The global impact of mental, neurological and substance use disorders (MNS) stand at 10% of the general population, but that percentage jumps to more than double at 22% in countries affected by humanitarian crises (8, 9). The majority (75%) of those affected by MNS are in low- and middle-income countries, especially in conflict and post-conflict settings, and lack access to effective mental health services (9, 10). With the high rates of mental health conditions in humanitarian settings, the current climatic crisis compounded by COVID-19 pandemic has only made the mental health situation worse (11). The ongoing climatic shocks has had profound impact on the Somali nation as a large section of the population maintain century old pastoralism lifestyle which is increasingly been threated by recurrent drought. According to UN estimates, over 6 million people been affected by the current drought (12).

The Eastern Mediterranean Region of WHO (EMRO) comprising of 21 countries across Horn of Africa (Somalia and Djibouti), North Africa, Middle East, South Asia (Pakistan and Afghanistan) with a population of nearly 700 million host half of the over 80 million displaced population in the world (11).

In Somalia, the protracted conflict, social unrest and ensued political instability since the fall of the central government in 1991 has left the country in a complex humanitarian situation. The profound impact of political instability, conflict, insecurity, coupled with resultant economic stagnation and poverty, has been a driving force behind the increased rates of mental illness and psychosocial distress across all segments of society, with vulnerable groups such as women and youth most affected (1, 13, 14).

A systematic review of mental health in Somalia (1) estimated that a third or more of the population experienced and or experiencing some form of mental illness. The review also identified that many of the psychosocial problems faced by the people stem from the effects of the protracted conflict, long-term displacements, economic turmoil and social deprivations. While three decades of unstable governance and the collapse of public institutions have worsened access to livelihoods, education, employment and healthcare services. To date, parts of Somalia especially the South-Central region continues to experience significant insecurity and violence. Thus, the continuous conflict and violence, exacerbated by recurring climatic shocks, have deeply fragmented the social fabric of the society and eroded people’s coping mechanisms, community resilience and contributing to the severe psychological distress (1, 6, 15).

Studies have shown significant mental health issues associated with conflict and refugee experiences in diverse settings (1, 15–17). The same is true for Somalis (18–20). Psychological experiences associated with living in conflict and crisis have been classified in the biomedical discourse under the categories of post-traumatic stress disorder (PTSD), anxiety, and depression. Somalis also acknowledge feelings of hopelessness, despair, anxiety, and anguish as part of their symptomatology, despite not explicitly labeling such experiences and symptoms as PTSD or depression (18). The after-effects of trauma among Somalis have also been described in somatic terms, with emphasis on headaches and other unexplained body pains (18, 21).

Despite this, few studies on Somali mental health exist, especially coming out of Somalia. The few studies that have been conducted bear grim results; for instance, a situational analysis by World Health Organization (WHO) in 2010 estimated that one in every three Somalis suffers from some form of mental illness, thus making Somalia one of the countries with the highest rate of mental illness in low-income countries (6). As such, mental illness is a significant health concern in Somalia, a country that has faced decades of political instability, civil unrest, and extreme violence. The prolonged conflict and violence in Somalia have exposed many people to trauma and stress, which are known risk factors for mental illness. Studies showed high exposure to traumatic events, displacement, and significant rates of mental health disorders (1, 6, 13).

Similarly, recent studies (1, 13, 17) showed significant increase in substance use, suicide, anxiety and depression among the Somali population. At the same time highlighting the near total absence of mental health services across the country. The country’s mental health services are close to non-existent with a strikingly low ratio of just 0.5 psychiatric beds per 100,000 population. In contrast, the WHO Eastern Mediterranean Region maintains a ratio of 6.4 beds/100000, and globally, the figure stands at 24 beds/100000 (22). Apart from a handful of inadequately staffed and under-resourced psychiatric hospitals, the country has no community-based mental health services (1, 6, 13). This dire situation leaves the population without adequate support and exacerbates the challenges faced by those suffering from mental health issues.

Equally impacted by the civil war is the loss of essential human resources in health care as most either fled the country or were victims of the war (1, 6). Hence the country is also facing a serious situation of acute and chronic shortage of health professionals. Such a shortage is more profound in the field of mental health where only a handful of professionals provide services across the country (1, 6).

Cultural beliefs and practices can also impact the perception and treatment of mental illness in Somalia. For example, there is widespread belief that mental illness as a result of the evil eye, jinn, or other supernatural forces, leading to stigmatization and a reluctance to seek professional help (1, 14). This drives a significant majority to seek traditional and alternative healing services or simply have those with mental illness abandoned or chained at home (1, 6, 14).

