A descriptive cross-sectional design was used in this study, which was conducted in Mogadishu, the capital city of Somalia, which has been affected by prolonged civil war and conflict over the last three decades. Mogadishu is now recovering from this long conflict and hosts most of the internally displaced persons (IDP) camps in Somalia. In this study, nine main maternal and child health centres (MCHs) were selected based on the population density of the nine Mogadishu districts, as per an unpublished document from the Banadir Regional Administration. Of these nine districts, three had high population density, three had low population density, and three had an average population density. All selected MCHs were funded by government and nongovernmental organisations to provide primary health care, including antenatal care, treatment of malnutrition, immunisation, basic laboratory diagnosis, family planning, and health education services, among others. Approximately 80 to 200 women visit these MCHs daily and are served by between 18 and 48 health-care professionals.

The study population comprised women who attended the nine selected MCH facilities during the study period. Each of the nine districts contributed a quota of 100 women, resulting in a total sample of 900 participants. A convenience sampling approach was used. In each district, recruitment was conducted at the primary Maternal and Child Health (MCH) facility, which is supported by both governmental and non-governmental organizations. The data were collected between August and September 2021 by four experienced data collectors, supervised by one of the first authors (GD). The data collectors received refreshment training on data collection specifically for this study, which was conducted by the authors (GD, FO), who also supervised their subsequent data collection activities. The questionnaire was available in Somali; however, as most participating women could not read or write in Somali, the data collectors read the questions aloud and recorded the participants' responses.

Ethical approval was obtained from the School of Public Health at Somali National University (SNU/SPHR/005/2021). Permission was sought from the district and health facility levels before contacting the MCHs. Information about the study was given to the health-care professionals working at each MCH. Together with one of the first authors (GD), the data collectors visited each MCH. The women visiting the MCHs were informed about the study individually. Those who expressed interest in participating and provided verbal consent were included in the study. In total, 900 women provided consent and completed the questionnaire with the help of the data collectors. One participant who did not complete the questionnaire was excluded from the analysis; thus, 899 women were included. All data were collected anonymously.

The data were collected through a self-report validated questionnaire in Somali. The dependent variables were the Patient Health Questionnaire-9 (PHQ-9) and the seven-item version of the Generalised Anxiety Disorder scale (GAD-7). The independent variables included the 13-item version of the SOC scale (SOC-13) and questions on sociodemographic factors, including age, number of children, marital status, pregnancy status, regular visits to MCHs, educational status, employment status, housing situation, and monthly household income.

All instruments were available in Somali except SOC-13, which we translated back and forth between English and Somali using the World Health Organisation guidelines for the Process of translation and adaptation of instruments (17).

Before data collection, the questionnaire was tested at another MCH that was not included among the MCHs selected for this study.

Descriptive statistics, including means, standard deviations (SD), ranges, frequencies, and proportions, were computed using STATA software (v.16; StataCorp LLC, College Station, TX, USA). Multicollinearity among the independent variables was assessed prior to the regression analysis, and no evidence of multicollinearity was detected. Because the outcome variables (PHQ-9 and GAD-7) showed right-skewed distributions, robust standard errors were applied to improve the reliability of the regression estimates. The prevalence of depressive and anxiety symptoms is illustrated in Figure 1.

Both univariate and multivariate linear regression analyses were conducted to examine the associations between the dependent variables (PHQ-9 and GAD-7) and the independent variables (SOC-13, perceived social support, and sociodemographic factors). Regression coefficients (β) and their corresponding 95% confidence intervals (CIs) are reported. Statistical significance was defined as a p-value < 0.05.

Table 1 shows that the mean levels for the PHQ-9 and GAD-7 scales were 4.72 (SD ± 5.15) and 4.79 (SD ± 4.87), respectively, which indicates the low mean levels of depressive and anxiety symptoms among Somali women in Somalia. When these symptoms were classified into mild, moderate and severe, however, the results showed that 323 (36%) and 344 (38%) women reported depressive and anxiety symptoms, respectively.

