Intimate partner violence and mental health outcomes among ever-married women in Mozambique

Abstract

Background

Intimate partner violence (IPV) is linked to adverse mental health outcomes. Despite its widespread prevalence and devastating consequences, the mental health impacts of IPV remain understudied in Mozambique. We examined the association between IPV and mental health outcomes among ever-married women in Mozambique.

Methods

This study adopted a cross-sectional design based on the most recent Mozambique Demographic and Health Survey (MDHS) conducted in 2022-23. The sample comprised of 3,957 ever-married women aged between 15 and 49 years. The outcome variable for our study is mental health outcomes specifically focusing on depressive and anxiety symptoms. The main independent variable is IPV, focusing on three forms: physical, sexual and emotional violence each measured as a binary variable (yes/no). A multivariate analysis using binary logistic regression was performed between IPV and mental health outcomes while controlling for other independent variables. Adjusted odds ratios (aOR) and their corresponding 95% confidence intervals (95% CI) were reported. All analyses were weighted to account for complex survey design.

Results

The prevalence of depressive symptoms was 11.5% (95% CI: 9.7–13.5) while that of anxiety symptoms was 29.9% (95% CI: 27.5–32.4). Ever-married women who experienced physical violence had significantly higher odds of depressive symptoms (aOR = 1.52, 95% CI: 1.01–2.30). Emotional violence was significantly associated with both depressive symptoms (aOR = 3.10; 95% CI: 2.08–4.61) and anxiety symptoms (aOR = 1.60; 95% CI: 1.22–2.11).

Conclusion

The findings highlight the need to integrate mental health services into IPV prevention and response programmes in Mozambique. Incorporating routine mental health screening within IPV services could enable early identification and timely support for victims.

Introduction

Intimate partner violence (IPV) remains a public health issue in sub-Saharan Africa and Mozambique in particular, with many women experiencing physical, sexual, and emotional abuse. At the same time, mental health challenges, including depression and anxiety, are increasingly prevalent, yet they remain under-researched particularly in relation to IPV. Among the general population, 19% reported depressive symptoms in the past year [1], and among female heads of households, 14% screened positive for depression [2], while 25% experienced generalized anxiety disorder (GAD) [3]. While the association between physical and sexual violence and mental health has been widely studied, emotional violence is often overlooked [4], despite evidence suggesting it may be even more detrimental to mental health than physical or sexual abuse [5, 6]. Evidence from South Africa, for instance, shows that women who experience only emotional or economic IPV report higher levels of depressive symptoms and are more likely to have suicidal thoughts compared to those subjected to physical or sexual IPV [4]. These mental health issues not only affect individual well-being but also hinder women’s ability to work, care for their families, and engage in community life, increasing the public health burden in Mozambique.

Mental health remains an overlooked outcome in the context of IPV in Mozambique. Existing studies on IPV have primarily focused on its predictors [7–9] or its impact on reproductive health [10], while research on mental health has examined various disorders [1, 11], and specific populations, such as individuals living with HIV [12, 13], displaced individuals due to armed conflict [14], rural communities [2], and mental health interventions [15, 16]. Few studies have examined the relationship between IPV and mental health. A study which investigated mental health outcomes among women who had experienced or perpetrated IPV, found higher levels of depression, anxiety, and somatization, however the sample was drawn from a clinical forensic setting, which limits the generalisability of the findings [17]. A recent study examined the prevalence and correlates of probable depression and anxiety symptoms in a national sample but did not include IPV in the analysis [18]. There is a need for further research to examine the mental health consequences of IPV at a national level, including emotional, physical and sexual forms of violence. This study contributes to the growing body of literature by examining these associations in a nationally representative sample of women in Mozambique.

Focusing on ever-married women in Mozambique is important due to their increased vulnerability to IPV and its adverse health outcomes. Women in Mozambique face gender inequalities, limited access to education and restricted economic opportunities, which not only increase their exposure to IPV but also hinder their ability to seek support or escape abusive relationships [7]. These challenges are perpetuated by socio-cultural norms that normalise violence against women, leaving many trapped in cycles of abuse. Understanding the association between IPV and mental health outcomes such as anxiety and depression symptoms, is essential for developing integrated interventions that combine mental health support with initiatives to prevent and respond to IPV. Such efforts are vital for achieving Sustainable Development Goal (SDG) 3, which aims to ensure good health and well-being for all, as addressing IPV and improving mental health are key components of this global agenda.