Research on mental illness in Somalia is limited, due in part to the ongoing conflict and access constraints, lack of resources, and stigma around mental health. One of the major challenges in studying mental illness in Somalia is the lack of reliable data, due to limited number of public and private health care facilities offering mental health services, and limited surveys, including data that should be routinely collected by the country’s health information system (1). However, a small number of studies conducted provides some insight into the nature of mental illness in Somalia. Most of these studies relied on small sample sizes, or were qualitative in nature, which limited the generalizability of their findings (6, 13, 23, 24).

Although there have been efforts to study and understand the mental health needs for the population in Somalia, there are no population wide studies to provide evidence on the mental health status of the population. This study bridges that gap by undertaking a study of a large size sample using a cross-sectional survey in South-Central Somalia, a first for the country.

The aim of this study was to provide population-based epidemiological data on mental health conditions. The specific objectives were:

This study was carried out in the three districts mentioned above (Baidoa, Dolow and Kismayo). Baidoa is located approximately 240 kilometers west of Mogadishu. It is characterized by rapid population growth, unplanned urbanization, influx of political refugees, long-standing history of drought and famine.

Kismayo is approximately 500 km southwest of Mogadishu. Following the collapse of central government that accompanied the civil war in 1991, various local militias fought for control of the city. The strategic importance of the port, the land fertility of the Juba region for agricultural cultivation and inter-clan conflict contributed to decades of decades of displacement and distress.

Dolow town, a border town of Somalia and Ethiopia that is located approximately 470 km northwest of Mogadishu. It has often suffered from strained relations between Ethiopia and Somalia, making its residents displaced multiple times. In the last three decades, the town further suffered the predicament of rival militias fighting over territories and invasions by the Ethiopian army (25, 26).

Common mental disorders (CMD) were considered as dependent variables while independent variables included socio-demographic factors such as (age, marital status, education level, family size, and occupation); experience of traumatic events (hunger, combat and witness of killing, lack of shelter and food), and substance use (alcohol, tobacco products, khat and sedatives).

The data was collected through face-to-face interviews using demographic questionnaire and pretested Somali version of Mini International Neuropsychiatric Interview (MINI) version 5.0 to generate sociodemographic, mental disorders, and substance related data. MINI is a short diagnostic tool developed for the assessment of ICD10 (psychiatric disorders). It can be administered in approximately 35–45 min by trained interviewers. The MINI 5.0 comprises of structured psychiatric assessment for all common mental disorders, severe mental health disorders, substance abuse, and suicide risk. This study used the English version of the MINI 5.0 and adapted it to the Somali context. Three authors of this paper, a mental health specialist, a psychologist, and a public health specialist, who are all of Somali origin, collaboratively adapted the English MINI 5.0 to Somali language and cultural context. The MINI questionnaire was then coded in Kobo Collect Software (KoboToolbox, Global) for use on Android operating systems (Google Corporation, Mountain View, CA, USA) by the study team including the research assistants and enumerators. After coding, the research team re-reviewed the Somali translated and adopted MINI 5.0 with focus on understandability, simplicity, and logical sequence of the questions. Eight research assistants (with bachelor in medicine or nursing) were recruited for this study. The research assistants attended a 2-day course on the correct use and administering of the MINI 5.0, in which the research assistants practiced administering and being screened with the MINI 5.0 among themselves.

Further, 60 enumerators (20 for each site) with a minimum bachelor’s degree were recruited for data collection. A two-day workshop was held for all the enumerators by the by the first author and the research assistants to ensure that the enumerators understand the administering of the instruments to the study participants. Enumerators received one-on-one supervision by the research assistants to ensure proper administration of the tool. Next, a pilot study of the MINI 5.0 was conducted, in which the questionnaire was tested for 2 days among the study participants with the supervision of the first author and research team. Enumerators who administered the MINI 5.0 in the piloting were asked for their thoughts on each question, such as, whether the question was obscure or inappropriate or difficult to understand and if so, how they suggested to improve each issue. Following the pilot implementation, the study team improved the MINI 5.0 based on the enumerators’ feedback. Eligible participants were referred to a trained enumerator, who administered the MINI 5.0 and the sociodemographic questionnaire.

The collected data was checked for completeness and consistency and exported into SPSS version 23 for analysis. Summary of descriptive statistics were calculated to explain the socio-demographic characteristics, experience of trauma events, substance related factors and CMD. To determine factors associated with common mental disorders, first bi-variable logistic regression was performed. Subsequently, significant variables in the bi-variable analysis (value of p <0.05) were further incorporated into the multi-variable logistic regression. Statistical significance was declared at p < 0.05 and the corresponding 95% confidence interval. Chi-square tests were used for categorical data to analyze the differences in levels of mental disorder proportions across socio-demographic characteristics.

