The Relationship Between Childhood Abuse and Unintended Pregnancy Among Married Women of Reproductive Age in Bangladesh

ABSTRACT

Childhood abuse has been associated with adverse medical, psychological, behavioral, and socioeconomic outcomes in adulthood. Despite this, limited research explores the connection between childhood abuse and unintended pregnancy during adulthood. Notably, existing studies have predominantly focused on high-income countries, leaving a significant gap regarding low- and middle-income nations. This study aims to investigate the impact of childhood physical, sexual, and psychological abuse on the prevalence of unintended pregnancies and explore the interaction effects of childhood abuse on unintended pregnancy occurrences. The cross-sectional survey study was conducted between October 2015 and January 2016 in the Chandpur District of Bangladesh. Data were collected from 426 married women aged 15–49 years who had at least one child of six months or younger. The assessment of child abuse pertains to the mother’s own experiences of childhood abuse and not abuse inflicted on her child. The prevalence of childhood physical, psychological, and sexual abuse was 37%, 26%, and 15%, respectively. About 25.1% of their most recent pregnancies were unintended. Notably, women with a history of childhood sexual abuse were twice as likely to experience unintended pregnancy compared to those without such a history. Furthermore, a dose-response connection was observed between the combined exposure of all categories of childhood abuse and a higher risk of unintended pregnancy. These findings highlight the need for targeted interventions, such as comprehensive sex education, accessible mental health support, and improved child protection frameworks, to address the potential repercussions of maternal childhood abuse and reduce the incidence of unintended pregnancies.

KEYWORDS:

• Childhood abuse
• sexual abuse
• unintended pregnancy
• adulthood outcomes
• maternal health
• low- and middle-income countries
• Bangladesh

Pregnancy is universally regarded as a moment of enormous joy and excitement for most women and their families because of the growing expectation of the pleasure a new offspring will bring to the family. Pregnancy, however, can also be a life event of trauma and anxiety for some women and their families because it may result in many crises for which they are unprepared (Islam, Broidy, Baird, & Mazerolle, 2017; Yildiz et al., 2017). Some women face unintended pregnancies, which encompass pregnancies that are either unwanted (i.e., not at all wanted) or mistimed (i.e., wanted but later), leading to complex emotional and societal dynamics. This divergence in experiences underscores the need for a comprehensive understanding of the factors influencing unintended pregnancies and their potential consequences.

In recent years, the global health community has increasingly recognized unintended pregnancy as a significant concern with far-reaching implications. A staggering 44% of all pregnancies worldwide are unintended, a phenomenon with varying outcomes – approximately 56% result in induced abortion, 32% in birth, and 12% in miscarriage (Bearak et al., 2020; Singh et al., 2018). Recent evidence suggests that unsafe abortions resulting from unintended pregnancies contribute significantly to maternal mortality, particularly in low- and middle-income countries (LMICs), where 97% of the abortions are performed in unsafe conditions due to legal restrictions (Ganatra et al., 2017; Say et al., 2014; Sedgh et al., 2012; Singh et al., 2018). Beyond immediate physical risks, unintended pregnancies have been associated with a range of adverse maternal and child health outcomes, including abortion-related morbidity and mortality, lower likelihood of breastfeeding, family dysfunction, compromised parent-child bonding, and infant mortality (Atzl et al., 2021; Boden et al., 2015; J. A. Hall et al., 2017; Sarder et al., 2021).

Previous studies have revealed a multitude of factors associated with unintended pregnancies, including socio-demographic characteristics such as younger age, limited education, poverty, minority status, urban residence, larger family size, alcohol abuse, inadequate access to contraception, and women’s restricted autonomy (Sarder et al., 2021; Zeleke et al., 2021). A growing body of research has illuminated the link between intimate partner violence (IPV) and unintended pregnancies, highlighting the intricate interplay between relational dynamics and reproductive health (Pallitto et al., 2013; Salazar & San Sebastian, 2014; Zakar et al., 2016).

