In November 2026, the WHO updated its Consolidated Guidelines on Tuberculosis Module 3 - Diagnosis with three major changes: a new class of near-point-of-care molecular tests (NPOC-NAATs) for sputum and tongue swabs, tongue swabs as a validated specimen type for low-complexity automated NAATs, and a sputum pooling strategy to stretch limited testing resources. The goal is to close a global diagnostic gap where roughly one in four TB patients never gets diagnosed. The WHO tuberculosis diagnosis guidelines 2026 aim to bring molecular testing closer to patients.

---

Why the WHO Tuberculosis Diagnosis Guidelines 2026 Matter Now

Getting an accurate, rapid TB and drug-resistance diagnosis is the gateway to starting treatment. Yet globally, about 25 percent of TB patients are never diagnosed or reported. That gap is widest where labs, trained staff, and cold-chain logistics are scarce. The 2026 WHO update tackles these barriers head-on by endorsing tools that work at the periphery - in primary care clinics, mobile units, and communities without conventional labs. These WHO tuberculosis diagnosis guidelines 2026 represent a major step forward for TB case finding worldwide.

---

Overview of Diagnostic Categories and Intended Use

Test Category	Primary Use	Specimen Types	Resistance Detection	Setting
NPOC-NAATs	Initial TB detection	Sputum, tongue swab	No (refer for DST)	No lab infrastructure needed
LC-aNAATs	Initial TB detection	Respiratory specimens, tongue swab	Yes (some platforms)	Low-complexity lab
Sputum pooling (LC-aNAATs)	Initial TB detection (resource-limited)	Pooled sputum (up to 4)	No (deconvolute if positive)	High-volume, low-positivity settings
TB culture + phenotypic DST	Drug resistance confirmation, treatment monitoring	Respiratory, other	Comprehensive	Reference lab
Smear microscopy	Treatment monitoring only	Sputum	No	Peripheral lab

Table 1. Diagnostic categories consolidated in the 2026 update, mapped to intended use, specimens, and setting requirements.

---

Update 1: New Near-Point-of-Care NAATs (NPOC-NAATs)

What Are NPOC-NAATs?

NPOC-NAATs are a newly defined class of nucleic acid amplification tests built for settings with no laboratory infrastructure. All reagents come pre-sealed in single-use disposable cartridges. The device needs only basic skills, at most one pipetting step (no precision pipetting), no more than three test steps, and runs on battery or UPS power at temperatures up to 50 degrees Celsius and 90 percent humidity. Results are reported automatically or manually.

The Pluslife MTB Nucleic Acid Test Card is the first product used to establish this category. All recommendations rest on evidence from this product; other brands need independent evaluation. The Pluslife MTB test card serves as the reference platform for the new NPOC-NAATs category.

Sputum NPOC-NAATs: Performance Data

Metric	Value	Evidence Base
Pooled sensitivity	85% (95% CI: 82-88%)	5 studies, 9 countries, 3,291 participants
Pooled specificity	98% (95% CI: 97-98%)	5 studies, 9 countries, 3,291 participants
vs. Smear microscopy: sensitivity difference	+22.0% (95% CI: +15.7 to +28.3%)	3 head-to-head studies
vs. Smear microscopy: specificity difference	-2.2% (95% CI: -2.9 to -1.5%)	3 head-to-head studies
vs. Xpert Ultra: sensitivity difference	-7.5% (95% CI: -11.6 to -3.5%)	5 head-to-head studies
vs. Xpert Ultra: specificity difference	+0.3% (95% CI: -1.2 to +1.8%)	5 head-to-head studies

Table 2. Sputum NPOC-NAATs diagnostic accuracy versus smear microscopy and Xpert Ultra.

Bottom line: Sputum NPOC-NAATs substantially outperform smear microscopy in sensitivity while keeping comparable specificity. They're moderately less sensitive than Xpert Ultra but bring molecular testing to places where Xpert can't operate. This Xpert Ultra comparison shows the trade-off between sensitivity and accessibility.

