Chapter 4 Discussion
4.5 Conclusion
In the trend of population aging, in order to further control the epidemic situation of TB, it is necessary to reduce the incidence of the elderly population. Through this study, we can further understand the importance of different risk factors. Apart from unchangeable determinants, such as age and sex, past history of lung disease and abnormal BMI have a significant impact on the risk of active TB in elders. In addition, due to the limits of the sample size, we cannot ascertain the real effect of novel risk factors. The association between novel risk factors and active TB should be further explored through future research. As society progression, more useful tools or methods will be developed for diagnosis or treatment. We hope through this study more research could put efforts in elders and care more about elderly in TB epidemiology.
Figures
Figure 1. TB incidence rate and standardized incidence rate (adjusted to the 2005 Taiwan population) in Taiwan (2005-2017) (Adapted from Taiwan CDC data[22-24])
72.5
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Cases per 105 population
Tuberculosis incidence rate in Taiwan (2005-2017)
incidence rate
standardized incidence rate
Figure 2. Age-specific TB incidence rate in Taiwan (2005-2018) (Adapted from Taiwan CDC data[26])
0
10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 ≥ 65
Age-specific TB incidence rate in Taiwan (2005-2018)
2005
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
10-14 4.7 4.8 3.8 4.5 3.8 3.5 4.1 3.2 2.7 2.2 2.0 2.0 1.5 1.7
15-19 23.9 24.0 21.4 19.2 22.3 18.6 18.4 15.5 15.2 15.8 12.4 10.1 9.4 8.0 20-24 32.7 30.8 28.3 29.0 22.5 23.5 23.1 19.8 19.2 15.3 14.0 13.5 12.2 10.3 25-29 32.3 26.9 25.5 28.1 22.9 24.5 21.1 20.5 17.1 16.2 15.9 12.3 12.5 10.5 30-34 35.3 29.6 27.3 25.9 22.2 21.5 22.4 19.3 19.4 19.0 15.5 15.6 13.3 12.2 35-39 36.5 34.3 30.8 31.5 27.7 24.6 24.8 24.3 22.7 20.0 17.2 15.6 15.8 13.4 40-44 46.1 43.7 39.8 38.2 31.7 33.3 29.7 29.1 27.8 24.2 21.1 17.9 18.9 15.3 45-49 56.0 52.1 48.5 44.5 42.1 39.6 39.4 37.2 34.0 32.0 28.9 26.8 24.2 20.8 50-54 71.1 68.0 58.7 56.6 52.6 51.9 49.1 48.5 41.4 42.4 38.4 37.4 34.9 31.1 55-59 90.3 83.7 79.5 70.4 67.0 63.5 60.9 60.2 52.7 52.2 48.5 47.9 45.0 37.1 60-64 142.6 120.1 109.0 100.0 87.9 86.0 75.1 77.0 65.0 67.9 66.7 61.9 54.6 49.1
≥ 65 373.0 338.5 314.9 306.4 284.7 279.3 258.8 246.1 227.2 215.8 199.6 187.0 169.3 150.3
Figure 3. Percentage of elderly TB cases in Taiwan (2005-2016) (Adapted from Taiwan CDC data[27])
51.0% 51.0%
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Percentage of elderly TB cases in Taiwan
Figure 4. Natural history of TB and risk factors in elderly population
Figure 5. Flow chart of study population exclusion process and the sample size All participants without permission for NHI claim data were excluded. The 2001 NHIS did not have elderly specific questionnaires. Elderly participants enrolled from 2001 NHIS were extracted from total participants by age. Main dataset and subgroup dataset only contained participants without missing information in the survey.
Tables
Table 1. ICD-9 code and inclusion criteria for comorbidities in National Health Insurance
Comorbidities ICD-9 code inclusion criteria
Diabetes 250 outpatient records ≥ 3 or inpatient
records ≥ 1 before the survey Lung disease 491, 492, 493, 494, 496 outpatient records ≥ 2 or inpatient
records ≥ 1 before the survey Chronic kidney disease 585
Table 2. Baseline descriptive characteristic of study population
Characteristic Mean (std.) / N (%)
(Total = 4,657)
Age (yrs.) 73.4 (6.05)
Body Mass Index (BMI) (kg/cm2) 23.9 (3.41)
Visits of medical centers 15.7 (34.02)
Sex
Male 2,495 (53.6%)
Female 2,162 (46.4%)
Education
Less than elementary school 1,343 (28.8%)
elementary school 1,914 (41.1%)
Widowed, divorced, separated, or others 1,445 (31.0%)
Lung disease NHIS 318 (6.84%)
NHIS & NHI 1,406 (30.2%)
Diabetes NHIS 832 (17.87%)
NHIS & NHI 1,095 (23.5%)
Chronic Kidney Disease NHIS 377 (8.09%)
NHIS & NHI 440 (9.5%)
Characteristic N (%) (Total = 4,657) NHIS (Survey year)
2001 1,192 (25.6%)
2005 1,425 (30.6%)
2009 2,040 (43.8%)
Note. The result is under consideration of sampling weight. Comorbidities (lung disease, diabetes, and chronic kidney disease) had two definitions. 1) answers of questionnaires (having the disease and is taught by health care providers/practitioners).