Recent attention has turned toward understanding the lasting impact of childhood adversities on health and well-being across the lifespan (Olsen, 2018). Extensive theoretical and empirical work demonstrates how adverse childhood experiences can shape outcomes in adolescence and adulthood, underscoring the need for nuanced investigation (Boynton-Jarrett et al., 2021; Greenfield, 2010). Evidence shows that childhood maltreatment – spanning sexual, psychological, and physical dimensions – contributes to perinatal mental health and adverse pregnancy outcomes (Islam et al., 2018; Leeners et al., 2014; McDonnell & Valentino, 2016; Olsen, 2018). It may be that childhood abuse can result in lasting psychological impacts, influencing relationship skills, decision-making autonomy, and intergenerational transmission of trauma. These factors may contribute to unintended pregnancies through disrupted contraceptive choices, limited negotiation power, and compromised mental health (Norman et al., 2012; Pereznieto et al., 2014; Sahle et al., 2021).

Violence against women and girls is deeply rooted in social and cultural norms and attitudes that legitimate male dominance and female subjugation and subordination (Ackerson & Subramanian, 2008; Schuler et al., 2008; Yllo, 2005). This patriarchal social system, supported by traditional beliefs, perpetuates the social acceptability of male control within families. Traditional attitudes in Bangladeshi culture justify a man’s violence on his partner in certain circumstances, such as a woman not obeying elders, going out without informing her partner, neglecting their children, arguing, refusing sex with him, or burning out food (Garcia-Moreno et al., 2006; National Institute of Population Research and Training National Institute of Population Research and Training (NIPORT), 2009). Despite Bangladesh’s ratification of international conventions and the implementation of laws such as the Domestic Violence Act (2010), the Prevention of Women and Child Repression Act (Ellsberg et al., 2000), the Cruelty to Women Ordinance (1983), and the Dowry Prohibition Act (1980; 1986), the enforcement of these measures has proven ineffective in eliminating violence against women and girls (Johnston & Naved, 2008). In a patriarchal social system, women facing abuse often experience reduced autonomy, restricted movement, and increased financial dependency on their male partners (Ellsberg et al., 2000; Smith & Martin, 1995), hindering their ability to make decisions for themselves and access maternal health care services (Koski et al., 2011).

Remarkably, the prevalence of childhood abuse and unintended pregnancies is disproportionately pronounced in LMICs like Bangladesh (Hillis et al., 2016; Sedgh et al., 2012). While research in high-income countries has delved into the association between childhood abuse and unintended pregnancy (Bellis et al., 2014; Drevin et al., 2020; K. S. Hall et al., 2019; Lukasse et al., 2015; Miller et al., 2012; Nelson & Lepore, 2013; Young-Wolff et al., 2021), a significant gap exists in the realm of LMICs. Moreover, the majority of existing studies have focused on specific categories of abuse, particularly sexual abuse, and specific pregnancy outcomes, such as teenage (Bellis et al., 2014) or adolescent pregnancies (Madigan et al., 2014; Noll et al., 2009). Notably, there persists a gap in understanding how various categories of childhood abuse collectively or independently influence unintended pregnancies, particularly within the context of married women of reproductive age (15–49) in Bangladesh. Bangladesh’s cultural norms, socioeconomic factors, and limited access to healthcare services could contribute to different patterns of unintended pregnancies compared to high-income nations. For instance, the societal emphasis on familial honor and the limited autonomy of women within traditional family structures might influence women’s decisions and control over contraception. Moreover, prevailing gender norms and disparities in education and economic opportunities could interact with childhood abuse experiences to impact women’s reproductive decisions.

To address the above key gaps in knowledge, this study aims to evaluate the impact of childhood physical, sexual, and psychological abuse on unintended pregnancies, and explore the interaction effects of different categories of childhood abuse on unintended pregnancy occurrences. By focusing on Bangladesh, a country with a high prevalence of childhood abuse and unintended pregnancies, this study aims to furnish evidence that could underpin targeted interventions and policies to mitigate the heightened risk of unintended pregnancies among individuals with a history of childhood abuse, thereby promoting reproductive health and overall well-being.

