Background: A negative link between unemployment and psychological health is well documented, yet little is known about the protective effect of continuous employment on psychological health.
Method: In this prospective population-based cohort study, the effect of continuous employment on psychological health was examined, using individuals born in Great Britain during a week of April 1970. Respondents (2901 men and 3288 women) who were employed at the age of 26 years, with a complete employment history between ages 26 and 30 years and having information about cohabitation, social class and psychological and physical health at age 30 years, were included in the analysis.
Results: Findings showed that continuous employment was associated with better psychological health in men. This effect was somewhat greater in those who showed evidence of poorer psychological health at the age of 26 years. In working women, cohabitation provided a protective effect on psychological health.
Conclusion: The findings show that, for men, staying in continuous employment despite experiencing poor psychological health may contribute to better psychological health.
Statistics from Altmetric.com
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.
Previous studies provide evidence of the deleterious influence of job insecurity and an unstable work history on psychological health. In contrast to the negative effect of unemployment, employment is known to be positively associated with psychological health.1 In addition, studies of re-employment have shown improved psychological health in formerly unemployed workers.2 3
The relationship between work and mental health is, however, bi-directional. Psychological ill health tends to force physically capable employees out of the labour force.4 Poor mental health is a major cause of economic loss in England5 and the United States.6 In May 2006, approximately 2.7 million working-aged people in the United Kingdom claimed incapacity benefit,7 for either physiological or psychological health reasons. Shiels et al8 reported that 28% of long-term incapacity claimants are estimated to have mild mental disorders such as depression and anxiety.
Although in some studies the combination of work and family strain is associated with poor physical and psychological health among women,9 others have shown health benefits of combining employment and family life over the longer term.10 Little is known, especially for women, about the specific protective effects from employment or whether these may vary according to the initial state of psychological health. The purpose of this study was to examine the effect of continuous employment on adult men and women’s psychological health using the 1970 British Cohort Study (BCS70).
Data were from the BCS70, an ongoing longitudinal study, targeting 17 200 individuals who were born in the United Kingdom during one week of April 1970.11 Those who were born in Northern Ireland were later excluded from the study. The analysis sample comprises a subsample of 2901 men and 3228 women from the BCS70 who were employed at the age of 26 years in 1996, had complete histories in the employment history data up to age 30 years in 2000, psychological health data at ages 26 and 30 years and cohabitation status, social class and physical health data at the age of 30 years.
Information about employment history was available between 1996 and 2000, when cohort members were aged from 26 to 30 years,12 giving employment status in each month. These data were used to define a group of respondents who were continuously in employment over this 4-year period. Periods of full-time or part-time employment, including self-employment, were defined as periods of stable employment. An employment history that included periods of being unemployed or out of the labour force as a result of taking up education, volunteer work, home care or sickness were treated as disrupted employment.
The data showed that most of those who were in employment were employed without any lapsed periods (see table 1), underlying the decision to dichotomise employment histories into a categorical variable. Individuals who were in full-time or part-time employment or self-employed during this entire period were identified to be in continuous employment, indicated by 1, as opposed to those who had gaps in their employment history (disrupted employment) indicated by 0.
Psychological health at ages 26 and 30 years
Information about individual psychological health was collected at the ages of 26 and 30 years, using the Malaise Inventory.13 The cut-off point of 7 or above suggested by Rodgers et al14 was used to identify the state of being psychologically less healthy, with the dichotomous variable, coded 1 for being psychologically less healthy and 0 for being psychologically healthy.
The effects from cohabitation status, physical health and social class at age 30 years were controlled in this study because we regarded these variables as confounders because of evidence of known associations with psychological health15 16 and employment.17–19 Cohabitation status was defined as either being married or living with a partner, indicated by 1, in contrast to those who responded that they were single, never married, separated or divorced, indicated by 0. Physical health was assessed according to self-reported information about being registered with a physical disability. Information about the nature of the physical disability was not disclosed. Those who reported having one or more registered physical disabilities were flagged for being compromised in their physical health, indicated by 1. We measured social position at age 30 years based on whether the cohort member’s last known occupation was manual, indicated by 1, or non-manual, indicated by 0. We used the partner’s last occupation if the information was not available.
