In our cohort of first-time fathers, just over half took parental leave before their child’s first birthday. The timing of fathers’ initial parental leave was similar for those receiving higher-level and basic benefits. Paid parental leave benefit levels were associated with fathers’ postpartum mental health during the first year after birth, even after accounting for history of mental health, income, and employment status. Fathers receiving basic benefits had higher odds of receiving any mental healthcare, being prescribed antidepressants or anxiolytics, and receiving outpatient care compared with those receiving higher-level benefits. Fathers who did not take paid parental leave during the first year had higher odds of receiving any mental healthcare, outpatient care, and hospitalisation. However, there was no difference in mental health outcomes between fathers receiving minimum-level and higher-level benefits. Swedish-born fathers who did not take leave in the first year had higher odds for all mental health outcomes compared with Swedish-born fathers receiving higher-level benefits. Similarly, foreign-born fathers who did not take paid parental leave had higher odds of outpatient care, while those from non-OECD regions also had higher odds of hospitalisation.
Research has shown that parental leave generosity in terms of remuneration is beneficial for parents’ postpartum mental health, yet most of these studies have focussed on paid versus unpaid leave or paid versus no leave. Our findings suggest that higher levels of parental leave benefits may protect fathers against poor mental health in the postpartum period even in context where all parents are entitled to paid leave, regardless of employment status. These findings are consistent with those observed for mothers in the same context (country and time period)15 and suggest that material resources are an important mechanism in the relationship between parental leave and postpartum mental health. Receiving more generous benefits may alleviate the financial burden of reduced income due to time-off work4, and create the conditions that allow fathers to take parental leave (i.e., afford time-off work), possibly for a longer time. Accordingly, this may reduce financial stress and support mental wellbeing. However, we acknowledge that parental leave benefit levels are contingent on prior income and employment, and that the observed associations may partly reflect underlying socioeconomic inequalities, given that poor mental health is socially patterned16. Nevertheless, the persistence of the associations between benefit levels and postpartum mental health after adjustment for income and employment status suggests that parental leave benefit levels may also be independently related to postpartum mental health, beyond broader socioeconomic disadvantage. The fact that eligibility for higher-level benefits is contingent on both income and employment status before childbirth, allows for some variation (i.e., there is not a perfect correlation between low income and receiving basic benefits), which makes our adjustments for these factors both feasible and meaningful. This finding is further supported by the IPWRA with overlap weights analyses restricted to fathers receiving basic or higher-level benefits. Furthermore, the analyses restricted to fathers with no prior history of mental health conditions can also lend additional support in the independent association of parental leave benefit levels on postpartum mental health, as socioeconomic disadvantage is associated with poorer mental health in the general population17,18. All these findings are consistent with those observed among first-time mothers in Sweden during the same study period using similar modelling strategies and sensitivity analyses15. However, because paid parental leave benefits are structurally linked to labour market conditions, the findings should not be interpreted as causal.
At the same time, although our analyses suggest an independent association from parental leave benefit levels, we observed no difference in mental health outcomes between fathers receiving minimum-level and higher-level benefits after adjusting for income and employment status. This finding is important, as days compensated at the minimum level may be used by fathers who are already in full-time employment to supplement their income, without necessarily engaging in the same level of caregiving as parents who take extended parental leave over consecutive days. This means that the health benefits of this group cannot be explained by parental leave use or by higher remuneration. Since the option to use these days to supplement income is likely more common among fathers with lower salaries and possibly poorer labour market conditions, these findings highlight the need for future quasi-experimental studies to disentangle the role of parental leave benefits themselves from the broader socioeconomic conditions associated with postpartum mental health.
Although our study focuses specifically on parental leave remuneration, it is important to acknowledge that the decision to take parental leave may depend largely on the level of compensation. As fathers typically earn higher salaries than women, taking parental leave often entails a loss of income in the absence of full compensation19. Consequently, fathers with weaker labour market attachment may delay taking leave until they are in established employment, or may result in mothers taking the majority, if not all, of the parental leave days during the postpartum period. Similarly, the capped remuneration for higher-level benefits of the Swedish system may create a financial disincentive for fathers with very high incomes to use parental leave. In both cases, fathers at the lower and higher income ends of the income distribution may be less able to benefit from the potential mental health advantages associated with parental leave use4 because of their labour market circumstances. In our study, fathers who did not take parental leave during the first postpartum year had higher odds of receiving any mental healthcare, outpatient care, and hospitalisation, which is consistent with this interpretation.
The close link between labour market participation and eligibility for higher-level benefits highlights the importance of situating postpartum mental health within a wider social determinants framework and by adopting a Health in All Policies approach20. Policies aiming to improve postpartum mental health among new fathers may therefore need to extend beyond parental leave remuneration alone and also address inequalities in the labour market. At the same time, the strong link between parental leave remuneration and labour market conditions, on the one hand, and between parental leave use and remuneration, on the other, highlights the need to consider how general socioeconomic inequalities shape research on parental leave and postpartum mental health for both parents. This issue is particularly important in Sweden given the increasing levels of labour market precarisation21, and in contexts with even stronger work-requirements for access to paid parental leave.
