Adverse childhood experiences and prenatal attachment in pregnant women

Abstract

BACKGROUND

Adverse childhood experiences (ACEs) have been demonstrated to exert a detrimental effect on emotional regulation, interpersonal relationships, and mental health, which may influence a woman’s ability to form a healthy prenatal attachment during pregnancy.

AIM

To investigate the relationship between ACEs and prenatal attachment levels among pregnant women.

METHODS

In this observational study, 367 women aged 18-35 years participated. The collection of data was conducted by means of the Personal Information Form, ACEs Scale, and Prenatal Attachment Inventory. The analyses were conducted utilizing IBM SPSS 25 software.

RESULTS

The mean age of the pregnant women who participated in the study was 27.08 ± 5.38 years (range 18-42 years), 4.98 ± 3.37 (1-24) years of marriage, 1.22 ± 1.15 (0-6) number of living children, 2.49 ± 1.37 (1-9) number of pregnancies, 34.99 ± 3.89 (26-40 weeks of gestation. A total ACE score mean of 1.70 ± 1.22 was obtained for pregnant participants of the study. This was accompanied by a mean Prenatal Attachment Inventory score of 63.53 ± 3.23. Multiple linear regression analysis was performed to identify the determinants of prenatal attachment levels in pregnant women. The established model was statistically significant (F = 115.184, P < 0.001), indicating that 56.0% of the variance in prenatal attachment was explained by the determinants.

CONCLUSION

Results from this study clearly demonstrate that ACEs negatively predict prenatal attachment levels. These findings emphasize that the psychological effects of ACEs during pregnancy need to be addressed more comprehensively.

Key Words: Adverse childhood experiences; Nurse; Pregnancy; Pregnant women; Prenatal attachment

Core Tip: It has been found that adverse childhood experiences (ACEs) significantly predict prenatal attachment levels in pregnant women. Pregnant women with higher ACE scores showed significantly lower prenatal attachment scores. Marital status, family type, number of living children, and ACEs explained 56.0% of the variance in prenatal attachment levels. Factors such as marital status, spouse’s employment status, pregnancy planning, and health problems experienced during pregnancy were found to be associated with both ACE and prenatal attachment scores. The findings emphasize the necessity of routine ACE screening and psychological support during pregnancy to strengthen the mother-fetus bond.

INTRODUCTION

During the period of gestation, the mother begins to form mental representations of her unborn child. This process of emotional bond formation between mother and fetus is referred to as prenatal attachment[1]. According to attachment theory, attachment experiences with caregivers during childhood influence the close relationships individuals form later in life and the relationship they develop with their baby when they become parents. This continuity, which suggests that attachment is transmitted across generations, can be scientifically examined through the Adult Attachment Interview, which allows for the assessment of adult attachment patterns[2]. Parents’ mental health also plays a role in the intergenerational transmission of attachment[3]. In particular, negative childhood experiences experienced by the mother affect behavioral problems in the child through insecure attachment and depression[4].

Research has shown that the interaction of pregnancy with a woman’s life history and identity and the quality of attachment processes related to pregnancy[5], play a role in facilitating the transition to motherhood[6] and are strongly associated with the mother-infant attachment relationship after birth[7]. Adverse experiences in childhood can have lasting effects on women’s emotional well-being during pregnancy and may, in turn, impair the development of prenatal attachment. Psychological distress experienced during pregnancy - particularly stress, anxiety, and depressive symptoms - has been consistently linked to lower levels of prenatal attachment[8].

Adverse childhood experiences (ACEs) refer to preventable traumatic events in childhood and include neglect, violence, parental separation, substance abuse, and mental health problems[9]. Although the effects of such traumas on pregnancy and birth outcomes are of limited knowledge, some research suggests that ACEs are associated with prematurity, pregnancy loss, low birth weight, unintended pregnancy, adverse mental health, intimate partner violence during pregnancy, substance abuse, gestational hypertension, and neonatal intensive care unit admissions[10-12].

Research on the impact of ACEs on prenatal attachment suggests that these experiences can have long-term effects on emotional and psychological health in adulthood, complicating mother-infant attachment[13,14]. In light of this information, the present study was conducted with the objective of examine the relationship between pregnant women’s ACEs and their prenatal attachment levels.

