Why it is plausible that health affects fertility

Mental or physical health problems are related to fertility intentions, childlessness and birth rates (e.g., Lazzari and Beaujouan 2025; Liu et al. 2024; Kravdal, Flatø and Torvik 2025). While, to some extent, such relationships are a result of joint determinants of health and fertility that have not been adequately controlled for, they may also reflect effects of health on fertility through a variety of causal pathways. For example, some chronic and severe health problems may adversely affect education and income, with implications for partnership formation and dissolution, and for a couple’s interest in having children.

Women and men with serious health issues may fear that having a child, or an additional child, would not be a joyful experience, that they would not be good enough parents, or that the strains of parenthood could make their condition worse. Additionally, there may be effects on fecundity and even on the effective use of contraception.

Even for individuals in good mental and physical health, childbearing can be a daunting experience. Some women experience particularly challenging pregnancies with, for example, severe nausea, health risks due to eclampsia, particularly difficult deliveries, or adverse mental or physical postnatal implications even after a relatively normal pregnancy and delivery. The latter include, for example, postnatal depression, pelvic pain, and perineal tears. These women and their partners may be less interested than others in having an additional child: even short-lived mental and physical health problems may be perceived as heavy burdens, and women may not want to take the risk of assuming them a second time. Some may not be physically able to have more children either.

Similarly, while it is often exhausting to care for very young children, some infants are more challenging for parents than others, for example because of their sleep patterns or colic. These issues may persist for several months, or even years, sometimes leading to health problems for the parents themselves. In such situations, some parents may not be very eager to have an additional child, who may affect them in the same way. 

New empirical evidence

Two studies based on nationwide Norwegian register data (Grundy and Kravdal 2025) and on the large Mother, Father and Child survey with added register data (Kravdal, Grundy and Hart 2026) provide evidence for these hypotheses. The latter investigation was based on information about health status around the time of the first birth for about 37,000 women, and around the time of the second birth for about 29,000. It was shown, for example, that mothers who experienced nausea in their first pregnancy were less likely than others to ever have a second child, and they also waited longer if they had one (see Table 1). Women who had pre-eclampsia or (more rarely) eclampsia in their second pregnancy had a reduced probability of progressing to a third birth, as did those who were on sick leave after week 30 of the second pregnancy. 

Furthermore, second- and third-birth probabilities were substantially reduced among women whose previous birth was difficult, as indicated by an emergency caesarean section, and more markedly so among those who had a planned caesarean section (but not when performed for a breech presentation). This situation may signal problems in pregnancy or underlying health problems not captured by other covariates. (The link between delivery mode and second births was investigated thoroughly in Grundy and Kravdal 2025.) Additionally, the probability of having a second child was low among women who had postnatal depression after the first birth or whose first child had colic or slept poorly. Sleep problems were also related to a reduced third-birth probability.  

Certain chronic somatic or mental health problems experienced by the mother or father before the most recent previous pregnancy were also associated with the probability of having additional children (not shown in the table).  

The value of this research

How useful is such knowledge about associations between health and fertility, which may partly reflect causal effects? In principle, it may help us predict the fertility implications of certain types of health deterioration or improvement. Having a more reasonable picture of future fertility levels may in turn be valuable from a planning perspective and for political discussions. Current evidence suggests that the apparent rise in mental disorders among young Norwegians in recent years – and the possible continuation of this trend – has had and will have a modest impact on the national fertility level (Kravdal, Flatø and Torvik 2025). However, it is also useful to be aware of a lack of aggregate-level implications, and there may be other implications relating to other settings and other types of health changes. 

Clearly, when steps are taken to improve the prevention or treatment of various diseases, or increase the quality of patient care, better health and well-being is a goal in itself. However, the evidence referred to above indicates that there may be spin-offs in the form of higher fertility. This means that, if there is concern about low fertility (Kravdal 2025), attempts to prevent or cure diseases, or to make life easier for individuals with chronic or temporary health problems – including during pregnancy and for some time after birth – may serve a dual purpose. This is in line with the conclusions of a recent White Paper to the Norwegian government about the causes and consequences of low fertility (NOU 2026). 

