Comparison of research diagnostic systems in an Edinburgh community sample

Four research diagnostic schemes are compared in one community sample. The prevalence of psychiatric disorder ranged from 8.7 per cent (ID-Catego, threshold and definite) through 13.7 per cent (RDC, probable and definite) to 20.3 per cent (Bedford, borderline and definite). The main comparison made is between the PSE/ID/Catego and SADS/RDC systems. Sixty-one per cent of cases are identified as such by both these schemes. There is poor agreement about labelling; only 56 per cent of cases of depression and 16.7 per cent of cases of anxiety are so diagnosed by both systems. A post hoc check list was used to identify Bedford cases; all bar one were found to fulfil RDC and PSE case criteria. The results are compared with those from other centres which have used the same diagnostic criteria in community studies.