Method

Data collection

A population-based cross-sectional study was carried out between October 2015 and January 2016 in two Sub-Districts (commonly known as Upazila) of the Chandpur District of Bangladesh among 426 married women, aged 15–49 years who had at least one child of six months or younger. The age range of participants (15–49) for this study was selected to capture a comprehensive representation of women in their reproductive years who might be at risk of unintended pregnancies. This age range is consistent with the World Health Organization’s definition of reproductive age and aligns with previous research examining reproductive health issues (Garcia-Moreno et al., 2006; NIPORT et al., 2009). Furthermore, focusing on married women allows us to explore the potential impact of childhood abuse on unintended pregnancies within the context of marital relationships and family dynamics.

A detailed data collection strategy has been described in detail elsewhere (Islam, Broidy, Baird, & Mazerolle, 2017; Islam, Broidy, Baird, & Mazerolle, 2017). Briefly, this study employed a multi-stage random sampling method by randomly selecting two Sub-Districts (Chandpur Sadar and Faridganj upazila) in the first stage, 10 unions (5 unions per sub-district) in the second, and 80 vaccination clinics (8 clinics per union) in the final stage. A list of union-wide vaccination clinics was gathered from the Upazila Health Center to choose the required vaccination clinics. The data were originally collected to examine experiences and consequences of intimate partner violence before pregnancy, during pregnancy, and following birth. Women who attended state-sponsored vaccination clinics to have their children vaccinated constitute the sampling frame. For this face-to-face survey, as many as 453 eligible women were approached for attaining the required sample size. An interviewer-administered structured questionnaire was utilized to learn about women’s experiences from childhood to adulthood. Assurance to the respondents was given that their involvement in this research was voluntary. They were also assured that their receipt of healthcare services from these clinics would not be affected by their participation or nonparticipation in the study.

Measures

Outcome: unintended pregnancy

Pregnancy intendedness in this study was evaluated only for those pregnancies culminating with live births. Pregnancy intendedness was assessed by asking respondents whether they had wanted the pregnancy at the time of conception, had wanted it later or had not wanted it at all. Women who reported their last pregnancy as wanted later or not at all were classified as unintended pregnancy (D’Angelo et al., 2004; Finer & Zolna, 2016; Oulman et al., 2015). Intended and unintended pregnancy were coded “0” and “1,” respectively.

Exposure: childhood abuse

As quite commonly used in the field (Charak et al., 2017; Falgares et al., 2018; Godbout et al., 2009; McKinney et al., 2009), three distinct dichotomous questions were asked to evaluate exposure to childhood physical, sexual, and psychological abuse. To ascertain childhood physical abuse, women were asked if anyone had ever beaten or physically mistreated them before 15 years of age (Godbout et al., 2009). To measure childhood sexual abuse, respondents were asked if anyone had ever been forced to have sex, be kissed, or forced to undress or perform sexual acts against their will before 15 years of age (Godbout et al., 2009). It should be noted here that this study has focused on contact child sexual abuse only. To evaluate childhood psychological abuse, respondents were asked if anyone had ever put them down or shouted hurtful words at them before 15 years of age (Godbout et al., 2009). Response categories were either no (0) or yes (1).

A composite variable with all categories of abuse was designed to examine the interaction effects. Eight resulting groups were coded as none (no abuse on all items) = 0, physical abuse only = 1, sexual abuse only = 2, psychological abuse only = 3, both physical and psychological = 4, both physical and sexual = 5, both sexual and psychological = 6, and all items = 7.

Control variables

We included a range of empirically associated control variables (Sarder et al., 2021; Zeleke et al., 2021) in this study. Age at last pregnancy was classified into three categories, adolescence (≤18 years = 0), young adulthood (19–24 years = 1), and adulthood (≥25 years = 2). Educational attainments were categorized into three groups: no formal education ( = 0), primary education ( = 1), and secondary and higher education ( = 2). Household monthly income was classified according to the national average of Bangladesh (8500 BDT ~ 109 USD) as ≤ 8500 BDT ( = 0) versus > 8500 BDT ( = 1). The area of residence was categorized as rural ( = 0) and urban ( = 1). The respondent’s employment status was categorized as unemployed ( = 0) versus employed ( = 1). Women’s marital duration was grouped into tertiles as: 1–3 years = 0, 4 years = 1, and ≥5 years = 2. Parity was classified as primiparous = 0 and multiparous = 1.