References

Grundy, Emily, and Øystein Kravdal. 2025. “Mode of delivery of the first child and progression to a second birth in Norway.” Population-E 80 (3-4), 381-404. https://shs.cairn.info/journal-population-2021-0-page-381?lang=en

Kravdal, Øystein. 2025. “Should we be concerned about low fertility? A discussion of six possible arguments.”  Demographic Research 53(14), 373-418. https://doi.org/10.4054/DemRes.2025.53.14

Kravdal, Øystein, Martin Flatø, and Fartein A. Torvik. 2025. “Fertility among Norwegian women and men with mental disorders.” European Journal of Population 41 (1): 17. https://doi.org/10.1007/s10680-025-09739-5

Kravdal, Øystein, Emily Grundy, and Rannveig K. Hart. 2026. “Health influences in second- and third-birth probabilities in Norway”. Population and Development Review. http://doi.org/10.1111/padr.70066

Lazzari, Ester, and Éva Beaujouan. 2025. “Self-assessed physical and mental health and fertility expectations of men and women across the life course.” Demography 62 (2): 543-569.https://doi.org/10.1215/00703370-11873109

Liu, Aoxing, Evelina T. Akimova, Xuejie Ding, Sakari Jukarainen, Pekka Vartiainen, Tuomo Kiiskinen, et al. 2024. “Evidence from Finland and Sweden on the relationship between early-life diseases and lifetime childlessness in men and women.” Nature Human Behaviour 8 (2): 276-287. https://doi.org/10.1038/s41562-023-01763-x

NOU 2026. “Policy for new generations – Causes, consequences and measures related to low birth rates. English summary”. NOU 2026:2. Oslo: Ministry of Families and Children.

The post Health around the time of first birth matters for future fertility appeared first on N-IUSSP.

Background

There are good reasons to expect that health problems may lead to lower fertility. Individuals who are exhausted because of health problems or who struggle with daily functioning may not want as many children as others, and those with particularly severe conditions may be less likely to enter partnerships. Additionally, some mental or physical diseases or the treatments for them may reduce fecundity. However, there are also mechanisms that could lead to a relatively large number of children, or quite early childbearing, among individuals with health issues. For example, to the extent that health problems make schooling more difficult, this could increase fertility, especially among women. So far, demographers who study high-income countries have taken little interest in the possible importance of health as a fertility determinant. 

Most women and men of reproductive age are in good health, but some suffer, for example, from chronic physical diseases with a severe impact on their lives, and in recent decades there have been indications of a quite sharp rise in depression, anxiety and other mental disorders among teenagers and young adults in many high-income countries (Krokstad et al. 2022; Zhou et al. 2025). Has this increase contributed negatively to fertility? This question is especially relevant for the Nordic countries, where fertility was higher than in most other rich countries for many years but has declined more markedly since 2009 (Eurostat 2025). 

Associations between mental disorders and fertility in Norway

In a recent analysis (Kravdal, Flatø and Torvik 2025), we used data from Norwegian registers for the years 2008-2018 to estimate whether individuals who had consulted a general practitioner or a specialist for a mental disorder in a certain year were less likely to have a child two years later than individuals who had not consulted. We found that women with depression had reduced first-, second- and third-birth rates (Table 1). According to a simulation where the calendar year was set to 2018, these lower birth rates correspond to a reduction in completed fertility from 1.60 births per woman (close to Norway’s total fertility rate in 2018) among those without any of the mental disorders under study, to 1.34 among those who – hypothetically – have depression throughout their reproductive period (Table 1, column 4). Depression is even more strongly linked to lower fertility among men: the corresponding numbers for them are 1.41 and 0.90. 

The associations between anxiety and fertility were of similar magnitude, while fertility was even lower among individuals with bipolar disorder, eating disorder or personality disorder. The simulated completed fertility was lowest among women and men with schizophrenia: 0.36 and 0.16, respectively. However, to the extent that individuals with mental disorders enter parenthood, many of the estimates suggested that they did so at a relatively early age.

It is important to note that these observed associations reflect the effects of mental disorders on fertility – through a variety of causal pathways – but may also be a result of individual and societal characteristics that influence not only the probability of developing a mental disorder but also fertility. We carried out additional analyses controlling for partnership, education and income. These factors may be joint determinants of mental health and fertility, and may also mediate the causal effects of mental disorders on fertility. Additionally, we controlled for unobserved characteristics shared by siblings, by comparing siblings with and without mental disorders. These steps made the associations between mental disorders and fertility weaker, but many of them were still significant. The remaining associations may partly reflect causal effects of mental disorder on fertility, operating through factors other than the above-mentioned control variables. For example, women and men with mental health problems may fear that raising a child will be too burdensome, and therefore not want a child, or another child. However, the remaining associations may also be linked to characteristics unique to one sibling, such as experiences early in life, that have implications for both mental health and fertility. It is hard to imagine a statistical analysis that is entirely free from this kind of confounding. 

The contribution to the national total fertility rate

In our paper (Kravdal, Flatø and Torvik 2025), we also showed that Norwegian fertility would have been higher in 2008-2018 in a hypothetical scenario without mental disorders, but not by much: only 0.05 extra children per woman, (depression and anxiety, the most prevalent conditions, are the main drivers of reduced fertility in our simulation). In other words, if we disregard the fact that the causal effects of mental disorders on fertility are likely smaller than the estimated associations, and if we assume hypothetically that the prevalence of the disorders increases from zero to the overall level observed in 2008-2018, fertility will be reduced by 0.05. Clearly, the increase in the actual prevalence from 2009 up to now is different from this, partly because the data do not capture all cases of mental disorders, and especially not the least severe. Everything considered, it is hard to believe that the rise in mental disorders has contributed more than modestly to the half-a-child reduction of the total fertility rate in Norway since 2009. 

Conclusion

The apparent increase in mental disorders is undoubtedly a challenge. Poor health reduces people’s wellbeing, partly because of its adverse effects on their education and income, which in turn have implications for society. Our analysis of Norwegian data shows that individuals with mental disorders also have fewer children than others, although the increasing prevalence of these problems is unlikely to have influenced the recent fertility decline in any significant way. In settings with less generous welfare policies, mental health may be more strongly linked to birth rates. All in all, this suggests that mental disorders probably deserve more attention from a demographic perspective than they have received thus far. 

References 

Eurostat (2025). Fertility statistics. Fertility statistics – Statistics Explained – Eurostat 

Kravdal, Ø., Flatø, M. & Torvik, F.A. (2025). Fertility among Norwegian women and men with mental disorders. European Journal of Population, online first  

Krokstad, S., Weiss, D.A., Krokstad, M.A., Rangul, V., Kvaløy, K., Ingul, J.M., Bjerkeset, O., Twenge, J. & Sund, E.R. (2022). Divergent decennial trends in mental health according to age reveal poorer mental health for young people: repeated cross-sectional population-based surveys from the HUNT Study, Norway. BMJ Open, 12(5), e057654.

Zhou, W., Hei, B., Liu, Z., Liu, Y., Ding, Z., & Li, M. (2025). Global temporal trends in depression incidence among women of childbearing age: A 30-year analysis and projections to 2030. Social Science & Medicine, 372, 118005.

The post Mental disorder and (reduced) fertility in Norway appeared first on N-IUSSP.

ART: A response to delayed parenthood

The past decades have seen a global trend toward delayed parenthood. Driven by factors such as investments in education, professional career ambitions and evolving societal norms, more people are seeking to conceive later in life, when their fecundity is lower. Yet, the increased use of assisted reproductive technology (ART) reflects not only a delayed pattern of family formation but also greater availability of ART treatments, lower stigma around its use and rising awareness of fertility options. Countries with robust healthcare systems and favourable policies have seen notable growth in ART-conceived births. As a result, ART has gained relevance and popularity.

However, undergoing ART remains an often costly and lengthy process – as women might need to undergo several cycles before eventually conceiving and giving birth to a child – and is not equally accessible to everyone. If we are to study the impact of ART on fertility rates we need a better understanding of who gives birth after ART.

Who uses ART? Not just a question of age 

While age is a key driver of ART use, it does not tell the whole story. Some groups rely on ART more than others, and not solely because they start their families later. For example, ART services are used relatively more frequently by highly educated individuals, possibly due to greater financial means, more flexible jobs which make it easier to accommodate the intense ART schedules and geographical proximity to fertility clinics. A large proportion of ART users are also first-time parents.

In a recent publication (Chanfreau, Goisis and Kravdal 2025), we show the importance of paying attention to this variation across population subgroups when calculating the contribution of ART to fertility. For example, although ART births have increased over time across all groups considered, the much larger contribution of ART to the fertility of highly educated couples compared to lower educated couples gets lost if we only look at the national average. Calculating the contribution at the national level gives the impression that ART makes a fairly small difference overall, because, on average, the need for fertility treatment at the population level is relatively limited. Figure 1, which refers to Norway in the years 1986–2018, illustrates this point visually. It shows, by women’s level of education, the ART total fertility rate (absolute contribution) and the ART contribution as a percentage of the fertility rate of each subgroup (relative contribution). The contribution of ART to fertility rates has increased over time for all educational subgroups, both in absolute and in relative terms. However, absolute/relative differences have not narrowed over time and the ART total fertility rate among highly educated women is at least double of that of lower educated women.

How much does ART actually boost fertility?

ART undeniably contributes to fertility rates, but by how much? Quantifying its impact is more complicated than simply counting births. Clearly, without ART, many couples who want children would remain involuntarily childless or with fewer children than desired. But others who experience difficulty conceiving and receive treatment might eventually conceive even without the help of ART. Moreover, it is unclear how one should interpret naturally conceived births that follow an ART birth. On the one hand, this situation may signal “overtreatment”, i.e. that the couple would have been able to conceive without ART, suggesting that the ART birth should not be included when calculating the contribution. On the other hand, the use of ART may have been essential for that birth and either the treatment, the hormonal changes during pregnancy or the reduced stress of achieving a much-wanted birth may have helped overcome the infertility and made it possible to have a naturally conceived child later. It would then be appropriate to consider the subsequent births as contributions from ART. 

We argue that whilst it might be difficult to precisely establish the contribution of ART to fertility rates, an important step in this direction can be taken by considering a range of scenarios and reflecting on how plausible they are in the context under study. 

The future of ART and fertility policy

As ART becomes more mainstream, governments and healthcare systems must adapt to its growing role. Policymakers face several critical questions: Should ART be publicly funded to ensure equal access? How should societies balance the benefits of ART with ethical considerations? And what measures are needed to prevent misinformation about ART’s success rates?

Decision-makers in countries concerned about declining birth rates may look to ART as part of a broad set of policies to slow down or reverse the decline. However, other policy elements would likely be more effective, such as supporting an attractive work-life balance, providing affordable childcare, and encouraging gender equality. 

The availability of ART itself may be influencing behaviour. In the context of delayed childbearing, some individuals or couples may see ART as a form of insurance, a reliable fallback option in case conception does not occur naturally. If people overestimate its success rates, such beliefs will contribute to reduced fertility. This raises important questions about how ART shapes reproductive choices, given the various structural factors that encourage postponement, and whether current policies support informed decision-making. 

Moreover, the persisting, and possibly increasing, educational stratification of the contribution of ART warrants further attention. We need to better understand whether and to what extent these patterns can be explained by differences in needs, or whether they indicate social inequalities in access, use or success of treatment. 

In conclusion, ART has transformed the reproductive landscape, offering hope to millions while raising new challenges. While ART has helped many individuals and couples achieve parenthood, it is not a panacea for declining fertility trends. Its role in shaping fertility trends will only grow, making it an essential area of study for demographers, policymakers, and healthcare providers alike.

References

Chanfreau Jenny, Goisis Alice, Kravdal Øystein (2025) Conceptualizing and Measuring the Contribution of Assisted Reproductive Technologies to Fertility Rates. Population and Development Review, online first, https://doi.org/10.1111/padr.70009.

Kravdal, Ø., Flatø, M. & Torvik, F.A. Fertility among Norwegian Women and Men with Mental Disorders. Eur J Population 41, 17 (2025). https://doi.org/10.1007/s10680-025-09739-5

The post The growing role of ART in fertility rates in Norway appeared first on N-IUSSP.

The dissolution process

Several studies have shown that partnered individuals have relatively good health (Kravdal, Wörn and Reme 2022), especially if the relationship quality is high (Robles et al. 2014). Unfortunately, for many couples, the relationship becomes less rewarding after some time together and they may even struggle with serious disagreements or conflicts. At some point, at least one of the partners may feel that life, after all, would be better outside the union and take the decision to separate. Presumably, most parents take the expected consequences for their children into account in such deliberations. 

After the break-up, there will obviously be some relief from the tension that triggered it, but new conflicts may arise, about dividing up property or the children’s living arrangements, for example. Additionally, there may be setbacks such as financial hardship, childcare issues, or distress because of moving to another area. In some cases, these problems may be persistent or even get worse over time, although the overall life situation may still be better – at least for some of the family members – than if the union had remained intact. In other families, the problems may wane as new sources of support and emotional satisfaction are found, and after some time life quality is perhaps not much lower than before the relationship problems started. This kind of temporary difficulty is often referred to in the divorce literature as a “crisis response”, as opposed to the more “chronic strain” that may be experienced (Amato 2000). 

Such changes in relationship quality and life circumstances during the years before and after a union dissolution may have implications for the health of both parents and children, and a huge number of studies have addressed this issue. However, many factors influence both union dissolution likelihood and health, and even the investigations that have taken this into account in relatively advanced ways have provided rather mixed evidence. 

Union dissolution and mental health in Norway

In a recent study, we estimated how marital separation and dissolution of consensual unions affected the health of Norwegian parents and their first child if the child was no older than 18 years at the break-up (Kravdal and Wörn 2023). The analysis was based on nationwide register data, and the annual number of consultations with general practitioners (GPs) between 2006 and 2018 was used as a health indicator. In total, households were observed over about 800,000 years. We controlled for all measured and unmeasured factors that are constant over time by comparing the number of consultations at different times before and after the break-up for the same individual. 

As one might expect, the number of consultations where the patient was diagnosed with a mental health problem increased among both mothers and fathers for some years before a union dissolution (Figure 1), when the relationship quality typically deteriorates. The numbers dropped immediately afterwards, although within the post-dissolution decade that could be studied, they remained higher than long before the dissolution (e.g. four years or more). In the long run, mothers appeared to be slightly more adversely affected than fathers.

Consultations for mental health problems before the dissolution also increased for children, although less so than for parents (also after taking into account the generally fewer consultations for such problems among children; not shown). At this stage, the children are possibly influenced by changes in parental behaviour caused by their relationship problems. Interestingly, this increase in consultations for mental health problems continued and even intensified after the break-up. The patterns were similar for daughters and sons. At first sight, this may suggest that, while a break-up comes as a (partly expected) relief to the parents, it harms the children. In other words, the children’s interests may not have been taken sufficiently into account in the decision-making. This is not necessarily the case, however, as these children might have experienced even worse health consequences had their parents stayed together. 
The general pattern did not vary much across married or cohabiting couples. 

Physical health

In contrast to the findings for mental health, there was almost no increase in consultations for non-infectious physical diseases before a dissolution for any of the family members. This is consistent with the hypothesis that effects on physical health are less immediate than effects on mental health (Hughes and Waite 2009). However, numbers of consultations started increasing later, for mothers especially, with a smaller increase or stability for the others. The upturn among mothers was not much smaller than their pre-dissolution increase in consultations for mental health problems, but because mothers in general consult GPs three times more often for physical than for mental illnesses, the increase in consultations for physical health problems is much smaller in relative terms. A dip could be seen among mothers in the break-up year, but this was entirely a result of musculoskeletal ailments (not shown), which perhaps were underdiagnosed that year or reported as mental health problems. 

The more adverse trend for mothers than fathers is noteworthy. Although one might expect mothers to have more economic problems after a break-up, and also to be more burdened by childrearing, other problems may be felt more strongly by fathers, and earlier studies have not given a clear picture about the difference between the two parents. The trends in consultations suggest that children’s physical health is also negatively affected by their parents’ break-up. This seems to hold especially for daughters, although gender differences are not always clear, and appear to be model-sensitive.

One might expect life changes before and after dissolution to affect social interaction patterns, and perhaps to increase vulnerability to infections because of sleep deprivation or other stress-induced health burdens. However, the overall conclusion was that the number of consultations for infections remained quite constant through the dissolution process (not shown). The clearest exception was a temporary surge in the break-up year and subsequent year for mothers.

Conclusion

In summary, our analysis indicates that the health of the entire family may be influenced by a dissolution process, although the break-up itself may give certain health benefits. It is important to note that even if the parents appear to be on a favourable health trajectory after the break-up, that is not necessarily the case for the children. 

References

Amato, Paul R. (2000). The consequences of divorce for adults and children. Journal of Marriage and Family 62(4): 1269-1287.

Hughes, Mary E., and Linda J. Waite (2009). Marital biography and health at mid-life.” Journal of Health and Social Behavior 50(3): 344-358.

Kravdal, Øystein, Jonathan Wörn, and Bjørn-Atle Reme (2022) Mental health benefits of cohabitation and marriage: A longitudinal analysis of Norwegian register data.  Population Studies (online first) www.tandfonline.com/doi/full/10.1080/00324728.2022.2063933

Kravdal, Øystein, and Jonathan Wörn. (2023). Mental and physical health trajectories of Norwegian parents and children before and after union dissolution. Population and Development Review online first. http://doi.org/10.1111/padr.12530Robles, Theodore F., Richard B. Slatcher, Joseph M. Trombello, and Meghan M. McGinn. (2014). Marital quality and health: a meta-analytic review,” Psychological Bulletin 140(1): 140-187.

The post Health before and after union dissolution among mothers, fathers and children appeared first on N-IUSSP.

Background

It is a strong desire for most of us – in periods of life when we are single – to enter a romantic relationship that is based on mutual love and that develops into a consensual union or marriage of persistently high quality. We expect such an intimate relationship to make our life better. However, should we expect the improvement in life quality to include health benefits? Is it possible that the emotional satisfaction derived from a good partnership, or, for example, the care and support provided by a partner in times of need, contribute to better health? 

Several studies have shown lower mortality and better health, especially mental health, for the married than the non-married (although the difference may vary with the quality of the relationship), and the mortality gap between marital status groups is increasing strongly in some countries (Kravdal et al. 2018). Recent research has also documented that cohabitants too enjoy some health advantages compared to those who are single, while it is less clear how they fare compared to married couples. Theoretically, one might expect smaller health benefits from cohabitation than from marriage, because cohabitants tend to have lower relationship quality than married couples (which may also be one of their reasons for not marrying), and tend to live more separate lives (Lyngstad et al. 2011; Wiik et al. 2012). 

However, patterns reported in the literature typically reflect a combination of causal effects of being married or cohabiting, compared to being single, and so-called selection or confounding. The latter means that people’s resources, health, lifestyle preferences and various other characteristics affect both the chance of forming a union, the type of union that is chosen, and later health and mortality. The magnitude of this selection component depends on the type of statistical analysis that is done, but it is never eliminated completely. 

A recent study of within-individual changes in mental health before and after cohabitation or marriage

In a recent study, we focused on how an individual’s annual number of consultations with a general practitioner (GP) for a mental health condition changes over time, before and after entry into cohabitation or marriage (Kravdal et al. 2022). With this kind of analysis, where the situation of the same person is compared before and after an event, all individual and community characteristics that are constant over time and may affect both health and the chance of forming a union are automatically controlled for. This is a fortunate, but rare possibility in research, made possible in our case by existence of the very complete and accurate Norwegian registers covering the entire population. In our analysis of the period 2006–2019, we also took into account age (people tend to be older after cohabitation or marriage than before it), and several other covariates, such as education, income and number of children (although, as it turned out, these covariates had only a limited impact on our main conclusions).  

We noticed a decline in GP consultations for mental health conditions for several years before the start of a cohabitation, although there is too much uncertainty in the estimates to be more specific about the timing of the decline (see Figure 1). While trends in use of health care also depend on the (possibly varying) tendency to seek help for health problems, the observed development likely reflects an improvement in mental health during this period. Advantages from being in a romantic relationship are probably key forces behind this improvement. However, one must also be open to the possibility that this and other patterns observed in the data also reflect unobserved characteristics that vary over time and affect both union formation and health, or just effects of health on union formation. 

The decline in GP consultations before start of cohabitation is particularly pronounced for those who later marry. More detailed analysis of this group revealed some further decline between the start of cohabitation and marriage, but – interestingly – not after marriage. In fact, among cohabitants who married after, for example, four years of cohabitation, the development over the next decade was not more favourable than among those still cohabiting at that time, despite the higher chance of dissolution among the latter. 

To summarize, starting cohabitation does in itself improve mental health. Most of the improvement has already taken place by then. Also, formalizing the union by marrying does not have any impact. However, marriage is a marker of a relatively favourable earlier improvement in mental health. 

Among couples marrying directly, who are increasingly uncommon in Norway, mental health improves more before marriage – especially over the last pre-marital years – than among those who marry after cohabitation, and also improves for a few years afterwards (not shown in the figure). This may partly reflect the fact that couples who marry directly have been together for a relatively short period at the time of marriage – and then experience mental health benefits of their relationship that prior cohabitants had already experienced when their cohabiting relationship began. 

The broader perspective

Although the Norwegian study indicates that mental health improvements are linked primarily to the dating period before a union is formed – and more marginally to union formation or marriage itself – there may, of course, be other health or welfare benefits of these transitions. In other words, when a couple starts cohabiting or cohabitants decide to marry because they think this best serves their interests, they may well be right in their expectation. It should also be noted that the effects of union formation may differ across countries, and that other patterns may be observed in societies with larger legal differences between cohabitation and marriage than in Norway, or with different welfare systems. 

References

Kravdal, Ø., E. Grundy, and K. Keenan (2018). The increasing mortality advantage of the married: The role played by education. Demographic Research 38 (20): 471–512.

Kravdal, Ø., J. Wörn, and B.A. Reme (2022) The importance of cohabitation and marriage for mental health – a longitudinal analysis. Population Studies, online first 

Lyngstad, T. H., T. Noack, T., and P. A. Tufte. 2011. Pooling of economic resources: A comparison of Norwegian married and cohabiting couples, European Sociological Review 27(5): 624–635.

Wiik, K. Aa., R. Keizer, and T. Lappegård, T. 2012. Relationship quality in marital and cohabiting unions across Europe, Journal of Marriage and Family 74(3): 389–398.

The post The mental health benefits of cohabitation and marriage in Norway appeared first on N-IUSSP.

Background

The relationship between the number of children a woman or man has and their chance of dying within a subsequent period has been analysed in several studies of contemporary rich populations, where few have more than four children. (See Kravdal 2019 for a discussion of the usefulness of such research). Three pathways contribute to the relationship:

• pregnancy: this involves biological mechanisms with importance for certain diseases risks;
• responsibility for and contact with children: this has implications for people’s lifestyle (broadly defined) and availability of support, with favourable or adverse effects on health and mortality;
• selection: people’s health, socioeconomic resources, attitudes and other characteristics influence both the number of children they have, and later health and mortality.

The first pathway is, of course, only relevant for women. However, if a relationship between number of children and a certain disease risk is observed only for women, it cannot be concluded that biological mechanisms are the only drivers, as lifestyle effects and selection mechanisms may differ by sex, or be relevant to only one. For example, a link between childbearing and use of oral contraception is relevant only for women, and may contribute to the reduced incidence of Crohn’s disease among women with a young child (Kravdal et al. 2020a).

It is widely agreed that childless people have higher all-cause mortality than one-child parents, while those with two or more children have the lowest mortality. The evidence is more varied for larger families. Some researchers have observed slightly elevated mortality for those with many children compared to two-child parents, while others have reported a small continuous decline as the number of children increases above two. In particular, studies focusing on men, or on cardiovascular disease (CVD) mortality rather than all-cause mortality, suggest a higher mortality for those with relatively many children. The variation in results reflects differences in the age groups considered and in the variables controlled for. Additionally, the lifestyle effects of childbearing, or the selection mechanisms, may differ between countries because of social policies or other factors.

A Norwegian register-based study

Such investigations take account of the fact that a person’s number of children is partly a result of education, and is also linked to marital status. In a recent analysis based on register data for the entire Norwegian population, we also controlled for repartnering, spouse’s education (if any), county of residence, and whether the municipality of residence was a city (Kravdal et al., 2020b). Additionally, like in some (but not all) earlier studies, we took account of the fact that people with many children are likely to have started their reproductive career earlier, which is also known to be linked to mortality. The negative association between mortality and number of children up to two was confirmed. For example, childless women were found to have 69% higher mortality than two-child mothers, while the corresponding figure for men was 48% (Figure 1). Mortality was significantly lower for those with more than two children, but the differences were small. However, a slightly higher mortality appeared for men with four or more children compared to those with two when age at first birth was not controlled for.

The picture looks somewhat different when the focus shifts to CVD mortality, which accounts for a large share of all-cause mortality. Men with four or more children had a 5% higher chance of dying from CVD than two-child fathers (Figure 1), and their chance of dying from stroke in particular was 12% higher. Among women, CVD mortality among those with four or more children was not significantly different from that of two-child mothers.

Possible causal pathways

In order to shed light on the mechanisms responsible for the relationship between number of children and CVD mortality, we analysed the association between number of children and three lifestyle factors believed to increase the CVD risk: smoking, high alcohol consumption, and physical inactivity. Additionally, we considered four health conditions assumed to have adverse effects: obesity, hypertension, high cholesterol level, and diabetes.

Differences in smoking and alcohol use did not appear to contribute to the high CVD mortality of childless people compared to two-child parents. However, all other CVD risk factors seemed to play a role: Childless people were more frequently physically inactive (men at least), and more likely to be obese and to have high cholesterol, hypertension or diabetes. Those with four or more children tend to smoke less and have lower alcohol consumption, which helps to lower their CVD mortality with respect to two-child parents, while higher likelihood of obesity and less physical activity have the opposite effect.

The underlying pathways are likely complex. In addition to affecting the three life style factors just mentioned, which in turn influence mortality partly through the four health conditions, childbearing may affect women’s health through biological mechanisms. Furthermore, other lifestyle factors and health conditions may also play a role, and a number of factors not taken into account in the analysis may affect both fertility and lifestyle, health or CVD mortality. Finally, causality in the observed relationship between number of children and some of the measured lifestyle factors and health conditions may also run in the opposite direction. For example, a pre-existing situation in young adulthood (such as diabetes), may have influenced fertility.

Given this potential complexity, the generally more negative association between number of children and CVD mortality for women than men does not necessarily invalidate the common idea that childbearing increases CVD mortality for women through biological mechanisms. Several other pathways may counteract such a mechanism.

Conclusion

To conclude, the main pattern is that CVD mortality and all-cause mortality in Norway fall as the number of children increases or, for the largest families, remain almost unchanged. However, there is an exception with respect to CVD mortality: for men who already have three children, the disadvantages caused by – or at least associated with – further childbearing more than outweigh the advantages. This is apparently linked to a higher likelihood of being overweight, partly linked to less physical activity.

References

Kravdal, Ø. (2019). What kind of individual-level effects of childbearing would we ideally be interested in learning about? The important distinction between expected, unexpected, varying and general effects. Journal of Population Research 36: 1-12

doi.org/10.1007/s12546-018-9218-7

link.springer.com/article/10.1007/s12546-018-9218-7

Kravdal, Ø., P. Magnus, B. Moum, and M. L. Høivik (2020a). Association of childbearing with a short-term reduced risk of Crohn disease in mothers. American Journal of Epidemiology, kwz285.

academic.oup.com

Kravdal, Ø., Aa. Tverdal, and E. Grundy (2020b). The association between parity, CVD mortality and CVD risk factors among Norwegian women and men. European Journal of Public Health, ckz235.

academic.oup.com

The post More children linked to lower mortality in contemporary Norway, with exceptions appeared first on N-IUSSP.

Databases of prescriptions in Nordic countries

In all Nordic countries, one of the newest additions to the registers are prescription databases, which include all purchases of prescription medicine, except those made by the institutionalized population (Furu et al. 2011). People who purchase medicine typically use it either for preventive purposes (e.g. statins to prevent heart diseases for those with a high cholesterol level), or as part of a treatment. Some medicines are used for multiple pathologies; others for one specific disease and are therefore good indicators of this disease. Researchers may, for example, use the sum of all different medicines purchased within a certain period as a proxy for general health, or select certain medicines as indicators of specific diseases or group of diseases (Kuo et al. 2011). Prescription databases linked to other registers are therefore potentially very useful to demographers interested in health issues.

On the negative side, however, use of prescription medication is not a perfect indicator of morbidity as it is influenced by individuals’ consulting behaviour and the prescribing behaviour of doctors, as well as by the presence of a particular medical condition. Medication data must therefore be used very cautiously. Three types of use stand out as particularly promising for demographers and health researchers. First, if some individuals purchase more medicines than others, this may be taken as an indication that the former have poorer health if there is little reason to expect a corresponding variation in the use of health care and thus the chance of being diagnosed and getting a prescription. As a special case of this, there are some diseases, such as diabetes among the young (Type I), that have a generally very high chance of being diagnosed. Second, and similarly, one may learn something about determinants of health by studying changes in purchases over part of a person’s life course, provided that there is not a corresponding change in health care use over time for that person. (Using such a longitudinal procedure, one also controls for unobserved characteristics of the individual that are time-invariant, and that with other study designs might confound observed relationships between health and the potential health determinant under consideration). Third, one may compare medication purchases with data on other good indicators of health to learn about differences in health care use.

An example of the use of the Norwegian Prescription Database

In a study of the latter type, we used data from the Norwegian Prescription Database (started in 2004) and some other registers to estimate  various types of models for the purchase of medicine for eight circulatory conditions by people aged 50-79 during 2004-2008 (for details, see Kravdal and Grundy 2014). Differences in medication purchase by marital status were compared with differentials in circulatory disease mortality during 2004-2007. We found that the unmarried had 1.4-2.8 times higher mortality from the four types of circulatory diseases considered (see Table 1), which fits with patterns reported elsewhere. The most plausible interpretation of this – which is also supported by a few survey-based studies of the prevalence of circulatory conditions from other countries – is that the unmarried have a larger burden of such diseases, which implies that they need more medication. Yet, our results showed that they purchased either less or only slightly more medication for these diseases and their risk factors (see Table 1). Use among the never-married was particularly low, largely reflecting a low chance of making a first purchase (not shown in the table).

These results suggest that the unmarried, and especially the never-married, to a larger extent than the married, have circulatory disorders that are undiagnosed or for which they for other reasons underuse medication. Inadequate use of these potentially very effective drugs in such a large population group would, of course, represent a serious public health challenge, so further research is needed to shed light on this phenomenon.

This finding, made possible by the new type of register data, may provide a partial explanation for the increasing marital status differences in all-cause mortality and mortality from circulatory diseases observed in several countries, despite the rising prevalence of consensual unions among the unmarried. The gap between the married and the never-married has increased the most. Possibly, one reason for the growing excess mortality among the never-married is that they are especially likely to underuse medication for circulatory disorders, and that the resulting health disadvantage due to this has increased in step with the introduction of new medicines which have contributed greatly to the general reduction in mortality from such diseases (Ford et al., 2007; O’Flaherty et al., 2013).

References:

Ford, E.S., Ajani, U.A., Croft, J.B., Critchley, J.A., Labarthe, D.R., Kottke, T.E., et al. (2007). Explaining the decrease in U.S. deaths from coronary disease, 1980-2000. New England Journal of Medicine, 356, 2388–2398.

Furu, K., Wettermark, B., Andersen, M., Martikainen, J.E, Almarsdottir, A.B., & Sørensen, H.T.(2011). The Nordic countries as a cohort for pharmacoepidemiological research. Basic & Clinical Pharmacology & Toxicology, 106, 86-94.

Kuo, R.N., Dong, Y.-H., Liu, J.-P., Chang, C.-H., Shau, W.-Y., & Lai, M.-S.(2011). Predicting healthcare utilization using a pharmacy-based metric with the WHO’s Anatomical Therapeutic Chemical Algorithm. Medical Care, 49, 1031-1039.

Kravdal, Ø. & Grundy E. (2014). Underuse of medication for circulatory disorders among unmarried women and men in Norway. BMC Pharmacology and Toxicology 15:65.

O’Flaherty, M., Buchan, I., & Capewell. (2013). Contributions of treatment and lifestyle to declining CVD mortality: why have CVD mortality rates declined so much since the 1960s? Heart, 99, 159-162.

The post Innovations in register data: the value of linkages to prescription databases illustrated by an analysis of underuse of medication among the unmarried appeared first on N-IUSSP.