World Health Organization · 2nd edition, 2025 · 126 pages
Standardized methods for national TB household cost surveys to measure and mitigate catastrophic financial burden on patients and families.
- Issuer
- World Health Organization
- Edition
- 2nd edition, 2025
- Audience
- National TB programmes, health economists, researchers, public health professionals, policy-makers in health and social protection sectors
- Evidence
- Normative guidance based on systematic review of patient cost surveys (2014), field-testing in 10 countries (2015–2017), experience from 37 national surveys (2015–2024), and expert consensus through Global Task Force meetings
- Reviewed
- 16 Oct 2026 · GMJ
What this guideline says
GMJ summary
This WHO normative guidance provides a comprehensive methodology for conducting national surveys to measure the costs incurred by people treated for tuberculosis and their households. The guidance updates the 2017 edition with revised terminology (replacing “patient cost surveys” with “household cost surveys”), expanded social protection assessment, and refined analytical approaches based on experience from 37 countries that completed surveys between 2015–2024.
The document describes a cross-sectional, facility-based survey design that measures direct medical costs, direct non-medical costs (transport, food, accommodation), and indirect costs (income loss) from TB symptom onset through treatment completion. The primary objective is to determine the proportion of TB-affected households experiencing catastrophic total costs, defined as costs exceeding 20% of annual household income—a key indicator in the WHO End TB Strategy target that no TB-affected households face catastrophic costs by 2030.
Beyond measurement, the guidance emphasizes translating survey findings into policy action through multisectoral stakeholder engagement. It provides detailed protocols for sampling (typically 25+ clusters with 10–40 participants each), data collection using electronic or paper questionnaires (approximately 60-minute interviews), analysis adjusting for clustering effects, and dissemination to inform health financing reforms, social protection expansion, and service delivery improvements to reduce the economic burden of TB on households.
Key points
For clinicians — doses and thresholds as stated in the guideline
- Survey population includes all people on TB treatment (DS-TB or DR-TB) in the NTP network who are ≥14 days into their current treatment phase, interviewed once in a cross-sectional design
- Recommended sample size typically requires ≥25 clusters with 10–40 participants per cluster using probability-proportional-to-size cluster sampling; simple random sampling or 100% sampling may be used in small countries
- Catastrophic total costs defined as direct costs (medical and non-medical) plus indirect costs (income loss) exceeding 20% of annual household income before TB onset; sensitivity analyses should explore 30% and 60% thresholds
- Direct costs include medical payments (consultations, diagnostics, medicines) and non-medical payments (transport, food, accommodation, nutritional supplements), net of reimbursements, from symptom onset through treatment completion
- Indirect costs estimated using output approach (self-reported income loss) for main analysis and human capital approach (time lost valued at hourly wage) for sensitivity analysis; caregiver time and costs should be included
- Annual household income estimated through self-reported income (most common in formal work sectors), consumption/expenditure (preferred in informal sectors), or regression based on asset ownership and dwelling characteristics
- Survey budget ranges US$25,300–US$346,726 (average US$138 per participant) with typical data collection period of 2–3 months; technical assistance, interviewer training, and ethics approval required before starting
- Results from 35 countries show 13–92% of households (pooled average 49%) experienced catastrophic total costs; 82% for DR-TB households, demonstrating need for health financing reforms and social protection interventions
Recommended methods for handling missing or inaccurate income data during analysis (Table 6.1)
Table extracted from the guideline
| Cost estimate | Data missing/requiring substitution | Employment status | Recommended action |
|---|---|---|---|
| Indirect costs – Output approach (actual income loss) | Household income before onset of TB, at diagnosis or during treatment | Employed before onset of TB | Use average household income for the three time periods based on data from other survey participants |
| Indirect costs – Output approach | Household income before onset of TB, at diagnosis or during treatment | Not employed before onset of TB | No substitution needed (replace missing values with 0) |
| Indirect costs – Human capital approach (time lost) | Hourly income of adult survey participant or caregiver of child | Employed before onset of TB | Use average wage from other survey participants, ideally from same (formal/informal) sector |
| Indirect costs – Human capital approach | Hourly income of adult survey participant or caregiver of child | Not employed before onset of TB | Use hourly national minimum wage |
| Indirect costs – Human capital approach | Hourly income of caregiver of adult survey participant | Information not collected | Use hourly national minimum wage |
| Total annual household income (denominator) | All household income data | Not relevant | Use regression to estimate income based on household assets and dwelling characteristics from other participants |
Source: World Health Organization, Consolidated guidance on tuberculosis data generation and use: Module 4: Surveys of costs faced by households affected by tuberculosis. Reproduced in summary form.
