Abstract

Background: Intimate partner violence (IPV) during pregnancy is a significant public health problem with adverse consequences for maternal and fetal outcomes. Understanding its magnitude and determinants is essential for effective screening and intervention within antenatal care. Objective: To determine the prevalence, forms, correlates, and predictors of IPV among pregnant women attending antenatal care at the Federal University Teaching Hospital, Lafia, Nigeria. Methods: A cross-sectional analytical study was conducted among 381 pregnant women to determine the prevalence and correlates of Intimate Partner Violence (IPV) using a structured questionnaire incorporating the Abuse Assessment Screen (AAS). IPV was defined as the occurrence of any emotional, physical, or sexual violence. Data on sociodemographic, obstetric, and household characteristics were collected alongside experiences of emotional, physical, and sexual IPV during pregnancy. Descriptive statistics summarized the data. Associations between categorical variables and IPV were assessed using chi-square tests, while Pearson correlation examined relationships among IPV types. Binary logistic regression identified independent predictors. Statistical significance was set at p < 0.05. Results: Among 379 respondents included in the final analysis, emotional violence was reported by 22.2%, physical violence by 16.6%, and sexual violence by 8.2%. Strong positive correlations existed among IPV types (r = 0.751 - 0.856, p < 0.001), indicating co-occurrence. At the bivariate level, age, educational level, and monthly income were significantly associated with IPV. In multivariable analysis, educational attainment (AOR = 7.64, 95% CI: 3.29 - 17.74; p < 0.001) and employment status (AOR = 6.34, 95% CI: 2.78 - 14.43; p < 0.001) were identified as independent predictors of IPV, while age, religion, and gestational age were not significant. Conclusion: IPV is common in this setting, with emotional abuse predominating and frequent co-occurrence of abuse types. Routine comprehensive screening and context-specific interventions that address women’s socioeconomic dynamics are recommended.

Keywords

Intimate Partner Violence, Pregnancy, Antenatal Care, Abuse, Predictors, Nigeria

Share and Cite:

1. Introduction

Gender-based violence (GBV), defined as violence perpetrated on the basis of socially ascribed gender differences, is a major global public health, human rights, and development issue [1] [2]. Intimate partner violence (IPV) is a subset of GBV and includes physical, sexual, and psychological abuse perpetrated by a current or former intimate partner [3] [4]. Violence against women has been recognized internationally as a violation of human rights. The United Nations Declaration on the Elimination of Violence Against Women defines it as any act of gender-based violence that results in, or is likely to result in, physical, sexual, or psychological harm or suffering to women [5].

Globally, IPV remains highly prevalent. Recent estimates indicate that physical and/or sexual intimate partner violence affects a substantial proportion of ever-partnered women worldwide [6]. During pregnancy, IPV remains a significant concern. A systematic review and meta-analysis reported that the global prevalence of IPV in pregnancy was approximately 25%, with prevalence estimates of 9.2% for physical violence, 18.7% for psychological violence, and 5.5% for sexual violence [7]. These findings highlight that pregnancy does not confer protection against violence and may, in some contexts, increase vulnerability.

Evidence from high-income countries further underscores the burden of IPV. In the United States, IPV contributes significantly to long-term health and economic burden, reinforcing its importance as both a clinical and public health issue [8]. In Europe, studies have also documented IPV among pregnant women. For example, research conducted in Spain demonstrated the occurrence of IPV during pregnancy using validated screening tools and highlighted associated risk factors [9]. Additionally, the Abuse Assessment Screen and its subsequent refinements have been developed and validated in Western settings to enhance the detection of abuse in healthcare environments [10] [11].

In African settings, the burden appears even higher. In Kenya, Morris et al. reported a lifetime IPV prevalence of 60.3% among women [12]. In Ethiopia, Berhanie et al. documented high levels of IPV both during pregnancy and across the life course [13]. In South Africa, Groves et al. found that more than one-fifth of women experienced IPV during pregnancy or the postpartum period [14]. Yohannes et al. also reported a high prevalence of IPV among pregnant women in Ethiopia and identified spousal occupation and alcohol use as significant predictors [15].

