Background In highly resource-limited settings, many clinics lack same-day microbiological testing for active tuberculosis (TB). In these contexts, risk of pretreatment loss to follow-up is high, and a simple, easy-to-use clinical risk score could be useful. Methods and findings We analyzed data from adults tested for TB with Xpert MTB/RIF across 28 primary health clinics in rural South Africa (between July 2016 and January 2018). We used least absolute shrinkage and selection operator regression to identify characteristics associated with Xpert-confirmed TB and converted coefficients into a simple score. We assessed discrimination using receiver operating characteristic (ROC) curves, calibration using Cox linear logistic regression, and clinical utility using decision curves. We validated the score externally in a population of adults tested for TB across 4 primary health clinics in urban Uganda (between May 2018 and December 2019). Model development was repeated de novo with the Ugandan population to compare clinical scores. The South African and Ugandan cohorts included 701 and 106 individuals who tested positive for TB, respectively, and 686 and 281 randomly selected individuals who tested negative. Compared to the Ugandan cohort, the South African cohort was older (41% versus 19% aged 45 years or older), had similar breakdown of biological sex (48% versus 50% female), and had higher HIV prevalence (45% versus 34%). The final prediction model, scored from 0 to 10, included 6 characteristics: age, sex, HIV (2 points), diabetes, number of classical TB symptoms (cough, fever, weight loss, and night sweats; 1 point each), and >14-day symptom duration. Discrimination was moderate in the derivation (c-statistic = 0.82, 95% CI = 0.81 to 0.82) and validation (c-statistic = 0.75, 95% CI = 0.69 to 0.80) populations. A patient with 10% pretest probability of TB would have a posttest probability of 4% with a score of 3/10 versus 43% with a score of 7/10. The de novo Ugandan model contained similar characteristics and performed equally well. Our study may be subject to spectrum bias as we only included a random sample of people without TB from each cohort. This score is only meant to guide management while awaiting microbiological results, not intended as a community-based triage test (i.e., to identify individuals who should receive further testing). Conclusions In this study, we observed that a simple clinical risk score reasonably distinguished individuals with and without TB among those submitting sputum for diagnosis. Subject to prospective validation, this score might be useful in settings with constrained diagnostic resources where concern for pretreatment loss to follow-up is high.

Background Clinical tuberculosis diagnosis and empiric treatment have traditionally been common among patients with negative bacteriologic test results. Increasing availability of rapid molecular diagnostic tests, including Xpert MTB/RIF and the new Xpert Ultra cartridge, may alter the role of empiric treatment. Methods We prospectively enrolled outpatients age > = 15 who were evaluated for pulmonary tuberculosis at three health facilities in Kampala, Uganda. Using sputum mycobacterial culture, interviews, and clinical record abstraction, we estimated the accuracy of clinical diagnosis relative to Xpert and sputum culture and assessed the contribution of clinical diagnosis to case detection. Results Over a period of 9 months, 99 patients were diagnosed with pulmonary tuberculosis and subsequently completed sputum culture; they were matched to 196 patients receiving negative tuberculosis evaluations in the same facilities. Xpert was included in the evaluation of 291 (99%) patients. Compared to culture, Xpert had a sensitivity of 92% (95% confidence interval 83–97%) and specificity of 95% (92–98%). Twenty patients with negative Xpert were clinically diagnosed with tuberculosis and subsequently had their culture status determined; two (10%) were culture-positive. Considering all treated patients regardless of Xpert and culture data completeness, and considering treatment initiations before a positive Xpert (N = 4) to be empiric, 26/101 (26%) tuberculosis treatment courses were started empirically. Compared to sputum smear- or Xpert-positive patients with positive cultures, empirically-treated, Xpert-negative patients with negative cultures had higher prevalence of HIV (67% versus 37%), shorter duration of cough (median 4 versus 8 weeks), and lower inflammatory markers (median CRP 7 versus 101 mg/L). Conclusion Judged against sputum culture in a routine care setting of high HIV prevalence, the accuracy of Xpert was high. Clinical judgment identified a small number of additional culture-positive cases, but with poor specificity. Although clinicians should continue to prescribe tuberculosis treatment for Xpert-negative patients whose clinical presentations strongly suggest pulmonary tuberculosis, they should also carefully consider alternative diagnoses.

