Assessment On Cause Of Death Statistics And Relevant Measures Research In The Urban And Rural Communities Of Anhui Province

Background: Cause of death statistics are one of the most important epidemiological data, and also critical for health policy planning, resources allocation,management and evaluation。3337 elderly (1736 65-year-old and above are from urban community who have stayed in Hefei for 5 years or more, 1600 60-year-old and above are form rural community in Yingshang country.) have been selected in Anhui Province in China in 2001 and 2003.This study aims at collect mortality date using verbal autopsy during the 5-year and 7-year follow-up .Because of large population and economic factors, it is difficult to set up and maintain mortality registration system in our country. Over the past 10 years the disease surveillance report shows that around 80% of people have died at home in rural areas, at the same time, at the lack of medical help, even some people did not go to hospital, and cause of death are just provided orally. Even the mortality date in urban areas also has many potential deficiencies. Given this background, statistics on registered causes of death in urban and rural China require careful evaluation if they are to be useful for epidemiological research and public health planningObjective: This study aims at using the tool of Verbal Autopsy, a systematic retrospective inquiry of the family members about the circumstances, events, symptoms, and signs of illness prior to death, to help determine the underlying cause of death and to classify the broad patterns of mortality. And then compare it with the present statistics to make clear the difference and the reason so as to provide scientific basis to the government's policy makers for emphasizing Mortality registration and the Quality. Method: Defined verbal autopsy (VA) procedures were used to derive underlying causes of death. k Measures of agreement between VA and registered diagnoses were computed. VA diagnoses were used as references to compute sensitivity and positive predictive values. All date were analysed by using the SPSS.The methods of descriptive and Consistency test were used.Result:(1)The distribution of cause of death in urban and rural community are neoplasms, cerebrovascular disease, cardiovascular diseases, respiratory disease, accident;The proportionate distribution of nervous system diseases in urban community is higher than that in rural community, and communicable diseases and accident are lower. (2) The main component of death place of urban community is patients'room, account for 67.8%,while the rate of that in rural community is 1.7%.The percentage of people without any health service is 11.7% among the death in rural community comparing with 6.0% in urban community. 73.0% of people in urban community can get health service provided by provincial level hospital. The compositions of hospitals that are accessible for rural person are country level hospital (26.9%), health centre (18.6%) and health room (7.9%). (3) The coincidence between cause of death statistics in urban community and VA is 78.5%, and that of neoplasms, nervous system diseases and accident are 100%. Kappa Scores indicate bad agreement for hypertensive disease, chronic obstructive pulmonary disease (COPD), pneumonia(Kappa<0.4); moderate agreement for ischemic heart disease, asthma, myocardial infarction, diabetes and cerebral thrombosis(0.75>Kappa≥0.4). only good agreement for cerebral hemorrhage(Kappa≥0.75); (4) The coincidence between cause of death statistics in rural community and VA is 64.8%, and that of nervous system diseases and accident are 100%. Kappa Scores indicate bad agreement for hypertensive disease, myocardial infarction, cor pulmonale and stroke(Kappa<0.4); moderate agreement for cerebral hemorrhage, cerebral infarction, cerebral thrombosis, coronary heart disease, COPD, lung cancer, colorectal cancer, pneumonia, asthma and diabetes(0.75 > Kappa≥0.4); Only good agreement for esophagus cancer, liver cancer and stomach cancer(Kappa≥0.75); (5) The results also suggest that cerebral thrombosis, diabetes, hypertensive disease and pneumonia are overreported and myocardial infarction and cerebral infarction are underreported in urbal community and asthma, cerebral thrombosis, hypertensive disease and coronary heart disease are overreported and myocardial infarction, pneumonia, cerebral infarction, cerebral hemorrhage and liver cancer are underreported in rural community.Conclusion: (1) The local registration system and VA report the same distribution of cause of death, which is neoplasms, cerebrovascular disease, cardiovascular diseases, respiratory disease, accident;(2) A significant degree of cross attribution of cause was found between the two data sources.and focus on hypertensive disease, COPD and ischemic heart disease. The statistics in rural community still show some misclafication of digestive and respiratory system neoplasms;(3) There are big differences of the health service accessibility between the urban and rural community, which may be the underlying risk factors of accuracy of death statistics; (4) To improve the accuracy of the cause of death statistics, it is important to carry out quality assessment of the cause of death statistics regularly, network construction of cause of death registration, promote network reporting, standardize the processes and functions.