Search

A multicentre study concerned with the health needs of the elderly population living in urban areas and coordinated by the Panamerican Health Organization was undertaken in 6 Latin-American countries. In Brazil, 1,602 elderly residents (of 60 years of age and over) in the District of S. Paulo, constituting a multistage random sample stratified by socioeconomic status, participated in a household survey using a multidemensional functional assessment questionnaire. The results showed a highly deprived population (70% had a per capita income of less than US $ 100 per month), living mostly in multigenerational households (59% were living with children and/or grandchildren), with a high prevalence of chronic physical illnesses (only 14% referred no illness) and psychiatric disorders (27% were considered psychiatric cases), a high proportion of them showing a loss of autonomy (47% needed help in performing at least one of the activities of daily living). The results are put into perspective concerning future needs in terms of specialized health services and social support for the growing population of elderly people in Brazil.

As a part of a survey to study the health and living conditions of the elderly population, a random sample of residents aged 65 and over are examined using the Clinical Interview Schedule (CIS) in order to evaluate their psychiatric status. The aim of this study is to evaluate this standard method of assessment as a case-identification instrument in our country. The schedule was completed by 91 subjects. It is easily administered, easily scored, and economical on time. Its completion rate is high. The weighted total scores (WTS) range from 0 to 48. Using the case criteria defined by Cooper & Schwarz (1982), 27 subjects (30%) are considered cases and 64 (70%) are regarded as non-cases. The sensitivity coefficients for the WTS are examined against the overall severity rating at different cut-off points. The optimum cut-off can be anywhere between 16 and 20 points. The WTS has higher validity coefficients to detect the following diagnostic categories (sensitivity, specificity): normals (100%,-); personality (100%, 92%) and affective disorders (100%, 75%). In general the CIS items are given low ratings. Psychotic symptoms are rarely found in this sample. One main problem arose: the item depersonalization is misunderstood by some patients probably because of interpreting it as an upsetting memory disturbance.

Considering that in most developing countries there are still no comprehensive lists of addresses for a given geographical area, there has always been a problem in drawing samples from the community, ensuring randomisation in the selection of the subjects. This article discusses the geographical stratification by socio-economic status used to draw a multistage random sample from a community-based elderly population living in a city like S. Paulo-Brazil. Particular attention is given to the fact that the proportion of elderly people in the total population of a certain area appeared to be a good discriminatory variable for such stratification. The validity of the stratification method is analysed in the light of the socio-economic results obtained in the survey.