Generating evidence to narrow the treatment gap for mental disorders in sub-Saharan Africa: rationale, overview and methods of AFFIRM.

There is limited evidence on the acceptability, feasibility and cost-effectiveness of task-sharing interventions to narrow the treatment gap for mental disorders in sub-Saharan Africa. The purpose of this article is to describe the rationale, aims and methods of the Africa Focus on Intervention Research for Mental health (AFFIRM) collaborative research hub. AFFIRM is investigating strategies for narrowing the treatment gap for mental disorders in sub-Saharan Africa in four areas. First, it is assessing the feasibility, acceptability and cost-effectiveness of task-sharing interventions by conducting randomised controlled trials in Ethiopia and South Africa. The AFFIRM Task-sharing for the Care of Severe mental disorders (TaSCS) trial in Ethiopia aims to determine the acceptability, affordability, effectiveness and sustainability of mental health care for people with severe mental disorder delivered by trained and supervised non-specialist, primary health care workers compared with an existing psychiatric nurse-led service. The AFFIRM trial in South Africa aims to determine the cost-effectiveness of a task-sharing counselling intervention for maternal depression, delivered by non-specialist community health workers, and to examine factors influencing the implementation of the intervention and future scale up. Second, AFFIRM is building individual and institutional capacity for intervention research in sub-Saharan Africa by providing fellowship and mentorship programmes for candidates in Ethiopia, Ghana, Malawi, Uganda and Zimbabwe. Each year five Fellowships are awarded (one to each country) to attend the MPhil in Public Mental Health, a joint postgraduate programme at the University of Cape Town and Stellenbosch University. AFFIRM also offers short courses in intervention research, and supports PhD students attached to the trials in Ethiopia and South Africa. Third, AFFIRM is collaborating with other regional National Institute of Mental Health funded hubs in Latin America, sub-Saharan Africa and south Asia, by designing and executing shared research projects related to task-sharing and narrowing the treatment gap. Finally, it is establishing a network of collaboration between researchers, non-governmental organisations and government agencies that facilitates the translation of research knowledge into policy and practice. This article describes the developmental process of this multi-site approach, and provides a narrative of challenges and opportunities that have arisen during the early phases. Crucial to the long-term sustainability of this work is the nurturing and sustaining of partnerships between African mental health researchers, policy makers, practitioners and international collaborators.

Clinical outcome in bipolar disorder in a community-based follow-up study in Butajira, Ethiopia.

OBJECTIVE: To determine the clinical outcome of bipolar disorder in a developing country setup. METHOD: After assessing 68 378 individuals, aged 15-49 years, in a double-sampling design in a rural community in Ethiopia, 312 patients with bipolar disorder were prospectively monitored with symptom rating scales and clinically for an average of 2.5 years. RESULTS: Overall, 65.9% of the cohort experienced a relapse--47.8% manic, 44.3% depressive and 7.7% mixed episodes--and 31.1% had persistent illness. Female gender predicted depressive relapse, while male gender predicted manic relapse. Being on psychotropic medication was associated with remission. CONCLUSION: This large community-based study confirms the relapsing nature of bipolar disorder and a tendency for chronicity. This may be partly because of lack of appropriate interventions in this setting; however, it may also indicate the underlying severity of the disorder irrespective of setting.

Short-term symptomatic and functional outcomes of schizophrenia in Butajira, Ethiopia.

BACKGROUND: Prospective outcome studies based on a community sample of mostly neuroleptic naive cases of schizophrenia are uncommon. OBJECTIVES: To describe short-term symptomatic and functional outcomes of schizophrenia, and potential predictors of outcome. METHODS: After a baseline assessment, 63 incident and 208 prevalent cases of schizophrenia were followed by a yearly clinical assessment for an average of 2.5 (range 1-4) years. Scores of negative symptoms and positive symptoms were used as indicators of symptomatic outcomes. SF-36 scores of physical and social functioning, and role limitation due to mental health problems were used as indicators of functional outcomes. Several variables were evaluated as potential predictors of outcome in random coefficient models. RESULTS: Functioning and other measures of health related quality of life were significantly diminished in cases as compared to the general population of the area at baseline and follow up. Of the socio-demographic and clinical factors evaluated, only lower negative and positive symptom scores were significantly associated with improvements in functioning. The level of functioning observed in cases from Butajira was lower than that reported for cases from developed countries. CONCLUSIONS: Our findings are not in accord with other outcome studies that have reported better functional outcome for cases of schizophrenia from developing countries.

