Background:Malignancy as a common cause of mortality needs to be investigated for its epidemiology and pattern of distribution in population. Data collection about cancer epidemiology is the first step in preventing and controlling this health problem. Cancer registries provide basic information for many epidemiological studies. In this study, reporting and notifying the rate of cancers to Fars province cancer registry center (FPCRC) were examined during 2004-5.Methods:10% of all admitted patients in Shiraz hospitals during 2004-5 were selected as studied group. So, 300 medical files were selected by multi-stage random sampling; then, then, 3 Shiraz University of Medical Sciences affiliated hospitals and 3 private hospitals were selected by cluster random sampling. Through medical records of patients diagnosed with cancer in 2005 in these hospitals, the study group was selected by systematic random sampling. These data were compared with data kept by FPCRC. Reporting and notification rates of different cancers to FPCRC were calculated.Findings:The reporting rate of cancer cases to FPCRC was 60%. The accuracy rate of those reported cases was 87.5%. The reporting rate of the cancers from SUMS affiliated hospitals was significantly lower comparing to other hospitals (p < 0.001).Conclusion:The reporting rates of cancers to FPCRC are obviously low and appropriate interventions are necessary.Key words:Cancer, medical report rate, Fars province, Iran.