We used logistic regression to examine the independent effects of continuous employment and psychological health at the age of 26 years on psychological health at 30 years. Reverse causality of psychological health at age 26 years affecting employment histories is expected; therefore, effects from continuous employment and psychological health at age 26 years were examined both independently and simultaneously. The effects were then adjusted for the confounders to assess whether there was a unique contribution of continuous employment to explaining the variation in psychological health at age 30 years. In addition, the moderating effect of continuous employment was tested formally by including the interaction term between continuous employment and the previous state of psychological health in the model.
The majority of cohort members were in good psychological health at the age of 26 and 30 years, were cohabiting, not disabled, were in a non-manual social class and had worked continuously from the age of 26 years. The mean number of months of non-employment during the 4-year period was 1.22 (SD 5.12, range 0–49) for men and 5.08 (SD 11.35, range 0–53) for women. Compared with men, women were less likely to be psychologically healthy at ages 26 and 30 years and were more likely to be cohabiting, in a non-manual social position and to have a disrupted employment history (see table 1). Gender differences in psychological health at age 30 years were not significant among those who had a sporadic employment history (χ2 = 1.00, df = 1, p = 0.32). Among those who had been in paid employment continuously, however, significantly more women than men were psychologically less healthy at age 30 years (χ2 = 39.06, df = 1, p<0.001).
In both men and women, there was a high degree of continuity in poor psychological health. Psychological ill health at age 26 years predicted psychological ill health at age 30 years independently of physical health status, social class, cohabitation status and employment history (see table 2). Although more women reported psychological ill health at the age of 26 and 30 years compared with men, the risk effect from psychological ill health at the age of 26 years was larger among men.
In men who were employed continuously from the age of 26 to 30 years, the odds of poor psychological health at age 30 years was reduced to less than half that found in men with disrupted employment. Because there could be a relationship between psychological health and future employment prospects, we adjusted for psychological health at age 26 years. The relationship of employment history to psychological health at age 30 years was not much attenuated in the multivariable model. This indicated that continuous employment protected men from psychological ill health, even when they had had a previous episode of psychological ill health and after adjustment for physical health, social class and cohabitation status. A stratified analysis confirmed that the reduction in the risk of psychological ill health at age 30 years was slightly greater (odds ratio (OR) 0.46, 95% CI 0.26 to 0.84) in those who were psychologically less healthy at age 26 years than in those who were healthy (OR 0.56, 95% CI 0.35 to 0.90) after adjusting for the confounders. A term was therefore fitted for the interaction between employment history and psychological health at age 26 years; the interaction term was non-significant.
In contrast, continuous employment did not protect women from psychological ill health at age 30 years. A stratified analysis by psychological health at age 26 years confirmed that continuous employment may have a small protective effect for women who were psychologically healthy at age 26 years (OR 0.78, 95% CI 0.57 to 1.07) only. In the women who were psychologically less healthy at age 26 years, continuous employment had no beneficial effect (OR 1.03, 95% CI 0.72 to 1.46). Testing the model containing a term for the interaction between employment history and psychological health at age 26 years suggested that women who were psychologically healthy may have benefited from continuous employment even though the interaction term did not reach formal significance.
Interestingly, the findings of this study showed that in women (but not men), those who were cohabiting were more likely to be psychologically healthy at age 30 years (see table 2). We further examined the possibility of a moderating effect of cohabitation on the relationship between continuous employment and psychological health at age 30 years to find whether the protective effect of continuous employment was significantly different between cohabiting and non-cohabiting women. No evidence of cohabitation moderating the effect of employment history was found (OR 1.15, 95% CI 0.69 to 1.91). This confirmed that for men, continuous employment protects against psychological ill health, whereas for women, it is living with a partner that is protective.
What is already known on this subject
Unemployment is related to deterioration and re-employment to improvements in psychological health
What this study adds
Continuous employment promoted men’s psychological health, regardless of their previous state of psychological health
Cohabitation status showed a protective effect on women’s psychological health independently of their employment history and previous psychological health
Detecting and treating psychological ill health in employed men and women would improve psychological health at the population level
Greater stability in employment would contribute to better psychological health among men
Our findings indicate that continuous employment protected men from psychological ill health. This protective effect was present regardless of their previous state of psychological health. The same effect was not, however, extended to women. Among women, their previous state of psychological health and their current cohabitation status were the most important predictors of their state of psychological health at the age of 30 years.
