We read Andrew Cheng's (Reference Cheng2001) editorial with much interest. We strongly agree that the development of cross-culturally comparable diagnostic interviews is a pressing need.

In a recent survey in our unit in Sri Lanka of 43 patients presenting with depressive disorder, one-third of these on presentation made a subjective complaint of a “burning sensation of the body” (literal translation) and related secondary distress and denied having most of the core depressive symptoms although the symptom manifestation was of a depressive disorder. Thus, finding semantic or psycholinguistic equivalence for psychiatric symptoms across cultures will be a challenging, albeit necessary, exercise.

We believe that the lack of valid diagnostic tools is an important factor in the limited capacity for psychiatric research in developing countries, which in turn contributes to the underrepresentation of such research in high-impact journals noted by Patel & Sumathipala (Reference Patel and Sumathipala2001).

A case in point is that in Sri Lanka the only validated psychiatric rating scales in the native languages are the Mini-Mental State Examination (MMSE) and the General Health Questionnaire (GHQ-30). Efforts at validating the Hospital Anxiety and Depression (HAD) scale (D. de Silva, personal communication, 2001) in Sinhala (the language of the majority) show that the sensitivity and specificity of such an instrument is low. This is noteworthy considering the fact that locally developed diagnostic instruments may not find ready acceptance in high-impact journals.