The COVID-19 pandemic magnified long-standing health disparities, showing that certain populations are at higher risk for effects of public health emergencies than others. The pandemic response also put demands on the nation's health departments and stretched their limited resources. In 2021, the Centers for Disease Control and Prevention launched the National Initiative to Address COVID-19 Health Disparities Among Populations at High-Risk and Underserved, Including Racial and Ethnic Minority Populations and Rural Communities (hereinafter, COVID-19 Health Disparities Grant) to reduce COVID-19 health disparities and advance health equity. Health departments in all 50 states, 50 localities, 5 territories, and 3 freely associated states were recipients of approximately $2.25 billion. This study explored the extent to which investments from the COVID-19 Health Disparities Grant, through the allocation of funds across 5 strategies, correspond to reported changes in recipient health departments' capacity to address the COVID-19 public health emergency and future emergencies as measured in the Health Department and Jurisdiction Capacity Survey in 2023. The survey measured capacity along 4 domains: workforce and human resources, interorganizational relationships, data and informational resources, and governance and planning. In total, 70 of 75 recipients who responded to the survey reported that they began with low capacity in at least 1 capacity domain and advanced their capacity during grant implementation. This study demonstrated the reported value of investments in health departments to build capacity and infrastructure to address health disparities and advance health equity to respond to future public health emergencies.

Characterization of serological responses to Plasmodium falciparum (Pf) is of interest to understand disease burden and transmission dynamics; however, their interpretation is challenging. Dried blood spots from 30,815 participants aged 6 months to 15 years from the 2018 Nigeria HIV/AIDS Indicator and Impact Survey were analyzed by multiplex bead-based assay to measure immunoglobulin G (IgG) to Pf-stage-specific MSP-1, AMA-1, GLURPR0, LSA-1, and CSP. These IgG levels were analyzed by principal component analysis (PCA). PC1 and PC2 scores explained 41% and 17% of the total variance, respectively. PC1 unit vectors represented seropositivity. PC2 unit vectors for blood-stage antigens were in opposite directions to liver-stage and sporozoite antigens. PC2 scores were correlated with MSP-1 positively (R = 0.52, P < 0.001) and CSP negatively (R=-0.65, P < 0.001) and may help identify areas with prior exposure but higher risk for increased infections or epidemics. PCA of Pf serology can provide summary scores to possibly inform future programmatic interventions.

Point-of-care antigen tests are an important tool for SARS-CoV-2 detection, but they are less clinically sensitive than real-time reverse-transcription PCR (RT-PCR), impacting their efficacy as screening procedures. Our goal in this study was to see whether we could improve this sensitivity by considering antigen test results in combination with other relevant information, namely exposure status and reported symptoms. In November of 2020, we collected 3,419 paired upper respiratory specimens tested by RT-PCR and the Abbott BinaxNOW antigen test at two community testing sites in Pima County, Arizona. We used symptom, exposure, and antigen testing data to evaluate the sensitivity and specificity of various symptom definitions in predicting RT-PCR positivity. Our analysis yielded 6 novel multi-symptom case definitions with and without antigen test results, the best of which overall achieved a Youden's J index of 0.66, as compared with 0.52 for antigen testing alone. Using a random forest as a guide, we show that this definition, along with our others, does not lose the ability to generalize well to new data despite achieving optimal performance in our sample. Our methodology is broadly applicable, and we have made our code publicly available to aid public health practitioners in developing or fine-tuning their own screening rules. Copyright The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.

Point-of-care antigen tests are an important tool for SARS-CoV-2 detection yet are less clinically sensitive than real-time reverse-transcription PCR (RT-PCR), impacting their efficacy as screening procedures. Our goal in this analysis was to see whether we could improve this sensitivity by considering antigen test results in combination with other relevant information, namely exposure status and reported symptoms. In November of 2020, we collected 3,419 paired upper respiratory specimens tested by RT-PCR and the Abbott BinaxNOW antigen test at two community testing sites in Pima County, Arizona. We used symptom, exposure, and antigen testing data to evaluate the sensitivity and specificity of various symptom definitions in predicting RT-PCR positivity. Our analysis yielded 6 novel multi-symptom case definitions with and without antigen test results, the best of which overall achieved a Youden's J index of 0.66, as compared with 0.53 for antigen testing alone. Using a random forest as a guide, we show that this definition, along with our others, does not lose the ability to generalize well to new data despite achieving optimal performance in our sample. Our methodology is broadly applicable, and our code is publicly available to aid public health practitioners in developing or fine-tuning their own case definitions.

BACKGROUND: Anecdotal reports of menstrual irregularities after receiving COVID-19 vaccines have been observed in post-authorisation and post-licensure monitoring. We aimed to identify and classify reports of menstrual irregularities and vaginal bleeding after COVID-19 vaccination submitted to a voluntary active surveillance system. METHODS: This observational cohort study included recipients of a COVID-19 vaccine who were aged 18 years and older and reported their health experiences to v-safe, a voluntary smartphone-based active surveillance system for monitoring COVID-19 vaccine safety in the USA, from Dec 14, 2020, to Jan 9, 2022. Responses to survey questions on reactions after vaccination were extracted, and a pre-trained natural language inference model was used to identify and classify free-text comments related to menstruation and vaginal bleeding in response to an open-ended prompt about any symptoms at intervals after vaccination. Related responses were further categorised into themes of timing, severity, perimenopausal and postmenopausal bleeding, resumption of menses, and other responses. We examined associations between symptom theme and respondent characteristics, including vaccine type and dose number received, solicited local and systemic reactions reported, and health care sought. FINDINGS: 63 815 respondents reported on menstrual irregularities or vaginal bleeding, which included 62 679 female respondents (1·0% of 5 975 363 female respondents aged ≥18 years). Common themes identified included timing of menstruation (70 981 [83·6%] responses) and severity of menstrual symptoms (56 890 [67·0%] responses). Other themes included menopausal bleeding (3439 [4·0%] responses) and resumption of menses (2378 [2·8%] responses). Respondents submitting reports related to menopausal bleeding were more likely to seek health care than were those submitting reports related to other menstruation and vaginal bleeding themes. INTERPRETATION: Reports of heterogeneous symptoms related to menstruation or vaginal bleeding after COVID-19 vaccination are being submitted to v-safe, although this study is unable to characterise the relationship of these symptoms to COVID-19 vaccination. Methods that leverage pretrained models to interpret and classify unsolicited signs and symptoms in free-text reports offer promise in the initial evaluation of unexpected adverse events potentially associated with use of newly authorised or licensed vaccines. FUNDING: Centers for Disease Control and Prevention.

