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

Abstract

ObjectiveTo systematically review and critically evaluate prediction models developed to predict tuberculosis (TB) treatment outcomes among adults with pulmonary TB.DesignSystematic review.Data sourcesPubMed, Embase, Web of Science and Google Scholar were searched for studies published from 1 January 1995 to 9 January 2020.Study selection and data extractionStudies 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.Results14 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.ConclusionsTB 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 registrationThe study was registered on the international prospective register of systematic reviews PROSPERO (CRD42020155782)

Impact of COVID-19 on Tuberculosis Case Detection and Treatment Outcomes in Sierra Leone

“…Additionally, we recorded significantly lower rates of loss to follow up and TB-related deaths during the COVID-19 pandemic. Our analysis of factors associated with treatment outcomes during the pre-and intra-COVID-19 periods identified the usual predictors of TB treatment success that have been described in the literature-i.e., younger age, newly diagnosed versus relapsed TB, pulmonary TB and HIV negative status [24][25][26]; however, the comparative success of monthly dispensing of anti-TB medications for self-administration at home over the traditional DOT model, though not entirely surprising, was an important and noteworthy finding in this setting.…”

Social and health factors associated with unfavourable treatment outcome in adolescents and young adults with tuberculosis in Brazil: a national retrospective cohort study

“…35 A systematic review of prediction models for pulmonary tuberculosis outcomes in adults also found that HIV was a common predictor of unfavourable treatment outcome. 36 Tubercu losis accounts for a quarter of HIVrelated deaths globally, and management and care of young people with tuberculosis-HIV coinfection is often challenging, especially among those with advanced HIV. 37,38 Improved access to well integrated tuberculosis and HIV services is crucial to improve outcomes for people living with HIV and diagnosed with tuberculosis, especially for affected adolescents and young adults.…”

Benchmarking Machine Learning Models to Assist in the Prognosis of Tuberculosis

“…Recently, Peetluk et al [23] published the first systematic review regarding models proposed to predict TB treatment outcomes. They followed the WHO definition of treatment outcomes for patients with TB i.e., treatment completion, cure, treatment success, treatment failure, death, loss to follow-up, and not evaluated.…”

“…None of the 16 cited papers that examined death as an outcome used machine learning; 11 used LR. It is important to note that Peetluk et al [23] do not classify LR as machine learning in their review as the LR analysis was used as a statistical methodology to understand the relationship between attributes and their prevalence. In the few machine learning studies identified, it was used primarily for predicting treatment completion [39] or unfavourable outcomes [40,41].…”

“…While it did not feature in Peetluk et al's systematic review [23], Kalhori et al [42] explored the use of machine learning to predict the outcome of a course of TB treatment. Using a data set of 6450 TB incidence from Iran in 2005, they compared six classifiers including DT, Bayesian networks, LR, MLP, Radial Basis Function, and SVM.…”