In Nigeria, IPV during pregnancy remains a major concern, with prevalence estimates ranging from 15.4% to 37.2% [16]-[20]. These variations reflect differences in study settings, methodologies, and sociocultural contexts but consistently highlight IPV as a persistent public health challenge.

Several factors have been associated with IPV, including low educational attainment, unemployment, financial dependency, substance abuse, childhood exposure to violence, and sociocultural norms that support male dominance [12] [18] [21]. Broader structural factors such as gender inequality and economic stress also contribute significantly [22] [23].

Despite this, studies on this important issue remain limited in Nigeria. This study, therefore, sought to determine the prevalence, correlates, and predictors of intimate partner violence among pregnant women accessing care in this facility.

1.1. General Objective

To determine the prevalence, correlates, and predictors of intimate partner violence during pregnancy, among women accessing care in the Federal University Teaching Hospital, Lafia.

1.2. Specific Objectives

1) To assess the prevalence and types of IPV among pregnant women in FUTH, Lafia.

2) To determine the correlates of the different forms of IPV among this cohort in FUTH, Lafia.

2. Methodology

2.1. Study Design

This was a cross-sectional analytical study [24].

2.2. Study Area

This study was undertaken in the antenatal clinic and obstetric service points of FUTH Lafia of Nasarawa State, Nigeria. Lafia is the capital city of Nasarawa State, one of the six states in the North-Central region of Nigeria. The state has a land mass of 26875.59 km², an estimated population of 1.9 million according to the 2006 population census, and a population density of about 67 persons per square kilometres [24]. The hospital is a major tertiary referral centre providing comprehensive maternal and child health services. The antenatal clinic runs four days per week with an average attendance of approximately 100 pregnant women per clinic day.

2.3. Study Population

The study population comprised pregnant women attending the antenatal clinic at FUTH Lafia during the study period.

2.3.1. Inclusion Criteria

1) Women who were pregnant and receiving care in the facility during the study period.

2) Women who met the criteria and had consented.

2.3.2. Exclusion Criteria

1) Patients who declined consent.

2) women who were too ill to participate.

3) Those with incomplete or unusable responses on key study variables.

2.4. Sample Size Determination

The sample size for this study was determined using the Cochran formula for cross-sectional studies [25].

$n= \frac{z^2\times p \left(1-p\right)}{d^2}$

where

n = required sample size;

Z = Z-score (1.96 for a 95% confidence level);

p = estimated prevalence from a previous study (34.3% or 0.34) [19];

d = margin of error (5% or 0.05) [25].

After adjustment for a 10% non-response rate, the final sample size was 381 participants.

2.5. Sampling Frame and Sampling Technique

A simple random sampling technique was employed to select study participants. The study was conducted over a defined recruitment period from August 1st to October 31^st, 2025, during routine antenatal clinic days.

The antenatal clinic operates four days per week, with an average daily attendance of approximately 100 pregnant women. On each clinic day, a daily sampling frame was constructed using the clinic attendance register, which served as the list of all eligible women present on that day.

From this list, eligible participants were assigned unique identification numbers, and simple random sampling was performed using a ballot method to select participants for inclusion in the study. This process was repeated on each clinic day until the required sample size was achieved.

Women who met the inclusion criteria but declined participation were excluded and replaced by selecting another participant using the same random sampling procedure from the remaining eligible pool. The number of non-respondents was minimal and did not significantly affect the sample size.

2.6. Data Collection Tools and Procedure

A structured, interviewer-administered questionnaire was utilized to collect data from participants. The instrument captured biodata, medical history, family history, and social history.

1) Sociodemographic and Clinical Data: A structured questionnaire was used to collect data on age, gender, education level, employment status, duration of pregnancy or postpartum duration, number of pregnancies, and number of live births.

