- Email: [email protected]
Outpatient antimicrobial stewardship: Optimizing patient care via pharmacist led microbiology review
ARTICLE IN PRESS American Journal of Infection Control 000 (2019) 1−5
Contents lists available at ScienceDirect
American Journal of Infection Control journal homepage: www.ajicjournal.org
Major Article
Outpatient antimicrobial stewardship: Optimizing patient care via pharmacist led microbiology review Bethany A. Wattengel PharmD, BCPS a, John A. Sellick DO, MS b, Kari A. Mergenhagen PharmD, BCPS, BCIDP a,* a b
Department of Pharmacy, Veteran’s Affairs Western New York Healthcare System, Buffalo, NY Department of Infectious Diseases, Veteran’s Affairs Western New York Healthcare System, Buffalo, NY
Key Words: Culture Antibiotic
Background: Cultures are often taken in the outpatient setting but results are not acted upon, leading to unnecessary re-presentations to the health care setting. Methods: This study was a prospective study with interventions made between January 1, 2018, and January 1, 2019. Cultures were reviewed to ensure appropriate antimicrobial coverage. The objective was to compare outcomes with accepted versus rejected interventions. Descriptive statistics were used to summarize data. Results: A total of 7,360 antibiotic orders were reviewed by the infectious diseases pharmacists. Pharmacists intervened on 20.1% (n = 194) of encounters with related cultures. Interventions were most frequent in the emergency department (42%). Ciprofloxacin required the most interventions (26%), followed by third-generation cephalosporins (22%). The intervention acceptance rate was 76%, which was associated with decreased rates of 30-day treatment failure (5% vs 28%, P < .001) and 30-day admission (0.7% vs 11%, P = .001), when interventions were accepted rather than rejected. Discussion: Approximately 20% of patients required intervention. Culture review services may be beneficial in a variety of outpatient settings. Outpatient stewardship literature is limited, and our study found a decrease in admission and treatment failure. Conclusions: Microbiology review and intervention positively impacted care for outpatients. Intervention was associated with significantly decreased rates of treatment failure and admission when interventions were accepted. Published by Elsevier Inc. on behalf of Association for Professionals in Infection Control and Epidemiology, Inc.
Antimicrobial stewardship programs are well established in the inpatient setting; however, progress has lagged in the outpatient setting. Recent development of outpatient programs can partially be attributed to the publishing of the White House’s national action plan for combating antibiotic-resistant bacteria, which called for a 50% decrease of inappropriate antibiotic use in the outpatient setting by 2020.1 Although no longer supported by the administration in the United States, this plan had spurred the development of outpatient antimicrobial stewardship programs including services, which review microbiology cultures. Outpatient use of antibiotics is often overlooked due to relatively low drug costs; however, it is an important cause of antimicrobial usage. Outpatient stewardship services are in their infancy. A targeted approach to outpatient antimicrobial stewardship includes review * Address correspondence to Kari A. Mergenhagen, PharmD, BCPS, BCIDP, Veteran’s Affairs Western New York Healthcare System, Pharmacy Department, 119, 3495 Bailey Ave, Buffalo, NY 14215 E-mail address: [email protected] (K.A. Mergenhagen). Conflicts of interest: None to report.
of microbiologic cultures to ensure appropriate follow through. Although studies are limited, most of the available literature pertains to pharmacist-led culture review services.2-6 According to the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA), pharmacists play a crucial role due to their knowledge of appropriate antimicrobial use, dosing, duration of therapy, and drug interactions.7 With approximately 270.2 million outpatient antibiotic prescriptions prescribed in the United States in 2016, and at least 30% judged to be inappropriate, outpatient stewardship is key, and pharmacists are poised to play an integral role.8,9 Most of the literature available regarding culture review services takes place in the emergency departments (EDs) of hospitals or urgent care centers.2-6 Such literature focuses on outcomes such as time to culture review and/or time to provider notification.2,5 Studies also evaluated the quantity of interventions made by pharmacists and compared it with that of other health care professionals such as nurses and physicians.5,6 Limited information, however, is available on the impact of such interventions on patient outcomes.6
https://doi.org/10.1016/j.ajic.2019.07.018 0196-6553/Published by Elsevier Inc. on behalf of Association for Professionals in Infection Control and Epidemiology, Inc.
ARTICLE IN PRESS 2
B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5
With a growing need for outpatient stewardship, data are needed to guide the development of new services to improve patient care. Often cultures are obtained in the outpatient setting but results are not acted upon, leading to unnecessary re-presentations to the health care setting. This study evaluates the utility of a pharmacist-led culture review service in the outpatient setting. Rates of pharmacist intervention will be assessed, as well as the impact of such interventions, including re-presentation to an outpatient location (eg, ED, primary care, and specialty clinic), subsequent admission to the hospital, and treatment failure requiring additional antibiotic therapy. METHODS Study design This study was a prospective chart review at the Veterans Affairs Western New York Healthcare System via the computerized patient record system with interventions made as needed. The study period was January 1, 2018, through January 1, 2019. The infectious diseases pharmacists received alerts when oral antibiotics for outpatient use were ordered. Special attention was paid to those antibiotics ordered in relation to cultures, thus forming the culture review service. Orders were reviewed no later than 12 hours post time of antibiotic order. Cultures were reviewed daily to ensure appropriate antimicrobial coverage and timely interventions. All cultures were taken before the antibiotic was dispensed. Antibiotic regimens were also evaluated for appropriateness in terms of dose, duration, antibiotic allergies, drug interactions, and QTc interval. Interventions therefore included changes to antibiotic regimens (eg, antibiotic choice, dose, and duration), diagnostic recommendations (cultures and tests), and patient counseling. Interventions were made via direct contact with providers (eg, phone, instant message, and electronic medical record notes with provider attachment for acknowledgment) and documented via notes in the electronic medical record. Providers entered their own orders based on recommendations from the stewardship team. The infectious diseases physician was available to review patients as needed. This study was approved as a quality assurance/ quality improvement project. Patient population Patients were included if they were 18 years of age and older, were prescribed an antibiotic in an outpatient setting, and had a related culture ordered. Patients were excluded if they were under the age of 18 or did not have a culture obtained in relation to an outpatient antibiotic order. Patients were also excluded if they were admitted to the hospital immediately following outpatient antibiotic order or culture. “Outpatient setting” included ED, primary care (on site, off site, and home based primary care), and outpatient clinics. Outcomes The primary objective of this study was to compare outcomes in patients with accepted interventions vs rejected interventions: 30-day all-cause mortality, 30-day representation rates, 30-day admission rates, and 30-day treatment failure. Thirty-day treatment failure was a composite outcome which included patient who represented to an outpatient location or were admitted within 30 days for additional antibiotic therapy. This study also evaluated rates of appropriate antibiotic based on culture results, location of patient presentation, type of culture, and number and type of interventions made.
