Health care index score and risk of death following tuberculosis diagnosis in HIV-positive patients

Benchmarking Machine Learning Models to Assist in the Prognosis of Tuberculosis

Tuberculosis (TB) is an airborne infectious disease caused by organisms in the Mycobacterium tuberculosis (Mtb) complex. In many low and middle-income countries, TB remains a major cause of morbidity and mortality. Once a patient has been diagnosed with TB, it is critical that healthcare workers make the most appropriate treatment decision given the individual conditions of the patient and the likely course of the disease based on medical experience. Depending on the prognosis, delayed or inappropriate treatment can result in unsatisfactory results including the exacerbation of clinical symptoms, poor quality of life, and increased risk of death. This work benchmarks machine learning models to aid TB prognosis using a Brazilian health database of confirmed cases and deaths related to TB in the State of Amazonas. The goal is to predict the probability of death by TB thus aiding the prognosis of TB and associated treatment decision making process. In its original form, the data set comprised 36,228 records and 130 fields but suffered from missing, incomplete, or incorrect data. Following data cleaning and preprocessing, a revised data set was generated comprising 24,015 records and 38 fields, including 22,876 reported cured TB patients and 1139 deaths by TB. To explore how the data imbalance impacts model performance, two controlled experiments were designed using (1) imbalanced and (2) balanced data sets. The best result is achieved by the Gradient Boosting (GB) model using the balanced data set to predict TB-mortality, and the ensemble model composed by the Random Forest (RF), GB and Multi-Layer Perceptron (MLP) models is the best model to predict the cure class.

Systematic review of prediction models for pulmonary tuberculosis treatment outcomes in adults

OBJECTIVE

To systematically review and critically evaluate prediction models developed to predict tuberculosis (TB) treatment outcomes among adults with pulmonary TB.

DESIGN

Systematic review.

DATA SOURCES

PubMed, Embase, Web of Science and Google Scholar were searched for studies published from 1 January 1995 to 9 January 2020.

STUDY SELECTION AND DATA EXTRACTION

Studies that developed a model to predict pulmonary TB treatment outcomes were included. Study screening, data extraction and quality assessment were conducted independently by two reviewers. Study quality was evaluated using the Prediction model Risk Of Bias Assessment Tool. Data were synthesised with narrative review and in tables and figures.

RESULTS

14 739 articles were identified, 536 underwent full-text review and 33 studies presenting 37 prediction models were included. Model outcomes included death (n=16, 43%), treatment failure (n=6, 16%), default (n=6, 16%) or a composite outcome (n=9, 25%). Most models (n=30, 81%) measured discrimination (median c-statistic=0.75; IQR: 0.68-0.84), and 17 (46%) reported calibration, often the Hosmer-Lemeshow test (n=13). Nineteen (51%) models were internally validated, and six (16%) were externally validated. Eighteen (54%) studies mentioned missing data, and of those, half (n=9) used complete case analysis. The most common predictors included age, sex, extrapulmonary TB, body mass index, chest X-ray results, previous TB and HIV. Risk of bias varied across studies, but all studies had high risk of bias in their analysis.

CONCLUSIONS

TB outcome prediction models are heterogeneous with disparate outcome definitions, predictors and methodology. We do not recommend applying any in clinical settings without external validation, and encourage future researchers adhere to guidelines for developing and reporting of prediction models.

TRIAL REGISTRATION

The study was registered on the international prospective register of systematic reviews PROSPERO (CRD42020155782).

Prognostic score to predict mortality during TB treatment in TB/HIV co-infected patients

Background

Estimating mortality risk during TB treatment in HIV co-infected patients is challenging for health professionals, especially in a low TB prevalence population, due to the lack of a standardized prognostic system. The current study aimed to develop and validate a simple mortality prognostic scoring system for TB/HIV co-infected patients.

Methods

Using data from the CDC’s Tuberculosis Genotyping Information Management System of TB patients in Texas reported from 01/2010 through 12/2016, age ≥15 years, HIV(+), and outcome being “completed” or “died”, we developed and internally validated a mortality prognostic score using multiple logistic regression. Model discrimination was determined by the area under the receiver operating characteristic (ROC) curve (AUC). The model’s good calibration was determined by a non-significant Hosmer-Lemeshow’s goodness of fit test.

Results

Among the 450 patients included in the analysis, 57 (12.7%) died during TB treatment. The final prognostic score used six characteristics (age, residence in long-term care facility, meningeal TB, chest x-ray, culture positive, and culture not converted/unknown), which are routinely collected by TB programs. Prognostic scores were categorized into three groups that predicted mortality: low-risk (<20 points), medium-risk (20–25 points) and high-risk (>25 points). The model had good discrimination and calibration (AUC = 0.82; 0.80 in bootstrap validation), and a non-significant Hosmer-Lemeshow test p = 0.71.

Conclusion

Our simple validated mortality prognostic scoring system can be a practical tool for health professionals in identifying TB/HIV co-infected patients with high mortality risk.

