Alternative approaches to tuberculosis diagnostics research: methods for estimating the incremental value of new tests

Tuberculosis (TB) remains a global health problem. Diagnosis is the critical first step for control of TB, and several promising tests have been developed, including the interferon-gamma release assay (IGRA). Unfortunately, TB diagnostics research is still focused on measures of test accuracy (i.e. sensitivity and specificity). There are limited data on the incremental value of new tests over and above conventional tests and their impact on clinical management. Test accuracy data, while necessary, are only surrogates for patient-important outcomes and cannot provide high quality evidence for policy-making. This manuscript-based PhD thesis focused on alternative approaches to evaluate the incremental value of new tests, with and without a gold standard. In the first manuscript, we performed a secondary data analysis of a study on 528 children evaluated for active TB in Cape Town, South Africa. Using TB culture as the gold standard, we measured the incremental value of the IGRA beyond patient demographics, clinical signs and conventional TB tests using the area under the receiver operating characteristic curve (AUC) as well as two newly-described measures based on risk probability: net reclassification improvement (NRI) and integrated discrimination improvement (IDI). All analyses showed that the IGRA did not have added value beyond clinical data and conventional tests for the diagnosis of active TB in hospitalized, smear-negative children. The use of multivariable analysis provided a useful approach to evaluate the incremental value of this new test as part of the diagnostic algorithm, rather than in isolation.In the second manuscript, we developed a methodology for estimating the incremental value of a test when no gold standard exists and true disease status is unknown, such as in the case of latent TB infection (LTBI). Using a Bayesian framework for latent class model estimation, we validated our proposed methods in a series of simulations and then applied these methods to calculate the AUC, NRI and IDI to measure the added value of the IGRA over the tuberculin skin test (TST) for diagnosis of LTBI in different settings. We showed that the magnitude of the AUC and IDI behaved as expected when we changed the true accuracy of the new test using simulated data. Furthermore, we showed that the added value of the new test decreased when conditional dependence between the new and standard tests was taken into account.Finally, the third manuscript was a primary data collection study at the Montreal Children's Hospital (MCH), which recently began implementing the IGRA for children with specific clinical indications. The aim of this study was to assess the impact of the IGRA on clinical management by asking pediatric respirologists to document how the IGRA result changed, if at all, their initial diagnostic and treatment decisions based on the TST and other available data in clinically-relevant subgroups. Our study of 399 children showed that pediatric respirologists used negative IGRA results to withhold preventive therapy in most low-risk children who were found through targeted screening programs and referred for a positive TST result. In contrast, in almost all TST-positive children who were evaluated as TB contacts, negative IGRA results did not change clinical management.While new technologies in the diagnostics pipeline offer great promise for TB control, limited resources mandate that we evaluate them in clinically-meaningful ways before their implementation into routine practice. This PhD thesis addressed the need for incremental value and clinical impact studies and offers insights into the comparative benefits and limitations of the various methods used.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.