Antimicrobial prescribing and stewardship in an ageing population
MetadataShow full item record
This item's downloads: 176 (view details)
Background: Antimicrobial resistance (AMR) is a global public health priority. This is primarily due to the rapid increase in AMR bacteria causing common healthcare-associated infections (HAI) and community-acquired infections (e.g. urinary tract infections and pneumonia). Inappropriate antimicrobial use (AMU) is an important cause of AMR. Of particular concern is AMR in long-term care facilities (LTCF), where 40-75% inappropriate prescribing occurs, leading to increased levels of multidrug resistanct (MDR) bacteria. Inappropriate AMU in LTCF includes prophylactic and broad-spectrum antimicrobial prescribing, which are contrary to local prescribing guidelines. Adverse drug events (ADEs), such as HAI as well as gastrointestinal, skin-related and neurological events, are often associated with antimicrobial use. Prevention and control of ADEs relating to antimicrobials are central to patient safety, which can be achieved through improved prescribing. Antimicrobial stewardship (AMS) represents an effective intervention for improving AMU, reducing ADEs and slowing the emergence and spread of resistant bacteria. This thesis aims to analyse antimicrobial prescribing in order to improve antimicrobial use and promote AMS in the LTCF in Ireland. In Ireland, point prevalence surveys (PPS) on HAI in LTCF (HALT) have been completed on four occasions since 2010, the most recent in 2016. However, in-depth understanding of the risk factors for HAI in LTCF remains limited due to lack of research. This thesis identifies the institutional and resident level risk factors associated with HAI and AMU from HALT PPS and the risk of ADEs relating to antimicrobials. This information can be useful for designing AMS interventions to improve AMU in Irish LTCF. Methods : This thesis is a combination of secondary data-analyses and has utilized data from several different sources. The main data sets used were from the HALT studies in Ireland. Chapter 3 analysed and compared the data from the 2013 and 2016 HALT studies to explore factors related to antimicrobial prophylaxis in LTCF in Ireland. Chapters 2 and 4 focused on multi-level analysis of the 2016 HALT data and aimed to identify LTCF-level and resident-level risk factors for AMU and HAI (Chapter 2) and to explore adherence to prescribing guidelines and identify risk factors associated with inappropriate prescribing in Irish LTCF (Chapter 4). It was necessary to collect additional data for these analyses, as the HALT data collected in May 2016 was only from those residents meeting the surveillance definition of HAI and/or AMU, and as such were not powered to perform multi-level analysis. Additional data on age, sex, the presence of a urinary catheter and disorientation was collected from all residents residing in LTCF in January 2017 and was matched with HALT 2016 data. Matching was performed to those on additional database by sex and age (closest in age) as well as urinary catheter use and disorientation; the case in the additional database was then replaced with the matched case from the original HALT 2016 database. The second data sets analysed were those from previously published randomised controlled trials (RCTs) on fluoroquinolones (FQs) (Chapter 5), which were used to perform a systematic review and meta-analysis of the risk of ADEs associated with FQs. The systematic review focused on comparing the risk of ADEs of FQs use to other antimicrobial agents used in primary care. The review followed the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) guidelines. Data management of review was carried out in Covidence and the meta-analysis was done in Revman. Thirdly, Chapter 6 analysed data from an RCT of amoxicillin for acute lower respiratory tract infections (RTI) from the GRACE study (Genomics to combat Resistance against Antibiotics in Community-acquired LRTI in Europe). A subgroup analysis of ADEs of amoxicillin was performed utilising the data collected as a part of this trial. Results : A total of 224 (10,044 residents) and 190 (9,318 