In this study in a poor rural community in southwest Uganda, the prevalence of probable SMD was 0.9%, similar to the 0.9% prevalence of moderate to serious mental disorders reported in Nigeria [14
], but much lower than the 5% prevalence reported in urban Ethiopia using a precursor of the screening tool used in this study [11
This study provides important new information on the community prevalence of SMD in a community rural Africa. There are two main limitations. Firstly, the use of one screening question for SMD runs the risk of assessing for a highly non-specific entity. Secondly, this question item has never been previously validated in this study population. The formation of the composite screening item was largely driven by the following considerations: the need for a question item that could be used by lay interviews; and the need to limit the number of questionnaire items in the survey. This question was however derived from an established WHO questionnaire (WHO-SRQ-25 tool) which has been used extensively in sub-Saharan Africa [10
Secondly, the results of this study show that this question had criterion validity. There is however a need to validate this question item which may have utility as screening tool for severe mental distress in the context of large population-based studies where: non-mental health professionals are used in data collection; where mental health may be competing with other health disciplines for space on study questionnaires. Secondly, because this was a cross-sectional study, it was not possible to tell the direction of association between SMD and the investigated factors.
The socio-demographic factors found to be associated with SMD in this study were increasing age and low socio-economic status. A study in Ethiopia reported a significantly increased risk of mental distress with increasing age and with indices of low socio-economic status [11
]. Two explanations have been offered for the association between the severe mental illness of schizophrenia and low socio-economic status. In the social causation theory, the socio-environmental factors associated with low socio-economic status (including more life events stressors, increased exposure to environmental and occupational hazards and infectious agents, poorer prenatal care and fewer support resources if stress does occur) are a cause of schizophrenia [26
]. The social selection, or drift, theory is that socio-economic status is a consequence of the disorder -the insidious onset of schizophrenia is believed to preclude elevating one's status or to cause a downward drift in status [26
On the behavioural factors, being a current smoker, having no or multiple sexual partners and having ever consulted a traditional healer were associated with SMD. After adjusting for age and sex there was a twofold increased odds of SMD among current regular cigarette smokers as compared to those who were not regular smokers. Both Lasser and colleagues (2010) and van Os and Kapur (2009) have reported higher rates of cigarette smoking among persons with mental illness as compared to population controls [27
]. It has been suggested that patients with the severe mental illness of schizophrenia use nicotine to help reduce cognitive deficits, negative symptoms or the neuroleptic side effects [27
]. The observed association between SMD and having multiple sexual partners in this study confirms what has previously been reported by other authors [29
]. The association between SMD and high risk sexual behavior has been attributed to factors associated with SMD such as cognitive processing difficulties, lack of planning, and poor social skills which place these patients at risk [29
]. The association between SMD and no sexual partners may reflect the severe social dysfunction associated with the severer end of the spectrum of SMD. The association between SMD and having previous contact with a traditional healer can be regarded as a form of health-seeking behavior for mental illness, a health seeking behavior that is in agreement with the predominant spiritual explanatory model for mental illness [4
]. As has been reported before in low income settings, only three patients (5.7%) with SMD were receiving formal health care for their problem in this study [14
On clinical factors, self-reported epilepsy was the only factor significantly associated with SMD. The strong association between self-reported epilepsy and SDM could be explained in two ways. Firstly, the confusional state commonly associated with the post-ictal phase of generalized seizures may lead to behavioural disturbances and hence epilepsy could be regarded as a cause of SMD. Secondly, the community may not have been able to adequately differentiate between epilepsy and SMD because of the possible overlap between SMD and epilepsy.