Tongue Swab NPOC-NAATs: Performance Data

Metric	Value	Evidence Base
Pooled sensitivity	76% (95% CI: 71-81%)	6 studies, 10 countries, 3,335 participants
Pooled specificity	99% (95% CI: 99-100%)	6 studies, 10 countries, 3,335 participants
Healthcare worker-collected: sensitivity	77% (70-82%)	6 studies, 2,501 participants
Healthcare worker-collected: specificity	99% (99-100%)	6 studies, 2,501 participants
Self-collected: sensitivity	75% (66-83%)	1 study, 834 participants
Self-collected: specificity	99% (98-100%)	1 study, 834 participants

Table 3. Tongue swab NPOC-NAATs diagnostic accuracy by collection method.

Important caveat: The evidence comes mainly from people who can produce sputum. Performance in patients who can't produce sputum - the very people who need tongue swabs most - hasn't been well validated. The tongue swab TB test offers a critical alternative for these patients.

Applicable Populations and Sample Collection

• Adults and adolescents with pulmonary TB signs/symptoms or positive screening

• HIV-positive individuals: Preferred strategy is simultaneous LC-aNAATs on respiratory and stool specimens plus urine LF-LAM. If LC-aNAATs unavailable, NPOC-NAATs may combine with LF-LAM. If no respiratory specimen obtainable, tongue swab NPOC-NAATs with LF-LAM.

• Children: Sputum/induced sputum recommendation extrapolated; tongue swab not recommended for extrapolation. Preferred pediatric strategy: simultaneous LC-aNAATs on respiratory and stool specimens (LF-LAM for HIV-positive). Pediatric TB diagnosis remains a challenge requiring dedicated research.

Sputum stays the preferred specimen because it's more accurate. Tongue swab collection requires:

1. Manufacturer-compatible swabs

1. Trained personnel or guided self-collection (limited self-collection data)

1. Standardized SOP with minimum 30-second collection time - shorter duration reduces sensitivity

1. Appropriate infection control and biosafety measures

Sputum collection quality directly impacts diagnostic accuracy, making proper technique essential.

Drug Resistance Testing Gap

Current NPOC-NAATs products don't provide rifampicin resistance results. All TB-positive specimens (leftover or newly collected) must be sent for drug susceptibility testing.

---

Update 2: Tongue Swab as New Specimen Type for LC-aNAATs

Recommendation Summary

• Respiratory specimens with LC-aNAATs: Strong recommendation, high certainty - initial diagnostic test instead of smear microscopy or culture

• Tongue swab with LC-aNAATs: Conditional recommendation, low certainty - when respiratory specimens cannot be obtained

Respiratory specimens include sputum (expectorated or induced), tracheal aspirate, and bronchoalveolar lavage. Tongue swab is explicitly not a respiratory specimen.

Respiratory Specimen LC-aNAATs Performance

Metric	Value	Evidence Base
Pooled sensitivity	90% (95% CI: 88-92%)	34 studies, 14,840 participants
Pooled specificity	95% (95% CI: 93-96%)	34 studies, 14,840 participants

Table 4. Respiratory specimen LC-aNAATs diagnostic accuracy.

Tongue Swab LC-aNAATs Performance

Metric	Value	Evidence Base
Pooled sensitivity	71% (95% CI: 67-75%)	11 studies, 10 countries, 6,124 participants
Pooled specificity	99% (95% CI: 98-99%)	11 studies, 10 countries, 6,124 participants
Healthcare worker-collected: sensitivity	70% (66-75%)	10 studies, 5,314 participants
Healthcare worker-collected: specificity	99% (97-99%)	10 studies, 5,314 participants
Self-collected: sensitivity	73% / 78%	2 studies, 810 participants
Self-collected: specificity	98% / 100%	2 studies, 810 participants

Table 5. Tongue swab LC-aNAATs diagnostic accuracy by collection method.

Applicable Populations

• Adults/adolescents with pulmonary TB signs/symptoms or positive screening, in any feasible setting

• Children: Tongue swab LC-aNAATs should not be used (limited evidence). Preferred: simultaneous LC-aNAATs on respiratory and stool specimens (LF-LAM for HIV-positive).

Drug Resistance Testing

For LC-aNAATs platforms without rifampicin resistance detection, all TB-positive patients should get DST. Even without tongue swab rifampicin resistance data, the guideline extrapolates the recommendation based on accuracy and equity.