Intimate partner violence (IPV)

The reason for including IPV before pregnancy as a control variable is that a range of research has documented the association between IPV and unintended pregnancy (Pallitto et al., 2013; Salazar & San Sebastian, 2014; Zakar et al., 2016). Women who reported that their intimate partner engaged in any of the following seven behaviors during the 12 months prior to pregnancy constitutes physical IPV before pregnancy: (1) pushing, shaking, or throwing something at her; (2) slapping; (3) twisting her arm or pulling her hair; (4) punching or hitting with a fist or something harmful; (5) kicking or dragging or physically assaulting her; (6) choking or burning; or (7) threatening or attacking with a knife, gun or any other weapon. Sexual IPV before pregnancy was indicated by a woman’s affirmative response to any item asking if she had been physically forced to have sexual intercourse even when she did not want to; if she had been involved in intercourse out of fear; or if she had been forced to perform other sexual acts that she found degrading or humiliating during the 12 months prior to pregnancy.

Analytical strategy

All analyses were conducted using IBM SPSS V24.0, with statistical significance defined as p < .05 (two-tailed). Descriptive statistics for the predictive factors and outcome measures were calculated. The links between the predictive factors and the risk of an unintended pregnancy were tested with the help of cross tabulations and the χ² test. The unadjusted relationships between the predictive factors and outcome variable were assessed through odds ratios (ORs) and 95% confidence intervals. The multicollinearity diagnostic check was performed by investigating variance inflation factors (VIFs <2.5), but no evidence of multicollinearity was noted. In this study, the confidence intervals of pregnancy intendedness were determined using the bootstrapping technique, a resampling method involving multiple pseudo-samples drawn from observed data. This approach allowed us to robustly estimate the variability and uncertainty surrounding the prevalence point estimates and derive confidence intervals reflecting the likely range within which the true prevalence exists with a specified level of confidence.

To assess the strength of any relationship between childhood abuse and unintended pregnancy, we estimated four adjusted multivariate logistic regression models – one for each category of childhood abuse to evaluate the independent influences of different categories of childhood abuse on unintended pregnancy, and one for a combination of all categories of childhood abuse with the control variables to investigate the influences of each type of childhood abuse, adjusting for the other. Finally, the interaction effects of multiple forms of childhood abuse on unintended pregnancy were evaluated by another adjusted multivariate logistic model with the composite child abuse variable.

Ethical considerations

Ethical approval was obtained for scientific and ethical integrity from the National research ethics committee of (BMRC/NREC/2013–2016/305) Bangladesh Medical Research Council and Griffith University Human Research Ethics Committee (CCJ/41/14/HREC) before initiation of the study. Due to the study’s sensitive nature and cultural context, verbal informed consent was obtained from respondents to ensure anonymity and to avoid any legal consequences. Interviews were conducted in a private room without being accompanied by children aged ≥3 years or family members to maintain confidentiality. Respondents with accompanying spouses or other adult family members were excluded from participation. Each respondent received a modest monetary “thank-you gift” (BDT 500 ~ USD 6.50) as appreciation for participation.

Results

Sample

The frequency distribution of pregnancy intendedness among the respondents is presented in Table 1. Among the respondents, 25.1% (n = 107) reported unintended pregnancies, with 12.0% (n = 51) being mistimed and 13.1% (n = 56) unwanted.

Table 1. Frequency distribution of pregnancy intendedness among Currently married women of Bangladesh (N = 426).

Pregnancy Intendedness	n	% (95% confidence intervals)^a
Became pregnant then (Intended)	319	74.9 (70.7–78.9)
Wanted to be pregnant but later date (Mistimed)	51	12.0 (8.9–15.0)
Did not want at all (Unwanted)	56	13.1 (10.1–16.4)
Unintended pregnancy (Unwanted & mistimed)	107	25.1 (21.1–29.3)

^acalculated using the bootstrapping technique.

Table 2 outlines sample descriptive and assesses rates of unintended pregnancy as a function of other variables. At the time of the survey, about one-quarter of the women were below 18 years during their last pregnancy, one-tenth of women had received no formal education, 38.3% of women had a household monthly income of ≤ BDT 8500, and very few women (3.3%) were engaged in a paid job. The majority of them (68.5%) were living in rural areas, and two out of five women had one living child.