Background New, sensitive diagnostic tests facilitate identification and investigation of milder forms of tuberculosis (TB) disease. We used community-based TB testing with the Xpert MTB/RIF Ultra assay (“Ultra”) to characterize individuals with previously undiagnosed TB and compare them to those from the same community who were diagnosed with TB through routine care. Methods We offered community-based sputum Ultra testing to adult residents of a well-defined area (population 34 000 adults) in Kampala, Uganda, via door-to-door screening and venue-based testing, then used detailed interview and laboratory testing to characterize TB-positive individuals. We compared these individuals to residents diagnosed with pulmonary TB at local health facilities and a representative sample of residents without TB (controls). Results Of 12 032 residents with interpretable Ultra results, 113 (940 [95% confidence interval {CI}, 780–1130] per 100 000) tested positive, including 71 (63%) positive at the lowest (trace) level. A spectrum of TB disease was observed in terms of chronic cough (93% among health facility–diagnosed cases, 77% among residents with positive community-based Ultra results at levels above trace, 33% among trace-positive community participants, and 18% among TB-negative controls), TB symptom prevalence (99%, 87%, 60%, and 38%, respectively), and C-reactive protein (75th percentile: 101 mg/L, 28 mg/L, 6 mg/L, and 4 mg/L, respectively). Community-diagnosed cases were less likely than health facility–diagnosed cases to have human immunodeficiency virus coinfection or previous TB. The specificity of Ultra was 99.4% (95% CI, 99.2%–99.5%) relative to a single spot sputum culture. Conclusions People with undiagnosed prevalent TB in the community have different characteristics than those diagnosed with pulmonary TB in health facilities. Newer diagnostic tests may identify a group of people with early or very mild disease.

Background: Challenges accessing nearby health facilities may be a barrier to initiating and completing tuberculosis (TB) treatment. We aimed to evaluate whether distance from residence to health facility chosen for treatment is associated with TB treatment outcomes. Methods: We conducted a retrospective cohort study of all patients initiating TB treatment at six health facilities in Kampala from 2014 to 2016. We investigated associations between distance to treating facility and unfavorable TB treatment outcomes (death, loss to follow up, or treatment failure) using multivariable Poisson regression. Results: Unfavorable treatment outcomes occurred in 20% (339/1691) of TB patients. The adjusted relative risk (aRR) for unfavorable treatment outcomes (compared to treatment success) was 0.87 (95% confidence interval [CI] 0.70, 1.07) for patients living ≥2 km from the facility compared to those living closer. When we separately compared each type of unfavorable treatment outcome to favorable outcomes, those living ≥2 km from the facility had increased risk of death (aRR 1.42 [95%CI 0.99, 2.03]) but decreased risk for loss to follow-up (aRR 0.57 [95%CI 0.41, 0.78]) than those living within 2 km. Conclusions: Distance from home residence to TB treatment facility is associated with increased risk of death but decreased risk of loss to follow up. Those who seek care further from home may have advanced disease, but once enrolled may be more likely to remain in treatment.

Background: Routine tuberculosis (TB) notifications are geographically heterogeneous, but their utility in predicting the location of undiagnosed TB cases is unclear. We aimed to identify small-scale geographic areas with high TB notification rates based on routinely collected data and to evaluate whether these areas have a correspondingly high rate of undiagnosed prevalent TB. Methods: We used routinely collected data to identify geographic areas with high TB notification rates and evaluated the extent to which these areas correlated with the location of undiagnosed cases during a subsequent community-wide active case finding intervention in Kampala, Uganda. We first enrolled all adults who lived within 35 contiguous zones and were diagnosed through routine care at four local TB Diagnosis and Treatment Units. We calculated average monthly TB notification rates in each zone and defined geographic areas of "high risk" as zones that constituted the 20% of the population with highest notification rates. We compared the observed proportion of TB notifications among residents of these high-risk zones to the expected proportion, using simulated estimates based on population size and random variation alone. We then evaluated the extent to which these high-risk zones identified areas with high burdens of undiagnosed TB during a subsequent community-based active case finding campaign using a chi-square test.