The third national TB prevalence survey of Cambodia, 2023: results and implications
General background and context
To reliably track the burden of tuberculosis (TB) disease from subnational to global levels, the ultimate goal is for all countries to rely on data routinely collected through a) national disease surveillance systems and b) national vital registration (VR) systems in which causes of death are coded according to the international classification of diseases (ICD). In countries with a relatively high burden of TB disease that do not yet have national disease notification systems or national VR systems with cause-of-death data that are of sufficiently high quality and coverage, national TB prevalence surveys are the best way to directly measure the burden of TB disease in the population (Section 1.4) (1).
National TB prevalence surveys can provide a reliable measurement of TB burden in a population at a given point in time. Repeat surveys allow assessment of how this burden has changed over time, including in response to interventions. Survey data also provide important insights that can help national TB programmes (NTPs) to identify ways to improve TB diagnosis and treatment, and to quantify and correct any underreporting of people diagnosed with TB through national disease surveillance systems.
National TB prevalence surveys typically focus on measuring the number of people aged 15 years and older who have bacteriologically confirmed pulmonary TB disease. This is in recognition of the limitations of available diagnostics, the realities of field operations and the relatively low rates of TB disease (per 100 000 population) in children and young adolescents (1,2). Results are then adjusted to estimate the overall prevalence of TB disease in the whole population (for all ages and including extrapulmonary TB disease).
Between 2007 and August 2025, a total of 36 national TB prevalence surveys in 32 countries (17 in Africa and 15 in Asia) were implemented using the screening and diagnostic methods recommended by the World Health Organization (WHO) (Section 1.4). During this period, five countries implemented repeat surveys: Cambodia, China, Myanmar, the Philippines and Viet Nam.
Cambodia is the first country to complete three national TB prevalence surveys this century
Cambodia, a lower-middle-income country, conducted its third national TB prevalence survey from June 2023 to May 2024, following surveys in 2002 and 2011 (3-5). In doing so, it became the first country to have successfully conducted three national TB prevalence surveys this century. The surveys span three global TB strategies: the DOTS Strategy (1994–2005) (6), the Stop TB Strategy (2006–2015) (7) and the End TB Strategy (2016–2035) (8). Cambodia is also one of the first countries to implement WHO’s most up-to-date guidance for prevalence surveys (1), using digital chest X-ray (CXR) with computer-aided detection (CAD), real-time digital data management and testing with Xpert® MTB/RIF Ultra (Xpert Ultra) (alongside liquid and solid culture) to optimize speed, accuracy and efficiency. Survey methods are summarized in Box 1 (below).
Strong, sustained reductions in TB burden in Cambodia over the past 2 decades
Together, Cambodia’s prevalence surveys confirm a large decline in TB burden between 2002 and 2023, with prevalence falling about 50% per decade over this period (Fig. 1). The estimated prevalence of TB in 2023 was 546 per 100 000 population (95% uncertainty interval [UI]: 461–631) for those aged 15 years and older. Based on all three surveys, in 2024, the estimated TB incidence rate was 272 (95% UI: 205–365) per 100 000 population (all ages), and the estimated mortality rate was 18.2 (9.9–30.1) per 100 000 population (Fig. 2). Moreover, despite substantial disruptions to notifications during the coronavirus disease (COVID-19) pandemic, Cambodia saw a 20% reduction in the annual TB incidence rate between 2015 and 2023, and a 25% reduction in the annual TB mortality rate during this period (Fig. 2).
Other key survey results are summarized in Box 2 (below).
Fig. 1 Prevalence of bacteriologically confirmed pulmonary TB among people aged ≥15 years in Cambodia, as measureda in three national TB prevalence surveys: 2002, 2011, 2023
Fig. 2 Estimated incidence and mortality of Cambodia, 2000–2024
(a) Incidence
(b) Mortality
Reasons for these large declines in TB disease burden are likely to be multifactorial, with potental drivers including: improvements in TB services, broader socioeconomic development and demographic change.