2) Abuse Assessment Screen is a validated screening instrument to establish various types of abuse [10] [11] [26] [27].

The AAS contains four questions on emotional, physical, and sexual abuse at any time during the woman’s life, within the previous year, and during pregnancy. The instrument also asks about the relationship with the aggressor, the frequency of the violence, and any fear of the perpetrator. The severity of physical violence is classified from levels 1 to 5: level 1 = slapping and/or pushing, with no injuries or persistent pain; level 2 = punching and/or kicking, with bruises, wounds or persistent pain, level 3 = beating, with severe bruises, burns and/or broken bones; level 4 = permanent head or internal injuries; level 5 = wounding with a weapon.

A positive response to any one of these questions indicates that the responder may be a victim of abuse.

Data were collected through interviewer-administered questionnaires using the Abuse Assessment Screen (AAS) and a structured sociodemographic instrument. Interviews were conducted in a private setting within the antenatal clinic to ensure confidentiality and minimize the risk of disclosure to accompanying partners or relatives.

Questionnaires were administered in English and, where necessary, translated into the Hausa language to ensure comprehension. Back-translation procedures were applied to maintain consistency of meaning.

Interviewers were trained healthcare personnel who received additional instruction on ethical conduct in IPV research, including maintaining neutrality, ensuring confidentiality, responding to distress, and avoiding re-traumatization. Training emphasized adherence to established ethical guidelines for violence research.

Women who disclosed experiences of abuse were handled with sensitivity and provided with immediate emotional support. Where appropriate, they were offered referral to available support services, including social welfare units, counseling services, and relevant non-governmental organizations that provide assistance for victims of domestic violence.

All data were anonymized, and no identifying information was linked to responses to protect participant confidentiality.

2.6.1. Study Protocol

Following the acquisition of written informed consent, eligible participants were enrolled in the study. The study questionnaire was pilot-tested by the researcher on 34 patients attending the ANC at the Federal University Teaching Hospital, Lafia, to assess its usability, administration duration, and feasibility. These were excluded from the final analysis. Findings from this pre-testing indicated that self-administration was unsuitable owing to participants’ heterogeneous educational levels, and administration of questionnaires lasted roughly 30 minutes. This adjustment ensured accurate data collection and accommodated the diverse literacy levels among the study population.

Individuals who consented to the study were assured of being attended to by the attending physicians or midwives as soon as the interview was over. Individuals who did not consent to the study were assured of normal care without any consequence. To ensure participant safety, only one eligible woman was interviewed at a time, and interviews were discontinued if privacy could not be guaranteed. Participants were informed that they could decline to answer any question or withdraw at any time without affecting their care.

The researchers and research assistants were versed in English and Hausa, and hence, the questionnaires were administered according to the respondent’s preference.

After the interview, those who agreed to be referred were handed over to the counselors attached to the antenatal clinic.

2.6.2. Study Duration

The study was carried out between August 1^st and October 30^th, 2025. Data analysis was conducted between November and December 2025.

2.6.3. Study Outcomes

1) Primary Outcome

IPV was defined as a composite binary outcome representing the occurrence of any emotional, physical, or sexual violence.

Intimate partner violence (IPV) was assessed using the Abuse Assessment Screen (AAS), a validated tool for detecting violence during pregnancy. The instrument captures emotional, physical, and sexual abuse occurring within a defined recall period.

In this study, IPV was operationalized as a binary outcome (Yes/No). A respondent was classified as having experienced IPV (IPV = 1) if she reported any form of abuse (emotional, physical, or sexual) during the index pregnancy. Respondents who reported no experience of any of these forms of abuse were classified as IPV = 0.

Each AAS item was scored dichotomously (Yes = 1, No = 0), and the composite IPV variable used for logistic regression analysis was derived by combining all three domains such that the presence of at least one positive response constituted IPV exposure.

