Antibiotics are prescribed when sick patients arrive at hospital, before doctors know what kind of bacteria is causing an infection and what antibiotics it might be resistant to. Even when a bacteria is identified as causing an infection, it is labelled as “resistant” (R) or “susceptible” (S) based on fairly arbitrary definitions. However, resistance is rarely all or nothing, mostly it is a degree with antibiotics finding it harder to kill some bacteria than others, rather than not working at all. Fear of treatment failure and resistance is a major driver for prescribing increasingly broad-spectrum antibiotics to ever larger groups of patients. In practice, the data supporting the clinical importance of “R” vs “S” for the initial antibiotics used is surprisingly scarce. One reason it may not be very important is because doctors regularly review patients during hospital and change their antibiotics for lots of reasons. So as long as patients get antibiotics initially, exactly what they are may not matter very much.
Within our anonymised linked database, we will therefore investigate the impact on clinical outcome of patients with bloodstream infections caused by one family of gram-negative bacteria receiving antibiotics with different amounts of drug susceptibility.
See publication: Mortality risks associated with empirical antibiotic activity in Escherichia coli bacteraemia: an analysis of electronic health records