Prescriber preference may influence prolonged antibiotic usage more than patient characteristics in long-term care facilities
- 1Department of Epidemiology & Public health and School of Pharmacy, University College Cork, Cork, Ireland;
- 2School of Pharmacy, University College Cork, Cork, Ireland
- Correspondence to: Aoife Fleming, Department of Epidemiology & Public health and School of Pharmacy, University College Cork, Cork, Ireland;
Implications for practice and research
Longer antimicrobial prescription durations are often driven by prescriber preference.
Antimicrobial stewardship strategies should focus on prescription durations and target the determinants of prescriber behaviours in order to improve long-term care facilities (LTCF) antimicrobial prescribing.
Recent point prevalence studies have revealed that antimicrobial use in LTCF is highly variable, with 1.4% of residents in Germany and Latvia to 19.4% of residents in Northern Ireland being prescribed an antimicrobial.1 Daneman et al2 conducted a point prevalence study in Ontario, Canada in 2009 and found that, on average 5.9% (range=2.24–10.8%) of residents were receiving an antibiotic. A wide variability in antimicrobial use across LTCF was reported, which was potentially driven by prescriber factors. In this 2010 study, the authors sought to further investigate this variability by examining the duration of antimicrobial prescription courses.
This was a retrospective cohort study of 630 LTCF with a total resident population of 66 901 in Ontario. Population-based administrative databases (prescription, long-term care and hospitalisations databases) were used to identify all adults aged 66 years or older who had received a systemic antibiotic treatment prescription in 2010. Details of the antimicrobial prescriptions, the residents and the prescribers were collected. The variation in prescription duration by prescriber was determined using logistic regression and was based on prolonged treatment courses lasting longer than 7 days. The findings are reported as OR.
In total, 50 061 residents received an antimicrobial treatment and the most commonly selected duration was 7 days. A mixed logistic regression model found that prescribers were an important factor in determining treatment duration (p<0.001). Resident factors were similar across all prescribers. Longer duration prescribers were more likely to select treatment durations of 10 or 14 days than short or average duration prescribers. The 75th centile prescribers were nearly four times more likely to prescribe prolonged treatment durations than 25th centile prescribers (OR=3.84).
The main finding of this study is that, certain prescribers are nearly four times more likely to prescribe a longer duration of antimicrobial prescription than other prescribers. This potentially increases the risk of inappropriate antimicrobial prescribing. A recent meta-analysis of different infection conditions, including urinary tract infection and pneumonia, found that short courses of antimicrobials are often as effective as longer courses.3
This well-constructed, large-scale study has taken advantage of the databases available to evaluate LTCF antimicrobial prescribing by linking prescription and long-term care records. The authors address the limitation of the low prevalence of diagnostic claims, which means that we cannot be certain for what indications the prolonged treatments were prescribed.
The results largely concur with those of other studies, which have identified that antimicrobial prescribing behaviour is influenced by social and attitudinal factors and beliefs.4 A previous study by Daneman et al2 found that only 43.9% of patients receiving antibiotics had a physician visit within 1 day of the first prescription. It is important that prescribers in LTCF are aware that their individual prescribing patterns can influence the overall antibiotic usage in their LTCF. A systematic review of trials to improve antimicrobial prescribing in LTCF concluded that the effect of interventions were modest and unsustained.5 The studies included did not address individual prescribing patterns; and only one of the four trials measured the effect of the intervention on prescription duration. It would be interesting for future research to evaluate the effect of feeding-back prescribing information on LTCF prescribers. As nurses are the main care-givers to LTCF residents, it is important that they are aware of these issues and that they monitor prescription durations.
This study is an important addition to the evidence and highlights the need for in-depth qualitative research to identify the determinants of antimicrobial prescribing in LTCF. This research also highlights that shortening antimicrobial prescription duration should be a focus of future antimicrobial stewardship strategies by all healthcare professionals in LTCF.