Effect of intimate partner violence on depression severity among women in informal settlements of Uganda

Abstract

Background

Preventive measures to disrupt COVID-19 transmission negatively impacted mental health, especially among women. Increasing cases of intimate partner violence (IPV) in socio-economically disadvantaged populations such as informal settlement dwellers were documented. We examined the longitudinal effect of IPV on depression symptoms among women in informal settlements during the COVID-19 onset in Kampala, Uganda.

Methods

This longitudinal study surveyed 148 adult women living in three purposively selected informal settings of Kampala, Uganda. Baseline and two follow-up surveys were conducted three weeks apart during the first COVID-19 wave. Our outcome was depression symptoms, a continuous variable score measured using a validated patient health questionnaire-9 tool. IPV data were collected using the modified conflict tactics scale. Two sample t-tests compared mean depression scores within each IPV category. Mixed effects linear models tested the independent association between any IPV exposure and depression symptom severity while controlling for random effects and interaction terms between IPV exposure and time points.

Results

Among 148 respondents (mean age 33.09 ± 9.33 years), 58.1% experienced at least one form of IPV. Mean depression scores declined from 8.74 (± 5.71) at baseline to 7.22 (± 5.26) at midpoint and 6.60 (± 6.13) at endline. At baseline, 27.03% had none/minimal depression and 2.7% had severe depression. In the mixed effects linear models, compared to baseline, depression symptom scores significantly decreased by 3.03 at endline (Coef. -3.03, 95%CI -4.80,-1.26, p = 0.001), controlling for confounders. IPV exposure at mid-point was associated with a non-significant 1.41-unit increase in depression symptoms (Coef. 1.41, 95%CI -1.05,3.88, p = 0.282). However, at endline, IPV exposure was significantly associated with a 2.79-unit increase in depression symptoms when compared to baseline (Coef. 2.79, 95%CI 0.32,5.27, p = 0.027).

Conclusions

The increase in depression severity among women exposed to IPV underscores the compounding effects of IPV during early phases of the pandemic. Public health responses to outbreaks should integrate mental health support and violence prevention strategies, particularly for women experiencing IPV. Future research should explore the long-term impact of the pandemic on mental health, especially in informal settlements.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-03397-2.

Introduction

Depression, a common mood disorder is the leading cause of disability globally contributing significantly to the global burden of disease and is projected to be the leading cause of mental illness [1]. According to the American Psychiatric Association and World Health Organization’s (WHO), major depressive episodes are characterized by depressed mood or loss of interest in activities, accompanied by psychological or somatic symptoms, persisting most of the day, nearly every day, for at least two weeks [2, 3]. Depressive disorders affect over 322 million people globally, about 4.4% of the world’s population being more common among females (5.1%) than males (3.6%) [4]. The WHO African region including Sub-Saharan Africa (SSA) countries have lower depression rates at 4.1% (± 0.05%), than the Americas at 4.8% (± 0.09%) and the European region at 4.9% (± 0.1%) respectively [4]. Sub-Saharan Africa (SSA) bears a considerable share of this burden, compounded by the region’s socio-economic challenges and limited mental health infrastructure [5]. Depression impairs daily functioning and quality of life, leading to increased morbidity and mortality, including higher risk of suicide. Furthermore, depression contributes to poor maternal health outcomes such as premature births, and reduced productivity which can perpetuate cycles of poverty and illness.

In Uganda, the prevalence of depression is estimated at 30.2% with higher depression rates reported in some sub-populations. For example, the prevalence of depression is highest among refugees in Uganda at 67.6%, followed by post-war survivors (36.0%), people living with HIV (PLWHIV) (28.2%), postpartum or pregnant mothers (26.9%) while 23.6% of children and adolescents have depression [6]. Depression prevalence in Uganda rose significantly during the COVID-19 pandemic, increasing from 29.3% pre-pandemic to 48.1% [7]. However, a study among refugees in informal settlements in Uganda found no significant change in depression levels during the same period [8].

