Discrepancy Results of Treponemal Tests: Exploring the Associated Risk Factors Using Machine Learning Technology Based on 18 Years of Electronic Medical Records and National Claims Data

Discrepancy Results of Treponemal Tests: Exploring the Associated Risk Factors Using Machine Learning Technology Based on 18 Years of Electronic Medical Records and National Claims Data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Discrepancy Results of Treponemal Tests: Exploring the Associated Risk Factors Using Machine Learning Technology Based on 18 Years of Electronic Medical Records and National Claims Data Hsin-Yao Wang, Ru-Fang Hu, Ting-Wei Lin, Wan-Ying Lin, Yu-Chiang Wang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3847949/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Syphilis is becoming more prevalent and is diagnosed primarily through serologic tests. Confirmatory treponemal tests (TTs), such as Treponema pallidum particle agglutination (TPPA) and fluorescent treponema antibody absorption (FTA-Abs), are typically used. Although only one TT is needed for syphilis diagnosis, multiple TTs are still common in the disease course. Discrepant TT results can cause confusion and delay treatment. We aimed to explore the clinical characteristics of patients with discrepant TT results and develop a machine learning based tool to evaluate the risk of TT discrepancies. Methods In the retrospective cohort study, we linked electronic health records from a medical centerand national claims records in Taiwan between January 2001 and September 2018. Medical histories and demographic characteristics were considered as variables of interest to identify risk factors and develop machine learning models to assess the risk of TT discrepancy. We also analyzed the association between syphilis treatments and discrepant TT results. Results Among 5,780 eligible patients tested for syphilis, 133 (2.30%) had discrepant TT results. Risk factors associated with discrepant TT results were HIV and AIDS (adjusted odds ratio [aOR] = 2.6, 95% CI 1.4-4.7), osteoarthritis (aOR = 3.3, 95% CI 1.3-7.1), and pregnancy (aOR = 5.0, 95% CI 1.8-11.6). Patients with a top 5% risk probability from the LighGBM model were 10 times more likely to have discrepant TT results. For cases with discrepant TT results, TPPA was more likely than FTA-Abs to turn negative after treatment (OR 9.18, 95% CI 2.05-41.07). Conclusions Identifying risk factors and developing a machine learning based decision support tool could significantly aid in interpreting serologic tests for syphilis and guide accurate diagnosis. Syphilis serologic tests treponemal tests discrepant results Figures Figure 1 Figure 2 Figure 3 Introduction Syphilis is a worldwide sexually transmitted disease that is widely known as the great imitator [1] It is transmitted through sexual intercourse and blood transfusion during the infection. Syphilis remains a global public health obstacle [2–5], with the World Health Organization (WHO) estimating that 6 million new cases of syphilis among people aged 15–49 years annually in 2012 [6]. Between 2014 and 2018, the number of reported syphilis cases in the US increased by about 80% [7]. Since the 2000s, there has been a growing number of isolated syphilis outbreaks in North America and Europe [8]. The World Health Organization (WHO) has set a goal of reducing the incidence of syphilis by 2030 [6]. In Taiwan, incidence rates have been increasing year by year, and clinical personnel are required to report syphilis cases to health authorities [9]. Typically, serologic tests are the important standards for diagnosing syphilis, and the conventional algorithm involves a screening nontreponemal test (NTT) followed by a confirmatory treponemal test (TT). Two widely used TTs are the Treponema pallidum particle agglutination assay (TPPA) and the fluorescent treponemal antibody absorption test (FTA-Abs) [10]. These tests play a fundamental role in confirming syphilis infection, and their simultaneous application has shown promising results in enhancing diagnostic accuracy. A meta-analysis evaluating the performance of TPPA and FTA-Abs simultaneously for syphilis diagnosis demonstrated enhanced sensitivity and specificity compared to using the tests separately [11]. Although only one TT is needed for syphilis diagnosis, multiple TTs are still common in the disease course. Patients could undergo testing with different TTs across various healthcare institutes. Different TTs could also be ordered simultaneously, particularly when physicians are less familiar with the TTs, or in areas where the cost of TTs isn't high (e.g., Taiwan). However, studies have reported that certain factors could interfere with TTs results and confuse syphilis diagnosis [12–14]. These studies revealed that advanced age, autoimmune diseases (e.g., systemic lupus, scleroderma), intravenous drug use, and pregnancy would interfere TT results [13]. The discrepant TT results would cause confusion and delay treatment. Therefore, there is a need for large-scale investigations that use real-world data and machine learning models to identify the risk of and risk factors associated with discrepant TT results. Hence, we analyzed the results of TTs using large-scale electronic health records from a tertiary medical center, and linked them with national wide claims data to identify the risk factors associated with discrepant TTs. We also developed a machine learning model and tool for distinguishing high-risk patients. The characteristics are valuable information for interpreting the TT test results and could result in early diagnosis, early treatment, and enhancing patient follow-up care by providing the risk of having discrepant TT results. Materials and Methods Study scheme Our study aimed to investigate the risk of and risk factors associated with discrepant TTs. We then developed a machine learning model and clinical decision support tool capable of evaluating the risk of TT discrepancy for individuals. Additionally, we studied the association between the types of TTs and the risk of TT discrepancy after anti-syphilis treatments. The study scheme is illustrated in Figure 1 . Data source A retrospective cohort study was conducted by linking the National Health Insurance Research Database (NHIRD) and Chang Gung Research Database (CGRD) [15]. The National Health Insurance (NHI) program covers more than 99% of the population in Taiwan and the NHIRD prospectively records the standardized data of healthcare services submitted to the NHI program [16]. The diagnoses are registered using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes before 2016, and ICD-10-CM after then. The NHIRD data is routinely validated by the NHI Bureau [17]. Chang Gung Research Database (CGRD) is a database that included de-identified electronic medical records (EMRs), including laboratory test results, from the largest group of healthcare providers in Taiwan [15]. To obtain a comprehensive medical history, we linked EMRs from CGRD with the NHIRD, using encrypted personal identification numbers. The data were anonymized and de-identified prior to analysis to protect patient privacy [18, 19]. The Chang Gung Medical Foundation Institutional Review Board approved this study (IRB no. 201801524B0) and granted waivers for patient consent. Research subjects identification This study included Taiwanese patients who underwent at least two confirmatory TTs at CGMHs from January 2001 to September 2018. The TTs include semi-quantitative T reponema pallidum particle agglutination assay (TPPA), semi-quantitative T reponema pallidum haemagglutination (TPHA), and qualitative fluorescent treponemal antibody-absorption (FTA-ABS). The TTs performed when individuals younger than 1 year were excluded due to the unstable test results. The weak, equivocal, or borderline test results and tests using cerebrospinal fluid specimens were also excluded. Finally, patients who had no outpatient or inpatient visits records in NHIRD are also excluded. Discrepancy test result definition The research subjects were divided into two groups based on their TT results. The discrepancy group was defined as those who had a positive TT result followed by a negative TT result. Other situations, such as the patient with all positive (consistent results) or all negative TT results (non-infected), or those with a negative TT result followed by a positive result (newly infected), were defined as a non-discrepancy group. For the discrepancy group, the index date was the date when the TT results changed from positive to negative, which could be due to false-positive for the prior test or false-negative results for the later test. For the non-discrepancy group, the index date was the last day of the TT. Age was calculated based on the date of birth and the index date. Comorbidities assessment Comorbidities were identified based on patients' history of diagnosis. The Elixhauser comorbidity index developed by the Healthcare Cost and Utilization Project was used to investigate general comorbidities [20–23]. Comorbidities associated with discrepancy in TT results, as identified in previous studies or by clinical experts, were included in the analysis ( Table S1 ) [24–26]. We used the dxpr package [27] to generate these comorbidities from diagnosis codes recorded in at least three encounters. Sensitivity Analys i s We conducted sensitivity analyses to evaluate the method used to identify comorbidities. To assess the impact of comorbidity definition, we redefined comorbidities as having at least one or two care encounters related to the condition. Univariable and multivariable analyses were conducted for both sensitivity analyses. Model establishment and evaluation The demographic characteristics and comorbidities that were significantly different between the two groups (p-value < 0.1 and the case number was greater than three [18]) were selected as independent variables for developing models to predict the risk of TT discrepancy. To robustly evaluate the model performance, we randomly divided the data into train and evaluation sets 100 times. In each training set, we fine-tuned the hyperparameters of the models using a 5-fold cross-validation approach. Finally, we use the evaluation set to evaluate the performance of prediction models [28–30]. We assessed the performance of the models using the area under the receiver operating characteristic curve (AUC). We used least absolute shrinkage and selection operator (Lasso) regression, stepwise logistic regression, eXtreme Gradient Boosting (XGBoost), and LightGBM to build the discrepant TTs prediction models. We estimated the feature importance using information gain in the LightGBM model. Positive predictive value (PPV) score of discrepancy TTs risk Based on the predictive probabilities generated by machine learning models, we calculated “positive predictive value (PPV) score” to better illustrate the risk of individuals experiencing discrepancy TTs. We sorted all cases in descending order of their predictive probabilities and divided them into 20 equal groups (vigintiles) based on their risk levels [31]. For each vigintile, the PPV score was calculated using the following formula: Using the first vigintile as the example, cases with top 5% risk of discrepant TTs were analyzed. To generate the PPV score for the first vigintile, we first divided the number of cases with discrepant TTs by the total number of cases in the first vigintile (numerator), then normalized it by the overall prevalence of discrepant TTs in the whole study population (denominator) to generate the PPV score. The PPV score provides a more explicit method for clinical caregivers to illustrate the risk of getting discrepant TTs in a random index case. Review of anti-syphilis treatments for the cases with discrepancy TTs We reviewed the cases with discrepant TTs to determine if they received anti-syphilis treatments. The status of treatment was classified as "had treatment" or "no treatment" based on the guidance of syphilis treatment [32]. Cases that received recommended or suboptimal treatments were classified as "had treatment," while those that received no treatment after being diagnosed with syphilis were labeled as "no treatment." We also recorded the time of treatment. We analyzed the association between the status of treatment and the types of discrepant TTs. Statistical analysis The differences in medians were tested using the Kruskal-Wallis test. Chi-square test and Fisher exact test were used for the univariate analysis of the categorical variables. To identify factors associated with TT discrepancy, we performed multivariable logistic regression with Firth's correction, which is a solution to the problem of separation caused by rare events. We reported the crude and adjusted odds ratios (OR) with their corresponding 95% confidence intervals (CI) estimated by the profile penalized likelihood method [33]. The analyses were performed in R (version 4.0.3) and SAS software version 9.4. All statistical tests were two-sided, and statistical significance was defined as P < 0.05. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline [34]. Results Population characteristics In total 11,135 patients were available for the study after linking data from CGRD with NHIRD. Of these, 5,780 patients met our inclusion criteria ( Figure S1 ). Among patients received at least two confirmatory tests, 133 (2.3%) had discrepant TT results, and 5,647 (97.7%) had non-discrepancy results. Patients with discrepant results (67.8, IQR=35.5) were found to be significantly older than those with non-discrepancy results (44.8, IQR=36.2, P<0.001, Table S2 ). There was no difference in the distribution of sex between the two groups. Treponemal test pattern for the cases with TTs discrepancy was shown in Table S3 . Most patients with tests discrepancy have TPPA-positive results followed by TPPA-negative results (94, 70.7%). Co m orbidities analysis The crude and adjusted odds ratios (aOR) adjusted by age and sex were shown in Table 1 . The risk factors associated with discrepant TT results were HIV and AIDS (aOR = 2.6, 95% CI 1.4-4.7), other and unspecified osteoarthritis (aOR = 3.3, 95% CI 1.3-7.1), and pregnancy (aOR = 5.0, 95% CI 1.8-11.6). In the sensitivity analysis, the crucial factors associated with discrepancy test results, including HIV and AIDS, and other and unspecified osteoarthritis, were identified as risk factors in both comorbidities definitions using one diagnosis ( Table S4) and two diagnoses ( Table S5), too. Table 1 . Association between comorbidities and discrepancy syphilis test results Comorbidities Odds Ratio (OR) Crude OR (95%CI) P value Adjusted OR a (95%CI) P value HIV and AIDS (Acquired immune deficiency syndrome) 1.68 (0.92-2.86) 0.070 2.63 (1.4-4.67) 0.004 * Deficiency anemias 0.82 (0.2-2.19) 0.733 0.62 (0.17-1.6) 0.361 Rheumatoid arthritis collagen vascular diseases 1.91 (0.74-4.06) 0.128 1.87 (0.75-3.92) 0.163 Congestive heart failure 2.14 (0.83-4.55) 0.075 1.31 (0.52-2.77) 0.540 Chronic pulmonary disease 2.13 (1.24-3.44) 0.004 * 1.49 (0.86-2.46) 0.146 Depression 1.87 (0.78-3.78) 0.115 1.44 (0.62-2.89) 0.366 Diabetes without chronic complications 1.84 (1.11-2.89) 0.012 * 1.13 (0.67-1.81) 0.638 Diabetes with chronic complications 1.75 (0.88-3.14) 0.080 1.14 (0.58-2.05) 0.688 Hypertension, uncomplicated 1.67 (1.13-2.42) 0.009 * 0.86 (0.56-1.32) 0.500 Hypertensive heart disease without heart failure 1.39 (0.49-3.12) 0.474 0.89 (0.33-1.96) 0.792 Liver disease 0.59 (0.23-1.24) 0.211 0.56\ (0.23-1.15) 0.124 Fluid and electrolyte disorders 4.46 (1.53-10.37) 0.002 * 2.57 (0.91-5.96) 0.071 Other neurological disorders 1.98 (1-3.55) 0.034 * 1.14 (0.57-2.09) 0.684 Peripheral vascular disease 3.82 (1.46-8.27) 0.002 * 2.35 (0.92-5.08) 0.071 Solid tumor without metastasis 0.55 (0.14-1.47) 0.310 0.4 (0.11-1.03) 0.059 Chronic peptic ulcer disease 1.47 (0.52-3.3) 0.404 1.14 (0.42-2.51) 0.770 Valvular disease 2.28 (0.8-5.17) 0.077 1.62 (0.59-3.6) 0.317 Arthropathy 2.58 (0.62-7.15) 0.114 1.92 (0.52-5.07) 0.290 Inflammatory bowel disease 1.5 (0.46-3.65) 0.429 1.6 (0.52-3.75) 0.372 Inflammatory spondylopathies 2.24 (0.78-5.05) 0.085 1.69 (0.62-3.75) 0.280 Other and unspecified osteoarthritis 5.18 (1.96-11.39) <0.001* 3.27 (1.27-7.14) 0.017 * Other and unspecified soft tissue disorders not elsewhere classified 2.07 (1.07-3.65) 0.019 * 1.33 (0.69-2.38) 0.371 Pneumonia 2.52 (1.17-4.78) 0.009 * 1.66 (0.78-3.19) 0.178 Polyosteoarthritis 3.41 (1.18-7.82) 0.010 * 2.07 (0.75-4.72) 0.148 Pregnancy 2.79 (0.97-6.35) 0.029 * 4.97 (1.76-11.6) 0.004 * Rheumatoid arthritis 3.16 (0.76-8.82) 0.057 2.6 (0.7-6.97) 0.137 Urticaria 2.04 (0.79-4.33) 0.096 1.9 (0.76-3.97) 0.155 Vasomotor and allergic rhinitis 1.44 (0.67-2.71) 0.299 1.33 (0.63-2.48) 0.424 a Adjusted for covariates: Age and Sex. * P < 0.05 T he Performance of Discrepancy TTs Prediction We identified demographic characteristics and comorbidities that showed a significant difference (P < 0.1) between the two groups and used them as variables to construct a prediction model for discrepant TT results. Among the four prediction algorithms tested, LightGBM significantly outperformed the others with an AUC of 0.705 (95% CI=0.697-0.713, P<0.001). The other three prediction algorithms, namely Lasso regression (AUC 0.657, 95% CI=0.648-0.666), stepwise logistic regression (AUC 0.669, 95% CI=0.66-0.678), and gradient boosting algorithm (AUC 0.676, 95% CI=0.666-0.687), had lower AUC values. The most important variables in the LightGBM model were HIV and AIDS, pregnancy, sex, and age ( Figure 2 ). PPV score of discrepancy TTs risk The prevalence of discrepancy in TTs among the research subjects was 0.02. Using the PPV score formula, we divided all the cases into 20 risk categories to indicate the risk of getting discrepancy TTs ( Figure 3 ). Most of the cases with discrepant TTs were classified into the first risk vigintile. In the first risk vigintile (with the top 5% risk probability), the risk of getting discrepancy TTs was ten times higher than the baseline frequency for the entire research subjects. Using the PPV score concept, we developed an application to estimate the risk of discrepancy in TTs ( Figure S2 ). Anti-syphilis treatments for the cases with discrepancy TTs As the NTT titer is known to decrease in response to syphilis treatments [35], we investigated the association between syphilis treatments on TTs discrepancy. Of the 133 patients with TTs discrepancy, 71 (53.4%) received antibiotic treatment for syphilis, while 62 (46.6%) did not receive treatment. Among those who received treatment, 57 patients treated before TTs discrepancy was observed. The results indicate that patients whose TPPA turned negative were at a higher risk than those with FTA-Abs to be associated with anti-syphilis treatment (OR=9.2, 95% CI= [2.0-41.1], Table 2 ). Table 2 . Association between TT types and syphilis treatment among the patients with discrepant TTs. TPPA turns negative: negative TPPA being recorded after a positive TT; FTA-Abs turns negative: negative FTA-Abs being recorded after a positive TT. TPPA turns negative FTA-Abs turns negative No treatment before negative 61 20 Had treatment before negative a 56 2 OR 9.2 (2.0-41.1) a One patient had both TPPA and FTA-Abs negative results after a positive TT. Discussion Theoretically, only one type of TT would be used in either traditional or reverse algorithm in syphilis diagnosis. However, in a real-world setting, multiple TTs with different analytical methods can be tested for a patient. First, different type of TTs can be tested along the entire disease course. Second, different types of TTs can be tested across different healthcare institutes. Finally, different types of TTs can be ordered at once when physicians are less familiar with the TTs. Briefly, multiple TTs are not uncommon for a patient. Due to the recent rapid increase in new syphilis cases [7, 36, 37] and the scarcity of research on discrepant results of syphilis confirmatory tests, it is important to identify the factors that influence TTs results. In this study, we used demographic information and comorbidities to determine the features associated with the discrepant results. Based on the multivariable analysis and discrepancy prediction model, the most relevant features are sex, age, HIV and AIDS, and pregnancy. We found that negative TPPAs after a positive TT were more strongly associated with syphilis treatment than FTA-Abs. This large-scale study investigated the risk comorbidities associated with TTs discrepancy and identified the TT type that could be affected by syphilis treatment. These results are valuable for the accurate diagnosis of syphilis when TTs are crucial tests for syphilis diagnosis. TTs discrepancy may be more common than previously thought, particularly in settings where the cost of testing is low (<10 USD, totally covered by national health insurance in Taiwan), and TTs are widely used in syphilis management. Although the test requests for TTs in our study did not always follow clinical guidance, our cohort was valuable for identifying cases with multiple TTs and TTs discrepancy. In total 2.3% of 5,780 cases with multiple TTs showed discrepancy. This finding raises concerns for syphilis diagnosis, as relying on a single TT may increase the risk of false negative results for true syphilis cases and false positive results for cases without syphilis. Such errors can have serious consequences, including missed diagnoses of syphilis in pregnant women or delayed diagnoses in patients with HIV/AIDS. Moreover, our study identified various patterns of TTs discrepancy, highlighting the complexity of interpreting these results. To improve syphilis management, it may be necessary to re-evaluate the frequency and criteria for testing TTs for screening or diagnosis. A previous study revealed that HIV and AIDS, and pregnancy, are known causes of false positive for treponemal tests [13], which is consistent with our findings ( Table 1 and Figure 2 ). In addition, we identified “unspecified osteoarthritis” as a risk factor associated with TTs discrepancy. The underlying mechanism between the unspecified osteoarthritis and TTs discrepancy has not been well explored. However, it is possible that the autoantibodies involved in unspecified osteoarthritis [38] could interfere with the testing process of TTs and cause discrepancies [39]. The large-scale nature of our study allowed us to observe TTs discrepancy in a more comprehensive way. The results could provide new insights for developing novel diagnostic tools for syphilis or investigating the underlying mechanisms. We combined claims data (NHIRD) with EMRs (CGRD) to examine the most important laboratory test results that are missing in claims data. This examination is the first investigation to develop machine learning models using multiple comorbidities and demographic factors to determine factors related to discrepant syphilis test results. The study results wcaill contribute to the epidemiological knowledge of TTs discrepancy and establish a baseline to evaluate the impact of syphilis. We also developed an online risk calculator based on the TTs discrepancy predictive model ( Figure S2 ). The PPV score instead of risk probability was used to indicate the risk of having TTs discrepancy ( Figure 3 ). The PPV score is a more easily understandable way to convey the concept of risk than risk probability. The PPV score describes the risk compared to the overall population and is important for both physicians and patients. A tool can only be useful in clinical practice if the information can be easily understood and communicated [40]. While TTs are considered as the “confirmatory test” for syphilis diagnosis, they are also considered “syphilis scar” because positive results typically last for lifetime even after anti-syphilis treatment. However, the TTs discrepancy we found implies that TTs are not as “confirmatory” as we believed for a long time [41]. We also investigated the impact of anti-syphilis treatment on TTs discrepancy. The results disclosed that TPPA is more likely than FTA-Abs to turn negative after treatment ( Table 2 ). The reasons why TPPA is more vulnerable than FTA-Abs would be complicated and have not yet been fully investigated. Previous study also found TPPA seroreversion after treatment [42]. One possible mechanism is related to the antigenicity of reagents. While only part of the antigens is coated onto the particles in TPPA test, the entire T. pallidum bacteria is used to capture antibodies in FTA-Abs [10]. Thus, antigenicity of FTA-Abs test would be more comprehensive and intact than that of TPPA because the whole surface antigens on T. pallidum are theoretically preserved in FTA-Abs. This comprehensive antigenicity could be helpful in detecting various anti-syphilis antibodies in serum, even when some antibodies disappear after treatment. However, the explanation is hypothesized on the basic concept of testing and antigenicity. There are many different commercial TTs, and different epitopes are used for testing [10]. Therefore, caution should be taken when extrapolating our observations to other TTs from different providers. This study has limitations. First, out-of-pocket services were not included in the diagnosis data obtained from NHIRD [43]. Second, we identified multiple comorbidities using diagnosis codes. However, in cases where individual blood test results were unavailable, doctors may have provided a tentative diagnosis [44]. Therefore, we defined the presence of comorbidities as having at least three medical encounters related to the condition. Third, disease stage is not available from the medical records. The impact of disease stage on TTs discrepancy cannot be analyzed. Including disease stages in medical records can help researchers gain a better understanding of the impact and relationship between these two factors. Last, the results may not be generalizable to other populations as we only included observations from the Linkou branch of CGMH. However, CGMH at Linkou, a major medical center located in Northern Taiwan with about 3,800 beds and serving nearly four million outpatients annually, represents one of the largest cohorts in Taiwan [45]. Despite these limitations, further research is needed to better understand the association between multiple comorbidities and discrepant TT results. This study portrays the association between demographic and multiple comorbidities data and discrepant TTs results. We linked EHRs with claims data to identify risk factors that may affect such test results. Our findings indicated that age, HIV and AIDS, pregnancy, and osteoarthritis were the influential factors associated with discrepant TTs results. These findings could serve as a reference for improving diagnostic decisions in syphilis preventive healthcare and guiding interventions in general. Syphilis remains a vital public health issue, and further research is necessary to improve healthcare and clinical conditions related to this disease. Declarations Ethics approval and consent to participate The Chang Gung Medical Foundation Institutional Review Board approved this study (IRB no. 201801524B0) and granted waivers for patient consent. Consent for publication Not applicable. Availability of data and materials All data used in the analyses is restricted due to the regulation of the law on the protection of patients’ data in accordance with local and institutional legal requirements; The process of applying for access to the data will be made available upon reasonable request to the corresponding author. Competing Interests The authors declare no conflict of interest. Funding This research was supported by the National Science and Technology Council, Taiwan (NSTC 111-2320-B-182A-002-MY2 and 111-2628-E-A49-026-MY3) and Chang Gung Memorial Hospital (CMRPG3M0851, CMRPG3L1011). The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Author Contributions YJT and RFH had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. YJT, RFH and HYW analyzed/interpreted the data, performed experiments, designed the study, and draft the manuscript. TWL, WYL, YCW, and JJL reviewed/edited the manuscript for important intellectual content and provided administrative, technical, or material support. YJT and HYW obtained funding and supervised the study. Acknowledgement We thank the Health and Welfare Data Science Center, Ministry of Health and Welfare Chang Gung sub-centers and the data Health and Welfare Data Science Center, Ministry of Health and Welfare (HWDC, MOHW) for administrative support, promoting Chang Gung medical system to national health database application to conduct value-added research. The authors also acknowledge the technical support of the Department of Management Information System, Chang Gung Memorial Hospital. This study is