Transnational Record Linkage for Tuberculosis Surveillance and Program Evaluation

Abstract

Objective. Pre-immigration tuberculosis (TB) screening, followed by post-arrival rescreening during the first year, is critical to reducing TB among foreign-born people in the United States. However, existing U.S. public health surveillance is inadequate to monitor TB among immigrants during subsequent years. We developed and tested a novel method for ascertaining post-U.S.-arrival TB outcomes among high-TB-risk immigrant cohorts to improve surveillance. Methods. We used a probabilistic record linkage program to link pre-immigration screening records from U.S.-bound immigrants from the Philippines ( n=422,593) and Vietnam ( n=214,401) with the California TB registry during 2000–2010. We estimated sensitivity using Monte Carlo simulations to account for uncertainty in key inputs. Specificity was evaluated by using a time-stratified approach, which defined false-positives as TB records linked to pre-immigration screening records dated after the person had arrived in the United States. Results. TB was reported in 4,382 and 2,830 people born in the Philippines and Vietnam, respectively, in California during the study period. Of these TB cases, records for 973 and 452 cases of people born in the Philippines and Vietnam, respectively, were linked to pre-immigration screening records. Sensitivity and specificity of linkage were 89% (90% credible interval [CrI] 83, 97) and 100%, respectively, for the Philippines, and 90% (90% CrI 83, 98) and 99.9%, respectively, for Vietnam. Conclusion. Electronic linkage of pre-immigration screening records to a domestic TB registry was feasible, sensitive, and highly specific in two high-priority immigrant cohorts. Transnational record linkage can be used for program evaluation and routine monitoring of post-U.S.-arrival TB risk among immigrants, but requires interagency data sharing and collaboration.