Introduction

Yaws, a neglected tropical disease (NTD) caused by Treponema pallidum subsp. pertenue, predominantly affects children in low- to upper-middle-income countries in the tropics [1–3]. If untreated, yaws can lead to severe health problems, progressing from initial skin ulcers to secondary and tertiary stages, resulting in chronic disfigurement and disability.

The epidemiological status of yaws indicates its persistence as a significant public health issue in certain populations. According to World Health Organization (WHO) data, yaws is endemic in at least 16 countries, with most cases reported from WHO’s Western Pacific Region [4]. In 2023, 222,652 suspected yaws cases were reported to WHO from 13 countries, although only 1,477 cases were confirmed in 9 countries. About 85% of all suspected cases were reported from two countries: Papua New Guinea (100,528) and Indonesia (88,646). However, laboratory tests are not yet routinely performed to confirm cases, highlighting challenges in accurate diagnosis and reporting [4].

In the 2021–2030 NTD road map [5], endorsed by the World Health Assembly in November 2020 [6], yaws is targeted for global eradication by 2030. WHO’s comprehensive yaws eradication strategy includes the use of azithromycin mass drug administration (MDA) [7], which has proven its efficacy, effectiveness, and safety for reducing the prevalence of yaws at population level [8,9]. This approach requires more than 90% coverage of the whole population in each of three MDA rounds in high endemic areas [10–12]. Recommended post-MDA impact assessment includes both passive and active surveillance [13]. Routine passive surveillance, in which patients seek care for their illnesses at health facilities, detects active yaws (i.e., symptomatic cases), though most affected children do not seek treatment. Active surveillance, involving outreach activities, facilitates the identification of active yaws cases in communities and schools, as well as the implementation of serosurveys to identify latent yaws (i.e., asymptomatic seroreactors). The criteria established by WHO for verification of interruption of transmission include a 3-year period, beginning two years after the last reported case of yaws, with (i) no new serologically confirmed active yaws cases, (ii) no new PCR-confirmed cases, and (iii) no evidence of new transmission events based on serosurveys (i.e., latent or active yaws in children aged 1–5 years born after the last reported case) [14]. WHO’s strategy also emphasizes the integration of yaws eradication efforts within broader NTD or skin-disease-focused strategies, to enhance resource optimization, intervention coverage, and surveillance efficiency [13].

The yaws eradication strategy, as previously published [7,13,15], does not specify parameters for the required serosurveys. Inadequate investigation could lead to biases, resource misuse, or missed cases, the last increasing the risk for a rebound in community transmission of yaws. Recommendations summarized in this policy platform provide a framework for national programmes to conduct effective serosurveys to assess the local interruption of yaws transmission.

Overview of the survey policy and objectives

In March 2023, the World Health Organization Global Neglected Tropical Diseases Programme (WHO/NTD) established an Informal WHO Working Group on Yaws Survey Protocol Development including independent experts from multiple disciplines. The group was asked to provide guidance for serological surveys to assess the interruption of transmission of yaws at national or local levels. This document summarizes the conclusions of the group.

Guidance was developed based on the WHO strategy for yaws eradication [7], statistical principles, expert opinion, and best practice, using the data and experiences available in 2023. Low-level transmission or absence of transmission should be demonstrated using seroprevalence surveys. These surveys must meet methodological standards to provide confidence for public health officials and policy-makers. Given that the agreed global public health target for yaws is eradication, a large sample size is encouraged wherever possible. It is not possible for a single survey to provide sufficient evidence to demonstrate the interruption of transmission at a local level or within any geographical region. Therefore, national programmes should aim to document multiple lines of evidence, including sequential surveys and evidence of active and passive surveillance being conducted. A single survey will most likely not be sufficiently powered to exclude low-level transmission, but sequential surveys over several years should provide the necessary level of confidence that local interruption of transmission has been achieved.

Survey protocols recommended in this document rely on active enrollment and testing of 1–5-years-olds in population-based surveys using fingerpick blood sample (“serosurveys”). These surveys aim to measure whether the prevalence of latent-yaws cases in a specific area, called the EU, is below 1%. Serosurveys should start at least 2 years after the last reported active case of yaws (i.e., 2 years after the provisional transmission interruption date) as a result of successive rounds of MDA (Fig 1) and/or other interventions. At least three serosurveys conducted between 2 and 10 years after zero case declaration are necessary to provide confidence that a state of no observed yaws transmission has been maintained. In cases of large EUs or when initial surveys show prevalence above 1%, more than three surveys might be required.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Overall roadmap of yaws eradication and scope of this policy guidance platform.