Findings from past studies have indicated the strong association of unemployment1 20 21 or re-employment2 3 with psychological health. Our findings add that in men, continuous employment promotes good psychological health even for those who have had previous episodes of poor psychological health. Descriptively, among those who were continuously employed, we found that women were psychologically less healthy at age 30 years compared with men. Our analysis through logistic regression confirmed that the benefit of continuous employment was not extended to women.
Although in general being in employment has a more positive effect on psychological health than being unemployed, some types of employment have been found to exert the same or even a more negative effect on psychological health than unemployment.22 Most of these findings have been for men only. In a previous study, a significantly higher level of job satisfaction among women disappeared when those women attained higher education, were in a professional occupation and worked in a male-dominated setting.23 Matthews et al24 found that women reported more negative work characteristics than did men in the 1958 cohort. Recent analysis of the British Household Panel Study25 26 reported that, in Great Britain, women were more satisfied with their jobs than men. Nevertheless, it is possible that work characteristics of women’s jobs may account for the reduction in the effect of continuous employment among women.
The negative effects of role strain9 22 and work or family-related conflicts27 on the psychological health of working women is well known. These findings might have led to the expectation that continuous employment could be associated with poorer mental health in cohabiting women. Cohabitation status did not, however, moderate the relationship between continuous employment and psychological health at the age of 30 years in this study. Moreover, continuous employment was associated with a slight increase in the risk of being psychologically less healthy at age 30 years in women who were psychologically less healthy at age 26 years. It is plausible that vulnerability to work-related stress may be greater in working women with poorer initial psychological health.
Although continuous employment did not significantly influence women’s psychological health at age 30 years, we did find that women’s cohabitation status contributed to the state of their current psychological health. This is consistent with studies showing that marital support28 or extended social ties29 are protective factors for women. Moreover, it has been found that working women in Great Britain were more likely than men to report good relations with management or colleagues, feelings of usefulness to society and of pleasure in being able to help other people.26 On the basis of our findings, it is possible that women’s main source of psychological wellbeing may be located more within their social networks both inside and outside the workplace, whereas that for men may be found predominantly within employment. More importantly, intermittent employment patterns may occur in women for quite different reasons than for men and may indicate the choice of some alternation between paid employment and care for children, with little or no negative implications for mental health. The contrasting relationship between continuous employment and psychological health in men and women is an example of the need for greater sensitivity to gender issues in epidemiology and public health more widely. Further investigation is needed to compare the significance and meaning of paid work for men and women in different types of employment and work-related environments.
The relationship between employment and psychological health is undoubtedly complex. In the “vitamin model”,22 a positive effect from features of the job environment is thought to have a threshold, indicating that employment may damage psychological health beyond a certain threshold of demand. We could only examine the effect of continuous employment during a 4-year period, given the availability of the data. The extent of the positive effect of continuous employment beyond 4 years is therefore unknown. Also, we only examined the psychological health of cohort members with complete data who were in employment in 1996 at the age of 26 years. The same positive effect might not have been found in those who became employed or re-employed some time between the ages of 26 and 30 years after a long period of non-employment. Further research is needed to examine employment histories into late adulthood so that the effects of periods of continuous employment can be compared within groups with various patterns of employment activities.
Psychological ill health at age 26 years was associated with a greatly increased risk of poor health at age 30 years among working men and women; however, the risk was greater among working men. This indicates the importance of the early detection and treatment of psychological ill health, especially among working men. The proportion of cohabiting or married women with children participating in paid work rose steadily between 1996 and 2000.30 During that period, the unemployment rate fell from 9.8% in 1996 to 5.8% in 2000 for men and from 6.3% to 4.6% in women. Whereas the increased participation of women in paid employment is welcome, our data indicate that there may be a growing need for psychological support for working people, especially for women with family responsibilities. The findings indicate that a negative chain of events leading from stable employment to disrupted employment may be broken by providing gender-specific support to working people who are psychologically distressed, so that they are able both to remain in the labour force and sustain a high level of psychological wellbeing.
Sweeps 5 and 6 of the 1970 British Cohort Study were deposited by the Centre for Longitudinal Study, Institute of Education. These data were accessed via the UK Data Archive. Neither the Centre for Longitudinal Studies nor the UK Data Archive bear any responsibilities for their further analysis or interpretation.
Funding: This research was funded by the Economic Social Research Council Capability and Resilience Network Grant RES-337-25-0001.
Competing interests: None declared.
Contributors: All authors participated in the design of the study and preparation of the manuscript. NC handled data analysis and is the guarantor.