based in part on data from the Chang Gung Research Database provided by the Chang Gung Memorial Hospital and National Health Insurance Research Database provided by the MOHW. The interpretation and conclusions contained herein do not represent the presentation of both institutions. References Çakmak SK, Tamer E, Karadağ AS, Waugh M (2019) Syphilis: A great imitator. Clin Dermatol 37:182–191. https://doi.org/https://doi.org/10.1016/j.clindermatol.2019.01.007 Spiteri G, Unemo M, Mårdh O, Amato-Gauci AJ (2019) The resurgence of syphilis in high-income countries in the 2000s: A focus on Europe. Epidemiol Infect 147:. https://doi.org/10.1017/S0950268819000281 Ghanem KG, Ram S, Rice PA (2020) The Modern Epidemic of Syphilis. New England Journal of Medicine 382:845–854. https://doi.org/10.1056/NEJMra1901593 de Souza RL, dos Santos Madeira LDP, Pereira MVS, et al (2020) Prevalence of syphilis in female sex workers in three countryside cities of the state of Pará, Brazilian Amazon. BMC Infect Dis 20:129. https://doi.org/10.1186/s12879-020-4850-1 Harmon ED, Robertson EW (2019) Syphilis: A growing concern. Nurse Practitioner 44:21–28. https://doi.org/10.1097/01.NPR.0000558159.61349.cb World Health Organization (2016) Global Health Sector Strategy on Sexually Transmitted Infections 2016–2021 Towards Ending STIs. https://apps.who.int/iris/bitstream/handle/10665/246296/?sequence=1. Accessed 5 Aug 2023 Ghanem KG, Ram S, Rice PA (2020) The Modern Epidemic of Syphilis. New England Journal of Medicine 382:845–854. https://doi.org/10.1056/NEJMra1901593 Kitayama K, Segura ER, Lake JE, et al (2017) Syphilis in the Americas: A protocol for a systematic review of syphilis prevalence and incidence in four high-risk groups, 1980-2016. Syst Rev 6:1–7. https://doi.org/10.1186/s13643-017-0595-3 Lai S-K, Chen C-M, Kuo H-W, Liu D-P (2017) Epidemiology of Syphilis and Gonorrhea in Taiwan_ 2005–2016. TAIWAN EPIDEMIOLOGY BULLETIN 33:460–468. https://doi.org/10.6525/TEB.20171219.33(24).001 Morshed MG, Singhb AE (2015) Recent trends in the serologic diagnosis of syphilis. Clinical and Vaccine Immunology 22:137–143. https://doi.org/10.1128/CVI.00681-14 Bristow CC, Klausner JD, Tran A (2020) Clinical Test Performance of a Rapid Point-of-Care Syphilis Treponemal Antibody Test: A Systematic Review and Meta-analysis. Clinical Infectious Diseases 71:S52–S57. https://doi.org/10.1093/cid/ciaa350 Felman YM, Nikitas JA (1980) Syphilis Serology Today. Arch Dermatol 116:84–89. https://doi.org/10.1001/archderm.1980.01640250086025 Forrestel AK, Kovarik CL, Katz KA (2020) Sexually acquired syphilis: Laboratory diagnosis, management, and prevention. J Am Acad Dermatol 82:17–28. https://doi.org/10.1016/j.jaad.2019.02.074 Hook EW (2017) Syphilis. The Lancet 389:1550–1557. https://doi.org/10.1016/s0140-6736(16)32411-4 Shao SC, Chan YY, Kao Yang YH, et al (2019) The Chang Gung Research Database—A multi-institutional electronic medical records database for real-world epidemiological studies in Taiwan. Pharmacoepidemiol Drug Saf 28:593–600. https://doi.org/10.1002/pds.4713 Lin LY, Warren-Gash C, Smeeth L, Chen PC (2018) Data resource profile: the National Health Insurance Research Database (NHIRD). Epidemiol Health 40:e2018062. https://doi.org/10.4178/epih.e2018062 Hsieh C-Y, Su C-C, Shao S-C, et al (2019) Taiwan’s National Health Insurance Research Database: past and future. Clin Epidemiol Volume 11:349–358. https://doi.org/10.2147/CLEP.S196293 Lin LY, Warren-Gash C, Smeeth L, Chen PC (2018) Data resource profile: the National Health Insurance Research Database (NHIRD). Epidemiol Health 40:e2018062. https://doi.org/10.4178/epih.e2018062 Hsieh C-Y, Su C-C, Shao S-C, et al (2019) Taiwan’s National Health Insurance Research Database: past and future. Clin Epidemiol Volume 11:349–358. https://doi.org/10.2147/CLEP.S196293 Walraven C Van, Quan H, Forster AJ (2009) A Modification of the Elixhauser Comorbidity Measures Into a Point System for Hospital Death Using Administrative Data. Med Care 47:626–633 Menendez ME, Neuhaus V, Van Dijk CN, Ring D (2014) The Elixhauser comorbidity method outperforms the Charlson index in predicting inpatient death after orthopaedic surgery. Clin Orthop Relat Res 472:2878–2886. https://doi.org/10.1007/s11999-014-3686-7 Elixhauser A, Steiner C, Harris D, Coffey R (1998) Comorbidity Measures for Use with Administrative Data Methods Defining Important Comorbidities. Med Care 36:8–27 Quan H, Sundararajan V, Halfon P, et al (2005) Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Med Care 43:1130–1139. https://doi.org/10.1097/01.mlr.0000182534.19832.83 van den Akker M, Buntinx F, Knottnerus JA (1996) Comorbidity or multimorbidity. European Journal of General Practice 2:65–70. https://doi.org/10.3109/13814789609162146 Van den Akker M, Buntix F, Metsemakers JFM, et al (1998) Multimorbidity in general practice: Prevalence, incidence, and determinants of co-occurring chronic and recurrent diseases. J Clin Epidemiol 51:367–375. https://doi.org/10.1016/S0895-4356(97)00306-5 Navickas R, Petric V-K, Feigl AB, Seychell M (2016) Multimorbidity: What do we know? What should we do? J Comorb 6:4–11. https://doi.org/10.15256/joc.2016.6.72 Tseng Y-J, Chiu H-J, Chen CJ (2021) dxpr : an R package for generating analysis-ready data from electronic health records—diagnoses and procedures. PeerJ Comput Sci 7:e520. https://doi.org/10.7717/peerj-cs.520 Batten AJ, Thorpe J, Piegari RI, Rosland AM (2020) A Resampling Based Grid Search Method to Improve Reliability and Robustness of Mixture-Item Response Theory Models of Multimorbid High-Risk Patients. IEEE J Biomed Health Inform 24:1780–1787. https://doi.org/10.1109/JBHI.2019.2948734 Liu Y, Chen P-HC, Krause J, Peng L (2019) How to Read Articles That Use Machine Learning: Users’ Guides to the Medical Literature. JAMA 322:1806–1816. https://doi.org/10.1001/jama.2019.16489 Uddin S, Khan A, Hossain ME, Moni MA (2019) Comparing different supervised machine learning algorithms for disease prediction. BMC Med Inform Decis Mak 19:1–16. https://doi.org/10.1186/s12911-019-1004-8 Wang H, Hung C, Chen C, et al (2019) Increase Trichomonas vaginalis detection based on urine routine analysis through a machine learning approach. Sci Rep 9:11074. https://doi.org/10.1038/s41598-019-47361-8 World Health Organization WHO guidelines for the treatment of treponema pallidum (Syphilis). Heinze G, Ploner M, Beyea J (2013) Confidence intervals after multiple imputation: combining profile likelihood information from logistic regressions. Stat Med 32:5062–76. https://doi.org/10.1002/sim.5899 Vandenbroucke JP (2007) Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. Ann Intern Med 147:W. https://doi.org/10.7326/0003-4819-147-8-200710160-00010-w1 Pandey K, Fairley CK, Chen MY, et al (2022) Changes in the Syphilis Rapid Plasma Reagin Titer Between Diagnosis and Treatment. Clinical Infectious Diseases. https://doi.org/10.1093/cid/ciac843 Kerani RP, Handsfield HH, Stenger MS, et al (2007) Rising rates of syphilis in the era of syphilis elimination. Sex Transm Dis 34:154–161. https://doi.org/10.1097/01.olq.0000233709.93891.e5 Huang S-Y, Hung J-H, Hu L-Y, et al (2018) Risk of sexually transmitted infections following depressive disorder. Medicine 97:e12539. https://doi.org/10.1097/md.0000000000012539 Camacho-Encina M, Balboa-Barreiro V, Rego-Perez I, et al (2019) Discovery of an autoantibody signature for the early diagnosis of knee osteoarthritis: Data from the Osteoarthritis Initiative. Ann Rheum Dis 78:1699–1705. https://doi.org/10.1136/annrheumdis-2019-215325 Park IU, Fakile YF, Chow JM, et al (2019) Performance of Treponemal Tests for the Diagnosis of Syphilis. In: Clinical Infectious Diseases. Oxford University Press, pp 913–918 Van de Velde S, Kunnamo I, Roshanov P, et al (2018) The GUIDES checklist: development of a tool to improve the successful use of guideline-based computerised clinical decision support. Implement Sci 13:86. https://doi.org/10.1186/s13012-018-0772-3 Young H (2000) Guidelines for serological testing for syphilis. Sex Transm Infect 76:403–405. https://doi.org/10.1136/sti.76.5.403 Bosshard PP, Graf N, Knaute DF, et al (2013) Response of treponema pallidum particle agglutination test titers to treatment of syphilis. Clinical Infectious Diseases 56:463–464 Chi C, Lee J, Tsai S, Chen W (2008) Out‐of‐pocket payment for medical care under Taiwan’s National Health Insurance system. Health Econ 17:961–975. https://doi.org/10.1002/hec.1312 Lin CC, Lai MS, Syu CY, et al (2005) Accuracy of diabetes diagnosis in health insurance claims data in Taiwan. Journal of the Formosan Medical Association 104:157–163. https://doi.org/10.29828/JFMA.200503.0002 Lin J-Y, Kang EY-C, Yeh P-H, et al (2020) Proposed measures to be taken by ophthalmologists during the coronavirus disease 2019 pandemic: Experience from Chang Gung Memorial Hospital, Linkou, Taiwan. Taiwan J Ophthalmol 10:80–86. https://doi.org/10.4103/tjo.tjo_21_20 Additional Declarations No competing interests reported. Supplementary Files SyphilisSupp.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3847949","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":266081733,"identity":"3f740c37-76dc-4fb9-80bb-6b2085be62a0","order_by":0,"name":"Hsin-Yao Wang","email":"","orcid":"","institution":"20/20 GeneSystems Inc.","correspondingAuthor":false,"prefix":"","firstName":"Hsin-Yao","middleName":"","lastName":"Wang","suffix":""},{"id":266081734,"identity":"45bdf872-3e73-4e5d-9bd7-b6afc22e5762","order_by":1,"name":"Ru-Fang Hu","email":"","orcid":"","institution":"Chang Gung University","correspondingAuthor":false,"prefix":"","firstName":"Ru-Fang","middleName":"","lastName":"Hu","suffix":""},{"id":266081735,"identity":"94334189-ccd6-4446-b597-da8c21046b55","order_by":2,"name":"Ting-Wei Lin","email":"","orcid":"","institution":"Chang Gung Memorial Hospital at Linkou","correspondingAuthor":false,"prefix":"","firstName":"Ting-Wei","middleName":"","lastName":"Lin","suffix":""},{"id":266081736,"identity":"debcc33d-a7bc-4baf-923c-28e39bed632c","order_by":3,"name":"Wan-Ying Lin","email":"","orcid":"","institution":"Syu Kang Sport Clinic","correspondingAuthor":false,"prefix":"","firstName":"Wan-Ying","middleName":"","lastName":"Lin","suffix":""},{"id":266081737,"identity":"f45433b5-1154-411b-9b03-2ae7298ec894","order_by":4,"name":"Yu-Chiang Wang","email":"","orcid":"","institution":"Brigham and Women’s Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yu-Chiang","middleName":"","lastName":"Wang","suffix":""},{"id":266081738,"identity":"cd890111-41f1-4278-818c-29ee697486a2","order_by":5,"name":"Jang-Jih Lu","email":"","orcid":"","institution":"Chang Gung Memorial Hospital at Linkou","correspondingAuthor":false,"prefix":"","firstName":"Jang-Jih","middleName":"","lastName":"Lu","suffix":""},{"id":266081739,"identity":"c4711511-8eeb-4a43-b593-d37ff11c74d3","order_by":6,"name":"Yi-Ju Tseng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYFCCA2BSDspjJkbLYTBpjNDCRlALxNzEBqK1GBw8f/jDxx216dslkp9JMFRYJzbI9xjg13LgMJvkzDPHc3fOSDOTYDiTntjAxkNYCzNv27HcDTdy2CQY2w4DtfBuIKSF+fPftmPpBmAt/4jTwiDN2FaTANHSQIQWyQOHzSR72w4YbjjzzNgi4Vi6cRtb/ge8WvhuHHz84WdbnbzB8eSHNz7UWMv2Mx9LwKtF4cYBEAWMTwGgQpBagjEp398AouoYGPgPEFI7CkbBKBgFIxUAAMlVTQkiK8eeAAAAAElFTkSuQmCC","orcid":"","institution":"National Yang Ming Chiao Tung University","correspondingAuthor":true,"prefix":"","firstName":"Yi-Ju","middleName":"","lastName":"Tseng","suffix":""}],"badges":[],"createdAt":"2024-01-09 10:29:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3847949/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3847949/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":49434144,"identity":"fda2f91d-a442-4d51-822c-1f24afb77aab","added_by":"auto","created_at":"2024-01-10 19:23:59","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":733326,"visible":true,"origin":"","legend":"\u003cp\u003eThe study scheme. The first part of the study investigates the associations between clinical features and treponemal tests (TTs) discrepancy. Risk factors that are linked to TTs discrepancy are identified, and an online risk calculator is developed to estimate the risk of TTs discrepancy. In addition to identifying risk factors, the study evaluates the association between TT types and treatment in cases with TTs discrepancy. TPPA: \u003cem\u003eTreponema pallidum \u003c/em\u003eparticle agglutination; FTA-Abs: fluorescent treponemal antibody-absorption.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3847949/v1/024a82f83b5d453aa99efd72.jpg"},{"id":49433676,"identity":"699e0cff-2a41-4c2c-af9c-a6f2f2547ef0","added_by":"auto","created_at":"2024-01-10 19:15:59","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":610047,"visible":true,"origin":"","legend":"\u003cp\u003eThe feature importance of LightGBM models. The feature importance is defined as the information gain, which represents the average of differences between the entropy before and after the split using the feature in trees. The middle line of boxes is the median. Dot points are single data point of each feature.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3847949/v1/d3e8d928c9660d0223836c57.jpg"},{"id":49433675,"identity":"c4c58f21-2001-4b5d-95bf-17275d9901a5","added_by":"auto","created_at":"2024-01-10 19:15:59","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":271261,"visible":true,"origin":"","legend":"\u003cp\u003ePPV scores of LightGBM model. PPV (positive predictive value) score was defined as the prevalence of discrepant test results in the vigintiles divided by the overall prevalence. For example, to generate the PPV score for the first vigintile, we first divided the number of cases with discrepant tests by the total number of cases in the first vigintile (numerator). Then normalized it by the overall prevalence of discrepant tests in the whole study population (denominator) to generate the PPV score.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3847949/v1/127cd6bde3db248604dda244.jpg"},{"id":49922818,"identity":"246a0184-95b1-4d4b-a0a8-44095508fcb3","added_by":"auto","created_at":"2024-01-21 05:07:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":972562,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3847949/v1/cd324cdf-4e78-4266-b709-612640d9c3fb.pdf"},{"id":49433678,"identity":"f8bdf894-2f18-42c0-837b-96b16aaaf07e","added_by":"auto","created_at":"2024-01-10 19:15:59","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":528071,"visible":true,"origin":"","legend":"","description":"","filename":"SyphilisSupp.docx","url":"https://assets-eu.researchsquare.com/files/rs-3847949/v1/5dd422015cb1165bb8a99cde.