2) Secondary Outcomes

i) The prevalence and distribution of specific forms of IPV, namely emotional, physical, and sexual violence.

ii) The bivariate associations between IPV and selected sociodemographic, obstetric, and household characteristics (including age, religion, educational level, employment status, parity, and income).

iii) The interrelationships among different forms of IPV.

iv) The independent predictors of IPV.

The recall period for IPV assessment was during the current pregnancy.

2.7. Statistical Analysis

Data were analysed using IBM SPSS Statistics version 25. Descriptive statistics were used to summarize participants’ sociodemographic and obstetric characteristics. Categorical variables were presented as frequencies and percentages.

The primary outcome variable, intimate partner violence (IPV), was defined as a binary variable (Yes/No), based on a positive response to any of the emotional, physical, or sexual violence items from the Abuse Assessment Screen (AAS).

Bivariate associations between independent variables and IPV were assessed using the Chi-square test. Variables with p-values < 0.20 at the bivariate level, as well as those considered clinically relevant, were included in the multivariable analysis.

A binary logistic regression model was fitted to identify independent predictors of IPV. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported. Model fitness was assessed using the Hosmer-Lemeshow goodness-of-fit test.

Due to sparse data and instability in some categorical variables (particularly income, spouse occupation, and number of children), these variables produced unreliable estimates and were excluded from the final interpretation. Only variables with stable and interpretable coefficients were retained in the final model.

Statistical significance was set at p < 0.05.

2.8. Ethical Consideration

Ethical approval was obtained from the Research Ethics Committee of the Federal University Teaching Hospital, Lafia. Participation was voluntary, and informed consent was obtained from all participants. Anonymity and confidentiality were strictly maintained throughout the study.

3. Results of IPV Analysis

Although 381 participants were recruited, 379 respondents with complete data were included in the final analysis. All analyses were conducted using the same complete-case sample (N = 379), ensuring consistency across Tables 1-3.

A total of 379 respondents were included in the final analysis. The majority were aged between 20 and 30 years. Most participants were married (99.2%) and predominantly Muslim (74.9%). A large proportion of respondents were unemployed (84.7%), and many reported low monthly income. Most participants had two or fewer children.

Table 1. Baseline sociodemographic, obstetric, and household characteristics of respondents (N = 379).

Table 2. Abuse assessment screen analysis.

Categories are not mutually exclusive; respondents may report more than one form of violence.

Table 3. Correlates of abuse types.

N = 379; **p < 0.01 (two-tailed).

Pearson correlation analysis demonstrated strong, positive, and statistically significant associations between all forms of IPV. Physical IPV was strongly correlated with emotional IPV (r = 0.856, p < 0.001), physical IPV was also strongly correlated with sexual IPV (r = 0.751, p < 0.001), and emotional IPV was strongly correlated with sexual IPV (r = 0.766, p < 0.001). These findings suggest that IPV often occurs as overlapping forms of abuse rather than as isolated events (Table 4).

Table 4. Association between sociodemographic characteristics and Intimate Partner Violence (IPV).

Age was recategorized for inferential analysis.

At the bivariate level, age, educational level, and monthly income were significantly associated with IPV, while religion, employment status, number of children, and spouse occupation were not (Table 5).

Multivariable logistic regression analysis identified employment status and educational level as significant predictors of intimate partner violence. Women who were employed had significantly higher odds of experiencing IPV compared to unemployed women (AOR = 6.34, 95% CI: 2.78 - 14.43, p < 0.001). Similarly, women with higher educational attainment had increased odds of IPV compared to those with lower education levels (AOR = 7.64, 95% CI: 3.29 - 17.74, p < 0.001).

Age, religion, and gestational age were not significantly associated with IPV after adjustment.

Table 5. Multivariable logistic regression of factors associated with IPV.