Generous parental leave system in terms of remuneration could help meet some of the Sustainable Development Goal (SDG), including improved health for both mothers and fathers (SDG 3: Good Health and Well-Being), while also promoting greater involvement in childcare and addressing gender role disparities (SDG 5: Gender Equality)22. Our study contributes to advance knowledge on men’s health highlighting the need to make fathers’ postpartum mental health more visible and creating opportunities for social interventions.
The main strength of our study lies in its use of data from national population registers thereby ensuring robust representativeness of the Swedish population23. This allowed us to capture to include 96% of fathers whose first child was born between 2007 and 2011. Additionally, our study encompassed a wide range of mental healthcare utilisation, ranging from milder problems to severe cases requiring hospitalisation. Lastly, all health data was derived from diagnoses or prescribed medication documented by medical professionals, effectively eliminating recall bias.
A key limitation of this study is the inability to fully rule out reverse causality between parental leave uptake and mental health outcomes in the first postpartum year. However, robustness analyses in which the temporal order of events was clearly established yielded similar patterns, strengthening the evidence that generous parental leave benefits may have a protective effect against the onset of postpartum mental health conditions. Nonetheless, since we measure postpartum mental health through healthcare utilisation (rather the onset of symptoms), the still the temporal order of events is difficult to measure.
Fathers can transfer all their leave days except for the reserved father’s quota. As a result, those experiencing moderate-to-severe mental disorders may forgo parental leave, with their partners extending their leave until the child becomes eligible for preschool. However, even after accounting for mental health before fatherhood or restricting the analysis to fathers without a history of mental health conditions, those who did not take paid leave still had higher odds of mental disorders.
It is important to acknowledge that the four parental leave benefit categories may reflect different underlying processes. The comparison between fathers receiving higher- versus basic-level benefits primarily reflects differences in eligibility for earnings-related parental leave benefits and benefit remuneration and is therefore the most straightforward to interpret. In contrast, the minimum-level and no parental leave categories reflect not only eligibility, but also decisions about whether and when to take parent leave during the child’s first year. These decisions may be influenced by family circumstances, including infant and maternal health, breastfeeding, and household finances. Consequently, estimates involving these groups may be more susceptible to residual confounding and should be interpreted with greater caution.
In Sweden, healthcare access is universal; however, use of healthcare services may still vary across population groups, particularly in relation to socioeconomic position. Our study did not include primary care data as this is not available at the total population level. Thus, we were unable to capture milder mental health conditions that may be managed without pharmacological intervention, for example, though psychological counselling or short-term sickness absence. In addition, our outcome for mild mental health issues may be underreported because men are more likely to postpone seeking medical care for mild mental health issues and tend to seek help only when the condition becomes more severe24. Nonetheless, we still observed an association between parental leave benefit levels and prescription antidepressants or anxiolytics.
Furthermore, it is important to note that our outcomes of specialist outpatient care and hospitalisation, reflect moderate-to-severe mental health conditions, are relatively rare in absolute terms, and should therefore be interpreted in the context of low baseline risk. Nonetheless, we believe these outcomes remain important despite their low prevalence. Severe mental health outcomes represent the most serious cases, and even small absolute differences may be meaningful at the population level. Moreover, rarer outcomes are less likely to be influenced by health-seeking behaviour or access to primary care compared with milder mental health conditions, and may therefore provide a more robust indicator of severe mental health burden.
Another limitation concerns our operationalisation of mental health, as we grouped all ICD-10 F diagnoses together, thereby combining a heterogeneous set of conditions (e.g., severe mental disorders, substance-related disorders and stress-related conditions). This aggregation may mask meaningful differences in associations across diagnostic categories, which could be particularly relevant when considering subgroup analyses such as migrants and non-migrants. However, this approach was necessary due to limited statistical power, which precluded disaggregation into more specific diagnosis groups. Nonetheless, one of our outcomes, prescription of antidepressants or anxiolytics, provides a more diagnostically specific proxy for common mental health conditions.
Moreover, income and employment status are reported annually, not monthly, in the LISA database25, meaning we measured these socio-economic variables in the year preceding childbirth. We believe this timing is appropriate, as parents generally establish themselves in the labour market before planning a family26, making these variables reflective of their financial and employment status prior to taking parental leave. Lastly, while we could not use a more detailed classification of birthplace, our categorisation is consistent with those used in prior studies27,28, ensuring comparability.
In our cohort of first-time fathers, generous parental leave benefits were associated with better postpartum paternal mental health compared to fathers receiving basic benefits or those who did not take leave. However, our study also discusses the importance to further disentangle the role of general socioeconomic inequalities in postpartum mental health in order to determine whether parental leave systems mitigate or reproduce inequalities in health.
Source: www.nature.com