2) answers of questionnaires or medical record of disease from NHI database.
Table 3. the number of TB cases and incidence rate (95% CI) distribution of
Illiterate or elementary school 16 98 (71-137)
Junior high school 27 210 (170-261)
Senior high school 8 172 (87-341)
College and above 16 384 (291-506)
Marital status
Never married 1 194 (23-1603)
Married or cohabiting 48 202 (171-239)
Widowed, divorced, separated, or other 18 156 (113-214) Lung disease
Variables TB case Incidence rate1 (95% CI) Employment
Employed 55 180 (154-211)
Unemployed 13 290 (194-436)
Personal income
Less than 5,000 27 177 (145-217)
5,000-15,000 19 111 (76-162)
More than 15,000 22 354 (272-460)
NHIS (survey year)
2001 36 183 (146-229)
2005 23 285 (230-353)
2009 8 72 (44-119)
Note. The result is under consideration of sampling weight.
1Incidence rate, the unit of incidence rate is per 100,000 person-year
Table 4. Univariate and multivariate Cox regression models for association between risk factors (variables) and risk of TB in main dataset
Variables TB cases
Variables TB cases (n=67)
Univariate Multivariate HR1 (95% CI) p-value aHR2 (95% CI) p-value Personal income
Less than 5,000 27 0.73 (0.42-1.29) 0.285 5,000-15,000 19 0.77 (0.42-1.43) 0.413 More than 15,000 22 Ref.
Note. The result is under consideration of sampling weight. The numbers of survival time were rounded to the nearest integer. The definition of comorbidities only included information from the NHIS.
1 Hazard Ratio
2 Adjusted Hazard Ratio
3 Body Mass Index (kg/cm2)
4 Widowed, divorced, separated, or other
Table 5. Analysis of multivariate Cox regression models for association between risk factors (variables) and risk of TB in main dataset
Variable Multivariate
Adjusted HR 95% CI p-value
NHIS survey year 0.96 (0.87-1.07) 0.470
Sex (Male) 3.07 (1.67-5.66) <.001
Age 1.08 (1.04-1.12) <.001
BMI1 0.86 (0.79-0.93) <.001
Lung Disease 1.60 (0.97-2.66) 0.067
Smoke 0.66 (0.39-1.14) 0.140
Note. The result is under consideration of sampling weight. The definition of comorbidities included information from the NHIS and NHI.
1 Body Mass Index (kg/cm2)
Table 6. Univariate Cox regression models for association between risk factors (variables) and risk of TB in subgroup dataset
Variables TB cases(n=35) Crude HR1 (95% CI) p-value Continuous
NHIS (Survey year) 0.79 (0.631-0.991) 0.0413
Age (yrs.) 1.09 (1.039-1.148) <.0001
BMI2 (kg/cm2) 0.82 (0.736-0.912) 0.0003
Visits of medical centers 1.00 (0.998-1.012) 0.177
Categorical
Variables TB cases(n=35) Crude HR (95% CI) p-value
Perceived changes in intake associated
with dental health 24 2.01 (0.975-4.154) 0.0585
Note. The result is under consideration of sampling weight. The definition of comorbidities included information from the NHIS and NHI.
1Hazard Ratio
2Body Mass Index (kg/cm2)
3Activity of Daily Livings
4Instrumental Activities of Daily Livings
5The Center for Epidemiologic Studies Depression Scale
Table 7. Adjusted hazard ratio of multivariate Cox regression models in subgroup dataset
Variable Univariate Multivariate
HRa 95% CIb p-value aHRc 95% CI p-value NHIS survey year 0.79 (0.63-0.99) 0.041 0.79 (0.63-1) 0.046 Sex (Male) 3.30 (1.46-7.47) 0.004 3.09 (1.36-7.04) 0.007 Age 1.09 (1.04-1.15) <.001 1.05 (1-1.11) 0.048 BMId 0.82 (0.74-0.91) <.001 0.85 (0.76-0.94) 0.002 Lung Disease 2.44 (1.25-4.74) 0.009 2.08 (1.05-4.1) 0.036 Smoke 1.74 (0.88-3.42) 0.110 0.91 (0.43-1.94) 0.813 Intake affected by
teeth health 2.01 (0.98-4.15) 0.059 1.62 (0.78-3.38) 0.196 Hire Caregiver 3.37 (0.85-13.36) 0.083 2.43 (0.59-10.07) 0.221 Note. Variables with p-value < 0.15 in univariate analysis were included. The definition of comorbidities included information from the NHIS and NHI.
a Hazard Ratio
b Confidence Interval
c adjusted Hazard Ratio
References
1. World Health Organization, Global tuberculosis report 2018. 2018.
2. Scobie, J., et al., Global AgeWatch Index 2015: insight report. 2015: HelpAge International.
3. Lonnroth, K., et al., Alcohol use as a risk factor for tuberculosis - a systematic review. BMC Public Health, 2008. 8: p. 289.
