Primary Care Antibiotic Prescribing and Infection-Related Hospitalisation

Abstract

Inappropriate prescribing of antibiotics has been widely recognised as a leading cause of antimicrobial resistance, which in turn has become one of the most significant threats to global health. Given that most antibiotic prescriptions are issued in primary care settings, investigating the associations between primary care prescribing of antibiotics and subsequent infection-related hospitalisations affords a valuable opportunity to understand the long-term health implications of primary care antibiotic intervention. A narrative review of the scientific literature studying associations between primary care antibiotic prescribing and subsequent infection-related hospitalisation was conducted. The Web of Science database was used to retrieve 252 potentially relevant studies, with 23 of these studies included in this review (stratified by patient age and infection type). The majority of studies (n = 18) were published in the United Kingdom, while the remainder were conducted in Germany, Spain, Denmark, New Zealand, and the United States. While some of the reviewed studies demonstrated that appropriate and timely antibiotic prescribing in primary care could help reduce the need for hospitalisation, excessive antibiotic prescribing can lead to antimicrobial resistance, subsequently increasing the risk of infection-related hospitalisation. Few studies reported no association between primary care antibiotic prescriptions and subsequent infection-related hospitalisation. Overall, the disparate results in the extant literature attest to the conflicting factors influencing the decision-making regarding antibiotic prescribing and highlight the necessity of adopting a more patient-focussed perspective in stewardship programmes and the need for increased use of rapid diagnostic testing in primary care.

Machine learning for the prediction of urosepsis using electronic health record data

“…[4] However, pervasive antibiotic administration for all UTI does not appreciably decrease the risk of urosepsis [14][15][16][17] and may increase poor health outcomes with increased rates of hospitalization due to antibiotic adverse events, post-infectious complications, and subsequent multidrug-resistant infections. [27][28][29] Thus, identification of those patients at highest risk for septic progression of UTI might allow for targeted antibiotic administration to those in whom the benefits outweigh the risks and improve antimicrobial stewardship.…”

“…It is also possible that antibiotic treatment itself may contribute to hospitalization. [27][28][29] As many as 20% of patients prescribed antibiotics will experience an adverse effect of the medication, from allergic reactions to dehydration from antibiotic-associated diarrhea or C. difficile colitis. In addition, frequent antibiotic use is associated with increased hospital admissions for infection-related complications in a dose-dependent manner.…”

“…In addition, frequent antibiotic use is associated with increased hospital admissions for infection-related complications in a dose-dependent manner. [27] Most of these readmissions are seen in the few days following antibiotic administration, and the rates of hospital admissions for infectious complications increase with larger cumulative antibiotic exposures. [28, 29] Future studies may need to account for cumulative antibiotic burden to further improve predictive modeling.…”

“…In fact, higher antibiotic burdens over time are associated with increased numbers of hospital admissions for a wide range of infections and antibiotic side effects. [27][28][29] This study sought to apply machine learning algorithms to predict the risk of urosepsis in patients presenting with a UTI in the outpatient setting, an area ripe for harnessing the potential of machine learning to improve patient care. We sought to derive a predictive algorithm for urosepsis risk, intentionally using a minimal number of features which could be obtained easily at the outpatient point-of-care to facilitate ease-of-use and ensure applicability across a range of care delivery settings.…”

“…In fact, higher antibiotic burdens over time are associated with increased numbers of hospital admissions for a wide range of infections and antibiotic side effects. [27–29]…”