This study examines the association between three forms of IPV: emotional, physical, and sexual violence and mental health outcomes, specifically anxiety and depressive symptoms, among ever-married women in Mozambique. It hypothesises that women who experience IPV are more likely to report adverse mental health outcomes, such as depression and anxiety symptoms, compared to those who do not experience IPV. Using nationally representative data from the Demographic and Health Survey (DHS), this research expands on previous studies that were limited to specific regions, such as Maputo, providing a more understanding of the association between IPV and mental health outcomes across the country. The findings will inform the development of integrated interventions that address both IPV and mental health.

Methods and data

Data source, study design and sample

This study used data from the recent Mozambique Demographic and Health Survey (MDHS) conducted in 2022-23. The DHS program is designed to collect and provide data on various demographic and health related indicators such as fertility, contraceptive use, maternal health, and child health in low and middle income countries (LMICs) [19]. The DHS are standardised surveys that use a cross-sectional design to collect data from LMICs. The DHS is a nationwide cross-sectional survey which utilises a two-stage stratified cluster sampling approach, dividing each region into rural and urban segments. In the first stage, primary sampling units (PSUs) were selected based on size and independently selected within each stratum. In the second stage, households were selected systematically with equal probability. The details of the methodology employed in the DHS are available online [20]. The study population of interest included ever-married women of reproductive ages 15–49 years who were sampled in the 2022-23 MDHS. The data included a module on mental health. This module is important because it enables research that provides data on the prevalence, risk factors, and effects of mental health conditions within the population. A sample of 3,957 ever-married women who responded to the IPV module in DHS formed the unit of analysis. The category “ever-married women” included those who were currently married, living with a partner, widowed, or separated at the time of the survey. All women who were not ever-married, who were not selected to answer questions in the domestic violence module, or who had missing data on the outcome variables were excluded from the study. The data was weighted to ensure national representativeness and to correct oversampling and undersampling. Figure 1 shows the sample derivation criteria for the unit of analysis.

Fig. 1.

Sample derivation criteria for the unit of analysis

Study variables and measurement

Outcome variable

The outcome variable for our study is mental health outcomes, focusing on depressive and anxiety symptoms. Depressive symptoms were assessed using the Patient Health Questionnaire (PHQ-9) consisting of 9 questions asked to women about their mental health status in the last two weeks before the survey. This is a nine-item depression scale used widely as a screening tool in healthcare and survey settings. Each of the 9 items is scored on a 4-point Likert-type scale from 0 (not at all) to 3 (nearly every day). Total scores range from 0 to 27. Similar to other studies, a cut-off score of 10 or higher indicate the presence of depressive symptoms coded as 1 and scores less than 10 coded as 0 indicating no presence of depressive symptoms [13, 21]. Anxiety symptoms were assessed using the General Anxiety Disorder (GAD-7) consisting of 7 questions asked to women. The scores range from 0 to 21, with a cut-off score of 6 or higher indicating the presence of anxiety symptoms as stated in the 2022-23 MDHS report. This was coded as 1 while scores less than 6 coded as 0.

Independent variables

The main independent variable for this study is IPV and more specifically, three forms of violence: physical, emotional and sexual [22, 23]. Each form of violence was constructed as a composite variable based on a series of questions that were used to measure its occurrence. The IPV module was administered to one randomly selected woman of reproductive age in each household. For this analysis, the experience of any form of physical or emotional or sexual violence in the past 12 months prior to the survey was used to measure IPV. Physical violence included any of the following acts: being pushed, slapped, punched with a fist, kicked, strangled, threatened by knife, and arm twisted. Therefore, experiencing any of the seven acts of physical violence was classified as having experienced physical violence. Emotional violence included any of the following acts ever been threatened of harm, insulted or made to feel bad, ever been humiliated in front of others. Sexual violence included any of the following acts ever been physically forced into unwanted sex and forced into unwanted sexual acts. The experience of any of the sexual acts qualified the respondent as having experienced sexual violence. All these variables had two response categories (yes/no) in the DHS. Therefore, the main independent variable IPV (Physical, Emotional or Sexual) were constructed as composite variables comprising of experiences of either physical or emotional or sexual violence.

