Collecting psychiatric resources utilisation data to calculate costs of care: a comparison between a service receipt interview and a case register

Background: Economic assessment of inter-ventions and policies is becoming increasingly common,in large measure because of the growing emphasis oncost containment within health care. Comprehensiveand reliable outcome and cost data are required to ad-vise policy makers and clinicians as to the best use oftheir limited resources. The process of costing can bebroken down into three connected tasks: the collectionof service receipt or utilisation data relative to individualclients or patients over a de®ned period; the costing ofeach of the services used; and the combination of thesetwo sets of information in order to calculate individualcosts. The aim of this study was to compare two meth-odologies of collecting data on individual service use ± acustomised interview schedule, ICAP, and the psychi-atric case register (PCR) ± and to calculate costs, testingthe extent of agreement between them. Method: Theagreement between the ICAP and the PCR costs mea-surement was evaluated using the concordance correla-tion coe cient qc. From all patients (n = 543) who hadatleast one contact with a psychiatrist or a psychologistduring the period October±December 1996, 339 patientswere interviewed using the ICAP. The overall number ofpatients in contact with the South-Verona CPS in thesame period was 630. Results: The agreement betweenthe two sources was very di€erent for each diagnosticgroup and each professional category. However, theoverall agreement on total costs was satisfactory(qc < 0.95). This result is probably due to the e€ect ofthe good agreement observed on more costly services:inpatient care and sheltered accommodation. Conclu-sion: The results suggested practical implications for theuse of the service receipt interview: interviewers shouldbe trained in order to avoid misinterpretation of thede®nitions given in the form; the sources of informationshould be clearly de®ned to tease out all the items ofservices provided for the users; the professionals (i.e.psychiatrists) could in¯uence the reliability of data col-lection by underestimating services provided by di€erentprofessionals (nurses, social workers, etc.). The ®ndingscon®rm that it is possible to use this approach when theaim is to estimate the whole cost of the services; how-ever, the importance of adopting adequate proceduresfor analysing the complexity of cost components shouldbe pointed out. Only a trained interviewer whothoroughly knows each component of the health andsocial services provided could guarantee an accuratedata collection.