What patient factors predict physicians' decision not to treat latent tuberculosis infection in tuberculosis contacts?

PLoS One. 2013 Sep 30;8(9):e76552. doi: 10.1371/journal.pone.0076552. eCollection 2013.


Objective: The study aimed to determine factors that are associated with physicians' decision to offer treatment for latent tuberculosis infection (LTBI) in contacts of patients with tuberculosis.

Methods: We performed a nested case-control study in a cohort of contacts of patients with pulmonary tuberculosis who had a tuberculin skin test (TST) ≥ 10 mm. Cases were those who were offered treatment for LTBI. Controls were randomly selected from those who were not offered treatment for LTBI by the reviewing physician. Odds ratios were estimated by multivariate logistic regression.

Results: There were 195 cases and 279 controls. The following factors were significantly (positively or negatively) associated with being offered LTBI treatment in the multivariate analysis: female gender (OR 2.9; 95% CI 1.6-5.5), TST conversion (OR 3.9; 2.0-7.9), TST > 20 mm (OR 4.1; 1.8-9.1, for TST of 21-30 mm and OR 7.9; 2.6-23.8, for TST >30 mm), sputum smear positive index case (OR 12.7; 4.5-36.1), being overseas-born and immigration more than 2 years ago (OR 0.1; 0.06-0.3), being a health care worker (OR 0.2; 0.1-0.6), being a non-household contact of the TB index case (OR 0.3; 0.2-0.6) and age >35 years (OR 0.2; 0.1-0.5 for age 35 to 54.9 years and OR 0.04; 0.01-0.2 for age ≥55 years). Previous BCG vaccine and chest x-ray findings were not significantly associated with physicians' decision to offer treatment for LTBI.

Conclusions: Most factors that influenced physicians' decisions on treatment for LTBI were based on evidence of an association with risk of developing TB or risk of having an adverse reaction to treatment for LTBI. However, the decreased likelihood of offering treatment for LTBI to people born overseas, men and health care workers, was apparently not based on any evidence of risk. Efforts should be made to ensure that these groups are given access to treatment for LTBI.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Case-Control Studies
  • Cohort Studies
  • Contact Tracing
  • Female
  • Humans
  • Latent Tuberculosis / drug therapy*
  • Logistic Models
  • Male
  • New South Wales
  • Odds Ratio
  • Practice Patterns, Physicians' / statistics & numerical data*
  • Risk Assessment / methods*
  • Sex Factors
  • Tuberculin Test
  • Tuberculosis / prevention & control*

Grant support

CCD is supported by a University of Sydney Postgraduate Award Scholarship. There was no other study funding. The funders of the research scholarship had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.