From 2002 onwards, the coverage and quality of TB services in Cambodia improved substantially. TB diagnosis and treatment was decentralized from hospitals and integrated into primary health care at clinic level. Standard 6-month and fixed-dose-combination treatment regimens were introduced. CXR services were expanded to become available in almost all districts, and active case-finding using digital CXR and Xpert assays was also established.
Broader transformational changes include a transition from a predominantly agrarian cash-based economy to an industrial society; improved communication and transportation infrastructure (which have also facilitated provision of and access to health care services); and improved educational services. These developments are evident in many of the Sustainable Development Goal (SDG) indicators associated with TB (Fig. 3).
In terms of demographic drivers, the past 2 decades have seen a decline in the proportion of the population aged 65 years and older. This trend includes a reduction in the number of survivors from the Khmer Rouge era (mid–late 1970s), who may have had a high prevalence of TB infection.
Fig. 3 Trends in selected Sustainable Developmental Goals indicators that are associated with TB incidence, Cambodia, 2000–2024 a,b
a Values are relative to those for 2000 (index value =100) and shown up to the latest year for which data are available.
b Estimates of the prevalence of undernutrition are for those aged ≥18 years. Estimates of the prevalence of smoking are for those aged ≥15 years.
Sources of data: United Nations Statistics Division SDG (9), World Bank Open Data (10), and the WHO Global Health Observatory (11).
How TB burden was distributed in the population
Among other notable findings from the 2023 survey, the prevalence of bacteriologically confirmed pulmonary TB among people aged 15 years and older was much higher among men than women: 806 (95% confidence interval [CI]: 675−963) per 100 000 population among men compared with 331 (95% CI: 254−432) per 100 000 population among women. Men accounted for two thirds of prevalent TB in the general population; hence, on average, for every three men with TB in Cambodia, only one was reported to the NTP. Almost 80% of people with TB were aged over 45 years, and TB prevalence in those aged over 65 years exceeded 1%. These findings highlight the need for further strengthening of TB case-finding, especially among men and the older population.
Compared with the previous survey in 2011, the most recent survey showed that much of the decline in the national prevalence of TB was driven by a corresponding decline in rural areas. Moreover, while prevalence in the 2011 survey was lower in urban (capital city) than rural areas, potentially reflecting better access to care, in the most recent survey this disparity has largely disappeared. This change may reflect increasing migration to urban areas, and resulting worsening of urban congestion.
Implications for symptom screening
The true public health significance of asymptomatic TB to the overall burden and transmission of TB is still being explored (12). In the third survey, 43% of people with TB did not report any cough during screening. This is substantially higher than the 23% found in the 2011 survey (4), possibly as a result of focused programmatic efforts to detect and treat people with symptomatic TB since the start of the century. Given such a low prevalence of cough among those with TB, systematic screening with CXR appears to be the next step for the proactive detection of people with TB in the community.
Health professionals should be aware of the high prevalence of asymptomatic TB, because this is a potential source of nosocomial infection. Hospital or clinic visits undertaken for any reason provide a good opportunity to make an early diagnosis of TB in settings of high prevalence or among high-risk individuals (including, for example, those who are underweight or have diabetes). Systematic TB screening in medical facilities with X-ray equipment should be appropriately guided and supported, with follow-up of those with CXR abnormalities who are bacteriologically negative (e.g. those with “healed TB” without TB treatment history).
Health care seeking behaviour needs further research
Even among those reporting symptoms, health care seeking behaviour was relatively limited. About half of those with cough for 2 weeks or longer had not sought care, much higher than the 10% during the 2011 survey. A similar trend between surveys is apparent in Myanmar (see Fig. 1.4.11 [Section 1.4]); this may reflect the reduced severity of symptoms among those with undiagnosed TB¸ which would be expected with improved provision of TB services. Among people who had sought care, 45% initially visited a private sector provider – a potential risk factor for a delayed diagnosis of TB. As in many TB prevalence surveys, these results suggest that a two-pronged approach is required to identify more people with TB: that is, identifying and addressing barriers to seeking care, and improving service linkages and awareness of TB among service providers.
