0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation | Less Is More

Prolonged Antibiotic Treatment in Long-term Care:  Role of the Prescriber FREE

Nick Daneman, MD, MSc; Andrea Gruneir, PhD; Susan E. Bronskill, PhD; Alice Newman, MSc; Hadas D. Fischer, MD, MSc; Paula A. Rochon, MD, MSc; Geoff M. Anderson, MD, PhD; Chaim M. Bell, MD, PhD
[+] Author Affiliations

Author Affiliations: Institute for Clinical Evaluative Sciences (Drs Daneman, Gruneir, Bronskill, Fischer, Rochon, Anderson, and Bell and Ms Newman); and Division of Infectious Diseases, Department of Medicine, Sunnybrook Health Sciences Centre (Dr Daneman), Women's College Research Institute, Women's College Hospital (Drs Gruneir and Rochon), Institute of Health Policy, Management, and Evaluation (Drs Daneman, Gruneir, Bronskill, Rochon, Anderson, and Bell), and Division of General Internal Medicine, Mount Sinai Hospital (Dr Bell), University of Toronto, Toronto, Ontario, Canada.


JAMA Intern Med. 2013;173(8):673-682. doi:10.1001/jamainternmed.2013.3029.
Text Size: A A A
Published online

Importance Given that most common bacterial infections can be treated with antibiotic courses of 7 or fewer days, reducing standard antibiotic treatment durations may be an avenue to curtailing antibiotic overuse in long-term care.

Objectives To describe the variability in the duration of antibiotic treatment courses in long-term care across resident recipients and prescribing physicians and to determine whether this variability is influenced by prescriber preference.

Design and Setting Province-wide retrospective analysis of residents of Ontario, Canada, long-term care facilities in 2010.

Participants All adults aged 66 years or older who received an incident treatment course with a systemic antibiotic while residing in an Ontario long-term care facility.

Main Outcome Measure Antibiotic treatment duration was examined across residents and prescribing physicians. The proportion of a physician's treatment courses that exceeded 7 days was used to classify short-, average-, and long-duration prescribers.

Results Of 66 901 long-term care residents from 630 long-term care facilities, 50 061 (77.8%) received an incident antibiotic treatment course (with 51 540 antibiotic courses prescribed). The most commonly selected antibiotic treatment course was 7 days (in 21 136 courses [41.0%]), but 23 124 (44.9%) exceeded 7 days. Among the 699 physicians responsible for 20 or more antibiotic treatment courses, the median (interquartile range) proportion of treatment courses beyond 7 days was 43.5% (26.9%-62.9%) (range, 0%-97.1%). Twenty-one percent of prescribers had a higher-than-expected proportion of prescriptions beyond the 7-day threshold. Patient characteristics were similar across short-, average-, and long-duration prescribers. A mixed logistic model confirmed that prescribers were an important determinant of treatment duration (P < .001), with a relative odds of prolonged prescription of 3.84 for 75th vs 25th percentile prescribers.

Conclusions and Relevance Antibiotic treatment courses in long-term care facilities are often prescribed for long durations, and this appears to be influenced by prescriber preference more than patient characteristics. Future trials should evaluate antibiotic stewardship interventions targeting prescriber preferences to systematically shorten average treatment durations to reduce the complications, costs, and resistance associated with antibiotic overuse.

Figures in this Article

Antibiotics are among the most frequently prescribed medications in long-term care facilities, with 6% to 10% of residents receiving these medications at any given time13 and 60% to 70% receiving at least 1 prescription over the course of a year.4,5 High rates of institutional antibiotic use are driving increased rates of antibiotic resistance, Clostridium difficile infection, antibiotic-related adverse events, and health care costs, yet up to half of antibiotic use in acute and long-term care institutions is unnecessary or inappropriate.57 Providing antimicrobial stewardship to optimize initiation of antimicrobials in the long-term care setting is especially challenging because older residents may have obscuring comorbidities, blunted febrile responses, and predominance of vague systemic symptoms (such as delirium) over localizing symptoms.5,8,9 Diagnosis of infections in long-term care facilities can be further confounded by a lack of on-site diagnostic testing equipment and the fact that many prescriptions are called by telephone order, often by on-call physicians, without a preceding physical examination.3,10

Given that it is frequently difficult to differentiate infected from uninfected patients at the time of antibiotic initiation in long-term care facilities, a potentially more feasible intervention to reduce antimicrobial use would be to shorten treatment duration.11 Indeed, the number one contributor to antibiotic overuse in the acute care setting is excessive length of treatment,12 and this may also be the case in long-term care. Randomized controlled trials have demonstrated that shorter course therapy (≤7 days) is equally effective as longer course treatment for the most common bacterial infections in long-term care residents, including cystitis,13 skin and soft tissue infections,14 and lower respiratory tract infections.1517 Antibiotic treatment that extends beyond clinical cure exposes patients to avoidable adverse events ranging from allergy to C difficile colitis, organ failure, and death. In addition, excessive treatment durations continue to expose the body's normal skin, gastrointestinal, and respiratory tract bacterial flora to selective pressure, thereby promoting the growth of antimicrobial-resistant bacterial populations that are then the potential cause of future difficult-to-treat infections.11,18

If antibiotic treatment duration in long-term care is a function of prescriber preference—as opposed to resident characteristics—then it may provide a target for antimicrobial stewardship interventions, such as standardized order sheets or audit-and-feedback to prescribers, which have been widely successful in acute care institutions.1922 In this population-based study, we sought to describe the variability in the duration of antibiotic treatment courses in long-term care across resident recipients and prescribing physicians and to determine whether this variability is influenced by prescriber preference. This type of fundamental work can serve as the basis for the future development of systems-based antimicrobial stewardship intervention strategies.

