Psychiatric morbidity in older people with moderate and severe learning disability. II: The prevalence study.

We present a prevalence study of psychiatric morbidity in people over 50 years of age with learning disability (LD), using a new semistructured clinical interview specifically for use with people who have LD (the 'PAS-ADD'). Assessment involved parallel interviewing of subject and informant, these two sets of information being combined to reach a final diagnosis using ICD-9 and DSM-III-R criteria. Detection of dementia involved interviews with informants, plus investigation of loss of cognitive function over a three-year period. The experimental sample was a mixed community and institutional group (n = 105), including, as far as possible, all people in a single administrative district (Oldham) matching the age and ability criteria. Prevalence of psychiatric disorder excluding dementia was 11.4% (n = 12), most of which were depression and anxiety. Seventy-five per cent of these cases were unknown to mental health services. However, immediate care staff were usually aware of the symptoms, although often unaware of their clinical significance. Prevalence of dementia was also 11.4% (n = 12), with a combined case prevalence of 21.0% (n = 22). The PAS-ADD proved a flexible interview, effective in use with people of varying linguistic level and intellectual ability: 61.9% (n = 65) of the sample were able to be interviewed, fully adequate clinical interviews being obtained with a group of 38 people whose mean IQ was only 39. In the remaining 38.1% (n = 40), diagnosis relied exclusively on informant data. Overall, the combination of subject and informant data was essential for sensitive case detection.