Informing antimicrobial stewardship: factors associated with inappropriate antimicrobial prescribing in primary care

Informing antimicrobial stewardship: factors associated with inappropriate antimicrobial... Abstract Background Antimicrobial stewardship (AS) programs promote the optimal use of antimicrobials and safe patient care. With most antimicrobials prescribed in the ambulatory setting, establishing benchmark data is imperative to gauge the impact of future AS initiatives. Objectives To determine the frequency of potentially inappropriate antimicrobial prescribing in primary care practices in Manitoba, Canada and to assess the association between potentially inappropriate antimicrobial prescribing and patient, prescriber and practice-related factors. Methods A retrospective cohort study using the Manitoba Primary Care Research Network repository of de-identified Electronic Medical Records from consenting primary care practices. Descriptive statistics and logistic regressions detailed patients with bacterial or viral infections of interest and antimicrobial prescriptions. Results Eighteen percent (n = 35 574) of primary care visits for common infections were associated with a potentially inappropriate antimicrobial prescription. Among antimicrobials prescribed to patients diagnosed with bacterial infections, 37.8% (n = 2168) had a potentially inappropriate antimicrobial prescribed and 19.6% (n = 1126) had an antimicrobial prescribed for a duration outside of guideline-based ranges. Female patients, younger age and less office visits were associated with potentially inappropriate antimicrobial prescribing for bacterial infections. Among physician visits for viral infection, 15.9% (n = 29 833) were associated with an antimicrobial prescription. Older patients, those with more comorbidity, more office visits and those who were seen in larger or rural practices, were associated with potentially inappropriate antimicrobial prescribing for viral infections. Conclusions High frequency of potentially inappropriate antimicrobial prescribing, especially in certain patient populations, suggests the need for coordinated community-based AS programs to optimize prescribing and improve patient care. Antibiotic, bacterial, bacterial infections, drug resistance, health care costs, primary health care, viral disease Background The emergence and spread of antimicrobial-resistant organisms highlights the need to preserve the efficacy of antimicrobials (1,2). Antimicrobial stewardship (AS) programs combine policies, education, surveillance and practice audit to optimize antimicrobial prescribing enhancing clinical outcomes, reducing sequelae and ensuring cost-effective therapy (1,2). AS programs reduced treatment duration, the incidences of Clostridium difficile colitis, antimicrobial-related adverse events, hospitalization duration and antimicrobial resistance (2,3). Development of effective AS strategies in primary care requires benchmarks for intervention evaluation. Most antimicrobial prescribing occurs in the ambulatory and outpatient settings. Antimicrobial prescribing occurs among 10% of adult and 20% of paediatric ambulatory visits, and anywhere from 25 to 68% of prescriptions are potentially inappropriate (not indicated, suboptimal antimicrobial spectrum) (3,4). It is crucial to understand rates of potentially inappropriate antimicrobial prescribing in community practices to develop and tailor suitable AS interventions. This study had two specific objectives aimed at informing community-based AS policies. The primary objective was to determine the frequency of potentially inappropriate antimicrobial prescribing (not indicated, suboptimal) in primary care practices in Manitoba, Canada. The secondary objective was to determine patient, prescriber and practice-related factors associated with potentially inappropriate (not indicated, suboptimal) antimicrobial prescribing. Methods This retrospective cohort study was conducted using the Manitoba Primary Care Research Network (MaPCReN) repository that collects and processes de-identified Electronic Medical Record (EMR) from consenting community-based primary care practices. At the time of the study, the MaPCReN contained administrative claims information extracted from 32 primary care clinics, 185 providers and 196 923 patients in Manitoba. All prescriptions generated in the EMR are captured by MaPCReN. In the medical office, prescriptions are printed or faxed to the patient’s pharmacy. Inclusion/exclusion criteria The Anatomical Therapeutic Chemical Classification System (ATC) divides drugs into different groups based on the organ and system in which they act, and their chemical, pharmacological and therapeutic properties (5). For the present study, prescription data were extracted from April 1999 to January 2016 for all antimicrobials starting with ATC code J01, ‘Antibacterial for Systemic Use’, in patients older than 1 year. All visits to a medical provider in Manitoba must include a tariff code and a single International Classification of Disease (ICD-9) in order for payment. In this study, diagnoses were matched to antimicrobial prescriptions within 7 days of the appointment (prior to, or after). If there were two visits within 7 days of the prescription, the one nearest to the antimicrobial prescription was considered. Prescription information was only included if it specified the antimicrobial prescribed and prescription duration. This method of identifying diagnosis and prescription data from administrative claims data has been validated within Manitoba, Canada and international studies (6–8). Diagnoses were separated into two groups: ‘Group 1’ consisted of bacterial infections (i.e. urinary tract infection (UTI), pharyngitis, skin/soft tissue infection, cellulitis and pneumonia), for which an antimicrobial prescription was likely indicated. Prescriptions between 30 June 2013 and 30 June 2015, linked to a Group 1 diagnosis, were deemed potentially inappropriate if they fell outside of the following evidence-informed guidelines: Infectious Diseases Society of America, Canadian Pediatric Society and American Academy of Pediatrics’ guideline-based standards (9–15). We included only recent antimicrobial prescriptions for bacterial infections (Group 1) to reflect current guidelines published between 2011 and 2014, as prior recommendations may have been different. These guidelines were chosen as they represent the ideal standard of care in Manitoba, Canada. When there was not uniform consensus between various North American guidelines, we considered a range of acceptable number of days as ‘appropriate’. A complete list of the Group 1 prescriptions was evaluated; what was considered ‘appropriate’ antimicrobial prescriptions and the range of acceptable duration can be found in Supplementary Table S1. ‘Group 2’ consisted of encounters with a diagnosis of a viral infection (i.e. acute sinusitis, acute laryngitis and trachealis, upper respiratory tract infection, bronchitis, acute rhinitis, nasopharyngitis and influenza) that typically do not require an antimicrobial (Table 1). Any prescription between April 1999 and January 2016, written within 7 days of a diagnosed Group 2 viral infection was considered potentially inappropriate. To account for antimicrobial treatment in response to a potentially coexisting bacterial infection, patients were excluded if they had a previous diagnosis of bronchiectasis, cystic fibrosis, chronic obstructive pulmonary disease or acute sinusitis (within 21 days prior to the encounter). Table 1. Number of potentially inappropriate antimicrobial prescriptions for bacterial infections in 3294 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) View Large Table 1. Number of potentially inappropriate antimicrobial prescriptions for bacterial infections in 3294 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) View Large Statistical analysis Descriptive analyses based on patient, prescriber