Research questions: Is there a significant difference between the demographic profile of pregnant women and their childhood adverse experiences and prenatal attachment levels? What are the mean scores of pregnant participants on the ACEs Scale? What are the mean scores of pregnant participants on the Prenatal Attachment Inventory (PAI)? Do negative childhood experiences of pregnant participants predict their prenatal attachment levels?

MATERIALS AND METHODS

Type of the study

This observational study was conducted in accordance with the Strengthening the Reporting of Observational studies in Epidemiology guidelines.

Setting and participants

This study was conducted with pregnant participants in the city center of Konya between July 22, 2023, and January 15, 2024. Data was collected using the snowball sampling method, in which participants invited people who met the research criteria to participate in the study. While snowball sampling facilitates access to the target population, it may create selection bias and limit the representativeness of the sample, as it may increase the likelihood of individuals with similar social circles and characteristics being included in the sample. The calculation of sample size was performed using the G*Power 3.1.9.2 program. In the reference study, the PAI score environment was taken into consideration[15]. The total number of pregnant women was calculated to be 367, with a margin of error of 5% (α = 0.05), a 1-β confidence interval of 0.95, and an effect size of d = 0.17. The study was terminated when the sample size was reached.

Inclusion criteria

The inclusion criteria including: (1) Having completed at least primary school; (2) Being between 18 yearsand 35 years of age; (3) At 28 weeks of gestation or later; (4) Carrying a single, healthy fetus; and (5) Willingness to communicate and cooperate with the researcher.

Exclusion criteria

The exclusion criteria including: (1) Pregnancy with a history of high-risk pregnancy; (2) Women who conceive with assisted reproductive techniques; (3) Presence of a chronic medical condition; (4) Diagnosis of any psychiatric disorder; and (5) Being a non-citizen.

Dependent and independent variables of the study

Independent variables: The independent variables of the study consisted of the sociodemographic characteristics of the participants and their childhood adverse experiences.

Dependent variables: PAI means of pregnant participations constitute the dependent variable.

Data collection tools

The collection of data was facilitated by the Personal Information Form, ACEs Scale, and PAI.

Personal Information Form: The researchers developed this form as a result of a literature review[16] which incorporated a total of 19 questions concerning the sociodemographic and obstetric characteristics of pregnant participations.

ACEs Scale: The Childhood Trauma Scale was developed by Centers for Disease Control and Prevention and Permanente in 1997, and reliability and validity studies were conducted by Gündüz et al[17]. The scale in question comprises 10 items designed to elicit information regarding childhood traumas. The total score range is from 0 to 10. The scale is not subject to a cut-off point. On the scale, the respondent is asked to check “yes” if he/she has experienced emotional abuse, physical abuse, emotional neglect, physical neglect, and violence against the mother frequently or very frequently until the age of 18, and if he/she has experienced the aforementioned situation even once in the sub-dimensions of sexual abuse or domestic dysfunction. The questions contain only “yes” options; otherwise, they are left blank. Conversely, the responses that are categorized as ‘yes’ are aggregated to derive the test’s score. The score thus serves as an indicator of the number of negative experiences the subject has accumulated. In the Turkish adaptation study, Cronbach’s alpha value was determined to be 0.74[18].

PAI: As stated by Yılmaz and Beji[19], the inventory originally developed by Muller[18] was adapted for use in Türkiye[20]. This scale, which is used to understand the feelings and thoughts of pregnant women and to evaluate their level of attachment to the fetus during the reproductive period, consists of 21 items. These items are designed to reliably measure the level of attachment that pregnant women feel toward their infants. Each item in the PAI is a four-point Likert-type scale with a score between 1 and 4. PAI, items are scored as “never” = 1, “sometimes” = 2, “often” = 3, “always” = 4. The maximum attainable score is 84, derived by multiplying 21 by itself four times, while the minimum score is 21, which is equivalent to multiplying 21 by 1. It is demonstrated that an elevated score on the scale is indicative of a higher level of prenatal attachment. In the Turkish adaptation study, the Cronbach alpha reliability coefficient of the PAI was found to be 0.84.