4. Rehm, J., et al., The association between alcohol use, alcohol use disorders and tuberculosis (TB). A systematic review. BMC Public Health, 2009. 9: p. 450.
5. Lonnroth, K., G. Roglic, and A.D. Harries, Improving tuberculosis prevention and care through addressing the global diabetes epidemic: from evidence to policy and practice. Lancet Diabetes Endocrinol, 2014. 2(9): p. 730-9.
6. Lee, P.H., et al., Tuberculosis and diabetes in low and moderate tuberculosis incidence countries. Int J Tuberc Lung Dis, 2018. 22(1): p. 7-16.
7. Lin, H.H., et al., Association between tobacco smoking and active tuberculosis in Taiwan: prospective cohort study. Am J Respir Crit Care Med, 2009. 180(5):
p. 475-80.
8. Gupta, K.B., et al., Tuberculosis and nutrition. Lung India, 2009. 26(1): p. 9-16.
9. Hochberg, N.S. and C.R. Horsburgh, Jr., Prevention of tuberculosis in older adults in the United States: obstacles and opportunities. Clin Infect Dis, 2013.
56(9): p. 1240-7.
10. Leung, C.C., et al., Smoking and tuberculosis among the elderly in Hong Kong.
Am J Respir Crit Care Med, 2004. 170(9): p. 1027-33.
11. Leung, C., et al., Diabetic Control and Risk of Tuberculosis: A Cohort Study.
Vol. 167. 2008. 1486-94.
12. Leung, C.C., et al., Passive smoking and tuberculosis. Arch Intern Med, 2010.
170(3): p. 287-92.
13. Chiang, C.Y., et al., The impact of national health insurance on the notification of tuberculosis in Taiwan. Int J Tuberc Lung Dis, 2002. 6(11): p. 974-9.
14. Lo, H.Y., et al., Completeness and timeliness of tuberculosis notification in Taiwan. BMC Public Health, 2011. 11: p. 915.
15. Lee, P.H., et al., Glycemic Control and the Risk of Tuberculosis: A Cohort Study.
PLoS Med, 2016. 13(8): p. e1002072.
16. Horton, K.C., et al., Sex Differences in Tuberculosis Burden and Notifications in Low- and Middle-Income Countries: A Systematic Review and Meta-analysis.
PLoS Med, 2016. 13(9): p. e1002119.
17. Lin, H.H., et al., Association of Obesity, Diabetes, and Risk of Tuberculosis: Two Population-Based Cohorts. Clin Infect Dis, 2018. 66(5): p. 699-705.
18. Pratt, R.H., et al., Tuberculosis in older adults in the United States, 1993-2008. J Am Geriatr Soc, 2011. 59(5): p. 851-7.
19. Monson, R.R., Observations on the healthy worker effect. J Occup Med, 1986.
28(6): p. 425-33.
20. Pearce, N., H. Checkoway, and D. Kriebel, Bias in occupational epidemiology studies. Occup Environ Med, 2007. 64(8): p. 562-8.
21. Lin, H.-H., et al., Exposure to Secondhand Smoke and Risk of Tuberculosis:
Prospective Cohort Study. PLOS ONE, 2013. 8(10): p. e77333.
22. Taiwan Centers of Disease Control. CDC Monitor. 2019; Available from:
https://daily.cdc.gov.tw/stoptb/CareMagChart.aspx.
23. Chan, T.-C., et al. Taiwan Infectious Disease Map. 2019; Available from:
http://ide.geohealth.tw/.
24. Taiwan Centers of Disease Control. Taiwan National Infectious Disease Statistics System. 2019; Available from: http://nidss.cdc.gov.tw/en/.
25. Harris, R.C., et al., Age-targeted tuberculosis vaccination in China and implications for vaccine development: a modelling study. The Lancet Global Health, 2019. 7(2): p. e209-e218.
26. Taiwan Centers for Disease Control. Taiwan National Infectious Disease Statistics System. 2019; Available from: http://nidss.cdc.gov.tw/en/.
27. Taiwan Centers for Disease Control. Statistics of Communicable Diseases and Surveillance Report. 2019; Available from: http://www.cdc.gov.tw/english/.