利用傾向分數探討老人流感疫苗施打記錄對預防急性冠心病成效之影響

Autor: 黃柏堯
Rok vydání: 2017
Druh dokumentu: 學位論文 ; thesis
Popis: 105
Acute coronary syndrome (ACS) is one of high morbidity and mortality disease worldwide. Several domestic and international studies have shown that influenza vaccine is an effective method to prevent acute coronary syndrome. In Taiwan, the vaccination is free for all adults aged≧65 years old since 2001. Vaccination effectiveness is affected by years, the popular state of the virus, the vaccine strains match, the annual vaccine status, past vaccination records and the individual status. The purpose of the study is to investigate the effect of the year’s exposure to the annual vaccination status and past vaccination record to prevent ACS. We used the data of one million people sampled by Longitudinal Health Insurance Database 2005 (LHID2005) in National Health Insurance Research Database, to sift out data from 2007 and 2008 for people aged over 68, to evaluate the protective effect of influenza vaccination on ACS and to investigate whether the influenza vaccination has protective effect to ACS occurs under different vaccination status in the past. The statistical method first uses the propensity score to analyze.The propensity score is a conditional probability, defined as the probability that the subject is under the control variable observed under it, accept the probability of being treated or assigned to a particular group. The most common way to estimate propensity score is logistic regression. According to the estimated propensity, there will be a one-to-one match between those who have been vaccinated and those who have not been vaccinated. Finally, we used Conditional Logistic Regression model to explore the effectiveness of the vaccine. The result shows that in the case of vaccine strains coincide, when the annual vaccination has a significant protective effect. But the vaccine strain doesn’t coincide when annual vaccination doesn’t has a significant protective effect. Regardless of whether annual vaccination was vaccinated. At least three years of continuous vaccination in the past has a significant protective effect.
Databáze: Networked Digital Library of Theses & Dissertations