Data collection method

Research data was collected between July 22, 2023, and January 15, 2024, using online platforms (WhatsApp) by sharing the survey link. The data collection process took approximately 10-15 minutes on average. The data collection form was designed so that participants could proceed to the questions after giving their consent on the first page. Women who agreed to participate completed the online survey form after completing the consent step. Data was collected using the snowball sampling method, in which participants invited people who met the research criteria to participate in the study. To prevent bias by preventing the same individuals from participating, only one survey submission per device was accepted.

Statistical analyses

Statistical analyses of the data were conducted using SPSS version 25.0. In descriptive analyses, frequency, percentage distribution, and mean ± SD were calculated to provide information about the general characteristics of the participants. The normal distribution of the scale scores, which are quantitative variables, was evaluated using skewness and kurtosis tests. In the context of descriptive tests, the t-test was employed for pair-wise comparisons, while one-way ANOVA was utilized for more than two comparisons, as per the suitability of the data for normal distribution. Multiple linear regression analysis was performed to examine the effect of negative childhood experiences on prenatal attachment in pregnant women. Significance was evaluated at the P < 0.05 level.

Ethics

This study was approved by the Ethics Committee of KTO Karatay University (Approval No. 2023/0005). After the purpose of the study was explained to the pregnant women, their consent (informed consent principle) was obtained in writing. Pregnant women participating in the study will be informed that information about them will not be shared with others. Permission was obtained from the authors of the ACEs Scales and PAI via e-mail. The study was conducted in accordance with the principles of the Declaration of Helsinki.

RESULTS

The mean age of the pregnant women was 27.08 ± 5.38 years, and the mean gestational week was 34.99 ± 3.89. Most participants were married, lived in a nuclear family structure, and were not employed, while the majority of their spouses were working. Most pregnancies were reported as planned, and regular antenatal follow-up was common. Detailed sociodemographic and obstetric characteristics are presented in Tables 1 and 2.

Table 1 Descriptive characteristics of pregnant women (n = 367), mean ± SD (minimum-maximum)/n (%).

Variables
Age	27.08 ± 5.38 (18-42)
Duration of marriage (year)	4.98 ± 3.37 (1-24)
Number of living children	1.22 ± 1.15 (0-6)
Number of pregnancies	2.49 ± 1.37 (1-9)
Gestation week	34.99 ± 3.89 (26-40)
Marital status
Single	12 (3.3)
Married	355 (96.7)
Family structure
Nuclear family	304 (82.8)
Extended family	63 (17.2)
Education status
High school and before	271 (73.8)
University and beyond	96 (26.2)
Employment status
Employee	81 (22.1)
Unemployed	286 (77.9)
Spouse’s employment status
Spouse employed	345 (94.0)
Spouse unemployed	22 (6.0)
Social security
Yes	292 (79.6)
No	75 (20.4)
Longest lived in
Village/town	81 (22.1)
City center	286 (77.9)
Monthly income status
Income exceeds expenses	44 (12.0)
Income equals expenses	233 (63.5)
Income less expenses	90 (24.5)

Table 2 Obstetric characteristics of pregnant women (n = 367).

Variables	n	%
Experiencing health problems during pregnancy
Yes	72	19.6
No	295	80.4
Regular check-ups during pregnancy
Yes	332	90.5
No	35	9.5
Pregnancy planning status
Planner	234	63.8
Happy without a plan	117	31.8
Unplanned and unprepared	16	4.4
Planned mode of delivery
Vaginal	268	73.0
Cesarean section	73	19.9
Other	26	7.1
Curettage
Yes	43	11.7
No	324	88.3
Abortus
Yes	59	16.1
No	308	83.9

The mean ACE total score of the participants was 1.70 ± 1.22, and the mean prenatal attachment score was 63.53 ± 3.23. ACE scores differed significantly according to several sociodemographic and pregnancy-related characteristics (Table 3). Higher ACE scores were observed among women with greater social and economic vulnerability and among those reporting unplanned or unprepared pregnancies. No significant differences were found for obstetric history variables such as curettage, miscarriage, or planned mode of delivery.

