Increasing emphasis is now being placed on the assessment of patient outcomes, both in evaluating medical interventions and in quality assurance initiatives. Clinicians, purchasers, managers and researchers need outcome measures that are valid, reliable and responsive. This paper describes the theory and practice underlying the development of outcome measures for two chronic conditions, asthma and diabetes, for application in ambulatory settings. Existing generic and condition-specific health status and health-related quality-of-life measures were administered to almost 1300 patients. The psychometric properties of these measures were examined to identify those that were of adequate validity and reliability in these population groups. Step-wise regression procedures were then used to identify a core set of scales that best predicted patients' general health perceptions, which could be used in measuring general health outcomes for each of these groups. These core sets consist of up to 40 items, spanning physical function, energy and vitality, emotional well-being and condition-specific aspects of health such as symptom control. Further analysis is being carried out to assess the responsiveness to change of these core item sets.