The impact of public health interventions on the future prevalence of ESBL-producing Klebsiella pneumoniae: a population based mathematical modelling study

Abstract

Background Future prevalence of colonization with extended-spectrum betalactamase (ESBL-) producing K. pneumoniae in humans and the potential of public health interventions against the spread of these resistant bacteria remain uncertain. Methods Based on antimicrobial consumption and susceptibility data recorded during > 13 years in a Swiss region, we developed a mathematical model to assess the comparative effect of different interventions on the prevalence of colonization. Results Simulated prevalence stabilized in the near future when rates of antimicrobial consumption and in-hospital transmission were assumed to remain stable (2025 prevalence: 6.8% (95CI%:5.4–8.8%) in hospitals, 3.5% (2.5–5.0%) in the community versus 6.1% (5.0–7.5%) and 3.2% (2.3–4.2%) in 2019, respectively). When overall antimicrobial consumption was set to decrease by 50%, 2025 prevalence declined by 75% in hospitals and by 64% in the community. A 50% decline in in-hospital transmission rate led to a reduction in 2025 prevalence of 31% in hospitals and no reduction in the community. The best model fit estimated that 49% (6–100%) of observed colonizations could be attributable to sources other than human-to-human transmission within the geographical setting. Conclusions Projections suggests that overall antimicrobial consumption will be, by far, the most powerful driver of prevalence and that a large fraction of colonizations could be attributed to non-local transmissions.

Modeling the transmission of antibiotic-resistant Enterobacterales in the community: a systematic review

“…Three models combining nosocomial and community transmission were used to compare the impact of interventions across these two settings. 22,24,27 MacFadden et al fitted a transmission model to observed proportions of ESBL production among E. coli over time in Sweden. 22 They predicted that a 20% decrease in antibiotic use in the community led to greater reductions in ESBL prevalence in both settings (community and hospital) than a same 20% decrease in hospital antibiotic use.…”

“…The other main type of data considered was aggregated counts of infections sensitive or resistant to a particular antibiotic, by unit of time at the region or country level. These data predominantly came from national surveillance systems, such as in Sweden or Switzerland, 22,27 or at the European level. 24,26 One study used aggregated infection cases from an intervention study.…”

“…Modeling their transmission within and between these two settings is thus highly relevant, and several of the reviewed studies did so. 22,24,27,28,32,38,39 However, only two studies used detailed antibiotic resistant infection data specific to the two environments. 22,27 In the future, extensive antibiotic resistance data collection in the community, through general practitioners or outpatient microbiological laboratories, should enable setting-specific model parametrization and therefore help to better characterize the contribution of each setting to transmission.…”

“…24 Salazar-Vizcaya et al predicted that a 50% decrease in consumption of all antibiotic classes in both hospitals and the community would lead to a small but significant decline in the prevalence of resistant K. pneumoniae in upcoming years in both settings, whereas a 50% decrease in the in-hospital transmission rate would lead to a decline only in hospital. 27 Empirical data for statistical inference of parameters.…”

The Contribution of Community Transmission to the Burden of Hospital-associated Pathogens: A Systematic Scoping Review of Epidemiological Models

“…Two studies (7%) with community transmission included zoonotic or foodborne transmission. 29,30 In terms of incorporating movement, we found that two studies (7%) utilized data to investigate the impacts of movement on colonization in the community setting between high and low-prevalence geographical regions 24,31 . Nine studies (31%) incorporated transfers between STACHs; among those studies, three of them modeled additional movement between STACHs and communities.…”