and practice factors associated with each prescription were performed. To identify factors that were statistically significantly associated with potentially inappropriate antimicrobial prescribing, a multilevel regression model was used. The unit of measure was the patient, with patient (i.e. comorbidity, age, gender, number of primary care office visits per year and socio-economic status), provider (i.e. country of graduation and age) and practice (i.e. size and location) covariates used to identify factors associated with an increased risk of potentially inappropriate prescriptions. Multivariable generalized linear mixed models were used to allow for correlation between patients within the same health care provider. Odds ratios and their confidence intervals are reported. SAS version 9.3 (SAS Institute, Cary, NC) was used for analyses. Results Between April 1999 and January 2016, the MaPCReN repository contained 192,601 primary care appointments for infections; 18.5% (n = 35 574) had an associated prescription for an antimicrobial. Of the antimicrobial prescriptions, 16.1% (n = 5741) were associated with a bacterial infection (Group 1) and 83.9% (n = 29 833) antimicrobial prescriptions were associated with a viral infection (Group 2). Female patients were more likely to have an appointment for an infection than male patients (62.0% versus 37.9%, respectively). Primary care appointments for an infection were for patients aged 1 to >89, with 48% (n = 89 871) of the appointments for adult patients aged 31–69. Patients younger than 20 years represented 13.9% (n = 25 903) of the primary care appointments. Young adults (aged 20–30) and seniors (70 or older) represented 13.2% and 24.8% of the primary care appointments for an infection, respectively. Group 1: Bacterial infections The MaPCReN repository provided a sample of 5741 antimicrobial prescriptions for Group 1. Within the Group 1 prescriptions, 37.8% (n = 2168) were potentially prescribed inappropriately and 19.6% (n = 1126) were prescribed for a duration outside of guideline-based ranges. Potentially inappropriate antimicrobial use was most frequent with macrolides and beta-lactams (Table 1). Macrolides made up the largest group of potentially inappropriately prescribed antimicrobials (50.4%, n = 1662). UTIs were the most frequently diagnosed condition to contain a potentially inappropriately prescribed macrolide (75.4%, n = 1367). Nitrofuran/imidazole derivatives (11.1%, n = 366) and fluoroquinolones (9.1%, n = 301) were the most likely antimicrobials to be prescribed for durations outside of guideline-based ranges (5–11). The majority of prescriptions in Group 1 were associated with a diagnosis of UTI (75.1%); of these 59.7% were potentially inappropriate (Table 2). An antimicrobial prescription for a UTI was more likely among female patients (92.7%, n = 3998) than male patients (7.4%, n = 317). A majority of these prescriptions were potentially inappropriate (59.7%, n = 2576). Forty percent (n = 1730) of antimicrobial prescriptions for UTI were written for an antimicrobial not considered a first-line treatment (Table 3). While the greatest number of antimicrobial prescriptions were for UTIs, the highest rate of potentially inappropriate antimicrobial prescribing in Group 1 was for pneumonia. When an incorrect duration was prescribed for pneumonia (n = 191), it was significantly more likely to have been prescribed for a shorter duration than guideline-based ranges. There were 102 one-day prescriptions (60.7%) compared to 8 (0.0004%) one-day prescriptions for UTI. The default prescription length in the EMRs used by providers in this study is 1 day. When 1-day prescriptions were removed, 20.9% (n = 66) of potentially inappropriate pneumonia prescriptions appeared to be too short. Macrolide prescriptions for pneumonia were frequently prescribed a shorter duration than guideline ranges (66.5%, n = 153). Table 2. Appropriate antimicrobial prescribing for diagnosed bacterial infections in 5745 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) View Large Table 2. Appropriate antimicrobial prescribing for diagnosed bacterial infections in 5745 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) View Large Table 3. Multivariate mixed-effects logistic regression model of bacterial infection prescriptions not guidelines based in 5746 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Bold numbers are significant at a P value of 0.05. View Large Table 3. Multivariate mixed-effects logistic regression model of bacterial infection prescriptions not guidelines based in 5746 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Bold numbers are significant at a P value of 0.05. View Large The majority (85.0%, n = 4880) of Group 1 antimicrobial prescriptions were written for female patients, younger patients and patients who visited the office less frequently (Table 3). Antimicrobial prescriptions for female patients were less likely to adhere to referenced guidelines compared to prescriptions for male patients (70.5% versus 77.1%, respectively). Female patients had a 1.17 times higher odds of receiving a potentially inappropriate antimicrobial prescription compared to male patients. An antimicrobial prescription for Group 1 was more likely among patients aged 51–69, than other age groups. However, there was a 7% decrease in the odds of potentially inappropriate prescribing for every 5 years of increase in patient age. Potentially inappropriate prescriptions were more likely among patients aged 16–30 (60%) and 31–50 (55%), versus patients aged 70 and older (33%). This represents a 27% absolute reduction in the risk of potentially inappropriate antimicrobial prescribing between patients aged 16–30 and those aged 70 and older. Additionally, for every increase in 5 office visits per patient, there was a 4% decrease in the odds of obtaining a potentially inappropriate antimicrobial prescription. Those who had less than 5 office visits per year were at an increased risk of potentially inappropriate antimicrobial prescribing compared to those who had greater than 10 office visits. Prescriber and practice factors were not associated with Group 1 potentially inappropriate antimicrobial prescribing (Table 3). Group 2: Viral infections There were 186 860 primary care visits for Group 2 in the study cohort. Of these, 29 833 had an associated prescription, suggesting 15.9% of patients received a potentially inappropriate prescription. Antimicrobial treatment for viral infections represented approximately 156 700 days of potentially unnecessary antimicrobial use. Antimicrobial prescriptions were largely prescribed for acute mild-to-moderate sinusitis (37.0%), bronchitis (30.4%) and upper respiratory tract infection (28.6%) (Table 4). Beta-lactams (44.9%) and macrolides (38.5%) were the leading antimicrobials prescribed for Group 2. Table 4. Antimicrobial prescriptions for viral diagnoses recorded in 186 860 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) View Large Table 4. Antimicrobial prescriptions for viral diagnoses recorded in 186 860 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) View Large In this group, the odds of receiving an antimicrobial prescription was greater among older patients, patients with more comorbidities, patients with more frequent office visits, patients in larger practices and patients in rural practices (Table 5). The odds of receiving an antimicrobial for an infection were 1.13 times higher for every 10 years increase in patient age. Patients aged 51–70 had the highest risk of receiving an antimicrobial prescription for a viral infection, accounting for 34.6% of all Group 2 prescriptions. Patients younger than 15 were least likely to receive a prescription for a viral infection (10.5%). Patients who received care at a rural practice (OR 1.47, CI 1.17–1.84) or a larger practice (OR 2.36, CI 1.76–3.16) had higher odds of receiving an antimicrobial prescription for a viral infection. Prescriber