STUDY DESIGN AND DATA SOURCES

We conducted a retrospective analysis of antibiotic treatment duration among a cohort of older residents of long-term care facilities. We then examined the extent of variability in the use of prolonged treatment durations (which we defined as >7 days) across physician prescribers in this setting. The study was conducted in Ontario, Canada, via linked population-based administrative databases at the Institute for Clinical Evaluative Sciences (ICES). These well-validated databases have been used extensively in prior research of medication use among older individuals.23,24 These data sets are linked via encrypted health care numbers and include (1) the Ontario Drug Benefit Program database, which contains detailed drug information for Ontario's greater than 1.5 million older adults (and all residents of long-term care facilities); (2) the Continuing Care Resident Reporting System Long-Term Care Database, which is created from the Resident Assessment Instrument Minimum Data Set 2.0, a mandated clinical assessment that must be completed on all residents at quarterly intervals and has been well validated for research purposes25; (3) the Registered Persons database, which contains demographic data for all of Ontario's 12.5 million residents who have ever had a valid health card number in Ontario's universal single-payer health care system; (4) the ICES Physicians Database, which records demographic and practice characteristics for each Ontario physician; (5) the Ontario Health Insurance Plan database, which includes physician billing claims for visits and procedures performed within Ontario; (6) the Canadian Institute for Health Information Discharge abstract database, which details all hospitalization events in the province; and (7) the National Ambulatory Care Reporting System database, which describes all emergency department visits.

RESIDENT SELECTION CRITERIA

The cohort consisted of all older adults (aged ≥66 years) who received an incident treatment course with a systemic antibiotic while residing in an Ontario long-term care facility from January 1 through December 31, 2010. We excluded residents who had received the same antibiotic within the preceding 90 days (to ensure that we were including only incident treatments), and we excluded residents who were discharged from the hospital within the previous 3 days (because their treatment duration may have been determined by hospital prescribers). Finally, we excluded the rare long-term care residents who lacked a Continuing Care Resident Reporting System Long-Term Care assessment during the preceding year, lacked a valid encrypted health card number, had no record in the Registered Persons Database, or had no identified prescriber associated with their antibiotic treatment course.

ANTIBIOTIC USE

Antibiotic prescriptions were ascertained from the Ontario Drug Benefit Program database, which includes information on drug name, dose, route, date of drug claim, days supplied, and prescription location. This database is both comprehensive (the Ontario Drug Benefit Program provides universal coverage of publicly funded medications for all Ontario adults older than 65 years) and accurate (concordance with pharmacy chart review exceeds 99%).26 Although the database records only the dispensing of medication, this should closely reflect actual receipt of medication given that the study population was residing in long-term care facilities where nursing staff could coordinate and administer their treatments. We included all systemic antibiotics administered via the enteral or parenteral route and excluded local and topical antibiotic treatments. Antibiotic treatments were grouped in the following pharmacologic classes (and subclasses): penicillins, cephalosporins (first generation, such as cephalexin, and second or third generation, such as ceftriaxone), fluoroquinolones (second generation, such as norfloxacin and ciprofloxacin, and third generation, such as levofloxacin and moxifloxacin), macrolides, sulfonamides, tetracyclines, lincosamides, nitrofurantoin, metronidazole, glycopeptides, and miscellaneous.1

The duration of each antibiotic treatment was also measured from the Ontario Drug Benefit Program. If there was another claim for the same drug within 3 days of the estimated completion of the prior claim (based on days supplied), then consecutive treatment courses were summed to determine the total duration of therapy.1 Antibiotic durations of 90 days or longer were collapsed into a single category as being indicative of chronic therapy or prophylaxis. As an additional descriptor, antibiotic duration was dichotomized into short course (≤7 days) and longer course (>7 days) treatment.

RESIDENT FACTORS

An extensive number of resident-level factors were collected from 1 year of preceding data from the Continuing Care Resident Reporting System Long-Term Care Database, as well as hospital, emergency department, and physician claims databases.1 The Continuing Care Reporting System Long-Term Care assessments are performed using the Resident Assessment Instrument Minimum Data Set 2.0, which provides a comprehensive, reliable, and internally consistent assessment of the functional status and care needs of long-term care residents.25

Resident factors that could potentially influence antibiotic needs included extensive demographic, health care use, comorbidity, functional status, and device use variables (Table 1).

Table Graphic Jump LocationTable 1. Characteristics of the 50 061 Ontario Long-Term Care Antibiotic Recipientsa
PRESCRIBER FACTORS

In addition, a number of prescriber-level characteristics were captured, including age, sex, Canadian vs foreign medical training, clinical specialty, years in practice, rural residence, and volume of long-term care and non–long-term care physician claims during the study year.

DESCRIPTIVE ANALYSES

Descriptive analyses were performed to examine the characteristics of antibiotic recipients in Ontario long-term care facilities, as well as the types and duration of antibiotics prescribed to these residents. The durations of antibiotic treatment courses were summarized and graphically displayed with stratification by antibiotic subclass. The proportion of antibiotic treatment courses exceeding 7 days was computed overall and for each long-term care prescriber.