Current guidelines for addressing depression advocate combining clinical treatments, like medications and cognitive-behavioral therapy, with community-based preventive strategies targeting poverty, violence, and social isolation [7, 8]. However, implementing these approaches in low-resource settings like Uganda remains challenging. Primary healthcare systems are poorly equipped to provide mental health services, and there is limited integration of mental healthcare into general healthcare services. Furthermore, the lack of tailored interventions that address the specific needs of women in informal settlements perpetuates their vulnerability to depression.

Women and marginalized populations, especially those in informal settlements characterized by poverty, overcrowding, and limited access to healthcare including mental health services, experience higher rates of mental health disorders due to continuous exposure to these individual- and environmental risk factors. These risk factors for depression in this population dwelling in informal settlement are compounded vulnerabilities such as economic dependency, domestic violence, and caregiving responsibilities [9–13]. Other risk factors for depression among Ugandan women include economic insecurity, food scarcity, and lack of access to healthcare services and substance abuse contributing to the high rates of depression in informal settlements [14, 15]. Structural factors such as low education levels, and exposure to adverse childhood experiences further compound the risk of depression. Additionally, societal stigma around mental health often prevents women from seeking help, worsening the prevalence and severity of untreated depression. Despite these challenges, addressing depression in Uganda remains constrained by systemic barriers [16]. Limited mental health infrastructure, shortage of trained mental health professionals, competing health priorities and individual and societal stigma towards mental health conditions significantly impede the effective management of depression [17–19]. Women in informal settlements are particularly underserved, as health services often fail to account for the unique socio-economic and cultural contexts that shape their experiences of depression.

Intimate partner violence (IPV) is a pervasive public health issue in Uganda, with significant implications for women’s mental health. 11% and 44% of Ugandan women report experiencing sexual and physical IPV in their lifetime, potentially increasing their risk for depression [20]. IPV is particularly prevalent in informal settlements, where socio-economic stressors like high unemployment levels, overcrowding, poor housing conditions, and poverty create an environment that may trigger and/or worsen spousal abuse. Furthermore, limited access to social support and legal resources may exacerbate the psychological impact of IPV. Studies in Uganda consistently shows a strong association between IPV and depression, with abused women exhibiting higher rates of depressive symptoms, impaired functioning, and suicidality compared to their non-abused counterparts [21, 22].

During the COVID-19 pandemic, reports from law enforcement and healthcare providers revealed potential exacerbation of the syndemic of IPV [23] and depression [24] especially among informal settlement dwellers [25]. Lockdowns, curfews, and movement restrictions, while necessary to disrupt the COVID-19 transmission, these measures also contributed to rising economic insecurity and social isolation - two key drivers of IPV [26]. Women in informal settlements faced disproportionate hardships, including loss of income and restricted access to support services, which intensified the frequency and severity of IPV. Simultaneously, these contextual situations during early pandemic phases created an environment that amplified stressors such as fear of infection, bereavement from rising mortality rates, and uncertainty about their family’s future [27]. Psychological distress from employment loss affects livelihoods and quality of life especially in the poorest homes’ triggers IPV with women being disproportionately affected [28]. Depression is more likely to occur in the context of IPV [22] especially among vulnerable poor women. Mental health service delivery is further constrained by reallocation of healthcare resources to prioritized pandemic-related needs, leaving many women without access to counseling or psychiatric care.

By examining the role of IPV in driving depression symptom severity among women in informal settlements, this study provides critical insights into how pandemics exacerbate existing disparities in mental health and IPV. Addressing these gaps requires an understanding of the unique risk factors and experiences of depression among women in informal settlements during the COVID-19 pandemic. Therefore, the primary objective of this study was to assess the longitudinal relationship between IPV and depression symptom severity during the COVID-19 pandemic among women in informal settlements of Kampala, Uganda. We hypothesized that exposure to IPV would be associated with increased severity of depression symptoms over time, in order to highlight the compounding impact of pandemic-related stressors on vulnerable populations.

Methods

Study design, and setting

This longitudinal study was designed to assess adult women (aged ≥ 18 years) and older living in informal settlements within Kampala district over a three-month period from June to August 2020. This study was conducted during the national lockdown and subsequent easing of restrictions aimed at curbing the importation and community transmission of SARS-CoV-2 infections. Kampala, the capital city of Uganda, is located in the central region of Uganda and is administratively operated by the Kampala Capital City Authority (KCCA). Kampala district accommodates one-third (31%) of Uganda’s urban population, with 57 informal settlements distributed across the city’s five administrative divisions: Central, Nakawa, Makindye, Lubaga, and Kawempe [29, 30].