residents) LTCF participated in HALT 2016 and 2013 respectively. In the 2016 study, the prevelance of HAI was 4.6% and AMU was 10.6% in Ireland. Of those on antimicrobials, 46% of the residents were on second-line antimicrobials. Out of the 224 LTCF participating in HALT 2016 that we contacted in January 2017, 80 LTCF provided additional information on 3,816 residents. Multi-level analysis identified between-LTCF variation in HAI and AMU (Chapter 2) as well as second-line AMU (Chapter 4). Less AMU was associated with: provision of medical care by the personal general practitioner (OR: 0.6; 95% confidence interval (CI): 0.7-1.0); increased number of healthcare assistants (OR: 0.9; 95% CI: 0.9-1.0); the presence of a coordinating physician (Odds ratio (OR): 0.3; 95% CI: 0.2-0.6); an AMS committee (OR: 0.2; 95%; CI: 0.1-0.6) and feedback on antimicrobial consumption (OR: 0.3; 95% CI: 0.1-0.6). Feedback on surveillance of infection prevention and control (IPC) practices (OR: 0.6; 95% CI: 0.3-1.0) were associated with less HAI. Providing education to prescribers on appropriate AMU and guideline use (OR=0.2; 95% CI: 0.1-0.7) was associated with less second-line AMU. However, a resident risk factor, the presence of a urinary catheter, increased the risk of HAI by 2.6 times and AMU by 2.2 times. The comparison of two HALT studies on prophylactic antimicrobials (Chapter 3) showed increased use of prophylactic antimicrobials in HALT 2016 compared to HALT 2013. Overall, more than 40% of the residents were on prophylactic antimicrobials. Prophylactic antimicrobials were most frequently prescribed for urinary tract-related conditions (72%), followed by respiratory tract (15%), and skin wound (8%). The main prescribed prophylactic agents were nitrofurantoin (39%), trimethoprim (41%) for urinary tract, macrolides (47%) for respiratory tract and macrolides and tetracycline (56%) for skin wound-related conditions. FQs accounted for 2-5% of prophylactic antimicrobials. Female residents and residents living in LTCF for more than a year were more often on prophylactic antimicrobials. The systematic review and meta-analysis (Chapter 5) identified that FQ use was significantly associated with central nervous system (CNS)-related ADEs (OR: 1.4; 95% CI: 1.1-1.7; P = 0.003; heterogeneity (I2) = 0%) and gastrointestinal (GI) tract-related ADEs (OR: 1.2; 95% CI: 1.1-1.4; P = 0.005; I2 = 80%). Withdrawal or discontinuation due to drug-related ADEs was higher with FQs (OR: 1.2; 95% CI: 1.0-1.4; P = 0.05; I2 = 5%) compared to other antimicrobials. The subgroup analysis (Chapter 6) identified that amoxicillin use caused a significantly higher proportion of any ADEs (diarrhoea or nausea or rash) (OR: 1.3; 95% CI: 1.1-1.6, number needed to harm (NNH) = 24) and of diarrhoea (OR: 1.4; 95% CI: 1.1-1.9, NNH = 29) compared to placebo group. No subgroup of patients were at increased risk of any ADEs, except for rash in males prescribed amoxicillin (interaction term 3.7; 95% CI: 1.2-11.4; OR of amoxicillin in males 2.8; 95% CI: 1.1-7.2). Conclusion : The collection of limited additional information on all residents from HALT 2016 participating LTCF increased the power of the study and allowed for multilevel analysis. The analysis showed substantial variation in HAI and AMU and identified significant associations at LTCF and resident level. Hence, the inclusion of all residents in the LTCF is recommended for future HALT PPS. Adherence to antimicrobial prescribing guidelines was low, with a high prevalence of prophylactic and second-line AMU in the LTCF. The use of broad-spectrum antimicrobials, specifically FQs, was associated with a higher risk of CNS- and GI-related ADEs. Therefore, an effective and sustainable AMS interventions programme is required for Irish LTCF. Such a programme would improve AMU and elderly care, as well as reducing the risk of ADEs and AMR. The identified associated risk factors for HAI, AMU, and ADEs should be considered while designing AMS interventions for LTCF.
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. Please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.
The following license files are associated with this item: