I’m very happy to share that my first paper has been accepted for publication 🎉
Together with my great co-author Nayeon Lim we look at early public childcare in Germany.
More details and the paper link below 👇
www.sciencedirect.com/science/arti...
Very nice article discussing my new paper studying the effects of NAFTA on mortality: www.nytimes.com/2026/03/13/u...
Full paper is here: users.nber.org/~notom/resea...
I’m guessing it’s Lawrence Bragg (b. 1890), who was awarded the prize in 1915 jointly with his father.
Paper in collaboration with @rannveigkhart.bsky.social, @hfsunde.bsky.social, @neilmdavies.bsky.social , and @torvik.bsky.social.
Key insights (2/2)
• 10 years later, employed grandmothers are 12% less likely to work full-time, compared to a 2% reduction for grandfathers. Women also see larger income drops.
• The gendered patterns in infections + employment suggest women still are more involved in informal childcare provision.
Key insights (1/2)
• Respiratory infections jump in the first years of grandparenthood: +56% for women, +31% for men
• Grandparents are slightly less likely to see a doctor for mental disorders (−4.5%) & cardiovascular issues (−3.3%)
• Grandmothers have fewer musculoskeletal-related visits (−3.8%)
Figure showing event-study plots centred around the birth of one's first grandchild. The left plot shows health changes, the right shows labour market changes (separated by grandparent gender).
The image shows the following abstract: The Cost of Caring: Gendered Health and Labour Market Effects of Grandparenthood While the effects of the transition to parenthood are well-researched, less is known about how the transition to grandparenthood affects health and labour market outcomes. Using comprehensive Norwegian register data covering the entire population born between 1950 and 1960, we examine the effects of first-born grandchildren born during 2007–2018. Employing event-study models with person-year records, we compare grandparents to not-yet grandparents. Our findings reveal a sharp increase in the likelihood of respiratory infections during the first two years of grandparenthood, with infections increasing by 56% for women and 31% for men. Additionally, grandparenthood modestly reduces the likelihood of doctor’s visits related to mental disorders (4.5%) and cardiovascular health (3.3%). Grandmothers also see a decline in musculoskeletal-related visits (3.8%). These health-related changes coincide with notable gendered effects on labour market participation. Ten years after the birth of their first grandchild, employed women are 12% less likely to hold full-time positions compared to a 2% reduction for men. Overall, our findings demonstrate that the transition to grandparenthood significantly reshapes health and economic outcomes for both women and men. The larger effects observed for women likely reflect their greater involvement in informal childcare provision. Our results underscore the intersection of health, family dynamics, and gendered labour market behaviours in late adulthood.
New preprint📈📉
What happens to health and work when people become grandparents? Using Norwegian register data on all individuals born 1950-1960, we use event-study models comparing grandparents to not-yet grandparents to track changes in health and labour supply.
🔗 www.ssrn.com/abstract=571...
Coverage of our new @nber.org working paper @upshot.nytimes.com @sangerkatz.bsky.social. Really appreciate the perspectives of the various folks they interviewed.
www.nytimes.com/2025/03/17/u...
Interested in the link between social media and mental health? Come work at NIPH (Oslo, Norway) as a PhD student! It’s a paid PhD position, you’ll have access to really good data, and a very nice work environment.
#SocSky #EpiSky #PublicHealth 🧪
🧪 How are risk factors like parental mental illness and low education distributed across families? This cool new paper explores how couples resemble each other — both before and (even more so) after meeting. A highly recommended read for those interested in intergenerational mobility + genetics.
a sense of purpose, and may increase one's social network (e.g. meeting other parents, teachers etc). These factors could be protective in terms of mental health. I want to emphasise though that we cannot conclude about mechanisms based on our data. (2/2)
Personally, I think both selection into parenthood and "lifestyle effects" of parenthood are possible explanations. It's not random who becomes a parent, parents typically are healthier. At the same time, parenthood also often results in a need for structure and routines in daily life, gives (1/2)
Good question! Unfortunately, from the data we cannot know why people do/don’t see their GP. However, it seems unlikely that children keep parents too busy to visit their doctor at all ages (incl 50+). Parents also had a higher risk of certain symptoms, suggesting they don’t avoid seeing their GP.
Absolutely! Unfortunately we don’t have health care records that stretch long enough to look at mental illness in early life. But we could (and did) control for marital status. Regardless of controls though, we cannot conclude about causality. So we see selection into parenthood as a key factor.