Table 3 Comparison of descriptive characteristics of pregnant women and mean Adverse Childhood Experiences scores (n = 367).

Variables	mean ± SD	Test value	P value
Marital status		t = 4.829	0.000
Single	3.33 ± 1.64
Married	1.65 ± 1.17
Family structure		t = -2.026	0.043
Nuclear family	1.64 ± 1.18
Extended family	1.98 ± 1.39
Education status		t = 1.184	0.237
High school and before	1.75 ± 1.24
University and beyond	1.57 ± 1.16
Employment status		t = -0.176	0.860
Employee	1.68 ± 1.26
Unemployed	1.71 ± 1.21
Spouse’s employment status		t = -3.567	0.000
Spouse employed	1.64 ± 1.17
Spouse unemployed	2.59 ± 1.65
Social security		t = -2.172	0.031
Yes	1.63 ± 1.16
No	1.97 ± 1.41
Longest lived in		t = 0.952	0.342
Village/town	1.81 ± 1.31
City center	1.67 ± 1.20
Planned mode of delivery		F = 1.254	0.287
Vaginal	1.66 ± 1.18
Caesarean section	1.64 ± 1.19
Other	1.88 ± 1.31
Experiencing health problems during pregnancy		t = 2.827	0.005
Yes	2.06 ± 1.45
No	1.61 ± 1.13
Regular check-ups during pregnancy		t = -2.392	0.017
Yes	1.65 ± 1.19
No	2.17 ± 1.44
Curettage		t = 1.176	0.240
Yes	1.91 ± 1.30
No	1.67 ± 1.21
Abortus		t = 1.588	0.113
Yes	1.93 ± 1.33
No	1.66 ± 1.20
Pregnancy planning status		F = 6.817	0.001
Planner	1.61 ± 1.18
Happy without a plan	1.74 ± 1.19
Unplanned and unprepared	2.75 ± 1.61

t: t-test; F: One-way ANOVA.

The associations between participants’ demographic and pregnancy-related characteristics and their PAI total scores are presented in Table 4. PAI scores showed significant differences across selected socioeconomic and pregnancy-related variables. Lower prenatal attachment scores were observed among women with greater vulnerability indicators, including lack of partner employment, presence of pregnancy-related health problems, and unplanned or unprepared pregnancies. No significant differences were found across other sociodemographic and obstetric variables.

Table 4 Comparison of socio-demographic characteristics and mean prenatal attachment inventory scores of pregnant women (n = 367).

Variables		n	mean ± SD	Test value
Marital status	Single	12	59.17 ± 2.69	t = -4.911; P = 0.000
	Married	355	63.68 ± 3.14
Family structure	Nuclear family	304	63.55 ± 3.20	t = 0.156; P = 0.876
	Extended family	63	63.48 ± 3.37
Education status	High school and before	271	63.42 ± 3.32	t = -1.094; P = 0.275
	University and beyond	96	63.84 ± 2.93
Employment status	Employee	81	63.48 ± 3.29	t = -0.166; P = 0.868
	Unemployed	286	63.55 ± 3.21
Spouse’s employment status	Spouse employed	345	63.65 ± 3.15	t = 2.730; P = 0.007
	Spouse unemployed	22	61.73 ± 3.84
Social security	Yes	292	63.65 ± 3.21	t = 1.367; P = 0.173
	No	75	63.08 ± 3.27
Longest lived in	Village/town	81	63.20 ± 3.19	t = -1.063; P = 0.289
	City center	286	63.63 ± 3.23
Monthly income status	Income exceeds expenses	44	63.61 ± 3.08	F = 1.730; P = 0.179
	Income equals expenses	233	63.73 ± 3.25
	Income less expenses	90	62.99 ± 3.21
Experiencing health problems during pregnancy	Yes	72	62.85 ± 3.30	t = -2.021; P = 0.044
	No	295	63.70 ± 3.19
Regular check-ups during pregnancy	Yes	332	63.64 ± 3.08	t = 1.916; P = 0.056
	No	35	62.54 ± 4.33
Pregnancy planning status	Planner	234	63.81 ± 3.16	F = 7.175; P = 0.001
	Happy without a plan	117	63.37 ± 2.96
	Unplanned and unprepared	16	60.75 ± 4.62
Curettage	Yes	43	63.47 ± 3.94	t = -0.149; P = 0.882
	No	324	63.54 ± 3.13
Abortus	Yes	59	63.17 ± 3.02	t = -0.946; P = 0.345
	No	308	63.60 ± 3.26
Planned mode of delivery	Vaginal	268	63.56 ± 3.21
	Caesarean section	73	63.21 ± 3.35	F = 0.989; P = 0.373
	Other	26	64.23 ± 3.03

