INTRODUCTION: Varicella-zoster virus (VZV) testing is increasingly needed for assessing immunity and diagnosis in the varicella vaccination era. VZV-specific immunoglobulin G (IgG) is recommended when assessing immunity; real-time polymerase chain reaction (PCR) is recommended for varicella or herpes zoster diagnosis. The study objective was to describe VZV serologic and virologic testing in U.S. clinical practice. METHODS: Patients with serologic (IgG, IgM) or virologic (PCR, culture) VZV testing were identified in five administrative data sources (∼11-100 million enrollees; 2016-2023). Descriptive analyses were used to examine VZV testing frequency, patient characteristics, and rates by test type. The top 20 diagnostic codes associated with VZV test type were used as a proxy for reason for testing. RESULTS: Across data sources, the highest proportion of VZV testing was for IgG (43%-92%); most was in females (79%-82%) and those aged 20-39 years (62%-70%). Rates of serologic testing were 50-60/10,000 persons. Frequency of VZV virologic testing was considerably lower; PCR testing rates were ∼1/10,000 persons. Diagnostic codes associated with IgG or virologic testing were primarily categorized as routine care or acute illness, respectively. IgM testing was up to 11% of tests, despite not being recommended for screening or diagnostic purposes. CONCLUSIONS: VZV serologic testing rates were 50-60 times higher than PCR. Serologic testing was more common among females and young adults, likely due to screening. Most VZV testing appeared relevant to clinical management; however, inappropriate IgM testing was identified. Appropriate testing is important to guide clinical and public health management for varicella and herpes zoster.

CONTEXT: Little is known about when and how the ICD-10-CM diagnosis code for Post-COVID Conditions (PCC; U09.9) is being used to document PCC. OBJECTIVES: To examine the use and characteristics of clinical coding for PCC. DESIGN: A retrospective cohort. SETTING: Transaction-level medical encounters, laboratory testing results, pharmacy claims, and medical claims for inpatient and outpatient care from the HealthVerity database. PARTICIPANTS: 382 400 US adults and children with private health insurance, Medicare, and Medicaid who had U09.9 code documented during October 1, 2021-June 30, 2023. OUTCOME MEASURES: Count of first use of the U09.9 code, (a) overall, over time, and proportion by provider type; (b) prevalence of PCC-associated incident conditions co-documented with U09.9; (c) number of documented SARS-CoV-2 infections preceding U09.9; (d) timing between infection and U09.9; (e) encounters during the 6 months following first use of U09.9. RESULTS: Overall, 0.6% of 65 556 068 patients had a PCC diagnosis code (64.6% female; 6 in 10 had ≥1 preexisting conditions). The highest count of new U09.9 codes occurred during Quarter 1 and Quarter 3 of 2022 and was documented by a variety of provider specialties. The most prevalent co-documented PCC-associated incident conditions were respiratory (13.4%) and malaise and fatigue (7.8%). Only 62% of patients had SARS-CoV-2 infection documented preceding U09.9; median time to PCC documentation was 17.0 days (interquartile range [IQR] = 5.0, 61.0). Patients with ≥1 encounters during which PCC was documented in the 6 months following their index encounter (n = 109 794) had, on average, 25.5 additional encounters (median = 14 [IQR = 7, 29]). CONCLUSIONS: Our study describes the sociodemographic characteristics, complex clinical manifestations, and high healthcare use of patients following a PCC diagnosis. These findings may inform efforts to identify and treat PCC, inform healthcare planning, and support efforts to educate clinicians about the definition of PCC and accurate application of the code.

BART-Survival is a Python package that allows time-to-event (survival) analyses in discrete-time using the non-parametric machine learning algorithm, Bayesian Additive Regression Trees (BART). BART-Survival combines the performance of the BART algorithm with the complementary data and model formatting required to complete the survival analyses. The library contains a convenient application programming interface (API) that allows a simple approach when using the library for survival analyses, while maintaining capabilities for added complexity when desired. The package is intended for analysts exploring use of flexible non-parametric alternatives to traditional (semi-)parametric survival analyses.

INTRODUCTION: Testing for immunity to measles, mumps, and rubella should include only IgG; IgM testing is appropriate only if acute illness is suspected. The appropriateness of measles, mumps, and rubella IgM testing was evaluated in a national administrative dataset. METHODS: Laboratory testing for measles, mumps, and rubella during 2019-2022 was analyzed in 2024 using HealthVerity administrative claims and laboratory data. IgG, IgM, and reverse-transcriptase polymerase chain reaction (RT-PCR) testing are described by year, demographics, and region. IgM testing was examined for appropriateness, defined as an IgM test combined with diagnostic codes indicative of acute illness. RESULTS: During 2019-2022, IgM testing represented a small proportion of serologic testing (measles: 3.3%, mumps: 2.4%, rubella: 2.1%) but appeared to be appropriately performed in only 15.4% of cases for measles, 32.8% of cases for mumps, and 10.2% of cases for rubella. IgM testing was more commonly performed for female patients, with the largest discrepancy seen for rubella (90.5% female vs 9.5% male). IgM for measles and mumps was more often performed appropriately for persons aged 0-19 years (37.6% and 60.1%) compared with persons aged 20-49 years (11.8% and 22.0%) and 50+ years (16.5% and 33.8%). CONCLUSIONS: The majority of IgM testing for measles, mumps, and rubella during this period appeared inappropriate. Clinicians and health systems could ensure that IgG testing alone is performed when evaluating for immunity through modifications to electronic medical records and commercial laboratories could ensure that providers are able to test for IgG alone when evaluating immunity.

