A randomized controlled trial of a novel, evidence-based algorithm for managing lower respiratory tract infection in a resource-limited setting

/ Abstract Lower respiratory tract infection (LRTI) is one of the most common reasons for hospitalization globally. Viral and bacterial LRTI present similarly, leading clinicians to overprescribe antibacterials for fear of missing a lethal bacterial infection or superinfection. However, emerging data from global cohorts indicate that viral LRTI is frequently more prevalent than bacterial LRTI in both children and adults. In low- or middle-income countries (LMICs), antibacterial overuse for viral LRTI is often worse given limited diagnostic capacity. Access to point- of-care (POC) diagnostic tests, which do not require laboratory infrastructure, may decrease antibacterial overuse for LRTI in LMICs. Locally relevant, evidence-based, cost-effective diagnostic algorithms for LRTI have not been systematically developed in LMICs. The objective of this proposal is to integrate multiple low- cost diagnostic tools (clinical predictors, POC pathogen tests, and POC biomarker tests) to develop and evaluate an LRTI diagnostic and treatment algorithm in a LMIC setting. We will use a large, existing, setting- specific biorepository of patients with LRTI to guide algorithm development. The following aims are proposed: 1) create an evidence-based algorithm for LRTI management by integrating clinical predictors, POC pathogen tests, and POC biomarker tests; 2) establish understanding, acceptability, and barriers to implementation of clinical algorithms for LRTI management among local physicians; and 3) evaluate an LRTI management algorithm in a stepped-wedge, cluster randomized trial at a single hospital in a LMIC. We will complete gold- standard testing and clinical adjudications of samples in our biorepository to identify etiology of infection. We will then construct decision trees by inputting 1) clinical predictors, 2) POC pathogen tests, and 3) POC biomarker tests to identify a potentially cost-effective algorithm that would reduce inappropriate antibacterial prescriptions. We will conduct focus group discussions with local physicians to identify barriers and facilitators to using clinical algorithms. Following algorithm development, we will reconvene focus groups to iterate on the algorithm and to determine appropriate methods for communicating and implementing the algorithm. We will then conduct a stepped-wedge cluster randomized trial to evaluate the algorithm. Patients admitted with LRTI will receive either 1) algorithm-directed care, or 2) usual care. To assess clinical outcomes and antibacterial duration concurrently in this trial, we will use the innovative Response Adjusted for Duration of Antibiotic Risk (RADAR) clinical trial design developed by the Antibacterial Resistance Leadership Group (ARLG). The expected outcome of this work is the development and evaluation of a LRTI diagnostic algorithm that uses local evidence and integrates multiple low-cost diagnostic tools. The long-term goal of this work is to translate these methods to other low-resource settings to combat the growing global crisis of antimicrobial resistance. Project Number: 3R01AI168420-04S1 | Fiscal Year: 2025 | NIH Institute/Center: National Institute of Allergy and Infectious Diseases (NIAID) | Principal Investigator: GAYANI TILLEKERATNE | Institution: DUKE UNIVERSITY, DURHAM, NC | Award Amount: $107,296 | Activity Code: R01 | Study Section: Clinical Informatics and Digital Health Study Section[CIDH] View on NIH RePORTER: https://reporter.nih.gov/project-details/3R01AI16842004S1