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Discrepancy Results of Treponemal Tests: Exploring the Associated Risk Factors Using Machine Learning Technology Based on 18 Years of Electronic Medical Records and National Claims Data","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSyphilis is a worldwide sexually transmitted disease that is widely known as the great imitator [1] It is transmitted through sexual intercourse and blood transfusion during the infection. Syphilis remains a global public health obstacle [2\u0026ndash;5], with the World Health Organization (WHO) estimating that 6 million new cases of syphilis among people aged 15\u0026ndash;49 years annually in 2012 [6]. Between 2014 and 2018, the number of reported syphilis cases in the US increased by about 80% [7]. Since the 2000s, there has been a growing number of isolated syphilis outbreaks in North America and Europe [8]. The World Health Organization (WHO) has set a goal of reducing the incidence of syphilis by 2030 [6]. In Taiwan, incidence rates have been increasing year by year, and clinical personnel are required to report syphilis cases to health authorities [9].\u003c/p\u003e\n\u003cp\u003eTypically, serologic tests are the important standards for diagnosing syphilis, and the conventional algorithm involves a screening nontreponemal test (NTT) followed by a confirmatory treponemal test (TT). Two widely used TTs are the \u003cem\u003eTreponema pallidum\u003c/em\u003e particle agglutination assay (TPPA) and the fluorescent treponemal antibody absorption test (FTA-Abs) [10]. These tests play a fundamental role in confirming syphilis infection, and their simultaneous application has shown promising results in enhancing diagnostic accuracy. A meta-analysis evaluating the performance of TPPA and FTA-Abs simultaneously for syphilis diagnosis demonstrated enhanced sensitivity and specificity compared to using the tests separately [11]. Although only one TT is needed for syphilis diagnosis, multiple TTs are still common in the disease course. Patients could undergo testing with different TTs across various healthcare institutes. Different TTs could also be ordered simultaneously, particularly when physicians are less familiar with the TTs, or in areas where the cost of TTs isn\u0026apos;t high (e.g., Taiwan). However, studies have reported that certain factors could interfere with TTs results and confuse syphilis diagnosis [12\u0026ndash;14]. These studies revealed that advanced age, autoimmune diseases (e.g., systemic lupus, scleroderma), intravenous drug use, and pregnancy would interfere TT results [13]. The discrepant TT results would cause confusion and delay treatment. Therefore, there is a need for large-scale investigations that use real-world data and machine learning models to identify the risk of and risk factors associated with discrepant TT results.\u003c/p\u003e\n\u003cp\u003eHence, we analyzed the results of TTs using large-scale electronic health records from a tertiary medical center, and linked them with national wide claims data to identify the risk factors associated with discrepant TTs. We also developed a machine learning model and tool for distinguishing high-risk patients. The characteristics are valuable information for interpreting the TT test results and could result in early diagnosis, early treatment, and enhancing patient follow-up care by providing the risk of having discrepant TT results.\u003c/p\u003e"},{"header":"Materials and Methods ","content":"\u003cp\u003e\u003cstrong\u003eStudy scheme\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur study aimed to investigate the risk of and risk factors associated with discrepant TTs. We then developed a machine learning model and clinical decision support tool capable of evaluating the risk of TT discrepancy for individuals. Additionally, we studied the association between the types of TTs and the risk of TT discrepancy after anti-syphilis treatments. The study scheme is illustrated in\u003cstrong\u003e\u0026nbsp;Figure 1\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData source\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA retrospective cohort study was conducted by linking the National Health Insurance Research Database (NHIRD) and Chang Gung Research Database (CGRD) [15]. The National Health Insurance (NHI) program covers more than 99% of the population in Taiwan and the NHIRD prospectively records the standardized data of healthcare services submitted to the NHI program [16]. The diagnoses are registered using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes before 2016, and ICD-10-CM after then. The NHIRD data is routinely validated by the NHI Bureau [17]. Chang Gung Research Database (CGRD) is a database that included de-identified electronic medical records (EMRs), including laboratory test results, from the largest group of healthcare providers in Taiwan [15].\u003c/p\u003e\n\u003cp\u003eTo obtain a comprehensive medical history, we linked EMRs from CGRD with the NHIRD, using encrypted personal identification numbers. The data were anonymized and de-identified prior to analysis to protect patient privacy [18, 19]. The Chang Gung Medical Foundation Institutional Review Board approved this study (IRB no. 201801524B0) and granted waivers for patient consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResearch\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003esubjects\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eidentification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study included Taiwanese patients who underwent at least two confirmatory TTs at CGMHs from January 2001 to September 2018. The TTs include semi-quantitative \u003cem\u003eT\u003c/em\u003e\u003cem\u003ereponema pallidum\u003c/em\u003e particle agglutination assay (TPPA), semi-quantitative\u003cem\u003e\u0026nbsp;T\u003c/em\u003e\u003cem\u003ereponema pallidum\u003c/em\u003e haemagglutination (TPHA), and qualitative fluorescent treponemal antibody-absorption (FTA-ABS). The TTs performed when individuals younger than 1 year were excluded due to the unstable test results. The weak, equivocal, or borderline test results and tests using cerebrospinal fluid specimens were also excluded. Finally, patients who had no outpatient or inpatient visits records in NHIRD are also excluded.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscrepancy test result definition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research subjects were divided into two groups based on their TT results. The discrepancy group was defined as those who had a positive TT result followed by a negative TT result. Other situations, such as the patient with all positive (consistent results) or all negative TT results (non-infected), or those with a negative TT result followed by a positive result (newly infected), were defined as a non-discrepancy group. For the discrepancy group, the index date was the date when the TT results changed from positive to negative, which could be due to false-positive for the prior test or false-negative results for the later test. For the non-discrepancy group, the index date was the last day of the TT. Age was calculated based on the date of birth and the index date.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eComorbidities assessment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eComorbidities were identified based on patients' history of diagnosis. The Elixhauser comorbidity index developed by the Healthcare Cost and Utilization Project was used to investigate general comorbidities [20–23]. Comorbidities associated with discrepancy in TT results, as identified in previous studies or by clinical experts, were included in the analysis (\u003cstrong\u003eTable S1\u003c/strong\u003e) [24–26]. We used the \u003cem\u003edxpr\u003c/em\u003e package [27] to generate these comorbidities from diagnosis codes recorded in at least three encounters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSensitivity Analys\u003c/strong\u003e\u003cstrong\u003ei\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted sensitivity analyses to evaluate the method used to identify comorbidities. To assess the impact of comorbidity definition, we redefined comorbidities as having at least one or two care encounters related to the condition. Univariable and multivariable analyses were conducted for both sensitivity analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModel establishment and evaluation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe demographic characteristics and comorbidities that were significantly different between the two groups (p-value \u0026lt; 0.1 and the case number was greater than three [18]) were selected as independent variables for developing models to predict the risk of TT discrepancy. To robustly evaluate the model performance, we randomly divided the data into train and evaluation sets 100 times. In each training set, we fine-tuned the hyperparameters of the models using a 5-fold cross-validation approach. Finally, we use the evaluation set to evaluate the performance of prediction models [28–30]. We assessed the performance of the models using the area under the receiver operating characteristic curve (AUC). We used least absolute shrinkage and selection operator (Lasso) regression, stepwise logistic regression, eXtreme Gradient Boosting (XGBoost), and LightGBM to build the discrepant TTs prediction models. We estimated the feature importance using information gain in the LightGBM model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePositive predictive value (PPV) score of discrepancy TTs risk\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on the predictive probabilities generated by machine learning models, we calculated “positive predictive value (PPV) score” to better illustrate the risk of individuals experiencing discrepancy TTs. We sorted all cases in descending order of their predictive probabilities and divided them into 20 equal groups (vigintiles) based on their risk levels [31]. For each vigintile, the PPV score was calculated using the following formula:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAfQAAACuCAYAAADAtP+vAAAgAElEQVR4Ae2dC5Usx7FFRUEYTMEcDEEYTMEMzEAMjEAIRMAEzMAc7lv7Pm/pKBTVn5memf6cXKuVVZnxPZkVJ6tn7uiHb21FoAgUgSJQBIrAwyPww8Nn0ASKQBEoAkWgCBSBbyX0boIiUASKQBEoAk+AQAn9CRaxKRSBIlAEikARKKF3DxSBIlAEikAReAIESuhPsIhNoQgUgSJQBIpACb17oAgUgSJQBIrAEyBQQn+CRWwKRaAIFIEiUARK6N0DRaAIFIEiUASeAIES+hMsYlMoAkWgCBSBIlBC7x4oAkWgCBSBIvAECJTQn2ARm0IRKAJFoAgUgRJ690ARKAJFoAgUgSdAoIT+BIvYFIpAESgCRaAIlNC7B4pAESgCRaAIPAECJfQnWMSmUASKQBEoAkWghN49UASKQBEoAkXgCRAooT/BIjaFIlAEikARKAIl9O6Bu0Xg119//fbDDz98//z88893G+cWGPES+3/+859t+mZjf/vb3777+fHHH29m81pD//73v7/H8K9//eta1d/kf/rpp29/+ctffru/5QXY/P3vf//N5D//+c/v8bI+xN5WBJ4FgRL6s6zkiTySGCVI+r/+9a/ffvnllxOaXz9FfMRKDo/UIJDPIlnWEUL8qgaRv5ccIfOPyIEDFbHNA+E//vGP7+NfhVn9FoGPQKCE/hGo3qFNC5jESKHz7e6j3yLfA4dk8R4bz6DLWrGGW4PQJ2Ftcs869t///vf74emawykHLjBtKwLPhEAJ/ZlW80Qufs2YIr65S/I5dy/XLbzfvnmo2QgLMnvv2/G9rPVb42CPgAFYXNr4RoBnoq0IPBMCJfRnWs0TufA2Mt9I8utsiqFv8X71yVsfhW8SBrIUQ+fo/VlkvvlbMJG36M63TMiKN0x88BU199mYmzrMcwghTvT4KLPlgU1s89lIkTFjwBbX2ciJ+NFnHhzxszUPScjpKzHxTVqswe7UNyT6xJ6fXEf8IZMYY3OL7wizozxcX/RorDG+MzdjEv9pCx3XiTjZE+DrfuFeG8aMP3wjzzXj+s3csaFu9u47deZ6Yg951ydjPrcfU7bXReDeECih39uKfFA8FDALnS4stBQ4CjLF0wILgVHcGKOw2pClQPKBiLi3UCODTWUsvtim+NI7hqwkiQ8aczmPHeKeJM8945KjMWNj5oGscsSc9pHXlj6IKWWIm/wYJx5ixbcxfw98/IcYkJGowURSx7Yxmt9cl2Huu28IbmvoYpMPsUpy5qyOeTqemCljT6zmSx7Gl2vrGKSIjPhpg35iR74eElKOdTG/6RusiIVxfU3s0UVmNvGdc9rBZjbkPEQwLq4p0+sicM8IlNDveXVuFNskIQqZpDMLMYU6CXqGQNGjgFIsbZt8EoFy2MYvTYKh6NsgDIqozbiz8HKNbYkJWa4Zy7blMQv0Zgu9JAByJS6bhDlJxXl6bEhQOY7/JAzmNpxSh2sIL2PKeWxiQxwlscRny3PDLO1yLf4zV3z6dqtMrpF2wADZbNxvY1t+5JV7TV8Zj/nOfaxPbMy5bX0u2Y/abF8E7hWBP1bBe42ycb0LAQoYhS0/WZTTOOS1kREyEoMFksKKPCSVBf2oyPpWjy18WMQhI2OUKJBhDNvZ0DE+fBILMh4UlDWuefBIOWxN++rTW+Txgx1iIwc+pxrY5iFAWXwlmXk4yJyVzZ51S4LOOWyKI+OSngTP2KWYpV2uiQvfc23FH5ltjVJ35oYuOtm2/MQmZT2E5Jpu+Wr7aG5bn0v2o3bbF4F7RaCEfq8rc8O4KGBZhE+Z3shReQuqBwPsQpBZYJG1kNLbuJZ4LNba8Q00SQg97PPJRnzqkRPEOUkD+ZkHpIReyk4yTD9cY1tfyBLLEbGmLrJJRMyZc/r3wDDxS1sbls5rM3HeCJZ4zOMUZtq1xxZ62RjzQMc4GM01YnzT3dZgywF991piw/6Zh6nNj/Fqw3t7ckLPZgxidLQflW9fBO4VgT8+rfcaZeN6FwIUKsn0lKGt4Kb8qeKZchtRUfR90ztFUmlnFl7mtrHU4do8kni2mM7ZIuaNrKa/vJcckmSZ178YMLYRVNri+hTm2kydLeZzeaZ+Xs/4iH3isR1esIHcPESaS2JwRLr4nvrc5zcs+pkxmcN22Nj23jamjfZF4JEQKKE/0mq9IVaLVZLbkRkJIgtuylqQcwwCm0V2ylGc883UmOhtvIlRgG1HxLiRE/YzZnwhhw3bJCfGN1vK02/kCEansDwiKDCC/LJtBJXzXIPJfCtV5tKctjwnZtrMPvNnfbhPnI/WCBvIZtzokf/EYCNd9NHN/YA+eeQ+Qg57c/+ZA3PsxWzb+lyyH9NGr4vAvSJQQr/XlblRXBTFrRBu5jfSSTkLn0WVe0gpiRN5CV2SngTIeBZiijUFPOX0RU8Rdk6SwQYfiIlPti0PCcaYkCd2xrWFj7RFHsQpiREHcSJ/1DL3JM1JUOizLsQKfkekxPoRAz7BIkmO+DNebRIDstq8BLMtH/WIj+u5zkdrhC18kx864EecfLCTTR/ImBu5ogveNn2x91gn9wPYaJMxc0YPGxOLbX3wd24/Gkf7InDPCJTQ73l13hmbxYvCZnE9ZdLiekoGmxQ/7FGAZ5FH1+KLPa63RmGGkLAD2XlIUJYiyzy+srBLLugxTzyzbXlQ7NFhzpjpjUFb+LVxDQmhRxxc57xy2ZMvstjLnLAxY9X2KbvEmHiLJ3FgM7EhDrBkPIntEswyB6/dP8TnocY5emIgT+KbuTFnLOwT76cc8RNvrov7x1z1hb3py3VlfMaJX2wnvrk+af/cfsy8e10E7hWBEvq9rkzjKgIPiIDE7aHJFDgQQK55yHGufREoArdB4GaEzsM6P771zFCnHPdTlnvlOL3bPL07l6dsZdoXgSLwNQj4xj2985bMM91WBIrAxyFwM0Ln5A3J+pUaJ3V/jubPu0zjEllO+BI9trJpV1851+siUAS+DgG/AvegzbMryfft/OvWpZ5fA4GbEbpvzj7Iwpck79ilsvxcLN/O1ee0v407374IFIGvQ4AD9/yGrWT+detRz6+DwM0InbdlyDsbp/ON0C+VhbQncfuzuO2XdNI31xSRLCz8Us08cCDjL+8QK9fZeONwfv5CDrawzzjXxOQv4tBnm3bmtxYp2+siUASKQBEoAtci8EcGvlY75DfyPfqq7VLZTY6xS75q9yt7idNvBfJNgTlIXJn55s+bBmStDr3ykLe/PcsYBE5sHGLoyd2GnKTP2JaXsu2LQBEoAkWgCLwFgZsROqQm0fpzM0gMMpvtUtlJfBDqpb9Y42EifWPP3771TT+Jl/iN1wNAzmMrY9c2Y8RF3rN5aNAv85L/lM17YsXuJR9ibSsCRaAIFIHXRuAmhC75JfnwNbUEnxBfIwvxcSigQZaQ5qXk5Ru6BJ0xcO1b8xz3nnnymSTNWOalH9/y1bcnZmNAFl1s+Nav3Ef0uR69vuxwVJyK07YHPuL5rM0icGsEbkLoktQlP9d+iyxJozd/Ln0ODN/SedudsR19e6BNiBi9bL7VE4tNH5P4mZfsLRAcciB3xtuKQBEoAkWgCNwSgZsQOsQHAV7SrpFN8oeAN9I855M3eogU/SR1SDaJedrZ5revzyFo7G/NbyMu/VYhbYCTB4Fz/Vvsp69eF4EiUASKwOMjcBNCh3D8WvkcJNfISuiQ2/xZ9ik/6OVX2r5ZJ/ERx7WEvr2151fqM6aN0DmUXPtNw7Tb+yJQBIpAESgCE4F3E7pfOfMb4efaNbLY8o346A34yB+EiY5v9PrNr7olZ2T44CsPJX6T4DxzvOWnDeY4GBwdNphHR2w4WBDX0c/bj/LpeBEoAkWgCBSBcwi8i9B9A4XUThEbQVwja9DqJIk6d6qHYCFO4+I639jRxSakjgw9b+sQsI3r/NqbQ8KMw/joj5q/mY+fLY4jvY5/PQLun/xGxTVn7ugg9/WRvy8C9z2H0bYiUAQeB4F3EfrjpNlIvwoBDkYcljw8QRJ825GHp6+K7ZxfYoS4549mOKQxfuogd872vc9z+ITY24pAEXgcBEroj7NWDxcphAgx8JH8+JYDUn8Esth+94JF8EdBD7cgVwTst1ZXqFS0CBSBL0aghP7FC/DM7iFyyHu+jfM7Bbzh5r86uEccjoibbxge4UDyVkyPDjJvtVe9IlAEPgeBEvrn4PxyXiTD7RcA+Qo7v7KGQPy5rV9vcwiAOJHzlwoFEZscFpjjwKAPvgXgzZIxrrHLz7+Ry5+DY4eff/tjAGxhY8rgl7nZGJsxpcyMg1zMj94DDjEQK3Ewhp55zZ/PH+WMngekPGSAI7Zn/Jfk7Y8UjNPcuMeuuNH7eyUzl/wdFddUO/TY8hdN5zo6f+Qr7fS6CBSB3xEoof+ORa9uiADFns/WKNQUcQiMBpFS4CEfSQmSghToHUNWElCXOT6QN3PYkcAZ554+CVIC9CBAjw762Yhnjmlf3ZTnesaBL/JjnJjxQyzo0zuGH+OHiCVKbDLuIYV7c+Y6ccI2Ddt8JObvg/+TRcbY6be8sTnXzvUBE3LhnphYS+xkLlxjAxl64zIObblu2EPGNXV+86WN9kWgCPwZgRL6nzHpyDsRsEBTzLcGQc0ijxxj822Oe+1IQEl2kCXEkA07EBLEMJsECulkQ0eic3wbU58cz7UtDsYyR+3Nbwe0fWnOSfLqYhtSpOnnkryxxRpl435iKqErp4/MhVzJOdtmCxkx3eanr7TX6yJQBP4fgT8+aUWlCNwAAUmIN8StQQwSjfOQL0V9kqpvacihJ9FA6pJF+mF8s6Mf9CGHbBJRHhS2MXTwOfXTltdHcRBbEjoEm2Smvv0lOSMLTmKjLgch8bw0b3SJJ4nfA5q2wAbSBgdJGD1zYd6GXK61ttK+svTOn/OVOr0uAkXg/xEooXcn3BwBiTYJUicSpW/dczzJgGtJSoKEbPhIYNOHpLK9neMLEso3SMbQmSRtDsZnv70JO5c9hESc2SSrJLNJeCl/ac7o4CsPCn5trb1L83Z9ch3EVOzBgPWbGG+5cCDJtdZWHgSMkd75c75Sp9dFoAj8PwJ/rDhFpQjcAIEjMsQ0RZ9iPQu6BJgkAXEotxHNFioHgHwjnDKT+JiHdPCV7Yi4N/3U85o4pk3JypyQnYSnPv2lOWOPuPKbCkg077e4t7y3tdvGMk6vsecBjLEtrnO2zs3rq30RKAJ/RqCE/mdMOvJOBCSufMvDpKRN0Z5tFnKIIQlpIzfIf75tT1KZfiaxEStvrzOmbcw35pnX9ME9h4pJ6JNAJTy/Xp52Ls15ymEv34qxe2ne4DnjnmuDPbBIH6wFPsDT5nrnIW2zpTz9Nj99pXyvi0AR+B2BEvrvWPTqRghQwP2K12IuyecbXLqzkCMPqUyS06YkAhlCmimHzCSV9MF1kqrEJ3lDjNqXAHMsiZN80vf0gz6+iIkPeeOHuG0cWJCDsLZ2Sc7oZVzguGF8ad5+M0GcHpa07wGLe+xl3FsuuabEhE0wI2dtYQM/2rrE14ZVx4pAEfj2rYTeXfAhCFCgIQeKN59JvtOphRwdrrcGCUAk2pMUlNXGkT5y6ECskjhjEAo2IXNI9GiMOfyjm2+i3xXiP8YBoSGLbXwkmSMu4YXqny7P5YyCxE9sR4eMS/P24MU6ZLwzFwnYgLdckCF/4sq1AmcwEZe5Xud86bN9ESgCf0SghP5HPHpXBN6NgIT0bkM1UASKQBG4AoES+hVgVbQIXIIAb+O84bYVgSJQBD4TgRL6Z6JdXy+BAF8x+7P4l0i4SRaBInAXCJTQ72IZGsSzIMDPs/nZMD87nj8bfpYcm0cRKAL3iUAJ/T7XpVEVgSJQBIpAEbgKgRL6VXBVuAgUgSJQBIrAfSJQQr/PdWlURaAIFIEiUASuQqCEfhVcFS4CRaAIFIEicJ8IlNDvc10aVREoAkWgCBSBqxAooV8FV4WLQBEoAkWgCNwnAiX0+1yXRlUEikARKAJF4CoESuhXwVXhIlAEikARKAL3iUAJ/T7XpVEVgSJQBIpAEbgKgRL6VXBVuAgUgSJQBIrAfSJQQr/PdWlURaAIFIEiUASuQqCEfhVcFS4CRaAIFIEicJ8IlNDvc10aVREoAkWgCBSBqxAooV8FV4WLQBEoAkWgCNwnAiX0+1yXRlUEikARKAJF4CoESuhXwVXhIlAEikARKAL3iUAJ/T7XpVEVgSJQBIpAEbgKgRL6VXBVuAgUgSJQBIrAfSJQQr/PdWlURaAIFIEiUASuQqCEfhVcFb5XBP75z39+++GHH75//v3vf99rmL/F9euvv37761//+lvMP//8829z5y7QNdep99NPP337y1/+cs5E5+8cgR9//PHb3//+9w+N8tQ++lDHw/h79yzPO8/Dv/71r2H5/bc8X9j+z3/+85sxnz3ivrdWQr+3FWk8b0bgH//4xzcK4b03C6kFCGJn7Jr2yy+/fC80Uw8yv8dCc01uXyHL3rFQH/XI/Pe//z0rhz5yb22QBzbmYe2t9k7puY++8hD83j3LcwReH5EDh6pZU9wDvETcWyuh39uKNJ43I0DBfQQy+9vf/vb97fzNiX779s1C/B4br6IL3uyNU41DFcVbIoYgfEO2gEMc4E6B9yDlN0O+weHn3r4h8QBpbomD+eTYq12zhqz/pc0Dl3vgUr3PkCuhfwbK9fEpCPBQfsZbzXuSsbi+93QPcVxThN4T8yPr+vYGcZ1qkLSE5xr5DQp6HAoo5Kxbrh3jSeDoehA45e+z5siJfXK0Vx7lEPxRePFWz9qfO/Clf/dUjt3LdQn9XlbiweKAOHkQKGYUDR4MrnmzyYKnXBZMUkUeWYsmhRB95LimeFIskaHHhzKMbQ8g49jFP3b4bIUcnxQ45JE5FQNv/Mide/MnNmWRz7c9l9Z8mPeTWCk3e3InX2LVNvEnBtjRJvLZwMC1QQZd4s2GfspgW7/omD/rqRxY21gvcjZG12yzAd7IJfbaoU8fxIq8/nNfIEcjVmIkLuZtxiIu9MR1rmEX2bR1pIOPUwRO/swbwzl59wh524gHPfLD3tGzpry9eejbPtcef8jlHuF+tqP1nXLc55on3j6X5nZqz2on9xTXfDI+c/JZSN/umdxvWQ/yedUOPQ2MHEsd5vBlDt+F//efazBKvVtel9BvieaL2OIB4aFy03M9CQAoplwWEuZ4YHgIfBB4GBnDFoWAcWwzxsPPw5wPbMKNbYoeutim8dBRBLNZIIwFP/rSPv542BnHH30WkbTHNQ+8MXJP3EeF27yxe2kjDz7GTJz4M0/tbPluByd0s0hhD5y0zzV4gyVjFl7k8MkY+dnwYb7kxTw+6KcNcBRLcE072EM+cxMvfNNYF/BFl4/28YsesWYztxw7d42PuW82nYntJoMt1oX4+Hi/yTJmHuZL/uAlplyTM3JidWSLcbBKPKcsc9jRH/HNNTm1vtMe99hDJ+MjDz4+K+pte5Y58mOODznQiI14wcKmPWzT8Mm8e1b8mMMWeyYb89jUR84Z/5zDjngpfy1G6t26L6HfGtEXsmeR4UGz+SB5T68cm97GwzILB3M8XBRTHmiaujxEts0HY9iTLJCdcj68GQexz4d8xqDfrSdO/CYGyEk4U+dagjGHjNmxWWjwOQsNeJBPNuS0t2HCPOM2saXfGuuV+WMbn6ydbbNhHsq41rmGzGEr42GMGME9fSA3Y9yKr/6OeuxOHDdZsZ3rkLLYypiIN7FKWa+3fMUmdSd+6mcv4W0xbja35/KS9U2fXrNGfLLhM59l5jesGQM76wA2tnzNYeaH7MR+i4d8jw5v2NjmtvV5K0aJzS2u//ik38JibbwMAhY0HiobBScfWMa3B5GHK4sTchJBFm8JJ33wsE8f2Jtj0y8PncUDX87n2+oWg7ltvfGlDeS24sE4Mc68N7uOzULBOHFTrGaj0EwyNB/znjqb/SlDvMhtzfwpqBRfcCDHuRabDfHXrkXce3oLNnlkI38wtpnnXIcNE3W2XjsTx02WnLZ1SFllZvwpk9dH+V76rKUtro/2CnPaTDKcz+Wl6zv9cs8emPsOAsWmbVsf9hHjyGbDVq45c0f5iXseCFiraXPmm/62uW193oNR+rvFdQn9Fii+qA0emlm4KfzbQzMfRB5YHsZsFph8CLE1yWTzwcM6izA+1bVQ45ePxWYW2i2GjHFeU2SwNxt+Z84Wqpn31PXemLe8wD6bsnnwcd6ciCeLtzpZYNXJfiuEzhOHmCKHjxkvsuAx854Ejv7MC1uMZzPuJG+Lau4di++GSdrLa+3g41wjpxnv1AFv9ho5bLhMeddqjuPnkmdt6rEecx8qg805x1pmnMhcsr7azB69XHPWJnNwHef6HK0beM/asuVADOCdsqwD8eSeQW7GmPFjI+Nnjnt0sr0Ho7Rzi+s/RnYLi7XxMgjwgOUJ/OihmQ+GRXM+yNjCZjYKAA+MbfOxFQbJ04f6qEho154Ysug4ftRvBdMYszCif2kM+trktT0LzRERpC3yYi2wcWk8+puFULtb/s7Za2OuN7Hk/tmKK/thko77xzzws63bVnyN6ajHDnGda+Y013jTYy8ai/txk2PsiKAufdam3Q1TZbCZ+2jbb5esr/ayF5/cN+Se90d7VqxyfY2Ntc82awtz+k5Z98ylB76tpmB7w2Mbyxg/87qE/ploP5EvCTML2vbQzAcDPYiFQpMPF9BM8mZsFiR95MPuWMI7i4IFIUkF/3lYQH8WzrS5XW8Ps6Qw8zOmzc42tsWMbTDJPNDdiAB/WUAtdOpu9mccYpt4p8yWPzp8bMQwY9ausSA715q9tRVsiGGSLus2yRJMrjmcEQN2yOlc2+KfOtM3a3fO9pYv+whszj1r07/PXu4BZdwLObftz0vWV5vZz70FXtv6bHgYRz4/1gxysplf7iHm3G8pC/ZzPTxQpB9tO+e9/bY+b8VIm7fsS+i3RPOFbG0PTT6IPEAUDR86HhAeHDb/JrcVrVkUgHfTleQsTj6M9Dbs8zBaVIiNBzyJZ4tB/aPeeCwe+k676l5LMBZdfNDIU39ggy/9WFTQ8ZCiP/KiGZuxaj8xwX7ixtwkz+/G/vcf5JnHlj7AVZ+MGbN2iT3X4n+m/kCmEgBy6KNjnNg3R3UhPObJTTlk0CcW9KeOuvbIYGcWfuezB29k3XM5x7X73vUBny3uqYfNme+lz9q0ZQzYAwN646HHl+uGLvOMIevzy9i59Z1+uRdLemxgb7ZtzyJjvuwXYmE9jS1tpA9kzQ15Ys6GL/el+wAd8sUOODAuHvpLLLC3rc9bMcr4bnVdQr8Vki9mxw2faVNAeJB4y8lCZ/GjR4aHZsrlw6lNHzgeKtvmgweVh5meB276V5eYmEMG2YwRmS0GdU/1FBBsHtlVl5y3wub81ouzOYEF19gCH5tYiTHjjInJUWzghi3nLYraxR6fo2bBQ9/8cr3QQ59iSY8c8Wfs2mY9sMGHvGnooQPG2uXeeXXBlXF65dwrjGOH9T1qyCLnZ9pPPWNSNue8liDE9ggb5e21nXm4B5ShNzf3Rc7ltbgghx2x2UgP/MkJWezTkNfGpTmohzy25p4yvm3POqdP9i+xEO/ch8SGffzkeiE3ZYmB3BjP3Hw+6DNO9orxZ51wfXI/vhUjc71lX0K/JZq1VQQOEKDAU1A2IjtQeZrhV837aRbwExOByLdnBOKGRNtOI1BCP41PZ4vATRDwDcGv9G5i9AGM8DYEoZ96O36ANBriJyBwtFde9dl5C+Ql9LegVp0icAUCfjXI28erNYsxGLQVgVMI+C1Wfn3uj2H6dn4Kud/nSui/Y9GrInBzBPy5PGT+iqRGIeYNnZ89thWBcwhwAPTn2uwbrpPgz+m/+nwJ/dV3QPMvAkWgCBSBp0CghP4Uy/i5SXBy7qcYdA90D3QPnN8Dn1mdS+ifiXZ9PQUC/NMXChn/rOXeG1/z87U3sd5bzP6zpUf5RUF+bMBvW390u4f9dYu1Yc99xO+N+MtzfD1vuwfMjOUr+xL6V6Jf3w+LAD/be4SfC1PoiBVi5xeMuL+XRrF/hEOReEHmn7XmX72/3rs2/oLb9k/QxPOtvb9o6b8n1w6Y3dP+Nq7P7Evon4l2fT0NAhSPjyhWtwTIX8jLP4xxS/uX2OIXmsDqKxqHGA4MX5n/qbwhn6Pf3n6E/XUqt/fOuXev+UVSDlyv/gt0JfT37rzqvxwCFBm+vp5vCPcEBDHyNkmcX9XAB0I9Iq2Pjou3TPK/hhQ+Oibt+5a5HTYeYX+Zx0f05M+B5pqDoN8IvPrfO/i6p/0jdkJtPh0CPNzznz5RDCGKfPtKufm12/x5IPfocqJHjyLANcVf8lEGufxZHQAjzzi6kgb682fBzPPGoG16DwHaZ8wYKGDEwNypRjzKEsd8K/HtBlv5OWVTDLDHNTGBI/r03GdDxhjIgZizAHuYSP9c0zK+jdBmfuJ/LkbjA+Ppl/vEiWtyVY545/ppD1nlxOHSWLSRffrVbu5ZbCODr9xf+k5bR1iljNfuOW0zDv6sH2Pkf25t1Jlrjw2b+yb3g77d76wR13NdkBOT7ImL5pjPqT7Jg7n3YqS9R+1L6I+6ci8SNw8uD7NFlQLGQ0+jYFgIp1zCQ1G04Khv4cIu8xQCSYixzYc2mUcW3xQmdCkm2LExRnx8KJTcUzTRnTGgZwzISPrayp48kZEILWTYnI2cZ+GbMtwTn/7JAx0JTpzEA3l9Ikde6KMnxvogJsaZnw3dbY44yM8CDsZ8thiRZdx41NEX8SAzG7kkzqwJsZDLUWMdze8tsUy7xKW9OUc85s1e4EN8uQboHGE17XHPWvBJrGYe4ne0NtrJWIyNeG3giIzY45fYcy+5d/SlLj1xoY/ebMY/57Cz4XkNRtPXI96X0B9x1V4wZoocRZjeZtHznl65HKMYQ1DZLC45jgO8pycAABbISURBVC6FhOJg23wwloSA7JSzYCdJzPi3GPS79crPwk7MiQu6p4riZtsxbFEYM+60zzh5JG7okr8FXFtHRZZ54p0FmCKNrzzQ4Afb2WaM4kJvk1Rm4Wd+s0lOp9qWH/KXxLLZZU9OvJTDV+JgLrnul2KlTXv3uPf2+HPNt7VBzj019xq629jE3nXKvbPFI2njbzZtzLltfd6K0fT5SPcl9EdarReOlSJA0bXoAAX3Sb6M8WDzyUbByWLIHPeMZ2HwNJ8+IJ3pA705ln4tfBY0ipDxp78thox7XhMfvjM+ZBibBdWimOQ47c1737aM2/m0vxVJY5gYg0kWb+3Rb3NgLckRi8WeXGzGmPmKY+Ji4d/yV37mqY+tB4OZ36WxXGpPOfa1ODC25XIJVtrLfjtkgW/629YGG+jO/edeJ0bbFi9z4p6y7A8ON9lY26MD1tHctj5vxShjebTrEvqjrdiLxjvJ20KSxR5okny4t7hkEWF8KyTzrWnzYRGfRIFfi6KFizE+FEiKYRLOUQyMHzUKFLayGWMSHPMW35Q9dy1Zp5z2JTMPJilzCpMZl3qsZ85pQ8xci4mz2CaW4L6RAraOGnkwv63L1DG2uYcujWXaO9qTyG2+Jokpcw6r6Zd79s/cQ9yzzra5No5vuu6ZXA9xUc9+e+bY06xBts2P89iY8W94vgcjfT1if7zjHzGbxvyUCEgq+Ua1FRIf4iy8FPutsOfpHdAoSMhJXIxtPrZi5duwhwsK8OZzLs5WzKZM3mMzSTBjJPdsp4piyuU1WM1iab4W/M3uhvFWZPW1rdMpefXo8QVu2TYctzhTh2vWF/Kah4EpJwZz/NJYpt6p/eGeS52Zy6VYpQ2vJ1bs2STUbW3UZf+RczZim+uxkS46yKW+z7XPjXa3fe7cdtjY8HwPRvp6xL6E/oir9mIxS5hJWhSGWYhn4UV+K9gbeW8FYCvYjFHEss2ithUYYsnCaTGjgF/atkJHkZzxYG+TPecHPKetaZ/5lBHjHMOPa5Fvbvp3znv6DX90IYdsxJhj4jhJgXVPvLVBnBnTRqDK2h8R1KWxaMcee3PvOrft67mWl2KlzezTFmt3tG6p4zW6ianPZa4Hshvpbs+c2Od6EBN+5npi1znyz7atz3swStuPdl1Cf7QVe8F4KSIUiWwUIopiFn2JggffYmWBTDkLETK2jYQ3HxAcsUAk2MQ+BShtWUwsStyjlzJbDMZy1BMPdvCrb2JJu+hyf1QUj2wzjs5mn1xt5ItP/OOHmCyoyFncXQtyd1w7Ym0OjHONXQmDMdY3DzzIECO2bYk1sspji9hojGmXcXKg6Z+xUw07fMxD3UtjmbbBCJ/4J34xQw78jU89/IAZsuRxCVbqzl5b7D9ywla2bW2cZz3cH8bCGDrZ9KEMc9t+33y5d5kjNnrXFHvYpmcPOH60PmDsum/7KWN+lusS+rOs5BPn4QObKfIw83AzJ6Hx0FJwGKcoUhAsJClnIUl7FFWKU7bpw6JAMdFP2k1dfFBQiAXbxqjMFoNzRz354A+bR3bRlUyJ99JmsZxxTxsbxvojNuWJFTyJk97iSzz4AhswZH1sXIsrOjmnHvbQt+kHe8Ruc+0YZy8YF9f6OIWhdugzP9dRvC6JJW1xjY3cG9ogF2LCXzZxlJyYO4dV6uc1ayRW+JuNWJifa4OccTPvwWKuB3Lsd8Z9Bhnb9rv2pi/00GccPeOkZ8z4jT3Xx3Vm7q0YafcR+xL6I65aYy4CJxCgIFL0rmkS+TU6lX0NBCBeiHQ2D00S7pzv/ecjUEL/fMzrsQh8KAIUX0j9msZbFW9vbUVgInC0N96yz6bt3t8WgRL6bfGstSLwpQj49WN+9XhJQBTn/Er3Ep3KvAYCHA7ZH+4per8Fcuw1kLj/LEvo979GjbAIXIQAb9gU3vmz53PKfGXKzyz5mt6f557T6fzrIMD+4C2d/eE+4Z6v4tvuC4ES+n2tR6MpAkWgCBSBIvAmBErob4Lt8ZT8LVNO2Oc+yLYVgSJQBIrAYyFQQn+s9frSaM8dBDp//rBUjIpR98Br7oHPKN4l9M9AuT6KQBEoAkWgCHwwAiX0Dwb4Xsz3K/d7WYnGUQSKQBH4GARK6B+Da60Wge8I5EHqUX4r+NFi5jfz/WtqfJ09/9Laqa2Irl+BTz1+k5t/NdBWBB4FgRL6o6xU43xYBPj33ZDGIzVivvavzX1FfhKyf1oWYr/2n97xz/xYn6kHmUPqbdch4H73oLT1yPjPJbf5HOtforsc/8eqMpfnVcki8B0B/m02xeOtjbdVSOI9jT/C8Wh/hQ3MHoHMwPW96yOhv2eNX0X3kueJ9WDPS8SQM/c0SZwDGLjn3z7wmyH/WA17sN+QXLezSujX4VXpB0LAvzV97R9aMUX/5xHvORBgi6JEsXqkRlGeX0HfW/y+nb8XW9b3vYeCe8PmI+K59HmCpCVz18hvUIiLQwGkzbrl2jGeBI6uB4GPyOcZbZbQn3FVm9Nvf9Uqv7qjYGSjyFDIkaEIZXHh7TR1vU59Co76FCIIcBKDbySnDhXo4R8byPs/w8BnxqRcFkziQR5ZiyZxYQs5rime5I4MPT6UYWw7sDCOXfxjh8+Ww8TwVAxieu7Nn9iUJY582xN/82HeT2Kl3OzJnXzJR9usWWKAHW0in008nCdO3yiVQwb8lcG2fhkzf9ZTObC2YY+cjdE122yAt+sj9tqhTx/kiYz+c18gRyNWYiSuzMtYzImeuM417CKbto508HGKwMmfeWM4J3/k55nHS+jPvLovnhsPP4VpaxZ1SYqeQpFFkeujYqS8xZqChez0p9xRQcMHRQ8SQ59rbVpciX/KIW/LOPFD3hQ/7GGDwsu4xRXCUSZ9aA/bFEvmsE2DDGZu2EDOWPCjL+0TAwTCODHRE8dREy9ipBH3UeE2b+xe2siDjzETJzGap3a2fImbWCRfYkQ3/WMPnLTPNXJgyZg6yOGTMWzasG2+2GUeH/TTBvGIJbimHey5tuYmXvimeRhBl4/28YtP18DYzM37S3p8zH2z6ZH3tg4piy3Whfj4eJ8yr35dQn/1HfDE+fPwW7wyTYukxdC5WcQocFsxophQPCko2SiK0x9FcbORelwbU9q0+KescpIKc8Q5iznj5INv4qWpCy62zQdj2Et8ppzkkHEQOxhkmzHk3Lw+heu0i+61BGMOGbNj88C1reWWX+K+YYIdCZWYxZZ+a6xX7gGJjrWzbTbMQxnXOteQuY00iZE80sd8FtA9ep70ufXYnc/EJkec+JzrkLLYStyIN7FK2Ve9LqG/6sq/QN4UiFnQSJsCw5xEJxSziFHotoKxFW5sbP6ObOjT3oKWRRXfSb7IzsLN2OZDIkgyMe70ARbTB/bm2PQL8Vio8eU8b9i2LQbntt740gZyxMNnNmLc1mfKeT/JknHiTlJWdltL1ygxVZ5+s5/zXBMvclszf0iNvQkO5DjXYrMh/tplbWZerDt5sS7ZkEt8Xbe5DhsmaWdea2d7BqcsOc14j2Rm/FPule9L6K+8+k+cu8Uryct0KahZwBiniFKwKIy2ee/4VnwsXrPYHNnQlj02Z+EmTt6+s23ktvmQfPLQsn3jsPmgsM4ijF+JyFzxy4e4IZCZ+xZD5jKvsYG92bb1Iq8t76nrvTFveYF9NmW3vYMsfsEysVXniOy1D7ZzTZ3TNvYl2RkvsuCR+5SxSeDoz7ywxXg2407y9mCR+Z16ntJeXmsHH+caOc14pw7PKHuNHDZcpvwr3v/56XlFFJrz0yEw31gywY0IZvE5VcA2Ut3I6JSNjIdrCho2bB4wstAyRzHLYm7ck3ywhc1sFMMsmpsPC3zakzwlokvzIoZ5SMl45vWGqzHOAn5pDPrY5LWdeCKPL/bIUQNz1iFz2+xPff3NNVVuy985e23gLxvx5P7Z9jj7AR/Z3D/YtW3rdup5Um/22CGuc82c5hpveuxFY3E/bnKvOna8a18Vkeb9FAhAXFlwM6lLip1FPd9StDELLyRI4ZrFUhvqHfUSZhY0C236n2TLHDmST8rhh/Ekb8Zm3vrIYu5YxmoBVW4jL/xPf/OQkja364krMpLCzM+YNjvb2BYztsFkkiN5zLXkPmOYOG32ZxzqiOOc3/JHh4+Nw8CMWbuZx1xr9hZ7FNyyQYqTdFm3SZannqe0l9fbASLnvd7id85+Psus3VwjZV+5L6G/8uo/ce4UIAoVRZhCl2RDIaDYMMdH0sivBiVjdCnA6FuIlUcXHewxT68s0Eo6+lB/wm6RTv+broRObNjE3ybHHAU9Dwgb4Wy6kpxvkeKQtrAPthZ98qLgJvFsMcy8573xiIO+06464D2LvHNbT4xJcuSpP7DBl37Ada4l+aJDIzf3gL60n5hgP3HbyFN9euTx4z5BlxzxZzNm7RJ7roVySabkpW/00TFO7OezgT44Mc86KIcMfogF/amjX3tksHPJGoE1su45bdi7710f99u5GNR/pb6E/kqr/UK5UgQoQBQKHnwKjI2iZBFxXhJJGYqRRcliwjwFhYLJHIUdexRY7iUC5PBJDMimf33YW6S9pzd+dLPQGTc9MsSiD+UspunT+JIcNh/kTK7mPv0bI77EAFl9O7/F4NypHgIBR3GfdtUlZwnWsXO9OJsTWHCNLQkSG2Ilxozhy3yJjT019wy4YcvYc89gA3t8jhrx4Ad988v10ga+sYMcMWXs2gY3bPAhbxp66ICxdrl3Xl1joFfOvYL8fJ7Us0cWOT/TvnL0xqRsznnNHkdObI+wUf6V+xL6K69+cy8CD4gABR4C2IjsAdO5KuRXzfsqkF5YuIT+wovf1IvAIyLAmy/EBrG/UvPNN795eaX8m+t5BEro5zGqRBEoAneCAF8B8zUzXwe/WvMg49fgr5Z/8z2PQAn9PEaVKAJF4A4Q8Ofy+bPdOwjr00Lw9wv4eXJbEdgQKKFvqHSsCBSBIlAEisCDIVBCf7AFa7jnEfA3Ztv//pvGxaJYPPMeOF8VXkOihP4a69wsi0ARKAJF4MkRKKE/+QI3vddBwDewR/oZq/+emn9bfO+NX0bj59j576HvJWbWnF8WbHttBEror73+zf6JEIBwIPVTf8jjHtPlD9M8wiGEwwexgjN/uOXUH4n5bJwh80fA8LNxeTV/JfRXW/Hm+yEI+BvY7/knRRAEb4Bvbf7BlUf7d8qQ5L3/kRjX9+gv1711za7Re+/+uMbXlDX/9+zvabP3t0eghH57TGvxxRCgyEFKfN7a/DfG7yEMbbw1hq/QAzu+VZh/RvUrYjnySYy8/RLnVzXX9j37462x32J/v9V39a5D4Ot26HVxVroI3CUC/t1vf35tn2/JvDnzb6f92StvWklgjqtLP7/OxQ9fqzLHwYECP79i5e3+1KGCOLCLDd+I/bfN2PYvr1HAHd/iQB/ZlDMW7JIPH2JMGXxk3iwoOCGLHBhhO2Nx0ZnnRwlikLb0yRhy2ASHzFM7sydGZYlj/rjCN1Ns5WfayXt0iAV7XBOTuNNzny3jRY98ch2xk765dl2I17lpFx8zP+5p52I0PmLRfvbo24ghY2QvuJeUaf85CJTQPwfnenliBCheFDuLZabKHMWOIkfBtbhTuLNBZnPMeclV+/T4QycbJDDHct5CCxnwwa6Egz3JzHHJIm1knFMOeQiAHMkFH/jcfGgTHWSQhezRnbkxRm58PEiAKbpggU98qEeM6CAzDxD6pSd+ZHzrpceGOKcs+SB/rhGf/rGFjrgbI/Ha9IkcMbuX5l5I3NW1B5cpzxw65Idfmuu+xYgs48ajDnrGtOFCLomze4Zc2j4fgRL652Nej0+GgEWQwjcbRZRimwXOopeyR2S8kQB6G/FsY+nDa2LKQq894srGPXLZiBOCyqZc6uODeMDGxn3KMG4sSbyM8bFJZokhcaUtcZqxaWP2yie5IrPFeIrQpt28x9Zc+7RPPuQxYyZ3cs52tD+Q2eQ99CWu+Elc0Z0xigu97dT+3mzOPaOd9h+PQAn94zGuhydHQEKbaUoE880OeQppNu4nuTDvW1bKWnSzWG9jqZPXFNws7NjBf5Iv8sikHGNbnBT1+YbIfeqKxfSBvYlP+lXPt0PyxB85MGcDO2zlmHNbD67I5yEBOcbyoMCYhJZ4bzZzTEyN27m0v5EucsjMvbCNpc0pD/4eCojFPZf4G2PmK46JC/NHJK38zNPY2n8uAn+sKp/ru96KwFMgkASUCVmw822HeclEWcl4yjFPIYXAslFEZ4G1YKfcdm0Rz8JunFnE0U3y4f4oTmJJUsYOulnkNx/GQp8NXclIwmCMD1j79XTqgBFvsZe2eeBAz8NDEhzj+MP3Nc18U0f75ETzYJIyGyZHuKOnfO4dx8TMt/uJs9jmuoP7xPFofxs3eeBrWxdl2n8OAtft0s+JqV6KwEMhQDGbJEACRyQLmSRJH8lhY7O9kdG5oiugEk2+yW5FXFJIopgHEWxKUkne2xvt5kNCMTZ6dT1wnMIm9cAkDxU5t11vuIrNJL5LsU0/5IteNvMV+83uhvEpDLSZfk4dAFIOX+CWbcNxwyp1uAY7DnbzMDDlev+xCJTQPxbfWn9yBCQ+CSjT3QqxpJFEeertchZTCjiFE9vZtrGc9xrSQzbbVsQnUZDnVrAl4CTBLW98QCDZNtKD5JJkNlv4SvLeDhXpZ7ueuCKD30nCjG+ym80cg9imrWl/EroYT71T+4O5Kb8ROm/heYgkVmLMMXHMvUxM5J9j5onffLt3bzvf/vMRKKF/Pub1+EQIWPAgHoobPYWNZmGFHL2HFCexUVQZRx+dLLJJAtiVkPGDrMQm6eTYd6fjP7OIM40udpIoJXTG+FC8iRv9JAfjSTfITpLZfJAbeUMk2MQ+cviziaGEwj16KbMdKtQ/6okPO/jVN7GkXXS5Jyb9H9mb4+hs9n07R558XXf8EBNrT4+c++DU/kB2ypMPdt0b2GLd3Jf4RoYY3ZuMJdbI8jH/bX/jw72cGE4sev95CJTQPw/renpSBCQiCriFz1QpihQ+Czzzs1E0laF4U1htEAlzfNRFBnsUbAopbRvTRvboacdx46fXHiRAPsg7LnFCIMRMk1C0RU+skonj04ckA6HoJ+2qR0+8iY++lWGeOK9p5Ik/9PiA37SLPQ82xHtpkxhn3NPGhrH+iE35U/sj5TN+1kpcIfN5IDHG3GtggmzuNXJ27eb+ZlwfpzC8FLfKvR+B656C9/urhSJQBIrAwyAAaUFw1zSJ/BqdyhaBWyBQQr8FirVRBIrAUyLAGyikfk3jbZ837LYi8NkIlNA/G/H6KwJF4CEQ8Otsv/q+NGgOAfNHDpfqVq4IvAeBEvp70KtuESgCT4kAb9gQ8/zZ87lk+Tk0P0/ma/r8+fQ5vc4XgVsgUEK/BYq1UQSKQBEoAkXgixEooX/xAtR9ESgCRaAIFIFbIFBCvwWKtVEEikARKAJF4IsRKKF/8QLUfREoAkWgCBSBWyBQQr8FirVRBIpAESgCReCLESihf/EC1H0RKAJFoAgUgVsgUEK/BYq1UQSKQBEoAkXgixEooX/xAtR9ESgCRaAIFIFbIFBCvwWKtVEEikARKAJF4IsRKKF/8QLUfREoAkWgCBSBWyBQQr8FirVRBIpAESgCReCLESihf/EC1H0RKAJFoAgUgVsgUEK/BYq1UQSKQBEoAkXgixEooX/xAtR9ESgCRaAIFIFbIFBCvwWKtVEEikARKAJF4IsRKKF/8QLUfREoAkWgCBSBWyBQQr8FirVRBIpAESgCReCLEfg/6TmwVjbfjzoAAAAASUVORK5CYII=\"\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eUsing the first vigintile as the example, cases with top 5% risk of discrepant TTs were analyzed. To generate the PPV score for the first vigintile, we first divided the number of cases with discrepant TTs by the total number of cases in the first vigintile (numerator), then normalized it by the overall prevalence of discrepant TTs in the whole study population (denominator) to generate the PPV score. The PPV score provides a more explicit method for clinical caregivers to illustrate the risk of getting discrepant TTs in a random index case.