The study controlled for the following predictor variables: age of a woman, education level, place of residence, household wealth status, religious denomination, number of children ever born, and employment status. These independent variables were selected based on their relevance in previous studies as predictors of mental health outcomes among women [24–26]. All covariates that had a p-value of less than 0.20 were considered for inclusion in the multivariable model. Multicollinearity was checked in all the predictor variables to separate the independent effects of the interrelated variables. The detailed description of the variables is presented in Table 1.

Table 1.

Definition of individual and household level predictor variables

No	Variable name	Definition and measurement
1	Age of a woman	Age of respondents at time of survey: range from 15–49 years
2	Place of residence	Place of residence where the woman lived: 1 = Urban; 2 Rural
3	Education level	Education level attained by the respondent: 0 = No education; 1 = Primary; 2 = Secondary; 3 = Higher
4	Household wealth status	Income classification of the respondent’s household: 1 = Poor; 2 = Middle; 3 = Rich
5	Employment status	Employment status of the respondent: 0 = Unemployed; 1 = Employed
6	Number of children ever born	Number of children ever born to the respondents: 0 = 0 Child; 1 = 1–2 children; 2 = 3–4 children; 3 = 5 or more children
7	Religious denomination	Religious denomination where a respondent belongs to; 1 = Catholic; 2 = Islamic; 3 = Zion; 4 = Evangelical; 5 = Anglican; 6 = Other
8	Experience of physical violence	Experience of physical violence (pushed, slapped, punched with a fist, kicked, strangled, threatened by knife, and arm twisted) by a respondent in the last 12 months prior to the survey: 0 = No, 1 = Yes
9	Experience of sexual violence	Experience of sexual violence (ever been physically forced into unwanted sex and forced into unwanted sexual acts) by a respondent in the last 12 months prior to the survey: 0 = No, 1 = Yes
10	Experience of emotional violence	Experience of emotional violence (Partner said or did something to humiliate you in front of others, partner threaten to hurt or harm you or someone you care about, partner insult you or make you feel bad about yourself) by a respondent in the last 12 months prior to the survey: 0 = No, 1 = Yes

Statistical analysis

Statistical analysis was conducted using Stata version 17 software; the analysis considered survey design and post-stratification weights within DHS datasets. To account for the survey design, the analysis utilised “svy” commands to adjust for the two-stage sampling of the DHS. To make sample estimates representative of the population, we used post-stratification weights from DHS datasets. The “svyset” command was used in Stata software to define the survey weights. Before performing analysis, the command “svyset cluster [pweight = sampleweight], strata (stratum)” incorporated weights, PSUs, and stratification, resulting in population-representative analysis estimates.

Statistical analysis was conducted at three levels: descriptive, bivariate and multivariable logistic regression. Descriptive analysis involved describing the prevalence of mental health outcomes among ever-married women. Bivariate analysis included the cross-tabulation using Pearson’s chi-square test of the main independent variable IPV (measured through experiences of physical, sexual and emotional violence) and other predictors with the two mental health outcomes (depressive and anxiety symptoms). Furthermore, multivariable analysis involved performing multiple binary logistic regression models to examine the association between IPV and mental health outcomes, while controlling for other independent variables. Adjusted odds ratios (AOR) with corresponding p-values were calculated. In this study, we analysed the association between IPV and the experience of depressive and anxiety symptoms separately. Univariate and multivariate logistic regression models were separately performed for each outcome.

Ethical consideration

Permission to use the country datasets for 2022-23 Mozambique DHS was sought from the DHS Program. Ethical approval was not needed for the secondary analysis as ICF International and national statistical agencies already met all ethical protocols during primary data collection.