Sample size and sampling process

We estimated a sample size of 148 women based on an assumed prevalence of 22.0% for physical, sexual, or emotional violence experienced by women from their current or past partners [20], with 80% power, a 95% confidence level, and a ± 5% margin of error. Participants were recruited from three purposively selected informal settlements in the Kawempe and Makindye divisions of Kampala district. Using probability-proportional-to-size sampling, we recruited 50 respondents from both Ki-Mombasa and Kabalagala slums and 48 from Katanga slum. Our sampling frame consisted of the village residents’ registers maintained by the Local Council (LC) I chairpersons of the respective informal settlements. In each settlement, we randomly selected the initial respondent and selected subsequent participants systematically from these registers. The enrollment response rate was 100%, with all 148 women recruited; follow-up rates were 93.2% at midline and 93.9% at endline.

Data collection

Pre-tested, structured questionnaires were used to collect data through one-on-one interviews conducted at household level by nine trained research assistants. Local leaders, including LC I chairpersons or committee members, facilitated introductions between the data collectors and respondents at their homes. We administered the questionnaires at baseline and conducted two follow-up assessments at three-week intervals over a three-month period during the first phase of COVID-19 in Uganda. Data were collected at these three time points during the first and/or second weeks of June, July, and August 2020. We selected this interval based on contextual and methodological considerations. This study period coincided with Uganda’s partial COVID-19 lockdown, which began on May 26, 2020, following the full national COVID-19 lockdown which began on March 18, 2020. The partial lockdown involved a gradual easing of restrictions, including the reopening of general shops and select non-essential businesses, resumption of public transport at half capacity in non-border districts, and limited use of private vehicles (up to three occupants). Motorcycle taxis (boda bodas) were permitted to carry passengers until 6:00 pm, and the national curfew was adjusted to 9:00 pm–5:30 am. During this time, schools began phased reopening, movement restrictions in border districts were lifted, international borders reopened, and religious gatherings of up to 70 people were allowed.

Study measures

Dependent variable

Our outcome variable was depression severity, a continuous variable measured using a 9-item, patient health questionnaire-9 (PHQ-9). Each item’s response options are on a four point Likert scale ranging from 0 to 3 with higher scores suggestive of increased severity of depression symptoms [31]. The validated PHQ-9 had an internal consistency (Cronbach’s alpha) of 0.81 for this study.

Key independent variable

We computed intimate partner violence (IPV) exposure as a composite categorical variable using a subset of validated items to measure (physical, emotional, and sexual violence) from the Domestic Violence module of the 2016 Uganda Demographic Health Survey [32]. IPV was defined as experiencing at least one form of physical, emotional, or sexual violence from a current or former partner (husband or cohabiting partner). The definition was adapted from previously validated measures [33, 34] but modified for this study to specifically capture IPV exposure within the three weeks preceding the interview date.

We computed a composite overall IPV exposure (any IPV exposure) by combining responses to any of physical, emotional, or sexual IPV experienced in the past two months. First, physical IPV was assessed by asking participants whether their current or former partner in the past two months had ever ’pushed’, ‘shook or threw something at them’, “slapped them’, ’fist punched or hit with something harmful’, ‘kicked or dragged’, ‘strangled or burned’, “threatened with a knife/gun or other weapon’ and/or ‘arm twisted or pulled your hair’ them. The final physical IPV variable was coded “Yes” if any physical IPV was reported and “No” otherwise. Secondly, emotional IPV items with yes/no responses asked participants whether their husband/partner (current or last) in the last two months had ever ’humiliated’, ‘threatened you with harm’, and/or ‘insulted or made you feel bad’. The final emotional IPV variable was coded ‘Yes’ when participants reported experiencing at least one emotional IPV item and ‘No’ when none of the emotional IPV items was experienced in the last two months. Thirdly, sexual IPV items with yes/no responses asked participants whether their husband/partner (current or last) in the last two months had ever ‘physically forced her into unwanted sex with him’, ‘forced her into other unwanted sexual acts’ and/or ‘physically forced the participant to perform sexual acts when she did not want to’. Any IPV exposure was coded “Yes” if the participant reported at least one form of IPV (physical, emotional, or sexual) and “No” if none were reported.

