Models that predict the prevalence of antimicrobial resistance worldwide could be used to prioritize surveillance efforts, a study finds. Antimicrobial resistance is a growing public health problem around the globe, and surveillance plays a key role in the development of mitigation strategies. But surveillance is expensive and therefore has been limited in low- and middle-income countries (LMIC), potentially further exacerbating antimicrobial resistance due to inappropriate antibiotic usage. To address this problem, Rik Oldenkamp, Antonio Cappuccio, and colleagues developed statistical models that use socioeconomic data to predict the prevalence of clinical antimicrobial resistance. First, the authors modeled the association between existing antimicrobial-resistance prevalence data and national socioeconomic profiles. Next, the authors used the models to estimate the prevalence of antimicrobial resistance in countries lacking surveillance data. The models showed high predictive accuracy for six out of nine priority pathogens. Estimates for two priority pathogens expanded coverage beyond the existing global prevalence map to include several billion additional people, mostly in LMIC, altogether representing 99% of the worldwide population. Moreover, the results suggest that the Middle East, Sub-Saharan Africa, and Pacific Islands should be prioritized for surveillance efforts. According to the authors, the study reveals a much-needed affordable method for systematically estimating the prevalence of antimicrobial resistance where resources are limited.
###
Article #20-13515: "Filling the gaps in the global prevalence map of clinical antimicrobial resistance," by Rik Oldenkamp, Constance Schultsz, Emiliano Mancini, and Antonio Cappuccio.
MEDIA CONTACTS: Rik Oldenkamp, University of Amsterdam, NETHERLANDS; e-mail: < r.oldenkamp@aighd.org >; Antonio Cappuccio, Icahn School of Medicine at Mount Sinai, New York, NY; e-mail: < antonio.cappuccio@mssm.edu >
Proceedings of the National Academy of Sciences