Syphilis epidemiology in Jordan: prevalence, incidence, and seroreversion from 15 years of laboratory-based data Authors: Issa Abu-Dayyeh, Hiam Chemaitelly, Ahmad Al Tibi, Mohammad Ghunaim, Thaer Hasan, Amid Abdelnour, Laith J. Abu-Raddad Published: 2026 ABSTRACT Syphilis is a health priority targeted for elimination by 2030, yet its global epidemiology remains insufficiently characterized. The study analyzed data from 25,630 adults aged ≥ 15 years who underwent syphilis testing in Jordan at Biolab Diagnostic Laboratories during 2010–2025. Cross-sectional and retrospective cohort analyses were used to estimate prevalence, temporal trends, patterns of rapid plasma reagin (RPR) and Treponema pallidum pallidum hemagglutination assay (TPHA) reactivity, associated risk factors, and incidence rate. Syphilis prevalence (ever RPR- or TPHA-reactive) was 0.9% (95% CI: 0.8–1.0%) overall and only 0.08% (95% CI: 0.05–0.10%) among those tested for visa-related medical screening. RPR reactivity was 0.5% (95% CI: 0.4–0.6%), and among RPR-reactive individuals, 79.2% (95% CI: 65.0–89.5%) were TPHA-reactive within 30 days. Outpatient or walk-in visits, doctor referrals, and laboratory referrals showed substantially higher reactivity than visa-related medical screening, whereas sex, governorate, insurance status, and testing year showed no significant associations. Cumulative syphilis incidence was 0.18% (95% CI: 0.04–0.72%) after two years of follow-up, with an incidence rate of 0.92 per 1,000 person-years (95% CI: 0.23–3.68). Syphilis prevalence and incidence in Jordan are relatively low, but with no evidence of decline over 15 years, indicating limited progress toward elimination targets.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-56491-9. Introduction Syphilis is a sexually transmitted infection (STI) caused by Treponema pallidum pallidum, a spirochete infection1,2. Symptoms of early syphilis are often mild or absent, contributing to underdiagnosis and delays in treatment2,3. When untreated, infection may progress to neurological, cardiovascular, or reproductive involvement, resulting in severe or potentially fatal complications1,2,4. Transmission occurs primarily through sexual contact2,4, but vertical transmission during pregnancy or childbirth can lead to congenital syphilis, a cause of fetal and neonatal morbidity and mortality worldwide5,6. The frequently asymptomatic presentation and persistent challenges in early detection continue to hinder effective control efforts2. Globally, the World Health Organization (WHO) estimated 8.0 million new syphilis infections in 2022, an incidence rate of 2.0 per 1,000 person-years, and a prevalence of active syphilis of 0.6%7. In the Middle East and North Africa (MENA), WHO estimated 443,000 incident cases in 2022, an incidence rate of 1.1 per 1,000 person-years, and a prevalence of 0.4%7, consistent with a pooled estimate of 0.48% in the general population from a recent systematic review8. The WHO Global Health Sector Strategy on STIs (2022–2030) targets a 90% reduction in overall syphilis incidence and a reduction in congenital syphilis incidence rate to fewer than 50 cases per 100,000 live births by 20309. Monitoring progress toward these global goals requires reliable, timely data on population-level prevalence and trends. In Jordan, a country of 11.4 million people10, the epidemiology of syphilis remains poorly defined due to the absence of surveillance systems, nationally representative population-based surveys, or a documented national program for routine antenatal syphilis screening8,11–13. Existing evidence is sparse and based largely on studies conducted in selected subpopulations with limited representativeness and narrow geographic and temporal coverage8,11,14–16. Reported active syphilis prevalence was 0.00% among both 1,089 symptomatic and 186 asymptomatic women attending gynecology and family-planning clinics between 2003 and 200514. Among blood donors, ever-infection prevalence was 0.00% during 2009–2013 in 62,933 donors15, while active infection prevalence was 0.02% among 365,029 donors during 2013–201816. More recent Ministry of Health blood-bank data indicate ever-infection prevalence of 0.019% in 2019, 0.012% in 2020, 0.014% in 2021, and 0.014% in 202211. To date, no study has assessed syphilis incidence, infection clearance, or temporal trends in Jordan, leaving important gaps in understanding the infection’s epidemiologic dynamics and transmission patterns. This study addresses these gaps by analyzing a large laboratory-based dataset comprising nearly 26,000 individuals tested for syphilis between 2010 and 2025. The analysis quantifies prevalence and trends in rapid plasma reagin (RPR) reactivity, reflecting active infection, and Treponema pallidum pallidum hemagglutination assay (TPHA) reactivity, reflecting ever infection, identifies factors associated with test reactivity, and estimates incidence and non-treponemal seroreversion rates. The study offers epidemiologic insight to guide national STI control strategies and should be viewed within the broader landscape of STI surveillance and research constraints in MENA17,18. Throughout the region, data quality remains limited, shaped by sociocultural sensitivities, restricted funding, and underdeveloped research infrastructure, alongside STI and sexual health programs that are narrow in scope and limited in scale8,19–22. These gaps hinder accurate burden estimation and impede monitoring of progress toward global elimination targets. By leveraging routinely collected electronic medical records and longitudinal laboratory data from a large diagnostic network, the study demonstrates a scalable, cost-efficient approach for generating population-relevant STI evidence in data-limited settings, helping to narrow epidemiologic knowledge gaps in the region. Methods Study population and data sources This study analyzed the syphilis testing database from Biolab Diagnostic Laboratories (Biolab) in Jordan for the period January 21, 2010, to July 27, 2025. Biolab, established in 2001, is a private accredited diagnostic laboratory network operating 28 branches across major urban centers in Jordan and provides nationwide diagnostic services to a diverse population, including walk-in clients and referrals from physicians, healthcare facilities, and other laboratories. The dataset includes all RPR and TPHA tests performed since the implementation of Biolab’s digital Laboratory Information Management System (LIMS) in 2010. Available data include patient demographics (sex, age, and insurance status), encounter type (outpatient/walk-in, doctor referral, visa-related medical screening, or laboratory referral), test location (Biolab branch), test date, and test type (RPR or TPHA). As a commercial diagnostic laboratory, Biolab performs tests according to physician or client requests and does not implement a fixed diagnostic or test-of-cure algorithm. In practice, RPR