OBJECTIVES: Economic evaluations are preferably performed from a societal perspective. This implies that participants report their healthcare utilization continuously during follow-up. Because this is a burden to participants, often resulting in missing values or withdrawal, researchers advocate collecting data discontinuously (i.e. in at least three months a year). This study aimed to compare costs of discontinuous measurement of healthcare utilization with those of continuous measurement, using several discontinuous measurement patterns and three imputation techniques: Individual Mean (IM), Last Observation Carried Forward (LOCF) and Next Observation Carried Backward (NOCB).  METHODS: We used continuous healthcare utilization data from a trial with twelve months’ follow-up and simulated several discontinuous measurement patterns combined with different imputation methods, to calculate simulated annual volumes and costs. Correlations and paired T-tests were used to compare simulated annual costs with actual annual costs from the continuous measurements.   RESULTS: Analyses confirmed that discontinuous measurements using cost diaries offer good estimates of annual health expenditures, but measurement patterns and imputation methods did influence the outcomes, as the correlations differed between methods. The best estimated annual costs were obtained by random cohort measurement, using three random cohorts, ensuring that at least a third of the participants were measuring costs each month, combined with IM imputation. Discontinuous measurement of health expenditures carries a small risk of missing infrequent expensive events, which may result in underestimation of annual costs. CONCLUSIONS: To reduce the burden on participants in future economic evaluation, we recommend calculating annual costs from discontinuous measurements in random cohorts, combined with IM imputation.