---

Update 3: Sputum Pooling Strategy for Resource-Limited Settings

Recommendation

When resources rule out individual testing: pooled sputum (up to 4 specimens) tested with LC-aNAATs becomes the initial diagnostic strategy instead of individual LC-aNAATs testing. Conditional recommendation, high-certainty evidence.

How Pooling Works

1. Pool up to 4 liquefied sputum specimens into one test pool

1. Test pool per manufacturer instructions

1. Pool negative - no further testing (all 4 presumed negative)

1. Pool positive - test each specimen individually to find the positive ones

Pooled Testing Performance

Metric	Value	Evidence Base
Pooled sensitivity	85% (95% CI: 81-88%)	7 studies, 7 countries, 14,741 participants
Pooled specificity	98% (95% CI: 98-99%)	7 studies, 7 countries, 14,741 participants
Sensitivity loss vs. individual testing	-3.1% (95% CI: -7.0 to +0.8%)	7 studies
Community setting sensitivity reduction	-3% to -33%	4 studies

Table 6. Sputum pooling (1:4) LC-aNAATs diagnostic accuracy.

When NOT to Use Pooling

• HIV-positive individuals (limited evidence)

• Children (limited evidence)

• Individuals at increased drug resistance risk where pooling may reduce equity and increase harm

• Settings with TB positivity rate >24% - cost savings diminish

Key populations should be prioritized for individual testing per local risk assessment. Active case finding programs should avoid pooling in community settings.

---

Counter-Intuitive Finding: Tongue Swabs Nearly Match Healthcare Worker Collection

The near-equivalence between healthcare worker-collected and self-collected tongue swabs (for both NPOC-NAATs and LC-aNAATs) is striking. For NPOC-NAATs: 77% vs 75% sensitivity. For LC-aNAATs: 70% vs 73-78% sensitivity. Specificity stays at 99% across both methods. This suggests that with proper guidance, patients can collect their own specimens reliably - a potential game-changer for community screening and cutting healthcare worker exposure.

---

Practical Recommendations

1. Adopt NPOC-NAATs as a first-line test in peripheral clinics, mobile units, and primary care centers without lab infrastructure - it replaces smear microscopy with better sensitivity. This smear microscopy replacement strategy brings molecular testing to the periphery.

1. Use tongue swabs when you can't get sputum - validated for both NPOC-NAATs and LC-aNAATs in adults/adolescents; enforce the 30-second collection SOP.

1. Reserve Xpert Ultra/LC-aNAATs for settings with lab capacity - higher sputum sensitivity, plus rifampicin resistance detection on most platforms.

1. Implement sputum pooling (1:4) only in high-volume, low-positivity settings (<24% positivity) - verify the manufacturer supports pooling; deconvolute positive pools right away.

1. Always refer TB-positive specimens for DST when the initial test doesn't detect rifampicin resistance - applies to all NPOC-NAATs and some LC-aNAATs platforms.

1. Don't use tongue swabs for children - evidence is insufficient; use respiratory + stool LC-aNAATs instead.

1. Check the WHO prequalification webpage for current approved products - brand names were removed from the guidelines; only WHO webpages have updated product lists.

1. Train staff on infection control for tongue swab collection - it can generate aerosols, so biosafety measures are required.

---

Limitations

• Tongue swab evidence base: Mostly from people who can produce sputum; true performance in sputum-scarce populations is unknown.

• Self-collection data: Limited to one NPOC-NAATs study (834 participants) and two LC-aNAATs studies (810 participants).

• NPOC-NAATs product base: Recommendations hinge on a single product (Pluslife MTB Nucleic Acid Test Card); other products need independent evaluation.

• Pooling in community settings: Sensitivity drops up to 33%; not recommended for active case-finding in communities.

• Pediatric extrapolation: Based on limited data; dedicated pediatric studies are needed.

• Cost-effectiveness: Not systematically evaluated in the guideline; local health economics should guide implementation.

• HIV co-infection: Preferred algorithms combine multiple tests; real-world feasibility of simultaneous testing at scale is unproven.