is commonly used as the initial non-treponemal test, and individuals with a reactive result are advised to undergo TPHA testing to confirm treponemal infection, although this ultimately depends on the decision of the patient and/or the requesting physician. PCR testing for syphilis was introduced only recently and is rarely requested, and dark-field microscopy is not available at Biolab. Each test is linked to a unique personal identifier, which is the registration number assigned to an individual at Biolab at the time of their first encounter for any test. All subsequent tests, encounters, and transactions at Biolab are linked and tracked under this identifier. For certain testing indications, such as visa-related medical screening, individuals are required to provide official government-issued identification, ensuring that the individual’s name is verified and fixed in the system. In addition, Biolab’s automated alert system flags potential duplicate records based on name or phone number, even if a patient number is not provided or a different patient number is used, thereby helping to minimize duplicate registrations. Syphilis testing captured in this database is performed for a wide range of clinical and non-clinical purposes, although the specific indications are not routinely documented in the laboratory record. The tested population is heterogeneous and includes individuals assessed for possible infection, such as those presenting with genital ulcers, rash, or neurological manifestations, as well as persons tested after exposure to suspected or confirmed cases or undergoing post-treatment serologic follow-up23. Non-clinical screening constituted the majority of syphilis testing in this database and included testing required as part of medical examinations for visa applications through specific embassies (referred to as visa-related medical screening), primarily those of the United States, Canada, Australia, and New Zealand. These examinations are typically conducted for individuals applying for immigration to other countries, work abroad, residency permits, or other travel purposes and are performed in designated laboratories approved to provide the required medical certification for visa processing, one of which is Biolab. Screening is also performed for other purposes, such as part of antenatal care or broader STI testing panels23,24. Self-initiated testing can also occur among individuals seeking to determine their infection status or to assess prior exposure. Laboratory referral testing refers to tests conducted for individuals referred to Biolab by other laboratories, as Biolab serves as a reference laboratory for other commercial laboratories. Such referrals may involve individuals with indeterminate or inconclusive test results requiring confirmatory testing using more advanced diagnostic methods available at Biolab. In other cases, laboratories may refer individuals because they do not perform syphilis testing themselves. These laboratory referral tests were included to characterize their patterns and implications within the database. Accordingly, they were incorporated in the analyses where appropriate, consistent with the study objective of examining the patterns and associations of all testing encounters and deriving epidemiologic insights through multiple complementary analyses, including subgroup analyses. Study designs and analyses A series of analyses (substudies) were conducted using methods developed by the authors for applying epidemiologic study designs to real-world electronic health record data, including electronic health records from Biolab25–27. Syphilis prevalence, defined pragmatically—given the structure of the database—as ever having an RPR- or TPHA-reactive result within a specified time period, was estimated among adults aged 15 years and older for the full study duration and at two-year intervals. Prevalence was further assessed by 10-year age groups and within the subgroup tested through visa-related medical screening, which was considered a proxy population approximating the general population of Jordan. Children younger than 15 years were excluded from all analyses because the study focused on syphilis as a sexually transmitted infection, and individuals in this age group may not yet have reached sexual debut. Additional quantitative measures included the proportion of individuals who were RPR- or TPHA-reactive among those tested with each assay; the proportion of RPR-reactive individuals who had a TPHA-reactive result within ± 30 days; and, conversely, the proportion of TPHA-reactive individuals who had an RPR-reactive result within the same interval. Factors associated with ever testing RPR- or TPHA-reactive were assessed using univariable and multivariable regression analyses. For individuals with at least one reactive RPR or TPHA result, variables corresponding to the earliest reactive test were included in the analysis, whereas for those who remained non-reactive throughout all testing, the variables from the first recorded non-reactive result were used. Syphilis incidence and non-treponemal seroreversion were assessed empirically using two retrospective cohort analyses. For the incidence analysis, which evaluated incident infection among individuals with no prior evidence of syphilis, eligibility required at least one non-reactive RPR or TPHA result followed by a subsequent test. Individuals were excluded if they had any RPR- or TPHA-reactive result preceding their first non-reactive test. Accordingly, the analysis captured first detected infections during follow-up and did not include reinfections among individuals with documented prior reactive tests. These individuals had evidence of prior infection and may have developed partial or short-term immunity to reinfection1. As a result, incidence estimates derived from such individuals may not accurately reflect the risk of infection in the broader population. Follow-up in the incidence analysis began at the date of the first non-reactive result and continued until the first reactive result or the last recorded non-reactive test if no reactive result occurred. Specifically, incident infection was defined as a reactive RPR following a non-reactive RPR or TPHA, and a reactive TPHA following a non-reactive TPHA. The incident infection date was assigned as the midpoint between the last non-reactive test and the subsequent first reactive test. Incidence was also assessed in the subcohort restricted to individuals who underwent visa-related medical screening at baseline. For the non-treponemal seroreversion analysis, seroreversion was defined as RPR seroreversion among individuals with a reactive RPR result. Eligibility required at least one reactive RPR test followed by a subsequent RPR test. Follow-up began at the date of the first RPR-reactive result and continued until the first RPR-non-reactive result, or until the last recorded RPR-reactive result