Gestational Diabetes Mellitus (GDM) is one of the most common and growing pregnancy complications, affecting 2–10 % of pregnancies in the United States (Centers for Disease Control and Prevention, n.d.). GDM has been linked with elevated risks of adverse health outcomes for both mother and infant (McIntyre et al., 2019). Individuals with GDM are at greater risk of developing Type 2 diabetes as well as cardiovascular disease (CVD) (McIntyre et al., 2019). Short-term adverse health outcomes for infants of GDM-diagnosed individuals include increased risk of macrosomia, hypoglycemia, and respiratory distress (Murray and Reynolds, 2020). Long-term adverse health outcomes for infants include increased risk of obesity and cardiometabolic diseases (Murray and Reynolds, 2020). Thus, identifying modifiable risk factors for GDM is critical for preventing GDM and promoting maternal and child health.

Recent evidence suggests psychosocial factors such as emotional distress and exposure to stressful conditions may play a role in the pathogenesis of Type 2 diabetes, likely through chronic allostatic load involving neuroendocrine, metabolic, and inflammatory pathways (Hackett and Steptoe, 2016; Chen et al., 2016). Furthermore, recent studies have reported that the stressors from different domains during pregnancy were associated with an increased risk of GDM, including financial- and work-related stress (Mason et al., 2016; Versteegen et al., 2021; Lissåker et al., 2022). Finally, findings from the Nurse Health Study II showed that stressful experiences in childhood and adolescent were also associated with a higher risk of GDM (Mason et al., 2016). Despite these findings, research on the interactive role of stress in both childhood and adulthood on GDM, through the life-course perspective, is lacking. Emerging evidence suggests that adverse psychosocial experiences during childhood, particularly adverse childhood experiences (ACEs), may moderate the effects of adult stressors on cardiometabolic outcomes, influencing associations between work-related stressors and hypertension and atherosclerosis (Matthews et al., 2022; Westerlund et al., 2012). However, studies specifically focusing on diabetes are lacking.

Thus, we aimed to elucidate the relationship between stressors from different life stages and their implications for GDM risk. We aimed to investigate how childhood stressors (i.e. ACEs) and maternal stressors (both work-related and non-work-related) interact to influence GDM risk in a racially diverse population of pregnant individuals in the United States. We hypothesize that ACEs and maternal stressors would be independently and jointly associated with GDM risk.

In this prospective study among racially diverse working pregnant individuals, we found that experiencing childhood abuse and maternal work stressors during pregnancy were independently and jointly associated with an increased risk of GDM. Specifically, participants who reported having maternal work stressors had 2.11 times the odds of developing GDM compared to those without work stressors. Also, participants who experienced childhood abuse (i.e., physical, sexual, or emotional abuse) had 1.49 times the odds of developing GDM compared to those who did not experience childhood abuse. Among individuals who experienced childhood abuse, those with maternal work stressors had 3.80 times the odds of developing GDM compared to working individuals without work stressors.

This examination of the individual and joint associations between childhood stressors and maternal adulthood stressors on GDM risk is an important contribution to the literature. Very few studies have separated pregnancy related stress by domains (Mason et al., 2016; Chen et al., 2020; Schoenaker et al., 2019). Our findings are consistent with results from a large prospective study in Sweden using the Job Exposure Matrix to measure work stress (Lissåker et al., 2022). Pregnant individuals with higher stress levels at work had 36–58 % increased risk of GDM compared to individuals with lower levels (Lissåker et al., 2022). In addition, work-related stress has been associated with a higher risk of T2D (Pena-Gralle et al., 2022) which share some common pathophysiological features with GDM, further corroborating the findings from the current study.

Unlike some other studies (Chen et al., 2020; Hosler et al., 2011; Silveira et al., 2014), we did not find the association between non-work stressors and GDM risk. This is likely due to the different study populations and types of stress captured by different studies. Given the impending re-submission of “work requirements” for Medicaid patients in several states (Sommers et al., 2019), our study provides additional evidence of exacerbated health outcomes for Medicaid beneficiaries working in high-stress occupations. Beyond the broader implications of our findings, it’s imperative to recognize the specific impact on predominantly income-eligible Medicaid populations. Individuals in this demographic face unique challenges such as limited access to healthcare services and other socioeconomic factors (Bellerose et al., 2022; Adams et al., 2005). Our study highlights the heightened maternal work stress levels in this population, potentially contributing to an increased risk of GDM. These findings underscore the urgency of developing targeted interventions and policies to improve maternal health outcomes within Medicaid-eligible populations.

Our results on the association between ACEs and GDM risk are consistent with two other cohort studies in the US (Mason et al., 2016; Versteegen et al., 2021). Mason et al. investigated associations of ACE subcategories with GDM and found significant positive associations of physical or sexual abuse with GDM risk using data from the US Nurses’ Health cohort (Mason et al., 2016). Another smaller prospective cohort study of 266 pregnant individuals performed by Versteegen et al., also demonstrated a positive association between higher ACE scores and GDM risk (Versteegen et al., 2021).

