WHO sets new standards for evidence behind tuberculosis diagnostics
WHO has published its first guidance on evidence generation for new tuberculosis diagnostics, a document aimed at tightening the path from laboratory innovation to policy adoption. The guidance is designed for researchers, developers, funders and other stakeholders who need to generate data that can support WHO guideline development on new TB tests, sample types and testing strategies.
The timing is clinically important. Diagnostic innovation in tuberculosis often moves faster than the evidence base needed to support real-world use, especially in settings where laboratory capacity is limited and the consequences of missed disease are severe. A test that performs well under ideal conditions can still fail to answer the questions that matter most at the bedside: who benefits, how the test fits into a care pathway, whether it is usable outside a tertiary center, and whether it can be deployed equitably.
Robust evidence is the hinge on which policy turns.
The guidance is the third in WHO’s series on evidence generation, following similar documents for new tuberculosis treatment regimens and for tuberculosis preventive treatment. In this case, the focus is diagnostic science: how to build studies that can actually inform global recommendations rather than simply add another accuracy estimate to the literature. WHO frames the document as a practical roadmap for work that can move from proof-of-concept to recommendations that countries can implement.
Diagnostic accuracy alone is not enough
The guidance places diagnostic accuracy within a broader clinical and programmatic frame. It addresses study design, participant selection criteria, sample size and reporting, and it extends beyond laboratory performance to outcomes that directly affect patients and health systems. Those outcomes include people-important health outcomes, values, cost-effectiveness, equity, acceptability and feasibility.
That breadth matters because tuberculosis diagnosis is rarely a one-step event. A test may be highly sensitive in a controlled study yet prove difficult to use in peripheral clinics, too expensive for routine deployment, or poorly suited to populations that carry the highest burden of disease. Evidence that stops at sensitivity and specificity can leave major unanswered questions about whether the test improves diagnosis, reaches vulnerable groups, or changes treatment decisions in a meaningful way.
WHO also introduces a new analytical framework that maps the role of a diagnostic across the clinical pathway and links that map to study design, population choice, methods and reporting. In practical terms, the framework forces investigators to think about where a test sits in the chain of care. A screening assay used at first contact in the community raises different questions from a confirmatory test used after referral, and a drug-resistance assay has a different evidentiary burden from a specimen-type study or a testing strategy comparison.
Study design gaps have limited past TB diagnostic evidence
WHO is explicit about the limitations that have weakened the evidence base for new TB diagnostics. Many studies have been methodologically incomplete, and too many have excluded vulnerable populations that stand to gain the most from better diagnostics. Others have focused narrowly on test accuracy while giving little attention to feasibility, acceptability, equity or cost. Even when data have been generated, reporting has often been incomplete or inconsistent, making studies difficult to combine and compare.
This pattern has direct clinical consequences. When study populations are too selective, the evidence may not reflect the people most likely to be tested in routine practice, including those facing barriers to care. When reporting is thin, systematic reviewers and guideline panels have less to work with, and promising technologies can stall despite biologic plausibility and technical performance. In a disease such as tuberculosis, where delayed diagnosis drives ongoing transmission and worse outcomes, evidence gaps are not a technical nuisance; they can shape access to care.
The guidance tries to solve that problem at the source by clarifying what should be measured and how it should be reported. It is written to help investigators design studies that are fit for policy use from the outset rather than trying to retrofit fragmented evidence after the fact.
Researchers and funders now have a more explicit template for what WHO needs before a new diagnostic can be judged ready for recommendation. That includes not only whether the test works, but also whether it can be used, scaled and accepted in the settings where tuberculosis is actually diagnosed.
Policy relevance extends beyond the laboratory
The document is also positioned alongside WHO’s target product profiles for tuberculosis diagnosis and detection of drug resistance, which together create a more structured path from innovation to implementation. Target product profiles define what an ideal or acceptable diagnostic should achieve; the new evidence-generation guidance explains how to produce the data needed to judge whether a candidate test meets those expectations.
This pairing is important for developers as well as policy makers. Diagnostic innovation often fails when product development and evidence planning happen on separate tracks. A test can be technically elegant and still be hard to adopt if the supporting evidence does not speak to the realities of health systems, costs, patient experience and equity. By defining the evidentiary requirements early, WHO is trying to reduce that disconnect.
Tereza Kasaeva, Director of the Department for HIV, Tuberculosis, Hepatitis and Sexually Transmitted Infections, said the guidance gives researchers, developers and funders a clear road map for generating the evidence WHO needs on innovative tuberculosis tests, sample types and testing strategies so better diagnostics can reach more people faster. The message behind that statement is straightforward: without better evidence, promising diagnostics can remain stranded between the bench and the clinic.
The remaining challenge is operational. The guidance lays out what strong evidence should look like, but the pace and quality of future TB diagnostic innovation will depend on whether investigators and sponsors build those standards into study design before the first patient is enrolled. The document itself does not answer how quickly that shift will happen, or whether upcoming studies will finally close the longstanding gaps in feasibility, equity and reporting that have limited the field for years.