age was also significant; every 10 years increase in prescriber age led to a reduction in the odds of prescribing for viral infections (OR 0.66, CI 0.48–0.90). Table 5. Multivariate mixed-effect logistic regression model of viral diagnoses prescriptions not guideline based in 29 833 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Bold numbers are significant at a P value of 0.05. View Large Table 5. Multivariate mixed-effect logistic regression model of viral diagnoses prescriptions not guideline based in 29 833 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Bold numbers are significant at a P value of 0.05. View Large Discussion This study establishes a baseline for quality improvement initiatives and is foundational to developing evidence-informed AS in primary care. Based on reviewed guidelines, 58% of identified Group 1 antimicrobial prescriptions were potentially inappropriate. In almost half of these an incorrect antimicrobial was prescribed. The highest rate of potentially inappropriate prescribing was among prescriptions written for adult pneumonia (64.7%). Although this study considered fluoroquinolones an appropriate treatment of pneumonia, further investigation may be warranted. Fifty-five percent of the pneumonia prescriptions in this study were written for a fluoroquinolone, which referenced guidelines consider second-line treatment. The use of fluoroquinolones should be monitored and increases in prescribing rates explored as their use is associated with antibacterial resistance (16). UTIs were potentially inappropriately treated in 60% of observed cases. Macrolides seemed to be frequently prescribed for UTI, a target area for AS programs since macrolides are typically not effective for the treatment of uropathogens. Of note, pelvic chlamydial infections may co-occur or be miscoded as an UTI. One-day macrolide prescriptions may be appropriately prescribed for the treatment of chlamydia. Only a small number of the macrolide prescriptions for UTIs (0.004%, n = 8) were for 1 day suggesting a small proportion of UTI diagnoses may have been correctly treating chlamydia. Inappropriate use of macrolides can produce resistance in respiratory tract flora within 4 days and persist for 6 months (9–15,17). Macrolide resistance in Canada is an increasing concern. A sample of Group A streptococcal isolates from throat swabs showed the macrolide resistance had increased from 2.1% to 14.4% in a 4-year period (18). A modest reduction in the amount of unnecessary macrolide use could potentially result in a significant reduction in the amount of resistance. A country-wide program in Finland reduced the use of erythromycin by 50%, which in turn reduced the resistance of group A streptococcal isolates from 17% to 9% (14). While evidence is limited, there are some studies demonstrating that shorter courses of therapy are equivalent to longer ones (19,20). Among adults hospitalized with pneumonia, stopping antibiotics after 5 days of therapy was not inferior to traditional, prescriber-determined longer courses of therapy (19). Our study demonstrated that in 15% of prescriptions with the appropriate antimicrobial the duration of treatment was longer than some guideline-based recommendations. Stewardship efforts targeting UTI and cellulitis prescriptions would be ideal targets given how common they are in primary care settings. Patient gender and age were associated with an increased rate of potentially inappropriate prescribing for Group 1 infections. Females were 1.6 times more likely to be prescribed an antimicrobial not in accordance with guidelines. Other research has reported a similar relationship. Serna et al. (21) found a 1.5 times increase in inappropriate antimicrobial prescriptions among females and younger patients (aged 15–19 years). Within the present study, although the number of prescriptions increased with age, younger adult patients aged 16–30 were more likely to have a potentially inappropriately prescribed antimicrobial for the bacterial infection. Patients older than 69 were 27% less likely to have a potentially inappropriate prescription for a bacterial infection compared to those younger than 31. This finding suggests further study to clarify underlying factors mitigating this association. Potentially inappropriate prescribing for viral infections was identified in 16% of the visits, representing 156 700 days of potentially unnecessary antimicrobial use. The highest rate of antimicrobial prescribing in Group 2 was for patients with acute respiratory infections. Similarly, a recent trial reported that 24.1% of adult primary care visits for an acute respiratory infection resulted in an antimicrobial prescription (22). Modest reductions in inappropriate prescribing could impact rates of antimicrobial resistance, patient safety and cost to the patients and health system. Within Group 2, our study found a higher rate of prescribing among older patients, those with a greater number of comorbidities and/or those who visited a primary care office more frequently. This finding is likely related to a perceived need for an antimicrobial following clinical assessment. In primary care settings, antimicrobials are often considered for clinical presentations that are viral due to concern of a secondary bacterial infection. Similarly, other studies have found that inappropriate prescribing of antimicrobials are related to comorbidity, older age and presentation in the emergency department reflecting concern for poor clinical outcomes and antimicrobial resistance (12). Our data suggest the need for improvement in antimicrobial prescribing, particularly in practices prescribing antimicrobials for greater than 16% of viral infection presentations. This study demonstrated that younger clinicians and those within larger rurally located practices were more likely to inappropriately prescribe an antimicrobial for a viral infection. This may reflect inexperience or time constraints for clinical assessment. There may also be hesitanecy to disappoint patients among younger providers establishing their practice. Recent studies have shown associations between physician time of day and inappropriate antimicrobial prescribing, which suggests workload and physician fatigue may account for this association (21–23). Rural regions of Manitoba have less physicians per population than urban areas which may exacerbate the challenge of physician fatigue related to increased workload. Health system planners and stakeholders should be included in the discussion regarding health workforce concerns. Limitations While this study represents a comprehensive sample of primary care appointments in Manitoba within the MaPCReN database, it does not include all Manitoba primary care appointments. Guidelines from Canada and the US were used as referenced standard; based on clinical presentation of the patient antimicrobial prescribing decisions may not follow these guidelines. Physician agreement with the referenced guidelines was not assessed within this study. The study is based on EMR data capture of structured data prescription and diagnostic information that did not describe clinical presentation of the patient. Since some of the prescriptions deemed inappropriate may be appropriate after considering clinical presentation we have used the terms ‘potentially inappropriate’ or ‘likely indicated’ throughout the study. Alterations to prescriptions after printing are also not captured in our data. This may have increased the number of ‘too short’ prescriptions and may not reflect the final dispensed medication. In Manitoba, EMR records contain one diagnosis code per visit to a health care provider, sent to Manitoba Health for the purposes of billing (6). Diagnosis chosen for billing was assumed to be correct. As well, since only one code is entered per visit it is possible that some granularity in the clinical presentation is overlooked. This study captured appointments for