What changed in this edition
Versus the previous edition
- Revised terminology throughout: "TB patient cost surveys" changed to "TB household cost surveys" and "TB patients" to "people treated for TB" to reflect person-centred language
- Added third survey objective: to determine use of existing social protection policies and recommend refinements based on survey results, with new section on health-related TB determinants (Chapter 2)
- Specified that costs from pre-diagnosis, diagnosis and treatment phases should be collected from all participants regardless of current treatment phase, with note to seek equal representation of intensive versus continuation phase participants during sampling
- Added guidance to explore costs when other household members are concurrently on TB treatment or TB preventive treatment, with estimates used in sensitivity analysis to guide future survey design
- Clarified that indirect costs should use output approach (self-reported income loss) for main analysis and human capital approach (time valuation) for sensitivity analysis, with caregiver time and costs included in both
- Expanded guidance on handling missing or inaccurate data during analysis, including methods to estimate household income using regression on asset ownership when income data unavailable (Table 6.1)
- Added weighting procedures for stratified cluster sampling (e.g., by drug-resistance status or urban/rural) and for unbalanced samples between treatment phases
- Included summary of data collection tools and platforms used in previous surveys (predominantly Ona platform for e-surveys since 2017, replacing paper-based approaches)
- Enhanced ethical considerations section with procedures for obtaining assent from child participants and protocol when written consent cannot be obtained
- Expanded Chapter 10 with concrete examples of policy changes resulting from survey findings in Ghana (health insurance enrollment), Lao PDR (nutritional support), and Viet Nam (social health insurance integration)
- Updated budget information based on 20 surveys 2015–2021: range US$25,300–US$346,726, average US$138 per participant, with detailed breakdown by WHO region
- Added analytical guidance on adjusting for intra-cluster correlation using logistic regression with robust standard errors, and weighting for incomplete cluster sampling
From the guideline
Verbatim
“Evidence from 35 TB household cost surveys for which results have been reported to WHO shows that the percentage of households that faced catastrophic total costs (defined as direct and indirect costs >20% of annual household expenditure or income) due to TB ranged from 13% to 92%, with a pooled average for all 35 countries, weighted for each country's number of notified cases, of 49%; the percentage was much higher for households affected by drug-resistant TB, with a pooled average of 82%.”
Editor’s perspective
Why this matters for Georgia and the region
Prof. Giorgi Pkhakadze, MD, MPH, PhD
Editor-in-Chief, Georgian Medical Journal · Chair, Public Health Institute of Georgia · WHO consultant, health and migration
As someone who has worked extensively on Georgia's health system reforms and TB programme strengthening, I recognize that this guidance addresses a critical gap in our understanding of the true burden of TB on Georgian families. While our National TB Programme provides nominally free diagnosis and treatment, and Georgia has made significant progress in UHC implementation since 2013, we lack systematic data on the out-of-pocket costs and income losses that TB patients and their households actually bear—costs that likely drive many into poverty despite official policies. The guidance's recommendation to use the output approach (measuring actual income loss) rather than only time valuation is particularly important for Georgia's context, where informal employment is substantial and wage-based calculations may underestimate true economic impact. The most actionable element for Georgian health authorities is the framework for multisectoral stakeholder engagement: our TB household cost survey should not be conducted in isolation by the TB programme, but rather involve the Ministry of Labour, Health and Social Affairs, the Social Service Agency, and the State Fund for Social Insurance from the outset, so that findings immediately translate into policy dialogue on expanding social protection coverage—whether through the existing poverty-targeted social assistance programme, targeted cash transfers, or transport voucher schemes. One realistic implementation challenge is that Georgia's TB notification numbers (approximately 2,000–2,500 annually) are relatively small compared to high-burden countries, which may require us to use 100% sampling or extended data collection periods rather than cluster sampling, and we must ensure that survey findings disaggregate costs for vulnerable populations including people living with HIV, migrants, and those in conflict-affected areas, where financial barriers may be most severe yet least visible in routine programme data.
