Bolton worked together with World Vision, a non-profit organization that set up a program to address the psychological aftermath of the Rwandan genocide. The genocide had claimed the lives of over 800.000 people in 1994. World Vision employed community-level interventions to deal with mental health issues that were the result of trauma. Bolton wanted to investigate the local validity of western mental illness concepts - that is, was a Western diagnosis of depression or PTSD accepted by the local community as a valid description of their response to trauma?

In order to determine the validity of the diagnoses, Bolton outlined six steps that were necessary for an emic approach to diagnosis.

1. Collection of data on local perceptions of mental health in the community.
2. Analysis of the data for evidence that they meet some of the Western indicators of mental problems - thus making comparison possible.
3. Use these data to adapt and translate existing questionnaires that measure these indicators.
4. Test the credibility of these questionnaires.
5. Use the validated instrument in a community-based survey.
6. Analyze the data to assess the local prevalence and characteristics of the selected mental health indicators.

Using these six steps, the researcher hoped to determine the extent to which local people experienced depression as a result of trauma.

The researchers carried out a case study on two rural areas in Rwanda.  As part of their case study, they used three interview styles: first, free listing provided a list of local terms for mental symptoms and disorders. Interviewers generated free lists by asking 40 local people to name all the problems that had resulted from the genocide and to briefly describe each one. Then, using inductive content analysis, they pulled out symptoms that were related to mental health.

As a result, the following lists of symptoms were obtained:

Next, key informant interviews provided more detailed information about these disorders. When mental health issues were identified, participants were asked for the names of people in the community who were knowledgeable of these problems - and further interviews were then carried out. Seven people were identified as knowledgeable; they were traditional healers and local leaders.

Finally, pile sorts confirmed the relationships among symptoms and disorders. After the interviews were carried out, a set of cards was created which included the mental health symptoms that were identified in the initial interview as well as the symptoms that identify depression as outlined in the DSM. The healers were asked to sort the cards based on similarity. Consistently, the healers and local leaders included three symptoms with the DSM diagnosis of depression: lack of trust in others, loss of intelligence, and mental instability.

They then used these local symptoms as part of their questionnaire which they developed to determine prevalence levels of depression in the community.

The researcher found that interviewees described the diagnostic symptoms of depression and post-traumatic stress disorder as results of the genocide and also described associated "local" symptoms not included in the established diagnostic criteria. They divided symptoms into a "mental trauma" syndrome that included the post-traumatic stress disorder symptoms and some depression and local symptoms (Guhahamuka), and a grief syndrome that included other depression and local symptoms (Agahinda gakabije).

After the questionnaire was established, 93 people were identified as having mental health issues in the community. When interviewed, 70 were diagnosed by local healers as showing the symptoms of agahinda gakabije. When given the newly standardized questionnaire, 30 of those then tested positive for signs of depression. The relationship between agahinda and depression was the same as that between grief and depression in the Western world. When the test was administered to the larger community, 368 adults were interviewed. 17.9 per cent met the DSM criteria for depression and 41.8 per cent described themselves as having agahinda gakabije.

There are several strengths to this study. An emic approach makes use of local resources in order to help with the diagnosis of mental health issues. By adopting this approach, clinicians consider how the various components of culture have shaped an individual’s health and help to determine appropriate treatment. The process is highly systematic, making use of data triangulation to strengthen the credibility of the findings.

There are several obstacles that have to be overcome in order to make sure that this approach is effective. First, trust needs to be established between the researcher and the local community. Second, there needs to be a well-trained local staff that understands data collection techniques. Several NGOs have access to people with the education level to carry out the work effectively. Lastly, there has to be a reliable means of translating from the local language to the language of the research team. Often back translation - that is, translating interviews back into the original language - is used to establish credibility.

The researcher himself has noted several problems with the research. First, it is reliant on determining which Western disorder most resembles the locally defined problems. The diagnosis of depression is then based on local definitions of symptoms, compared to the symptoms as defined in the DSM. There is no outside verifiability of an actual diagnosis of depression - except by drawing comparisons to diagnoses in the West, which may also be flawed.

Emic approaches are also limited to the community that is studied, and thus has very limited generalizability. For example, in Uganda, the researchers identified local symptoms that were more similar to depression than in Rwanda.

The main limitation of cross-cultural psychiatry, of course, is that it fails to recognize that cultures are dynamic, complex social constructs which defy easy definition or measurement. As this study only took into account two rural communities in Rwanda, it is possible that their symptoms are a direct result of their experience in the genocide, which may be different from other parts of Rwanda.

Bolton, Paul (April, 2001). Local perceptions of the mental health effects of the Rwandan genocide. Journal of Nervous & Mental Disease, 189, pp 243 - 248.

Bolton, Paul (September, 2002). Prevalence of depression in rural Rwanda based on symptom and functional criteria. Journal of Nervous & Mental Disease, 190, pp 631 - 637.