Susan Hasanpour-Heidari
1, Nastaran Jafari-Delouei
1, Nesa Shokoohifar
1, Seyed Mehdi Sedaghat
2, Abbas Moghaddami
2, Reza Hosseinpour
2, Mohammad Poorabbasi
3, Masoomeh Gholami
3, Shahryar Semnani
1, Mohammad Naeimi-Tabiei
4, Mohammad Reza Honarvar
2, Abdolreza Fazel
5, Arash Etemadi
6,7, Freddie Bray
8, Gholamreza Roshandel
1*1 Golestan Research Center of Gastroenterology and Hepatology, Golestan University of Medical Sciences, Gorgan, Iran
2 Deputy of Public Health, Golestan University of Medical Sciences, Gorgan, Iran
3 Death Registry Unit, Deputy of Public Health, Golestan University of Medical Sciences, Gorgan, Iran
4 Cancer Research Center, Golestan University of Medical Sciences, Gorgan, Iran
5 Omid Cancer Research Center, Omid Preventive Medicine and Health Promotion Center, Golestan University of Medical Sciences, Gorgan, Iran
6 Division of Cancer Epidemiology & Genetics, National Cancer Institute, Bethesda, MD, USA
7 Digestive Oncology Research Center, Digestive Disease Research Institute, Tehran University of Medical Sciences, Shariati Hospital, Tehran, Iran
8 Section of Cancer Surveillance, International Agency for Research on Cancer, Lyon, France
Abstract
Background: We aimed to evaluate completeness and accuracy of the Golestan Death Registry (GDR) to identify cancer-related
causes of death (CCoD).
Methods: The GDR data (2004-2015) were compared with cancer data collected from clinical/pathological sources (the considered
gold standard) by the Golestan Population-Based Cancer Registry (GPCR). Using a linkage method, matched cases, including
subjects with CCoD and those with ill-defined cause of death (ICoD) (garbage codes), were identified and entered into the final
analysis as study subjects. The completeness (percentage of study subjects with CCoD) and accuracy (number of subjects with
correct CoD from the total number of study subjects) of the GDR were calculated.
Results: In total, 3,766 matched cases were enrolled. Overall, the completeness and accuracy of the GDR for identifying CCoD
were 92.7% and 53.2%, respectively. There were variations by cancer site and age group, with completeness and accuracy highest
for brain cancer (96.3%) and leukaemia (79.8%) while the lowest accuracy was observed for colorectal cancer (29.9%). The
completeness and accuracy of GDR was higher in patients aged under 60 years (95.7% and 53.6%, respectively). We also found
higher completeness (93.7%) and accuracy (55.8%) in residents of rural areas.
Conclusion: Linkage of death registry data with cancer registry data can be a significant resource for evaluating quality of the death
registry data. Our findings suggested that completeness of the GDR for identifying CCoD is reasonable, but accuracy is relatively
low. Access to clinical and pathological data from other sources and enhanced training of death certifiers can improve the present
situation