The early identification of clusters of persons with tuberculosis (TB) that will grow to become outbreaks creates an opportunity for intervention in preventing future TB cases. We used surveillance data (2009-2018) from the United States, statistically derived definitions of unexpected growth, and machine learning techniques to predict which clusters of genotype-matched TB cases are most likely to continue accumulating cases above expected growth within a 1-year follow-up period. We developed a model to predict which clusters are likely to grow on a training and testing dataset that was generalizable to a validation dataset. Our model shows that characteristics of clusters were more important than the social, demographic, and clinical characteristics of the patients in those clusters. For instance, the time between cases before unexpected growth was identified as the most important of our predictors. A faster accumulation of cases increased the probability of excess growth being predicted during the follow-up period. We demonstrated that combining the characteristics of clusters and cases with machine learning can add to existing tools to help prioritize which clusters may benefit most from public health interventions. For example, consideration of an entire cluster, not only an individual patient, may assist in interrupting ongoing transmission.

BACKGROUND: Optimized symptom-based COVID-19 case definitions that guide public health surveillance and individual patient management in the community may assist pandemic control. METHODS: We assessed diagnostic performance of existing cases definitions (e.g. influenza-like illness, COVID-like illness) using symptoms reported from 185 household contacts to a PCR-confirmed case of COVID-19 in Wisconsin and Utah, United States. We stratified analyses between adults and children. We also constructed novel case definitions for comparison. RESULTS: Existing COVID-19 case definitions generally showed high sensitivity (86-96%) but low positive predictive value (PPV) (36-49%; F-1 score 52-63) in this community cohort. Top performing novel symptom combinations included taste or smell dysfunction and improved the balance of sensitivity and PPV (F-1 score 78-80). Performance indicators were generally lower for children (< 18 years of age). CONCLUSIONS: Existing COVID-19 case definitions appropriately screened in household contacts with COVID-19. Novel symptom combinations incorporating taste or smell dysfunction as a primary component improved accuracy. Case definitions tailored for children versus adults should be further explored.

INTRODUCTION: Efficient management of study drug inventory shipments is critical to keep research sites enrolling into multisite clinical treatment trials. A standard manual drug-management process used by the Tuberculosis Trials Consortium (TBTC), did not accommodate import permit approval timelines, shipment transit-times and time-zone differences. We compared a new web-based solution with the manual process, during an international 34-site clinical trial conducted by the TBTC and the AIDS Clinical Trials Group (ACTG); TBTC Study 31/ACTG A5349. MATERIAL AND METHODS: We developed and implemented a technological solution by integrating logistical and regulatory requirements for drug importation with statistical simulations that estimated stock-out times in an online Drug Management Module (DMM). We measured the average shipment-related drug stock-outs and time to drug availability, to assess the efficiency of the DMM compared to the manual approach. RESULTS: An Interrupted Time-Series (ITS) analysis showed a 15% [p-value = 0.03; 95% C.I. (-28.8%, -2.0%)] reduction in average shipment-related study drug stock-out after DMM implementation. The DMM streamlined the restocking process at study sites, reducing median transit-time for sites associated with a depot by 2 days [95% C.I. (-3.0, -1.0)]. Under the DMM, study drugs were available for treatment assignment on the day received, compared to one day after receipt under the manual process. DISCUSSION: The DMM provided TBTC's Data and Coordinating Center and site staff with more efficient procedures to manage and consistently maintain study drug inventory at enrolling sites. This DMM framework can improve efficiency in future multicenter clinical trials. TRIAL REGISTRATION: This trial was registered with ClinicalTrials.gov (Identifier: NCT02410772) on April 8, 2015.

IMPORTANCE: Criterion-standard specimens for tuberculosis diagnosis in young children, gastric aspirate (GA) and induced sputum, are invasive and rarely collected in resource-limited settings. A far less invasive approach to tuberculosis diagnostic testing in children younger than 5 years as sensitive as current reference standards is important to identify. OBJECTIVE: To characterize the sensitivity of preferably minimally invasive specimen and assay combinations relative to maximum observed yield from all specimens and assays combined. DESIGN, SETTING, AND PARTICIPANTS: In this prospective cross-sectional diagnostic study, the reference standard was a panel of up to 2 samples of each of 6 specimen types tested for Mycobacterium tuberculosis complex by Xpert MTB/RIF assay and mycobacteria growth indicator tube culture. Multiple different combinations of specimens and tests were evaluated as index tests. A consecutive series of children was recruited from inpatient and outpatient settings in Kisumu County, Kenya, between October 2013 and August 2015. Participants were children younger than 5 years who had symptoms of tuberculosis (unexplained cough, fever, malnutrition) and parenchymal abnormality on chest radiography or who had cervical lymphadenopathy. Children with 1 or more evaluable specimen for 4 or more primary study specimen types were included in the analysis. Data were analyzed from February 2015 to October 2020. MAIN OUTCOMES AND MEASURES: Cumulative and incremental diagnostic yield of combinations of specimen types and tests relative to the maximum observed yield. RESULTS: Of the 300 enrolled children, the median (interquartile range) age was 2.0 (1.0-3.6) years, and 151 (50.3%) were female. A total of 294 met criteria for analysis. Of 31 participants with confirmed tuberculosis (maximum observed yield), 24 (sensitivity, 77%; interdecile range, 68%-87%) had positive results on up to 2 GA samples and 20 (sensitivity, 64%; interdecile range, 53%-76%) had positive test results on up to 2 induced sputum samples. The yields of 2 nasopharyngeal aspirate (NPA) samples (23 of 31 [sensitivity, 74%; interdecile range, 64%-84%]), of 1 NPA sample and 1 stool sample (22 of 31 [sensitivity, 71%; interdecile range, 60%-81%]), or of 1 NPA sample and 1 urine sample (21.5 of 31 [sensitivity, 69%; interdecile range, 58%-80%]) were similar to reference-standard specimens. Combining up to 2 each of GA and NPA samples had an average yield of 90% (28 of 31). CONCLUSIONS AND RELEVANCE: NPA, in duplicate or in combination with stool or urine specimens, was readily obtainable and had diagnostic yield comparable with reference-standard specimens. This combination could improve tuberculosis diagnosis among children in resource-limited settings. Combining GA and NPA had greater yield than that of the current reference standards and may be useful in certain clinical and research settings.