Covariates

All independent variables assessed were categorical. These included the respondent’s age (18–24, 25–34, 35–44, 45–64 years), level of formal education attained (none, primary and secondary or above), whether the participant is currently married (yes/no), household size (number of people living in respondents’ home), occupation (employed/self-employed, unemployed/student), and survey timepoints (baseline, mid-point, and Endline). For ‘adherence to COVID-19 preventive measures’, we developed a composite count variable from four binary items assessing WHO recommended COVID-19 preventive behaviors namely: maintaining at least two meters distance from non-household members, wearing a face mask outside the home, frequent handwashing after contact with surfaces or people, and regular use of hand sanitizer. Based on changes between baseline and endline responses, we generated a 0–3 adherence scale: 0 = non-adherent at both time points, 1 = shifted from adherence to non-adherence, 2 = shifted from non-adherence to adherence, and 3 = consistently adherent. For analysis, due to low frequencies (fewer than five cases) in categories 0 and 2, the scale was recoded into a binary variable: (i) Remained adherent - adhered to all four measures at both time points, and (ii) Poorly adherent - all other adherence patterns.

Statistical analysis

Categorical variables at baseline were described using frequencies and their respective percentages. Continuous variables were described using means and respective standard deviations (SD). Student’s t-tests were used to assess differences in depression scores between IPV and no IPV exposure groups. One-way analysis of variance (ANOVA) with Bonferroni posthoc correction for multiple comparisons tested for significant differences in depressive scores between timepoints. We used Cohen’s D tests to estimate effect sizes in changes of depression severity scores between survey time-points. Simple linear regression models were conducted at unadjusted analysis to test the bivariate associations between the key independent variable and covariates and the outcome (depression symptom severity). We subsequently run a forward fitting, mixed effects regression model while adjusting for potential confounders to test the independent association between IPV and depression severity. We used a forward-fitting, multilevel mixed-effects linear regression model sequentially adding variables with statistical significance (p < 0.05) in unadjusted analyses, theoretical relevance, and prior literature on depression and IPV. This adjusted model controlled for age, years of education, marital status, employment, number of household members, and adherence to COVID-19 preventive measures as random-effect parameters, and interaction terms between IPV and survey timepoints as fixed effects. To assess for differential attrition, we compared baseline sociodemographic characteristics and IPV exposure between participants who completed both baseline and endline visits (retained) and those who had a baseline visit but were missing at the endline visit (lost to follow-up). We used independent sample t-tests to test for differences in continuous variables and Fisher’s exact test for categorical variables between the two groups. Data were analyzed using Stata/MP v18.5 and statistical significance considered at p < 0.05.

Results

Table 1. This study included 148 participants with a mean age of 33.09 years (SD ± 9.33). Most were aged 25–34 years (37.84%), with a mean household size of 4.87 (SD ± 2.81). Over half (54.05%) had between one to seven years of schooling, and 50.68% were currently married. Employment status showed 51.35% were self-employed, 27.7% were unemployed and 35.81% had no formal education. IPV exposure was reported by 58.11% of all study participants.

Table 1.

Baseline socio-demographic characteristics, N = 148

	Freq.	%
Characteristic
Age category (years)
18–24	28	18.92
25–34	56	37.84
35–44	48	32.43
45–64	16	10.81
Age (years)
Mean ± SD	148	33.09 ± 9.33
Number of household members
Mean ± SD	148	4.87 ± 2.81
Depression scores, Mean ± SD
Baseline	148	8.74 ± 5.71
Mid-point	138	7.22 ± 5.26
Endline	139	6.60 ± 6.13
Baseline depression severity
None or minimal	40	27.03
Mild	45	30.41
Moderate	35	23.65
Moderately severe	24	16.22
Severe	4	2.70
Years in school
None	53	35.81
> 7 years	15	10.14
1–7 years	80	54.05
Currently married
No	73	49.32
Yes	75	50.68
Employment
Formal	31	20.95
Self-employed	76	51.35
None	41	27.70
Sexual IPV
No	125	84.46
Yes	21	14.19
Physical IPV
No	87	58.78
Yes	61	41.22
Emotional IPV
No	78	52.70
Yes	70	47.30
Any IPV exposure
No	62	41.89
Yes	86	58.11
Adherence to COVID-19 preventive measures
Poorly adherent	50	33.78
Remained adherent	98	66.22

