Measurement of resource utilisation in cancer clinical studies – tools, issues and challenges

Inclusion of economic evaluations alongside cancer clinical trials necessitates the collection and analysis of resource utilisation and cost data alongside outcomes. The purpose of this paper is to describe and discuss the measurement of cost in clinical studies, particularly resource utilisation. Cost data collection can be conducted retrospectively through linkage of treatment data with claims data, such as Medicare, or by patient recall (questionnaires). Prospective approaches include the patient diary. Measures and data collection tools are usually modified by researchers to fit the purpose and target population of their specific study. There is strong agreement on the inclusion of direct medical and non-medical costs in economic evaluations. The balance of opinion is that inclusion of indirect costs is appropriate; but agreement on exactly ‘which indirect costs’ and in ‘what context’ differs. However, narrow study perspectives mean that inter-sectoral resources are often overlooked. In addition to the two cancer-specific instruments included in the Database of Instruments for Resource-Use Measurement, there are numerous resource utilisation measurement tools used in a broad range of clinical research with heterogeneous intervention characteristics and outcome measures. Despite this, very few studies report validated cost/resource use instruments. Further, many cost analyses ignore long-term care costs, non-medical costs borne by patients and important costs incurred in other sectors, such as social services. There is no ‘gold standard’ for resource utilisation instruments and the agenda for future research is lengthy. For example, many issues such as recall length, accuracy in recall of medical terms and medicines, specificity versus comprehensiveness of the instrument and missing data, remain to be addressed. Innovation in mobile technology will likely revolutionise data collection and may overcome many of the existing barriers to robust measurement of resource utilisation for cancer clinical trials and improve societal decision making.