BACKGROUND: The clinical and epidemiological significance of HIV-associated Mycobacterium tuberculosis bloodstream infection (BSI) is incompletely understood. We hypothesised that M tuberculosis BSI prevalence has been underestimated, that it independently predicts death, and that sputum Xpert MTB/RIF has suboptimal diagnostic yield for M tuberculosis BSI. METHODS: We did a systematic review and individual patient data (IPD) meta-analysis of studies performing routine mycobacterial blood culture in a prospectively defined patient population of people with HIV aged 13 years or older. Studies were identified through searching PubMed and Scopus up to Nov 10, 2018, without language or date restrictions and through manual review of reference lists. Risk of bias in the included studies was assessed with an adapted QUADAS-2 framework. IPD were requested for all identified studies and subject to harmonised inclusion criteria: age 13 years or older, HIV positivity, available CD4 cell count, a valid mycobacterial blood culture result (excluding patients with missing data from lost or contaminated blood cultures), and meeting WHO definitions for suspected tuberculosis (presence of screening symptom). Predicted probabilities of M tuberculosis BSI from mixed-effects modelling were used to estimate prevalence. Estimates of diagnostic yield of sputum testing with Xpert (or culture if Xpert was unavailable) and of urine lipoarabinomannan (LAM) testing for M tuberculosis BSI were obtained by two-level random-effect meta-analysis. Estimates of mortality associated with M tuberculosis BSI were obtained by mixed-effect Cox proportional-hazard modelling and of effect of treatment delay on mortality by propensity-score analysis. This study is registered with PROSPERO, number 42016050022. FINDINGS: We identified 23 datasets for inclusion (20 published and three unpublished at time of search) and obtained IPD from 20, representing 96.2% of eligible IPD. Risk of bias for the included studies was assessed to be generally low except for on the patient selection domain, which was moderate in most studies. 5751 patients met harmonised IPD-level inclusion criteria. Technical factors such as number of blood cultures done, timing of blood cultures relative to blood sampling, and patient factors such as inpatient setting and CD4 cell count, explained significant heterogeneity between primary studies. The predicted probability of M tuberculosis BSI in hospital inpatients with HIV-associated tuberculosis, WHO danger signs, and a CD4 count of 76 cells per muL (the median for the cohort) was 45% (95% CI 38-52). The diagnostic yield of sputum in patients with M tuberculosis BSI was 77% (95% CI 63-87), increasing to 89% (80-94) when combined with urine LAM testing. Presence of M tuberculosis BSI compared with its absence in patients with HIV-associated tuberculosis increased risk of death before 30 days (adjusted hazard ratio 2.48, 95% CI 2.05-3.08) but not after 30 days (1.25, 0.84-2.49). In a propensity-score matched cohort of participants with HIV-associated tuberculosis (n=630), mortality increased in patients with M tuberculosis BSI who had a delay in anti-tuberculosis treatment of longer than 4 days compared with those who had no delay (odds ratio 3.15, 95% CI 1.16-8.84). INTERPRETATION: In critically ill adults with HIV-tuberculosis, M tuberculosis BSI is a frequent manifestation of tuberculosis and predicts mortality within 30 days. Improved diagnostic yield in patients with M tuberculosis BSI could be achieved through combined use of sputum Xpert and urine LAM. Anti-tuberculosis treatment delay might increase the risk of mortality in these patients. FUNDING: This study was supported by Wellcome fellowships 109105Z/15/A and 105165/Z/14/A.

OBJECTIVE: To compare the prevalence of tuberculosis infection and disease in household contacts of patients with bacteriologically confirmed tuberculosis disease and contacts of non-bacteriologically confirmed disease in western Kenya. METHODS: We enrolled newly diagnosed index patients and their household contacts from March 2014 to June 2016. All contacts were evaluated with a symptom questionnaire, tuberculin skin test (TST), and HIV test. Clinical evaluation and sputum testing were performed for those with symptoms, positive TST result, or HIV infection. RESULTS: We enrolled 1155 contacts of 330 index patients with bacteriologically confirmed tuberculosis and 192 contacts of 55 index patients with non-bacteriologically confirmed tuberculosis. 3.5% of contacts of patients with bacteriologically confirmed tuberculosis were diagnosed with tuberculosis, whereas no contacts of index patients with non-bacteriologically confirmed tuberculosis were. Of those diagnosed with tuberculosis disease, 58.5% reported symptoms, 34.1% reported no symptoms but had positive TST results, and 7.3% had neither symptoms nor positive TST but were HIV-positive. Among 872 contacts with a TST result, 50.9% of contacts of index patients with bacteriologically confirmed tuberculosis and 41.0% of contacts of index patients with non-bacteriologically confirmed tuberculosis had a positive result (prevalence ratio = 1.16, 95% confidence interval 0.92-1.48). CONCLUSION: In a high-burden setting, tuberculosis disease was more prevalent among contacts of patients with bacteriologically confirmed tuberculosis than contacts of patients with non-bacteriologically confirmed disease. TST was feasible to perform and helped to detect cases that would have been missed had only symptomatic contacts been evaluated.