FUNNEL PLOTS

After excluding prescribers who generated less than 20 prescriptions during the year, we generated funnel plots to determine whether the interprescriber variation in the use of prolonged treatment courses (>7 days) was greater than that expected by random chance.27 The funnel plot provides a graphical representation of variations in the proportion of antibiotic treatment courses that exceeded 7 days (y-axis) as a function of prescriber volume (x-axis).27 Control limits were generated using exact binomial confidence intervals for the expected proportion of antibiotic treatments exceeding 7 days (standardized to the mean proportion of prolonged prescriptions among the total population of prescribers). To examine the consistency of findings across common infectious syndromes for which short-duration treatment is well established, separate funnel plots were generated for the 2 subgroups of antimicrobial classes that are typically used as urinary anti-infectives (second-generation fluoroquinolones, nitrofurantoin, and trimethoprim and/or sulfonamides) or respiratory anti-infectives (third-generation fluoroquinolones and macrolides).

COMPARISON OF SHORT-, AVERAGE-, AND LONG-DURATION PRESCRIBERS

The funnel plots were used to distinguish average-duration prescribers (whose proportion of prolonged treatments was within the 3-SD control limits) from short- and long-duration prescribers. Short-duration prescribers were defined as those whose proportion of prolonged treatments was below the lower 3-SD control limit; long-duration prescribers were defined as those whose proportion of prolonged treatments was above the upper 3-SD control limit. Prescriber classification as short, average, or long duration was compared for urinary and respiratory anti-infectives (by calculation of percentage agreement). The characteristics of short-, average-, and long-duration prescribers were compared, as were the characteristics of the residents treated by these prescribers.

TEST OF PRESCRIBER EFFECT ON PROLONGED TREATMENT DURATION

Finally, we directly evaluated the hypothesis that individual prescriber practice is a predictor of prolonged treatment duration through a mixed logistic model with the outcome of prolonged treatment duration (>7 days) at the resident level. The model adjusted for all resident-level covariates (listed in the “Resident Factors” subsection) and included a random effect for provider. The likelihood ratio test of the variance component of the provider effect provided the statistical test of hypothesis. A summary measure of the impact of provider was estimated by the relative odds of prolonged antibiotics prescribed by the 75th percentile vs the 25th percentile of prescribers.

All analyses were performed using SAS statistical software, version 9.2 (SAS Institute, Inc). Patient confidentiality was maintained via encrypted health card numbers and strict adherence to privacy protocols of the ICES. The study was approved by the research ethics board of Sunnybrook Health Sciences Centre.

DESCRIPTION OF LONG-TERM CARE RESIDENTS

During the study year, 50 061 of 66 901 long-term care residents (74.8%) received an incident antibiotic treatment course and were included in our study. These residents lived in 630 distinct long-term care facilities, with a median of 120 beds per facility (interquartile range, 77-160) and mostly situated in urban areas (73.2%). As expected, the residents were of advanced age (median age, 86 years), were predominantly women (72.0%), exhibited a high prevalence of dementia (56.7%), and required assistance with most activities of daily living (Table 1).

ANTIBIOTIC TREATMENTS AND DURATIONS AMONG LONG-TERM CARE RESIDENTS

The incident prescriptions included 51 540 individual antimicrobial agents (greater than the number of long-term care residents due to occasional use of combination therapy). The most commonly prescribed antibiotic classes were second-generation fluoroquinolones, penicillins, third-generation fluoroquinolones, first-generation cephalosporins, and sulfonamides (Table 2). Seven days was the most commonly selected treatment duration in 21 136 treatment courses (41.0%), but 23 124 (44.9%) exceeded 7 days and only 7280 (14.1%) were less than 7 days. Prolonged treatment courses were common for all antibiotic subclasses (Figure 1).

Place holder to copy figure label and caption
Graphic Jump Location

Figure 1. Distribution of antibiotic treatment durations. The most common antibiotic treatment duration was 7 days in 21 136 courses (41.0%), but 23 124 (44.9%) exceeded 7 days and only 7277 (14.1%) were less than 7 days. Color segments represent different antibiotic classes and subclasses.

Table Graphic Jump LocationTable 2. Most Frequently Used Antibiotic Classes Among Ontario Long-Term Care Residents
PRESCRIBERS OF ANTIBIOTICS IN LONG-TERM CARE FACILITIES

In total, 2601 different physicians prescribed antibiotics to long-term care residents during the study year, but a smaller number of high-volume prescribers generated the majority of antibiotic treatment prescriptions. Approximately one-fifth of the prescribers (561 [21.6%]) were responsible for four-fifths of the long-term care antibiotic treatment courses (41 007 [79.6%]). Of the 707 unique prescribers responsible for providing at least 20 antibiotic treatments to long-term care residents, 699 could be linked to the physician database. These prescribers were mostly men (579 [82.8%]) and family/general practitioners (683 [97.7%]) with a median (interquartile range) of 31 (24-39) years of practice experience. Across these long-term care prescribers, the median (interquartile range) proportion of antibiotic treatment courses that exceeded 7 days was 43.5% (26.9%-62.9%); the range was 0% to 97.1%.

FUNNEL PLOTS

The funnel plot in Figure 2 displays the proportion of antibiotic treatment courses that exceeded 7 days for each individual prescriber as a function of the total number of antibiotic treatments by that prescriber. If treatment duration were randomly distributed, 99.8% of prescribers would be within 3-SD control limits and would be classified as average-duration prescribers. However, only 402 long-term care prescribers (57.5%) were within this range. There were large numbers of high outliers (145 [20.7%]) who were classified as long-duration prescribers; similarly, there were large numbers of low outliers (152 [21.7%]) who were classified as short-duration prescribers.