Table 2. In the unadjusted model, IPV exposure was associated with a 2.93-unit increase in depression symptom severity compared to no IPV exposure (Coef. 2.93, 95%CI: 1.20,4.67; p = 0.001). Over time, depression symptoms scores decreased by 1.86 units at mid-point (Coef.-1.86, 95%CI: -3.67,-0.07; p = 0.042) when compared to baseline depression symptom scores. Depression symptom scores decreased by 2.84 units at endline (Coef.-2.84, 95%CI: -4.61,-1.08; p = 0.002) when compared to baseline depression symptom scores. An interaction term revealed a 2.66-unit increase in depression severity at endline among those with IPV exposure compared to baseline (Coef. 2.66, 95%CI: 0.17, 5.15; p = 0.036). Sexual IPV was associated with a 3.38-unit higher depression severity score (Coef. 3.38, 95%CI: 0.77,5.99; p = 0.012), physical IPV with a 3.63-unit increase (Coef. 3.63, 95%CI: 1.83,5.42; p < 0.001), and emotional IPV with a 3.24-unit increase (Coef. 3.24, 95%CI: 1.45,5.03; p < 0.001), compared to no IPV exposure. Age, marital status, household size, and employment status were not associated with significant changes in depression symptom severity.

Table 2.

Mixed effects linear regression analyses

	Outcome: Depression symptom severity
	Unadjusted model		Adjusted model*
Characteristic	Coef (95%CI)	p-value	Coef (95%CI)	p-value
Any IPV exposure
No	Ref		Ref
Yes	2.93 (1.20, 4.67)	0.001	2.93 (1.22, 4.64)	0.001
Timepoint
Baseline	Ref		Ref
Mid-point	-1.86 (-3.67, -0.07)	0.042	-1.77 (-3.55, -0.01)	0.052
Endline	-2.84 (-4.61, -1. 08)	0.002	-3.03 (-4.80, -1.26)	0.001
Any IPV#Timepoint
Any IPV exposure#Baseline	Ref		Ref
Any IPV exposure#Mid-point	1.43 (-1.06, 3.91)	0.259	1. 41 (-1.05, 3.88)	0.261
Any IPV exposure#Endline	2.66 (0.17, 5.15)	0.036	2.79 (0.32, 5.27)	0.027
Sexual IPV
No	Ref
Yes	3.38 (0.77, 5.99)	0.012
Physical IPV
No	Ref
Yes	3.63 (1.83, 5.42)	< 0.001
Emotional IPV
No	Ref
Yes	3.24 (1.45, 5.03)	< 0.001
Age category, years
> 35	Ref
18–35	-0.62 (-2.56, 1.32)	0. 531
Years of education
> 7	Ref
1–7	-0.38 (-3.70, 2.94)	0.823
None	0.49 (-1.52, 2.50)	0.630
Currently married
No	Ref
Yes	0.78 (-1.08, 2.64)	0.409
Number of household members	-0.03 (-0.36, 0.30)	0.858
Employment
Formal	Ref
Self-employed	-0.36 (-2.78, 2.06)	0.768
None	-0.16 (-2.87, 2.54)	0.905
Adherence to COVID-19 preventive measures
Poorly adherent	Ref
Remained adherent	0.83 (-0.40, 2.05)	0.185

*Age, education, marital status, employment, number of household members, and adherence to COVID-19 preventive measures were random-effect parameters controlled for