Although there has been an increased attentiveness on examining long-term, cumulative effects of risk factors on chronic diseases through a life-course framework, studies in psychosocial health and its impact on chronic diseases across the life-course are still limited to hypertension (Repetti et al., 2002), coronary heart disease (Repetti et al., 2002; Lynch and Smith, 2005) and type 2 diabetes (Lynch and Smith, 2005; Ben-Shlomo and Kuh, 2002). This study represents another important contribution to our understanding in this area by examining the GDM risk among working pregnant individuals. By exploring the intricate interplay between childhood abuse, maternal work stressors, and the risk of GDM among a racially diverse cohort of pregnant individuals, our findings shed light on potential pathways underlying these associations. One plausible explanation for the moderating effect of ACEs on the relationship between maternal work stressors and GDM risk is the vulnerability hypothesis (Belsky and Pluess, 2009). According to this hypothesis, individuals who have experienced early life adversity may exhibit heightened vulnerability to the detrimental effects of stressors encountered later in life. Specifically, childhood abuse may expose individuals to the impact of subsequent stressors, such as work-related stressors during pregnancy, thereby amplifying their susceptibility to developing GDM. This vulnerability could manifest through dysregulation of physiological stress response systems, including alterations in neuroendocrine and immune pathways, ultimately contributing to the development of GDM (Phillips and Jones, 2006). Another potential pathway through which ACEs may contribute to the association between maternal work stressors and GDM is via mediation. Adult stressors could serve as mediators in the pathway from ACEs to GDM. However, our mediation analysis did not support this hypothesis (results not shown but available upon request). Further research exploring the underlying mechanisms through which ACEs heighten the effects of maternal work stressors on GDM risk is warranted to elucidate these pathways and inform preventive interventions targeting vulnerable populations.

Our study has several strengths. First, this is a prospective study and included racially diverse pregnant individuals, particularly Black (47 %) and Hispanic (16 %) pregnant individuals who are under-represented in many other studies. Second, GDM was clinically measured based on laboratory results with high validity (Obstet. Gynecol., 2018). Lastly, we collected extensive sociodemographic and lifestyle information adjusting for several confounders of GDM.

Our study has several limitations worth mentioning. First, it is important to indicate our study is observational in nature. Additionally, while we included many potential confounders, certain lifestyle factors such as physical activity, sleep, and diet, which may also influence GDM risk (Zhang and Ning, 2011), were not assessed or included in our analysis. This omission is acknowledged as a limitation. All reports of ACEs are collected in adulthood thus introducing the potential for recall bias, however evidence has shown the ACEs questionnaire as a valid measure of early life adversity even if collected retrospectively (Hardt and Rutter, 2004). Although we asked participants to report stressors during the pregnancy, the questionnaire was administered around 30 weeks of gestation. Since GDM screening occurred between 24 and 30 weeks gestation, we cannot ascertain the diagnosis of GDM would influence participants’ responses. However, given we asked them to report stressful life events during pregnancy and not their perceived stress levels, the influence of GDM diagnosis would be minimal. Furthermore, we had a relatively large number of participants who did not answer questions related to ACEs and maternal stressors, which is common and are likely due to the sensitive nature of the questions. We thus treated “non-response” as a separate category in the data analysis. In our sensitivity analyses, findings did not change substantially after conducting complete case analysis and multiple imputation. In addition, the study sample size may be underpowered to estimate a dose-response relationship between ACE scores and GDM risk given we only had 12 participants with ACE scores equal or above 4 (a more common cutoff applied in other studies). Lastly, the exclusion of non-working participants limits the generalizability of our findings to working pregnant individuals, as our study was specifically designed to examine the risk of GDM in this group, hypothesizing that work-related stressors uniquely impact GDM risk.

Our study underscores the importance of addressing maternal work stressors and childhood adversity in reducing the risk of GDM, particularly among vulnerable populations such as Medicaid beneficiaries. Public health efforts should focus on developing interventions aimed at improving access to affordable healthcare services before and after pregnancy, creating supportive work environments, addressing social determinants of health, and providing education and resources for stress management during pregnancy. Additionally, interventions targeting childhood adversity through early identification and intervention programs may help mitigate the long-term health consequences, including GDM risk. By addressing these psychosocial stressors across the life course, public health initiatives can improve maternal and child health outcomes and reduce the burden of GDM on healthcare systems.

Among US working pregnant individuals, having maternal work stressors was associated with a higher risk of GDM, which was further moderated by the childhood experience of abuse. Further investigation is warranted to confirm our findings and investigate work-related stressors before pregnancy, as well as strategies to mitigate work-related stress during pregnancy.