A cross-sectional survey of antimicrobial stewardship strategies in UK hospitals

Abstract

Antimicrobial stewardship programmes describe strategies to optimize antimicrobial prescribing and utilization, minimize resistance and improve patient outcomes. Strategies in hospitals are usually implemented by multidisciplinary antimicrobial teams (AMTs). The objective of this study was to describe the profile and activities of AMTs within hospitals in the United Kingdom (UK). All hospitals within the UK (n=836) were included, and a prepiloted questionnaire was mailed to the 'Director of Pharmacy'. Non-respondents were mailed up to two reminder questionnaires at two-weekly intervals. Main outcome measures are as follows: existence and remit of the AMTs; availability of antimicrobial-prescribing policies, aims, scope and methods of dissemination; and monitoring and feedback provided on antimicrobial policy adherence. Response rate was 33% (n=273). Completed questionnaires analysed were n=226. Eighty-two (n=186) of respondents indicated the presence of an AMT within the hospital, with 95% of these (n=177) reporting an antimicrobial pharmacist as part of the team. All AMTs (n=186) were involved in development of an antimicrobial policy and almost all (99% n=184) promoted adherence and restricting use of specific antimicrobials (97% n=180). Ninety-eight per cent of respondents (n=222) reported the availability of a local antimicrobial-prescribing policy within the hospital with this disseminated mainly through the hospital intranet (98% n=217). Adherence to policy was measured mainly through audits measuring the appropriateness of antimicrobial use against the local policy (76% n=169). Hospitals in England (P=0·010), tertiary care hospitals (P=0·021) and bed capacity >500 (P<0·001) were more likely to have an AMT, as were hospitals with an accident and emergency department (P<0·001), an infectious diseases unit (P=0·019) and a microbiology department (P<0·001). Audits to measure policy adherence were more likely (P<0·001) if an AMT was present. The only variable retained in bivariate logistic regression was the presence of a microbiology unit, with an odds ratio of 14·1 (95% CI 6·02-33·33, P<0·001). Although most respondents reported an antimicrobial-prescribing policy, less had an AMT. Despite recent government and regional initiatives, further improvements in antimicrobial stewardship are still required.