Table 2. Maternal characteristics, reported childhood abuse and unintended pregnancy (N = 426).

Independent variables	n (%)	Pregnancy intendedness (%)			Unadjusted OR (95% CI)
		Intended	Unintended	p-value
Age at index pregnancy
≤18	106 (24.9)	81.1	18.9	<.001	1.00
19–24	187 (43.9)	81.3	18.7		0.99 (0.54–1.82)
≥25	133 (31.2)	60.9	39.1		2.76 (1.52–5.02)
Respondent’s education
No formal education	35 (8.2)	65.7	34.3	.008	1.00
Primary	104 (24.4)	65.4	34.6		1.02 (0.45–2.27)
Secondary and higher	287 (67.4)	79.4	20.6		0.50 (0.23–1.06)
Household monthly income
≤8500 BDT	163 (38.3) (33.3)	65.0	35.0	<.001	1.00
>8500 BDT	263 (61.7)	81.0	19.0		0.44 (0.28–0.68)^a
Employment status
Unemployed	412 (96.7)	74.5	25.5	.34	1.00
Employed	14 (3.3)	85.7	14.3		0.49 (0.11–2.21)
Area of residence
Rural	292 (68.5)	74.3	25.7	.69	1.00
Urban	134 (31.5)	76.1	23.9		0.91 (0.56–1.46)
Parity
Primiparous	175 (41.1)	88.6	11.4	<.001	1.00
Multiparous	251 (58.9)	65.3	34.7		4.11 (2.41–7.01)^a
Marriage duration, years
1–3	131 (30.8)	85.5	14.5	.001	1.00
4	37 (8.7)	81.1	18.9		1.34 (.53–3.58)
≥5	258 (60.6)	68.6	31.4		2.70 (1.55–4.69)^a
Childhood physical abuse
No	269 (63.1)	76.2	23.8	.41	1.00
Yes	157 (36.9)	72.6	27.4		1.21 (0.77–1.89)
Childhood sexual abuse
No	364 (85.4)	77.2	22.8	.008	1.00
Yes	62 (14.6)	61.3	38.7		2.22 (1.21–3.77)^b
Childhood psychological abuse
No	317 (74.4)	76.7	23.3	.15	1.00
Yes	109 (25.6)	69.7	30.3		1.43 (0.88–2.32)
Any childhood abuse
No	121 (28.4)	83.5	16.5	.01	1.00
Yes	305 (71.6)	71.5	28.5		2.02 (1.17–3.46)^b
Physical IPV before pregnancy
No	201 (47.2)	84.1	15.9	<.001	1.00
Yes	225 (52.8)	66.7	33.3		2.64 (1.65–4.22)^a
Sexual IPV before pregnancy
No	336 (78.9)	77.4	22.6	.022	1.00
Yes	90 (21.1)	65.6	34.4		1.80 (1.09–2.98)^c
Childhood Abuse Index
None	189 (44.4)	79.9	20.1		1.00
Only physical abuse	88 (20.7)	73.9	26.1		1.41 (.78–2.55)
Only sexual abuse	23 (5.4)	60.9	39.1		2.56 (1.03–6.35)^c
Only psychological abuse	46 (10.8)	73.9	26.1		1.40 (.66–2.96)
Both physical and psychological	41 (9.6)	75.6	24.4		1.28 (.58–2.84)
Both physical and sexual abuse	17 (4.0)	76.5	23.5		1.22 (.38–3.96)
Both sexual and psychological	11 (2.6)	54.5	45.5		3.31 (.96–11.43)^c
All	11 (2.6)	45.5	54.5		4.77 (1.38–16.46)^c

Note: OR odds ratio, CI confidence intervals, IPV intimate partner violence.

^ap < .001; ^bp < .01; ^cp < .05.

Among the abuse categories, the rate of childhood physical, psychological, and sexual abuse was 37%, 26% and 15%, respectively. About 72% of women reported exposure to any childhood abuse. While the lowest prevalence of unintended pregnancy (20.1%) was reported by respondents who had not experienced any childhood abuse, the highest rates (54.5%) were reported by women who experienced all categories of childhood abuse together. Women who were exposed to only childhood sexual abuse reported higher unintended pregnancies (39.1%) compared to women who were not victimized.

