Globally, the overuse of antibiotics has led to the spread of antimicrobial resistance, and this is a growing threat to child health. Antibiotics are frequently used to treat children with fever and respiratory symptoms, even though most of these illnesses will resolve on their own.
The measurement of C-reactive protein (CRP) in the blood, produced in response to inflammation, has been shown to decrease antibiotic use in children with fever and respiratory illness in health care facilities. But in many resource-constrained settings, patients first seek care outside the formal health sector, often from lay community health workers.
A team led by Emily Ciccone, MD, MHS, and Ross Boyce, MD, MSc, members of the UNC Institute for Global Health and Infectious Diseases, and Edgar M. Mulogo of the Department of Community Health at Mbarara University of Science and Technology, with researchers from the Gillings School of Global Public Health’s Departments of Biostatistics and Epidemiology, designed a study to determine the impact of CRP on antibiotic use for pediatric acute respiratory illness (ARI) in rural Uganda. Published in PLOS Medicine, this is the first study of C-reactive protein testing by community health workers, who, in this region, are volunteers with limited medical training.
Designing the Study
Designed as a stepped wedge cluster randomized trial, the study involved 65 community health workers (CHW) in 15 villages in rural western Uganda. Researchers wanted to compare the Integrated Community Care Management (iCCM) algorithm currently used by workers in the region with an amended study algorithm that included a CRP measurement for guiding antibiotic treatment decisions among children presenting with fever and respiratory symptoms. 
The villages were stratified according to altitude, distance to the main health facility in the subcounty, and size, and randomly assigned to one of five groups of three villages each. All villages started the study in a control period, evaluating children using the iCCM algorithm. Each month, one group of villages switched to evaluating children with the study intervention algorithm, called the STAR Sick Child Job Aid. Children aged two months to five years, presenting to the community health workers with fever and cough, were eligible. Follow-up assessments were conducted seven days after the initial visit. Altogether, the study enrolled 1,280 children, 1,220 of whom were included in the analyses.
The Results
The primary outcome of interest was the proportion of children who were given or prescribed an antibiotic at the initial visit. As secondary outcomes, researchers assessed adverse effects seven days after the visit, including persistent fever and clinical failure. They also accessed perceived improvement by each child’s caregiver. The study found that point-of-care CRP testing by community health workers significantly decreased antibiotic use.
“The community health workers were able to implement and interpret the CRP test as part of an algorithm for evaluation and management of children experiencing cough and/or fast-breathing, decreasing antibiotic use by 25% as compared to the current standard of care,” said Ciccone. “Notably, this decrease was not associated with an increased risk of clinical failure, although this outcome occurred infrequently in both groups.”
“As community health workers are often the first providers children in rural, resource-limited settings see when ill, CRP testing could have a significant impact on unnecessary antibiotic use for acute respiratory illness. These findings therefore support expanded, equitable access to point-of-care diagnostics for antibiotic stewardship.”
The results also provide rationale for larger scale studies of CRP testing incorporated into clinical algorithms that focus on safety outcomes, implementation, and cost-effectiveness.
Other authors included: Di Hu, MS; John Preisser, PhD; Caitlin A. Cassidy, MS; Lydiah Kabugho, BDS;Baguma Emmanuel, BCR; Georget Kibaba; Fred Mwebembezi, BIT; Jon Juliano, MD, MSPH; and Edgar Mulogo, MPH, MSc, PhD.