---

Bottom Line

The 2026 WHO TB diagnosis guideline update delivers three practical tools to close the diagnostic gap: NPOC-NAATs bring molecular testing to the periphery, tongue swabs offer a viable alternative when sputum fails, and sputum pooling stretches limited test cartridges in high-volume settings. None is a silver bullet - each has specific population restrictions and evidence gaps - but together they expand the diagnostic toolkit for the settings that need it most. Success hinges on matching the right tool to the right setting, maintaining specimen quality (especially the 30-second tongue swab collection), and making sure every TB-positive result triggers reflex drug susceptibility testing. The WHO Module 3 diagnosis updates mark a turning point for global TB control. The WHO tuberculosis diagnosis guidelines 2026 provide a roadmap for countries to close the diagnostic gap. The WHO tuberculosis diagnosis guidelines 2026 also emphasize equity in access to these new tools. Each confidence interval in the evidence tables reflects the precision of the pooled estimates.

---

FAQ

What are NPOC-NAATs and how do they differ from existing molecular tests?

NPOC-NAATs are a new WHO-defined class of near-point-of-care nucleic acid amplification tests that operate without laboratory infrastructure. Unlike Xpert Ultra or other LC-aNAATs, they use pre-sealed single-use cartridges, require only basic skills and minimal pipetting, run on battery power, and tolerate temperatures up to 50 degrees Celsius and 90% humidity.

Which specimen types are validated for NPOC-NAATs?

Sputum and tongue swabs are validated for NPOC-NAATs in adults and adolescents. Sputum remains preferred due to higher sensitivity (85% vs 76%). Tongue swabs are recommended when sputum cannot be obtained, provided a standardized 30-second collection protocol is followed.

Can NPOC-NAATs detect rifampicin resistance?

No. Current NPOC-NAATs products do not provide rifampicin resistance results. All TB-positive specimens must be referred for separate drug susceptibility testing.

How accurate are tongue swabs compared to sputum for LC-aNAATs?

Tongue swab LC-aNAATs show 71% pooled sensitivity (95% CI: 67-75%) and 99% specificity (95% CI: 98-99%), versus 90% sensitivity and 95% specificity for respiratory specimens. Healthcare worker and self-collection yield similar accuracy.

Are tongue swabs recommended for children?

No. The guideline explicitly states tongue swabs (both NPOC-NAATs and LC-aNAATs) should not be used for children due to insufficient evidence. The preferred pediatric strategy is simultaneous LC-aNAATs on respiratory and stool specimens, with LF-LAM for HIV-positive children.

What is the sputum pooling strategy and when should it be used?

Up to four liquefied sputum specimens are combined into one pool and tested with LC-aNAATs. If the pool is negative, all four are considered negative. If positive, each specimen is tested individually. Use only in high-volume, low-positivity settings (<24% TB positivity) where individual testing is not feasible.

Does pooling reduce diagnostic sensitivity?

Yes. Pooled sensitivity is 85% (95% CI: 81-88%) versus individual testing, representing a 3.1% average reduction (95% CI: -7.0 to +0.8%). In community settings, sensitivity loss ranges from 3% to 33%, making pooling unsuitable for active case-finding in communities.

Which populations should not receive pooled testing?

HIV-positive individuals, children, and those at increased drug resistance risk where pooling could reduce equity and increase harm. These groups should receive individual testing prioritized by local risk assessment.

What happened to the brand names in the guidelines?

WHO removed all brand and product names that evolved with prequalification expansion. Readers are directed to WHO webpages for the current list of recommended and prequalified technologies. The Pluslife MTB Nucleic Acid Test Card is noted only as the index product for NPOC-NAATs category establishment.

How do the three web annexes organize diagnostic technologies?

Web Annex A covers technologies for initial TB detection. Web Annex B covers technologies for subsequent anti-TB drug resistance detection. Web Annex C covers technologies for TB infection detection. This replaces the previous scattered annex structure.

What is the global TB diagnostic gap these updates aim to close?

Approximately 25% of TB patients globally remain undiagnosed and unreported. The new tools target resource-limited settings where conventional laboratory infrastructure is absent, aiming to bring accurate molecular diagnosis closer to patients.

Can self-collected tongue swabs replace healthcare worker collection?

Evidence shows comparable accuracy: NPOC-NAATs self-collection sensitivity 75% (66-83%) vs 77% (70-82%) for healthcare worker collection; LC-aNAATs self-collection 73-78% vs 70%. However, self-collection data remains limited and requires standardized guidance and infection control measures.