common infections listed in Group 1 or Group 2 conditions and assumed the diagnosis entered by the provider was accurate. Some diagnoses, such as otitis media, were not included in the study due to controversy over appropriate treatment and the use of delayed prescriptions. This study assumed linkage between an antimicrobial prescription and the most recent encounter; there may have been times when this assumption was not correct. Utilizing the MaPCReN database enabled the present study to assess the reason for prescription but did not assess prescription dispensation or patient adherence. This is a retrospective cohort study using regression analysis to assess associations; causality reasoning is limited by this approach. Conclusion This study seeks to further our understanding of potentially inappropriate antimicrobial prescribing in Canadian primary care by providing a description of suboptimal antimicrobial prescribing for bacterial infections (i.e. inappropriate antimicrobial choice or inappropriate duration) and for viral infections. Our findings suggest optimal targets for AS interventions should be directed at antimicrobial prescribing for viral infections, especially among rural patients. We also found that management of pneumonia and UTIs offer reasonable targets for improvement. Collaboration with primary care stakeholders is vital for promoting practice change and establishing dedicated stewardship resources for optimal antimicrobial use. This research helps us to define a baseline rate of potentially inappropriate prescribing; an important preliminary step to designing AS community interventions. Supplementary material Supplementary material is available at Family Practice online. Declarations Funding: Operational funding from the Department of Family Medicine, Max Rady College of Medicine, Rady Faculty of Health Sciences, University of Manitoba was allowed for the collection, processing and maintenance of data holding used in this study. Ethical approval: Ethical approval for this study was obtained from the Health Research Ethics Board at the University of Manitoba. Conflict of interest: none. References 1. Feazel LM , Malhotra A , Perencevich EN et al. Effect of antibiotic stewardship programmes on Clostridium difficile incidence: a systematic review and meta-analysis . J Antimicrob Chemother 2014 ; 69 : 1748 – 54 . Google Scholar CrossRef Search ADS PubMed 2. MacDougall C , Polk RE . Antimicrobial stewardship programs in health care systems . Clin Microbiol Rev 2005 ; 18 : 638 – 56 . Google Scholar CrossRef Search ADS PubMed 3. Shapiro DJ , Hicks LA , Pavia AT , Hersh AL . Antibiotic prescribing for adults in ambulatory care in the USA, 2007–09 . J Antimicrob Chemother 2014 ; 69 : 234 – 40 . Google Scholar CrossRef Search ADS PubMed 4. Hersh AL , Shapiro DJ , Pavia AT , Shah SS . Antibiotic prescribing in ambulatory pediatrics in the United States . Pediatrics 2011 ; 128 : 1053 – 61 . Google Scholar CrossRef Search ADS PubMed 5. World Health Organization . The Anatomical Therapeutic Chemical Classification System with Defined Daily Doses (ATC/DDD) . 2003 . http://www.who.int/classifications/atcddd/en/ (accessed on 10 July 2017 ). 6. Katz A , Halas G , Dillon M , Sloshower J . Describing the content of primary care: limitations of Canadian billing data . BMC Fam Pract 2012 ; 13 : 7 . Google Scholar CrossRef Search ADS PubMed 7. Kern DM , Davis J , Williams SA et al. Validation of an administrative claims-based diagnostic code for pneumonia in a US-based commercially insured COPD population . Int J Chron Obstruct Pulmon Dis 2015 ; 10 : 1417 – 25 . Google Scholar CrossRef Search ADS PubMed 8. Mangione-Smith R , Wong L , Elliott MN , McDonald L , Roski J . Measuring the quality of antibiotic prescribing for upper respiratory infections and bronchitis in 5 US health plans . Arch Pediatr Adolesc Med 2005 ; 159 : 751 – 7 . Google Scholar CrossRef Search ADS PubMed 9. Gupta K , Hooton TM , Naber KG et al. ; Infectious Diseases Society of America; European Society for Microbiology and Infectious Diseases . International clinical practice guidelines for the treatment of acute uncomplicated cystitis and pyelonephritis in women: a 2010 update by the Infectious Diseases Society of America and the European Society for Microbiology and Infectious Diseases . Clin Infect Dis 2011 ; 52 : e103 – 20 . Google Scholar CrossRef Search ADS PubMed 10. Subcommittee on Urinary Tract Infection, Steering Committee on Quality Improvement and Management . Urinary tract infection: clinical practice guidelines for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months . Pediatrics 2011 ; 128 : 595 – 610 . CrossRef Search ADS PubMed 11. Robinson J , Finlay J , Lang M , Bortolussi R . Urinary tract infection in infants and children: diagnosis and management . Paediatr Child Health 2014 ; 19 : 315 – 319 . Google Scholar CrossRef Search ADS PubMed 12. Shulman ST , Bisno AL , Clegg HW et al. Clinical practice guideline for the diagnosis and management of group A streptococcal pharyngitis: 2012 update by the Infectious Diseases Society of America . Clin Infect Dis 2012 ; 55 : 1279 – 82 . Google Scholar CrossRef Search ADS PubMed 13. Stevens D , Bisno A , Chambers H et al. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America . Clin Infect Dis 2014 ; 59: 147–59. 14. Le Saux N , Robinson JL ; Canadian Paediatric Society, Infectious Diseases and Immunization Committee . Uncomplicated pneumonia in healthy Canadian children and youth: practice points for management . Paediatr Child Health 2015 ; 20 : 441 – 50 . Google Scholar CrossRef Search ADS PubMed 15. Mandell LA , Wunderink RG , Anzueto A et al. ; Infectious Diseases Society of America; American Thoracic Society . Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults . Clin Infect Dis 2007 ; 44 ( suppl 2) : S27 – 72 . Google Scholar CrossRef Search ADS PubMed 16. Summary Safety Review-Fluoroquinolones-Assessing the potential risk of persistent and disabling side effects . Health Canada . http://www.hc-sc.gc.ca/dhp-mps/medeff/reviews-examens/fluoroquinolones2-eng.php (accessed on 18 April 2017 ). 17. Seppälä H , Klaukka T , Vuopio-Varkila J et al. The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland . N Engl J Med 1997 ; 337 : 441 – 6 . Google Scholar CrossRef Search ADS PubMed 18. Katz KC , McGeer AJ , Duncan CL et al. Emergence of macrolide resistance in throat culture isolates of group a streptococci in Ontario, Canada, in 2001 . Antimicrob Agents Chemother 2003 ; 47 : 2370 – 2 . Google Scholar CrossRef Search ADS PubMed 19. Uranga A , España PP , Bilbao A et al. Duration of antibiotic treatment in community-acquired pneumonia . JAMA Intern Med 2016 ; 176 : 1257 – 65 . Google Scholar CrossRef Search ADS PubMed 20. Hepburn MJ , Dooley DP , Skidmore PJ et al. Comparison of short-course (5 days) and standard (10 days) treatment for uncomplicated cellulitis . Arch Intern Med 2004 ; 164 : 1669 – 74 . Google Scholar CrossRef Search ADS PubMed 21. Serna MC , Real J , Ribes E et al. [Factors determining antibiotic prescription in primary care] . Enferm Infecc Microbiol Clin 2011 ; 29 : 193 – 200 . Google Scholar CrossRef Search ADS PubMed 22. Meeker D , Linder JA , Fox CR et al. Effect of behavioral interventions on inappropriate antibiotic prescribing among primary care practices: a randomized clinical trial . JAMA 2016 ; 315 : 562 – 70 . Google Scholar CrossRef Search ADS PubMed 23. Linder JA , Doctor JN , Friedberg MW et al. Time of day and the decision to prescribe antibiotics . JAMA Intern Med 2014 ; 174 : 2029 – 31 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Family Practice Oxford University Press

Informing antimicrobial stewardship: factors associated with inappropriate antimicrobial prescribing in primary care

Loading next page...
 
/lp/ou_press/informing-antimicrobial-stewardship-factors-associated-with-32svN9o5o6
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0263-2136
eISSN
1460-2229
D.O.I.
