and the demand-discretion model of job stress: tests of additive and interactive effects in two samples

    Journal of Vocational Behavior

    (1994)
  • L.R Archer et al.

    The relationship between residents’ characteristics, their stress experiences and their psychosocial adjustment at one medical school

    Academic Medicine

    (1991)
  • Aiken, L.S., West, S.G., 1991. Multiple Regression: Testing and Interpreting Interactions. Sage Publications, Newbury...
  • R.M Baron et al.

    The moderator-mediator variable distinction in social psychological research: conceptual, strategic and statistical considerations

    Journal of Personality and Social Psychology

    (1986)
  • P.M Bentler

    Comparative fit indexes in structural models

    Psychological Bulletin

    (1990)
  • P.M Bentler et al.

    Significance tests and goodness of fit in the analysis of covariance structures

    Psychological Bulletin

    (1980)
  • K Berkoff et al.

    Pediatric house staff’s psychological response to call duty

    Developmental and Behavioral Pediatrics

    (1991)
  • Bollen, K.A., Scott Long, J., 1993. Introduction. In: Bollen, K.A., Scott Long J. (Eds.), Testing Structural Equation...
  • N.P.G Boumans et al.

    Some problems concerning the measurement of job characteristics in nursing work

    The European Work and Organizational Psychologist

    (1993)
  • Browne, M.W., Cudeck, R., 1993. Alternative ways of assessing model fit. In: Bollen, K.A., Scott Long, J. (Eds.),...
  • Burke, R. J., Greenglass, E. R. 1987. Work and family. In: Cooper, C. L., Robertson, J. T. (Eds.), International Review...
  • Burke, R. J., 1988. Some antecedents and consequences of work-family conflict. Journal of Social Behavior and...
  • R.J Burke

    Stressful events, work–family conflict, coping, psychological burnout and well-being among police officers

    Psychological Reports

    (1994)
  • P.S Butterfield

    The stress of residency: a review of the literature

    Archives of Internal Medicine

    (1988)
  • Cox, T., Ferguson, E., 1991. Individual differences, stress and coping. In: Cooper, C.L., Payne R. (Eds.), Personality...
  • Dirken, J.M., 1969. Arbeid en Stress [Work and Stress]. Groningen: Wolters...
  • van Deursen, C.G.L., Gründemann, R.W.M., 1995. Werksituatie, Gezondheid en Ziekteverzuim in de Contract-catering...
  • Evers, A., 1995. Meetinstrumenten voor Arbeidsomstandigheden, Stress en Welzijn [Measurement Instruments of...
  • J Firth-Cozens

    Sources of stress and ways of coping in junior house officers

    Stress Medicine

    (1989)
  • M.R Frone et al.

    Antecedents and outcomes of work–family conflict: testing a model of the work–family interface

    Journal of Applied Psychology

    (1992)
  • A Gazendam et al.

    De validiteit van de VOEG onderzocht bij een groep depressieve vrouwelijke patiënten in de huisartspraktijk [The validity of the VOEG investigated among a group of depressive female patients in a general practitioner’s office]

    Nederlands Tijdschrift voor Psychologie

    (1987)
  • G.H Gordon et al.

    Comparison of mood changes and satisfaction among first-year medical residents in three programs

    Journal of Medical Education

    (1987)
  • J.H Greenhaus et al.

    Sources of conflict between work and family roles

    Academy of Management Review

    (1985)
  • C.A Higgins et al.

    Work–family conflict: a comparison of dual career and traditional career men

    Journal of Organizational Behavior

    (1992)
  • J.B Herman et al.

    Working men and women: inter- and intrarole conflict

    Psychology and Women Quarterly

    (1977)
  • C.K Holahan et al.

    Conflict between major life roles: women and men in dual career couples

    Human Relations

    (1979)
  • House, J.S., 1981. Work Stress and Social Support. Addison Wesley, Reading,...
  • S.E Jackson et al.

    After-effects of job-related stress: families as victims

    Journal of Occupational Behavior

    (1982)
  • Jansen, B., 1987. Rota-Risk-Profile-Analysis. In: Oginsky, A., Pokorski J., Ruttenfranz, J. (Eds.), Contemporary...
  • M.E Jansen et al.

    Verkorte versies van de VOEG-schaal [Shortened versions of the VOEG-questionnaire]

    Gedrag & Samenleving

    (1981)
  • S.M Jex et al.

    Behavioral consequences of job-related stress among resident physicians: the mediating role of psychological strain

    Psychological Reports

    (1991)
  • Cited by (217)

    • Mentoring in radiology: An asset worth exploring!

      2022, European Journal of Radiology
    • Work-family spillover stress predicts health outcomes across two decades

      2020, Social Science and Medicine
      Citation Excerpt :

      Researchers have established links between NWFS and poorer physical and emotional health conditions including anxiety, depression, psychosomatic health complaints, sleep disruption, fatigue, poor sleep, obesity, and increased risk for chronic illness (Allen et al., 2000; Buxton et al., 2016; Demerouti and Geurts, 2004; Frone, Russell and Cooper, 1997; Geurts et al., 1999; Grzywacz, 2000; Grzywacz et al., 2002; Williams et al., 2006). Moreover, researchers have found that NWFS accounts for the relationship between reported job stress and adverse health outcomes in a cross-sectional study (Geurts et al., 1999; Geurts et al., 2003). These findings imply that the resulting lingering effects of work stress compound over time, yet are limited by the lack of longitudinal information (Geurts et al., 2003).

    • Learning climate and quality of Italian training courses in gynecology and obstetrics

      2019, European Journal of Obstetrics and Gynecology and Reproductive Biology
    View all citing articles on Scopus
    View full text