Results

Description of the characteristics of the sampled ever-married women

The percent distribution of sample characteristics of all ever-married women who were included in the analysis are presented in Table 2. The age distribution showed a concentration in the younger adult years, with ever-married women aged 20–24 years forming the largest group at 20.5%, closely followed by those aged 25–29 at 19.9%. The 15–19 age group represented the smallest portion at 9.3%. A significant rural-urban divide was evident, with 66.1% of participants residing in rural areas and 33.9% in urban settings. Education levels varied widely among participants, with primary education being the most common at 44.7%. Those without formal education constituted 32.2%, while higher education was least represented at 2.2%. The wealth status of participants showed a relatively even distribution between poor and rich categories at 40.1% and 39.5% respectively, with 20.4% falling in the middle category. In terms of employment, a significant majority of 68.1% reported not working, compared to 31.9% who were employed. The study also examined experiences of emotional violence among the respondents. Emotional violence was the most reported form at 18.8%, followed by physical violence at 15.6% while sexual violence was reported by 4.3% of the women.

Table 2.

Percent distribution of selected background characteristics of ever-married women (15–49 years) MDHS 2022-23

Background Characteristics	Frequency	Percent
Age
15–19	350	9.3
20–24	771	20.5
25–29	750	19.9
30–34	527	14.0
35–39	551	14.6
40–44	430	11.4
45–49	392	10.4
Place of residence
Urban	1278	33.9
Rural	2494	66.1
Education level
None	1215	32.2
Primary	1685	44.7
Secondary	788	20.9
Higher	83	2.2
Wealth status
Poor	1513	40.1
Middle	769	20.4
Rich	1490	39.5
Employment status
Not working	2569	68.1
Working	1202	31.9
Children ever born
0	257	6.8
1–2	1372	36.4
3–4	1155	30.6
5+	988	26.2
Religious denomination
Catholic	1154	30.6
Islamic	804	21.3
Zion	468	12.4
Evangelical/Pentecostal	990	26.2
Other	356	9.5
Experience of physical violence
No	3185	84.4
Ye	587	15.6
Experience of sexual violence
No	3610	95.7
Yes	161	4.3
Experience of emotional violence
No	3063	81.2
Yes	708	18.8

Distribution of ever-married women by experience of depression or anxiety symptoms

Table 3 presents the distribution of sampled ever-married women who exhibited symptoms of depression or anxiety symptoms in Mozambique by various background factors. Overall, the prevalence of depressive and anxiety symptoms was 11.5% and 29.9%, respectively. Ever-married women aged 20–29 and 45–49 were most affected by depressive symptoms (both 13.3%) and those aged 40–44 were the least affected (6.7%). For anxiety symptoms, women aged 25–29 were most affected (32.6%) while those 15–19 were the least affected (25.1%).

Table 3.

Percent distribution of ever-married women who experienced depressive or anxiety symptoms in the last 12 months prior to survey by background factors, MDHS 2022-23

Background Characteristics	Depressive symptoms (n = 3,957)		Anxiety symptoms (n = 3,957)
	N	%	N	%
Age		Ns		Ns
15–19	40	11.4	86	25.1
20–24	103	13.3	252	32.3
25–29	95	12.7	245	32.6
30–34	52	9.9	141	27.0
35–39	63	11.3	164	29.9
40–44	29	6.7	122	28.3
45–49	52	13.3	119	30.1
Place of residence		Ns		Ns
Urban	128.7	10.1	368	28.8
Rural	304.2	12.2	760	30.5
Education level		*		*
None	160	13.2	415	34.0
Primary	205	12.2	469	27.8
Secondary	63	8.0	227	29.0
Higher	4	4.8	17	20.1
Wealth status		**		***
Poor	233	15.4	541	35.6
Middle	72	9.4	204	26.8
Rich	128	8.6	383	25.7
Employment status		**		***
Not working	335	13.0	837	32.6
Working	98	8.1	291	24.2
Children ever born		Ns		Ns
0	26	10.1	72	28.1
1–2	185	13.5	429	31.4
3–4	135	11.7	342	29.4
5+	87	8.8	286	28.9
Religious denomination		***		***
Catholic	169	14.7	436	37.7
Islamic	140	17.4	298	37.1
Zion	23	4.9	77	16.4
Evangelical/Pentecostal	72	7.3	229	23.2
Other	29	13.1	88	25.8
Experience of physical violence		***		***
No	310	9.7	843	26.5
Yes	123	20.9	285	47.8
Experience of sexual violence		*		***
No	403	11.2	1049	29.1
Yes	30	18.4	79	47.5
Experience of emotional violence		***		***
No	267	8.7	824	26.9
Yes	166	23.4	304	42.7
Total	433	11.5	1128	29.9

