SURVEYING AFRICAN CARIBBEAN ELDERS IN THE COMMUNITY - IMPLICATIONS FOR RESEARCH ON HEALTH AND HEALTH-SERVICE USE

We established a population-based register of African Caribbean people aged 65 and older by door knocking every other household within an el ectoral ward. The remaining households within the ward were either app roached or eliminated on the basis of information from neighbours. The number of residents aged 65 and older found during our door knocking was fewer than those specified by the 1991 census for the black Caribb ean and white ethnic categories, although the percentage of the census accounted for was higher for black Caribbeans. The discrepancy betwee n our register and the 1991 census was inversely proportional to age. Based on information provided at the doorstep, there was no significan t difference between African Caribbean and white elders in extent of G P registration and in time since last attendance at a GP appointment. We conclude that door knocking provides a feasible method for developi ng a register of African Caribbean elders residing within a defined ge ographical area. However, care must be taken in enumerating the young elderly.