ORCID 0000-0001-7609-4515 · Perspective written for the GMJ Guidelines Library, 16 Oct 2026
Applicability
Georgia and the South Caucasus
This guidance has direct relevance for Georgia and neighboring South Caucasus countries (Armenia, Azerbaijan) as well as Eastern Europe, where TB remains a public health concern with drug-resistant TB rates above global averages. Georgia’s National TB Programme, operating within the country’s Universal Health Coverage framework established in 2013, would benefit from implementing this survey methodology to quantify the financial burden on TB-affected households and identify gaps in existing social protection coverage. Given Georgia’s mixed formal-informal employment structure, the guidance’s flexibility in measuring household income through multiple approaches (self-reported income, consumption/expenditure, or asset-based estimation) is particularly applicable. Countries in the region should consider that while TB diagnosis and treatment may be nominally free through national programmes, this survey methodology would reveal the often-hidden costs of transport, time loss, and nutritional needs that drive catastrophic expenditure. Implementation would require coordination between Ministries of Health, Labour and Social Affairs, and engagement with the Social Service Agency to map existing social protection mechanisms (disability benefits, poverty-targeted assistance) and assess whether TB-affected households can access them. The estimated budget of US$138 per participant suggests a national survey in Georgia (with approximately 2,000–2,500 TB notifications annually) would require US$70,000–100,000, feasible within national TB programme or Global Fund allocations.
Questions this guideline answers
Quick answers
What is the definition of catastrophic total costs due to TB and how does it differ from the SDG indicator on catastrophic health expenditures?
Catastrophic total costs due to TB are defined as direct costs (medical and non-medical) plus indirect costs (income loss) that exceed 20% of annual household income before TB onset, measured specifically for TB-affected households from symptom onset through treatment completion. This differs from SDG indicator 3.8.2, which measures only direct out-of-pocket health expenditures (for all health conditions, not TB-specific) exceeding 10% or 25% of total household expenditure for the general population. The TB indicator includes indirect costs (lost income), non-medical costs (transport, food, accommodation), and focuses only on households with a TB-diagnosed member receiving treatment in the NTP network, making it incomparable to the broader SDG catastrophic health expenditure measure.
What is the recommended sampling approach and minimum sample size for a national TB household cost survey?
Cluster sampling using probability-proportional-to-size (PPS) method is recommended for most countries, with a minimum of 25 clusters and 10–40 participants per cluster. Clusters are typically health facilities treating TB, selected based on their TB notification numbers. The exact sample size depends on the expected proportion experiencing catastrophic costs (often assumed 20–30% for DS-TB, 50% for DR-TB), desired precision (typically ±5%), and design effect from clustering. For small countries with few TB treatment facilities, 100% sampling of all eligible patients attending all facilities during a defined period may be more appropriate. Simple random sampling from electronic TB registers is theoretically attractive but has not been widely tested. Stratified sampling by drug-resistance status (DS-TB vs DR-TB) or geography (urban vs rural) can improve representation of important subgroups but requires larger overall sample sizes.
How should indirect costs (income loss) be measured and which approach should be used for calculating the End TB Strategy indicator?