Ns Not significant

* p < 0.05

** p < 0.01

*** p < 0.001

By educational level, ever-married women with no formal education were most affected by both depressive (13.2%) and anxiety (34.0%) symptoms. Those with higher education were the least affected (4.8% for depression and 20.1% for anxiety). In terms of wealth status, poor ever-married women were most affected by both depressive (15.4%) and anxiety (35.6%) symptoms. Middle-class ever-married women were least affected by both depressive (9.4%) and anxiety (26.8%) symptoms. Furthermore, those from rich households were least affected by depressive symptoms (8.6%) and anxiety symptoms (25.7%).

In terms of employment status, non-working ever-married women were more affected by both depressive (13.0%) and anxiety (32.6%) symptoms compared to those working (8.1% and 24.2%, respectively). Ever-married women who experienced violence showed notably higher rates of mental health issues compared to those who did not. Among those who experienced physical violence, 20.9% were affected by depressive symptoms and 47.8% by anxiety symptoms, compared to 9.7% and 26.5% respectively for those who did not experience physical violence. For sexual violence, 18.4% were affected by depressive symptoms and 47.5% by anxiety symptoms, versus 11.2% and 29.1% for those who did not experience such violence. In case of emotional violence, 23.4% were affected by depressive symptoms and 42.7% by anxiety symptoms, compared to 8.7% and 26.9% for those who did not experience emotional violence.

Determinants of experiences of depression or anxiety among ever-married women

Table 4 presents the determinants of depressive and anxiety symptoms among ever-married women in Mozambique. After controlling for covariates, we observed that among ever-married women, several factors were significantly associated with mental health symptoms. These factors included age, wealth status, employment status, religious denomination, and experiences of physical, sexual, and emotional violence. Ever-married women aged 45–49 had significantly higher odds of experiencing depressive symptoms (aOR = 2.26; 95% CI: 1.07–4.79) and anxiety symptoms (aOR = 1.84; 95% CI: 1.11–3.05) compared to those aged 15–19. In terms of wealth status, middle-class ever-married women had significantly lower odds of experiencing both depressive symptoms (aOR = 0.61; 95% CI: 0.38–0.98) and anxiety symptoms (aOR = 0.71; 95% CI: 0.53–0.95) compared to those who belonged to poor households. Ever-married women who reported being employed had lower odds of experiencing anxiety (aOR = 0.77; 95% CI: 0.63–0.94) compared to non-working ever-married women. In terms of religious denomination, compared to Catholic ever-married women, Zion adherents had significantly lower odds of experiencing both depressive symptoms (aOR = 0.30; 95% CI: 0.16–0.56) and anxiety symptoms (aOR = 0.35; 95% CI: 0.24–0.53).

Table 4.

Multivariate logistic regression analysis of the associations between the characteristics of ever-married women and symptoms of depression and anxiety, MDHS 2022-23