Still from Table 2 among participants with any IPV exposure, depression symptom severity increased by 2.93 units when compared to those not exposed to any IPV (Coef. 2.93, 95%CI: 1.22,4.63; p = 0.001), controlling for potential confounders. Over successive survey rounds (time point), depression symptom severity scores decreased by 3.03 at endline (Coef. -3.03, 95%CI: -4.80,-1.26; p = 0.001) when compared to baseline, after controlling for potential confounders. Regarding the moderating effect of survey round, IPV-exposed individuals had a non-significant 1.41-unit increase in depression symptom severity scores at mid-point (Coef. 1.41, 95%CI:-1.05,3.88; p = 0.261) and a significant 2.79-unit increase at endline (Coef. 2.79, 95%CI: 0.32,5.27; p = 0.027) when compared to baseline. Additionally, Supplementary Table 2 shows descriptive mean depression scores across IPV typologies and timepoints, while Fig. 1 illustrates these temporal trends. Analyses in Supplementary Table 3 showed no significant differences between baseline sociodemographic characteristics, IPV exposure and loss-to-follow up.

Fig. 1.

Depression symptom scores by IPV typology over time

Discussion

This study examined the longitudinal effect of IPV on severity of depression symptoms among women in informal settlements during the first phase of the COVID-19 pandemic in Uganda. Overall, key findings revealed that: (i) at baseline, three out of five participants experienced at least one form of IPV, (ii) depression symptom severity decreased over time in our study population but remained significantly higher among women exposed to at least one IPV typology, and (iii) the interaction between IPV exposure and survey timepoint showed that depression severity increased significantly among women with IPV exposure at the endline when compared to baseline. This significant interaction effect highlights the compounding nature of IPV during pandemic-related public health emergencies. Although depression scores generally declined in our study population, women who experienced IPV continued to experience elevated levels of depression symptoms, particularly at the endline. These findings emphasize that IPV not only contributed to the depression burden at baseline but may also intensify or prolong psychological distress over time, particularly during periods of socio-economic distress like the COVID-19 pandemic.

Our findings add to the existing body of evidence on the intersection of IPV and mental health in resource-constrained settings. While previous studies in Uganda have documented the prevalence of IPV and its impact on mental health [33, 34], our research uniquely highlights the exacerbating effects of IPV on depression symptoms over time during the first phase of the COVID-19 pandemic in Uganda. In addition, the current findings expand our understanding of the challenges faced by socioeconomically disadvantaged women in informal settlements, supports prior explanations on how chronic stress related to the COVID-19 pandemic specifically IPV may interact to influence the severity of depression in the long-term if not adequately and promptly/timely addressed. Our findings align with prior research indicating that IPV is a strong determinant of depression among women [33, 34]. A prior study conducted in central Uganda reported similar relationships between emotional IPV, physical IPV and depression symptoms, emphasizing the pervasive impact of IPV on women’s mental health.

However, our findings differ from some recent Ugandan studies focused on informal settlements during the COVID-19 pandemic. For example, Logie et al. (2022) reported varying but generally high proportions of refugee adolescents and youth with depression in Kampala’s informal settlements, however, they did not identify IPV as a key contributing factor [35]. In contrast, our study highlights IPV as a persistent, and compounding stressor that increased depression symptom severity over time. Similarly, a previous study by Wafula et al. (2023) found significant associations between recent COVID-19 diagnoses and psychological distress among informal settlement residents [36]. This previous study highlights a broader burden of pandemic-related mental health challenges that extends beyond women in informal settlements. However, our findings add to this knowledge by identifying IPV as a modifier of depression symptoms. In a prior study, Anguzu et al. (2022) found that women in informal settings of Kampala, Uganda, who adhered to COVID-19 preventive measures experienced increased exposure to IPV [37]. However, the study did not assess the long-term mental health consequences. Our study fills this gap by demonstrating how IPV exposure during COVID-19 restrictions contributes to worsening depression trajectories over time.

Finally, Miller et al. (2021) found associations between alcohol use, IPV, and poor mental health among women in Wakiso, central Uganda, echoing the compounding role of IPV [38]. However, their cross-sectional design limited inferences on depression symptom progression. In contrast, our longitudinal approach offers deeper understanding of how IPV may exacerbate and sustain depression and potentially other psychological distresses during public health emergencies. It is probable that observed differences in findings across prior studies may be due to variations in study populations, temporal framing, and the types of psychosocial stressors examined. This underscores the need for stratified, context-specific mental health interventions that address IPV in informal settlements.