TCT, total community treatment; TTT, total targeted treatment, S1, initial serosurvey; S2,= first follow-up survey; S3, second follow-up survey; S4, third follow-up survey (optional); S5, fourth follow up survey (optional).

https://doi.org/10.1371/journal.pntd.0012899.g001

In line with the strategic shift from vertical to integrated NTD programmes [6], yaws programmes are encouraged to conduct surveillance alongside other planned activities (such as surveys for other NTDs). As such, the guidance herein on survey design should be interpreted as broad guidance. Programmes are encouraged to find opportunities to incorporate yaws testing into existing surveys, in which case the sampling strategy and sample size can be adjusted to facilitate integration and maximize efficiency.

A Technical Advisory Group is recommended to be formed at the start of serosurveillance to guide the national program on the timing, design, and interpretation of these surveys. This group should include experts in yaws, epidemiology, clinical practice, and statistics. A definition of the Technical Advisory Group and other clinical and epidemiological terms employed in this policy guidance are provided in the S1 Appendix.

Choice of serological tests

Table 1 lists various serological tests available for yaws, each of which has advantages and disadvantages [16–18]. None of these tests can distinguish yaws from syphilis; therefore, result interpretation may in some cases require careful clinical assessment [19]. The selection of diagnostic tools depends on availability, budget, and how well they fit with other disease control programs which act as platforms. Typically, all these tests have high accuracy when used on individuals suspected of having active yaws. However, in low-prevalence populations, particularly for the detection of latent cases in a post-MDA setting, they are prone to producing more false-positive results due to the lower positive predictive power. Box 1 summarizes the list of recommendations regarding the use of serological tests, including essential features for result interpretation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. List of available serological tests for trepanomatoses, including yaws [16–18].

https://doi.org/10.1371/journal.pntd.0012899.t001

Delineation of evaluation units

EUs must be selected by considering local yaws epidemiology, programmatic feasibility, and population size:

1. i) Epidemiology: The risk of yaws should be relatively homogenous within each EU. Factors such as historic yaws data, geography, population density, socio-economic status, and the degree of remoteness or rurality could be considered for assessing risk homogeneity. Neighboring communities and districts not known to have previously been endemic should also be considered for inclusion in serosurveys, especially if they have similar risk factors as the endemic area. In areas with a history of many cases of yaws or in sub-populations that could sustain yaws transmission independently of the rest of the population (e.g., island communities or mobile populations), relatively smaller EUs may be necessary. While more costly, defining smaller EUs with the same sample size leads to greater sampling density and reduces the chance that serosurveys fail to detect evidence of infection that is present in the population.
2. ii) Feasibility: The selected EU should correspond to a geographical area where public health interventions can be effectively delivered: ideally a province, district or sub-district. An advantage of using districts as the EU is that doing so facilitates alignment with population-based surveys conducted for other indications. Due to limited resources for single-disease assessments, national yaws programmes need to be flexible and should try to integrate with other serosurveys, such as trachoma surveys [20], transmission assessment surveys for lymphatic filariasis [21], vaccine-preventable disease surveys, community- or school-based nutritional surveys or population-based HIV impact assessments as much as possible. The national programmes must carefully consider these epidemiological and practical factors when defining the EUs for yaws serosurveys and deciding on appropriate integration opportunities.
3. iii) Population size: The population size of each EU should be carefully considered to facilitate effective survey administration and efficient resource allocation. For a 30-cluster survey, it is recommended that EUs maintain a maximum population of 500,000 and have a similar size within a given country. While a precise minimum size has not been established, total population for surveys usually range from 100,000 to –250,000 [22], with lower sizes (20,000–50,000) suggested to facilitate decision-making [7].

Yaws-specific initial serosurveys

Selection of primary sampling units and households within a PSU

The initial serosurvey targets children aged 1–5 years who reside in the EU. PSUs are the smallest administrative units within the EU, such as villages, communities, neighborhoods, or census enumeration areas, for which a complete list is available. Table 2 lists the required sample sizes for different scenarios: in EUs with ≤30 PSU (all PSUs are included) and in EUs with >30 PSUs (cluster sampling is recommended).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Number of children to be tested for yaws-specific initial serosurvey.

https://doi.org/10.1371/journal.pntd.0012899.t002

Following this, Fig 2 provides a flowchart for decision-making in the selection of PSUs and households within each PSU.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Algorithm for determining the number and selecting primary sampling units and households.

https://doi.org/10.1371/journal.pntd.0012899.g002

In small EUs with ≤30 PSUs, all PSUs need to be included in the survey. For the selection of households, a household sampling interval should be established. The survey team and local leaders should use this interval to plan a route that covers all households. At each selected household, the survey team will enroll all children aged 1–5 years until all households in the plan are accounted for. The S1 Appendix provides guidelines and an example for estimating the household sampling interval.

