OBJECTIVES: Accurate measurement of resource use is required for economic evaluations alongside clinical trials.  Patient-completed questionnaires are commonly employed as a means of collecting data; however, concerns over data quality persist, and there is little certainty about best practice.  This review collates the evidence concerning the validity and reliability of resource-use measures based on patient recall with the aim of aiding health economists in developing better measures. METHODS: A search strategy incorporating terms covering healthcare resources, utilisation, patient-reported measures and validation/reliability concepts was applied to the MEDLINE, EMBASE and PsycINFO bibliographic databases.  Studies were included if they reported original research to inform costing studies, and were about patient or proxy self-reports of direct healthcare-related resource use in which a comparator (to assess validity or reliability) was specified.  Studies were excluded if they were not in English or if they assessed general population surveys.  Reference and citation lists of included studies were hand searched to identify additional studies.  Data on study and population characteristics, type of instrument, recall period and sample size were extracted.  Results and conclusions concerning the validity and reliability of reports of types of resource use consumed (e.g.medication, inpatient stays) were also extracted. RESULTS: 13367 abstracts were identified as potentially relevant through the database searches.  Following abstract and full-text screening, 60 articles were deemed relevant, with a further 9 identified through hand searching.  The majority focused on adults (60/69), and nearly half originated from north America (31/69). Emerging themes suggest that better accuracy is achieved when patients are answering questions about inpatient and specialist care. CONCLUSIONS: There is only a limited amount of validity and reliability information available to inform best practice for resource-use measurement in clinical trials. This ongoing review will identify the gaps, giving a clearer view of where research efforts should be concentrated.