Loading…
Loading grant details…
| Funder | EUNICE KENNEDY SHRIVER NATIONAL INSTITUTE OF CHILD HEALTH & HUMAN DEVELOPMENT |
|---|---|
| Recipient Organization | Socios En Salud Sucursal Peru |
| Country | Peru |
| Start Date | Sep 01, 2024 |
| End Date | Aug 31, 2026 |
| Duration | 729 days |
| Number of Grantees | 1 |
| Roles | Principal Investigator |
| Data Source | NIH (US) |
| Grant ID | 10784569 |
SUMMARY Children and adolescents exposed to tuberculosis (TB) in their homes (i.e. household contacts) are at high risk for developing active TB disease if they do not receive TB preventive treatment (TPT). The management of child and adolescent household contacts is hampered by the low sensitivity of bacteriological diagnostic tools and
limited accessibility of chest-x ray (CXR) in primary care settings of low- and middle-income countries. Primary care physicians, who typically manage household contacts, will not prescribe TPT if they cannot decide whether a contact has TB disease. There is a need for tools to help primary care physicians distinguish between child
and adolescent household contacts who require further evaluation for active TB disease and those who can safely initiate TPT immediately. CXR is important for detecting incipient lesions of TB disease, but capacity for CXR is limited in primary care settings in low- and middle-income countries. Even in those places where CXR
can be accessed, there is not enough capacity among primary care physicians to correctly interpret the radiological findings. Our proposal will (1) derive a risk score to predict CXR abnormalities among child and adolescent household contacts and (2) evaluate the performance of computer-aided detection (CAD) software
for detecting CXR abnormalities in children and adolescents contacts. The risk score, meant to aid clinicians in places with limited CXR capacity, will be derived using data from a previous cohort study of 6001 contacts 0-19-years old. We will consider 21 candidate predictors that can be collected easily during a simple clinical evaluation.
Primary care physicians will be able to use this risk score to identify children and adolescents who should be referred for further evaluation with CXR, and can prescribe TPT to the rest. For settings that have access to CXR but where primary care physicians lack expertise in reading CXRs for TB, we will evaluate the performance of
three CAD software systems among children and adolescents: CAD4TB version 7 (Delft Imaging), qXR version 3.0 (qure.ai), and INSIGHT CXR version 3.1.0.0 (Lunit Inc). These systems have largely been validated in adults but not children and younger adolescents. Using 1600 stored radiographs from a large-scale community
screening program, we will perform a receiver operating characteristic analysis for each CAD software using a pulmonologist’s decision of CXR abnormality as the reference standard. We will compare performance among systems and between children and adolescents. This research will contribute to the development of simple tools
that facilitate the clinical decision-making of primary care physicians to prescribe TPT. Increasing the prescription of TPT to child and adolescent household contacts will reduce the incidence of TB and TB-associated mortality among children and adolescents worldwide, helping to achieve the goals of End TB Strategy.
Socios En Salud Sucursal Peru
Complete our application form to express your interest and we'll guide you through the process.
Apply for This Grant