Maternal History of Adverse Childhood Experiences and Subsequent Infant Paternal Involvement

Background of the Study

Adverse childhood experiences (ACE) are traumatic events occurring before age 18 (0–17 years), reflecting different types of abuse, neglect, and household dysfunction.^[1] In the United States, 61% of adults reported experiencing at least one ACE, with females more affected than males.^[2] Exposure to ACEs correlates with various health risks, including depression, alcoholism, stroke, diabetes, and asthma.^[2] Mothers experiencing ACEs are linked to increased risks of adverse birth outcomes such as low birth weight and preterm birth, contributing to infant health disparities.^[3,4]

A father’s support during pregnancy can positively impact maternal behavior and child outcomes, such as improved prenatal care and decreased infant mortality.^[5–8] The Father’s involvement is measured through co-parenting and relationship with the mother, financial and emotional investment, quality of time spent, and expressed willingness to rear children, using self-reports, spouse/other reports, and interviews.^[9] In research, marriage or paternity acknowledgment (PA) has been used as proxies for paternal involvement. In the absence of direct data on paternal involvement, being married or establishing acknowledgment of paternity at birth generally suggests the presence of a partner and paternal involvement during the perinatal period.

Legal paternity (legal establishment of the identity of a child’s father) is the determination of the related rights and obligations of the father to the child. In the United States, it’s routine for children of married parents, while unmarried mothers can establish it voluntarily at birth through subsequent parental confirmation or as mandated by the court when requested by either parent. Legal paternity ensures children receive benefits from their fathers, such as health insurance and inheritance. The absence of paternal information on birth records or no PA often indicates nonmarital births and limited paternal engagement.

Studies found that fathers of infants with paternity legally acknowledged at birth tend to be more involved in their children’s lives and are more likely to support their children financially than fathers who establish paternity later or not at all.^[10,11] Acknowledging paternity at birth lowered infant mortality, preterm birth, and low birth weight risk, particularly for unmarried women with higher ACEs.^[12] Voluntary legal establishment of paternity is associated with longer exclusive breastfeeding duration, potentially decreasing maternal risk of breast and ovarian cancer, type 2 diabetes, high blood pressure, and infant risk of infections and adulthood obesity.^[13,14] Missing paternity was also strongly related to involvement with the child protection system, such as investigations or the termination of parental rights.^[15] Paternity status at birth may also be linked to early maltreatment outcomes, with variations by maternal race and ethnicity.^[16]

Maternal ACEs have been shown to biologically alter stress responses, potentially due to long-term changes in hypothalamic–pituitary–adrenal (HPA) axis functioning, leading to heightened sensitivity to stress and negative cues.^[17] This, in turn, may influence maternal behavior and the dynamics within relationships, possibly influencing paternal engagement. This biological alteration, coupled with the psychological challenges such as trust and trauma, which ACEs often precipitate, may compound difficulties in creating stable relationships vital for paternity establishment.^[18] Together, these biological and psychosocial effects of ACEs intricately weave into the fabric of maternal behavior, potentially influencing the acknowledgment of paternity.

Objectives of the Study

The Monroe County Mothers and Babies Health Survey (MBHS) is a comprehensive research initiative aimed at assessing the health and well-being of expectant mothers and infants in Monroe County, New York, USA. Extensive data collection and analysis provide valuable insights into maternal and infant health outcomes in the region.^[19] Utilizing the substantial availability of prenatal data and maternal history of ACEs within this cohort as well as paternity status at birth, we conducted an exploratory investigation to examine the potential association between maternal history of ACEs and the establishment of legal paternity for infants. It’s important to note that this study was exploratory.

Study Subjects

Our study utilizes the MBHS database (N = 1,879), encompassing data from single-birth mothers in Monroe County, NY, surveyed to understand health outcomes and care access from 2015 to 2017. Surveys, consisting of 200 questions, offered in English and Spanish, were completed either electronically via REDCap or by mail. The response rate averaged 80 monthly over the 24-month data collection period, with balanced participation across income levels. The completeness of electronic records was checked and documented. Paper-based responses were processed through double-data entry. Survey data were linked with birth certificates and census tract data. For this study, the subjects included in the main analyses were mothers who reported their information (e.g., demographics, pregnancy intention, health behavior during pregnancy) as well as their baby’s father’s information (i.e., father age and father education) (1,556 [83%] of 1,879).

Study Variables

Statistical Analysis

Descriptive statistics for maternal and paternal characteristics were reported as frequencies and percentages (N [%]) for categorical variables and mean value and standard deviation (Mean [SD]) for continuous variables. The association between ACEs and the risk of CM-PA was assessed using logistic regressions because the outcome is dichotomous. In the main analysis, PA was treated as a binary variable (vPA vs. CM-PA), excluding the No-PA group. In the sensitivity analysis, the No-PA group was included and combined with the CM-PA group, leading to the classification of PA into the binary comparison of vPA versus suboptimal PS (suboptimal PS = No-PA + CM-PA). This approach implies that in the main analysis, the father’s age and education variables were included, whereas in the sensitivity analysis, these variables were omitted due to missing data. This omission was necessary to prevent the exclusion of No-PA individuals who had missing data for these variables, as subjects with any missing value(s) were excluded from the analysis. For each ACE encoding method, which included a continuous total ACE score, three binary domains, and ten binary itemized questions, the analyses proceeded as follows: an initial unadjusted examination of the association between ACE and PA status was conducted, followed by a subsequent adjusted analysis controlling for covariates selected in a backward stepwise manner. A significance threshold of p < 0.1 was employed to determine the retention of variables in the model during each backward step. All analyses were performed with the statistical software SAS (9.4, SAS Institute, Cary, NC).