One-year mortality of HIV-positive patients treated for rifampicin- and isoniazid-susceptible tuberculosis in Eastern Europe, Western Europe, and Latin America

OBJECTIVES

The high mortality among HIV/tuberculosis (TB) coinfected patients in Eastern Europe is partly explained by the high prevalence of drug-resistant TB. It remains unclear whether outcomes of HIV/TB patients with rifampicin/isoniazid-susceptible TB in Eastern Europe differ from those in Western Europe or Latin America.

METHODS

One-year mortality of HIV-positive patients with rifampicin/isoniazid-susceptible TB in Eastern Europe, Western Europe, and Latin America was analysed and compared in a prospective observational cohort study. Factors associated with death were analysed using Cox regression modelsRESULTS:: Three hundred and forty-one patients were included (Eastern Europe 127, Western Europe 165, Latin America 49). Proportions of patients with disseminated TB (50, 58, 59%) and initiating rifampicin + isoniazid + pyrazinamide-based treatment (93, 94, 94%) were similar in Eastern Europe, Western Europe, and Latin America respectively, whereas receipt of antiretroviral therapy at baseline and after 12 months was lower in Eastern Europe (17, 39, 39%, and 69, 94, 89%). The 1-year probability of death was 16% (95% confidence interval 11-24%) in Eastern Europe, vs. 4% (2-9%) in Western Europe and 9% (3-21%) in Latin America; P < 0.0001. After adjustment for IDU, CD4 cell count and receipt of antiretroviral therapy, those residing in Eastern Europe were at nearly 3-fold increased risk of death compared with those in Western Europe/Latin America (aHR 2.79 (1.15-6.76); P = 0.023).

CONCLUSIONS

Despite comparable use of recommended anti-TB treatment, mortality of patients with rifampicin/isoniazid-susceptible TB remained higher in Eastern Europe when compared with Western Europe/Latin America. The high mortality in Eastern Europe was only partially explained by IDU, use of ART and CD4 cell count. These results call for improvement of care for TB/HIV patients in Eastern Europe.

Major differences in organization and availability of health care and medicines for HIV/TB coinfected patients across Europe

OBJECTIVES

The aim of the study was to investigate the organization and delivery of HIV and tuberculosis (TB) health care and to analyse potential differences between treatment centres in Eastern (EE) and Western Europe (WE).

METHODS

Thirty-eight European HIV and TB treatment centres participating in the TB:HIV study within EuroCoord completed a survey on health care management for coinfected patients in 2013 (EE: 17 respondents; WE:21; 76% of all TB:HIV centres). Descriptive statistics were obtained for regional comparisons. The reported data on health care strategies were compared with actual clinical practice at patient level via data derived from the TB:HIV study.

RESULTS

Respondent centres in EE comprised: Belarus (n = 3), Estonia (1), Georgia (1), Latvia (1), Lithuania (1), Poland (4), Romania (1), the Russian Federation (4) and Ukraine (1); those in WE comprised: Belgium (1), Denmark (1), France (1), Italy (7), Spain (2), Switzerland (1) and UK (8). Compared with WE, treatment of HIV and TB in EE are less often located at the same site (47% in EE versus 100% in WE; P < 0.001) and less often provided by the same doctors (41% versus 90%, respectively; P = 0.002), whereas regular screening of HIV-infected patients for TB (80% versus 40%, respectively; P = 0.037) and directly observed treatment (88% versus 20%, respectively; P < 0.001) were more common in EE. The reported availability of rifabutin and second- and third-line anti-TB drugs was lower, and opioid substitution therapy (OST) was available at fewer centres in EE compared with WE (53% versus 100%, respectively; P < 0.001).

CONCLUSIONS

Major differences exist between EE and WE in relation to the organization and delivery of health care for HIV/TB-coinfected patients and the availability of anti-TB drugs and OST. Significant discrepancies between reported and actual clinical practices were found in EE.

Multi-drug-resistant tuberculosis in HIV positive patients in Eastern Europe

Observational data from Eastern Europe on the management and outcome of multi-drug-resistant tuberculosis (MDR TB) in HIV positive populations remain sparse in the English-language literature. We compared clinical characteristics and outcomes of 55 patients who were diagnosed with HIV and MDR TB in Eastern Europe between 2004 and 2006 to 89 patients whose Mycobacterium tuberculosis isolates were susceptible to isoniazid and rifampicin. Patients with HIV and MDR TB were young and predominantly male with high rates of intravenous drug use, imprisonment and hepatitis C co-infection. Eighty-four per cent of patients with MDR TB had no history of previous TB drug exposure suggesting that the majority of MDR TB resulted from transmission of drug-resistant M. tuberculosis. The use of non-standardized tuberculosis treatment was common, and the use of antiretroviral therapy infrequent. Compared to those with susceptible tuberculosis, patients with MDR TB were less likely to achieve cure or complete tuberculosis treatment (21.8% vs. 62.9%, p < 0.0001), and they were more likely to die (65.5% vs. 27.0%, p < 0.0001). Our study documents suboptimal management and poor outcomes in HIV positive patients with MDR TB. Implementation of WHO guidelines, rapid TB diagnostics and TB drug susceptibility testing for all patients remain a priority in this region.