INTRODUCTION: Phase 2 clinical trials of tuberculosis treatment have shown that once-daily regimens in which rifampin is replaced by high dose rifapentine have potent antimicrobial activity that may be sufficient to shorten overall treatment duration. Herein we describe the design of an ongoing phase 3 clinical trial testing the hypothesis that once-daily regimens containing high dose rifapentine in combination with other anti-tuberculosis drugs administered for four months can achieve cure rates not worse than the conventional six-month treatment regimen. METHODS/DESIGN: S31/A5349 is a multicenter randomized controlled phase 3 non-inferiority trial that compares two four-month regimens with the standard six-month regimen for treating drug-susceptible pulmonary tuberculosis in HIV-negative and HIV-positive patients. Both of the four-month regimens contain high-dose rifapentine instead of rifampin, with ethambutol replaced by moxifloxacin in one regimen. All drugs are administered seven days per week, and under direct observation at least five days per week. The primary outcome is tuberculosis disease-free survival at twelve months after study treatment assignment. A total of 2500 participants will be randomized; this gives 90% power to show non-inferiority with a 6.6% margin of non-inferiority. DISCUSSION: This phase 3 trial formally tests the hypothesis that augmentation of rifamycin exposures can shorten tuberculosis treatment to four months. Trial design and standardized implementation optimize the likelihood of obtaining valid results. Results of this trial may have important implications for clinical management of tuberculosis at both individual and programmatic levels. TRIAL REGISTRATION: NCT02410772. Registered 8 April 2015,https://www.clinicaltrials.gov/ct2/show/NCT02410772?term=02410772&rank=1.

OBJECTIVE: The Centers for Disease Control and Prevention (CDC) coordinates a labor-intensive process to measure the prevalence of autism spectrum disorder (ASD) among children in the United States. Random forests methods have shown promise in speeding up this process, but they lag behind human classification accuracy by about 5%. We explore whether more recently available document classification algorithms can close this gap. MATERIALS AND METHODS: Using data gathered from a single surveillance site, we applied 8 supervised learning algorithms to predict whether children meet the case definition for ASD based solely on the words in their evaluations. We compared the algorithms' performance across 10 random train-test splits of the data, using classification accuracy, F1 score, and number of positive calls to evaluate their potential use for surveillance. RESULTS: Across the 10 train-test cycles, the random forest and support vector machine with Naive Bayes features (NB-SVM) each achieved slightly more than 87% mean accuracy. The NB-SVM produced significantly more false negatives than false positives (P = 0.027), but the random forest did not, making its prevalence estimates very close to the true prevalence in the data. The best-performing neural network performed similarly to the random forest on both measures. DISCUSSION: The random forest performed as well as more recently available models like the NB-SVM and the neural network, and it also produced good prevalence estimates. NB-SVM may not be a good candidate for use in a fully-automated surveillance workflow due to increased false negatives. More sophisticated algorithms, like hierarchical convolutional neural networks, may not be feasible to train due to characteristics of the data. Current algorithms might perform better if the data are abstracted and processed differently and if they take into account information about the children in addition to their evaluations. CONCLUSION: Deep learning models performed similarly to traditional machine learning methods at predicting the clinician-assigned case status for CDC's autism surveillance system. While deep learning methods had limited benefit in this task, they may have applications in other surveillance systems.

OBJECTIVE: To evaluate the utility of a broad and non-specific symptom screen for identifying people with undiagnosed HIV infection. DESIGN: Secondary analysis of operational data collected during implementation of a cluster-randomized trial for tuberculosis case detection. METHODS: As part of the trial, adults reporting cough, fever, night sweats, weight loss, or difficulty breathing of any duration in the past month were identified in health facilities and community-based mobile screening units in western Kenya. Adults reporting any symptom were offered HIV testing. We analysed the HIV testing data from this study, using modified Poisson regression to identify predictors of new HIV diagnoses among adults with symptoms and initially unknown HIV status. RESULTS: We identified 3,818 symptomatic adults, referred 1424 (37%) for testing, of whom 1065 (75%) accepted, and 107 (10%) were newly diagnosed with HIV. The prevalence of new HIV diagnoses was 21% (95% CI: 17-25%) among those tested in health facilities and 5% (95% CI 4-7%) among those tested in mobile units. More men were diagnosed with HIV than women despite fewer men being screened. People who reported 4-5 symptoms were over twice as likely to be diagnosed with HIV compared to those reporting 1-3 symptoms (adjusted prevalence ratio [aPR] in health facilities = 2.58, 95% CI, 1.65-4.05; aPR in mobile units = 2.63, 95% CI, 1.37-5.03). CONCLUSION: We observed a high yield of new HIV diagnoses among adults identified by active application of a broad symptom screen. Integrated tuberculosis and HIV screening using could help close the detection gap for both conditions.

SETTING: Efficient tuberculosis (TB) active case-finding strategies are important in settings with high TB burdens and limited resources, such as those in western Kenya. OBJECTIVE: To guide efforts to optimize screening efficiency, we identified the predictors of TB among people screened in health facilities and communities. DESIGN: During February 2015-June 2016, adults aged >/=15 years reporting any TB symptom were identified in health facilities and community mobile screening units, and evaluated for TB. We assessed the predictors of TB using a modified Poisson regression with generalized estimating equations to account for clustering according to screening site. RESULTS: TB was diagnosed in 484 (20.3%) of 2394 symptomatic adults in health facilities and 39 (3.4%) of 1424 in communities. In health facilities, >10% of symptomatic adults in all demographic groups had TB, and no predictors were associated with a >/=2-fold increased risk. In communities, the independent predictors of TB were male sex (adjusted prevalence ratio [aPR] = 4.26, 95%CI 2.43-7.45), HIV infection (aPR 2.37, 95%CI 1.18-4.77), and household TB contact in the last 2 years (aPR 2.84, 95%CI 1.62-4.96). CONCLUSION: Our findings support the notion of general TB screening in health facilities and evaluation of the adult household contacts of TB patients.

