Abstract (EN):
Administrative data is of great value for the administration of health services. These databases are the base for hospital financing and for the process of contracting between hospitals and regional health authorities. This data is also used for the production of indicators describing the quality of care and measuring hospital performance. In this context, the variable Diagnosis Related Groups (DRG) plays a central role, because its codes are used to distinguish/select various types of episodes, for instance, the obstetric ones. Since the assignment of a hospitalization episode to a DRG is based on information extracted from a set of variables, the existence of possible coding or data quality problems associated with these variables can influence the outcome of the grouping process. In this article we highlight some of these problems associated with obstetric episodes, aiming primarily to alert to the possibility of the loss of some information, when episodes are counted using only DRG codes. As a result we present a new approach for the counting of obstetric episodes, the ObsCount algorithm, which can be used in the calculation of indicators in this medical specialty. © 2011 AISTI.
Language:
Portuguese
Type (Professor's evaluation):
Scientific
No. of pages:
6