Childhood abuse and unintended pregnancy: bivariate findings

Significant associations were found between unintended pregnancies and factors such as maternal age during last pregnancy, educational attainment, household income, marital duration and number of living children (Table 2). Notably, childhood sexual abuse, as well as physical IPV prior to pregnancy, were significantly linked to higher rates of unintended pregnancies (Table 2).

Childhood abuse and unintended pregnancy: multivariate findings

Table 3 presents multivariate analyses focusing on the link between childhood abuse categories and unintended pregnancy. Each form of childhood abuse was investigated independently, and then collectively. Models 1–3 show that only childhood sexual abuse was significantly connected with the risk of unintended pregnancy. With the introduction of all three categories of childhood abuse, controlling for control variables in Model 4, the odds ratio for childhood sexual abuse and physical IPV before pregnancy remained significant. More specifically, women who were exposed to childhood sexual abuse were twice as likely (AOR: 2.07, 95% CI [1.07, 4.03]) to experience unintended pregnancies compared to non-abused women.

Table 3. Adjusted odds ratios for the Association between Childhood Abuse and unintended pregnancy.

Independent variable	Unintended Pregnancy, AOR (95% CI)
	Model 1	Model 2	Model 3	Model 4
Childhood physical abuse
No	1.00			1.00
Yes	1.01 (0.62–1.64)			0.86 (.52–1.44)
Childhood psychological abuse
No		1.00		1.00
Yes		1.30 (0.75–2.25)		1.10 (0.62–1.96)
Childhood sexual abuse
No			1.00	1.00
Yes			2.52 (1.32–4.80)^b	2.07 (1.07–4.03)^c
Physical IPV before pregnancy
No				1.00
Yes				1.88 (1.03–3.44)^c
Sexual IPV before pregnancy
No				1.00
Yes				1.60 (0.90–2.86)
−2 log likelihood	420.45	419.19	412.37	403.92
R^2 (Cox & Snell)	.13^a	.13^a	.15^a	.16^a
R^2 (Nagelkerke)	.20^a	.20^a	.22^a	.24^a
Overall model prediction rate	76.5%	76.5%	77.5%	77.04%

Note: AOR adjusted odds ratio; CI confidence interval; IPV intimate partner violence; ^ap < .001; ^bp < .01; ^cp < .05.

Model 1: Adjusted for age at last pregnancy, education, income, employment, residence, parity, marriage duration, physical and sexual IPV.

Model 2 & 3: Adjusted for the above control variables.

Model 4: Influence of childhood physical, psychological and sexual abuse adjusted for above control variables.

Interaction effects of multiple childhood abuse

Table 4 displays findings of a multivariate logistic regression model exploring the interaction between different categories of childhood abuse and unintended pregnancy. Findings indicated that women who experienced childhood sexual abuse were three times more likely to report unintended pregnancies (AOR: 2.88, 95% CI [1.03, 8.11]) than non-abused women. Furthermore, those exposed to all three forms of childhood abuse together were approximately five times more likely (AOR: 4.83, 95% CI [1.26, 18.56]) to have unintended pregnancies compared to those without childhood abuse experiences.

Table 4. Interaction effects of different forms of childhood abuse on unplanned pregnancy (N = 426).

Childhood Abuse Index	Unintended Pregnancy, AOR (95% CI)^a
None	1.00
Only physical abuse	1.19 (.63–2.27)
Only sexual abuse	2.88 (1.03–8.11)^c
Only psychological abuse	1.33 (.58–3.06)
Both physical and psychological	.99 (.41–2.37)
Both physical and sexual abuse	1.17 (.31–4.35)
Both sexual and psychological	4.04 (1.04–15.74)^c
All	4.83 (1.26–18.56)^c

Note: AOR adjusted odds ratio; CI confidence interval; ^cp < .05.

^aAdjusted for age at last pregnancy, marriage duration, education, income, employment, area of residence, parity, physical and sexual IPV before pregnancy.