OBJECTIVE: To evaluate the utility of a broad and non-specific symptom screen for identifying people with undiagnosed HIV infection. DESIGN: Secondary analysis of operational data collected during implementation of a cluster-randomized trial for tuberculosis case detection. METHODS: As part of the trial, adults reporting cough, fever, night sweats, weight loss, or difficulty breathing of any duration in the past month were identified in health facilities and community-based mobile screening units in western Kenya. Adults reporting any symptom were offered HIV testing. We analysed the HIV testing data from this study, using modified Poisson regression to identify predictors of new HIV diagnoses among adults with symptoms and initially unknown HIV status. RESULTS: We identified 3,818 symptomatic adults, referred 1424 (37%) for testing, of whom 1065 (75%) accepted, and 107 (10%) were newly diagnosed with HIV. The prevalence of new HIV diagnoses was 21% (95% CI: 17-25%) among those tested in health facilities and 5% (95% CI 4-7%) among those tested in mobile units. More men were diagnosed with HIV than women despite fewer men being screened. People who reported 4-5 symptoms were over twice as likely to be diagnosed with HIV compared to those reporting 1-3 symptoms (adjusted prevalence ratio [aPR] in health facilities = 2.58, 95% CI, 1.65-4.05; aPR in mobile units = 2.63, 95% CI, 1.37-5.03). CONCLUSION: We observed a high yield of new HIV diagnoses among adults identified by active application of a broad symptom screen. Integrated tuberculosis and HIV screening using could help close the detection gap for both conditions.

Syndromic surveillance detects and monitors individual and population health indicators through sources such as emergency department records. Automated classification of these records can improve outbreak detection speed and diagnosis accuracy. Current syndromic systems rely on hand-coded keyword-based methods to parse written fields and may benefit from the use of modern supervised-learning classifier models. In this paper, we implement two recurrent neural network models based on long short-term memory (LSTM) and gated recurrent unit (GRU) cells and compare them to two traditional bag-of-words classifiers: multinomial naive Bayes (MNB) and a support vector machine (SVM). The MNB classifier is one of only two machine learning algorithms currently being used for syndromic surveillance. All four models are trained to predict diagnostic code groups as defined by Clinical Classification Software, first to predict from discharge diagnosis, and then from chief complaint fields. The classifiers are trained on 3.6 million de-identified emergency department records from a single United States jurisdiction. We compare performance of these models primarily using the F1 score, and we measure absolute model performance to determine which conditions are the most amenable to surveillance based on chief complaint alone. Using discharge diagnoses, the LSTM classifier performs best, though all models exhibit an F1 score above 96.00. Using chief complaints, the GRU performs best (F1=47.38), and MNB with bigrams performs worst (F1=39.40). We also note that certain syndrome types are easier to detect than others. For example, chief complaints using the GRU model predicts alcohol-related disorders well (F1=78.91) but predicts influenza poorly (F1=14.80). In all instances, the RNN models outperformed the bag-of-words classifiers suggesting deep learning models could substantially improve the automatic classification of unstructured text for syndromic surveillance.

Drug overdose is the leading cause of unintentional injury-associated death in the United States. Among 70,237 fatal drug overdoses in 2017, prescription opioids were involved in 17,029 (24.2%) (1). Higher rates of opioid-related deaths have been recorded in nonmetropolitan (rural) areas (2). In 2017, 14 rural counties were among the 15 counties with the highest opioid prescribing rates.* Higher opioid prescribing rates put patients at risk for addiction and overdose (3). Using deidentified data from the Athenahealth electronic health record (EHR) system, opioid prescribing rates among 31,422 primary care providers(dagger) in the United States were analyzed to evaluate trends from January 2014 to March 2017. This analysis assessed how prescribing practices varied among six urban-rural classification categories of counties, before and after the March 2016 release of CDC's Guideline for Prescribing Opioids for Chronic Pain (Guideline) (4). Patients in noncore (the most rural) counties had an 87% higher chance of receiving an opioid prescription compared with persons in large central metropolitan counties during the study period. Across all six county groups, the odds of receiving an opioid prescription decreased significantly after March 2016. This decrease followed a flat trend during the preceding period in micropolitan and large central metropolitan county groups; in contrast, the decrease continued previous downward trends in the other four county groups. Data from EHRs can effectively supplement traditional surveillance methods for monitoring trends in opioid prescribing and other areas of public health importance, with minimal lag time under ideal conditions. As less densely populated areas appear to indicate both substantial progress in decreasing opioid prescribing and ongoing need for reduction, community health care practices and intervention programs must continue to be tailored to community characteristics.

OBJECTIVE: To assess the performance of symptom-based screening for tuberculosis (TB), alone and with chest radiography among people living with HIV (PLHIV), including pregnant women, in Western Kenya. DESIGN: Prospective cohort study. METHODS: PLHIV from 15 randomly-selected HIV clinics were screened with three clinical algorithms [World Health Organization (WHO), Ministry of Health (MOH), and "Improving Diagnosis of TB in HIV-infected persons" (ID-TB/HIV) study], underwent chest radiography (unless pregnant), and provided two or more sputum specimens for smear microscopy, liquid culture, and Xpert MTB/RIF. Performance of clinical screening was compared to laboratory results, controlling for the complex design of the survey. RESULTS: Overall, 738 (85.6%) of 862 PLHIV enrolled were included in the analysis. Estimated TB prevalence was 11.2% (95% CI, 9.9-12.7). Sensitivity of the three screening algorithms was similar [WHO, 74.1% (95% CI, 64.1-82.2); MOH, 77.5% (95% CI, 68.6-84.5); and ID-TB/HIV, 72.5% (95% CI, 60.9-81.7)]. Sensitivity of the WHO algorithm was significantly lower among HIV-infected pregnant women [28.2% (95% CI, 14.9-46.7)] compared to non-pregnant women [78.3% (95% CI, 67.3-86.4)] and men [77.2% (95% CI, 68.3-84.2)]. Chest radiography increased WHO algorithm sensitivity and negative predictive value to 90.9% (95% CI, 86.4-93.9) and 96.1% (95% CI, 94.4-97.3), respectively, among asymptomatic men and non-pregnant women. CONCLUSIONS: Clinical screening missed approximately 25% of laboratory-confirmed TB cases among all PLHIV and more than 70% among HIV-infected pregnant women. National HIV programs should evaluate the feasibility of laboratory-based screening for TB, such as a single Xpert MTB/RIF test for all PLHIV, especially pregnant women, at enrollment in HIV services.