10.1093/fampra/cmx118
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Antimicrobial stewardship (AS) programs promote the optimal use of antimicrobials and safe patient care. With most antimicrobials prescribed in the ambulatory setting, establishing benchmark data is imperative to gauge the impact of future AS initiatives. Objectives To determine the frequency of potentially inappropriate antimicrobial prescribing in primary care practices in Manitoba, Canada and to assess the association between potentially inappropriate antimicrobial prescribing and patient, prescriber and practice-related factors. Methods A retrospective cohort study using the Manitoba Primary Care Research Network repository of de-identified Electronic Medical Records from consenting primary care practices. Descriptive statistics and logistic regressions detailed patients with bacterial or viral infections of interest and antimicrobial prescriptions. Results Eighteen percent (n = 35 574) of primary care visits for common infections were associated with a potentially inappropriate antimicrobial prescription. Among antimicrobials prescribed to patients diagnosed with bacterial infections, 37.8% (n = 2168) had a potentially inappropriate antimicrobial prescribed and 19.6% (n = 1126) had an antimicrobial prescribed for a duration outside of guideline-based ranges. Female patients, younger age and less office visits were associated with potentially inappropriate antimicrobial prescribing for bacterial infections. Among physician visits for viral infection, 15.9% (n = 29 833) were associated with an antimicrobial prescription. Older patients, those with more comorbidity, more office visits and those who were seen in larger or rural practices, were associated with potentially inappropriate antimicrobial prescribing for viral infections. Conclusions High frequency of potentially inappropriate antimicrobial prescribing, especially in certain patient populations, suggests the need for coordinated community-based AS programs to optimize prescribing and improve patient care. Antibiotic, bacterial, bacterial infections, drug resistance, health care costs, primary health care, viral disease Background The emergence and spread of antimicrobial-resistant organisms highlights the need to preserve the efficacy of antimicrobials (1,2). Antimicrobial stewardship (AS) programs combine policies, education, surveillance and practice audit to optimize antimicrobial prescribing enhancing clinical outcomes, reducing sequelae and ensuring cost-effective therapy (1,2). AS programs reduced treatment duration, the incidences of Clostridium difficile colitis, antimicrobial-related adverse events, hospitalization duration and antimicrobial resistance (2,3). Development of effective AS strategies in primary care requires benchmarks for intervention evaluation. Most antimicrobial prescribing occurs in the ambulatory and outpatient settings. Antimicrobial prescribing occurs among 10% of adult and 20% of paediatric ambulatory visits, and anywhere from 25 to 68% of prescriptions are potentially inappropriate (not indicated, suboptimal antimicrobial spectrum) (3,4). It is crucial to understand rates of potentially inappropriate antimicrobial prescribing in community practices to develop and tailor suitable AS interventions. This study had two specific objectives aimed at informing community-based AS policies. The primary objective was to determine the frequency of potentially inappropriate antimicrobial prescribing (not indicated, suboptimal) in primary care practices in Manitoba, Canada. The secondary objective was to determine patient, prescriber and practice-related factors associated with potentially inappropriate (not indicated, suboptimal) antimicrobial prescribing. Methods This retrospective cohort study was conducted using the Manitoba Primary Care Research Network (MaPCReN) repository that collects and processes de-identified Electronic Medical Record (EMR) from consenting community-based primary care practices. At the time of the study, the MaPCReN contained administrative claims information extracted from 32 primary care clinics, 185 providers and 196 923 patients in Manitoba. All prescriptions generated in the EMR are captured by MaPCReN. In the medical office, prescriptions are printed or faxed to the patient’s pharmacy. Inclusion/exclusion criteria The Anatomical Therapeutic Chemical Classification System (ATC) divides drugs into different groups based on the organ and system in which they act, and their chemical, pharmacological and therapeutic properties (5). For the present study, prescription data were extracted from April 1999 to January 2016 for all antimicrobials starting with ATC code J01, ‘Antibacterial for Systemic Use’, in patients older than 1 year. All visits to a medical provider in Manitoba must include a tariff code and a single International Classification of Disease (ICD-9) in order for payment. In this study, diagnoses were matched to antimicrobial prescriptions within 7 days of the appointment (prior to, or after). If there were two visits within 7 days of the prescription, the one nearest to the antimicrobial prescription was considered. Prescription information was only included if it specified the antimicrobial prescribed and prescription duration. This method of identifying diagnosis and prescription data from administrative claims data has been validated within Manitoba, Canada and international studies (6–8). Diagnoses were separated into two groups: ‘Group 1’ consisted of bacterial infections (i.e. urinary tract infection (UTI), pharyngitis, skin/soft tissue infection, cellulitis and pneumonia), for which an antimicrobial prescription was likely indicated. Prescriptions between 30 June 2013 and 30 June 2015, linked to a Group 1 diagnosis, were deemed potentially inappropriate if they fell outside of the following evidence-informed guidelines: Infectious Diseases Society of America, Canadian Pediatric Society and American Academy of Pediatrics’ guideline-based standards (9–15). We included only recent antimicrobial prescriptions for bacterial infections (Group 1) to reflect current guidelines published between 2011 and 2014, as prior recommendations may have been different. These guidelines were chosen as they represent the ideal standard of care in Manitoba, Canada. When there was not uniform consensus between various North American guidelines, we considered a range of acceptable number of days as ‘appropriate’. A complete list of the Group 1 prescriptions was evaluated; what was considered ‘appropriate’ antimicrobial prescriptions and the range of acceptable duration can be found in Supplementary Table S1. ‘Group 2’ consisted of encounters with a diagnosis of a viral infection (i.e. acute sinusitis, acute laryngitis and trachealis, upper respiratory tract infection, bronchitis, acute rhinitis, nasopharyngitis and influenza) that typically do not require an antimicrobial (Table 1). Any prescription between April 1999 and January 2016, written within 7 days of a diagnosed Group 2 viral infection was considered potentially inappropriate. To account for antimicrobial treatment in response to a potentially coexisting bacterial infection, patients were excluded if they had a previous diagnosis of bronchiectasis, cystic fibrosis, chronic obstructive pulmonary disease or acute sinusitis (within 21 days prior to the encounter). Table 1. Number of potentially inappropriate antimicrobial prescriptions for bacterial infections in 3294 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) View Large Table 1. Number of potentially inappropriate antimicrobial prescriptions for bacterial infections in 3294 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) Antimicrobial (N = 3294) Potentially inappropriate antimicrobial Duration outside guideline-based ranges Macrolides 1491 (45.3%) 171 (5.2%) Beta-lactams 425 (12.9%) 178 (5.4%) Nitrofuran derivatives/imidazole derivatives 85 (2.6%) 366 (11.1%) Fluoroquinolones 43 (1.3%) 301 (9.1%) Sulfonamides and trimethoprim 55 (1.7%) 4 (0.01%) Tetracyclines 69 (2.1%) 106 (3.2%) Total 2168 (65.8%) 1126 (34.2%) View Large Statistical analysis Descriptive analyses based on patient, prescriber and practice factors associated with each prescription were performed. To identify factors that were statistically significantly associated with potentially inappropriate antimicrobial prescribing, a multilevel regression model was used. The unit of measure was the patient, with patient (i.e. comorbidity, age, gender, number of primary care office visits per year and socio-economic status), provider (i.e. country