Background Characteristics	(n = 3957)			(n = 3957)
	Depressive symptoms			Anxiety symptoms
Age	AOR	95% CI	p-value	AOR	95% CI	p-value
15–19	Ref			Ref
20–24	1.12	(0.61–2.06)	0.710	1.42	(0.97–2.09)	0.075
25–29	1.28	(0.65–2.52)	0.471	1.56	(1.03–2.36)	0.035*
30–34	1.36	(0.66–2.81)	0.395	1.40	(0.90–2.16)	0.131
35–39	1.75	(0.79–3.85)	0.161	1.77	(1.15–2.70)	0.009***
40–44	0.95	(0.42–2.13)	0.905	1.66	(0.99–2.79)	0.055
45–49	2.26	(1.07–4.79)	0.033*	1.84	(1.11–3.05)	0.018*
Place of residence
Urban	Ref			Ref
Rural	0.80	(0.44–1.43)	0.456	0.78	(0.58–1.06)	0.112
Education level
None	Ref			Ref
Primary	1.05	(0.74–1.48)	0.774	0.85	(0.66–1.09)	0.202
Secondary	0.71	(0.44–1.16)	0.175	1.01	(0.73–1.39)	0.965
Higher	0.48	(0.10–2.14)	0.033*	0.70	(0.36–1.35)	0.284
Wealth status
Poor	Ref			Ref
Middle	0.61	(0.38–0.98)	0.042*	0.71	(0.53–0.95)	0.020**
Rich	0.58	(0.31–1.10)	0.101	0.62	(0.43–0.89)	0.010**
Employment status
Not working	Ref			Ref
Working	0.79	(0.56–1.10)	0.171	0.77	(0.63–0.94)	0.012**
Children ever born
0	Ref			Ref
1–2	1.35	(0.76–2.41)	0.300	1.11	(0.78–1.60)	0.558
3–4	0.98	(0.50–1.92)	0.959	0.95	(0.63–1.45)	0.821
5+	0.60	(0.31–1.17)	0.137	0.86	(0.55–1.33)	0.491
Religious denomination
Catholic	Ref			Ref
Islamic	1.35	(0.89–2.05)	0.148	1.05	(0.78–1.41)	0.752
Zion	0.30	(0.16–0.56)	0.000***	0.35	(0.24–0.53)	0.000***
Evangelical/Pentecostal	0.49	(0.30–0.78)	0.003**	0.54	(0.38–0.76)	0.000***
Other	0.26	(0.05–1.29)	0.100	0.64	(0.28–1.46)	0.287
Experience of physical violence
No	Ref			Ref
Ye	1.52	(1.01–2.30)	0.045*	2.11	(1.55–2.90)	0.000***
Experience of sexual violence
No	Ref			Ref
Yes	1.28	(0.72–2.27)	0.391	1.56	(1.05–2.33)	0.028**
Experience of emotional violence
No	Ref			Ref
Yes	3.10	(2.08–4.61)	0.000***	1.60	(1.22–2.11)	0.033*

Ref  Reference, AOR Adjusted Odds Ratio

* p < 0.05

** p < 0.01

*** p < 0.001

Furthermore, Evangelical or Pentecostal ever-married women had lower odds of depressive symptoms (aOR = 0.49; 95% CI: 0.30–0.78) and anxiety symptoms (aOR = 0.54, 95% CI: 0.38–0.76). Finally, regarding the experiences of violence, ever-married women who experienced physical violence had higher odds of experiencing depressive (aOR = 1.52, 95% CI: 1.01–2.30) and anxiety symptoms (aOR = 2.11, 95% CI: 1.55–2.90). Additionally, those who experienced emotional violence had significantly higher odds of both depressive symptoms (aOR = 3.10; 95% CI: 2.08–4.61) and anxiety symptoms (aOR = 1.60; 95% CI: 1.22–2.11).

Discussion

Our findings indicate that the prevalence of depressive symptoms was 11.5%, while anxiety symptoms were reported at 29.9%. This prevalence is comparable to findings from other Southern African countries with similar socio-demographic contexts. For instance, a study in Zimbabwe reported a 15.4% prevalence of depressive symptoms [27], while in Malawi, 9.5% of women screened positive for antenatal depressive symptoms [28]. Although slightly lower than Zimbabwe’s, the prevalence in Mozambique is higher than that reported in Malawi. Our findings also show slightly higher levels of depressive and anxiety symptoms compared to a recent study in Mozambique, which reported a prevalence of 10% and 11%, respectively [18]. These differences may be partly explained by variations in sample characteristics. Specifically, our study included ever-married women who responded to the domestic violence module, a subgroup that may be more vulnerable to mental health challenges due to their exposure to traumatic experiences and stressors associated with IPV.

In both high- and lower-income countries women exposed to IPV are at an increased risk for mental health challenges. Our study indicates that ever-married women who experienced physical violence had a higher likelihood of reporting depressive symptoms compared to ever-married women who did not experience it. These findings reinforce existing evidence, suggesting that physical violence is a key risk factor for depression in women [24–26]. For instance, a study in Ethiopia found a link between physical violence and depressive episodes [24]. Also, studies in high income countries such as Sweden highlighted that women who experienced physical or sexual violence were at a higher risk of reporting the symptoms of depression than women who were not exposed [26]. These findings suggest that this relationship is not unique to one setting but is also observed across diverse populations and settings. Other studies have shown that any type of intimate partner violence is associated with various mental health issues such as depression, posttraumatic stress disorder and suicidal ideation [29, 30]. The findings of our study could be attributed to Mozambique having a high prevalence of IPV, with cultural norms that tolerate or justify physical violence against women. Women experiencing IPV may feel trapped in abusive relationships due to social expectations and financial dependence, which can contribute to chronic stress and emotional distress, thereby increasing the risk of depression and anxiety symptoms.