The reduction in overall depression scores over time likely reflects the potential effects of several factors during and after the first phase of the COVID-19 pandemic period. First, easing of the COVID-19 pandemic-related restrictions, could have improved household income/economic activity and social support. However, the persistent and worsening depression symptoms among IPV-exposed women in informal settlements suggests that IPV acts as a chronic stressor, may have negated any benefits from reduction in external stressors. This finding is consistent with prior studies which showed that IPV not only directly contributes to psychological distress but also limits women’s ability to recover from external adversities [33]. Additionally, prior studies from Uganda during the COVID-19 pandemic highlight several exacerbating factors for depression, including increased economic hardships, food insecurity, and restricted access to mental healthcare and social support services due to lockdowns [39–41]. The compounding effects of pandemic-related stressors, such as job loss and heightened household tensions/worsening relationships among intimate partners, have also been linked to rising depression rates among vulnerable populations, particularly women [40]. These challenges may have disproportionately affected IPV survivors, amplifying their psychological burden during and after the pandemic.

Secondly, another critical factor that could have contributed to worsening depression symptoms during the COVID-19 pandemic in Uganda was the significant disruption of healthcare services, including mental health care. Reports indicate that the diversion of resources to pandemic management and the restriction of movement negatively affected physical access particularly transport to healthcare centers [42] to seek and obtain mental health support, leaving many individuals without professional help. Societal stigma associated with seeking mental health services, exacerbated by misinformation, and fear surrounding COVID-19, further deterred individuals from seeking healthcare [13, 43, 44]. Additionally, the school closures and communal spaces disrupted social networks that often serve as informal support systems, particularly for women, heightening feelings of isolation and despair. The combined effects of these systemic and social disruptions likely compounded the psychological toll on vulnerable populations, such as those exposed to IPV, resulting in a worsening of depression symptoms.

Another significant contributor to worsening depression during the COVID-19 pandemic in Uganda was financial constraints associated with pandemic-related unemployment due to job losses. The prevailing uncertainty during the first COVID-19 phase was driven by fear of COVID-19 infection, income loss, and limited access to essential services which disproportionately affected residents of informal settlements. Despite being the central tenet to preventing community transmission of COVID-19, several preventive guidelines such as stay-at-home policies were restrictive. While our findings suggest increases in IPV and depression in the early COVID-19 phase, the short follow-up period limits our ability to determine whether these reflect lasting changes or short-term stress responses. Prior studies reveal that financial constraints, challenging shared decision making in households, and strict adherence to COVID-19 guidelines may have heightened tensions within homes, leading to increased cases of IPV [37, 45]. The pandemic also disrupted communal and religious gatherings, which are integral to social support systems in Uganda. For many individuals, particularly women, these gatherings provide a critical buffer against psychological distress, and their absence intensified feelings of isolation [46, 47]. Furthermore, cultural norms that discourage discussing mental health issues openly or seeking formal help added another layer of complexity, as many individuals resorted to coping in isolation, worsening their mental health outcomes [48]. These dynamics highlight the interplay of structural, social, and cultural factors that exacerbated depression during this period.

Study implications

The study underscores the profound impact of IPV on the mental health of women in informal settlements, especially after the early COVID-19 pandemic phases. Future research studies should examine the role of social and economic interventions in mitigating IPV effects on mental health. In terms of practice, the study highlights the need for integrating IPV screening and mental health support into primary healthcare services. Community health workers, who are often the first point of contact in informal settlements, should be trained to identify IPV survivors and provide trauma-informed care. Additionally, scalable interventions, such as the WHO’s Problem Management Plus (PM+), could be adapted to address the dual burden of IPV and depression in such settings [8]. From a policy perspective, there is an urgent need to prioritize IPV prevention and mental health support within national health and social welfare strategies. This includes enforcing laws against IPV, establishing safe shelters for survivors, and addressing structural inequalities that exacerbate vulnerability, such as housing instability and limited access to healthcare [49]. Moreover, public awareness campaigns aimed at reducing stigma around mental health and IPV can empower women to seek help.