Place holder to copy figure label and caption
Graphic Jump Location

Figure 2. Funnel plot to determine whether variability in average treatment durations by individual prescribers is greater than can be expected by random chance. The CIs for the funnel plot are generated using exact binomial CIs for the expected proportion of treatments exceeding 7 days (standardized to the population average). Each dot indicates 1 of the 699 prescribers responsible for more than 20 individual antibiotic treatments. There were more long-duration outlier prescribers above 3-SD CIs (black dots) and short-duration outlier prescribers below 3-SD CIs (gray dots) than expected by random chance.

Funnel plots for the subgroup of urinary anti-infectives (Figure 3A) and respiratory anti-infectives (Figure 3B) also demonstrated a greater number of short- and long-duration outlier prescribers than could be expected by random chance. There was 62% agreement in the classification of individual physicians as short-, average-, or long-duration prescribers for urinary and respiratory antibiotics (higher than the expected agreement of 44% if there was no association of physician treatment duration preference across these 2 types of antibiotics).

Place holder to copy figure label and caption
Graphic Jump Location

Figure 3. Funnel plots for the subgroup of urinary anti-infectives (A) and respiratory anti-infectives (B). A, Funnel plot for subgroup of most common urinary anti-infectives (ciprofloxacin, trimethoprim-sulfamethoxazole, and nitrofurantoin). Each dot indicates 1 of the 306 prescribers responsible for more than 20 individual urinary anti-infective treatments. There were more long-duration outlier prescribers above 3-SD CIs (black dots) and short-duration outlier prescribers below 3-SD CIs (gray dots) than expected by random chance. B, Funnel plot for subgroup of most common respiratory anti-infectives (levofloxacin, moxifloxacin, clarithromycin, and azithromycin). Each dot indicates 1 of the 190 prescribers responsible for more than 20 individual respiratory anti-infective treatments. There were more long-duration outlier prescribers above 3-SD CIs (black dots) and short-duration outlier prescribers below 3-SD CIs (gray dots) than expected by random chance.

MOST COMMON TREATMENT DURATIONS BY PRESCRIBER CATEGORY

Long-duration prescribers compared with average- and short-duration prescribers were less likely to select treatment durations of 5 days (2% vs 8% vs 12%) or 7 days (22% vs 42% vs 59%) (P < .001 for both). Conversely, longer-duration prescribers were more likely to select treatment durations of 10 days (23% vs 22% vs 10%) or 14 days (32% vs 13% vs 5%) (P < .001 for both).

COMPARISON OF SHORT-, AVERAGE-, AND LONG-DURATION PRESCRIBERS

Short-, average-, and long-duration prescribers exhibited similar demographic characteristics, including age, sex, place of graduation, clinical specialty, and years in practice (Table 3). The characteristics of residents treated by short-, average-, and long-duration prescribers were also similar, suggesting that prescribing tendencies were not driven by differences in patient demographic characteristics, comorbidities, or care needs (Table 4).

Table Graphic Jump LocationTable 3. Characteristics of Short-, Average-, and Long-Duration Prescribersa
Table Graphic Jump LocationTable 4. Characteristics of Residents Treated by Short-, Average-, and Long-Duration Prescribersa
PERCENTAGE OF ANTIBIOTIC DAYS ATTRIBUTABLE TO PRESCRIBER PREFERENCE

The average treatment durations used by long-, average-, and short-duration prescribers were 11.6, 9.1, and 7.5 days, respectively. If long-duration prescribers adopted the prescribing profile of average prescribers, their total antibiotic days prescribed would decrease by 22% and the overall antibiotic days in long-term care would decrease by 7% (28 411 of 410 690). If long- and average-duration prescribers both adopted the prescribing profile of short prescribers, their total antibiotic days would decrease by 35% and 17%, respectively; the overall antibiotic days in long-term care would decrease by 19% (78 235 of 410 690).

TEST OF PRESCRIBER EFFECT ON PROLONGED TREATMENT DURATION

In a mixed logistic model accounting for all measured resident-level factors, the prescriber was significantly associated with the likelihood of a resident receiving a prolonged treatment duration beyond 7 days (P < .001). The relative odds of a prolonged treatment duration was 3.84 for a 75th percentile prescriber compared with a 25th percentile prescriber.

Although most common infectious diseases can be treated effectively with shorter courses of antibiotics, our study demonstrates that nearly half of long-term care antibiotic treatment courses extend beyond 1 week. The use of prolonged treatment courses is common for all antibiotic classes, including those often used for indications with the strongest evidence supporting shorter-duration treatment (such as urinary and respiratory tract infections). Prolonged antibiotic treatment durations appear to be influenced by individual prescriber preference, with more long-duration outlier prescribers than can be explained by random chance or variation in patient characteristics and with 75th percentile prescribers approximately 4 times more likely to prescribe prolonged treatment durations than the 25th percentile prescribers.

