Bidirectional Associations BetweenBronchiectasis and Tuberculosis, A National Health Insurance Database Study from Taiwan

• Main Authors: Pao-Jen Hsu, 許寶仁
• Other Authors: Tzu-Chi Lee
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/37548326468883586664

碩士 === 高雄醫學大學 === 公共衛生學研究所 === 101 === Background and Objective: Tuberculosis (TB) is a global infectious disease which makes remarkable morbidity and mortality in people. Screening for TB is indicated for groups with high risk to decrease disease transmission. There was no study assessing the effect of bronchiectasis on TB and bidirectional associations between these two diseases. To investigate this relationship, we performed a retrospective cohort study with the National Health Insurance Research Database (NHIRD) in Taiwan. Methods: New subjects were identified from the NHIRD with a primary diagnosis of bronchiectasis (ICD-9: 494) and TB (ICD-9: 010-018) respectively between 2002 and 2007. In the main direction of the research, the subjects were verified by inpatient diagnosis of bronchiectasis and/or at least 3 months diagnosis of bronchiectasis in outpatient service. There were 1,752 new adult bronchiectasis cases and 10,512 gender-, age-, urbanization- and insurance premium- matched subjects. All were followed until December 31, 2008. Competing risk adjusted Cox regression was utilized to reveal the effect of bronchiectasis on TB by controlling gender, age, urbanization, insurance premium, steroid use, diabetes mellitus, renal diseases, malignancies, chronic obstructive pulmonary disease and death. TB was defined as those with prescriptions of anti-TB agents for at least 2 months and with medical service claim of ICD-9 codes, for at least 2 months. In the other direction, the similar method was applied to investigate the effect of TB on bronchiectasis by controlling gender, age, urbanization, insurance premium, lung cancer, receipt of organ transplants, rheumatoid arthritis, Crohn’s disease, autosomal dominant polycystic kidney disease and death. Results: In the main direction, 99 subjects claimed TB in a total of 12,264 subjects, including 37 (2.11%) within the bronchiectasis group and 62 (0.59%) within the non-bronchiectasis group. The average follow-up time was 4.12±1.78 (mean±SD) years. The life time TB incidence for bronchiectasis group was 544.3 (95% CI at 383.2-750.2) per 105 person-year, and for non- bronchiectasis group was 141.6 (95% CI at 108.6-181.6) per 105 person-year. After adjusting for competing risk of death and confounding factors, bronchiectasis cases increased TB risk by 267% (HR=3.67; 95% CI=2.34-5.77, p<0.001). In the other direction, TB cases increased bronchiectasis risk by 501% (HR=6.01; 95% CI=4.34-8.32, p<0.001). Conclusions: Bronchiectasis has been quantified as a risk factor on TB in Taiwan after controlling the competing risk of death and confounding factors. In addition, TB also has been verified as a risk factor on bronchiectasis, a well-known relationship, in this retrospective cohort study. There are bidirectional associations between these two diseases.