4. Discussion

This study provides important insight into the determinants of intimate partner violence (IPV) among pregnant women attending antenatal care in North-Central Nigeria. The findings show that educational attainment and employment status emerged as independent predictors of IPV, as shown in Table 5, while age group, religion, gestational age, spouse’s occupation, monthly income, and number of children were not significant after adjustment. These results reinforce the complex and context-dependent nature of IPV in pregnancy and suggest that the relationship between women’s socioeconomic position and IPV is not always straightforward.

The prevalence of IPV observed in this study confirms that violence during pregnancy remains a major public health concern. Table 2 shows that emotional violence was the most commonly reported form (22.2%), followed by physical (17.1%) and sexual violence (8.2%). These findings are consistent with global evidence indicating that IPV affects a substantial proportion of women worldwide and that pregnancy does not necessarily protect women from abuse [1] [2]. Similar patterns have been reported across sub-Saharan Africa, where IPV during pregnancy remains common and often under-recognized within routine maternity services [3] [4]. In Nigeria, previous studies have also documented substantial levels of IPV during pregnancy, with prevalence estimates broadly comparable to those found in the present study [16]-[19].

A key finding of this study is the strong positive correlation among emotional, physical, and sexual IPV, indicating that these forms of violence frequently co-occur rather than occur in isolation. As demonstrated in Table 3, correlation coefficients ranged from 0.751 to 0.856 (p < 0.001), reflecting a high degree of overlap between IPV subtypes. This supports the conceptualization of IPV as a multidimensional construct and justifies the use of a composite outcome measure in the regression analysis. Similar patterns of co-occurrence have been reported in previous studies, where emotional abuse often precedes or accompanies physical and sexual violence [4] [5].

At the bivariate level, age, educational level, and monthly income were significantly associated with IPV, as presented in Table 4, while religion, employment status, number of children, and spouse occupation were not. However, in the multivariable model, only educational attainment and employment status remained significant predictors, as shown in Table 5, highlighting the influence of confounding factors and the importance of adjusted analyses.

A notable finding of this study is that higher educational attainment was associated with increased odds of IPV. Table 5 further demonstrates that women with higher educational levels had significantly greater odds of experiencing IPV. This appears counterintuitive, as education is often considered protective against violence [11]. However, this finding has been reported in settings where women’s education may disrupt traditional gender roles without corresponding changes in male attitudes or societal norms [11] [12]. Increased autonomy and assertiveness among educated women may generate tension in relationships where patriarchal structures remain dominant, thereby increasing the risk of IPV.

Similarly, employment status emerged as an independent predictor of IPV. Findings in Table 5 indicate that employed women had higher odds of experiencing IPV compared to unemployed women. Although employment is typically viewed as empowering, it may, in certain sociocultural contexts, challenge established power dynamics within households and provoke conflict [11]-[13]. This highlights the complex interplay between economic empowerment and IPV risk, suggesting that employment alone may not be sufficient to protect women without broader shifts in gender norms and partner dynamics.

By contrast, age group, religion, and gestational age were not independently associated with IPV after adjustment, as also reflected in Table 5, despite some associations observed at the bivariate level. This indicates that these factors may be mediated by other socioeconomic or relational variables. The absence of an independent association with religion suggests that IPV transcends religious boundaries and is more strongly influenced by structural and interpersonal factors than by religious affiliation.

It is important to note that some variables, including spouse’s occupation, monthly income, and number of children, were associated with unstable estimates due to sparse data in some categories and were therefore not emphasized in the final regression model. This limitation is evident when comparing the bivariate findings in Table 4 with the multivariable results in Table 5 and underscores the challenges of modeling complex social phenomena with highly stratified categorical variables.

The implications of these findings are important for both clinical practice and policy. First, routine IPV screening should be integrated into antenatal care using validated tools such as the Abuse Assessment Screen [14] [15]. Second, interventions should move beyond individual-level approaches to address relationship dynamics and structural determinants of violence [11]-[13]. Third, programs aimed at women’s empowerment through education and employment should be complemented by efforts to engage male partners, promote gender equity, and support healthy conflict resolution within households.

