Longevity loss among cured tuberculosis patients and the potential value of prevention

S. Hoger, K. Lykens, S. F. Beavers, D. Katz, T. L. Miller

Research output: Contribution to journalArticlepeer-review

15 Scopus citations

Abstract

BACKGROUND: Evidence of substantial, quantifiable and preventable burdens of mortality hazard even after anti-tuberculosis treatment and cure would be a compelling, concrete, and useful measure of the value of prevention.

METHODS: We compared years of potential life lost between a cohort of 3 933 cured tuberculosis (TB) patients and 9 166 persons with latent tuberculous infection. We constructed a regression model to predict the expected years of potential life lost in each cohort and for demographic subgroups.

RESULTS: Among decedents, a history of fully treated TB is associated with a predicted average 3.6 more years of potential life loss than a comparable population without active TB. Greater longevity losses were predicted among those identified as White and Hispanic than among Black and Asian counterparts.

CONCLUSION: We found significant differences in predicted longevity of treated TB survivors relative to a similar group without active TB. These excess losses are substantial: a total of 14 158 life-years or the equivalent of more than 188 75-year lifespans. These findings illustrate an important opportunity cost associated with each preventable TB case - an average of 3.6 potential years of life. We conclude that substantial preventable mortality burdens remain despite adequate anti-tuberculosis treatment, a compelling rationale for more widespread and systematic use of prevention.

Original languageEnglish
Pages (from-to)1347-1352
Number of pages6
JournalInternational Journal of Tuberculosis and Lung Disease
Volume18
Issue number11
DOIs
StatePublished - 1 Nov 2014

Keywords

  • Mortality
  • Premature death
  • Regression

Fingerprint Dive into the research topics of 'Longevity loss among cured tuberculosis patients and the potential value of prevention'. Together they form a unique fingerprint.

Cite this