Predicting Risk of Tuberculosis (TB) Disease in People Who Migrate to a Low-TB Incidence Country: Development and...

Area of Science:

• Epidemiology
• Public Health
• Biostatistics

Background:

• Tuberculosis (TB) incidence is notably higher in migrants to low-incidence countries like Canada.
• Current TB screening and prevention programs for migrants face cost challenges.
• There is a need for effective risk stratification tools to guide TB screening for foreign-born populations.

Purpose of the Study:

• To develop and validate a TB risk-prediction model for foreign-born permanent residents in Canada.
• To inform TB screening decisions and optimize resource allocation for TB programs.
• To improve TB risk stratification in migrant populations resettling in low-incidence settings.

Main Methods:

• A proportional baselines landmark supermodel was developed using health administrative data from British Columbia and Ontario.
• The model incorporated demographic, TB exposure, and medical covariates.
• Model performance was assessed using discrimination and calibration statistics in a separate validation cohort.

Main Results:

• The study included over 1.6 million individuals across two Canadian provinces.
• The model demonstrated good predictive ability with 2- and 5-year concordance statistics of 0.77 in the validation cohort.
• Calibration-in-the-large values indicated acceptable model accuracy for predicting TB risk.

Conclusions:

• A validated TB risk-prediction model is available online (https://tb-migrate.com).
• This model can enhance TB risk stratification for migrants in low-incidence countries.
• The tool has the potential to inform TB screening policies and guidelines for immigrant populations.