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eReview of\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eanti-syphilis treatments for the cases with\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ediscrepancy TTs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe reviewed the cases with discrepant TTs to determine if they received anti-syphilis treatments. The status of treatment was classified as \"had treatment\" or \"no treatment\" based on the guidance of syphilis treatment [32]. Cases that received recommended or suboptimal treatments were classified as \"had treatment,\" while those that received no treatment after being diagnosed with syphilis were labeled as \"no treatment.\" We also recorded the time of treatment. We analyzed the association between the status of treatment and the types of discrepant TTs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe differences in medians were tested using the Kruskal-Wallis test. Chi-square test and Fisher exact test were used for the univariate analysis of the categorical variables. To identify factors associated with TT discrepancy, we performed multivariable logistic regression with Firth's correction, which is a solution to the problem of separation caused by rare events. We reported the crude and adjusted odds ratios (OR) with their corresponding 95% confidence intervals (CI) estimated by the profile penalized likelihood method [33]. The analyses were performed in R (version 4.0.3) and SAS software version 9.4. All statistical tests were two-sided, and statistical significance was defined as P \u0026lt; 0.05. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline [34].\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003ePopulation characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn total 11,135 patients were available for the study after linking data from CGRD with NHIRD. Of these, 5,780 patients met our inclusion criteria (\u003cstrong\u003eFigure S1\u003c/strong\u003e). Among patients received at least two confirmatory tests, 133 (2.3%) had discrepant TT results, and 5,647 (97.7%) had non-discrepancy results. Patients with discrepant results (67.8, IQR=35.5) were found to be significantly older than those with non-discrepancy results (44.8, IQR=36.2, P\u0026lt;0.001, \u003cstrong\u003eTable S2\u003c/strong\u003e). There was no difference in the distribution of sex between the two groups. Treponemal test pattern for the cases with TTs discrepancy was shown in \u003cstrong\u003eTable S3\u003c/strong\u003e. Most patients with tests discrepancy have TPPA-positive results followed by TPPA-negative results (94, 70.7%).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCo\u003c/strong\u003e\u003cstrong\u003em\u003c/strong\u003e\u003cstrong\u003eorbidities\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eanalysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe crude and adjusted odds ratios (aOR) adjusted by age and sex were shown in \u003cstrong\u003eTable 1\u003c/strong\u003e. The risk factors associated with discrepant TT results were HIV and AIDS (aOR = 2.6, 95% CI 1.4-4.7), other and unspecified osteoarthritis (aOR = 3.3, 95% CI 1.3-7.1), and pregnancy (aOR = 5.0, 95% CI 1.8-11.6). In the sensitivity analysis, the crucial factors associated with discrepancy test results, including HIV and AIDS, and other and unspecified osteoarthritis, were identified as risk factors in both comorbidities definitions using one diagnosis (\u003cstrong\u003eTable S4)\u0026nbsp;\u003c/strong\u003eand two diagnoses (\u003cstrong\u003eTable S5),\u003c/strong\u003e too.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003cstrong\u003e.\u003c/strong\u003e Association between comorbidities and discrepancy syphilis test results\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"42.42424242424242%\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eComorbidities\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"57.57575757575758%\" colspan=\"4\"\u003e\n \u003cp\u003e\u003cstrong\u003eOdds Ratio (OR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.54545454545455%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCrude OR\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(95%CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.363636363636363%\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"29.09090909090909%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdjusted OR \u003csup\u003ea\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(95%CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20%\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHIV and AIDS (Acquired immune deficiency syndrome)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.68\u003c/p\u003e\n \u003cp\u003e(0.92-2.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.070\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e2.63\u003c/p\u003e\n \u003cp\u003e(1.4-4.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.004\u0026nbsp;*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDeficiency anemias\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003cp\u003e(0.2-2.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.733\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003cp\u003e(0.17-1.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.361\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRheumatoid arthritis collagen vascular diseases\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.91\u003c/p\u003e\n \u003cp\u003e(0.74-4.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.128\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.87\u003c/p\u003e\n \u003cp\u003e(0.75-3.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.163\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCongestive heart failure\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.14\u003c/p\u003e\n \u003cp\u003e(0.83-4.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.075\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.31\u003c/p\u003e\n \u003cp\u003e(0.52-2.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.540\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eChronic pulmonary disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.13\u003c/p\u003e\n \u003cp\u003e(1.24-3.44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.004\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.49\u003c/p\u003e\n \u003cp\u003e(0.86-2.46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.146\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDepression\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.87\u003c/p\u003e\n \u003cp\u003e(0.78-3.78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.115\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.44\u003c/p\u003e\n \u003cp\u003e(0.62-2.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.366\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiabetes without chronic complications\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.84\u003c/p\u003e\n \u003cp\u003e(1.11-2.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.012\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.13\u003c/p\u003e\n \u003cp\u003e(0.67-1.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.638\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiabetes with chronic complications\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.75\u003c/p\u003e\n \u003cp\u003e(0.88-3.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.080\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.14\u003c/p\u003e\n \u003cp\u003e(0.58-2.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.688\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypertension, uncomplicated\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.67\u003c/p\u003e\n \u003cp\u003e(1.13-2.42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.009\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003cp\u003e(0.56-1.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypertensive heart disease without heart failure\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.39\u003c/p\u003e\n \u003cp\u003e(0.49-3.12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.474\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003cp\u003e(0.33-1.96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.792\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eLiver disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003cp\u003e(0.23-1.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.211\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e0.56\\\u003c/p\u003e\n \u003cp\u003e(0.23-1.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.124\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eFluid and electrolyte disorders\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e4.46\u003c/p\u003e\n \u003cp\u003e(1.53-10.37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.002\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e2.57\u003c/p\u003e\n \u003cp\u003e(0.91-5.96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOther neurological disorders\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.98\u003c/p\u003e\n \u003cp\u003e(1-3.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.034\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.14\u003c/p\u003e\n \u003cp\u003e(0.57-2.09)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.684\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePeripheral vascular disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e3.82\u003c/p\u003e\n \u003cp\u003e(1.46-8.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.002\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e2.35\u003c/p\u003e\n \u003cp\u003e(0.92-5.08)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSolid tumor without metastasis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003cp\u003e(0.14-1.47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.310\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e0.4\u003c/p\u003e\n \u003cp\u003e(0.11-1.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.059\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eChronic peptic ulcer disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.47\u003c/p\u003e\n \u003cp\u003e(0.52-3.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.404\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.14\u003c/p\u003e\n \u003cp\u003e(0.42-2.51)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.770\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eValvular disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.28\u003c/p\u003e\n \u003cp\u003e(0.8-5.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.077\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.62\u003c/p\u003e\n \u003cp\u003e(0.59-3.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.317\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eArthropathy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.58\u003c/p\u003e\n \u003cp\u003e(0.62-7.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.114\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.92\u003c/p\u003e\n \u003cp\u003e(0.52-5.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.290\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eInflammatory bowel disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.5\u003c/p\u003e\n \u003cp\u003e(0.46-3.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.429\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.6\u003c/p\u003e\n \u003cp\u003e(0.52-3.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.372\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eInflammatory spondylopathies\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.24\u003c/p\u003e\n \u003cp\u003e(0.78-5.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.085\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.69\u003c/p\u003e\n \u003cp\u003e(0.62-3.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.280\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOther and unspecified osteoarthritis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e5.18\u003c/p\u003e\n \u003cp\u003e(1.96-11.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e3.27\u003c/p\u003e\n \u003cp\u003e(1.27-7.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.017\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOther and unspecified soft tissue disorders not elsewhere classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.07\u003c/p\u003e\n \u003cp\u003e(1.07-3.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.019\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.33\u003c/p\u003e\n \u003cp\u003e(0.69-2.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.371\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePneumonia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.52\u003c/p\u003e\n \u003cp\u003e(1.17-4.78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.009\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.66\u003c/p\u003e\n \u003cp\u003e(0.78-3.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.178\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePolyosteoarthritis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e3.41\u003c/p\u003e\n \u003cp\u003e(1.18-7.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.010\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e2.07\u003c/p\u003e\n \u003cp\u003e(0.75-4.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.148\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePregnancy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.79\u003c/p\u003e\n \u003cp\u003e(0.97-6.35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.029\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e4.97\u003c/p\u003e\n \u003cp\u003e(1.76-11.