Purpose: The Health Utilities Index Mark 3 (HUI3) and Short Form-6D (SF-6D) are widely used health-related quality of life instruments that allow the calculation of quality adjusted life years (QALYs). The HUI3 accounts for more health states than the SF-6D (972,000 versus 18,000), but lacks domains captured by the SF-6D such as social functioning. In economic evaluations of tuberculosis (TB) control strategies, differences in how well the HUI3 and SF-6D discriminate between health states may lead to differences in cost-effectiveness estimates which are difficult to predict. Our objective was to examine the impact of using HUI3 utility values, compared to SF-6D values, on estimates of the incremental cost-effectiveness ratio (ICER) for providing standard latent tuberculosis infection (LTBI) treatment to contacts of active TB patients, versus no treatment. Methods: We administered the HUI3 and SF-6D questionnaires to TB patients recruited through British Columbia Centre for Disease Control TB clinics. We then incorporated each set of utility values into separate Markov model-based evaluations of the incremental cost per QALY gained of offering LTBI treatment (versus not) to tuberculin skin test-positive TB contacts. Uncertainty in the ICER estimates was assessed using probabilistic analysis. Results: In the LTBI patients, the mean HUI3 utility value, 0.90 (N = 93, SD = 0.19), and the mean SF-6D value, 0.82 (N = 91, SD = 0.12), were significantly different (p < 0.0001). In the active TB patients, the mean HUI3 value, 0.76 (N = 69, SD = 0.30), and the mean SF-6D value, 0.70 (N = 66, SD = 0.16), were not significantly different (p = 0.38). The mean utility values between LTBI and active TB groups were significantly different whether measured with the HUI3 (p = 0.0011) or the SF-6D (p < 0.0001). Using HUI3 or SF-6D utilities, LTBI treatment dominated no treatment. The HUI3-based ICER was greater than the SF-6D-based ICER by $1.72 per QALY gained. Conclusion: The choice between using HUI3- or SF-6D-derived utility values does not have an important impact on the estimated ICER of providing LTBI treatment to TB contacts compared to no treatment.