The Xpert(R) MTB/RIF assay is both sensitive and specific as a diagnostic test. Xpert also reports quantitative output in cycle thresholds (Ct), which may provide a dynamic measure of sputum bacillary burden when used longitudinally. We evaluated the relationship between Xpert Ct trajectory and drug exposure during TB treatment to evaluate the potential utility of Xpert Ct for treatment monitoring. We obtained serial sputum samples from patients with smear-positive pulmonary TB consecutively enrolled at ten international clinical trial sites participating in Study 29X, a CDC-sponsored TB Trials Consortium study evaluating the tolerability, safety, and antimicrobial activity of rifapentine at daily doses up to 20 mg/kg. Xpert was performed at weeks 0, 2, 4, 6, 8, and 12. Longitudinal Ct data were modeled using a nonlinear mixed effects model, in relation to rifapentine exposure (area under the concentration-time curve [AUC]). The rate of change of Ct was higher in subjects receiving rifapentine compared to that in subjects receiving standard dose rifampin. Moreover, rifapentine exposure, but not assigned dose, was significantly associated with rate of change in Ct (p = 0.02). The estimated increase in Ct slope for every additional 100 mcg*h/mL of rifapentine drug exposure (as measured by AUC) was 0.11 Ct/week (95% CI 0.05 - 0.17). Increasing rifapentine exposure is associated with faster rate of change of Xpert Ct, indicating faster clearance of MTB DNA. These data suggest that the quantitative outputs of the Xpert MTB/RIF assay may be useful as a dynamic measure of TB treatment response.

RATIONALE: The interferon gamma release assays and tuberculin skin tests are used to support the diagnosis of both latent and active tuberculosis. However, we previously demonstrated that a negative tuberculin test in active tuberculosis is associated with disseminated disease and death. It is unknown whether the same associations exist for interferon gamma release assays. OBJECTIVES: To determine the association between these tests and site of tuberculosis and death among persons with active tuberculosis. METHODS: We analyzed interferon gamma release assay and tuberculin test results for all persons with culture-confirmed tuberculosis reported to the US National Tuberculosis Surveillance System from 2010-2014. We used logistic regression to calculate the association between these tests and site of disease and death. MEASUREMENTS AND MAIN RESULTS: 24,803 persons with culture-confirmed tuberculosis had either of these test results available for analysis. Persons with a positive tuberculin test had lower odds of disseminated disease (i.e., miliary or combined pulmonary and extrapulmonary disease), but there was no difference in the odds of disseminated disease with a positive interferon gamma release assay. However, persons who were positive to either of these tests had lower odds of death. An indeterminate interferon gamma release assay result was associated with greater odds of both disseminated disease and death. CONCLUSIONS: Despite perceived equivalence in clinical practice, interferon gamma release assay and tuberculin test results have different associations with tuberculosis site yet similar associations with the risk of death. Furthermore, an indeterminate interferon gamma release assay result in a person with active tuberculosis is not unimportant, and rather carries greater odds of disseminated disease and death. Prospective study may improve our understanding of the underlying mechanisms by which these tests are associated with disease localization and death.

BACKGROUND: Diagnosis followed by effective treatment of tuberculosis (TB) reduces transmission and saves lives in persons living with HIV (PLHIV). Sputum smear microscopy is widely used for diagnosis, despite limited sensitivity in PLHIV. Evidence is needed to determine the optimal diagnostic approach for these patients. METHODS: From May 2011 through June 2012, we recruited PLHIV from 15 HIV treatment centers in western Kenya. We collected up to three sputum specimens for Ziehl-Neelsen (ZN) and fluorescence microscopy (FM), GeneXpert MTB/RIF (Xpert), and culture, regardless of symptoms. We calculated the incremental yield of each test, stratifying results by CD4 cell count and specimen type; data were analyzed to account for complex sampling. RESULTS: From 778 enrolled patients, we identified 88 (11.3%) laboratory-confirmed TB cases. Of the 74 cases who submitted 2 specimens for microscopy and Xpert testing, ZN microscopy identified 25 (33.6%); Xpert identified those plus an additional 18 (incremental yield = 24.4%). Xpert testing of spot specimens identified 48 (57.0%) of 84 cases; whereas Xpert testing of morning specimens identified 50 (66.0%) of 76 cases. Two Xpert tests detected 22/24 (92.0%) TB cases with CD4 counts <100 cells/muL and 30/45 (67.0%) of cases with CD4 counts ≥100 cells/mul. CONCLUSIONS: In PLHIV, Xpert substantially increased diagnostic yield compared to smear microscopy and had the highest yield when used to test morning specimens and specimens from PLHIV with CD4 count <100 cells/muL. TB programs unable to replace smear microscopy with Xpert for all symptomatic PLHIV should consider targeted replacement and using morning specimens.

Mycobacterium africanum is endemic to West Africa and causes tuberculosis (TB). We reviewed reported cases of TB in the United States during 2004-2013 that had lineage assigned by genotype (spoligotype and mycobacterial interspersed repetitive unit variable number tandem repeats). M. africanum caused 315 (0.4%) of 73,290 TB cases with lineage assigned by genotype. TB caused by M. africanum was associated more with persons from West Africa (adjusted odds ratio [aOR] 253.8, 95% CI 59.9-1,076.1) and US-born black persons (aOR 5.7, 95% CI 1.2-25.9) than with US-born white persons. TB caused by M. africanum did not show differences in clinical characteristics when compared with TB caused by M. tuberculosis. Clustered cases defined as >2 cases in a county with identical 24-locus mycobacterial interspersed repetitive unit genotypes, were less likely for M. africanum (aOR 0.1, 95% CI 0.1-0.4), which suggests that M. africanum is not commonly transmitted in the United States.

