Systematic review on ultrasound patterns of TB disease

A systematic review investigating the literature for sonographic findings associated with TB. The aim is to describe the spectrum of sonographic TB disease as well as identify possible targets for novel point-of-care ultrasound protocols.  

Economic & Equity Analysis of Centralized vs Decentralized TB Diagnosis

While TB testing has improved greatly in the past decades, there is still a need to improve equitable, accessible, cost-effective, and affordable diagnostic testing approaches for TB.  This project investigates the impact of decentralized versus centralized diagnostic approaches on economic and health outcomes and equitable access from a health system and patient perspective.  We looked at data from 4200 patients in two high burden countries (Mozambique and Tanzania) in an effort to perform a cost-effectiveness and a benefit incidence analysis, map needs, and carry out qualitative research. 

Decentralized TB diagnostic algorithms will become increasingly feasible with the introduction of innovative diagnostic tests, such as the TrueNat test, which solves some of the bottlenecks by bringing the diagnostic system closer to patients in more peripheral clinics.

Parallel to this, the WHO recently recommended centralized assays for TB diagnosis that allows for multi-disease diagnosis (e.g., HIV, hepatitis) and high-throughput testing which could generate economies of scale. According to the WHO, such platforms could enhance system efficiency, provide cost savings, expand patient access, and ultimately improve quality of care.

These diagnostic advancements raise new questions about whether to pursue a centralized or decentralized strategy. To inform this decision making, this project provides the evidence required beyond clinical/diagnostic effectiveness, looking into relevant policy issues such as costs in relation to benefits, the potential for financial risk protection, equity in access to diagnostics and resulting care, and affordability of the technology at national level.

A qualitative evaluation of preferences of healthcare providers, patients, and decision makers for the implementation of decentralized TB testing is also ongoing as part of the CORE trial.

TB-CAPT
Close the gap, increase Access, Provide adequate Therapy

To facilitate implementation of TB testing, TB-CAPT provides evidence for impactful implementation for TB and TB/HIV co-infection diagnostic strategies in multiple clinical trials in South Africa, Mozambique, and Tanzania. These trials aim to evaluate the most promising new technologies, inform policy change and implementation strategies on a regional, national, and global level as well as to consolidate and strengthen human capacity and in-country infrastructure for conducting clinical trials, implementation research and plans for new diagnostic methods. Three trials from part of TB-CAPT: the CORE trial on decentralized molecular diagnosis, the XDR trial on extended resistance testing and the HIV trial, on rapid, comprehensive diagnostics testing for most vulnerable and hospitalized people living with HIV. The project was initiated by Claudia while at FIND and is coordinated by FIND. The Heidelberg team contributes primarily on an economic and benefit incidence analysis in partnership with David Dowdy at John Hopkins University.

FIND

An agent-based simulation study of diagnostic interventions to address the diagnostic gap in hypertension

Cardiovascular diseases (CVD) are the leading cause of mortality globally with almost a third of all annual deaths worldwide. For CVDs hypertension is the leading modifiable risk factor, however, the necessity of at least one repeated measurement for a final diagnosis and the different accuracies of an operator-dependent blood pressure measurement constitute a barrier in access to treatment and management of the disease. We investigated the impact of scaling up diagnostics in reducing hypertension-induced morbidity and mortality in low-and-middle-income-countries by an agent-based simulation study. We clearly showed that increasing the coverage of screening for hypertension and reducing complexity of the screening is more important than reducing the diagnostic accuracy in lowering the global cardiovascular disease burden compared to the current standard of care. The manuscript is under review at PLOS Medicine. 

RECOVER

A Randomized Open label Phase-II Clinical Trial with or without Infusion of Plasma from Subjects after Convalescence of SARS-CoV-2 Infection in High-Risk Patients with Confirmed Severe SARS-CoV-2 Disease. This is a study investigating the potential value of plasma in subgroups in which plasma has not been previously investigated. This includes patients with severe COVID-19 who have cancer, immunosuppression, laboratory risk factors of advanced age. We also investigate the impact on antibody neutralization before and after plasma donation to identify possible mechanisms or markers of protection. This work is aimed to identify therapeutic options for risk patients in the evolving pandemic landscape, where plasma is an adaptable therapy as opposed to e.g. monoclonal antibody therapy. 

Assessing acceptability and feasibility of urine-based LAM testing

In this qualitative study addressing needs and feasibility, we assessed the possible implementation of using urine-based LAM testing.  The study consisted of 42 semi-structured interviews with patients, health care providers, lab technicians, and decision makers in Malawi and Zambia from September 2020 to March 2021.  The assessment was conducted using the health equity implementation framework, which guided the presentation of the results.  Some possible barriers included the limited availability of sanitary facilities, change in agency, and the need of CD4 counts in HIV patients as a prerequisite.  The decentralization of diagnostics was considered possible with FujiLAM due to low infrastructure requirements. The easy testing and sample method as well as the test requiring no electricity or maintenance makes it a promising alternative in low- and middle-income countries.

TB hostmarker analysis – using machine learning

Within TB testing, there is still a need for a TB point of care triage test with 90% sensitivity and 70% specificity or better.  This design centered analysis will address the question: Can we develop such a triage test with at most 3 host-biomarkers? In order to answer this, we will develop a method using machine learning expertise, efficient python programming, parallelization of programs, and computer cluster usage with the hopes of improving TB testing.