Discussion

The present study sought to investigate the relationship between childhood abuse and the occurrence of unintended pregnancy among married women of reproductive age. Findings from this study shed light on the complex interplay of childhood abuse and unintended pregnancy, emphasizing the need for comprehensive policies and interventions to address this critical issue. However, it is important to note that the narrative should align more closely with the results and delve into the cultural, social, and policy factors that may contribute to these findings.

This study revealed that 36.9% of women reported experiencing childhood physical abuse, 14.6% reported childhood sexual abuse, and 25.6% reported experiencing psychological abuse during their early years. A recent systematic review indicates that the global prevalence of childhood physical abuse ranges from 6.9% to 73.8%, sexual abuse from 5.7% to 40.2%, and psychological abuse from 8.8% to 62.9% (Moody et al., 2018). Furthermore, findings from three systematic reviews suggest that the prevalence of childhood sexual abuse among girls is estimated to be between 2.6% to 69% (Barth et al., 2013; Brunton & Dryer, 2021; Stoltenborgh et al., 2015). Our prevalence estimates align with documented rates, yet they are relatively lower when compared to those of other Asian countries. These differences might be attributed to differences in definitions, methodology, sample characteristics, the geographical region, and cultural beliefs and values (Barth et al., 2013; Stoltenborgh et al., 2015). However, the prevalence of childhood abuse in this study demonstrates that many children’s lives are touched by childhood abuse and neglect.

In agreement with previous studies linking women’s experience of childhood abuse and the risk of unintended pregnancy (Drevin et al., 2020; K. S. Hall et al., 2019; Lukasse et al., 2015; Miller et al., 2012; Tang et al., 2018; Young-Wolff et al., 2021), our findings demonstrate a significant dose-response connection between sexual abuse in childhood and unintended pregnancy in adulthood. In addition, our analyses showed significant interactions of childhood abuse categories. Results from this study indicate that women who experienced combinations of different categories of childhood abuse had more than four times higher likelihood of having unintended pregnancies than those with no exposure to childhood abuse, even after controlling for socio-demographic and other psychosocial characteristics. Consistent with other research (Drevin et al., 2020; Young-Wolff et al., 2021), the risk of unintended pregnancy elevated with the interaction of all forms of childhood abuse, indicating the independent influences of abuse categories to be additive.

A rigorous investigation of the causal pathways between childhood abuse and the risk for unintended pregnancies is warranted to develop effective and responsive intervention strategies to prevent unintended pregnancies. Previous studies have documented that women who were exposed to childhood abuse are more likely than non-victims to have sexually risk-taking behaviors, multiple sexual partners, unprotected sex, an adolescent pregnancy, and inconsistent use of contraceptives (Fergusson et al., 2013; Steel & Herlitz, 2005; Tang et al., 2018; Wang et al., 2019). Since, the survey instruments of this study did not cover information regarding women’s risk-taking behaviors or use of contraceptive around the time of the last pregnancy, this research was not able to demonstrate any supporting evidence of these intermediary factors. However, we speculate that childhood abuse may be linked to the development of negative self-models as unworthy, incompetent, powerless, and overly self-reliant in adulthood (Corrales et al., 2016; Wright et al., 2009). Evidence further suggests that women who experienced childhood abuse may downplay their own needs and views, exhibit significantly higher sensitivity to others’ needs and wishes, and less ability to communicate own needs and views in adulthood (Kunst et al., 2019; Maas et al., 2019). This diminishing sense ultimately may contribute to prevent them from developing social competence of autonomous decision-making in initiating or negotiating contraceptive usage during adolescence and adulthood. In addition, due to a direct influence of sexual IPV, some women may unintentionally become pregnant, and husbands’ controlling behavior may limit abused women’s ability to control fertility (Pallitto et al., 2013).

The cultural and social context in Bangladesh undoubtedly influences the complex interplay between childhood abuse and unintended pregnancy. The pervasive influence of patriarchal societal norms and gender inequalities prevailing in the region can exacerbate the vulnerability of women to childhood abuse and limit their agency in family planning decisions (Islam et al., 2018, 2021; Islam, Broidy, Baird, & Mazerolle, 2017). Furthermore, societal stigma may hinder women from seeking help or discussing reproductive health issues openly.

