The Saudi National Cancer Registry (NCR), a population-based registry, showed an age standardized incidence rate (ASR) for all cancers among Saudi population of 80.9/100,000 during its first year of operation (1994). Because ASR was low, we estimated rate of under-registration with a capture-recapture method and assessed the validity of reported data from the major public referral hospital in Riyadh, the Riyadh Medical Complex (RMC).
Methodology
We compared cancer cases from three data sources: medical records (MR), which were the original source of NCR cases, pathology reports (PR), and death certificates (DC). We estimated missing cancer cases using the log-linear modeling of Fienberg with a correction to correct for interdependency among sources. To assess validity of the data we reabstracted records of 10% (47/476) of previously reported cases from RMC.
Results
A total of 811 cancer cases were reported through the three sources: MR had 611 (75.3%), PR 639 (78.8%) and DC 204 (25.2%). Fitting a series of log-linear modeling to the three sources of data, the three sources were found to be statistically dependent. The number of missed cases calculated by the three sources was 403, giving an estimation of total cancer cases to be 1214 (95% confidence interval=866-1562). Using these 1214 estimated cases, the estimated ascertainment rates were 50% for MR, 53% for PR, 17% for DC, and 67% for the aggregated registry. In the validity assessment, 17% of records were incorrectly filed and were not cancer cases. Among the remaining 83%, major disagreement about the abstracted and reabstracted data was found to be highest for stage of disease (44%) followed by histology code and behavior (25.6%). Minor disagreements were most common for date of diagnosis (36%) and grade (36%). Overall agreement was highest for laterality (95%) followed by primary site codes (90%) and basis of diagnosis (85%) whereas date of diagnosis and stage had the lowest agreement percentage. Agreement of tumor description variables (site, histology, behavior, and stage) was 57%.
Conclusion
Data reported from RMC in 1994 indicate that the NCR will require substantial improvements in both completeness of reporting and data quality at the hospital level. Use of multiple data sources and estimation of missed cases will help ensure completeness of case registration.