of graduation and age) and practice (i.e. size and location) covariates used to identify factors associated with an increased risk of potentially inappropriate prescriptions. Multivariable generalized linear mixed models were used to allow for correlation between patients within the same health care provider. Odds ratios and their confidence intervals are reported. SAS version 9.3 (SAS Institute, Cary, NC) was used for analyses. Results Between April 1999 and January 2016, the MaPCReN repository contained 192,601 primary care appointments for infections; 18.5% (n = 35 574) had an associated prescription for an antimicrobial. Of the antimicrobial prescriptions, 16.1% (n = 5741) were associated with a bacterial infection (Group 1) and 83.9% (n = 29 833) antimicrobial prescriptions were associated with a viral infection (Group 2). Female patients were more likely to have an appointment for an infection than male patients (62.0% versus 37.9%, respectively). Primary care appointments for an infection were for patients aged 1 to >89, with 48% (n = 89 871) of the appointments for adult patients aged 31–69. Patients younger than 20 years represented 13.9% (n = 25 903) of the primary care appointments. Young adults (aged 20–30) and seniors (70 or older) represented 13.2% and 24.8% of the primary care appointments for an infection, respectively. Group 1: Bacterial infections The MaPCReN repository provided a sample of 5741 antimicrobial prescriptions for Group 1. Within the Group 1 prescriptions, 37.8% (n = 2168) were potentially prescribed inappropriately and 19.6% (n = 1126) were prescribed for a duration outside of guideline-based ranges. Potentially inappropriate antimicrobial use was most frequent with macrolides and beta-lactams (Table 1). Macrolides made up the largest group of potentially inappropriately prescribed antimicrobials (50.4%, n = 1662). UTIs were the most frequently diagnosed condition to contain a potentially inappropriately prescribed macrolide (75.4%, n = 1367). Nitrofuran/imidazole derivatives (11.1%, n = 366) and fluoroquinolones (9.1%, n = 301) were the most likely antimicrobials to be prescribed for durations outside of guideline-based ranges (5–11). The majority of prescriptions in Group 1 were associated with a diagnosis of UTI (75.1%); of these 59.7% were potentially inappropriate (Table 2). An antimicrobial prescription for a UTI was more likely among female patients (92.7%, n = 3998) than male patients (7.4%, n = 317). A majority of these prescriptions were potentially inappropriate (59.7%, n = 2576). Forty percent (n = 1730) of antimicrobial prescriptions for UTI were written for an antimicrobial not considered a first-line treatment (Table 3). While the greatest number of antimicrobial prescriptions were for UTIs, the highest rate of potentially inappropriate antimicrobial prescribing in Group 1 was for pneumonia. When an incorrect duration was prescribed for pneumonia (n = 191), it was significantly more likely to have been prescribed for a shorter duration than guideline-based ranges. There were 102 one-day prescriptions (60.7%) compared to 8 (0.0004%) one-day prescriptions for UTI. The default prescription length in the EMRs used by providers in this study is 1 day. When 1-day prescriptions were removed, 20.9% (n = 66) of potentially inappropriate pneumonia prescriptions appeared to be too short. Macrolide prescriptions for pneumonia were frequently prescribed a shorter duration than guideline ranges (66.5%, n = 153). Table 2. Appropriate antimicrobial prescribing for diagnosed bacterial infections in 5745 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) View Large Table 2. Appropriate antimicrobial prescribing for diagnosed bacterial infections in 5745 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) Diagnosis n Prescription in accordance with guidelines Reason prescription was inappropriate Yes No Incorrect antibiotic Duration longer than guidelines Duration shorter than guidelines Pneumonia 474 158 (33.3%) 316 (65.8%) 125 (26.4%) 23 (4.9%) 168 (35.4%) Urinary tract infection 4315 1739 (40.3%) 2576 (59.7%) 1730 (40.1%) 836 (19.4%) 10 (0.2%) Pharyngitis 349 249 (71.3%) 100 (28.7%) 68 (19.5%) 0 (0%) 32 (9.2%) Skin/soft tissue infection (cellulitis) 607 293 (48.3%) 314 (51.7%) 256 (42.2%) 49 (8.1%) 9 (1.5%) Total 5745 2443 (42.5%) 3302 (57.5%) 2179 (37.9%) 908 (15.7%) 219 (3.8%) View Large Table 3. Multivariate mixed-effects logistic regression model of bacterial infection prescriptions not guidelines based in 5746 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Bold numbers are significant at a P value of 0.05. View Large Table 3. Multivariate mixed-effects logistic regression model of bacterial infection prescriptions not guidelines based in 5746 patients who had a primary care visit between 2013 and 2015 at a Manitoba Primary Care Research Network participating clinic Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Variable (n = 5746) Odds ratio P value 95% Confidence interval Patient factors Sex (female versus male) 1.174 0.0346 1.012–1.362 Patient age (per 5 years increase) 0.9373 <0.0001 0.9131–0.9620 Number of comorbid conditions (per 1 comorbidity increase) 1.0139 0.7825 0.9190–1.1187 Office visit frequency (per 5 visit increase) 0.9577 0.0173 0.9242–0.9924 Socioeconomic status (per 1 unit increase in SES) 1.0983 0.2532 0.9351–1.2900 Prescriber factors Country of graduation (other versus Canada) 0.919 0.2025 0.807–1.046 Physician age (per 5 years increase) 1.0214 0.6075 0.9422–1.1073 Practice factors Practice location (rural versus urban/suburban) .0933 0.2035 0.838–1.038 Practice size (per 100 patient increase) 0.9813 0.3350 0.9443–1.0197 Bold numbers are significant at a P value of 0.05. View Large The majority (85.0%, n = 4880) of Group 1 antimicrobial prescriptions were written for female patients, younger patients and patients who visited the office less frequently (Table 3). Antimicrobial prescriptions for female patients were less likely to adhere to referenced guidelines compared to prescriptions for male patients (70.5% versus 77.1%, respectively). Female patients had a 1.17 times higher odds of receiving a potentially inappropriate antimicrobial prescription compared to male patients. An antimicrobial prescription for Group 1 was more likely among patients aged 51–69, than other age groups. However, there was a 7% decrease in the odds of potentially inappropriate prescribing for every 5 years of increase in patient age. Potentially inappropriate prescriptions were more likely among patients aged 16–30 (60%) and 31–50 (55%), versus patients aged 70 and older (33%). This represents a 27% absolute reduction in the risk of potentially inappropriate antimicrobial prescribing between patients aged 16–30 and those aged 70 and older. Additionally, for every increase in 5 office visits per patient, there was a 4% decrease in the odds of obtaining a potentially inappropriate antimicrobial prescription. Those who had less than 5 office visits per year were at an increased risk of potentially inappropriate antimicrobial prescribing compared to those who had greater than 10 office visits. Prescriber and practice factors were not associated with Group 1 potentially inappropriate antimicrobial prescribing (Table 3). Group 2: Viral infections There were 186 860 primary care visits for Group 2 in the study cohort. Of these, 29 833 had an associated prescription, suggesting 15.9% of patients received a potentially inappropriate prescription. Antimicrobial treatment for viral infections represented approximately 156 700 days of potentially unnecessary antimicrobial use. Antimicrobial prescriptions were largely prescribed for acute mild-to-moderate sinusitis (37.0%), bronchitis (30.4%) and upper respiratory tract infection (28.6%) (Table 4). Beta-lactams (44.9%) and macrolides (38.5%) were the leading antimicrobials prescribed for Group 2. Table 4. Antimicrobial prescriptions for viral diagnoses recorded in 186 860 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) View Large Table 4. Antimicrobial prescriptions for viral diagnoses recorded in 186 860 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) Diagnosis Associated appointments Prescriptions Acute rhinitis 2447 (1.3%) 135 (0.5%) Acute mild-to-moderate sinusitis 27 728 (14.8%) 11 043 (37%) Acute laryngitis and tracheitis 5469 (2.9%) 382 (1.3%) Upper respiratory tract infection 100 458 (53.8%) 8542 (28.6%) Bronchitis 28 164 (15.1%) 9030 (30.2%) Nasopharyngitis 17 458 (9.3%) 403 (1.4%) Influenza 5145 (2.8%) 298 (1%) Total 186 860 (100%) 29 833 (15.9%) View Large In this group, the odds of