Among the 323 women with depressive symptoms, 114 (35%) had mild symptoms, 153 (49%) had moderate symptoms, and 56 (17%) had severe symptoms. Similarly, among the 344 women who had anxiety symptoms, 168 (49%) had mild symptoms, 122 (36%) had moderate symptoms, and 54 (16%) had severe symptoms (see Figure 1).

Univariate regression analyses showed that several sociodemographic factors were positively associated with depression scores (Table 2). The number of children was associated with a 0.242 increase in depression score (β = 0.242, 95% CI: 0.125 to 0.358). Being single was associated with a higher depression score (β = 2.155, 95% CI: 0.638 to 3.670) compared to married participants.

Living in rented housing was associated with a 2.06 higher depression score (β = 2.056, 95% CI: 1.13 to 2.97), and living with relatives was associated with a 5.11 higher score (β = 5.11, 95% CI: 4.06 to 6.17). Higher household income, $100–$199 (β = 0.70, 95% CI: 0.05 to 1.35), $200–$299 (β = 6.61, 95% CI: 5.26 to 7.97), and ≥ $300 per month (β = 10.0, 95% CI: 9.35 to 10.65), was associated with higher depression scores.

Anxiety symptoms also showed a strong positive association with depression (β = 0.86, 95% CI: 0.818 to 0.917), whereas a higher sense of coherence was associated with lower depression scores (β = −0.259, 95% CI: −0.289 to −0.229). Surprisingly, having no formal education was associated with lower depression scores (β = −3.29, 95% CI: −4.34 to −2.24). Other variables, such as pregnancy status, MCH visits, and social support, showed negative but non-significant associations.

In the multivariate model, only living with relatives (β = 0.937, 95% CI: 0.294 to 1.758), higher household income ($200–$299 and ≥$300), and greater anxiety symptoms (β= 0.724, 95% CI: 0.648 to 0.799) remained significantly associated with higher depression scores. Sense of coherence retained a protective effect.

Univariate analyses for anxiety revealed a similar pattern (Table 3). Having more children (β = 0.202, 95% CI: 0.095 to 0.309) was associated with higher anxiety symptoms. Additionally, compared with participants living in their own homes, higher anxiety scores were observed among those renting (β = 2.023, 95% CI: 1.04 to 3.00), living with relatives (β = 4.67, 95% CI: 3.58–5.77), and living in IDP camps (β = 1.76, 95% CI: 0.56 to 2.96) were all associated with increased anxiety symptoms.

Higher income, $200–$299 (β = 6.19, 95% CI: 4.80 to 7.58) and ≥$300 per month (β = 10.18, 95% CI: 9.23 to 11.12), were respectively associated with increased anxiety symptoms. In contrast, a higher sense of coherence (β = −0.21, 95% CI: −0.244 to −0.184) and having no formal education (β = −2.816, 95% CI: −3.84 to −1.79) were associated with lower levels of anxiety symptoms.

After adjustment, living in refugee camps (β = 0.992, 95% CI: 0.123 to 1.86), higher income $200–$299 (β = 1.456, 95% CI: 0.704 to 2.207) and ≥$300 (β = 3.065, 95% CI: 2.206 to 3.924), and higher depression scores (β = 0.706, 95% CI: 0.660 to 0.753) remained positively associated with anxiety. Only the sense of coherence was negatively associated with anxiety (β = −0.031, 95% CI: −0.055 to −0.0062).

This study has several important strengths. It is among the few investigations to examine depressive and anxiety symptoms among Somali women in Mogadishu and draws on a relatively large sample distributed across nine districts, improving geographic representation. The use of standardized and previously validated measures adds methodological rigor, and the study contributes valuable evidence from a context where empirical mental health research remains limited.

Nonetheless, several limitations must be acknowledged. First, the study relied on self-reported questionnaires, which are inherently sensitive to recall issues, social desirability, and cultural nuances in symptom reporting. Because data collection took place in MCH facilities, some women may have intentionally overestimated their mental health concerns in hopes of accessing additional support, while others may have minimized symptoms due to mental health stigma. These opposing pressures complicate the interpretation of prevalence estimates.