Our study findings are in keeping with prior literature28,29 that suggests that physicians develop consistent prescribing patterns that may be independent of patient treatment needs. For example, when equivalent generic and trade name medications are available, the likelihood of prescribing the more expensive trade name drug is a matter of prescriber habit.28 With respect to antibiotic medications in particular, the influence of prescribing behavior on the decision to initiate antibiotics has been studied among 722 patients with nasal or sinus symptoms presenting to 80 general practitioners in Belgium.29 Although some patient factors (such as physical findings and diagnostic labels) were predictive of antibiotic prescription, the strongest independent predictor of an antibiotic prescription was the individual antibiotic prescribing rate, a measure of the physician's tendency to prescribe antibiotics.29 This individual prescribing rate was not linked to specific physician demographic characteristics. Similarly, our study indicates that physician preference influences the decision of how long to continue such antibiotic treatment courses once they are started.

Given that antibiotic courses of 7 days or less are as effective as longer treatment duration for the majority of common bacterial infections13-17 and that selection of treatment duration appears to be influenced by prescriber preference, efforts to alter these behaviors may offer a reasonable avenue to reduce antibiotic use in long-term care facilities. Interventions such as standard order sets for antibiotics30,31 or audit-and-feedback for excessive antibiotic treatment durations21,22 could potentially lead to large reductions in antimicrobial use and should be the focus of future research. These interventions need not be targeted specifically at long-duration outlier prescribers because the median proportion of prolonged treatments was 44% across the entire population of long-term care prescribers.

Our study has important limitations inherent to the use of administrative databases, but the availability of an accurate, comprehensive, and independent drug database has minimized the likelihood of misclassifying antibiotic use or duration.1,26 The low prevalence of physician diagnostic claims for infection among long-term care residents receiving antibiotic treatments implies that we cannot definitively determine which residents may have rare indications for more prolonged treatment (such as osteomyelitis or endocarditis). However, extensive prior work has shown that urinary tract, respiratory tract, and skin and soft tissue infections account for the majority of bacterial infections in these facilities4,5,10; moreover, our subgroup analyses for urinary and respiratory antibiotics corroborated our main findings. Our data sources do not capture prescribers' rationales for treatment duration selection, but prospective data collection would inevitably introduce Hawthorne effects on these behaviors. Further work in this area should incorporate historical prescribing patterns of physicians to characterize preferences more fully. We believe that these study limitations are outweighed by study strengths, including the use of population-based data (with no selection bias from voluntarily participating physicians) derived from well-validated data sets that are rich in clinical details, across a jurisdiction with a large sample size.

Antibiotic treatment courses in long-term care facilities were prescribed for long durations and appeared to be influenced by prescriber preference more so than resident characteristics. Future trials should evaluate interventions to systematically reduce average treatment durations and thereby reduce the costs, complications, and resistance associated with antibiotic overuse in these facilities.

Correspondence: Nick Daneman, MD, MSc, Division of Infectious Diseases, Department of Medicine, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Bayview Ave, Toronto, ON M4N 2M5, Canada (nick.daneman@sunnybrook.ca).

Accepted for Publication: November 2, 2012.

Published Online: March 18, 2013. doi:10.1001/jamainternmed.2013.3029

Author Contributions: Drs Daneman and Bell had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Daneman, Gruneir, Bronskill, Fischer, Rochon, Anderson, and Bell. Analysis and interpretation of data: Daneman, Gruneir, Bronskill, Newman, Anderson, and Bell. Drafting of the manuscript: Daneman, Fischer, Rochon, and Bell. Critical revision of the manuscript for important intellectual content: Daneman, Gruneir, Bronskill, Newman, Rochon, Anderson, and Bell. Statistical analysis: Daneman, Gruneir, Bronskill, and Newman. Obtained funding: Rochon and Anderson. Administrative, technical, and material support: Gruneir and Fischer. Study supervision: Anderson and Bell.

Conflict of Interest Disclosures: Dr Fischer reports that she worked at Bayer Inc Canada in 2003 and 2004.

Funding/Support: This study was conducted at ICES, which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. This work was supported by an Interdisciplinary Capacity Enhancement Grant (HOA-80075) from the Canadian Institutes of Health Research (CIHR) Institute of Gender and Health and the CIHR Institute of Aging and by a CIHR Team Grant (OTG-88591) from the CIHR Institute of Nutrition, Metabolism, and Diabetes; a CIHR clinician scientist award (Dr Daneman); a CIHR and Canadian Patient Safety Institute chair in Patient Safety and Continuity of Care (Dr Bell); salary support from the Team Grant (OTG-88591) from the CIHR Institute of Nutrition, Metabolism, and Diabetes (Dr Gruneir); and a CIHR New Investigator Award in the Area of Aging (Dr Bronskill).

Role of the Sponsors: No endorsement by ICES or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred.

Disclaimer: The opinions, results, and conclusions reported in this article are those of the authors and are independent from the funding sources.

