Adolescence has been described as a period of tremendous emotional upheaval and change [1–4]. The transition from childhood to adulthood involves major physical, psychological, cognitive and social transformations [5–8] which may be stressful to the adolescent. These transformational challenges are often associated with emotional turmoil including depression. Indeed a recent review of the mental health burden among children and adolescents world wide indicate that 10–20 % of them in the general population will suffer from at least one mental disorder in a given year [9]. The commonest of these mental health problems is unipolar depressive disorder which has been reported to be associated with a myriad of complications including impaired academic and social functioning and accounting for 40·5 % of disability adjusted life years (DALYs) caused by mental and substance use disorders [10], risky behaviours [11] as well as increased mortality rates through suicide [12].

Considerable literature points to the high prevalence of depression amongst adolescents [13–15]. School based studies of adolescent depression have reported various mean scores ranging between 2.6 and 3.6 % [16–18]. The variation in rates has been attributed to the great diversity in research instruments and methodologies.

The majority of studies documenting adolescent mental problems such as depression are from developed countries. The few studies conducted in sub Saharan African countries that have documented adolescent depression rates indicate estimates of 15.3–37 % among Egyptian students [19, 20] 6.9–23.8 % among Nigerian student populations [21]. In these studies depression has been associated with female gender, alcohol use, poor family functioning, large family size [21], childhood adversities such as emotional neglect [22] and frequent health services use.

Prior studies in Uganda have focused on mental health problems of adolescents in highly vulnerable and marginalised populations such as war traumatised individuals [23] and persons living with human immune deficiency virus (HIV) infection [24]. Further, studies on mental health issues among secondary school students in Uganda have mostly focused on alcohol and substance use problems. In the present study, we use data from four secondary schools to explore the prevalence of depressive symptoms in school-going adolescents. We sought to answer the following questions: What is the prevalence of depressive symptoms in school-going adolescents aged 13–16 years in central Uganda? And to what extent are socio-demographic factors, alcohol/substance use, chronic physical illness, chronic medication use and orphan hood associated with depressive symptoms in this age range?

This study contributes to the research literature on prevalence and factors associated with depression symptoms among school-going adolescents in sub-Saharan Africa. The prevalence estimate of depression symptoms in this study of 21 % is high and is likely to impair the adolescents’ ability to achieve academically and other areas of functioning. The prevalence of 21 % falls within the range of prevalence estimates obtained from previous studies conducted in both developing [14, 19] and developed countries that used depression screening instruments [33–35]. Likewise the prevalence rate of depressive disorder of 2.9 % that we found in this study is similar to what has been reported in studies conducted in the United States where a formal diagnosis of depression has been made among study samples of adolescents [36]. In this study, Kessler and colleagues analyzed data from 10,123 school-going adolescents in the age range of 13–17 years and found a prevalence rate of depressive disorder of 2.6 %. The high rates of depressive symptoms may also be due to general psychosocial distress resulting from general hardships in living, school related stress and poverty while the low rates of Major depressive disorder could be explained by the factors that promote resilience. In our study the research participants were secondary school students, and some of them could have come from high social economic class which has been found to be protective against depressive illness. Indeed Klassen et al. in their study on resilience in former Ugandan child soldiers, found that 27.6 % showed posttraumatic resilience as indicated by the absence of posttraumatic stress disorder, depression as well as clinically significant behavioural and emotional problems. This was attributed to better socioeconomic situation in the family, and more perceived spiritual support among other factors [37]. On the other hand, one would think that the low rates of depression (as measured by MINI KID) could have been a consequence of the selection bias as 35 students out of 109 students who had scored ≥19 points on the CDI were not interviewed. However these students may have left school for other reasons such as poverty, peer influence (Table 2).

