Can we rely on public data as a source of information for cancer registry in developing countries?

Although a "hospital-based cancer registry" is important in improving patient care, a "population-based cancer registry" with emphasis on epidemiology is important in allocating health care resources and prioritizing public health programs. Because of its reliance on retrieved clinical and para-clinical documents, there is some limitation in registering all cancer incidents in this system, especially in developing countries. In this study we examined the possibility of using public data as a complementary source of information for recording cancers in a population-based cancer registry.

Along with the annual census in rural areas, a survey was performed in Golestan province in March 2004 to identify public awareness about cancer incidents in the community. Individuals were questioned about history of cancer in their close relatives during the last two years. Those who reported cancer in their relatives were also asked to name the main organ of involvement. A similar list was retrieved from the cancer registry at the Ministry of Health in Gorgan, and cases with upper GI (esophagus and gastric) cancer diagnosis from 21 March 2002 through 20 March 2004 were selected for this study. Finally, these two lists were compared for examining accuracy of the collected data.

We included 137 cases in our study with rural residence and known addresses. Only 35 (25.5%) cases were reported by the relatives and among them only 20 (57.1%) relatives correctly reported the tumor location. Although we found a difference in accurate reporting of cancer incidents by year of diagnosis (more correct cases reported during the second versus the first year), the difference was not statistically significant between the two years.

In this study, we examined the possibility of using public awareness about cancer incidents as a complementary source of information for a population-based cancer registry. We found that this approach is not ideal for reducing limitations. Therefore, we recommend a nationwide cancer registry to record all cancer-related information at the time of diagnosis. This strategy will reduce the need for performing retrospective surveys in collecting cancer-related information.