if seroreversion was not observed. The seroreversion date was assigned as the midpoint between the last reactive test and the subsequent first non-reactive test. Given the limited sample size and the skewed distribution of follow-up duration, this analysis should be interpreted as exploratory, and the resulting estimates should be viewed as hypothesis-generating rather than definitive. In the above analyses, apart from children, no additional exclusions were applied. Any exclusions in specific substudies reflected eligibility criteria for those analyses. For example, in the analysis assessing syphilis incidence, individuals with only one RPR or TPHA test result were excluded because at least two test results, indicating follow-up of the individual, are required to assess incident infection. In these substudies, all available data were also included and no subsampling was performed. Accordingly, the sample size was determined by data availability rather than by study design, and a priori sample size calculations were not required. There were also no missing data in the analyzed database, as only tests with complete information were extracted at Biolab and included in the database. Laboratory methods Syphilis RPR testing was performed using the RPR-carbon card test (CHEMELEX S.A., Polígono Industrial Can Castells, Barcelona, Spain; catalog no. 40130/40131), a manual non-treponemal flocculation assay that employs carbon-coated lipid antigen particles to facilitate macroscopic visualization of reagin antibody–antigen reactions. Serum was placed in the designated test circle, one drop of antigen suspension was added, and the mixture was spread using a disposable stirrer. The card was then rotated at approximately 100 rpm for 8 min, after which results were interpreted macroscopically under bright illumination. Following manufacturer guidance, visible flocculation of carbon particles was classified as reactive, whereas its absence indicated a non-reactive result. Syphilis TPHA testing was performed using a treponemal assay. From January 25, 2010 through December 8, 2020, testing was conducted with the TPHA Lorne kit (Lorne Laboratories Ltd., Berkshire, United Kingdom; catalogue no. 043100A), after which the TPHA Linear assay (Cromatest TPHA, Linear Chemicals S.L.U., Barcelona, Spain; catalog no. 2520005) was implemented beginning in January 2021. Both assays are manual treponemal hemagglutination tests that detect total antibodies against T. pallidum pallidum through agglutination of sensitized erythrocytes coated with treponemal antigen. Testing and interpretation followed the manufacturers’ protocols, with titers ≥ 1:80 classified as reactive. All laboratory testing was conducted at Biolab Diagnostic Laboratories using standardized procedures to ensure consistency, accuracy, and reliability across assays. Oversight The study used a retrospective analysis of routinely collected laboratory data extracted from the Biolab LIMS. The dataset was de-identified prior to sharing and analysis and contained no personal identifiers. Because the study relied exclusively on existing laboratory records generated during routine clinical and screening activities and involved no direct contact with individuals or collection of additional data, obtaining individual informed consent was not feasible. The study therefore met standard criteria for a waiver of informed consent under institutional review board policies. Ethical approval, including the waiver of informed consent, was granted by the institutional review boards at Biolab (IRB-Bio/7-2024/01) and Weill Cornell Medicine–Qatar (23–00024). The prevalence analysis and the two cohort analyses were conducted in accordance with the Declaration of Helsinki and reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Tables S1–S3 of the Supplementary Material). Statistical analysis Population characteristics were summarized using frequency distributions and measures of central tendency. Proportion measures, such as prevalence, were calculated within each analytic stratum, and corresponding 95% confidence intervals (CIs) were estimated using the exact binomial method. Each individual contributed only once per analytic period to avoid repeated inclusion in the numerator or denominator. Most testing occurred in Amman governorate, which accounts for roughly half of Jordan’s population28 and contributed over 98% of all tests, more than 70% of which were conducted for visa-related medical screening. In contrast, visa-related medical screening constituted less than 1% of tests performed outside Amman. Given these extensive regional differences in testing volume and purpose, applying geographic sampling weights25,26 would yield unreliable estimates and likely introduce bias, and was therefore deemed methodologically inappropriate for this study. Associations with ever testing RPR- or TPHA-reactive were assessed using chi-square tests and logistic regression modeling. Variables associated with the outcome at a significance level of p ≤ 0.20 in univariable analyses were included in the multivariable model. Statistical significance in the final model was defined as p ≤ 0.05. Results were presented as odds ratios (ORs) and adjusted odds ratios (aORs) with corresponding 95% CIs and p-values. Cumulative syphilis incidence was defined as the proportion of individuals with a baseline non-reactive result who developed a reactive result during follow-up, with the first reactive result occurring ≥ 30 days after the baseline non-reactive test. The 30-day interval was applied to avoid misclassifying infections acquired shortly before follow-up—before antibodies became detectable by the assays—as incident cases23,29. Cumulative syphilis RPR seroreversion was defined as the proportion of individuals with a baseline RPR-reactive result who subsequently became RPR-non-reactive, with seroreversion confirmed ≥ 30 days after the baseline reactive test. The ≥ 30-day threshold was used because seroreversion occurring very soon after a reactive result may indicate a false outcome in one of the two tests, rather than true serologic clearance1,23,29,30. In routine clinical and epidemiologic settings, genuine RPR seroreversion is typically observed only after several months1,23,29,30. A sensitivity analysis was also conducted excluding two cases with non-reactive results occurring just after 30 days of follow-up, as these were close to the 30-day threshold and may reflect unexpectedly rapid seroreversion1,23,29,30. Both cumulative incidence and seroreversion were estimated using the Kaplan–Meier estimator method31. Syphilis incidence and seroreversion rates, and their corresponding 95% CIs, were estimated using Poisson log-likelihood regression models with the Stata stptime command. Decimal places for proportions, incidence, and seroreversion measures were reported according to what is appropriate or conventional for each metric, typically retaining at least one meaningful decimal digit. Statistical analyses were performed using Stata/SE version 19.0 (Stata Corporation, College Station, TX, USA). Results Study population The selection of the study population from the syphilis testing database for inclusion in the different analyses is summarized in Fig. 1. The dataset comprised 27,824 syphilis tests (26,759 RPR tests and 1,065 TPHA tests) conducted for 25,759 individuals between January 21, 2010, and July 27, 2025. Fig. 1Flowchart illustrating the population selection process for the different substudies among 25,759 individuals tested for syphilis at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025. RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay.*These samples overlap because some individuals underwent both RPR testing and TPHA testing. Flowchart illustrating the population selection process for the different substudies among 25,759 individuals tested for syphilis at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025. RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay.*These samples overlap because some individuals underwent both RPR testing and TPHA testing. A total of 152 tests were conducted for 129 children younger than 15 years (Fig. 1). Of these, 145 were RPR tests and 7 were TPHA tests. All RPR tests were non-reactive. Among the 7 TPHA tests, 2 were performed in infants younger than 1 year and were reactive, while the remaining 5, conducted in older children, were non-reactive. These tests among children were excluded from all subsequent analyses. The final analytic dataset included 27,672 syphilis tests conducted for 25,630 individuals, comprising 26,614 RPR tests for 25,101 individuals and 1,058 TPHA tests for 1,007 individuals. The characteristics of these individuals are summarized in Table 1. The median age was 32.0 years, 52.6% were female, and 98.6% resided in the Amman governorate. On average, 1,784 syphilis tests were conducted annually (1,716 RPR tests and 68 TPHA tests), with some fluctuation in testing volume over time (Table 1). Fewer than 1% of individuals had insurance coverage for the test. Most tests (72.4%) were performed for visa-related medical screening, followed by outpatient or walk-in visits (16.9%), referrals from other laboratories (6.6%), and doctor referrals (4.0%). Testing frequency varied across individuals, with 5.3% undergoing more than one test during the study period. Table 1Characteristics of 25,630 individuals aged ≥ 15 years tested with RPR or TPHA at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025.Characteristics*RPRTPHATotalN (%)†N (%)†N (%)†Number of individuals25,101 (100)1,007 (100)25,630 (100)Median age (interquartile range)—years32.0 (25.0–43.0)33.0 (28.0–42.0)32.0 (25.0–43.0)Age (years)15–191,875 (7.5)19 (1.9)1,886 (7.4)20–298,385 (33.4)320 (31.8)8,553 (33.4)30–397,006 (27.9)367 (36.4)7,222 (28.2)40–493,495 (13.9)184 (18.3)3,578 (14.0)50–592,174 (8.7)75 (7.4)2,211 (8.6)60–691,335 (5.3)28 (2.8)1,345 (5.2)70+831 (3.3)14 (1.4)835 (3.3)SexFemale13,296 (53.0)415 (41.2)13,473 (52.6)Male11,805 (47.0)592 (58.8)12,157 (47.4)GovernorateAmman24,769 (98.7)948 (94.1)25,261 (98.6)Aqaba39 (0.2)3 (0.3)41 (0.2)Balqaa60 (0.2)7 (0.7)62 (0.2)Irbid125 (0.5)21 (2.1)138 (0.5)Karak0 (0.0)2 (0.2)2 (0.0)Maan6 (0.02)0 (0.0)6 (0.0)Madaba2 (0.0)0 (0.0)2 (0.0)Tafila6 (0.02)2 (0.2)6 (0.0)Zarqa94 (0.4)24 (2.4)112 (0.4)Year of first test2010159 (0.6)14 (1.4)171 (0.7)2011560 (2.2)22 (2.2)573 (2.2)2012703 (2.8)39 (3.9)729 (2.8)2013828 (3.3)90 (8.9)891 (3.5)20141,677 (6.7)60 (6.0)1,715 (6.7)20152,310 (9.2)44 (4.4)2,338 (9.1)20163,215 (12.8)125 (12.4)3,258 (12.7)20172,581 (10.3)125 (12.4)2,634 (10.3)20181,999 (8.0)94 (9.3)2,046 (8.0)20192,021 (8.1)55 (5.5)2,060 (8.0)20201,147 (4.6)47 (4.7)1,180 (4.6)2021959 (3.8)38 (3.8)984 (3.8)20221,723 (6.9)64 (6.4)1,751 (6.8)20232,301 (9.2)67 (6.7)2,331 (9.1)20242,048 (8.2)72 (7.1)2,083 (8.1)2025870 (3.5)51 (5.1)886 (3.5)InsuranceNo24,939 (99.4)989 (98.2)25,462 (99.3)Yes162 (0.6)18 (1.8)168 (0.7)Encounter typeOutpatient/walk-in4,236 (16.9)350 (34.8)4,342 (16.9)Doctor referral988 (3.9)161 (16.0)1,029 (4.0)Visa-related medical screening18,556 (73.9)17 (1.7)18,556 (72.4)Laboratory referral1,321 (5.3)479 (47.6)1,703 (6.6)Proportion withOnly 1 test23,757 (94.6)972 (96.5)24,282 (94.7)2 tests1,226 (4.9)27 (2.7)1,228 (4.8)3 tests89 (0.4)4 (0.4)89 (0.3)4 tests18 (0.1)2 (0.2)18 (0.1)5 + tests11 (0.04)2 (0.2)13 (0.1)Infection statusAt least one reactive test114 (0.5)162 (16.1)219 (0.9)All non-reactive tests24,987 (99.5)845 (83.9)25,411 (99.1)RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay.*Covariates for each individual—including age, sex, governorate, insurance status, and encounter type—were defined at the time of their first test.†Percentages may not always sum precisely to 100% for a given variable because of rounding. Table 1Characteristics of 25,630 individuals aged ≥ 15 years tested with RPR or TPHA at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025.Characteristics*RPRTPHATotalN (%)†N (%)†N (%)†Number of individuals25,101 (100)1,007 (100)25,630 (100)Median age (interquartile range)—years32.0 (25.0–43.0)33.0 (28.0–42.0)32.0 (25.0–43.0)Age (years)15–191,875 (7.5)19 (1.9)1,886 (7.4)20–298,385 (33.4)320 (31.8)8,553 (33.4)30–397,006 (27.9)367 (36.4)7,222 (28.2)40–493,495 (13.9)184 (18.3)3,578 (14.0)50–592,174 (8.7)75 (7.4)2,211 (8.6)60–691,335 (5.3)28 (2.8)1,345 (5.2)70+831 (3.3)14 (1.4)835 (3.3)SexFemale13,296 (53.0)415 (41.2)13,473 (52.6)Male11,805 (47.0)592 (58.8)12,157 (47.4)GovernorateAmman24,769 (98.7)948 (94.1)25,261 (98.6)Aqaba39 (0.2)3 (0.3)41 (0.2)Balqaa60 (0.2)7 (0.7)62 (0.2)Irbid125 (0.5)21 (2.1)138 (0.5)Karak0 (0.0)2 (0.2)2 (0.0)Maan6 (0.02)0 (0.0)6 (0.0)Madaba2 (0.0)0 (0.0)2 (0.0)Tafila6 (0.02)2 (0.2)6 (0.0)Zarqa94 (0.4)24 (2.4)112 (0.4)Year of first test2010159 (0.6)14 (1.4)171 (0.7)2011560 (2.2)22 (2.2)573 (2.2)2012703 (2.8)39 (3.9)729 (2.8)2013828 (3.3)90 (8.9)891 (3.5)20141,677 (6.7)60 (6.0)1,715 (6.7)20152,310 (9.2)44 (4.4)2,338 (9.1)20163,215 (12.8)125 (12.4)3,258 (12.7)20172,581 (10.3)125 (12.4)2,634 (10.3)20181,999 (8.0)94 (9.3)2,046 (8.0)20192,021 (8.1)55 (5.5)2,060 (8.0)20201,147 (4.6)47 (4.7)1,180 (4.6)2021959 (3.8)38 (3.8)984 (3.8)20221,723 (6.9)64 (6.4)1,751 (6.8)20232,301 (9.2)67 (6.7)2,331 (9.1)20242,048 (8.2)72 (7.1)2,083 (8.1)2025870 (3.5)51 (5.1)886 (3.5)InsuranceNo24,939 (99.4)989 (98.2)25,462 (99.3)Yes162 (0.6)18 (1.8)168 (0.7)Encounter typeOutpatient/walk-in4,236 (16.9)350 (34.8)4,342 (16.9)Doctor referral988 (3.9)161 (16.0)1,029 (4.0)Visa-related medical screening18,556 (73.9)17 (1.7)18,556 (72.4)Laboratory referral1,321 (5.3)479 (47.6)1,703 (6.6)Proportion withOnly 1 test23,757 (94.6)972 (96.5)24,282 (94.7)2 tests1,226 (4.9)27 (2.7)1,228 (4.8)3 tests89 (0.4)4 (0.4)89 (0.3)4 tests18 (0.1)2 (0.2)18 (0.1)5 + tests11 (0.04)2 (0.2)13 (0.1)Infection statusAt least one reactive test114 (0.5)162 (16.1)219 (0.9)All non-reactive tests24,987 (99.5)845 (83.9)25,411 (99.1)RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay.*Covariates for each individual—including age, sex, governorate, insurance status, and encounter type—were defined at the time of their first test.†Percentages may not always sum precisely to 100% for a given variable because of rounding. Characteristics of 25,630 individuals aged ≥ 15 years tested with RPR or TPHA at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025. RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay. *Covariates for each individual—including age, sex, governorate, insurance status, and encounter type—were defined at the time of their first test. †Percentages may not always sum precisely to 100% for a given variable because of rounding. Syphilis prevalence The selection of the study population included in the syphilis prevalence analysis is presented in Fig. 1. Overall syphilis prevalence, defined as ever testing RPR- or TPHA-reactive, was 0.9% (95% CI: 0.8–1.0%; 219/25,630). Among individuals with at least one visa-related medical screening test, prevalence was substantially lower at 0.08% (95% CI: 0.05–0.10%; 15/18,577). Temporal trends in prevalence across 2-year intervals are shown in Fig. 2A. Prevalence fluctuated modestly over time, with estimates remaining largely stable and overlapping across intervals within the 95% CIs. Age-specific prevalence across 10-year age groups is shown in Fig. 2B. Prevalence was lowest among individuals younger than 20 years (0.05%; 95% CI: 0.001–0.30%), increased progressively with age, and peaked at 1.2% (95% CI: 1.0–1.5%) among those aged 30–39 years. Thereafter, prevalence declined with increasing age, reaching 0.4% (95% CI: 0.2–0.8%) among individuals aged 60 years or older. Corresponding analyses stratified by sex are presented in Fig. S1 and show patterns broadly consistent with the overall estimates. Fig. 2Syphilis prevalence, defined as ever having an RPR- or TPHA-reactive result, among 25,630 adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025, by (A) 2-year time intervals and (B) 10-year age groups. RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay. Syphilis prevalence, defined as ever having an RPR- or TPHA-reactive result, among 25,630 adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025, by (A) 2-year time intervals and (B) 10-year age groups. RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay. Temporal prevalence trends by 2-year intervals and age-specific prevalence across 10-year age groups among individuals with at least one visa-related medical screening test are illustrated in Figs. S2A and S2B. Prevalence remained low, at ≤ 0.2% across all intervals and age groups. Reactivity patterns for RPR and TPHA Among the 25,101 individuals tested by RPR, 114 had at least one reactive result, yielding an RPR reactivity of 0.5% (95% CI: 0.4–0.6%; Table 1). Of these, 48 also had a TPHA test performed within ± 30 days of the reactive RPR result. Among these individuals, 38 were TPHA-reactive, indicating active syphilis, while the remaining 10 were TPHA-nonreactive, indicating likely false-positive RPR results. The proportion of TPHA-reactive individuals among those who were RPR-reactive, that is the proportion with active syphilis, was 79.2% (95% CI: 65.0–89.5%). Among the 1,007 individuals tested by TPHA, 162 had at least one reactive result, corresponding to a TPHA reactivity of 16.1% (95% CI: 13.9–18.5%; Table 1). Of these, 65 also underwent RPR testing within ± 30 days of the reactive TPHA result. Among these individuals, 38 were RPR-reactive, indicating active syphilis, while the remaining 27 were RPR-nonreactive, suggesting past or treated infection. The proportion of RPR-reactive individuals among those who were TPHA-reactive, that is the proportion with active syphilis, was 58.5% (95% CI: 45.6–70.6%). Associations with ever testing RPR- or TPHA-reactive The associations with ever testing RPR- or TPHA-reactive are summarized in Table 2. Individuals aged 15–19 years had significantly lower odds of testing reactive compared with those aged 20–29 years (aOR: 0.13; 95% CI: 0.02–0.97). No significant differences were observed between individuals aged 20–29 years and those in older age groups. Table 2Associations with syphilis infection, defined as ever testing RPR- or TPHA-reactive, among 25,630 adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025.CharacteristicsTestedEver RPR or TPHA reactiveUnivariable regression analysisMultivariable regression analysis* N N (%)p-valueOR (95% CI)p-valueF test p-value†aOR (95% CI)p-value‡Age (years)20–29§8,55387 (1.0)< 0.0011.00< 0.0011.0015–191,8861 (0.1)0.05 (0.01–0.37)0.0030.13 (0.02–0.97)0.04630–397,22287 (1.2)1.19 (0.88–1.60)0.2620.88 (0.65–1.20)0.42140–493,57836 (1.0)0.99 (0.67–1.46)0.9560.87 (0.59–1.30)0.51150–592,21117 (0.8)0.75 (0.45–1.27)0.2891.08 (0.63–1.86)0.77260+2,1809 (0.4)0.40 (0.20–0.80)0.0101.18 (0.58–2.41)0.645Sex0.4790.479Female13,473130 (1.0)1.00----Male12,157107 (0.9)0.91 (0.70–1.18)0.479----GovernorateAmman25,261221 (0.9)< 0.0011.00< 0.0011.00Balqaa622 (3.2)3.78 (0.92–15.55)0.0661.70 (0.41–7.11)0.466Irbid1382 (1.4)1.67 (0.41–6.77)0.4750.46 (0.11–1.88)0.280Aqaba/Karak/Maan/Madaba/Tafila575 (8.8)10.89 (4.31–27.54)< 0.0011.99 (0.77–5.11)0.155Zarqa1127 (6.2)7.55 (3.47–16.42)< 0.0011.16 (0.53–2.57)0.707InsuranceNo25,462236 (0.9)0.6541.000.632----Yes1681 (0.6)0.64 (0.09–4.59)0.657----Encounter typeVisa-related medical screening18,55614 (0.1)< 0.0011.00< 0.0011.00Outpatient/walk-in4,34280 (1.8)24.86 (14.08–43.91)< 0.00123.97 (13.41–42.84)< 0.001Doctor referral1,02931 (3.0)41.14 (21.82–77.58)< 0.00139.93 (21.06–75.71)< 0.001Laboratory referral1,703112 (6.6)93.23 (53.37-162.89)< 0.00199.71 (56.19-176.94)< 0.001Proportion withOnly 1 test24,280216 (0.9)< 0.0011.000.0021.002 tests1,23015 (1.2)1.38 (0.81–2.33)0.2352.93 (1.68–5.10)< 0.0013 + tests1206 (5.0)5.86 (2.55–13.47)< 0.0014.71 (2.00-11.14)< 0.001Year of the test¶25,630237 (0.9)NA0.99 (0.96–1.03)0.8590.859----aOR, adjusted odds ratio; CI, confidence interval; NA, not applicable; OR, odds ratio; RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay.*McFadden’s Pseudo R² in the multivariable model was 23.6%. Pseudo R² is a descriptive measure of model fit in logistic regression, reflecting the improvement of the fitted model over a null model based on log-likelihood values, but it does not represent the proportion of variance explained46. Higher values indicate better relative fit, and values greater than 0.2 are interpreted as indicative of a well-fitted model46.†Covariates with p-value ≤ 0.2 in the univariable analysis were included in the multivariable analysis.‡Covariates with p-value ≤ 0.05 in the multivariable analysis were considered as showing evidence for an association with syphilis infection.§The 20–29 years age group was chosen as the reference because it included the largest number of individuals.