Indirect costs should be estimated using two approaches: (1) the output approach, which uses self-reported household income at three time points (before TB onset, at diagnosis, during treatment) to calculate actual income loss, and (2) the human capital approach, which values time lost seeking care by multiplying hours spent by the individual's hourly wage. The output approach should be used for the main analysis of the End TB Strategy indicator because it represents actual income protection—what households value most and what social protection interventions aim to preserve. The human capital approach should be explored in sensitivity analysis. Both approaches should include caregiver time and costs for those accompanying the person with TB. For individuals without reported income, the output approach uses zero (no substitution), while the human capital approach uses national minimum wage to value time lost. Neither approach fully captures all indirect costs (e.g., reduced productivity while working, time of other family members caring for children), so results should be presented as a range.
What are the main policy actions that countries have taken based on TB household cost survey findings?
Countries have implemented three main types of policy responses based on survey findings: (1) Health service delivery changes—reducing hospitalization duration, decentralizing diagnostic services closer to patients, strengthening community-based TB care (implemented in Mongolia, Philippines, Uganda, Papua New Guinea); (2) Health financing reforms—fast-tracking TB patients into national health insurance schemes without premium payment (Ghana, Kenya), subsidizing insurance premiums (Viet Nam), integrating TB services into insurance benefit packages (Lao PDR, Myanmar, Philippines, Thailand, Viet Nam); and (3) Social protection expansion—enhancing nutritional support for TB patients (Democratic Republic of Congo, Fiji, Indonesia, Kenya, Mali, Nigeria, Zimbabwe), expanding TB-specific cash transfers or vouchers, linking TB patients to existing non-TB social protection programs (Brazil, Ghana, Indonesia, Uganda), and strengthening job protection legislation (Brazil, Indonesia, Kenya, Mongolia, Myanmar, Nigeria). Viet Nam's experience demonstrates comprehensive action: survey findings showing 63% catastrophic cost rates led to full TB service integration into Social Health Insurance by 2022 and creation of the PASTB charity providing financial support to nearly 9,000 patients.
Related on GMJ
Interconnected resources
Quality & accreditation
Other guidelines on this topic
- Consolidated guidance on tuberculosis data generation and use: Module 3 – National tuberculosis prevalence surveys
- Social and behaviour change interventions for contraception and family planning
- Responding to the health financing emergency: immediate measures and longer-term shifts
- Guidance on evidence generation on new regimens for tuberculosis preventive treatment
- Guiding principles for prioritizing malaria interventions in resource-constrained country contexts to achieve maximum impact, 2nd ed.
- Issuer
- World Health Organization
- Type
- Guidance (normative)
- Edition
- 2nd edition, 2025
- Replaces
- Tuberculosis patient cost surveys: a handbook, 2017
- Status
- Current
- Published
- 2025-06-19
- Languages
- English
- Scope
- Global, all settings with TB treatment programmes
- Audience
- National TB programmes, health economists, researchers, public health professionals, policy-makers in health and social protection sectors
- Pages
- 126
- Licence
- CC BY-NC-SA 3.0 IGO
- ISBN
- 978-92-4-010457-0
- Evidence
- Normative guidance based on systematic review of patient cost surveys (2014), field-testing in 10 countries (2015–2017), experience from 37 national surveys (2015–2024), and expert consensus through Global Task Force meetings
- Next review
- Not specified
- Record history
- Created 16 Oct 2026 · updated 16 Oct 2026 · next check Apr 2027
World Health Organization. Consolidated guidance on tuberculosis data generation and use. Module 4. Surveys of costs faced by households affected by tuberculosis. Geneva: World Health Organization; 2025. Licence: CC BY-NC-SA 3.0 IGO.GMJ Guidelines Library records are original summaries written by the GMJ editorial team from the issuer’s published document and reviewed by a physician. They do not replace the full guideline. The document is the copyright of its issuer and is linked, not hosted. Spotted an error or a newer edition? Tell the Guidelines Desk.
Was this article helpful?
