Influenza vaccination and medical care utilization for the community elderly in Taiwan

• Main Authors: Pei-Fen Chen, 陳珮芬
• Other Authors: Weng-Foung Huang
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/13718127947877619603

碩士 === 國立陽明大學 === 衛生福利研究所 === 92 === Can a country afford to annually vaccinate its entire aging population with a flu vaccine that may or may not be necessary? Taiwan’s Department of Health’s free influenza vaccination program for the elderly, those at high risk for contracting the disease, was implemented in 1998. In 2001, however, the program was expanded to cover all people 65 years old or older, not just those in that high risk alone, but also those considered healthy. Administering the vaccine annually makes it a considerable financial burden. As the aged grow in proportion to the rest of the nation, the effectiveness of the influenza program and whether to temper that age-based population (65 years or older) with some risk-based criteria has become a recently more popular research topics. It has been shown that one’s medical history often determines whether he or she opts for vaccination, and, therefore, making it a self-selected choice. Previous studies of this program have only used multiple linear regression to analyze the relationship of flu vaccination and medical care expenditures, and found greater medical costs in the vaccinated group than the unvaccinated group. Some researchers suggested using instrumental variables in the statistical methods to clarify to what extent the self-selection is involved and to make pseudo-randomization to compensate for that bias. Some studies have only used multiple linear regression to analyze this relationship, and they will underestimate the effectiveness of the flu vaccination. To solve this problem, this study uses the instrumental variable method to measure the effectiveness of the influenza vaccination program at reducing utilization of medical health care services. Study samples were taken from the Year 2000 National Health Interview Surveys administered to people 65 years old or older. After excluding those who died before the study period and those who did not agree to have their information linked to the National Health Insurance Database, we were left with a sample of 1,700 (mean age 73 years, SD+/- 6.3 years). In our study, multiple regression was introduced in two stages. In the first stage, the instrumental variable was used to predict every sample’s probability to vaccinate. In the second, the probabilities were used in a regression model for pseudo-randomization. Logistic regression was used to analysis death rate and to estimate the probability of visiting an outpatient department. Poisson regression was used when determining the frequency of the administration; linear regression when determining length-of-stay, administration fee, outpatient department visiting and outpatient fees. We found the influenza vaccination program for the elderly to reduce the death rate (OR=0.01). While the program did not significantly reduce the general admission rates, it did reduce fees paid for hospital inpatient services (NT$ 588,886 per person year, P<0.05) and the length-of-stay (119.94 days per person year, P<0.05). Outpatient services saw no such reductions. The influenza vaccination program greatly reduced hospitalization expenses for the high-risk group, patients 75 years old or older, whose inpatient services alone were reduced by NT$ 2,188,806 per person year. Based upon these findings, the government should increase the vaccination rate for very old people.