Socio-demographic correlates of bipolar disorder in Butajira, rural Ethiopia.

OBJECTIVES: To describe the socio-demographic correlates of bipolar disorder and their interactions. DESIGN: Cross sectional study. SETTING: Rural population in Ethiopia. SUBJECTS: We conducted a door-to-door survey of a predominantly rural population of close to 70,000 individuals. A total of 315 cases of bipolar disorder were identified by use of the SCAN and clinical assessment. RESULTS: The lifetime prevalence of biopolar disorder in this population was 0.5% (0.6% for men and 0.4%) for women. Adjusted odds ratios show that males had 80% higher risk of bipolar disoder: OR (95% CI) = 1.81 (1.42, 2.32). Those aged 25-34 years had 45% higher risk than those aged under 25: adjusted OR (95% CI) = 1.45 (1.03, 2.06). Area of residence and educational level were not associated with the disorder. The association of marital status with bipolar disorder was modified by age and sex. Among males the odds of bipolar disorder among the married was 3.6 times higher than those who have never married. Among the age group of 15 to 24 years, those married had a 84% higher risk of disease, while those previously married had a 55% increase. On the other hand the association between marital status and bipolar disorder is reversed in older age groups, with those who have never married having a higher risk compared to those married. CONCLUSION: The study shows that the prevalence of bipolar disorder in this population is within ranges of prevalence reported in the literature, although on the lower side. It also shows that age, sex and marital status are associated with bipolar disorder and these variables interacts with each other.

Prevalence and clinical characteristics of bipolar I disorder in Butajira, Ethiopia: a community-based study.

BACKGROUND: Bipolar disorders have been extensively studied in the high-income countries but community-based studies from low-income countries are very rare. The main objectives of the current study are to estimate the lifetime prevalence of bipolar I disorder in the general population of the Butajira district in Ethiopia and to characterize the onset and course of the disorder in a predominantly treatment naïve population. METHOD: Cases were identified by a door-to-door screening of the district's entire adult population aged 15 to 49 years (N=83,387), where 68,378 were successfully screened. CIDI and key informant method were used in the first stage of screening followed by confirmatory SCAN interviews. RESULTS: Three hundred fifteen cases were identified and complete information could be collected for 295 individuals. Of these, 55.3% were males, 83.1% were from a rural area, and 70.2% were illiterate. Lifetime prevalence of bipolar I disorder was estimated to be 0.6% for males and 0.3% for females. The mean age of cases was 29.5 years, with no significant sex difference. The mean age of first recognition of illness was 22.0 years; for men 22.3 years and for women 21.2 years. The mean age at onset of manic phase of the illness was found to be 22.0 years (22.5 for men and 21.4 for women). The mean age at onset of depressive phase was 23.4 years (24.1 for men and 22.5 for women). There was no significant sex difference in the age of onset of manic or depressive phases. In 22.7% of the cases bipolar I illness started with a depressive episode and in 77.3% of the cases it started with a manic episode. Two or more episodes of the illness were reported by 64.1%. Over half of the study subjects (55.9%) had never sought any help from modern healthcare sector, and only 13.2% had ever been admitted to psychiatric hospital. During the survey 7.1% of the cases were undergoing treatment. A previous suicide attempt was reported by 8.1% of the males and 5.4% of the females. CONCLUSION: The overall lifetime prevalence and age of onset are within the range of findings from other studies in Western countries. In contrast to most previous studies, prevalence of the disorder among females was half of that among males. Our finding that prevalence of this disorder among males and females appeared to be different from many other studies warrants further research.

Onset and clinical course of schizophrenia in Butajira-Ethiopia--a community-based study.