Although the PHQ-9 and GAD-7 instruments have been used for Somali-speaking populations, the constructs they measure originate from Western psychiatric frameworks. Somali understandings of mental distress are often rooted in social, spiritual, or situational explanations rather than biomedical ones. Somalis have also been described as a people who use metaphor and poetry to convey emotional and social experiences. This rich expressive tradition may not translate seamlessly into the symptom-focused format of structured screening tools, potentially influencing how participants interpret and respond to individual items.

The use of convenience sampling in MCH clinics further limits generalizability, as women seeking maternal or child health services may differ from those who do not access such facilities. Additionally, the study was conducted during the COVID-19 pandemic, a period marked by heightened uncertainty and stress. It is therefore possible that the levels of depressive and anxiety symptoms reported here are elevated compared with what might have been observed prior to the pandemic. Moreover, women presenting to MCH clinics may already experience higher stress levels due to health concerns for themselves or their children, which could contribute to increased symptom reporting. Finally, given the study's cross-sectional design, causal interpretations cannot be made regarding the associations observed.

This study provides important insights into the mental health of Somali women living in Mogadishu, revealing that depressive and anxiety symptoms affect more than one-third of the population. These findings reflect broader patterns observed in conflict-affected settings and underscore the persistent mental burden facing women in environments marked by insecurity, displacement, and socioeconomic strain. Unlike prior studies highlighting factors such as marital status or social support, this study found limited associations with these variables after adjustment, suggesting that broader structural and contextual stressors may play a more influential role in shaping mental health outcomes.

Housing instability and displacement emerged as key factors. Living with relatives was associated with higher depression levels, while residing in IDP camps was linked with elevated anxiety symptoms. Higher income, unexpectedly, was also associated with increased symptoms of both depression and anxiety, emphasising the complexity of economic pressures in fragile settings. Additionally, while lower education is commonly identified as a risk factor for poor mental health, this study instead found an inverse association in some cases, indicating that educational attainment may influence mental health differently in this context.

Although a generally high level of sense of coherence (SOC) was observed, and SOC remained protective in all models, it did not prevent the notable levels of depressive and anxiety symptoms found. This suggests that internal coping resources alone are insufficient to offset the sustained external stressors faced by women in this post-conflict environment.

Taken together, these findings highlight the multifaceted nature of mental health challenges among Somali women and point to the need for context-sensitive, structural, and community-based interventions. Future research should explore additional determinants of mental health, including trauma exposure, household dynamics, and gender-specific stressors, and consider longitudinal designs to understand better how ongoing instability and systemic barriers shape mental health over time. Furthermore, given the cultural nuances in how mental distress is expressed and understood in Somali communities, future research should also prioritize culturally grounded assessment methods and qualitative approaches to complement standardized screening tools.

This study highlights a critical need for integrated mental health care for Somali women, particularly in conflict-affected settings such as Mogadishu. The high prevalence of moderate to severe depression and anxiety among the study population underscores the urgent necessity for accessible, culturally appropriate, and sustainable mental health services.

First, primary health care and maternal and child health (MCH) centres should serve as key entry points for identifying and supporting women with common mental disorders (CMDs). Training health care workers at these facilities in basic mental health care, including screening, psychoeducation, and brief interventions, can bridge the existing service gap and reduce stigma associated with seeking mental health support.

Second, special attention must be given to women in internally displaced persons (IDP) camps, who exhibited significantly higher anxiety symptoms. Mental health interventions targeting this population should account for the trauma, instability, and insecurity these women face. Community-based psychosocial programs that address trauma, promote safety, and build resilience are essential in these settings.

Finally, the Somali government must take concrete steps to integrate and sustainably finance mental health services into the broader health system. This includes training healthcare professionals, investing in mental health infrastructure, and reducing reliance on international aid by developing local funding mechanisms.