Conclusions
Multipartner collaboration with political support can lead to a successful and highly informative national TB prevalence survey. The findings of the 2023 survey in Cambodia show that the country has made great strides in reducing the number of people suffering and dying from TB over the past 2 decades. Building on the large investments and efforts already made to find and cure people with TB, sustained efforts linked to the SDGs and multisectoral partnerships are required to reduce the burden even further, towards the ultimate goal of ending TB.
Box 1
Methods and further technical details
The primary objectives of the survey were to estimate the prevalence of bacteriologically confirmed pulmonary TB among the general population aged 15 years and older, and to assess the trend in prevalence between the surveys implemented in 2011 and 2023. Worldwide, this was the first repeat national TB prevalence survey to be completed after the COVID-19 pandemic.
Survey design and methods followed the recommendations of the latest WHO guidance on national TB prevalence surveys (1). All survey participants were screened for symptoms by interview and by digital CXR examination. Participants with any of the following – cough of at least 2 weeks, a CAD (CAD4TB version 7) score above a threshold of 30 or any radiological abnormality as adjudged by a human reader – were asked to submit one spot and one morning sputum specimen for testing with Xpert Ultra. If either of these two specimens was positive, then two additional specimens were collected for liquid culture (mycobacterial growth indicator tube, MGIT™) and solid culture (Löwenstein–Jensen, LJ) testing at the Institut Pasteur du Cambodge (IPC). Solid culture testing was undertaken to allow for a direct comparison with the second national TB prevalence survey.
To measure a statistically significant decline of at least 30% in the prevalence of culture positive TB between the second and third surveys, it was estimated that a sample size of 44 520 was required. Multistage clustered design was implemented with sampling population proportional to size strategy ; this resulted in 84 randomly selected clusters (with a target cluster size of 530 people) across three strata (rural, urban and Phnom Penh) throughout the country. The eligible population (those invited to participate in the survey) was defined as people aged 15 years and older who had been resident in the selected cluster for at least 2 weeks before the survey.
Data were entered in the field using REDCap software and tablet computers (13). Best-practice analytical methods were used to estimate TB prevalence, accounting for clustered sampling, non-participation and other missing data (1).
The survey was implemented by the NTP of Cambodia, through the National Center for Tuberculosis and Leprosy Control (CENAT), along with the Research Institute of TB /Japan Anti-TB Association (RIT/JATA) and IPC (as co-principal investigators). Other technical partners were the Cambodia Anti-TB Association, Center for Health and Social Development Organization, and WHO. Funding was provided by four agencies: the Department of Foreign Affairs and Trade, Australia; The Global Fund ; the United States Agency for International Development (USAID); and RIT/JATA.
Box 2
Key results
Detailed results from the 2023 survey and more detailed comparisons of these results with the two previous surveys will be published in an upcoming survey report and peer-reviewed journals.
Key numbers from the 2023 survey are summarized here:
A total of 65 632 people of all ages from 15 410 households were enumerated during the survey, across 84 clusters. Of these, 45 890 (69.9%) were eligible and invited to participate in the survey based on age (≥15 years) and residency status.
Of those who were eligible, 34 836 people (75.9%) agreed to participate. The participation rate was higher for females (79.7%) than males (71.3%).
Of all participants, 4381 (12.6%) screened positive for TB by both screening methods; among these participants, 4372 (99.8%) submitted two sputum specimens for Xpert Ultra testing.
A total of 310 participants were Xpert Ultra positive; among these participants, 307 (99%) submitted samples for culture.