Study strengths and limitations

Our longitudinal study enabled the observation of temporal changes in depression symptoms and IPV exposure. The use of a validated tool (PHQ-9) to measure depression adds reliability to the findings, and the focus on informal settlements provides insights into a marginalized population often overlooked in research. However, there are some study limitations. Although our three-week follow-up interval was pragmatic for the study context, in-depth insights of the long-term trajectory of depression in relation to IPV exposure would require longer follow-up periods. Additionally, the brief three-week follow-up interval enabled capture of the immediate psychological impact of IPV exposure during this early COVID-19 pandemic phase, ensure participant safety, maintain follow-up feasibility, and minimize attrition lockdowns and changing movement restrictions.

We used purposive sampling to efficiently reach informal settlement residents during the first COVID-19 phase, which may have introduced selection bias. Changes in depression and IPV observed over the brief three-week interval should be interpreted cautiously, as this period may be too short to reflect substantial shifts in depression symptoms or violence exposure. Self-reported data on IPV and depression could be subject to social desirability bias, potentially underestimating the true prevalence. Although our study achieved high follow-up rates (93.2% at midline and 93.9% at endline), modest attrition could still introduce attrition bias. However, we examined baseline differences between participants retained versus those lost to follow-up at any time during the study period and found no significant differences with sociodemographic characteristics or IPV exposure, suggesting minimal bias due to differential attrition.

Moreover, because participants were purposively sampled from selected informal settlements, the findings may have limited external validity and may not fully represent all women living in informal settlements across Kampala or other similar urban areas. Additionally, the current study did not account for the potential influence of other stressors, such as pre-existing mental health conditions or substance use, which could have confounded the results. Also, our study findings should be interpreted with caution given the limited COVID-19-specific data analyzed. Lastly, the short follow-up period may not capture the ‘longer’ term mental health effects of IPV and pandemic-related stressors.

Recommendations

First, multi-sectoral approaches that integrate mental health and violence response and prevention services are essential to address the interconnected issues of IPV and mental health. Other key support services essential for integrated care include social protection, and legal services. Second, interventions should be culturally tailored, considering the stigma and socio-cultural norms surrounding IPV and mental health in Uganda. Third, enhancing access to education and economic opportunities for women can reduce their vulnerability to IPV and its psychological consequences.

Conclusion

This study highlights the significant burden of IPV on mental health among women in informal settlements of Uganda in the post COVID-19 pandemic period. Our study suggests that IPV exposure is associated with higher depression symptom severity, and the interaction effects that increases the impact of IPV on depression severity over time. Addressing both IPV and depression will require a multi-sectoral approach, involving health, education, and community development sectors. Future research should evaluate the effectiveness of integrating IPV and mental health prevention programs on improving outcomes for vulnerable populations.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ANOVA

Analysis of variance

COVID-19

Coronavirus disease 2019

GBV

Gender-based violence

IPV

Intimate partner violence

KCCA

Kampala capital city authority

LC

Local council

MakRIF

Makerere university research and innovations fund

PHQ-9

Patient health questionnaire-9

PLWHIV

People living with HIV

PM+

Problem management plus

SD

Standard deviation

SSA

Sub-saharan africa

WHO

World health organization

Author contributions

RA conceptualized this study. RA, AK, SK and JNM contributed to the study design, supervised study implementation, and data analysis. RA, AK, SK, CA, ARS, JDG, LDC and JDG contributed to the interpretation of the study findings. RA drafted the initial manuscript. All authors reviewed and approved the final manuscript.

Funding

This study was funded by the Government of Uganda through the Makerere University Research and Innovations Fund (MakRIF). The funder had no role in the design, decision to publish or preparation of the manuscript.

Data availability

The dataset generated and analyzed during the current study is not publicly available due data sharing restriction established by the research ethics committee but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This research was reviewed and approved by the Higher Degrees and Research Ethics Committee of Makerere University School of Public Health. Confidentiality of participant information was guaranteed using an anonymized database. Informed consent was obtained from all study participants. To protect respondents disclosing IPV, female research assistants conducted one-on-one interviews, and confidentiality was strictly upheld. Participants reporting IPV were offered immediate no-cost access to a police hotline and referred to psychosocial support services through the gender-based violence (GBV) Prevention Network. We referred participants with symptoms of probable depression to government health facilities for further evaluation. The study adhered to the principles of the Declaration of Helsinki. Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.