IMPORTANCE: Elimination of tuberculosis (TB) disease in the US hinges on the ability of tests to detect individual risk of developing disease to inform prevention. The relative performance of 3 available TB tests-the tuberculin skin test (TST) and 2 interferon-γ release assays (IGRAs; QuantiFERON-TB Gold In-Tube [QFT-GIT] and SPOT.TB [TSPOT])-in predicting TB disease development in the US remains unknown. OBJECTIVE: To compare the performance of the TST with the QFT-GIT and TSPOT IGRAs in predicting TB disease in high-risk populations. DESIGN, SETTING, AND PARTICIPANTS: This prospective diagnostic study included participants at high risk of TB infection (TBI) or progression to TB disease at 10 US sites between 2012 and 2020. Participants of any age who had close contact with a case patient with infectious TB, were born in a country with medium or high TB incidence, had traveled recently to a high-incidence country, were living with HIV infection, or were from a population with a high local prevalence were enrolled from July 12, 2012, through May 5, 2017. Participants were assessed for 2 years after enrollment and through registry matches until the study end date (November 15, 2020). Data analysis was performed in June 2023. EXPOSURES: At enrollment, participants were concurrently tested with 2 IGRAs (QFT-GIT from Qiagen and TSPOT from Oxford Immunotec) and the TST. Participants were classified as case patients with incident TB disease when diagnosed more than 30 days from enrollment. MAIN OUTCOMES AND MEASURES: Estimated positive predictive value (PPV) ratios from generalized estimating equation models were used to compare test performance in predicting incident TB. Incremental changes in PPV were estimated to determine whether predictive performance significantly improved with the addition of a second test. Case patients with prevalent TB were examined in sensitivity analysis. RESULTS: A total of 22 020 eligible participants were included in this study. Their median age was 32 (range, 0-102) years, more than half (51.2%) were male, and the median follow-up was 6.4 (range, 0.2-8.3) years. Most participants (82.0%) were born outside the US, and 9.6% were close contacts. Tuberculosis disease was identified in 129 case patients (0.6%): 42 (0.2%) had incident TB and 87 (0.4%) had prevalent TB. The TSPOT and QFT-GIT assays performed significantly better than the TST (PPV ratio, 1.65 [95% CI, 1.35-2.02] and 1.47 [95% CI, 1.22-1.77], respectively). The incremental gain in PPV, given a positive TST result, was statistically significant for positive QFT-GIT and TSPOT results (1.64 [95% CI, 1.40-1.93] and 1.94 [95% CI, 1.65-2.27], respectively). CONCLUSIONS AND RELEVANCE: In this diagnostic study assessing predictive value, IGRAs demonstrated superior performance for predicting incident TB compared with the TST. Interferon-γ release assays provided a statistically significant incremental improvement in PPV when a positive TST result was known. These findings suggest that IGRA performance may enhance decisions to treat TBI and prevent TB.

Introduction: Prevention of tuberculosis disease through diagnosis and treatment of latent tuberculosis infection is critical for achieving tuberculosis elimination in the U.S. Diagnosis and treatment of latent tuberculosis infection in safety-net primary care settings that serve patients at risk for tuberculosis may increase uptake of this prevention effort and accelerate progress toward elimination. Optimizing tuberculosis prevention in these settings requires measuring the latent tuberculosis infection care cascade (testing, diagnosis, and treatment) and identifying gaps to develop solutions to overcome barriers. We used electronic health record data to describe the latent tuberculosis infection care cascade and identify gaps among a network of safety-net primary care clinics. Methods: Electronic health record data for patients seen in the OCHIN Clinical Network, the largest network of safety-net clinics in the U.S., between 2012 and 2019 were extracted. electronic health record data were used to measure the latent tuberculosis infection care cascade: patients who met tuberculosis screening criteria on the basis of current recommendations were tested for tuberculosis infection, diagnosed with latent tuberculosis infection, and prescribed treatment for latent tuberculosis infection. Outcomes were stratified by diagnostic test and treatment regimen. Results: Among 1.9 million patients in the analytic cohort, 43.5% met tuberculosis screening criteria, but only 21.4% were tested for latent tuberculosis infection; less than half (40.4%) were tested using an interferon-gamma release assay. Among those with a valid result, 10.5% were diagnosed with latent tuberculosis infection, 29.1% of those were prescribed latent tuberculosis infection treatment, and only 33.6% were prescribed a recommended rifamycin-based regimen. Conclusions: Electronic health record data can be used to measure the latent tuberculosis infection care cascade. A large proportion of patients in this safety-net clinical network are at high risk for tuberculosis infection. Addressing identified gaps in latent tuberculosis infection testing and treatment may have a direct impact on improving tuberculosis prevention in primary care clinics and accelerate progress toward elimination. © 2023