In keeping with findings from previous studies, the prevalence of depressive symptoms was more than twice as common in girls as in boys. The excess of affected girls is seen in epidemiological as well as clinical samples, and is robust across different methods of assessment. Previous researchers have explained that sex differences in rates of depression are therefore unlikely to be merely due to differences in help-seeking or reporting of symptoms [38]. Although the reasons for this post-pubertal-onset sex difference are not fully understood, recent studies indicate that this difference is probably due to some combination of age-related changes in biological or social circumstances [39, 40].

The significant association between psychosocial stressors such as being a double orphan, living in a child headed household, and the presence of significant depressive symptoms is not surprising as such stressors have been reported to be significantly associated with adolescent depression and suicidality [41]. In South Africa, Cluver and colleagues found that acquired immuno-deficiency syndrome (AIDS)—orphaned children showed higher depression, anxiety, and post-traumatic stress disorder (PTSD) scores when compared with other-orphans and non-orphans [42]. El-Missiry and colleagues, studied depression in adolescent girls in Egypt using the CDI and found that presence of significant depression symptoms was associated with psychosocial stressors such as, quarrelsome family atmosphere, socioeconomic status, and negative life events [19].

The association between alcohol and drug use and the presence of depressive symptoms in this study is consistent with findings from previous studies [43, 44]. We noted a trend towards greater likelihood of alcohol and drug use in participants with significant depression symptoms than in those without. However, as our data are of a cross-sectional nature, it is not possible to make any inferences about whether the depression symptoms led to alcohol use or vice versa. Thus, longitudinal studies are needed to address this issue.

Consistent with findings from a systematic review of 340 studies investigating the relationship between depressive symptoms in children and adolescents with chronic physical illness [45], the adolescents who reported the presence of a chronic physical illness were more likely to have significant depression symptoms than those who did not report such an illness. Previous researchers have explained that the myriad of complex challenges associated with chronic disease conditions may interfere with regular school attendance [46–48], lead to peer rejection which may have detrimental effects on their self-concept [49, 50] and may result in inappropriate parental attitudes and behaviors, which may impair psychological well-being [51].

This study has limitations. First, as the study sample consisted of school-going adolescent in one district we cannot generalize our findings other districts elsewhere in Uganda or other sub-Saharan developing countries. Second, this study did not assess for parental factors and other factors such as coping styles or social support all of which have been associated with adolescent depression in previous studies. Third, the absence of collateral information may maximize effects of recall bias. Fourth, information was collected on exposures and outcomes simultaneously, thus causal relationships are difficult to establish. Fifth, the study did not include those who left school for a variety of reasons yet those who left school could have done so for reasons of depression. Indeed 35 students out of 109 students who had scored ≥19 points on the CDI were not interviewed with the MINI KID as they had left school and this could have affected the prevalence rates. Consequently, this study will only give clues as to whether certain factors may or may not be potential etiological factors of depression symptoms in school-going adolescents in central Uganda. Therefore, studies with better epidemiological design such as the case–control study can be used to investigate risk factor for depression in school-going adolescents.

Despite these limitations, this study, to our knowledge, provides the first prevalence estimates of depression symptoms among a sample of school going-adolescents in a non-conflict region in Uganda. Our study has important implications for school health programs in particular the integration of mental health issues into the school health education and health services. First, school health programs need to embrace locally adapted simple tools to measure depression which will enable us to distinguish depressive symptoms from clinical syndromes of depression because management strategies are different. Second, there is a need to offer stress management programs in which stressful situations among adolescent can be addressed before they affect emotional well-being, this research provides an important first step into current understanding of depression among school-going adolescents, which could be useful in designing school interventions for depression. Thirdly, mental health education for all stakeholders in the education sector must be scaled up to enhance early diagnosis and early interventions.

Significant depression symptoms are highly prevalent among this sample of school-going adolescents and may progress to full-blown depressive disorders. Integration of culturally sensitive psychological interventions to prevent and treat depression among school-going adolescents is desperately needed. There is great need for a child and adolescent mental health policy that will be used to plan for mental health services in schools.