BACKGROUND: TB remains a major public health concern, even in low-incidence countries like the USA and the UK. Over the last two decades, cases of TB reported in the USA have declined, while they have increased substantially in the UK. We examined factors associated with this divergence in TB trends between the two countries. METHODS: We analysed all cases of TB reported to the US and UK national TB surveillance systems from 1 January 2000 through 31 December 2011. Negative binominal regression was used to assess potential demographic, clinical and risk factor variables associated with differences in observed trends. FINDINGS: A total of 259 609 cases were reported. From 2000 to 2011, annual TB incidence rates declined from 5.8 to 3.4 cases per 100 000 in the USA, whereas in the UK, TB incidence increased from 11.4 to 14.4 cases per 100 000. The majority of cases in both the USA (56%) and the UK (64%) were among foreign-born persons. The number of foreign-born cases reported in the USA declined by 15% (7731 in 2000 to 6564 in 2011) while native-born cases fell by 54% (8442 in 2000 to 3883 in 2011). In contrast, the number of foreign-born cases reported in the UK increased by 80% (3380 in 2000 to 6088 in 2011), while the number of native-born cases remained largely unchanged (2158 in 2000 to 2137 in 2011). In an adjusted negative binomial regression model, significant differences in trend were associated with sex, age, race/ethnicity, site of disease, HIV status and previous history of TB (p<0.01). Among the foreign-born, significant differences in trend were also associated with time since UK or US entry (p<0.01). INTERPRETATION: To achieve TB elimination in the UK, a re-evaluation of current TB control policies and practices with a focus on foreign-born are needed. In the USA, maintaining and strengthening control practices are necessary to sustain the progress made over the last 20 years.

SETTING: Even among persons who have completed a course of treatment for their first tuberculosis (TB) episode, patients with a history of TB are at higher risk for having TB. OBJECTIVE: To describe factors from the initial TB episode associated with recurrent TB among patients who completed treatment and remained free of TB for at least 12 months. DESIGN: During 1993-2006, US TB cases stratified by birth origin were examined. Cox proportional hazards regression was used to assess the association of factors during the initial episode with recurrence at least 12 months after treatment completion. RESULTS: Among 632 US-born patients, TB recurrence was associated with age 25-44 years (adjusted hazard ratio [aHR] 1.77, 99% confidence interval [CI] 1.02-3.09, attributable fraction [AF] 1-34%), substance use (aHR 1.57, 99%CI 1.23-2.02, AF 8-22%), and treatment supervised by health departments (aHR 1.42, 99%CI 1.03-1.97, AF 2-28%). Among 211 foreign-born patients, recurrence was associated with human immunodeficiency virus infection (aHR 2.24, 99%CI 1.27-3.98, AF 2-9%) and smear-positive TB (aHR 1.56, 99%CI 1.06-2.30, AF 3-33%). CONCLUSION: Factors associated with recurrence differed by origin of birth, and might be useful for anticipating greater risk for recurrent TB among certain patients with a history of TB.

SETTING: Drug susceptibility testing (DST) is recommended in Kenya to identify multidrug-resistant tuberculosis (MDR-TB) in persons registered for tuberculosis (TB) retreatment. DST is performed at a central laboratory with a two-step growth-based process and a regional laboratory with a simultaneous molecular- and growth-based process. OBJECTIVE: To compare proportions of retreatment cases who underwent DST and turnaround times for hospitals referring to the central vs. regional laboratory. DESIGN: Cases were persons registered for TB retreatment from 1 January 2012 to 31 December 2013. Records of 11 hospitals and 7 hospitals referring patients to the regional and central laboratories, respectively, were reviewed. RESULTS: Respectively 238/432 (55%) and 88/355 (25%) cases from hospitals referring to the regional and central laboratories underwent DST. The mean time from case registration to receipt of DST results and initiation of MDR-TB treatment was quicker in hospitals referring to the regional laboratory. The time required for the transportation of specimens, specimen testing and receipt of DST results at hospitals was shorter for the regional laboratory (P < 0.05). CONCLUSION: Testing was faster and more complete at hospitals referring to the regional laboratory. A comprehensive review of MDR-TB detection in Kenya is required to increase the proportion of cases receiving DST.

BACKGROUND: Sputum culture conversion is often used as an early microbiological endpoint in phase 2 clinical trials of tuberculosis treatment on the basis of its assumed predictive value for end-of-treatment outcome, particularly in patients with drug-susceptible tuberculosis. We aimed to assess the validity of sputum culture conversion on solid media at varying timepoints, and the time to conversion, as prognostic markers for end-of-treatment outcome in patients with multidrug-resistant (MDR) tuberculosis. METHODS: We analysed data from two large cohort studies of patients with MDR tuberculosis. We defined sputum culture conversion as two or more consecutive negative cultures from sputum samples obtained at least 30 days apart. To estimate the association of 2 month and 6 month conversion with successful treatment outcome, we calculated odds ratios (ORs) and 95% CIs with random-effects multivariable logistic regression. We calculated predictive values with bivariate random-effects generalised linear mixed modelling. FINDINGS: We assessed data for 1712 patients who had treatment success, treatment failure, or who died. Among patients with treatment success, median time to sputum culture conversion was significantly shorter than in those who had poor outcomes (2 months [IQR 1-3] vs 7 months [3 to ≥24]; log-rank p<0.0001). Furthermore, conversion status at 6 months (adjusted OR 14.07 [95% CI 10.05-19.71]) was significantly associated with treatment success compared with failure or death. Sputum culture conversion status at 2 months was significantly associated with treatment success only in patients who were HIV negative (adjusted OR 4.12 [95% CI 2.25-7.54]) or who had unknown HIV infection (3.59 [1.96-6.58]), but not in those who were HIV positive (0.38 [0.12-1.18]). Thus, the overall association of sputum culture conversion with a successful outcome was substantially greater at 6 months than at 2 months. 2 month conversion had low sensitivity (27.3% [95% confidence limit 16.6-41.4]) and high specificity (89.8% [82.3-94.4]) for prediction of treatment success. Conversely, 6 month sputum culture conversion status had high sensitivity (91.8% [85.9-95.4]), but moderate specificity (57.8% [42.5-71.6]). The maximum combined sensitivity and specificity for sputum culture conversion was reached between month 6 and month 10 of treatment. INTERPRETATION: Time to sputum culture conversion, conversion status at 6 months, and conversion status at 2 months in patients without known HIV infection can be considered as proxy markers of end-of-treatment outcome in patients with MDR tuberculosis, although the overall association with treatment success is substantially stronger for 6 month than for 2 month conversion status. Investigators should consider these results regarding the validity of sputum culture conversion at various timepoints as an early predictor of treatment efficacy when designing phase 2 studies before investing substantial resources in large, long-term, phase 3 trials of new treatments for MDR tuberculosis. FUNDING: US Agency for International Development, US Centers for Disease Control and Prevention, Division of Intramural Research of the US National Institute of Allergy and Infectious Diseases, Korea Centers for Disease Control and Prevention.