receiving an antimicrobial prescription was greater among older patients, patients with more comorbidities, patients with more frequent office visits, patients in larger practices and patients in rural practices (Table 5). The odds of receiving an antimicrobial for an infection were 1.13 times higher for every 10 years increase in patient age. Patients aged 51–70 had the highest risk of receiving an antimicrobial prescription for a viral infection, accounting for 34.6% of all Group 2 prescriptions. Patients younger than 15 were least likely to receive a prescription for a viral infection (10.5%). Patients who received care at a rural practice (OR 1.47, CI 1.17–1.84) or a larger practice (OR 2.36, CI 1.76–3.16) had higher odds of receiving an antimicrobial prescription for a viral infection. Prescriber age was also significant; every 10 years increase in prescriber age led to a reduction in the odds of prescribing for viral infections (OR 0.66, CI 0.48–0.90). Table 5. Multivariate mixed-effect logistic regression model of viral diagnoses prescriptions not guideline based in 29 833 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Bold numbers are significant at a P value of 0.05. View Large Table 5. Multivariate mixed-effect logistic regression model of viral diagnoses prescriptions not guideline based in 29 833 primary care appointments between 1999 and 2016 at a Manitoba Primary Care Research Network participating clinic Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Variable Odds ratio P value 95% Confidence intervals Patient factors Sex (female versus male) 1.05 0.05 0.99–1.11 Patient age (per 10 years increase) 1.13 0.007 1.03–1.24 Number of comorbid conditions (per 1 comorbidity increase) 1.11 0.0007 1.07–1.17 Office visit frequency (per 5 visit increase) 1.12 <0.0001 1.08–1.22 Prescriber factors Country of graduation (other versus Canada) 1.01 0.96 0.79–1.29 Prescriber age (per 10 years increase) 0.66 0.009 0.48–0.90 Practice factors Practice location (rural versus urban/suburban) 1.47 0.0009 1.17–1.84 Practice size (patient count < 1235 versus practices ≥ 1235) 2.36 <0.0001 1.76–3.16 Bold numbers are significant at a P value of 0.05. View Large Discussion This study establishes a baseline for quality improvement initiatives and is foundational to developing evidence-informed AS in primary care. Based on reviewed guidelines, 58% of identified Group 1 antimicrobial prescriptions were potentially inappropriate. In almost half of these an incorrect antimicrobial was prescribed. The highest rate of potentially inappropriate prescribing was among prescriptions written for adult pneumonia (64.7%). Although this study considered fluoroquinolones an appropriate treatment of pneumonia, further investigation may be warranted. Fifty-five percent of the pneumonia prescriptions in this study were written for a fluoroquinolone, which referenced guidelines consider second-line treatment. The use of fluoroquinolones should be monitored and increases in prescribing rates explored as their use is associated with antibacterial resistance (16). UTIs were potentially inappropriately treated in 60% of observed cases. Macrolides seemed to be frequently prescribed for UTI, a target area for AS programs since macrolides are typically not effective for the treatment of uropathogens. Of note, pelvic chlamydial infections may co-occur or be miscoded as an UTI. One-day macrolide prescriptions may be appropriately prescribed for the treatment of chlamydia. Only a small number of the macrolide prescriptions for UTIs (0.004%, n = 8) were for 1 day suggesting a small proportion of UTI diagnoses may have been correctly treating chlamydia. Inappropriate use of macrolides can produce resistance in respiratory tract flora within 4 days and persist for 6 months (9–15,17). Macrolide resistance in Canada is an increasing concern. A sample of Group A streptococcal isolates from throat swabs showed the macrolide resistance had increased from 2.1% to 14.4% in a 4-year period (18). A modest reduction in the amount of unnecessary macrolide use could potentially result in a significant reduction in the amount of resistance. A country-wide program in Finland reduced the use of erythromycin by 50%, which in turn reduced the resistance of group A streptococcal isolates from 17% to 9% (14). While evidence is limited, there are some studies demonstrating that shorter courses of therapy are equivalent to longer ones (19,20). Among adults hospitalized with pneumonia, stopping antibiotics after 5 days of therapy was not inferior to traditional, prescriber-determined longer courses of therapy (19). Our study demonstrated that in 15% of prescriptions with the appropriate antimicrobial the duration of treatment was longer than some guideline-based recommendations. Stewardship efforts targeting UTI and cellulitis prescriptions would be ideal targets given how common they are in primary care settings. Patient gender and age were associated with an increased rate of potentially inappropriate prescribing for Group 1 infections. Females were 1.6 times more likely to be prescribed an antimicrobial not in accordance with guidelines. Other research has reported a similar relationship. Serna et al. (21) found a 1.5 times increase in inappropriate antimicrobial prescriptions among females and younger patients (aged 15–19 years). Within the present study, although the number of prescriptions increased with age, younger adult patients aged 16–30 were more likely to have a potentially inappropriately prescribed antimicrobial for the bacterial infection. Patients older than 69 were 27% less likely to have a potentially inappropriate prescription for a bacterial infection compared to those younger than 31. This finding suggests further study to clarify underlying factors mitigating this association. Potentially inappropriate prescribing for viral infections was identified in 16% of the visits, representing 156 700 days of potentially unnecessary antimicrobial use. The highest rate of antimicrobial prescribing in Group 2 was for patients with acute respiratory infections. Similarly, a recent trial reported that 24.1% of adult primary care visits for an acute respiratory infection resulted in an antimicrobial prescription (22). Modest reductions in inappropriate prescribing could impact rates of antimicrobial resistance, patient safety and cost to the patients and health system. Within Group 2, our study found a higher rate of prescribing among older patients, those with a greater number of comorbidities and/or those who visited a primary care office more frequently. This finding is likely related to a perceived need for an antimicrobial following clinical assessment. In primary care settings, antimicrobials are often considered for clinical presentations that are viral due to concern of a secondary bacterial infection. Similarly, other studies have found that inappropriate prescribing of antimicrobials are related to comorbidity, older age and presentation in the emergency department reflecting concern for poor clinical outcomes and antimicrobial resistance (12). Our data suggest the need for improvement in antimicrobial prescribing, particularly in practices prescribing antimicrobials for greater than 16% of viral infection presentations. This study demonstrated that younger clinicians and those within larger rurally located practices were more likely to inappropriately prescribe an antimicrobial for a viral infection. This may reflect inexperience or time constraints for clinical assessment. There may also be hesitanecy to disappoint patients among younger providers establishing their practice. Recent studies have shown associations between physician time of day and inappropriate antimicrobial prescribing, which suggests workload and physician fatigue may account for this association (21–23). Rural regions of Manitoba have less physicians per population than urban areas which may exacerbate the challenge of physician fatigue related to increased workload. Health system planners and stakeholders should be included in the discussion regarding health workforce concerns. Limitations While this study represents a comprehensive sample of primary care appointments in Manitoba within the MaPCReN database, it does not include all Manitoba primary care appointments. Guidelines from Canada and the US were used as referenced standard; based on clinical presentation of the patient antimicrobial prescribing decisions may not follow these guidelines. Physician agreement with the referenced guidelines was not assessed within this study. The study is based on EMR data capture of