Unintended pregnancies, arising from a history of childhood abuse, may have profound consequences for both child development and maternal well-being. Research suggests that maternal stress and trauma can influence fetal development and contribute to adverse birth outcomes (Coussons-Read, 2013; Lautarescu et al., 2020). Long-term psychological effects on mothers may affect their ability to provide a nurturing environment for their children (Islam et al., 2018). Therefore, addressing the issue of unintended pregnancies in the context of childhood abuse is crucial for promoting positive child development and maternal mental health.

The observed connection between childhood abuse and unintended pregnancy underscores the need for comprehensive policies and programs that address both issues concurrently. Preventive strategies should not only focus on reproductive health and family planning education but also integrate measures to prevent and address childhood abuse. Collaboration between healthcare providers, social services, and policymakers is crucial to create a holistic approach that supports survivors of abuse and empowers women to make informed decisions about their reproductive health. Efforts to strengthen legal frameworks, raise awareness about abuse prevention, and enhance access to mental health services could play a pivotal role in mitigating both childhood abuse and unintended pregnancy. To break the cycle of intergenerational transmission of childhood maltreatment risk, implementing robust parenting education programs and mental health support for survivors are crucial. Furthermore, targeted programs that empower women to assert control over their reproductive choices and negotiate contraceptive use within marital relationships are essential.

This study has several limitations to consider. Firstly, its cross-sectional design prevents establishing definitive causal links between childhood abuse and unintended pregnancies. Longitudinal research is needed for a better grasp of temporal dynamics. Exploring whether abuse during specific developmental stages is associated with higher vulnerability to unintended pregnancies can unveil nuanced connections, contributing to a more refined comprehension of the complex nexus between early-life adversities and reproductive health. Such investigation can offer valuable insights to inform targeted preventive interventions. Secondly, reliance on retrospective self-reports of childhood abuse might introduce recall bias. Future studies could include prospective data collection methods to improve the accuracy and reliability of abuse reporting. In addition, the measurement of abuse relied on single binary questions rather than standardized tools, possibly affecting reporting accuracy. Using validated instruments in future research would strengthen the precision of abuse assessment. Thirdly, the study only assessed pregnancy intendedness in live births, possibly underestimating true rates of unintended pregnancy. Future research should consider expanding the scope to include all pregnancies, regardless of outcome, providing a more comprehensive picture of unintended pregnancy prevalence. Data collection occurred seven years ago, potentially not reflecting current sociopolitical changes in Bangladesh that influence unintended pregnancies. Finally, findings are based on a sample of married women within a patriarchal culture, where marriage might limit contraceptive autonomy. Future investigations should diversify the study population to encompass varied demographic groups, ensuring a more inclusive and representative exploration of the relationship between childhood abuse and unintended pregnancies.

Despite these limitations, this research includes a significant contribution to the empirical literature on the understanding of the relationship between childhood abuse and unintended pregnancy using a large sample (N = 426) of married women that provided an opportunity to study various categories of abuse separately. This is the first known study to investigate how childhood abuse enhances the risk of unintended pregnancy in a developing country context, especially in South Asia. It is one of the very limited studies to incorporate all types of childhood abuse to explore this relationship while adjusting several empirically relevant covariates.

Conclusion

This study contributes valuable insights into the nuanced relationship between childhood abuse and unintended pregnancies among married women in Bangladesh. The interplay of cultural, social, and policy factors underscores the complexity of these issues and emphasizes the need for comprehensive and multidimensional interventions that address cultural norms, promote gender equity, and provide accessible reproductive healthcare services. Such comprehensive efforts hold the potential to alleviate the burden of childhood abuse and unintended pregnancies and contribute to the overall well-being of women and their families in Bangladesh and other LMICs.

Ethical standards and informed consent

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation [institutional and national] and with the Helsinki Declaration of 1975, as revised in 2000. Verbal informed consent was obtained from all participants to be included in the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes on contributors

Md Jahirul Islam

Md Jahirul Islam, Ph.D. is an Adjunct Research Fellow at Griffith Criminology Institute, Griffith University, Australia. His research focus includes intimate partner violence, child abuse and neglect, and criminology.