Job loss during pregnancy and the risk of miscarriage and stillbirth

Alessandro Di Nallo

Selin Köksal

Alessandro Di Nallo and Selin Köksal discuss their research looking at job loss during pregnancy and the risks of miscarriage and stillbirth. 

Pregnancy loss, an issue often overlooked, affects a significant portion of pregnancies, as one in five to one in six gestations end before delivery. While the prevalence of pregnancy loss is predominantly attributed to the factors like advanced maternal age and genetic anomalies, the influence of other elements has not been extensively explored.

Research indicates that stressful life events, including financial instability, extended work hours, and economic downturns, may increase the risk of miscarriage or stillbirth. However, there has been a lack of knowledge on whether losing a job during pregnancy contributes to this increased risk. This aspect is particularly pertinent in the UK context, where the impact of job loss can be more severe due to the relatively brief duration and limited amount of unemployment benefits compared to other OECD countries.

Pregnancy loss may have profound and far-reaching consequences. We know that losing a pregnancy is associated with an increased risk of long-term depression and anxiety. They are also costly for healthcare systems as the financial impact of miscarriages is estimated almost top half a billion annually in the UK.

Our analysis

We used data from the first twelve waves of Understanding Society and examined 8,142 pregnancies that occurred between 2009 and 2021. Our focus was to detect the events of miscarriage or stillbirth, and to identify cases where women or their partners experienced an involuntary job loss, such as dismissals or redundancies.

Miscarriages (pregnancy loss before the 24th week of gestation) represented approximately 11.5% of all pregnancies in our sample, though this number might be an underestimate. Often, pregnancy losses occur within the first month when women have not yet realized their pregnancy. Additionally, about 0.45% of these pregnancies resulted in stillbirths (pregnancy loss after the 24th week of gestation).

Understanding Society data offers detailed information on the month of conception and the timing of a job loss, reported by women or their partners. This granularity allowed us to accurately detect the cases where the job loss preceded or coincided with the month of pregnancy loss. Our analysis specifically focused on involuntary job loss, differentiating it from other forms of employment termination, such as contract completion, illness, or parental leave.

We acknowledge that various factors might influence both the risk of experiencing job loss and the risk of an involuntary pregnancy end. To address this issue, we considered a range of other variables such as socio-economic background, age, parenthood status, educational level, self-reported health status, history of previous miscarriage, marital status at the time of conception, cohabitation with a partner, the partner’s age and employment status, overall household income in our statistical analyses.

Our results

Our findings revealed that women who experience a job loss during pregnancy have higher risk of losing their pregnancies. In cases where there is no job loss, the probability of pregnancy loss stands at about 12%. However, this risk increases up to 17% when the woman or her partner had lost their job.

A figure displaying Predicted probability for risk of miscarriage or stillbirth for pregnancies affected and not affected by woman’s or her partner’s job loss. (A–C) The results obtained from Model 1 (baseline + woman’s demographic characteristics and prior miscarriage), Model 2 (Model 1 + woman’s SES and partner’s characteristics), and Model 3 (Model 2 + partner’s SES and household income), respectively. The 95% CIs are shown.

Predicted probability for risk of miscarriage or stillbirth for pregnancies affected and not affected by woman’s or her partner’s job loss. (A–C) The results obtained from Model 1 (woman’s demographic characteristics and prior miscarriage), Model 2 (Model 1 + woman’s SES and partner’s characteristics), and Model 3 (Model 2 + partner’s SES and household income), respectively. The 95% CIs are shown.

To validate these findings, we replicated the analyses using various employment scenarios: an involuntary job loss, an expected end of a contract, a job separation due to another reason (illness or an unspecified motivation, for instance), and no change in employment. We found that voluntary job termination, or those not classified as involuntary, didn’t have the same influence on pregnancy loss risk as involuntary job loss did.

We also considered the timing of job loss by looking at: losing a job between 12 months and 1 month before conception; losing a job during pregnancy; losing a job after pregnancy (between birth and 12 months later); losing a job at another time or not at all. Again, our findings indicated that job loss during pregnancy is a significant risk factor for pregnancy loss, highlighting that it’s the timing rather than job-related stress experienced beforehand to be critical.

Finally, we examined woman’s job loss separately from her partner’s. The results here showed us that both women’s and their partners’ loss of work were associated with an increased risk of pregnancy loss.

What the results could mean

Our findings support the hypothesis that a job loss, particularly when it was unexpected or involuntary, as in the form of dismissal or redundancy, can significantly decrease the probability of carrying a pregnancy to term. This finding can be explained by the greater amount of stress associated with sudden and involuntary job losses. This hypothesis is compatible with previous research that has established a link between the release of stress hormones, such as cortisol, and the increased risk of pregnancy loss, further reinforcing the connection between psychological stress and reproductive outcomes.

A complicated picture

We explored several different scenarios in our study, but it’s worth noting there might be more to the story. For example, couples dealing with health issues could face a double burden of job instability and pregnancy complications. Moreover, couples from lower socio-economic backgrounds, could be more vulnerable to the ups and downs of precarious jobs.

We acknowledge that our study was limited in terms of sample size. So, it would be helpful to replicate our findings in larger samples, possibly by making use of birth records or hospital data. Plus, getting a clearer picture of pregnancies that don’t go full-term, especially those ending in abortion – which are not always reported in surveys – could really give us a fuller understanding of what’s going on.

Nevertheless, Understanding Society data provided us with a unique opportunity. It allowed us to distinguish between different types of job loss (i.e. whether it was anticipated or involuntary), an aspect that is often overlooked in previous studies, which mainly focused on unemployment rates at the national or subnational level.

What are the implications?

Data infrastructures. Many countries, including the UK, lack comprehensive data on miscarriages. More importantly, what’s often missing is a detailed socio-demographic background information that could help us better understand the underlying causes of miscarriages and stillbirths. Hence, building a data infrastructure which brings together pregnancy loss records and individual socio-demographic information would certainly allow researchers to fully uncover the complexity behind pregnancy losses.

Context matters. The UK welfare state is designed to keep people out of poverty, rather than to fully replace their income. Thus, unemployment benefits in the UK are relatively low compared to the rest of Europe. If our research was replicated in other countries, with different levels of welfare benefits, we could see whether more generous welfare regimes are better at mitigating the adverse consequences of job loss on pregnancy outcomes. It could be that a more generous welfare setting is better at helping people overcome the economic and psychological stress of a job loss.

This blog post is based on research which can be read in this paper.

About the authors

Alessandro Di Nallo is a research scientist at the Max Planck Institute for Demographic Research in Rostock and research fellow of the Dondena Centre for Research on Social Dynamics and Public Policy, Bocconi University in Milan. He is a social demographer interested in the life course and reproductive health.

Selin Köksal is an assistant professor in demography at London School of Hygiene and Tropical Medicine. She is a social demographer interested in fertility and reproductive health. Her current research explores the social and biological factors shaping reproductive experiences and their further consequences over the life course.

Leave a Reply

Your email address will not be published. Required fields are marked *