Table 1 summarizes sociodemographic information collected from the study’s respondents in terms of their residence, age groups, and household characteristics among other variables. The demographic details represent both the non-displaced populations and internally displaced persons (IDPs). Of importance is that most participants were below the age of 34 represented 62.9% of the total study population of 713 while 37% were over 35 years.

In respect to levels of education, we looked at various categories as shown in Table 1, and our findings show a low level of formal education, i.e., 39% of the respondents stated that they could not read or write even in their own language while 41% have some form of Islamic education from attending local Quranic schools known as Dugsi in the Somali language. Dugsi typically involves writing and memorizing Qur’an, the Islamic holy book. Whereas 17.5% attended elementary or high school, and only 2% had acquired college or university education.

We looked further at marital status of the participants and the findings showed 50% were married while 30% reported being single and the remaining classified as either widowed (6%) or divorced (13%). In addition, the majority 78% of the respondents came from a family with five children or more. Overall, significant group differences (p < 0.05) were observed in all the variables in relation to the study site; i.e., non-displaced, displaced communities, family size, and age as well as education levels.

We assessed mental health disorders using the various ICD 10 categories. (Table 2) summarizes the prevalence of the most common mental disorders as identified. Overall panic disorder, generalized anxiety disorders, post-traumatic and stress related conditions, and mood disorders appeared to be the most prevalent conditions. Importantly, panic disorders (39%), generalized anxiety disorders (35%), major depressive disorders (32.1%), and PTSD (30%) are the four most common complaints among the study population, while overall, a staggering 78.1% of the participants endorsed experiencing some form of mental or psychological conditions.

Results in Table 3A show sociodemographic variables (age, sex, marital status) and (Table 3B) (occupation, residence, site). In our analysis, there was an observed statistically significant results based on age, sex, and marital status (p < 0.05) in the presence of manic disorder. Sex and age had a statistically significant (p < 0.05) difference with the presence of hypomania, manic episodes, panic disorder with agoraphobia (current), mood disorder with psychotic features (current), psychotic disorder (current), and psychotic disorder (lifetime). Age and marital status showed a significant difference (p < 0.05) with depressive episodes (recurrent). In addition, age also appeared statistically significance difference (p < 0.05) with mood disorder with psychotic features (lifetime).

Marital status is a moderating variable (p < 0.05) in panic disorder without agoraphobia (current), PTSD (current), and generalized anxiety disorder (current). There was no observable significant difference by age, sex, and marital status in major depressive episode (current), dysthymia (current), and suicide risk.

In Table 3B, largely, significant differences by site and occupation (p < 0.05) were observed in three disorders, namely, panic, post-traumatic stress, and psychotic disorder. Psychotic disorder (lifetime) was statistically significant different with occupation at p < 0.05. There were significant site differences (p < 0.05) in most disorders among participants who were from Dolow and Kismayo with lower proportion of mental disorders compared to those from Baidoa.

In this study, the overall prevalence of common mental disorders (CMD) was found to be 78.1%. It was higher in males 335 (47%) than females 222 (31%). In bivariable analysis, being male, aged between 15–24 or 25–34 years, unemployed, having family size above or equal 10 members and experienced starvation were associated with CMD. According to the multivariable logistic regression analysis, variable with value of p of less than 0.5 were considered as statistically significant associated with CMD. Accordingly, sex, marital status, occupation, family size, experience of ill-health without access to medical care, experiences of combat situation, bodily harm, starvation, use of sedatives or sleeping pills, khat, and tobacco products had significant associations with CMD. The odds of experiencing CMD for unemployed participants were 1.90 times higher (AOR = 1.90, 95% CI = 1.18–3.06) than employed. Those who had family size of equal or more than 10 members were 3.3 times more likely to develop CMD (AOR = 3.3, 95% CI = 1.90–5.72) compared with those who had family size of 1–4 members. The odds of having CMD among study participants who experienced starvation was 3.46 times (AOR = 3.46, 95% CI = 2.23–5.37) higher as compared to their counterparts. Participants who use khat were 5.87 times more likely to have CMD (AOR = 5.87, 955 CI, 1.75–19.65) than participants who did not use. Participants who were physically harmed or beaten were 95% more likely to have CMD (AOR = 1.95, 95% CI = 1.23–3.09) compared to those who did not experience the same (Table 4).

In this table, we looked at the co-morbidity between various mental health conditions ie anxiety, depression and PTSD and substance use. More than one third of the study participants (42.2%) experienced starvation. Similarly, one third of the participants (35.3%) reported that they had been exposed to physical bodily harm or beatings. Of the total of 713 participants, 277 (38.8%) reported using tobacco products, and 265 (37.2%) reported using sedatives or sleeping pills. In addition, most of the study participants (81.9%) reported that they have experienced ill health without access to medical care (see Table 5).