Outcomes of Tuberculosis contact tracing and predictors of success: a 10-year retrospective cohort analysis

ABSTRACT

Background: The Covid-19 pandemic risks disruption to diagnosis and treatment of Tuberculosis (TB) globally, jeopardising the 2035 eradication target. Between 1990-2010 40.9% of contacts did not complete Tuberculosis (TB) contact screening in Birmingham, UK. Understanding screening outcome success is urgently needed to guide future resource allocation. Aim: To evaluate changes made to TB screening since 2010. To identify predictors of contact screening non-completion, and of screening outcomes. Methods: A retrospective cohort analysis of all index and contact patients in Birmingham between 2011-2020, with separation of Covid-19 data, and stratification of contacts by Pulmonary TB (PTB) or Extra-Pulmonary TB (EPTB) index infection. Univariate and multiple logistic regression models were used to identify predictors of screening completion and clinical outcome. Results: 3,255 index cases and 27,820 contacts were identified. Screening non-completion has improved from 40.9% of contacts to 25% since 2010. Contacts were less likely to complete screening if they were >65 years (P=0.001) had no BCG (P<0.001), were male (P<0.001 PTB, P=0.02 EPTB), had had TB themselves (P<0.001 PTB, P=0.025 EPTB), were a close contact (P<0.001), or were from the Indian subcontinent (PTB only, P=0.019). Contacts were significantly more likely to require treatment for TB if they were born outside the UK (P<0.001), were <65 years (P<0.001 PTB, P=0.01 EPTB), if they were male, close contacts, or of Black ethnicity (all P=0.001, PTB only). Conclusions: Changes to contact screening since 2010 have substantially improved screening completion rates. Significant predictors of screening non-completion exist, which warrant further investigation and targeted screening support.