Our findings show that ever-married women who experienced emotional violence had an increased likelihood of having depressive symptoms and anxiety symptoms compared to ever-married women who did not experience any form of violence. These findings are consistent with existing literature which has shown a link between experiences of emotional violence and mental health challenges in women [4, 31]. For instance, a study in Ethiopia, indicated that emotional violence, particularly severe emotional violence, had the strongest association with depressive episodes [24]. A study in the US highlighted that emotional abuse, more than physical or sexual abuse, was the only form of IPV that had long-term effects on mental health [31]. In other words, emotional abuse has a lasting psychological impact, making individuals more likely to experience long-term depression. The effects of emotional violence on mental health have implications for women across various stages of life, including pregnancy and the postnatal period [32, 33]. Emotional violence is linked to psychological distress, which in turn increases the risk of perceived stress [34]. The constant exposure to such abuse can erode self-esteem, create feelings of worthlessness, and contribute to persistent worry and fear, which increases the likelihood of mental health disorders.

Due to the cross-sectional nature of the data used, we were not able to determine whether IPV occurred before or after the onset of depressive or anxiety symptoms. As a result, the study cannot confirm a cause-and-effect relationship between IPV and poor mental health outcomes. There is a possibility of reverse causality, rather than IPV leading to mental health issues, it is theoretically possible that ever-married women with pre-existing mental health conditions, such as depression or anxiety, may be more vulnerable to experiencing IPV. Longitudinal studies are required to establish causal relationships. The data rely on self-reported measures of IPV and mental health, which may be subject to social desirability bias and underreporting particularly in contexts where IPV is stigmatised or normalised. Although we adjusted for various covariates, there remains a possibility of residual confounding due to unmeasured variables that were not captured in the dataset. Furthermore, the use of logistic regression in this analysis may have inflated the estimated associations, particularly because the outcomes of interest are considered to be relatively common since the prevalence is greater than 10%. This is due to the fact that odds ratios can overstate the magnitude of effect compared to risk ratios when the outcome is not rare. Despite these limitations, the study provides insights into the relationship between IPV and ever-married women’s mental health outcomes.

Conclusion

In conclusion, this study reveals that experiences of IPV, particularly physical and emotional violence were statistically significantly associated with depressive symptoms among ever-married women. Furthermore, ever-married women who experienced emotional violence were more likely to report anxiety symptoms. These findings highlight the need for integrated mental health and gender-based violence interventions to support ever-married women affected by IPV. There are several pathways through which IPV may impact mental health, including chronic stress from exposure to violence, social isolation resulting from controlling or abusive relationships, and learned helplessness, where repeated trauma diminishes a woman’s sense of agency and coping ability. Integrating routine mental health screening into IPV response services could facilitate early identification and support for victims.

Authors’ contributions

MM: Conceptualization, Writing – original draft. MP & MH: Data curation, Methodology, Formal analysis, Writing – original draft. TS: Investigation, Conceptualization, Review & editing.

Funding

This work is based on the research supported by the South African Medical Research Council (SAMRC) and National Research Foundation (NRF) of South Africa (Grant Numbers: 150571).

Data availability

The study used secondary data obtained from Demographic and Health Survey available at https://dhsprogram.com/Data/.

Declarations

Ethics approval and consent to participate

We sought permission to use the 2022-23 Mozambique DHS secondary dataset from the DHS Program. The Demographic and Health Survey data has been anonymised prior to its public release. Ethical approval was not needed for the secondary analysis as ICF International and national statistical agencies already met all ethical protocols during primary data collection. Further information on DHS data and ethical standards can be found at: https://dhsprogram.com/methodology/protecting-the-privacy-of-dhs-survey-respondents.cfm.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.