Daneman N, Gruneir A, Newman A,  et al.  Antibiotic use in long-term care facilities.  J Antimicrob Chemother. 2011;66(12):2856-2863
PubMed   |  Link to Article
Pakyz AL, Dwyer LL. Prevalence of antimicrobial use among United States nursing home residents: results from a national survey.  Infect Control Hosp Epidemiol. 2010;31(6):661-662
PubMed   |  Link to Article
Warren JW, Palumbo FB, Fitterman L, Speedie SM. Incidence and characteristics of antibiotic use in aged nursing home patients.  J Am Geriatr Soc. 1991;39(10):963-972
PubMed
D’Agata E, Mitchell SL. Patterns of antimicrobial use among nursing home residents with advanced dementia.  Arch Intern Med. 2008;168(4):357-362
PubMed   |  Link to Article
Nicolle LE, Bentley DW, Garibaldi R, Neuhaus EG, Smith PW.SHEA Long-Term Care Committee.  Antimicrobial use in long-term care facilities.  Infect Control Hosp Epidemiol. 2000;21(8):537-545
PubMed   |  Link to Article
Rotjanapan P, Dosa D, Thomas KS. Potentially inappropriate treatment of urinary tract infections in two Rhode Island nursing homes.  Arch Intern Med. 2011;171(5):438-443
PubMed   |  Link to Article
Hughes JM. Preserving the lifesaving power of antimicrobial agents.  JAMA. 2011;305(10):1027-1028
PubMed   |  Link to Article
McGeer A, Campbell B, Emori TG,  et al.  Definitions of infection for surveillance in long-term care facilities.  Am J Infect Control. 1991;19(1):1-7
PubMed   |  Link to Article
Castle SC, Norman DC, Yeh M, Miller D, Yoshikawa TT. Fever response in elderly nursing home residents: are the older truly colder?  J Am Geriatr Soc. 1991;39(9):853-857
PubMed
Katz PR, Beam TR Jr, Brand F, Boyce K. Antibiotic use in the nursing home: physician practice patterns.  Arch Intern Med. 1990;150(7):1465-1468
PubMed   |  Link to Article
Rice LB. The Maxwell Finland Lecture: for the duration-rational antibiotic administration in an era of antimicrobial resistance and Clostridium difficile Clin Infect Dis. 2008;46(4):491-496
PubMed   |  Link to Article
Hecker MT, Aron DC, Patel NP, Lehmann MK, Donskey CJ. Unnecessary use of antimicrobials in hospitalized patients: current patterns of misuse with an emphasis on the antianaerobic spectrum of activity.  Arch Intern Med. 2003;163(8):972-978
PubMed   |  Link to Article
Lutters M, Vogt-Ferrier NB. Antibiotic duration for treating uncomplicated, symptomatic lower urinary tract infections in elderly women.  Cochrane Database Syst Rev. 2008;(3):CD001535
PubMed
Hepburn MJ, Dooley DP, Skidmore PJ, Ellis MW, Starnes WF, Hasewinkle WC. Comparison of short-course (5 days) and standard (10 days) treatment for uncomplicated cellulitis.  Arch Intern Med. 2004;164(15):1669-1674
PubMed   |  Link to Article
Dimopoulos G, Matthaiou DK, Karageorgopoulos DE, Grammatikos AP, Athanassa Z, Falagas ME. Short- versus long-course antibacterial therapy for community-acquired pneumonia: a meta-analysis.  Drugs. 2008;68(13):1841-1854
PubMed   |  Link to Article
Rafailidis PI, Pitsounis AI, Falagas ME. Meta-analyses on the optimization of the duration of antimicrobial treatment for various infections.  Infect Dis Clin North Am. 2009;23(2):269-276
PubMed   |  Link to Article
el Moussaoui R, de Borgie CA, van den Broek P,  et al.  Effectiveness of discontinuing antibiotic treatment after three days versus eight days in mild to moderate-severe community-acquired pneumonia: randomised, double blind study.  BMJ. 2006;332(7554):1355
PubMed  |  Link to Article   |  Link to Article
Eckburg PB, Bik EM, Bernstein CN,  et al.  Diversity of the human intestinal microbial flora.  Science. 2005;308(5728):1635-1638
PubMed   |  Link to Article
Dellit TH, Owens RC, McGowan JE Jr,  et al; Infectious Diseases Society of America; Society for Healthcare Epidemiology of America.  Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship.  Clin Infect Dis. 2007;44(2):159-177
PubMed   |  Link to Article
Kaki R, Elligsen M, Walker S, Simor A, Palmay L, Daneman N. Impact of antimicrobial stewardship in critical care: a systematic review.  J Antimicrob Chemother. 2011;66(6):1223-1230
PubMed   |  Link to Article
Elligsen M, Walker SA, Pinto R,  et al.  Audit and feedback to reduce broad-spectrum antibiotic use among intensive care unit patients: a controlled interrupted time series analysis.  Infect Control Hosp Epidemiol. 2012;33(4):354-361
PubMed   |  Link to Article
Solomon DH, Van Houten L, Glynn RJ,  et al.  Academic detailing to improve use of broad-spectrum antibiotics at an academic medical center.  Arch Intern Med. 2001;161(15):1897-1902
PubMed   |  Link to Article
Rochon PA, Tu JV, Anderson GM,  et al.  Rate of heart failure and 1-year survival for older people receiving low-dose beta-blocker therapy after myocardial infarction.  Lancet. 2000;356(9230):639-644
PubMed   |  Link to Article
Rochon PA, Stukel TA, Bronskill SE,  et al.  Variation in nursing home antipsychotic prescribing rates.  Arch Intern Med. 2007;167(7):676-683
PubMed   |  Link to Article
Mor V. A comprehensive clinical assessment tool to inform policy and practice: applications of the minimum data set.  Med Care. 2004;42(4):(suppl)  III50-III59
PubMed
Levy AR, O’Brien BJ, Sellors C, Grootendorst P, Willison D. Coding accuracy of administrative drug claims in the Ontario Drug Benefit database.  Can J Clin Pharmacol. 2003;10(2):67-71
PubMed
Spiegelhalter DJ. Funnel plots for comparing institutional performance.  Stat Med. 2005;24(8):1185-1202
PubMed   |  Link to Article
Hellerstein JK. The importance of the physician in the generic versus trade-name prescription decision.  Rand J Econ. 1998;29(1):108-136
PubMed   |  Link to Article
De Sutter AI, De Meyere MJ, De Maeseneer JM, Peersman WP. Antibiotic prescribing in acute infections of the nose or sinuses: a matter of personal habit?  Fam Pract. 2001;18(2):209-213
PubMed   |  Link to Article
Durbin WA Jr, Lapidas B, Goldmann DA. Improved antibiotic usage following introduction of a novel prescription system.  JAMA. 1981;246(16):1796-1800
PubMed   |  Link to Article
Echols RM, Kowalsky SF. The use of an antibiotic order form for antibiotic utilization review: influence on physicians' prescribing patterns.  J Infect Dis. 1984;150(6):803-807
PubMed   |  Link to Article

