This study examined trajectories of internalising and externalising problems from adolescence to young adulthood in young adults with different work-family life courses. The primary finding is that trajectories of internalising and externalising problems of women in the inactive work-family life course diverged from other work-family life courses during young adulthood. Differences in the internalising and externalising problem scores increased during young adulthood with women in the inactive work-family life course reporting higher scores in comparison with the work-family life courses active with children and continuous education and work for internalising and externalising problems respectively.

The inactive work-family life course, although not exclusive in experiencing periods of inactivity, represents women who were inactive for over half of the observed period, meaning they were neither in education nor working for more than five years during young adulthood. Our findings on higher scores of mental health problems among the inactive work-family life course are in line with previous research showing that being in NEET (not in education, employment or training) in young adulthood is associated with experiencing mental health problems [20]. A potential explanatory mechanism could be that young women with a history of long spells of inactivity do not achieve age-specific transitions such as finishing education or finding a job. Previously, it has been shown that work-family life courses characterised by postponing important transitions during young adulthood, such as a transition from education to work, were associated with worse well-being and mental health in young adulthood [4]. Another explanatory mechanism might be related to the transition to parenthood. One-third of the women in the inactive work-family life course became mothers and the combination of inactivity and parenthood may put these women in an especially vulnerable position and decrease their mental health. For instance, a recent study examining the impact of employment on well-being in German mothers showed that their mental health improved after the transition from parental leave into employment [21]. However, the observed relationship between inactivity in young adulthood and higher scores of mental health problems can also operate in reverse, i.e. experiencing mental health problems in adolescence and young adulthood has been shown to decrease participation in educational and employment status [22]. Further research is needed to shed light on mechanisms and causal pathways.

While the inactive work-family life course in women displayed higher scores of both internalising and externalising problems compared with the other work-family life courses and the effect sizes of the mean differences were between small and medium [23], the average scores remained below the borderline clinical threshold, i.e. their scores did not meet the criteria for significant mental disorders. Even though the inactive work-family life course remained below the borderline threshold, this group of women had consistently higher scores of mental health problems throughout young adulthood, and the difference was at its largest at age 29. For some women in this group, preventive support measures could be beneficial for their future mental health.

The mental health trajectories of all five remaining work-family life courses among women were relatively similar to one another. The overall negligible differences in mental health trajectories between these other five work-family life courses could be a sign of loosening strict expectations of what young adults need to achieve in their 20s. In light of this finding and previous research showing more individualised pathways into adulthood [2], it may not be harmful for mental health to postpone or forfeit some age-specific transitions, as long as spells of inactivity do not dominate the work-family life course. It is also possible that differences in mental health between work-family life courses become larger only later in the life course. A longer follow-up time is needed to examine how the work-family life courses further develop and whether differences in the mental health trajectories emerge later in the life course.

We did not find differences in trajectories of mental health problems between work-family life courses of men. A potential explanation may be that men in a work-family life course labelled as inactive may have not exhibited a distinct level of inactivity compared to other life courses. While men in the inactive life course spent over half of the observed period in inactive states, their engagement in education or work at the end of young adulthood might have mitigated the potential impact on mental health, making them less distinguishable from the other work-family life courses. Also, the exclusion of men with lower education and prior mental health problems due to the loss to follow-up might limit the representation of inactive men in our sample, potentially underestimating the association between the inactive work-family life course and worse mental health. Another explanation may be that men more often underreport mental health problems. Research has shown the problematic impact of conformity to traditional gender norms on underreporting and expression of symptoms of mental health problems among men [24]. Although we did not find differences in mental health trajectories of men, we did observe differences in externalising problems at age 29 between work-family life courses continuous education and work and education and work to work. The effect size of the mean difference was small (Cohen’s d = 0.43) [23]. Further research is needed to investigate how mental health develops in young men experiencing different education-to-work transitions. Additionally, our study highlights the need to consider gender differences when examining the relationship between work, family, and mental health trajectories during young adulthood.

Finally, the findings of the present study are consistent with and contribute to our previous research in the following ways. Women in work-family life courses characterised by an early transition into parenthood reported on average higher scores of externalising problems during adolescence, and men in the long education work-family life course reported higher scores of internalising problems during adolescence, which is in line with the findings from our previous study [13]. A finding emerging from the present study is the higher average score of externalising problems during adolescence among men in the early work life course. This is in line with one of our previous studies showing that experiencing externalising problems in adolescence is associated with low or medium educational attainment at age 19 [25]. An important insight provided by this study is the observation of diverging mental health trajectories during young adulthood, particularly in women. Thus, this study underscores the importance of examining mental health problems beyond mere point estimates and considering the developmental trajectory of mental health problems during the life course.

This study has several strengths. We used detailed information on respondents’ education, work and parenthood histories which made it possible to examine combined work-family life courses during young adulthood. Additionally, we linked mental health trajectories based on data collected at seven mental health assessments spanning a period of 18-year follow-up to the identified work-family life courses. By doing so, we were able to go beyond analysing the associations between work-family life courses and mental health assessed at one or two points in time. This approach allowed us to describe how mental health developed during work-family life courses.

Our findings also need to be interpreted in the light of some limitations. Firstly, our sample comprises 44% of the baseline sample included in TRAILS, resulting in a few of the identified work-family life courses containing fewer than 50 people. Respondents lost to follow-up were most likely to be men, lower educated and had a lower parental SES. Respondents who did not report their work and family histories were more likely to experience internalising and externalising problems at the last two measurement waves. Thus, we might have underestimated the severity of mental health problems in work-family life courses characterised by longer spells of inactivity and lower education due to the loss to follow-up of respondents with more severe mental health problems, especially among men. Secondly, the family domain in our typology of work-family life courses only covers parenthood and does not include other family characteristics such as partnership or residency because detailed data on these aspects was not available. By not including these potentially important predictors of mental health, we may have grouped respondents with substantially different work-family experiences together in the same work-family life course and thus be missing important associations between work-family experiences and mental health.

Our findings show that, among women, differences in mental health trajectories between the inactive work-family life course and the other life courses started to increase in young adulthood. This indicates an important time window for implementing policies and interventions to support young adults in the inactive work-family life course. In order to understand how young people in the inactive work-family life course perceive their situation, what their needs are and how they can be supported, qualitative studies are needed. Furthermore, future research may examine whether the timing of key transitions in work-family life courses is associated with changes in mental health. Such research could contribute to identifying entry points for work-family interventions to support young adults’ mental health.