Ethical Approval

The Monroe County MBHS was designated as surveillance and quality improvement by the University of Rochester Institutional Research Subject Review Board (RSRB). Therefore, IRB oversight was not required for data collection. However, approval was obtained from the University RSRB for these secondary data analyses.

Sociodemographic Characteristics

Among the 1,879 mothers in the cohort, 1,556 qualified for the main analysis due to complete data on maternal and paternal variables. Among these, 279 (17.9%) had a CM-PA, while 1,277 (82%) had vPA for their infants. Mothers with CM-PA were more likely to be Black, Hispanic, have lower income, education, and younger age at pregnancy. They had higher rates of smoking, marijuana use, urban residence, unplanned pregnancies, elevated stress levels, and lower social support [Table 1].

Maternal History of ACEs and Infants’ Paternity Status at Birth

Variables that were consistently retained by backward selection included maternal race, income, birthplace, parity, pregnancy intention, stress, tobacco and marijuana use, social support, and paternal education. After adjustment, a one-point increase in maternal ACE total score was associated with 14% higher odds of CM-PA (OR = 1.14; 95% CI, 1.03–1.27). Maternal experience of household dysfunction was linked to 83% higher odds of CM-PA (OR = 1.83; 95% CI, 1.23–2.71) while living with a problem drinker or drug user during childhood was associated with 70% higher odds of CM-PA (OR = 1.70; 95% CI, 1.09–2.65) [Table 3].

In the sensitivity analysis, 1,665 mothers with complete maternal data were included, representing a larger sample than in the main analysis. This increase, by n = 109, was due to the exclusion of the father’s age and education variables, allowing more subjects to meet the complete data criteria. The decision to combine CM-PA with No-PA rather than with vPA was guided by the observed similarities between the CM-PA and No-PA groups, including maternal age, income, pregnancy stress, social support, tobacco use, ACE experience, and ACE total score [Table 2].

We found that compared to mothers of infants with vPA, mothers of suboptimal PS infants were more likely to be Black, Hispanic, have lower income and education, and be younger at pregnancy. They were also more likely to have a history of smoking and marijuana use, urban residence, unplanned pregnancies, higher stress levels, and lower social support [Table 4]. Variables consistently retained through backward selection encompassed the mother’s race, education, income, birthplace, resilience, parity, pregnancy intention, stress, tobacco and marijuana use, and social support. After adjusting for these factors, a one-point increase in maternal ACE total score was associated with 13% higher odds of suboptimal PS (OR = 1.13; 95% CI, 1.02–1.25). Mothers who experienced household dysfunction had 83% higher odds of suboptimal PS (OR = 1.83; 95% CI, 1.26–2.65), and those with parents separated or divorced had 57% higher odds of suboptimal PS (OR = 1.57; 95% CI, 1.10–2.24) [Table 5].

Limitations and Strengths of the Study

This study is limited by its reliance on existing secondary data, missing broader measures of paternal characteristics and involvement. Future research should include both parents’ perspectives to fully capture paternal engagement. Another limitation is the recall bias from the retrospective survey, despite efforts to minimize this through standardized surveys. Caution is advised when extending findings beyond a sample of majority white, educated individuals. Finally, despite adjusting for known confounders, there could still be unaccounted variables affecting the results.

Nevertheless, this study has several strengths. First, we have a large sample size (N = 1,556) that allowed for the detection of a small effect with decent power. Second, the results of our study were robust across sensitivity analyses, which consistently confirmed that ACEs were associated with a higher risk of no paternity establishment and/or court-mandated paternity establishment. Third, we adjusted for a comprehensive set of covariates that were associated with paternity establishment, which helped tease out the confounding effects as much as possible. Fourth, we utilized a secondary dataset to explore the relationship between maternal ACE and offspring’s paternity acknowledgment status at birth in a cost-efficient and time-saving manner, laying the foundation for subsequent confirmatory research. Fifth, to our knowledge, this is among the few studies that examine the association between maternal ACEs and infants’ paternity establishment, helping to fill a knowledge gap.

Key Messages

• Mothers of infants with CM-PA were more likely to be Black or Hispanic, had lower income and education, had higher substance use and traumatic stress, and had lower social support.

• An elevated maternal ACE score was associated with higher odds of infants experiencing CM-PA.

• Infants of mothers who experienced family dysfunction were more likely to have CM-PA.

• Infants of mothers who lived with a problem drinker or drug user during childhood were more likely to have court-enforced paternity affidavits.