BACKGROUND: There are reports on favourable course and outcome of schizophrenia in lowincome countries. The aim of the present study was to examine onset and clinical course of the illness in a community-based sample in rural Ethiopia based on crosssectional information. METHOD: A two-stage survey was carried out in Butajira-Ethiopia, a predominantly rural district. Altogether 68,378 individuals aged 15-49 years were CIDI-interviewed, of whom 2,159 were identified as cases according to the CIDI interview with regard to psychotic or affective disorders. Key informants identified another group of 719 individuals as being probable cases and a total of 2,285 individuals were SCAN-interviewed. The present paper reports on cases with schizophrenia. RESULTS: There were 321 cases of schizophrenia giving an estimated lifetime prevalence of 4.7/1,000). Of the cases,83.2% (N = 267) were males. Mean age of first onset of psychotic symptoms for males was 23.8 (sd 8.6) compared to 21.0 (sd 7.8) for females (P = 0.037; 95 %CI 0.16-5.47). Over 80% had negative symptoms and over 67% reported continuous course of the illness. Less than 10% had a history of previous treatment with neuroleptic medication. About 7% were vagrants, 9 % had a history of assaultive behaviour,and 3.8% had attempted suicide. The male to female ratio was nearly 5:1. CONCLUSION: This large community-based study differs from most previous studies in terms of higher male to female ratio, earlier age of onset in females and the predominance of negative symptoms.

Schizophrenia: illness impact on family members in a traditional society--rural Ethiopia.

BACKGROUND: Studies have consistently shown that both the subjective and objective dimensions of burden among family members of schizophrenia patients and other psychiatric disorders are prevalent. However, as most of these reports were from western societies, we lack information on the subject in developing countries. METHOD: The study was conducted within the framework of the ongoing epidemiological study of course and outcome of schizophrenia and bipolar disorders in a rural population of 15-49 years of age. Three hundred and one cases of schizophrenia and their close relatives participated in the study. RESULTS: Family burden is a common problem of relatives of cases with schizophrenia. Financial difficulty is the most frequently endorsed problem among the family burden domains (74.4 %). Relatives of female cases suffered significantly higher social burden (Z = 2.103; p = 0.036). Work (Z = 2.180; p = 0.029) and financial (Z = 2.088; p = 0.037) burdens affected female relatives more often than males. Disorganised symptoms were the most important factors affecting the family members in all family burden domains. Prayer was found to be the most frequently used coping strategy in work burden (adj. OR = 1.99; 95 % CI = 1.08-3.67; p = 0.026). CONCLUSION: Negative impact of schizophrenia on family members is substantial even in traditional societies such as those in Ethiopia where family network is strong and important. The scarce existing services in the developing countries should include family interventions and support at least in the form of educating the family members about the nature of schizophrenia illness and dealing with its stigma and family burden.

Socio-demographic correlates of depressive disorder in Butajira, rural Ethiopia.

OBJECTIVES: To describe the socio-demographic correlates of depressive disorder and their interactions. DESIGN: Cross sectional study. SETTING: Rural population in Ethiopia. SUBJECTS: Two hundred and twelve cases of depressive disorders. MAIN OUTCOME MEASURE: Clinically confirmed depressive disorder. METHODS: We conducted a door-to-door survey of a predominantly rural population of close to 70,000 individuals. Cases of depressive disorder were identified by use of the SCAN (Schedule for Clinical Assessment in Neuropsychiatry) and clinical assessment. RESULTS: Age was associated with the disorder: those aged 35 and above had over twice the risk of those aged 24 or below: adjusted odds ratio (95% Confidence interval) = 2.24 (1.38, 3.64). On aggregate, those previously married (separated/divorced/widowed) had an almost two-fold increased risk of the disorder compared to the singles [Adj. OR (95% CI) = 1.93 (1.00, 3.70)]. As a result of effect modification by age, being married was positively associated with depressive disorders [Adj. OR (95% CI) = 3.52 (1.96, 6.32)] among the younger age group (16 to 24 years), but this association was reversed in older (25 to 34 years) age groups [Adj. OR (95% CI) = 0.50 (0.29, 0.88)]. CONCLUSION: The study shows that age and marital status are associated with depressive disorders independently of each other and that age modifies the association of marital status with depressive disorder.

Neurological soft signs (NSS) in 200 treatment-naïve cases with schizophrenia: a community-based study in a rural setting.

Several studies have reported neurological soft signs (NSS) to be common in individuals with schizophrenia. The majority of these studies are based on clinical samples exposed to neuroleptic treatment. The present study reports on 200 treatment-naïve and community-identified cases of schizophrenia and 78 healthy individuals from the same area, evaluated using the Neurological Evaluation Scale (NES). The median NES score was 5.0 for cases of schizophrenia and 1.5 for healthy subjects. The impairment rate of NSS in cases with schizophrenia was 65.0% against 50.0% in healthy subjects, and the difference was statistically significant (chi2 = 5.30; df = 1; P < 0.021). NSS abnormality is as common in treatment-naïve cases as reported in many studies in those on neuroleptic medication. There was no significant relation between the NSS impairment and duration of illness, remission status, positive symptoms, negative symptoms and disorganized symptoms.

An evaluation of two screening methods to identify cases with schizophrenia and affective disorders in a community survey in rural Ethiopia.

BACKGROUND: Detecting cases with psychiatric disorders in the general population is costly and it is not clear which is the method of choice for community surveys in low-income countries. AIM: To compare the performance of a standardized diagnostic layman interview instrument--the Composite International Diagnostic Interview (CIDI 2.1) versus the Key Informant method in identifying cases with schizophrenia and major affective disorders in a community survey. METHOD: Both screening methods were tested against an expert interview--the Schedule for Clinical Assessment in Neuropsychiatry (SCAN 2.1) in a rural district in Ethiopia with 25,632 inhabitants. RESULT: CIDI identified 524 and key informants 192 individuals as probable cases who were invited for a further SCAN interview. Seventy-two individuals were identified by both methods. Of those identified as probable cases by either method, a total of 481 volunteered the SCAN interviews. The Key Informant method alone detected more cases of schizophrenia, 59 vs. 29 for CIDI, whereas CIDI alone detected more cases of affective disorders, 45 vs. 30. Key informants performed better in detecting chronic cases. CONCLUSION: For community surveys, which aim at identifying cases with major mental disorders in low-income countries like Ethiopia, the combined use of both CIDI and the Key Informant method is recommended.

Perception of stigma among family members of individuals with schizophrenia and major affective disorders in rural Ethiopia.

BACKGROUND: Many studies from the Western world have reported on stigmatisation of people with mental illnesses and its negative consequences, but few studies have addressed the issue in traditional rural societies. The present study aimed to estimate the extent and socio-demographic distribution of stigma as perceived by relatives of mentally ill individuals in rural Ethiopia. METHOD: A total of 178 relatives of individuals who were diagnosed as suffering from schizophrenia or major affective disorders in a community-based survey were interviewed using the Family Interview Schedule. RESULTS: About 75% of the respondents perceived that they were stigmatised or had experienced some sort of stigma due to the presence of mental illness in the family, 42% were worried about being treated differently and 37% wanted to conceal the fact that a relative was ill. Those from the older age group (45+) and urban residents were more likely to perceive stigma as a major problem, but otherwise differences were few between socio-demographic groups. The illness was attributed to supernatural forces by 27% and praying was suggested as a preferred method to deal with the problem by 65%. CONCLUSION: Stigma was found to be a common problem, with few differences between socio-demographic groups or between types of mental disorder. Beliefs about causes differ from those held by professionals. Popular beliefs and attitudes must be taken into account when planning for intervention.

Use of psychiatric rating instruments in Ethiopia.

It is quite fascinating to see the increasing use of psychiatric rating instruments in Ethiopia in the last 3 decades, almost parallel with the growing interest world-wide. A review of those instruments used in Ethiopia was made to evaluate their applicability and quality, and to recommend ways of ensuring standard use. In this paper, the effect of culture and translation are discussed and practical ways of using the instruments in field situation, and culturally sensitive and applicable ways of translation are endorsed. Their use in general medical settings, is also stressed.

Khat induced psychosis and its medico-legal implication: a case report.

It has been established that khat plant leaves (Catha edulis forsk) contain an active psychostimulant substance known as Cathinone that is similar in structure and pharmacological activity to amphetamine. There have been no previously reported cases of khat induced psychosis in Ethiopia despite heavy consumption of khat in most parts of the country. We report here a case of brief and episodic psychosis attributed to heavy khat chewing. We maintain that khat leaf is a substance of abuse and that Khat chewing has the potential to complicate psychiatric conditions and forensic events.