Background and Aim: Undercounting is a common problem in surveillance systems and registries. One of the procedures has been used for assessing sensitivity of a surveillance system or completeness of a registry is capture-recapture method.The objective of the present study was to estimate the number of deaths due to road traffic injuries applying capture-recapture method and using three data sources: police, legal medicine organization and hospital.
Material and Methods: All of the deaths due to traffic injuries occurred within Kerman district in the year 2000 were derived from police, Shahid Bahonar hospital and Kerman legal medicine data sources. Matching cases between the lists was based on three characteristics: first name, family name and date of accident. Loglinear model was used for statistical analysis.
Results: The total number of identified cases was 471 the best fitted loglinear model estimated the actual number of deaths as 596 (CI 95%: 543-686). Based on Iranian Statistical Center estimates, the population of Kerman district in the year 2000 has been 644673 so the cause-specific mortality rate of traffic injuries is estimated as 92 (CI 95%: 84-107) per 100,000 population. Therefore the proportion of deaths registered in police, legal medicine and Shahid Bahonar hospital are 16%, 58% and 48% respectively. The coverage of total deaths (471) is about 79%.
Conclusion: The findings showed that none of the data sources had enough coverage of all deaths due to traffic injuries. Capture-recapture estimates can help for obtaining better estimates.