🧪
🙋🏻♀️ I’m an interdisciplinary social scientist with an econ/psychology background. www.fhi.no/en/ab/depart...
Also happy to be included! I work on socio-economic inequalities in health in Norway.
🙋🏻♀️ (if there’s still room!) I do research on the drivers of mental health inequalities using population registry data📉
Overall, parenthood appears to be an important indicator of mental health inequalities in the Norwegian population. Its growing importance may suggest stronger selection into parenthood across time and cohorts.
#MedSky #PublicHealth #MentalHealth
Key insights (2/2):
• From 2006 to 2019, the mental health gap between parents and the childless has widened significantly
• Not all diagnoses and symptoms were negatively associated with parenthood; parents had higher odds of certain adverse mental health outcomes
Key insights (1/2):
• Sibling- and twin-matched analyses were largely consistent with population-based results
• The difference in mental health was especially large for men and among those with only compulsory education
Parenthood, Mental Disorders, and Symptoms Through Adulthood: A Total Population Study ML Andersen*, HF Sunde, RK Hart, FA Torvik *Corresponding author(s). E-mail(s): MariaLyster.Andersen@fhi.no; Abstract During recent decades, parenthood has declined in many Western countries. Simultaneously, mental disorders have become more prevalent. We investigated the link between parenthood and mental health in the entire Norwegian-born population aged 31 to 80 from 2006 to 2019 (n=2,234,087). We used logistic regression models on national register data and included sibling- and twin-matched analyses to address unobserved confounders. Parenthood was associated with a lower risk of mental disorders, including depressive and anxiety disorders. For symptoms related to mental disorders, fathers had a reduced risk, while mothers had a slightly elevated risk. Mental health disparities between parents and non-parents were greater among men than among women and persisted across adulthood, before reducing at older ages. Our main findings were largely consistent in sibling- and twin-matched designs. The disparity between parents and non-parents increased over the study period, suggesting stronger selection into parenthood. Our findings highlight parenthood as a significant indicator of mental health inequalities, with its importance growing over time. Keywords: Mental health, parenthood, childlessness, social inequality, fertility JEL Classification: I14, J13, J16
Two figures. Top figure (a) shows the prevalence of mental disorders and symptoms by parent-status, age, and gender. X-axis is age, and y-axis is prevalence in percent. The prevalence curves for mental disorders are higher for the childless than for parents. For symptoms, the prevalence curve is higher for childless men than fathers. For women, the two lines are much closer. The curve for childless women is higher until approximately age 40, then the two curves are similar until age 70. After 70, the prevalence curve for mothers is above the curve for childless women. Bottom figure (b) shows estimated odds ratios for a variety of logistic and conditional logistic regression models. For mental disorders, all estimated odds ratios are below 1, meaning that parents have a lower likelihood of having a mental disorder. For symptoms, the estimated odds ratios vary, sometimes above 1 (for women) and sometimes below 1 (for men). The magnitude of the estimates differ between the models of symptoms and disorders. For disorders, estimated odds ratios are in the range 0.44-0.71, while for symptoms the range is 0.80-1.19.
New preprint 📈📉
🧠👶🏻 We explored the link between parenthood and mental health in over 2M Norwegians using primary health care records. We found that parents generally experienced fewer mental disorders, even after accounting for factors like education and marital status.
🔗 doi.org/10.1101/2024...
Hello all Oslo-people:
Next Wednesday (20th of Nov), my colleagues and I will be at Kulturhuset and present research findings about social differences and mental health. The event is open to all (but will be in Norwegian): www.fhi.no/om/kurs-og-k...
Recently accepted to #REStud, ``The Child Penalty Atlas" quantifies the unequal effects of parenthood on men and women’s employment in 134 countries.
From Henrik Kleven, Camille Landais & Gabriel Leite Mariante:
www.restud.com/the-child-pe...
#Fertility, #Childcare, #Children, #Youth
🙋🏻♀️ And thanks for making this!!
Some articles from @reveconstudies.bsky.social are more relevant than ever before:
"In all countries, respondents greatly overestimate the total number of immigrants, think immigrants are culturally and religiously more distant from them, and economically weaker"
academic.oup.com/restud/artic...
Many Stata users will be familiar with the -motivate- command. Only today did I learn of the -motivatedolly- command.