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.004\u0026nbsp;*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRheumatoid arthritis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e3.16\u003c/p\u003e\n \u003cp\u003e(0.76-8.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.057\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e2.6\u003c/p\u003e\n \u003cp\u003e(0.7-6.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.137\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eUrticaria\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e2.04\u003c/p\u003e\n \u003cp\u003e(0.79-4.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.096\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.9\u003c/p\u003e\n \u003cp\u003e(0.76-3.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.155\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"43.29896907216495%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVasomotor and allergic rhinitis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.587628865979383%\"\u003e\n \u003cp\u003e1.44\u003c/p\u003e\n \u003cp\u003e(0.67-2.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e0.299\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.49484536082474%\"\u003e\n \u003cp\u003e1.33\u003c/p\u003e\n \u003cp\u003e(0.63-2.48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.34020618556701%\"\u003e\n \u003cp\u003e0.424\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003csup\u003ea\u003c/sup\u003e Adjusted for covariates: Age and Sex. * P \u0026lt; 0.05\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eT\u003c/strong\u003e\u003cstrong\u003ehe Performance of\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eDiscrepancy TTs\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Prediction\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe identified demographic characteristics and comorbidities that showed a significant difference (P \u0026lt; 0.1) between the two groups and used them as variables to construct a prediction model for discrepant TT results. Among the four prediction algorithms tested, LightGBM significantly outperformed the others with an AUC of 0.705 (95% CI=0.697-0.713, P\u0026lt;0.001). The other three prediction algorithms, namely Lasso regression (AUC 0.657, 95% CI=0.648-0.666), stepwise logistic regression (AUC 0.669, 95% CI=0.66-0.678), and gradient boosting algorithm (AUC 0.676, 95% CI=0.666-0.687), had lower AUC values. The most important variables in the LightGBM model were HIV and AIDS, pregnancy, sex, and age (\u003cstrong\u003eFigure 2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePPV score of discrepancy TTs risk\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe prevalence of discrepancy in TTs among the research subjects was 0.02. Using the PPV score formula, we divided all the cases into 20 risk categories to indicate the risk of getting discrepancy TTs (\u003cstrong\u003eFigure 3\u003c/strong\u003e). Most of the cases with discrepant TTs were classified into the first risk vigintile. In the first risk vigintile (with the top 5% risk probability), the risk of getting discrepancy TTs was ten times higher than the baseline frequency for the entire research subjects. Using the PPV score concept, we developed an application to estimate the risk of discrepancy in TTs (\u003cstrong\u003eFigure S2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnti-syphilis treatments for the cases with\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ediscrepancy TTs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs the NTT titer is known to decrease in response to syphilis treatments [35], we investigated the association between syphilis treatments on TTs discrepancy. Of the 133 patients with TTs discrepancy, 71 (53.4%) received antibiotic treatment for syphilis, while 62 (46.6%) did not receive treatment. Among those who received treatment, 57 patients treated before TTs discrepancy was observed. The results indicate that patients whose TPPA turned negative were at a higher risk than those with FTA-Abs to be associated with anti-syphilis treatment (OR=9.2, 95% CI= [2.0-41.1], \u003cstrong\u003eTable 2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e. Association between TT types and syphilis treatment among the patients with discrepant TTs. TPPA turns negative: negative TPPA being recorded after a positive TT; FTA-Abs turns negative: negative FTA-Abs being recorded after a positive TT.\u003c/p\u003e\n\u003ctable border=\"1\" ellspacing=\"0\" cellpadding=\"0\" width=\"602\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTPPA turns negative\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFTA-Abs turns negative\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo treatment before negative\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHad treatment before negative \u003csup\u003ea\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR 9.2 (2.0-41.1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003csup\u003ea\u003c/sup\u003e One patient had both TPPA and FTA-Abs negative results after a positive TT.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eTheoretically, only one type of TT would be used in either traditional or reverse algorithm in syphilis diagnosis. However, in a real-world setting, multiple TTs with different analytical methods can be tested for a patient. First, different type of TTs can be tested along the entire disease course. Second, different types of TTs can be tested across different healthcare institutes. Finally, different types of TTs can be ordered at once when physicians are less familiar with the TTs. Briefly, multiple TTs are not uncommon for a patient. Due to the recent rapid increase in new syphilis cases [7, 36, 37] and the scarcity of research on discrepant results of syphilis confirmatory tests, it is important to identify the factors that influence TTs results. In this study, we used demographic information and comorbidities to determine the features associated with the discrepant results. Based on the multivariable analysis and discrepancy prediction model, the most relevant features are sex, age, HIV and AIDS, and pregnancy. We found that negative TPPAs after a positive TT were more strongly associated with syphilis treatment than FTA-Abs. This large-scale study investigated the risk comorbidities associated with TTs discrepancy and identified the TT type that could be affected by syphilis treatment. These results are valuable for the accurate diagnosis of syphilis when TTs are crucial tests for syphilis diagnosis.\u003c/p\u003e\n\u003cp\u003eTTs discrepancy may be more common than previously thought, particularly in settings where the cost of testing is low (\u0026lt;10 USD, totally covered by national health insurance in Taiwan), and TTs are widely used in syphilis management. Although the test requests for TTs in our study did not always follow clinical guidance, our cohort was valuable for identifying cases with multiple TTs and TTs discrepancy. In total 2.3% of 5,780 cases with multiple TTs showed discrepancy. This finding raises concerns for syphilis diagnosis, as relying on a single TT may increase the risk of false negative results for true syphilis cases and false positive results for cases without syphilis. Such errors can have serious consequences, including missed diagnoses of syphilis in pregnant women or delayed diagnoses in patients with HIV/AIDS. Moreover, our study identified various patterns of TTs discrepancy, highlighting the complexity of interpreting these results. To improve syphilis management, it may be necessary to re-evaluate the frequency and criteria for testing TTs for screening or diagnosis.\u003c/p\u003e\n\u003cp\u003eA previous study revealed that HIV and AIDS, and pregnancy, are known causes of false positive for treponemal tests [13], which is consistent with our findings (\u003cstrong\u003eTable 1\u003c/strong\u003e and \u003cstrong\u003eFigure 2\u003c/strong\u003e). In addition, we identified \u0026ldquo;unspecified osteoarthritis\u0026rdquo; as a risk factor associated with TTs discrepancy. The underlying mechanism between the unspecified osteoarthritis and TTs discrepancy has not been well explored. However, it is possible that the autoantibodies involved in unspecified osteoarthritis [38] could interfere with the testing process of TTs and cause discrepancies [39].\u003c/p\u003e\n\u003cp\u003eThe large-scale nature of our study allowed us to observe TTs discrepancy in a more comprehensive way. The results could provide new insights for developing novel diagnostic tools for syphilis or investigating the underlying mechanisms. We combined claims data (NHIRD) with EMRs (CGRD) to examine the most important laboratory test results that are missing in claims data. This examination is the first investigation to develop machine learning models using multiple comorbidities and demographic factors to determine factors related to discrepant syphilis test results. The study results wcaill contribute to the epidemiological knowledge of TTs discrepancy and establish a baseline to evaluate the impact of syphilis. We also developed an online risk calculator based on the TTs discrepancy predictive model (\u003cstrong\u003eFigure S2\u003c/strong\u003e). The PPV score instead of risk probability was used to indicate the risk of having TTs discrepancy (\u003cstrong\u003eFigure 3\u003c/strong\u003e). The PPV score is a more easily understandable way to convey the concept of risk than risk probability. The PPV score describes the risk compared to the overall population and is important for both physicians and patients. A tool can only be useful in clinical practice if the information can be easily understood and communicated [40].\u003c/p\u003e\n\u003cp\u003eWhile TTs are considered as the \u0026ldquo;confirmatory test\u0026rdquo; for syphilis diagnosis, they are also considered \u0026ldquo;syphilis scar\u0026rdquo; because positive results typically last for lifetime even after anti-syphilis treatment. However, the TTs discrepancy we found implies that TTs are not as \u0026ldquo;confirmatory\u0026rdquo; as we believed for a long time [41]. We also investigated the impact of anti-syphilis treatment on TTs discrepancy. The results disclosed that TPPA is more likely than FTA-Abs to turn negative after treatment (\u003cstrong\u003eTable 2\u003c/strong\u003e). The reasons why TPPA is more vulnerable than FTA-Abs would be complicated and have not yet been fully investigated. Previous study also found TPPA seroreversion after treatment [42]. One possible mechanism is related to the antigenicity of reagents. While only part of the antigens is coated onto the particles in TPPA test, the entire \u003cem\u003eT. pallidum\u003c/em\u003e bacteria is used to capture antibodies in FTA-Abs [10]. Thus, antigenicity of FTA-Abs test would be more comprehensive and intact than that of TPPA because the whole surface antigens on \u003cem\u003eT. pallidum\u003c/em\u003e are theoretically preserved in FTA-Abs. This comprehensive antigenicity could be helpful in detecting various anti-syphilis antibodies in serum, even when some antibodies disappear after treatment. However, the explanation is hypothesized on the basic concept of testing and antigenicity. There are many different commercial TTs, and different epitopes are used for testing [10]. Therefore, caution should be taken when extrapolating our observations to other TTs from different providers.\u003c/p\u003e\n\u003cp\u003eThis study has limitations. First, out-of-pocket services were not included in the diagnosis data obtained from NHIRD [43]. Second, we identified multiple comorbidities using diagnosis codes. However, in cases where individual blood test results were unavailable, doctors may have provided a tentative diagnosis [44]. Therefore, we defined the presence of comorbidities as having at least three medical encounters related to the condition. Third, disease stage is not available from the medical records. The impact of disease stage on TTs discrepancy cannot be analyzed. Including disease stages in medical records can help researchers gain a better understanding of the impact and relationship between these two factors. Last, the results may not be generalizable to other populations as we only included observations from the Linkou branch of CGMH. However, CGMH at Linkou, a major medical center located in Northern Taiwan with about 3,800 beds and serving nearly four million outpatients annually, represents one of the largest cohorts in Taiwan [45]. Despite these limitations, further research is needed to better understand the association between multiple comorbidities and discrepant TT results.\u003c/p\u003e\n\u003cp\u003eThis study portrays the association between demographic and multiple comorbidities data and discrepant TTs results. We linked EHRs with claims data to identify risk factors that may affect such test results. Our findings indicated that age, HIV and AIDS, pregnancy, and osteoarthritis were the influential factors associated with discrepant TTs results. These findings could serve as a reference for improving diagnostic decisions in syphilis preventive healthcare and guiding interventions in general. Syphilis remains a vital public health issue, and further research is necessary to improve healthcare and clinical conditions related to this disease.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Chang Gung Medical Foundation Institutional Review Board approved this study (IRB no. 201801524B0) and granted waivers for patient consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data used in the analyses is restricted due to the regulation of the law on the protection of patients\u0026rsquo; data in accordance with local and institutional legal requirements; The process of applying for access to the data will be made available upon reasonable request to the corresponding author.