t: t-test; F: One-way ANOVA.

Multiple linear regression analysis was performed to identify the determinants of prenatal attachment levels in pregnant women. The established model was statistically significant (F = 115.184, P < 0.001), indicating that 56.0% of the variance in prenatal attachment was explained by the determinants. The results showed that marital status (B = 0.065, β = 0.076, P = 0.036), family type (B = -0.030, β = -0.073, P = 0.039), number of children living in the household (B = -0.035, β = -0.078, P = 0.026), and ACEs (B = -0.092, β = -0.736, P < 0.001) were significant predictors of prenatal attachment levels. Accordingly, it was observed that those who were married, those living in extended families, and those with three or more children had high prenatal attachment. Furthermore, it was determined that a one-unit increase in the negative childhood experiences of pregnant women resulted in a 0.092-unit decrease in their prenatal attachment (Table 5).

Table 5 Multiple linear regression analysis on the effect of pregnant women’s adverse childhood experiences on prenatal attachment (n = 367).

Independent variables	B	SE	β	t	P value	95%CI
Constant	3.174	0.038		83.534	< 0.001	3.100-3.249
Marital status (reference: Single)	0.065	0.031	0.076	2.101	0.036	0.004-0.126
Family type (reference: Extended family)	-0.030	0.014	-0.073	-2.073	0.039	-0.058 to -0.002
Number of living children (reference: 3 and above)	-0.035	0.016	-0.078	-2.228	0.026	-0.067 to -0.004
ACE	-0.092	0.005	-0.736	-20.237	< 0.001	-0.101 to -0.083

r = 0.748; R² = 56.0%; adjusted R² =0.55 (F = 115.184, P < 0.001, Durbin-Watson: 2.046). ACE: Adverse childhood experience; CI: Confidence interval.

DISCUSSION

This research was conducted to examine the relationship between pregnant women’s ACEs and their prenatal attachment levels. The study findings reveal that women’s negative experiences during childhood are associated with various socio-demographic factors such as marital status, family structure, spouse’s employment status, social security status, pregnancy planning, and frequency of antenatal care; while women’s pre-natal attachment levels are associated with factors such as marital status, spouse’s employment status, pregnancy planning, and experiencing health problems during pregnancy. The study’s findings align with the existing literature[21-23].

The mean ACE score identified in our sample indicates that ACEs are observed at a significant level among pregnant women. Similarly, a study found that 58.3 per cent of pregnant women had experienced at least one ACEs[24]. This parallelism between the findings suggests that early life experiences form an important background for understanding the psychosocial structure of expectant mothers. Therefore, it may be useful to consider childhood experiences in psychosocial risk assessments conducted during pregnancy. From a clinical perspective, the findings suggest that ACE scores may help identify pregnant women who are more likely to need additional psychosocial support. Recognizing a history of childhood adversity during routine prenatal care could enable health professionals to offer early and preventive support, with the aim of protecting maternal mental well-being and strengthening prenatal attachment.

Our findings indicate that 56.0% of prenatal attachment variance is explained by predictors (marital status, family type, number of living children, ACE). According to our findings, pregnant women living in extended families with three or more children have higher levels of prenatal attachment. These findings differ from some studies indicating that family structure and number of children may negatively affect prenatal bonding by increasing the burden on the mother. However, these results only become meaningful when evaluated in the context of the sociocultural characteristics of the sample and the perceived level of social support. It can be said that living in an extended family structure may increase perceived social support by enabling the sharing of care and household responsibilities, and that this situation may strengthen prenatal bonding.

