Background. The methodology for measuring the needs of patients with severe mental illness is now well established through the MRC Needs for Care Assessment Schedule and its modification in the form of the Cardinal Needs Assessment. This paper reports the rationale and construction of a relatives' version of the Cardinal Needs Schedule and looks at preliminary data reporting on reliability and validity.

Methods. Potential problem areas for relatives were identified from the literature. The criteria determining Cardinal Needs for each problem included objective threshold, carer concern, and carer cooperation. The reliability of the Schedule was assessed in a study whereby 27 relatives of patients with established schizophrenic illness completed two Schedules administered by two independent raters within a short time period.

Results. The Schedule was acceptably reliable for most areas of need although there were some difficulties associated with the cooperation criteria. Concurrent measures of relatives and patients including EE, relative distress and patient psychopathology indicated that the Relatives' Cardinal Needs Schedule shows acceptable validity when used with a sample of 45 relatives.

Conclusions. The paper suggests that the Schedule may prove to be a useful tool for both clinicians and researchers interested in establishing and evaluating family interventions.