A pilot project was undertaken in Liverpool UK, to investigate whether data useful for both general practice and public health purposes could be collected by primary care personnel using a computerized semi-structured interview (AGE-PC) for assessment of patients aged 75 and over. A high degree of variation between practices was found in the proportion of elderly assessed with AGE-PC largely related to motivation to undertake assessments and methods employed to approach patients. Considerable differences between and within practices were evident in staff attitudes to the need for detailed formal assessments and the use of computers. Several aspects of the AGE-PC package itself were identified as requiring modification including streamlining data transfer methods, adding facilities for analysis and giving users a degree of choice over content of the assessment. This technique does offer a potentially efficient means of collecting very detailed standardized data for intra- or inter-practice analysis while at the same time generating a comprehensive report for individual patients. However, in the context of the over-75 assessments it was too time-consuming. Possible future strategies are discussed such as using an initial screening stage with a self-report questionnaire to reduce numbers needing full assessment and the use of the package for the community care assessments. The pilot project highlighted some of the potential advantages and disadvantages of collecting data from primary health care for public health purposes.(ABSTRACT TRUNCATED AT 250 WORDS)