This study adds to the growing Nigerian and African evidence base on IPV in pregnancy [12]-[19]. It highlights those socioeconomic transitions, particularly in relation to education and employment, which may shape vulnerability to violence in ways that are context-specific and socially mediated. Addressing IPV in this setting, therefore, requires multilevel interventions that combine antenatal screening, psychosocial support, male partner engagement, and broader community strategies aimed at transforming inequitable gender norms.

Intimate partner violence remains prevalent among pregnant women in this setting, with significant co-occurrence of abuse types. Integrating routine IPV screening into antenatal care and implementing partner-focused, socioeconomically informed interventions are essential for reducing IPV and improving maternal health outcomes in Nigeria.

4.1. Strengths

This study provides a comprehensive assessment of IPV among a representative sample of pregnant women in North-Central Nigeria, using validated instruments and robust statistical analyses, including chi-square analysis, correlation analysis, and logistic regression. The simultaneous evaluation of emotional, physical, and sexual IPV allows for a nuanced understanding of the co-occurrence of abuse forms. Furthermore, the study incorporates both descriptive and inferential analyses, highlighting associations and predictors, which can inform targeted interventions.

4.2. Weaknesses

1) The cross-sectional design limits causal inferences between sociodemographic factors and IPV. 2. Self-reported data on abuse may be subject to social desirability and recall biases, potentially leading to underreporting.

2) Additionally, certain cells in the Chi-square analyses had low expected counts, indicating sparse data in some categories.

3) The study was conducted at a single tertiary hospital, which may limit the generalizability of findings to other settings.

5. Conclusions

This study demonstrates that intimate partner violence (IPV) is a common experience among pregnant women attending antenatal care in North-Central Nigeria, with emotional abuse being the most prevalent form and substantial co-occurrence across different IPV types. The findings further reveal that women’s educational attainment and employment status are significant predictors of IPV after adjustment, underscoring the complex and context-dependent relationship between socioeconomic factors and vulnerability to violence.

These results highlight that IPV in pregnancy is not only prevalent but also multifaceted, requiring a comprehensive approach to identification and management. Routine screening for IPV should be integrated into antenatal care services using validated tools, with appropriate referral pathways for affected women. In addition, interventions should extend beyond individual-level responses to address broader relational and societal dynamics, including gender norms, power imbalances, and partner engagement.

Overall, addressing IPV in this setting will require coordinated, multilevel strategies that combine clinical screening, psychosocial support, community sensitization, and policy-driven efforts aimed at promoting gender equity and reducing violence against women.

6. Recommendations

Based on the findings of this study, the following actions are recommended:

1) Routine IPV Screening: Integrate structured IPV screening into antenatal care using validated tools such as the Abuse Assessment Screen, with emphasis on privacy and confidentiality.

2) Capacity Building for Healthcare Providers: Train antenatal care providers on identification, safe inquiry, and appropriate response to IPV, including trauma-informed care approaches.

3) Strengthening Referral Systems: Establish and formalize referral pathways linking health facilities to counseling services, social welfare units, legal aid, and community-based support organizations.

4) Male Partner Engagement: Develop community-based interventions that involve male partners to address harmful gender norms and promote healthy relationship dynamics.

5) Policy Integration: Incorporate IPV indicators into maternal health monitoring frameworks and national reproductive health policies to improve accountability and resource allocation.

6) Further Research: Conduct longitudinal and multi-centre studies to better understand causal pathways and improve generalizability.

Authors’ Contributions

Conception and Study design: PEA; LIC; EFA; LAA.

Study Protocol Design: PEA; LIC; MCM; SOM; DBB; RM; SCO; OA; EFA; LAA.

Manuscript Review: PEA; LIC; MCM; SOM; DBB; RM; SCO; OA; EFA; LAA.

Manuscript Approval: PEA; LIC; MCM; SOM; DBB; RM; SCO; OA; EFA; LAA.

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

References