structured data prescription and diagnostic information that did not describe clinical presentation of the patient. Since some of the prescriptions deemed inappropriate may be appropriate after considering clinical presentation we have used the terms ‘potentially inappropriate’ or ‘likely indicated’ throughout the study. Alterations to prescriptions after printing are also not captured in our data. This may have increased the number of ‘too short’ prescriptions and may not reflect the final dispensed medication. In Manitoba, EMR records contain one diagnosis code per visit to a health care provider, sent to Manitoba Health for the purposes of billing (6). Diagnosis chosen for billing was assumed to be correct. As well, since only one code is entered per visit it is possible that some granularity in the clinical presentation is overlooked. This study captured appointments for common infections listed in Group 1 or Group 2 conditions and assumed the diagnosis entered by the provider was accurate. Some diagnoses, such as otitis media, were not included in the study due to controversy over appropriate treatment and the use of delayed prescriptions. This study assumed linkage between an antimicrobial prescription and the most recent encounter; there may have been times when this assumption was not correct. Utilizing the MaPCReN database enabled the present study to assess the reason for prescription but did not assess prescription dispensation or patient adherence. This is a retrospective cohort study using regression analysis to assess associations; causality reasoning is limited by this approach. Conclusion This study seeks to further our understanding of potentially inappropriate antimicrobial prescribing in Canadian primary care by providing a description of suboptimal antimicrobial prescribing for bacterial infections (i.e. inappropriate antimicrobial choice or inappropriate duration) and for viral infections. Our findings suggest optimal targets for AS interventions should be directed at antimicrobial prescribing for viral infections, especially among rural patients. We also found that management of pneumonia and UTIs offer reasonable targets for improvement. Collaboration with primary care stakeholders is vital for promoting practice change and establishing dedicated stewardship resources for optimal antimicrobial use. This research helps us to define a baseline rate of potentially inappropriate prescribing; an important preliminary step to designing AS community interventions. Supplementary material Supplementary material is available at Family Practice online. Declarations Funding: Operational funding from the Department of Family Medicine, Max Rady College of Medicine, Rady Faculty of Health Sciences, University of Manitoba was allowed for the collection, processing and maintenance of data holding used in this study. Ethical approval: Ethical approval for this study was obtained from the Health Research Ethics Board at the University of Manitoba. Conflict of interest: none. References 1. Feazel LM , Malhotra A , Perencevich EN et al. Effect of antibiotic stewardship programmes on Clostridium difficile incidence: a systematic review and meta-analysis . J Antimicrob Chemother 2014 ; 69 : 1748 – 54 . Google Scholar CrossRef Search ADS PubMed 2. MacDougall C , Polk RE . Antimicrobial stewardship programs in health care systems . Clin Microbiol Rev 2005 ; 18 : 638 – 56 . Google Scholar CrossRef Search ADS PubMed 3. Shapiro DJ , Hicks LA , Pavia AT , Hersh AL . Antibiotic prescribing for adults in ambulatory care in the USA, 2007–09 . J Antimicrob Chemother 2014 ; 69 : 234 – 40 . Google Scholar CrossRef Search ADS PubMed 4. Hersh AL , Shapiro DJ , Pavia AT , Shah SS . Antibiotic prescribing in ambulatory pediatrics in the United States . Pediatrics 2011 ; 128 : 1053 – 61 . Google Scholar CrossRef Search ADS PubMed 5. World Health Organization . The Anatomical Therapeutic Chemical Classification System with Defined Daily Doses (ATC/DDD) . 2003 . http://www.who.int/classifications/atcddd/en/ (accessed on 10 July 2017 ). 6. Katz A , Halas G , Dillon M , Sloshower J . Describing the content of primary care: limitations of Canadian billing data . BMC Fam Pract 2012 ; 13 : 7 . Google Scholar CrossRef Search ADS PubMed 7. Kern DM , Davis J , Williams SA et al. Validation of an administrative claims-based diagnostic code for pneumonia in a US-based commercially insured COPD population . Int J Chron Obstruct Pulmon Dis 2015 ; 10 : 1417 – 25 . Google Scholar CrossRef Search ADS PubMed 8. Mangione-Smith R , Wong L , Elliott MN , McDonald L , Roski J . Measuring the quality of antibiotic prescribing for upper respiratory infections and bronchitis in 5 US health plans . Arch Pediatr Adolesc Med 2005 ; 159 : 751 – 7 . Google Scholar CrossRef Search ADS PubMed 9. Gupta K , Hooton TM , Naber KG et al. ; Infectious Diseases Society of America; European Society for Microbiology and Infectious Diseases . International clinical practice guidelines for the treatment of acute uncomplicated cystitis and pyelonephritis in women: a 2010 update by the Infectious Diseases Society of America and the European Society for Microbiology and Infectious Diseases . Clin Infect Dis 2011 ; 52 : e103 – 20 . Google Scholar CrossRef Search ADS PubMed 10. Subcommittee on Urinary Tract Infection, Steering Committee on Quality Improvement and Management . Urinary tract infection: clinical practice guidelines for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months . Pediatrics 2011 ; 128 : 595 – 610 . CrossRef Search ADS PubMed 11. Robinson J , Finlay J , Lang M , Bortolussi R . Urinary tract infection in infants and children: diagnosis and management . Paediatr Child Health 2014 ; 19 : 315 – 319 . Google Scholar CrossRef Search ADS PubMed 12. Shulman ST , Bisno AL , Clegg HW et al. Clinical practice guideline for the diagnosis and management of group A streptococcal pharyngitis: 2012 update by the Infectious Diseases Society of America . Clin Infect Dis 2012 ; 55 : 1279 – 82 . Google Scholar CrossRef Search ADS PubMed 13. Stevens D , Bisno A , Chambers H et al. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America . Clin Infect Dis 2014 ; 59: 147–59. 14. Le Saux N , Robinson JL ; Canadian Paediatric Society, Infectious Diseases and Immunization Committee . Uncomplicated pneumonia in healthy Canadian children and youth: practice points for management . Paediatr Child Health 2015 ; 20 : 441 – 50 . Google Scholar CrossRef Search ADS PubMed 15. Mandell LA , Wunderink RG , Anzueto A et al. ; Infectious Diseases Society of America; American Thoracic Society . Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults . Clin Infect Dis 2007 ; 44 ( suppl 2) : S27 – 72 . Google Scholar CrossRef Search ADS PubMed 16. Summary Safety Review-Fluoroquinolones-Assessing the potential risk of persistent and disabling side effects . Health Canada . http://www.hc-sc.gc.ca/dhp-mps/medeff/reviews-examens/fluoroquinolones2-eng.php (accessed on 18 April 2017 ). 17. Seppälä H , Klaukka T , Vuopio-Varkila J et al. The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland . N Engl J Med 1997 ; 337 : 441 – 6 . Google Scholar CrossRef Search ADS PubMed 18. Katz KC , McGeer AJ , Duncan CL et al. Emergence of macrolide resistance in throat culture isolates of group a streptococci in Ontario, Canada, in 2001 . Antimicrob Agents Chemother 2003 ; 47 : 2370 – 2 . Google Scholar CrossRef Search ADS PubMed 19. Uranga A , España PP , Bilbao A et al. Duration of antibiotic treatment in community-acquired pneumonia . JAMA Intern Med 2016 ; 176 : 1257 – 65 . Google Scholar CrossRef Search ADS PubMed 20. Hepburn MJ , Dooley DP , Skidmore PJ et al. Comparison of short-course (5 days) and standard (10 days) treatment for uncomplicated cellulitis . Arch Intern Med 2004 ; 164 : 1669 – 74 . Google Scholar CrossRef Search ADS PubMed 21. Serna MC , Real J , Ribes E et al. [Factors determining antibiotic prescription in primary care] . Enferm Infecc Microbiol Clin 2011 ; 29 : 193 – 200 . Google Scholar CrossRef Search ADS PubMed 22. Meeker D , Linder JA , Fox CR et al. Effect of behavioral interventions on inappropriate antibiotic prescribing among primary care practices: a randomized clinical trial . JAMA 2016 ; 315 : 562 – 70 . Google Scholar CrossRef Search ADS PubMed 23. Linder JA , Doctor JN , Friedberg MW et al. Time of day and the decision to prescribe antibiotics . JAMA Intern Med 2014 ; 174 : 2029 – 31 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

Family PracticeOxford University Press

Published: Dec 10, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off