In large EUs with >30 PSUs, cluster sampling of PSUs is needed, which involves randomly selecting a subset of PSUs to survey. The steps displayed in Fig 2 for cluster sampling in large EUs are further described in the S1 Appendix.

In all cases, survey teams should be asked to record the GPS coordinates of the selected households in each of the PSUs visited. These data, combined with the serosurvey results, will guide subsequent follow-up surveys.

Yaws initial serosurvey integrated into existing programs

Integration of serosurveys into existing programs is an opportunity to optimize public health resources, with potentially important benefits for the population. To assess whether an existing survey can be adapted to integrate yaws transmission interruption assessments, consider the survey scale (ensuring it represents the entire EU) and the demographics of the target population. School-based surveys generally miss preschool-age children, who are crucial for yaws assessments, making household-based surveys more suitable due to their broader age coverage. National programmes may wish to explore potential platforms for integration considering the key recommendations in Box 1.

For example, in trachoma impact and surveillance surveys that include all residents of selected households aged 1 year and above, yaws testing could include all selected 1–5-year-olds, with consideration given to adding households to exceed 1000 individuals aged 1–5 years [23]. In the case of lymphatic filariasis Transmission Assessment Surveys, typically targeting children aged 6–7 years, children aged 1–5 years in all selected households may be invited to undergo yaws testing. In a lymphatic filariasis IDA Impact Surveys, which target adults aged ≥20 years, yaws testing could extend to all households selected, enrolling children aged 1–5 years. Other survey platforms may include those for vaccine-preventable diseases, non-communicable diseases, or malaria. Table A in S1 Appendix provides detailed information regarding ongoing survey platforms that could integrate yaws serosurveys.

Follow-up surveys

The initial serosurveys aim to provide a geographically representative estimate of yaws seroprevalence across the entire EU. However, the design does not take into account the spatial heterogeneity in yaws prevalence. Model-based geostatistical methods yield more precise predictions of EU-wide prevalence and quantify within-EU heterogeneity by incorporating spatial correlation and environmental covariate information from georeferenced data collected during the initial serosurvey [24] (further explanation and theoretical framework for model-based geostatistics are provided in the S1 Appendix). Whereas initial surveys may be integrated with other surveys, follow-up surveys may need to be tailored based on baseline findings and may not be as amenable to integrating.

It is suggested that the national programmes consider the serosurvey objectives, the local context, the initial survey (Survey ‘S1’ in Fig 1) results, and the recommendations in Box 1 to inform the follow-up survey design (Survey ‘S2’, ‘S3’, etc. in Fig 1). The aims of these follow-up surveys may vary according to the estimated seroprevalence in the EU at the initial serosurvey. Examples of possible aims for follow-up surveys are provided in the S1 Appendix.

At least two follow-up surveys should be conducted to demonstrate with high confidence that the threshold has been achieved and maintained over time and that there are no signs of re-emergence. While these surveys may be conducted annually, it may be preferable to space them out over several years to enhance the detection of disease re-emergence.

Handling positive tests during serological surveys

When asymptomatic children in a village test positive for yaws during a serosurvey, it is important to understand whether they have current infection. Fig 3 summarizes the decision algorithm for classifying a case as a true current yaws infection versus past treated cases.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Testing and decision algorithm for determining infection in individual positive results during serological survey.

Trep: Treponemal test; Non-trep: Non-treponemal test; POS: Positive NEG: Negative. ^aGiven the reactivity of serological tests to syphilis, this algorithm does not apply to children with a history of congenital syphilis.

https://doi.org/10.1371/journal.pntd.0012899.g003

In the field, individuals should be tested using a treponemal RDT, which should then be verified with the non-treponemal line of a DPP assay to distinguish current infections from past treated cases using a finger prick. In the laboratory, positive treponemal tests can be confirmed with RPR tests, which require a venous sample. High RPR titers (≥1:8) suggest current infection, either active or latent; low titers (≤1:4) should be interpreted variably, based on birthdate: past treated infection (i.e., serofast) in children born before the zero-case declaration, and recent infection in those born after the zero-case declaration. Treatment with azithromycin can be initiated upon treponemal RDT positivity since recalling patients for treatment may be difficult in many environments. However, confirmatory testing is important for interpreting programmatic results and making informed decisions. Actions to follow include:

1. For confirmed current infections, intensified serological sampling (i.e., repeated testing to the same individual for investigating serological progression and sampling household contacts to investigate local transimission) and active clinical case searches should be conducted, with PCR testing used for clinical cases whenever feasible.
2. For potential serofast cases, where children have low titers, but are known to have been treated, interviews with parents or guardians can provide the child’s history and symptoms.
3. In children with low RPR titers who were not previously treated, appropriate treatment for yaws should be administered and intensified sampling and clinical searches should be reinitiated.
4. If no new cases are found despite intensified sampling, the initial findings may be false positive test results. This is especially likely as the predictive power of a positive test decreases with yaws prevalence decrease.
5. Identifying additional positive cases during intensified sampling can lead to an overestimation of the true area-wide prevalence. A statistician should analyze the data considering the sampling strategy, but if expert analysis is unavailable, survey analyses can exclude results from areas of intensified sampling for a less biased estimate.

Recommendations for serosurvey results interpretation and decision-making

The goal of analyzing the results of yaws serosurveys is to compare the sampled prevalence with the target threshold of 1%. To determine whether the true prevalence in the population is likely to be below this threshold, it is common to report the upper bound of the one-sided 95% confidence limit for true prevalence. If this upper limit is less than 1%, it suggests that the true prevalence is probably below the target threshold. The analysis should account for the survey design; in particular, it is important to account for clustering of positives if a cluster-based survey design was used.

If any initial or follow-up surveys find yaws prevalence above 1%, two additional years of serosurveys and active case searches should be added to the surveillance period. If latent yaws cases (asymptomatic seroreactors) are still found during this period in children aged 1–5 years, a new treatment intervention (TTT or TCT) should be considered, followed by three more rounds of serosurveys. If active yaws cases (symptomatic cases) are detected and confirmed by serology and PCR during active case searches, a new intervention (TTT or TCT) should be deployed, and surveillance should continue until zero active cases are achieved, followed by seroprevalence surveys starting anew. The intervention choice should be made based on the WHO guidelines in force [4]; however, consideration of evolving evidence regarding the effectiveness of TCT followed by TTT versus repeated TCT rounds is recommended [10,25]. In case TTT or TCT is needed, close monitoring of treatment failures is essential to identify macrolide resistances [11,26].

Of note, decisions regarding the epidemiological progress of yaws in a given area, should be make by combining the information from serosurveys and passive and active surveillance, when applicable.

Conclusion

This policy paper for yaws serosurveys provides guidance on a comprehensive approach to sequential serosurveys required to confirm the interruption of yaws transmission over time. Ensuring sufficient sample sizes and systematic PSU selection, coupled with georeferenced data, enhances the precision of prevalence estimates and facilitates targeted interventions. Positive serological results should be confirmed with additional testing, and high-titer reactions should trigger intensified surveillance and intervention, including PCR testing for symptomatic individuals.

While all recommendations in this paper are evidence-based and stem from extensive experience in yaws management, caution is advised when applying them. First, the sample size calculations and recommendations reflect the current global standard but should be adapted to local contexts by experts; larger sample sizes are encouraged when possible, given the goal of yaws’ eradication. Second, no single survey can confirm transmission interruption in a population. Health ministries should gather multiple lines of evidence, such as surveillance records and consecutive serosurveys over several years, to ensure confidence in transmission interruption. Third, some critical recommendations lack robust evidence but are provided to assist initial program development. As more data emerges from yaws serosurveys, guidance may be updated; users should stay informed of any revisions or new recommendations.

For local yaws elimination efforts to succeed, sustained political support, targeted funding, ongoing training for health workers, maintaining passive surveillance, and active testing are indispensable. Equally crucial is engaging communities with high-level awareness about yaws and implementing and maintaining robust monitoring and evaluation frameworks to assess the effectiveness of intervention and surveillance strategies. Advancing the goal of yaws eradication globally also requires continuous research and global partnership and inter-sectoral collaboration among stakeholders.

Supporting information

Acknowledgments

We acknowledge the valuable contributions of all researchers who participated in the initial discussions that shaped the development of this Policy Forum. We thank Ms. Laia Bertran and Dr. Gerard Carot-Sans (Fight Infectious Diseases Foundation, Hospital Germans Trias i Pujol, Spain) for their editorial support.

AWS, DAD, PP, AY, ZL, MYTB, KHS, and KAB are staff members of the World Health Organization. RCS is a staff member of the Pan American Health Organization. The authors alone are responsible for the views expressed in this article and they do not necessarily represent the views, decisions, or policies of the institutions with which they are affiliated. All members of the Informal WHO Working Group on Yaws Survey Protocol Development and invited experts who contributed to this work in their individual capacity completed a “Declarations of interests for WHO experts” form. The WHO Secretariat assessed the interests declared by the experts and found no significant interests warranting the exclusion of any expert.