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the National Science and Technology Council, Taiwan (NSTC 111-2320-B-182A-002-MY2 and 111-2628-E-A49-026-MY3) and Chang Gung Memorial Hospital (CMRPG3M0851, CMRPG3L1011). The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYJT and RFH had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. YJT, RFH and HYW analyzed/interpreted the data, performed experiments, designed the study, and draft the manuscript. TWL, WYL, YCW, and JJL reviewed/edited the manuscript for important intellectual content and provided administrative, technical, or material support. YJT and HYW obtained funding and supervised the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the Health and Welfare Data Science Center, Ministry of Health and Welfare Chang Gung sub-centers and the data Health and Welfare Data Science Center, Ministry of Health and Welfare (HWDC, MOHW) for administrative support, promoting Chang Gung medical system to national health database application to conduct value-added research. The authors also acknowledge the technical support of the Department of Management Information System, Chang Gung Memorial Hospital. This study is based in part on data from the Chang Gung Research Database provided by the Chang Gung Memorial Hospital and National Health Insurance Research Database provided by the MOHW. The interpretation and conclusions contained herein do not represent the presentation of both institutions.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003e\u0026Ccedil;akmak SK, Tamer E, Karadağ AS, Waugh M (2019) Syphilis: A great imitator. Clin Dermatol 37:182\u0026ndash;191. https://doi.org/https://doi.org/10.1016/j.clindermatol.2019.01.007\u003c/li\u003e\n\u003cli\u003eSpiteri G, Unemo M, M\u0026aring;rdh O, Amato-Gauci AJ (2019) The resurgence of syphilis in high-income countries in the 2000s: A focus on Europe. Epidemiol Infect 147:. https://doi.org/10.1017/S0950268819000281\u003c/li\u003e\n\u003cli\u003eGhanem KG, Ram S, Rice PA (2020) The Modern Epidemic of Syphilis. New England Journal of Medicine 382:845\u0026ndash;854. https://doi.org/10.1056/NEJMra1901593\u003c/li\u003e\n\u003cli\u003ede Souza RL, dos Santos Madeira LDP, Pereira MVS, et al (2020) Prevalence of syphilis in female sex workers in three countryside cities of the state of Par\u0026aacute;, Brazilian Amazon. BMC Infect Dis 20:129. https://doi.org/10.1186/s12879-020-4850-1\u003c/li\u003e\n\u003cli\u003eHarmon ED, Robertson EW (2019) Syphilis: A growing concern. Nurse Practitioner 44:21\u0026ndash;28. https://doi.org/10.1097/01.NPR.0000558159.61349.cb\u003c/li\u003e\n\u003cli\u003eWorld Health Organization (2016) Global Health Sector Strategy on Sexually Transmitted Infections 2016\u0026ndash;2021 Towards Ending STIs. https://apps.who.int/iris/bitstream/handle/10665/246296/?sequence=1. Accessed 5 Aug 2023\u003c/li\u003e\n\u003cli\u003eGhanem KG, Ram S, Rice PA (2020) The Modern Epidemic of Syphilis. New England Journal of Medicine 382:845\u0026ndash;854. https://doi.org/10.1056/NEJMra1901593\u003c/li\u003e\n\u003cli\u003eKitayama K, Segura ER, Lake JE, et al (2017) Syphilis in the Americas: A protocol for a systematic review of syphilis prevalence and incidence in four high-risk groups, 1980-2016. Syst Rev 6:1\u0026ndash;7. https://doi.org/10.1186/s13643-017-0595-3\u003c/li\u003e\n\u003cli\u003eLai S-K, Chen C-M, Kuo H-W, Liu D-P (2017) Epidemiology of Syphilis and Gonorrhea in Taiwan_ 2005\u0026ndash;2016. TAIWAN EPIDEMIOLOGY BULLETIN 33:460\u0026ndash;468. https://doi.org/10.6525/TEB.20171219.33(24).001\u003c/li\u003e\n\u003cli\u003eMorshed MG, Singhb AE (2015) Recent trends in the serologic diagnosis of syphilis. Clinical and Vaccine Immunology 22:137\u0026ndash;143. https://doi.org/10.1128/CVI.00681-14\u003c/li\u003e\n\u003cli\u003eBristow CC, Klausner JD, Tran A (2020) Clinical Test Performance of a Rapid Point-of-Care Syphilis Treponemal Antibody Test: A Systematic Review and Meta-analysis. Clinical Infectious Diseases 71:S52\u0026ndash;S57. https://doi.org/10.1093/cid/ciaa350\u003c/li\u003e\n\u003cli\u003eFelman YM, Nikitas JA (1980) Syphilis Serology Today. Arch Dermatol 116:84\u0026ndash;89. https://doi.org/10.1001/archderm.1980.01640250086025\u003c/li\u003e\n\u003cli\u003eForrestel AK, Kovarik CL, Katz KA (2020) Sexually acquired syphilis: Laboratory diagnosis, management, and prevention. J Am Acad Dermatol 82:17\u0026ndash;28. https://doi.org/10.1016/j.jaad.2019.02.074\u003c/li\u003e\n\u003cli\u003eHook EW (2017) Syphilis. The Lancet 389:1550\u0026ndash;1557. https://doi.org/10.1016/s0140-6736(16)32411-4\u003c/li\u003e\n\u003cli\u003eShao SC, Chan YY, Kao Yang YH, et al (2019) The Chang Gung Research Database\u0026mdash;A multi-institutional electronic medical records database for real-world epidemiological studies in Taiwan. Pharmacoepidemiol Drug Saf 28:593\u0026ndash;600. https://doi.org/10.1002/pds.4713\u003c/li\u003e\n\u003cli\u003eLin LY, Warren-Gash C, Smeeth L, Chen PC (2018) Data resource profile: the National Health Insurance Research Database (NHIRD). Epidemiol Health 40:e2018062. https://doi.org/10.4178/epih.e2018062\u003c/li\u003e\n\u003cli\u003eHsieh C-Y, Su C-C, Shao S-C, et al (2019) Taiwan\u0026rsquo;s National Health Insurance Research Database: past and future. Clin Epidemiol Volume 11:349\u0026ndash;358. https://doi.org/10.2147/CLEP.S196293\u003c/li\u003e\n\u003cli\u003eLin LY, Warren-Gash C, Smeeth L, Chen PC (2018) Data resource profile: the National Health Insurance Research Database (NHIRD). Epidemiol Health 40:e2018062. https://doi.org/10.4178/epih.e2018062\u003c/li\u003e\n\u003cli\u003eHsieh C-Y, Su C-C, Shao S-C, et al (2019) Taiwan\u0026rsquo;s National Health Insurance Research Database: past and future. Clin Epidemiol Volume 11:349\u0026ndash;358. https://doi.org/10.2147/CLEP.S196293\u003c/li\u003e\n\u003cli\u003eWalraven C Van, Quan H, Forster AJ (2009) A Modification of the Elixhauser Comorbidity Measures Into a Point System for Hospital Death Using Administrative Data. Med Care 47:626\u0026ndash;633\u003c/li\u003e\n\u003cli\u003eMenendez ME, Neuhaus V, Van Dijk CN, Ring D (2014) The Elixhauser comorbidity method outperforms the Charlson index in predicting inpatient death after orthopaedic surgery. Clin Orthop Relat Res 472:2878\u0026ndash;2886. https://doi.org/10.1007/s11999-014-3686-7\u003c/li\u003e\n\u003cli\u003eElixhauser A, Steiner C, Harris D, Coffey R (1998) Comorbidity Measures for Use with Administrative Data Methods Defining Important Comorbidities. Med Care 36:8\u0026ndash;27\u003c/li\u003e\n\u003cli\u003eQuan H, Sundararajan V, Halfon P, et al (2005) Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Med Care 43:1130\u0026ndash;1139. https://doi.org/10.1097/01.mlr.0000182534.19832.83\u003c/li\u003e\n\u003cli\u003evan den Akker M, Buntinx F, Knottnerus JA (1996) Comorbidity or multimorbidity. European Journal of General Practice 2:65\u0026ndash;70. https://doi.org/10.3109/13814789609162146\u003c/li\u003e\n\u003cli\u003eVan den Akker M, Buntix F, Metsemakers JFM, et al (1998) Multimorbidity in general practice: Prevalence, incidence, and determinants of co-occurring chronic and recurrent diseases. J Clin Epidemiol 51:367\u0026ndash;375. https://doi.org/10.1016/S0895-4356(97)00306-5\u003c/li\u003e\n\u003cli\u003eNavickas R, Petric V-K, Feigl AB, Seychell M (2016) Multimorbidity: What do we know? What should we do? J Comorb 6:4\u0026ndash;11. https://doi.org/10.15256/joc.2016.6.72\u003c/li\u003e\n\u003cli\u003eTseng Y-J, Chiu H-J, Chen CJ (2021) \u003cem\u003edxpr\u003c/em\u003e : an R package for generating analysis-ready data from electronic health records\u0026mdash;diagnoses and procedures. PeerJ Comput Sci 7:e520. https://doi.org/10.7717/peerj-cs.520\u003c/li\u003e\n\u003cli\u003eBatten AJ, Thorpe J, Piegari RI, Rosland AM (2020) A Resampling Based Grid Search Method to Improve Reliability and Robustness of Mixture-Item Response Theory Models of Multimorbid High-Risk Patients. IEEE J Biomed Health Inform 24:1780\u0026ndash;1787. https://doi.org/10.1109/JBHI.2019.2948734\u003c/li\u003e\n\u003cli\u003eLiu Y, Chen P-HC, Krause J, Peng L (2019) How to Read Articles That Use Machine Learning: Users\u0026rsquo; Guides to the Medical Literature. JAMA 322:1806\u0026ndash;1816. https://doi.org/10.1001/jama.2019.16489\u003c/li\u003e\n\u003cli\u003eUddin S, Khan A, Hossain ME, Moni MA (2019) Comparing different supervised machine learning algorithms for disease prediction. BMC Med Inform Decis Mak 19:1\u0026ndash;16. https://doi.org/10.1186/s12911-019-1004-8\u003c/li\u003e\n\u003cli\u003eWang H, Hung C, Chen C, et al (2019) Increase Trichomonas vaginalis detection based on urine routine analysis through a machine learning approach. Sci Rep 9:11074. https://doi.org/10.1038/s41598-019-47361-8\u003c/li\u003e\n\u003cli\u003eWorld Health Organization WHO guidelines for the treatment of treponema pallidum (Syphilis).\u003c/li\u003e\n\u003cli\u003eHeinze G, Ploner M, Beyea J (2013) Confidence intervals after multiple imputation: combining profile likelihood information from logistic regressions. Stat Med 32:5062\u0026ndash;76. https://doi.org/10.1002/sim.5899\u003c/li\u003e\n\u003cli\u003eVandenbroucke JP (2007) Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. Ann Intern Med 147:W. https://doi.org/10.7326/0003-4819-147-8-200710160-00010-w1\u003c/li\u003e\n\u003cli\u003ePandey K, Fairley CK, Chen MY, et al (2022) Changes in the Syphilis Rapid Plasma Reagin Titer Between Diagnosis and Treatment. Clinical Infectious Diseases. https://doi.org/10.1093/cid/ciac843\u003c/li\u003e\n\u003cli\u003eKerani RP, Handsfield HH, Stenger MS, et al (2007) Rising rates of syphilis in the era of syphilis elimination. Sex Transm Dis 34:154\u0026ndash;161. https://doi.org/10.1097/01.olq.0000233709.93891.e5\u003c/li\u003e\n\u003cli\u003eHuang S-Y, Hung J-H, Hu L-Y, et al (2018) Risk of sexually transmitted infections following depressive disorder. Medicine 97:e12539. https://doi.org/10.1097/md.0000000000012539\u003c/li\u003e\n\u003cli\u003eCamacho-Encina M, Balboa-Barreiro V, Rego-Perez I, et al (2019) Discovery of an autoantibody signature for the early diagnosis of knee osteoarthritis: Data from the Osteoarthritis Initiative. Ann Rheum Dis 78:1699\u0026ndash;1705. https://doi.org/10.1136/annrheumdis-2019-215325\u003c/li\u003e\n\u003cli\u003ePark IU, Fakile YF, Chow JM, et al (2019) Performance of Treponemal Tests for the Diagnosis of Syphilis. In: Clinical Infectious Diseases. Oxford University Press, pp 913\u0026ndash;918\u003c/li\u003e\n\u003cli\u003eVan de Velde S, Kunnamo I, Roshanov P, et al (2018) The GUIDES checklist: development of a tool to improve the successful use of guideline-based computerised clinical decision support. Implement Sci 13:86. https://doi.org/10.1186/s13012-018-0772-3\u003c/li\u003e\n\u003cli\u003eYoung H (2000) Guidelines for serological testing for syphilis. Sex Transm Infect 76:403\u0026ndash;405. https://doi.org/10.1136/sti.76.5.403\u003c/li\u003e\n\u003cli\u003eBosshard PP, Graf N, Knaute DF, et al (2013) Response of treponema pallidum particle agglutination test titers to treatment of syphilis. Clinical Infectious Diseases 56:463\u0026ndash;464\u003c/li\u003e\n\u003cli\u003eChi C, Lee J, Tsai S, Chen W (2008) Out‐of‐pocket payment for medical care under Taiwan\u0026rsquo;s National Health Insurance system. Health Econ 17:961\u0026ndash;975. https://doi.org/10.1002/hec.1312\u003c/li\u003e\n\u003cli\u003eLin CC, Lai MS, Syu CY, et al (2005) Accuracy of diabetes diagnosis in health insurance claims data in Taiwan. Journal of the Formosan Medical Association 104:157\u0026ndash;163. https://doi.org/10.29828/JFMA.200503.0002\u003c/li\u003e\n\u003cli\u003eLin J-Y, Kang EY-C, Yeh P-H, et al (2020) Proposed measures to be taken by ophthalmologists during the coronavirus disease 2019 pandemic: Experience from Chang Gung Memorial Hospital, Linkou, Taiwan. Taiwan J Ophthalmol 10:80\u0026ndash;86. https://doi.org/10.4103/tjo.tjo_21_20\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Syphilis, serologic tests, treponemal tests, discrepant results ","lastPublishedDoi":"10.21203/rs.3.rs-3847949/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3847949/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSyphilis is becoming more prevalent and is diagnosed primarily through serologic tests. Confirmatory treponemal tests (TTs), such as \u003cem\u003eTreponema pallidum\u003c/em\u003e particle agglutination (TPPA) and fluorescent treponema antibody absorption (FTA-Abs), are typically used. Although only one TT is needed for syphilis diagnosis, multiple TTs are still common in the disease course. Discrepant TT results can cause confusion and delay treatment. We aimed to explore the clinical characteristics of patients with discrepant TT results and develop a machine learning based tool to evaluate the risk of TT discrepancies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the retrospective cohort study, we linked electronic health records from a medical centerand national claims records in Taiwan between January 2001 and September 2018. Medical histories and demographic characteristics were considered as variables of interest to identify risk factors and develop machine learning models to assess the risk of TT discrepancy. We also analyzed the association between syphilis treatments and discrepant TT results.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong 5,780 eligible patients tested for syphilis, 133 (2.30%) had discrepant TT results. Risk factors associated with discrepant TT results were HIV and AIDS (adjusted odds ratio [aOR] = 2.6, 95% CI 1.4-4.7), osteoarthritis (aOR = 3.3, 95% CI 1.3-7.1), and pregnancy (aOR = 5.0, 95% CI 1.8-11.6). Patients with a top 5% risk probability from the LighGBM model were 10 times more likely to have discrepant TT results. For cases with discrepant TT results, TPPA was more likely than FTA-Abs to turn negative after treatment (OR 9.18, 95% CI 2.05-41.07).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIdentifying risk factors and developing a machine learning based decision support tool could significantly aid in interpreting serologic tests for syphilis and guide accurate diagnosis.\u003c/p\u003e","manuscriptTitle":"Discrepancy Results of Treponemal Tests: Exploring the Associated Risk Factors Using Machine Learning Technology Based on 18 Years of Electronic Medical Records and National Claims Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-10 19:15:54","doi":"10.21203/rs.3.rs-3847949/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"883ab889-128c-4108-b5a8-70ae04b81fa6","owner":[],"postedDate":"January 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-01-21T04:59:44+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-10 19:15:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3847949","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3847949","identity":"rs-3847949","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}