Antibiotic prescribing and non-prescribing in nursing home residents with signs and symptoms ascribed to urinary tract infection (ANNA): study protocol for a cluster randomized controlled trial

Jeanine J S Rutten,Johannes C Van Der Wouden,Debby L Gerritsen,Ruth B Veenhuizen,Martin Smalbrugge,Philip D Sloane,Cees M P M Hertogh,Stephanie Natsch,Laura W Van Buul,Suzanne E Geerlings

doi:10.1186/s12877-020-01662-0

Abstract

BackgroundAntibiotic overprescribing for suspected urinary tract infection (UTI) in nursing homes (NHs) is common. Typical clinical scenarios in which antibiotics are inappropriately prescribed include response to nonspecific signs and symptoms and/or a positive urine test in the absence of symptoms referable to the urinary tract. These and other scenarios for inappropriate antibiotic prescribing were addressed in a recent international Delphi study which resulted in the development of a decision tool for the empiric treatment of UTI in frail older adults. The aim of the current study is to implement this decision tool, by integrating it into the electronic health record (EHR) and providing education on its content and use, and to evaluate its effect on appropriate antibiotic prescribing. An additional aim is to evaluate the quality of the intervention and the implementation process.MethodsA cluster Randomized Controlled Trial (cRCT) is conducted in sixteen NHs and aims to include 897 residents diagnosed with suspected UTI. NHs in the intervention group use the EHR-integrated decision tool, and receive education for physicians and nursing staff; in the control group care as usual is provided. Data is collected through case report forms within the EHR at the day of diagnosis and at 3, 7, and 21 days thereafter. The primary outcome is appropriate antibiotic prescribing for suspected UTI at the day of diagnosis. Secondary outcomes include the course of symptoms, alternative diagnoses, treatment changes, complications, hospitalization, and mortality. Data on total antibiotic prescribing are additionally collected in the participating NHs 12 months before and during the study. Finally, the process evaluation combines cRCT data with questionnaires and qualitative interviews with NH professionals.DiscussionThis is the first cRCT to evaluate the recently developed, international decision tool for empiric treatment of suspected UTI in NH residents. Study findings will elucidate the effect of the intervention on appropriate antibiotic prescribing for suspected UTI, and provide insight into the applicability of the decision tool in NHs in general and in specific subgroups of NH residents. With this study we aim to contribute to antibiotic stewardship efforts in long-term care.Trial registrationThe ANNA study was registered at the Netherlands Trial Register on 26 February 2019, with identification number NTR NL7555.

Highlights

Antibiotic overprescribing for suspected urinary tract infection (UTI) in nursing homes (NHs) is common
Study findings will elucidate the effect of the intervention on appropriate antibiotic prescribing for suspected UTI, and provide insight into the applicability of the decision tool in NHs in general and in specific subgroups of NH residents
We describe facilitators and barriers that may explain differences in the use and implementation of the intervention. These facilitators and barriers are extracted from descriptive questionnaire analysis and from deductive thematic analysis of interviews. This protocol paper described the design of a cluster Randomized Controlled Trial (cRCT) to evaluate whether an electronic health record (EHR)-integrated decision tool in combination with education results in an increase in appropriate antibiotic prescribing for NH residents with suspected UTI

Summary

Introduction

Antibiotic overprescribing for suspected urinary tract infection (UTI) in nursing homes (NHs) is common. Typical clinical scenarios in which antibiotics are inappropriately prescribed include response to nonspecific signs and symptoms and/or a positive urine test in the absence of symptoms referable to the urinary tract. These and other scenarios for inappropriate antibiotic prescribing were addressed in a recent international Delphi study which resulted in the development of a decision tool for the empiric treatment of UTI in frail older adults. Diagnosing UTIs is challenging in the NH population because presenting symptoms are often not typical and cognitive disabilities can impede communication of experienced complaints This diagnostic uncertainty, in combination with other factors such as expectations of residents and their family members, drives antibiotic prescribing to be ‘better safe than sorry’ [2]. Urine testing is considered only useful to rule out UTI when the results are negative, but otherwise should not influence treatment decisions [4]

Objectives

Methods

Findings

Conclusion