Of the 195 people with bacteriologically confirmed TB identified from the survey, 27 (13.8%) reported symptoms during screening and 191 (97.9%) screened positive on CXR.
TB prevalence per 100 000 population aged 15 years and older was estimated as 546 (95% CI: 461−631) for bacteriologically confirmed TB. Prevalence was much higher among men than women: 806 (95% CI: 675−963) per 100 000 population for men compared with 331 (95% CI: 254−432) per 100 000 population for women. Prevalence was highest in those aged 55−64 years.
Of the 754 (2.2%) survey participants who reported the presence of a cough of 2 weeks or more, only 382 (50.7%) had sought care. Public health facilities were the most common first point of care (51% of participants who reported this symptom and sought care), followed by private health facilities (20%) and pharmacies (25%). Care was sought by more females (56%) than males (44%).
Half of participants with symptoms did not seek care; those not seeking care tended to be younger and male.
Of the 195 bacteriologically confirmed TB cases, 28 (14.4%) reported a cough of at least 2 weeks during screening. Of these 28 people, 27 were not on anti-TB treatment; of these 27 people, 11 (41%) had sought care for their symptoms, and the remaining 16 (59%) had not taken any action.
References
World Health Organization. WHO consolidated guidelines on tuberculosis data generation and use. Module 3: national tuberculosis prevalence surveys. Geneva, Switzerland: World Health Organization; 2025 (https://iris.who.int/handle/10665/381191). License: CC BY-NC-SA 3.0 IGO.
Background document 3. National TB Prevalence Surveys: are they still relevant for estimation of TB disease burden in the period 2025–2030? In: WHO/Global Task Force on TB Impact measurement [website]. Geneva, Switzerland: World Health Organization; 2024 (https://cdn.who.int/media/docs/default-source/hq-tuberculosis/global-task-force-on-tb-impact-measurement/meetings/2024-09/background-documents/background3-nationaltbprevalencesurveys.pdf?sfvrsn=17fb7f57_7).
Mao TE, Okada K, Yamada N, Peou S, Ota M, Saint S et al. Cross-sectional studies of tuberculosis prevalence in Cambodia between 2002 and 2011. Bull World Health Organ. 2014;92:573-81. doi: https://doi.org/10.2471/blt.13.131581.
Ministry of Health Cambodia. Report of the second national tuberculosis prevalence survey, 2011. Phnom Penh: CENAT; 2012 (https://openjicareport.jica.go.jp/pdf/12120325.pdf).
Ministry of Health Cambodia. Final report of the national TB prevalence survey, Cambodia 2002. Phnom Penh: CENAT; 2005.
World Health Organization. What is DOTS? : a guide to understanding the WHO-recommended TB control strategy known as DOTS. Geneva, Switzerland: World Health Organization; 1999 (https://iris.who.int/handle/10665/65979).
Uplekar M, Stop TB Partnership, World Health Organization. The Stop TB Strategy: Building on and Enhancing DOTS to Meet the TB-related Millennium Development Goals. Geneva, Switzerland: 2006 (https://iris.who.int/handle/10665/69241).
World Health Organization. The End TB Strategy. Geneva, Switzerland: World Health Organization; 2015 (https://iris.who.int/handle/10665/331326).
United Nations Statistics Division SDG API [website]. UNSD; 2025 (https://unstats.un.org/sdgapi/swagger/).
World Bank open data [website]. World Bank Group; 2025 (https://data.worldbank.org/).
The Global Health Observatory [website]. World Health Organization; 2025 ( https://covid.comesa.int/data/gho/info/gho-odata-api).
Falzon D, Miller C, Law I, Floyd K, Arinaminpathy N, Zignol M et al. Managing tuberculosis before the onset of symptoms. Lancet Respir Med. 2024; ( https://doi.org/10.1016/S2213-2600(24)00372-2).
REDCap (Research Electronic Data Capture) [website]. ( https://projectredcap.org).