¶The test year was defined as the year of the first reactive RPR or TPHA result for individuals with at least one reactive test, or the year of the first non-reactive test for those with only non-reactive results. Table 2Associations with syphilis infection, defined as ever testing RPR- or TPHA-reactive, among 25,630 adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025.CharacteristicsTestedEver RPR or TPHA reactiveUnivariable regression analysisMultivariable regression analysis* N N (%)p-valueOR (95% CI)p-valueF test p-value†aOR (95% CI)p-value‡Age (years)20–29§8,55387 (1.0)< 0.0011.00< 0.0011.0015–191,8861 (0.1)0.05 (0.01–0.37)0.0030.13 (0.02–0.97)0.04630–397,22287 (1.2)1.19 (0.88–1.60)0.2620.88 (0.65–1.20)0.42140–493,57836 (1.0)0.99 (0.67–1.46)0.9560.87 (0.59–1.30)0.51150–592,21117 (0.8)0.75 (0.45–1.27)0.2891.08 (0.63–1.86)0.77260+2,1809 (0.4)0.40 (0.20–0.80)0.0101.18 (0.58–2.41)0.645Sex0.4790.479Female13,473130 (1.0)1.00----Male12,157107 (0.9)0.91 (0.70–1.18)0.479----GovernorateAmman25,261221 (0.9)< 0.0011.00< 0.0011.00Balqaa622 (3.2)3.78 (0.92–15.55)0.0661.70 (0.41–7.11)0.466Irbid1382 (1.4)1.67 (0.41–6.77)0.4750.46 (0.11–1.88)0.280Aqaba/Karak/Maan/Madaba/Tafila575 (8.8)10.89 (4.31–27.54)< 0.0011.99 (0.77–5.11)0.155Zarqa1127 (6.2)7.55 (3.47–16.42)< 0.0011.16 (0.53–2.57)0.707InsuranceNo25,462236 (0.9)0.6541.000.632----Yes1681 (0.6)0.64 (0.09–4.59)0.657----Encounter typeVisa-related medical screening18,55614 (0.1)< 0.0011.00< 0.0011.00Outpatient/walk-in4,34280 (1.8)24.86 (14.08–43.91)< 0.00123.97 (13.41–42.84)< 0.001Doctor referral1,02931 (3.0)41.14 (21.82–77.58)< 0.00139.93 (21.06–75.71)< 0.001Laboratory referral1,703112 (6.6)93.23 (53.37-162.89)< 0.00199.71 (56.19-176.94)< 0.001Proportion withOnly 1 test24,280216 (0.9)< 0.0011.000.0021.002 tests1,23015 (1.2)1.38 (0.81–2.33)0.2352.93 (1.68–5.10)< 0.0013 + tests1206 (5.0)5.86 (2.55–13.47)< 0.0014.71 (2.00-11.14)< 0.001Year of the test¶25,630237 (0.9)NA0.99 (0.96–1.03)0.8590.859----aOR, adjusted odds ratio; CI, confidence interval; NA, not applicable; OR, odds ratio; RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay.*McFadden’s Pseudo R² in the multivariable model was 23.6%. Pseudo R² is a descriptive measure of model fit in logistic regression, reflecting the improvement of the fitted model over a null model based on log-likelihood values, but it does not represent the proportion of variance explained46. Higher values indicate better relative fit, and values greater than 0.2 are interpreted as indicative of a well-fitted model46.†Covariates with p-value ≤ 0.2 in the univariable analysis were included in the multivariable analysis.‡Covariates with p-value ≤ 0.05 in the multivariable analysis were considered as showing evidence for an association with syphilis infection.§The 20–29 years age group was chosen as the reference because it included the largest number of individuals.¶The test year was defined as the year of the first reactive RPR or TPHA result for individuals with at least one reactive test, or the year of the first non-reactive test for those with only non-reactive results. Associations with syphilis infection, defined as ever testing RPR- or TPHA-reactive, among 25,630 adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan from January 21, 2010, to July 27, 2025. aOR, adjusted odds ratio; CI, confidence interval; NA, not applicable; OR, odds ratio; RPR, rapid plasma reagin; TPHA, Treponema pallidum pallidum hemagglutination assay. *McFadden’s Pseudo R² in the multivariable model was 23.6%. Pseudo R² is a descriptive measure of model fit in logistic regression, reflecting the improvement of the fitted model over a null model based on log-likelihood values, but it does not represent the proportion of variance explained46. Higher values indicate better relative fit, and values greater than 0.2 are interpreted as indicative of a well-fitted model46. †Covariates with p-value ≤ 0.2 in the univariable analysis were included in the multivariable analysis. ‡Covariates with p-value ≤ 0.05 in the multivariable analysis were considered as showing evidence for an association with syphilis infection. §The 20–29 years age group was chosen as the reference because it included the largest number of individuals. ¶The test year was defined as the year of the first reactive RPR or TPHA result for individuals with at least one reactive test, or the year of the first non-reactive test for those with only non-reactive results. Encounter type showed a strong association with syphilis reactivity. Compared with visa-related medical screening, outpatient or walk-in visits were associated with 23.97-fold higher odds of testing reactive (95% CI: 13.41–42.84), doctor referrals with 39.93-fold higher odds (95% CI: 21.06–75.71), and laboratory referrals with 99.71-fold higher odds (95% CI: 56.19–176.94). Individuals who underwent two syphilis tests had nearly three-fold higher odds of testing reactive compared with those tested once (aOR: 2.93; 95% CI: 1.68–5.10), while those tested three or more times had almost five-fold higher odds (aOR: 4.71; 95% CI: 2.00–11.14). No significant associations were observed with sex, governorate, insurance status, or year of testing. Syphilis incidence The selection of the study population included in the retrospective cohort analysis of syphilis incidence is presented in Fig. 1. The characteristics of the 1,350 individuals in the cohort (1,340 with an RPR-non-reactive baseline test and 10 with a TPHA-non-reactive baseline test) are summarized in Table S4. The median age was 33.0 years, 54.4% were male, and nearly all resided in the Amman governorate. Most baseline tests were conducted for visa-related medical screening (76.2%), followed by outpatient or walk-in visits (18.9%) and doctor referrals (< 5%). During follow-up, two individuals tested reactive ≥ 30 days after their baseline test (Table S5). The median follow-up duration was 1.28 years (interquartile range (IQR): 0.82–1.81), totaling 2,170.62 person-years. Cumulative incidence was 0.18% (95% CI: 0.04–0.72%) after 2 years of follow-up (Fig. 3A). The syphilis incidence rate was 0.92 per 1,000 person-years (95% CI: 0.23–3.68). Fig. 3Kaplan–Meier analyses of cumulative syphilis incidence and syphilis non-treponemal seroreversion among adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan between January 21, 2010, and July 27, 2025. (A) Cumulative syphilis incidence in the retrospective incidence cohort, including 1,350 adults with a baseline non-reactive RPR or TPHA result and at least one subsequent syphilis test. (B) Cumulative non-treponemal seroreversion in the retrospective seroreversion cohort, including 10 adults with a baseline RPR-reactive result and at least one subsequent RPR test. Tables beneath each curve show the numbers at risk over follow-up. *In panel B, the final seroreversion event occurred at the last observed time point, affecting the shape of the Kaplan–Meier estimator. The seroreversion date was assigned as the midpoint between the last reactive test and the