OBJECTIVE: to assess the quality of information from the SIP-CLAP/OPAS in maternity hospitals in Belo Horizonte. METHODS: completeness was assessed by analyzing a systematic random sample of 562 SIP forms and the corresponding medical records, collected while care was being given at two maternity hospitals in 2004. A sub-sample of 20% was used to evaluate the reliability of the program's electronic database. Proportions were calculated for completeness before and after recovery of information available on medical records, along with the kappa index and the intra-class correlation coefficient (ICC), with a level of significance of 5%, for analysis of agreement. RESULTS: the completeness of SIP variables while care was being given was on average 72% in Hospital 1 and 86% in Hospital 2. The average increase after recovering the information in medical records was of 18% and 7%, respectively. A good level of agreement was found at both hospitals. CONCLUSIONS: although the SIP-CLAP represents an alternative way of monitoring perinatal hospital care, there are still some problems regarding it's adequate use, as completeness of the data evaluated during care was not completely satisfactory. Investments are necessary to improve the program's use in hospitals, in order to obtain essential indicators to qualify perinatal hospital care.
Health services evaluation; Information systems; Perinatal care; Infant mortality