RATIONALE: Rifapentine has potent activity in mouse models of tuberculosis chemotherapy but its optimal dose and exposure in humans are unknown. OBJECTIVES: We conducted a randomized, partially blinded dose-ranging study to determine tolerability, safety, and antimicrobial activity of daily rifapentine for pulmonary tuberculosis treatment. METHODS: Adults with sputum smear-positive pulmonary tuberculosis were assigned rifapentine 10, 15, or 20 mg/kg or rifampin 10 mg/kg daily for 8 weeks (intensive phase), with isoniazid, pyrazinamide, and ethambutol. The primary tolerability end point was treatment discontinuation. The primary efficacy end point was negative sputum cultures at completion of intensive phase. MEASUREMENTS AND MAIN RESULTS: A total of 334 participants were enrolled. At completion of intensive phase, cultures on solid media were negative in 81.3% of participants in the rifampin group versus 92.5% (P = 0.097), 89.4% (P = 0.29), and 94.7% (P = 0.049) in the rifapentine 10, 15, and 20 mg/kg groups. Liquid cultures were negative in 56.3% (rifampin group) versus 74.6% (P = 0.042), 69.7% (P = 0.16), and 82.5% (P = 0.004), respectively. Compared with the rifampin group, the proportion negative at the end of intensive phase was higher among rifapentine recipients who had high rifapentine areas under the concentration-time curve. Percentages of participants discontinuing assigned treatment for reasons other than microbiologic ineligibility were similar across groups (rifampin, 8.2%; rifapentine 10, 15, or 20 mg/kg, 3.4, 2.5, and 7.4%, respectively). CONCLUSIONS: Daily rifapentine was well-tolerated and safe. High rifapentine exposures were associated with high levels of sputum sterilization at completion of intensive phase. Further studies are warranted to determine if regimens that deliver high rifapentine exposures can shorten treatment duration to less than 6 months. Clinical trial registered with www.clinicaltrials.gov (NCT 00694629).

SETTING: Twenty tuberculosis (TB) clinics in the United States and Canada. OBJECTIVE: To evaluate the efficacy and safety of a 6- month intermittent regimen of rifampin (RMP), pyrazinamide (PZA) and ethambutol (EMB) in human immunodeficiency virus (HIV) negative patients with culture-confirmed pulmonary or extra-pulmonary tuberculosis and either isoniazid (INH) resistance or INH intolerance. DESIGN: Patients were enrolled in a single-arm clinical trial to receive intermittent dosing after at least 14 initial daily doses of RMP+PZA+EMB. Treatment was continued twice (BIW) or thrice weekly (TIW) per physician/patient preference for a total of 6 months, with 2 years of follow-up for relapse after treatment. RESULTS: From 1999 to 2004, 98 patients were enrolled, 78 with reported INH resistance and 20 with INH intolerance. BIW dosing was used in 77 and TIW in 21. Study treatment was completed in 73 (74%). Reasons for discontinuation were hepatic adverse events (n = 12), other adverse effects (n = 3) and other reasons (n = 10). Failure (n = 1) and relapse (n = 2) occurred in 3 (3.5%, 95%CI 1.2-9.8) of 86 patients eligible for efficacy analysis, all occurring in patients with cavitary, acid-fast bacilli smear-positive pulmonary TB. CONCLUSIONS: Intermittent RMP+PZA+EMB appears to be effective in HIV-negative patients, but the regimen is poorly tolerated, possibly due to the prolonged use of PZA. Alternative regimens of lower toxicity are needed.

We conducted a prospective study to determine which solid medium is the most reliable overall and after two months of therapy to detect Mycobacterium tuberculosis complex (MTB). MTB isolation and contamination rates on LJ and Middlebrook 7H10 and 7H11 agar with and without selective antibiotics were examined in a single laboratory and compared against a constructed reference standard and MGIT 960 results. Of 50 smear positive adults with pulmonary TB enrolled, 45 successfully completed standard treatment. Two spot sputum specimens were collected before treatment and at week 8 and one spot specimen each at weeks 2, 4, 6, and 12. The MTB recovery rate among all solid media for pre-treatment specimens was similar. After 8 weeks, selective (S) 7H11 had the highest positivity rate. Latent class analysis was used to construct the primary reference standard. The 98.7% sensitivity of 7H11S (95% Wilson confidence interval 96.4%-99.6%) was highest among the 5 solid media (P = 0.003 by bootstrap); the 82.6% specificity of 7H10S (95% CI 75.7%-87.8%) was highest (P = 0.098). Our results support 7H11S as the medium of choice. Further studies in different areas where recovery and contamination are likely to vary, are recommended.