Figures

Place holder to copy figure label and caption
Graphic Jump Location

Figure 1. Distribution of antibiotic treatment durations. The most common antibiotic treatment duration was 7 days in 21 136 courses (41.0%), but 23 124 (44.9%) exceeded 7 days and only 7277 (14.1%) were less than 7 days. Color segments represent different antibiotic classes and subclasses.

Place holder to copy figure label and caption
Graphic Jump Location

Figure 2. Funnel plot to determine whether variability in average treatment durations by individual prescribers is greater than can be expected by random chance. The CIs for the funnel plot are generated using exact binomial CIs for the expected proportion of treatments exceeding 7 days (standardized to the population average). Each dot indicates 1 of the 699 prescribers responsible for more than 20 individual antibiotic treatments. There were more long-duration outlier prescribers above 3-SD CIs (black dots) and short-duration outlier prescribers below 3-SD CIs (gray dots) than expected by random chance.

Place holder to copy figure label and caption
Graphic Jump Location

Figure 3. Funnel plots for the subgroup of urinary anti-infectives (A) and respiratory anti-infectives (B). A, Funnel plot for subgroup of most common urinary anti-infectives (ciprofloxacin, trimethoprim-sulfamethoxazole, and nitrofurantoin). Each dot indicates 1 of the 306 prescribers responsible for more than 20 individual urinary anti-infective treatments. There were more long-duration outlier prescribers above 3-SD CIs (black dots) and short-duration outlier prescribers below 3-SD CIs (gray dots) than expected by random chance. B, Funnel plot for subgroup of most common respiratory anti-infectives (levofloxacin, moxifloxacin, clarithromycin, and azithromycin). Each dot indicates 1 of the 190 prescribers responsible for more than 20 individual respiratory anti-infective treatments. There were more long-duration outlier prescribers above 3-SD CIs (black dots) and short-duration outlier prescribers below 3-SD CIs (gray dots) than expected by random chance.

Tables

Table Graphic Jump LocationTable 1. Characteristics of the 50 061 Ontario Long-Term Care Antibiotic Recipientsa
Table Graphic Jump LocationTable 2. Most Frequently Used Antibiotic Classes Among Ontario Long-Term Care Residents
Table Graphic Jump LocationTable 3. Characteristics of Short-, Average-, and Long-Duration Prescribersa
Table Graphic Jump LocationTable 4. Characteristics of Residents Treated by Short-, Average-, and Long-Duration Prescribersa