subsequent first non-reactive test. Kaplan–Meier analyses of cumulative syphilis incidence and syphilis non-treponemal seroreversion among adults aged ≥ 15 years tested at Biolab Diagnostic Laboratories in Jordan between January 21, 2010, and July 27, 2025. (A) Cumulative syphilis incidence in the retrospective incidence cohort, including 1,350 adults with a baseline non-reactive RPR or TPHA result and at least one subsequent syphilis test. (B) Cumulative non-treponemal seroreversion in the retrospective seroreversion cohort, including 10 adults with a baseline RPR-reactive result and at least one subsequent RPR test. Tables beneath each curve show the numbers at risk over follow-up. *In panel B, the final seroreversion event occurred at the last observed time point, affecting the shape of the Kaplan–Meier estimator. The seroreversion date was assigned as the midpoint between the last reactive test and the subsequent first non-reactive test. In the subcohort restricted to individuals tested for visa-related medical screening at baseline, no incident infections were identified, yielding a cumulative incidence of 0% (95% CI: 0–0.30%) after 2 years of follow-up and an incidence rate of 0 per 1,000 person-years (95% CI: 0–2.33). Exploratory analysis of syphilis seroreversion The selection of the study population included in the exploratory retrospective cohort analysis of syphilis seroreversion is presented in Fig. 1. The characteristics of the 10 individuals in the cohort are summarized in Table S6. The median age was 35.5 years, four were female, and eight resided in the Amman governorate. Baseline testing was distributed across encounter types, with six individuals tested during outpatient or walk-in visits, two through physician referrals, and two as part of visa-related medical screening. During follow-up, five individuals tested RPR-non-reactive ≥ 30 days after their baseline test (Table S7). Examination of the time interval between the baseline RPR-reactive result and the subsequent RPR-non-reactive result did not suggest a pattern indicative of a test-of-cure algorithm (Table S7). Follow-up duration varied, with a median of 2.48 months (IQR: 0.60–6.20), totaling 42.86 person-months. Only two individuals underwent additional RPR testing after their RPR-non-reactive result, and one of these reverted to reactivity. Cumulative seroreversion reached 100% (95% CI: 54.1–100%) after 15 months of follow-up (Fig. 3B). The syphilis seroreversion rate was 11.7 per 100 person-months (95% CI: 4.9–28.0). In the sensitivity analysis excluding the two cases with non-reactive results occurring just after 30 days of follow-up (31 and 35 days; Table S7), cumulative seroreversion was 100% (95% CI: 29.2–100%) after 15 months of follow-up, and the seroreversion rate was 7.2 per 100 person-months (95% CI: 2.3–22.3). Discussion Overall results, including those from visa-related medical screening that may approximate the general population, suggest that syphilis infection rates in Jordan are relatively low, both within the context of the MENA region and relative to global levels7,8,32,33. These findings are consistent with earlier studies from Jordan11,14–16, recent regional evidence8, and global analyses32. The comparatively low rates may reflect the sexually conservative context18,34, expanding opportunities for screening and treatment such as among pregnant women despite the absence of a documented national program for routine antenatal syphilis screening12,13, and the widespread use of antibiotics, which may inadvertently clear concurrent, often asymptomatic, syphilis infections32,35–37. Globally, syphilis prevalence has exhibited both declines and increases over the past three decades32,38, while evidence suggests declining rates in MENA8,32,39. Increases reported in some countries outside MENA have often been linked to rising sexual risk behavior following the introduction of HIV antiretroviral therapy and pre-exposure prophylaxis1,38,40,41—interventions with relatively low coverage in MENA’s nascent HIV epidemics42. In contrast, no temporal trends in syphilis prevalence were observed in Jordan over the decade and a half covered by this study. This finding indicates that additional efforts will be needed for Jordan to meet WHO’s target of a 90% reduction in syphilis incidence by 20309. Test encounter type showed a strong association with syphilis reactivity. The highest rates were observed among laboratory referrals, consistent with the practice of confirming reactive or borderline in-house results at accredited external laboratories such as Biolab. Outpatient or walk-in visits and doctor referrals also exhibited substantially higher reactivity than visa-related medical screening, possibly reflecting underlying risk factors or clinical suspicion, including recent high-risk sexual activity or the emergence of signs or symptoms suggestive of infection. Moreover, individuals undergoing repeat testing had higher reactivity rates, probably for similar reasons. The study yielded additional findings. Individuals aged 15–19 years had markedly lower syphilis prevalence than older adults, possibly reflecting lower sexual risk behavior in this age group, including absent or less frequent sexual activity. Beyond this group, prevalence did not differ across adult age categories. No differences in prevalence were observed by sex, governorate, or insurance status. Lastly, seroreversion reached 100% among those RPR-reactive after 15 months of follow-up, indicating a decline in reactivity over time, likely following treatment of infection1,23,29,30. This study has limitations. Real-world analyses of this type are inherently constrained by their retrospective nature and reliance on electronic health record data of individuals who underwent syphilis testing, rather than prospective designs specifically tailored to estimate epidemiologic measures. Consequently, the study population may not be representative of the wider population of Jordan, introducing the potential for selection bias. In addition, STI infrastructure in Jordan remains limited, and we are not aware of standardized syphilis protocols—such as diagnostic or test-of-cure algorithms—that are consistently applied or the extent of their implementation that could inform study design and analyses. Visa-related medical screening was considered the testing modality most likely to approximate the general population and serve as a proxy for it. In contrast, the other encounter types—outpatient or walk-in visits, doctor referrals, and laboratory referrals—are more likely to involve individuals tested because of clinical suspicion, symptoms, or confirmatory testing needs and are therefore less likely to reflect the general population. Moreover, Jordanians have high emigration rates, with a large proportion of citizens living abroad, and visa-related screening affects a broad segment of the population seeking employment, residency, or migration opportunities outside Jordan. Notably, approximately 75% of all syphilis tests in this database were conducted as part of visa-related medical screening. Nevertheless, this group should be interpreted as a proxy population rather than a representative sample, and the observed prevalence may underestimate the true infection level if individuals who anticipate a positive result avoid visa or immigration applications that require medical screening. The lack of a specific implemented protocol for diagnosing syphilis, confirmatory testing, and titration data restricted analyses to reactivity based on RPR and TPHA without definitively distinguishing active from past infection. Temporal changes in testing practices, healthcare utilization, and access to diagnostic services over the 15-year study period may have influenced observed trends. The data were captured through electronic health records and lacked detailed clinical and behavioral information, thereby introducing the potential for information bias. This precluded assessment of symptom status, infection stage, treatment history, and behavioral risk factors such as sexual practices, partner characteristics, and exposure history. Although Biolab operates nationally and receives samples from multiple regions of Jordan, the data originate from a single laboratory network and therefore do not capture all syphilis testing conducted in the country. Other providers, including hospital-based laboratories and other private diagnostic centers, also perform testing. Differences in testing practices across laboratories or regions could therefore introduce selection bias and limit the generalizability of the findings. In addition, differences in the availability of testing data across regions and by insurance status may have limited the precision and representativeness of the analyses in detecting associations with these variables. While repeat tests are linked through a unique patient identifier, duplicate registration remains possible, although it is likely minimized by Biolab’s automated alert system25,26. In some situations, however, such as when individuals request STI test panels, a person may theoretically wish to conceal their identity and therefore provide a false name to initiate a new registration, resulting in a new Biolab number and a potential duplicate registration. Because government-issued identification may not be requested for STI panel screening, such duplicates may not always be detectable. The extent to which this may occur and affect the presented estimates is unknown. Syphilis testing relied on RPR and TPHA assays, which provide complementary but biologically distinct information, and each has recognized limitations, including the potential for false-positive and false-negative results1,23,29,43–45. The sensitivity of RPR varies by stage of infection and may be reduced in very early or late disease, while specificity may be affected by other infections or autoimmune conditions1,23,29. In addition, technical phenomena such as the prozone effect can rarely produce false-negative results such as in samples with very high antibody titers43,45. Conversely, TPHA detects treponemal antibodies and therefore reflects current or past infection rather than active disease1,23,29. Treponemal assays such as TPHA may also cross-react with antibodies to other treponemal infections (e.g., yaws or bejel) and therefore cannot always distinguish venereal syphilis from these infections where they occur23. The less-than-perfect concordance observed between RPR and TPHA results in this study is therefore consistent with the known diagnostic characteristics of these assays rather than indicative of poor test performance. Accordingly, epidemiologic outcomes were defined pragmatically to align with available data, which may introduce some imprecision or bias. For instance, prevalence was defined as ever having an RPR- or TPHA-reactive result within a specified period rather than through a cross-sectional assessment using a single standardized test per individual. Despite these pragmatic definitions, alternative reasonable analytic choices would likely have produced similar conclusions, as supported by sensitivity and additional analyses. For example, estimating prevalence by randomly selecting a single test for individuals with multiple records produced a value of 0.8% (95% CI: 0.7–1.0%), closely matching the primary estimate of 0.9% (95% CI: 0.8–1.0%). Finally, the seroreversion analysis was exploratory and its findings should be interpreted with caution. Implementation of this analysis required methodological assumptions dictated by the structure of the available data. In particular, testing occurred at irregular intervals without a standardized follow-up or test-of-cure protocol, treatment dates were not available, and seroreversion could only be identified when a reactive RPR was followed by a subsequent non-reactive result. Consequently, the timing of seroreversion was interval-censored and was approximated by assigning the midpoint between the last reactive and first non-reactive test. The analysis was further limited by the small cohort size and the skewed distribution of follow-up duration. This study has strengths. It utilized a large laboratory database with national coverage spanning 15 years of data. Diagnostic testing was conducted using reliable assays, ensuring consistent and comparable results. The study was large in scale and scope, integrating multiple epidemiologic analyses and designs—including cross-sectional and cohort approaches—to assess prevalence, temporal trends, reactivity patterns, associations with infection, incidence, and seroreversion, thereby enabling a comprehensive characterization of syphilis epidemiology in the country. To our knowledge, this study represents the most detailed analysis of syphilis in Jordan to date and provides an evidence base to inform national STI surveillance and control strategies, such as strengthening laboratory-based surveillance, identifying trends and patterns in infection, and highlighting key attributes of syphilis epidemiology and the need to address it. In conclusion, syphilis prevalence and incidence in Jordan are relatively low, with no evidence of temporal change over 15 years of observation. However, the lack of a downward trend indicates that additional efforts are required to place the country on a trajectory consistent with the WHO 2030 syphilis elimination target. By leveraging routinely collected data from a large national diagnostic network, this study highlights the value of laboratory-based surveillance in data-limited settings. Strengthening integrated STI surveillance, expanding targeted testing among higher-risk populations, and enhancing antenatal screening are key priorities to support national control efforts and advance progress toward syphilis elimination. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1