This study is a follow up of a cohort of patients with severe mental illness attending traditional healers. Follow up was conducted at both three and six months for several reasons. First, the three months follow up was important to reduce the risk of missing out participants whose symptoms could have remitted within this three months[11, 36]. Secondly to avoid high attrition rate and finally, a six months of a follow up period was considered valuable for continued assessment of patients whose symptoms had potential for remitting given a longer treatment period.
In light of the local context were patients ability to return for review are a function of various factors such as lack of transport, the follow up rate of 85.5% at three months and 76.6% at six months was satisfactory. There was no significant difference in key baseline socio-demographic or clinical variables between those who were successfully traced and those lost to follow up.
There was a general trend of reduction in symptom scales at 3 and 6 months follow up. The percentage reduction was greatest in the PANSS scale followed by YMRS, and MADRS scale. However, in terms of dichotomous caseness, patients with mania had most recovery followed by patients with psychotic depression and lastly patients with schizophrenia.
Over 80% of the subjects used traditional healing and biomedical services concurrently with the majority being patients with mania. The greatest concurrent use occurred in the first three months of follow up. At bivariate level of analysis, poor economic status, early age of onset, longer duration of symptoms, previous episodes, positive family history, current illness severity, co morbidity and combined use of biomedical services and traditional healing were significantly associated with outcome of psychosis. At multivariate level, only combined use of biomedical services and traditional healing and being in debt remained significant.
To our knowledge, there are hardly any controlled studies of outcome of treatment of patients with severe mental illness who use traditional healing practices. Contemporary biomedical models which most health workers subscribe to, equate health and sickness with normal and abnormal biological functioning respectively. This approach assumes that symptoms can be adequately characterized in terms of measurable changes in the basic biological processes in the human body and reduces all health or illnesses phenomena to these processes. By extension, conventional western medicine argues that human mind can be described in terms of neurobiology. Biomedical psychiatry has endorsed this model and the corollary that normal pathology states of the mind are reduced to basic neuropsychological or neuroanatomical processes. In this broad context, it has been argued that the claims of traditional healing practices are of capricious and lacks scientific rigor, diagnostic accuracy and that outcome measures are not clear or objective. Lack of control measures in those studies is stated as another drawback.
In this study, we attempted to overcome these problems. The patients were carefully diagnosed according to DSM IV diagnostic criteria by using the MINI plus. In addition to use of symptom outcome measures, we used CGI and GAF as objective measures of impairment and the study was prospective. The principal difficulty and limitation in this study, however, was that we could not control the subjects. At the time of the study, no sensible or ethical solution to the problem of control of the subjects was evident to us. Traditional healing practices are not officially accepted as a form of treatment for severe mental illnesses although 4 in 5 Ugandans visit traditional healers. On the other hand, there is concern that controlled trials may not capture the full richness and diversity of traditional healing practices and that randomisation may undermine the representation of the therapeutic encounter. Attempts to compare those patients who go to traditional healing practices with those who go for biomedical services were limited by the design of the study. On the other hand, since severe mental illnesses are relatively enduring, comparisons between different periods of the patients’ lives offer a suitable means of obtaining a control. Thus, comparisons were made at three and six months with first contact with the patient.
Of particular interest in the present study is the finding that patients who combined biomedical services and traditional healing were less likely to be cases at three months follow up but more likely to be cases at six months follow up. Alternative explanations for the improvement at three months did not seem plausible. Those that were considered included spontaneous remission, selective attrition and a combination of use of western health facility and traditional healing.
Spontaneous remission seemed an unlikely explanation for the results. All patients were diagnosed with MINI plus for DSM-IV diagnosis thus eliminating acute psychoses. For affective psychosis, spontaneous remission may occur between 3 to 6 months. This study registered a positive change at 3 months, but not at 6 months.
A possible explanation for improvement of our patients might be selective patient attrition. There was no evidence that the more disturbed patients had dropped out of the study. Comparisons were made between the patients who completed the study and those who dropped out. No significant differences were found on any measure. There was a tendency for the drop outs to be females, born in Busoga but married in another district with a diagnosis of mania. However, this did not reach statistical significance. Furthermore; the drop out rate was low at 14.50% at three and 23.1% at six months.
Combination treatment seems to be the most plausible explanation for the improvement at 3 months.
An increasing number of studies have shown that psychological treatment combined with medication work additively on different complementary aspects of illness resulting in clinical benefits over medication alone[40–42]. Psychological treatment helps social functioning whereas medication controls abnormalities of mood and thought. Although in our study we did not include the kind of treatment these patients got from the traditional healers, previous research reports have indicated that traditional healers are good at dealing with psychosocial issues, thus offering psychological aspects[44, 45].
At six months follow up, combined use of both biomedical services and traditional healing was more likely in the cases. This may be explained by the need to achieve improvement by those who had not yet registered it. Hence the continued use of both biomedical services and traditional healing.
Our finding of over 20% to nearly 40% reduction in symptom scales scores is higher than a similar study carried out in India that reported 18.90% reduction in symptoms. The authors attributed their observed clinical improvement to the cultural power of residency in the healing temple and a supportive, non threatening and a reassuring setting since their patients had not had any western treatment. It could be that our patients had both medications from the western health facilities and the psychosocial input from the traditional healers thus offering better outcome.
Although ours is the first study to use standard clinical assessments to try and evaluate the outcome of traditional healing practices, our findings are only suggestive and not conclusive owing to the limitations of our methods and therefore the results should be interpreted with caution. First the numbers of individual psychotic illnesses were few. This could have exaggerated the percentage reduction. However, the number of 20 people is adequate for a statistically significant measuring of the difference before compared to after the intervention.
Secondly, it was not possible for us to have patients go for only traditional healing or western facility since in reality both are used concurrently[48, 49] .This makes it hard to maximise methodological rigour and minimise the intrusiveness of the research .
Despite efforts to explain the improvement in our patients, the other possible explanations like natural course of schizophrenia and mania may still be the reason for the improvements that was seen. A longer-term follow-up, at least for 12 months with a larger sample size would help to address these possible explanations in order to have a more conclusive report about the effects of traditional healing in psychosis.
Nevertheless, such a research has a useful role in helping to assess needs and resources for developing locally relevant mental health programmes. In terms of research, a lot remains to be done in this area of traditional healing in relation to psychosis.