References

Daneman N, Gruneir A, Newman A,  et al.  Antibiotic use in long-term care facilities.  J Antimicrob Chemother. 2011;66(12):2856-2863
PubMed   |  Link to Article
Pakyz AL, Dwyer LL. Prevalence of antimicrobial use among United States nursing home residents: results from a national survey.  Infect Control Hosp Epidemiol. 2010;31(6):661-662
PubMed   |  Link to Article
Warren JW, Palumbo FB, Fitterman L, Speedie SM. Incidence and characteristics of antibiotic use in aged nursing home patients.  J Am Geriatr Soc. 1991;39(10):963-972
PubMed
D’Agata E, Mitchell SL. Patterns of antimicrobial use among nursing home residents with advanced dementia.  Arch Intern Med. 2008;168(4):357-362
PubMed   |  Link to Article
Nicolle LE, Bentley DW, Garibaldi R, Neuhaus EG, Smith PW.SHEA Long-Term Care Committee.  Antimicrobial use in long-term care facilities.  Infect Control Hosp Epidemiol. 2000;21(8):537-545
PubMed   |  Link to Article
Rotjanapan P, Dosa D, Thomas KS. Potentially inappropriate treatment of urinary tract infections in two Rhode Island nursing homes.  Arch Intern Med. 2011;171(5):438-443
PubMed   |  Link to Article
Hughes JM. Preserving the lifesaving power of antimicrobial agents.  JAMA. 2011;305(10):1027-1028
PubMed   |  Link to Article
McGeer A, Campbell B, Emori TG,  et al.  Definitions of infection for surveillance in long-term care facilities.  Am J Infect Control. 1991;19(1):1-7
PubMed   |  Link to Article
Castle SC, Norman DC, Yeh M, Miller D, Yoshikawa TT. Fever response in elderly nursing home residents: are the older truly colder?  J Am Geriatr Soc. 1991;39(9):853-857
PubMed
Katz PR, Beam TR Jr, Brand F, Boyce K. Antibiotic use in the nursing home: physician practice patterns.  Arch Intern Med. 1990;150(7):1465-1468
PubMed   |  Link to Article
Rice LB. The Maxwell Finland Lecture: for the duration-rational antibiotic administration in an era of antimicrobial resistance and Clostridium difficile Clin Infect Dis. 2008;46(4):491-496
PubMed   |  Link to Article
Hecker MT, Aron DC, Patel NP, Lehmann MK, Donskey CJ. Unnecessary use of antimicrobials in hospitalized patients: current patterns of misuse with an emphasis on the antianaerobic spectrum of activity.  Arch Intern Med. 2003;163(8):972-978
PubMed   |  Link to Article
Lutters M, Vogt-Ferrier NB. Antibiotic duration for treating uncomplicated, symptomatic lower urinary tract infections in elderly women.  Cochrane Database Syst Rev. 2008;(3):CD001535
PubMed
Hepburn MJ, Dooley DP, Skidmore PJ, Ellis MW, Starnes WF, Hasewinkle WC. Comparison of short-course (5 days) and standard (10 days) treatment for uncomplicated cellulitis.  Arch Intern Med. 2004;164(15):1669-1674
PubMed   |  Link to Article
Dimopoulos G, Matthaiou DK, Karageorgopoulos DE, Grammatikos AP, Athanassa Z, Falagas ME. Short- versus long-course antibacterial therapy for community-acquired pneumonia: a meta-analysis.  Drugs. 2008;68(13):1841-1854
PubMed   |  Link to Article
Rafailidis PI, Pitsounis AI, Falagas ME. Meta-analyses on the optimization of the duration of antimicrobial treatment for various infections.  Infect Dis Clin North Am. 2009;23(2):269-276
PubMed   |  Link to Article
el Moussaoui R, de Borgie CA, van den Broek P,  et al.  Effectiveness of discontinuing antibiotic treatment after three days versus eight days in mild to moderate-severe community-acquired pneumonia: randomised, double blind study.  BMJ. 2006;332(7554):1355
PubMed  |  Link to Article   |  Link to Article
Eckburg PB, Bik EM, Bernstein CN,  et al.  Diversity of the human intestinal microbial flora.  Science. 2005;308(5728):1635-1638
PubMed   |  Link to Article
Dellit TH, Owens RC, McGowan JE Jr,  et al; Infectious Diseases Society of America; Society for Healthcare Epidemiology of America.  Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship.  Clin Infect Dis. 2007;44(2):159-177
PubMed   |  Link to Article
Kaki R, Elligsen M, Walker S, Simor A, Palmay L, Daneman N. Impact of antimicrobial stewardship in critical care: a systematic review.  J Antimicrob Chemother. 2011;66(6):1223-1230
PubMed   |  Link to Article
Elligsen M, Walker SA, Pinto R,  et al.  Audit and feedback to reduce broad-spectrum antibiotic use among intensive care unit patients: a controlled interrupted time series analysis.  Infect Control Hosp Epidemiol. 2012;33(4):354-361
PubMed   |  Link to Article
Solomon DH, Van Houten L, Glynn RJ,  et al.  Academic detailing to improve use of broad-spectrum antibiotics at an academic medical center.  Arch Intern Med. 2001;161(15):1897-1902
PubMed   |  Link to Article
Rochon PA, Tu JV, Anderson GM,  et al.  Rate of heart failure and 1-year survival for older people receiving low-dose beta-blocker therapy after myocardial infarction.  Lancet. 2000;356(9230):639-644
PubMed   |  Link to Article
Rochon PA, Stukel TA, Bronskill SE,  et al.  Variation in nursing home antipsychotic prescribing rates.  Arch Intern Med. 2007;167(7):676-683
PubMed   |  Link to Article
Mor V. A comprehensive clinical assessment tool to inform policy and practice: applications of the minimum data set.  Med Care. 2004;42(4):(suppl)  III50-III59
PubMed
Levy AR, O’Brien BJ, Sellors C, Grootendorst P, Willison D. Coding accuracy of administrative drug claims in the Ontario Drug Benefit database.  Can J Clin Pharmacol. 2003;10(2):67-71
PubMed
Spiegelhalter DJ. Funnel plots for comparing institutional performance.  Stat Med. 2005;24(8):1185-1202
PubMed   |  Link to Article
Hellerstein JK. The importance of the physician in the generic versus trade-name prescription decision.  Rand J Econ. 1998;29(1):108-136
PubMed   |  Link to Article
De Sutter AI, De Meyere MJ, De Maeseneer JM, Peersman WP. Antibiotic prescribing in acute infections of the nose or sinuses: a matter of personal habit?  Fam Pract. 2001;18(2):209-213
PubMed   |  Link to Article
Durbin WA Jr, Lapidas B, Goldmann DA. Improved antibiotic usage following introduction of a novel prescription system.  JAMA. 1981;246(16):1796-1800
PubMed   |  Link to Article
Echols RM, Kowalsky SF. The use of an antibiotic order form for antibiotic utilization review: influence on physicians' prescribing patterns.  J Infect Dis. 1984;150(6):803-807
PubMed   |  Link to Article

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 4

Related Content

Customize your page view by dragging & repositioning the boxes below.

See Also...
Articles Related By Topic
Related Collections
PubMed Articles
Antibiotic prophylaxis in COPD: Why, when, and for whom? Pulm Pharmacol Ther Published online May 11, 2014.;
Antioxidant supplementation for lung disease in cystic fibrosis. Cochrane Database Syst Rev Published online Aug 7, 2014.;
JAMAevidence.com

Users' Guides to the Medical Literature
Clinical Resolution

Users' Guides to the Medical Literature
Clinical Scenario