Abstract:
Objective To evaluate the completeness of acute myocardial infarction (AMI) incidence reports in Chronic Disease Incidence Surveillance System (CDISS) of Tianjin between 2007 and 2015 using DisMod II (DisMod stands for DISease MODelling, a software developed by World Health Organization).
Methods We extracted the data on AMI incidence and mortality among residents in Tianjin between 2007 and 2015 from CDISS and Death Registry System; we also collected data on remission rate for AMI patients in Tianjin published by Institute of Health Metrics and Evaluation. Then we calculated mortality/incidence ratios (M/I) by original data and DisMod Ⅱ. The difference of M/I values between the above two methods were tested, and then underreporting rate of AMI incidence in CDISS was estimated.
Results The DisMod-adjusted M/I values of AMI were all less than 1 (ranging from 0.49 to 0.97) during the period. The directly calculated M/I values for the population aged ≥ 65 years were significantly greater than the DisMod-adjusted M/I values (all P < 0.05). Compared to the reported average annual number of AMI incidence (8 563), a greater average annual number (11 837) of AMI incidence was estimated by DisMod calculation, especially for the number of AMI incidents aged 65 years and above. The estimated average annual underreporting rate of CDISS was 26.4% for all AMI incidents and the rate was 42.9% for the AMI incidents aged 65 years and above.
Conclusion There may be some underreporting AMI incidents in CDISS of Tianjin, especially for those aged 65 years and above and calculations with DisMod software could be used to evaluated the completeness of AMI disease surveillance system.