According to our findings, the prenatal attachment levels of pregnant women who are married were found to be high, consistent with the literature. Being married can support the mother’s psychological well-being during pregnancy through protective factors such as spousal support, emotional sharing and financial security. For women with more than three children, it is thought that previous experiences of motherhood facilitate adaptation to pregnancy, reduce uncertainty and anxiety, and contribute to the earlier and stronger adoption of maternal identity. According to this information, it is understood that the effect of marital status, family structure and number of children on prenatal bonding should be considered not only in terms of quantitative load but also within the framework of contextual factors such as social support, cultural norms and maternal experience.

The transition to motherhood is a complex process that requires women to adapt physically, emotionally, and psychologically. The main tasks include adjusting to body changes, establishing a maternal identity, and developing a strong bond with the baby. However, for women who have experienced emotional neglect during childhood, these adaptations can be more challenging[25]. Research indicates that women with a history of abuse experience greater difficulty bonding with their babies compared to those without such a history[26]. A study has reported that childhood trauma complicates the transition to parenthood and prevents expectant mothers from establishing a secure bond with their fetuses[27]. Stark Stigger et al[20] emphasized that childhood trauma negatively affects maternal-fetal attachment and highlighted the importance of identifying pregnant women with such histories during the prenatal period and providing appropriate support mechanisms. Childhood trauma can weaken attachment quality by undermining parental trust and maternal mental health. ACEs are associated with depression and anxiety during pregnancy and are linked to increased risk of both prenatal and postnatal attachment difficulties[27-31]. A review of the literature reveals that some studies indicate that ACEs do not have a direct effect on prenatal attachment, but rather that this relationship is shaped through indirect pathways[27,32] In a multinational cohort study (n = 1185), it was reported that prenatal depression completely mediated the relationship between the mother’s ACEs and prenatal attachment[33]. Similarly, in community samples, experiencing maltreatment in childhood predicted greater mental health problems, but it was noted that the variable associated with prenatal attachment and parental trust was the mother’s mental health rather than the maltreatment itself[27]. Furthermore, a comprehensive mediation meta-analysis revealed that maternal depression and insecure attachment are consistent mediating variables in the relationship between the mother’s ACEs and the child’s emotional development[34]. The findings of our study are consistent with the existing literature and also suggest that the mother’s ability to manage and process past negative experiences during pregnancy plays an important role in establishing a healthy bond with her baby. These findings contribute to a better understanding of the impact of emotional neglect on the mothering process and highlight the need for supportive interventions in this field.

In conclusion, the study findings strongly support the claim that ACEs negatively affect prenatal attachment levels. The interplay of maternal mental health, attachment security, and the potential for intergenerational trauma emphasizes the importance of early identification and intervention for expectant mothers with a history of trauma. Addressing these issues can promote healthier maternal-fetal relationships and improve outcomes for both mothers and their children.

This study has several limitations. Firstly, due to its cross-sectional design, it is not possible to draw causal inferences. The observed percentage is clinically meaningful and indicates a strong relationship between the variables, but the temporal direction and cause-and-effect relationships cannot be determined. Longitudinal studies are recommended in the future to clarify the nature and direction of this relationship. The use of the snowball sampling technique in the study imposes limitations. Although the snowball sampling method facilitates access to participants, it may have increased the likelihood of selecting individuals from similar social networks, thereby creating selection bias and limiting the representativeness of the sample.

CONCLUSION

The findings of this study suggest a meaningful relationship between ACEs and prenatal attachment levels. This highlights the importance of considering women’s early life experiences within prenatal care and supports the use of trauma-informed approaches, such as ACE screening and targeted psychosocial support for those at higher risk. Greater awareness among health professionals may help enable earlier recognition and supportive intervention during pregnancy. Since the study used a cross-sectional design, the results should be interpreted as associations rather than causal effects. Future longitudinal studies would help to better understand how this relationship develops over time and how it may influence longer-term maternal and infant outcomes.

ACKNOWLEDGEMENTS

The authors would like to thank the pregnant women who participated in the study.