Definitions A drug-bug mismatch was defined as discordance between prescribed antibiotic therapy and culture susceptibility report. A drugbug mismatch also included cases of bacteria with known resistance mechanisms to prescribed antibiotic therapy. Such cases included patients with cultures positive for SPICE organisms (Serratia, Providencia, indole positive Proteus, Citrobacter, and Enterobacter) who were prescribed a cephalosporin when viable alternatives existed due to concern for inducible resistance via the AmpC gene.10 Last, if a culture was negative for growth and antibiotic therapy was prescribed, this was considered a drug-bug mismatch. A drug-bug match was defined as concordance between prescribed antibiotic therapy and culture susceptibility report. In cases where susceptibilities were not reported, cases were considered to be a drug-bug match if the organism report is predictably susceptible to prescribed antibiotic therapy. Accepted intervention occurred when recommendations to change a component of the antibiotic regimen (antibiotic choice, dose, or duration), discontinue an antibiotic, add an antibiotic, or order diagnostic tests or cultures were acted upon in accordance with the stewardship team’s recommendations. Rejected intervention occurred when recommendations to change a component of the antibiotic regimen (antibiotic choice, dose, or duration), discontinue an antibiotic, add an antibiotic, or order diagnostic tests or cultures were declined by providers. Statistical analysis Descriptive statistics were generated using JMP Pro software (version 14; SAS Institute Inc, Cary, NC). Data were presented as a mean and SD when normally distributed. A bivariate analysis was used to compare outcomes in patients with accepted interventions versus rejected interventions. RESULTS A total of 7,360 antibiotic orders were reviewed by the infectious diseases clinical pharmacists, including 965 encounters with related cultures (Fig 1). Time spent reviewing electronic medical records and antibiotic orders varied. Pharmacist review took as little as 5 minutes for some orders without cultures to as many as 30 minutes for those with cultures that required intervention. Pharmacists intervened in 20.1% (n = 194) of patient encounters with related cultures. Results of this study specifically focus on the 194 patient encounters that required pharmacist intervention with regard to the culture review service. The majority of antibiotic prescriptions that required intervention were from the ED (42.3%) and primary care (38.7%), with the remaining 19% being from various outpatient specialty clinics (Table 1). Most patients were older white males, with an average age of approximately 70 years. Antibiotic use varied, however, the most common antibiotics prescribed for patients requiring intervention were ciprofloxacin (26.3%), third-generation cephalosporins (22.2%), and sulfamethoxazole/trimethoprim (18%) (Table 2). The most frequent indication for use was urinary tract infection (71.1%) and therefore urine was the most common type of culture reviewed. An average of 3 days was required for the microbiology laboratory to finalize cultures. Thirtyone percent of cultures grew gram-negative organisms, 25.3% gram positive, 18.6% were mixed with both gram negatives and gram positives, and 18.6% of cultures were negative for growth. Patients were treated with appropriate antibiotic therapy in accordance with culture results in 22.2% of cases; there was drug-bug mismatch in 72.7% of cases. Approximately 5% (n = 10) of the cultures contained results
ARTICLE IN PRESS B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5
3
Fig 1. Inclusion and exclusion criteria.
that could not confirm whether prescribed therapy provided appropriate coverage due to lack of complete microbiological work up. Of the patients with confirmed drug-bug match or mismatch (n = 184), the highest degree of drug-bug mismatch occurred in gram negative cultures (29.1%). When comparing drug-bug mismatch to drug-bug match, mismatch occurred more often in mixed cultures with both gram negative and gram positive organisms (20.6% mismatch vs 14.0% match) and negative cultures without bacterial growth that did not require antibiotic therapy (25.5% mismatch vs 0% match) (P = .0014). A total of 36 cultures were mixed,
containing both gram positives and negatives. Patients with mixed culture results were prescribed an antibiotic that did not provide adequate coverage of culture results 80.6% (n = 29) of the time; however, only 58.3% of recommended interventions were accepted. In the majority of these cases, neither bacterium (gram positive or negative) was covered appropriately by the prescribed antibiotic. Patients with mixed cultures were more likely to experience 30-day treatment failure when interventions were rejected rather than accepted (75.0% vs 25.0%, P = .030). Accepted vs rejected interventions
Table 1 Accepted vs rejected interventions: demographic information
Location ED Primary care Specialty Location Main Rural Age Body mass index Creatinine clearance Male Race African American White Other Antibiotic allergy Beta-lactam allergy ED, emergency department.
Accepted (n = 148, 76.3%)
Rejected (n = 46, 23.7%)
65 (43.9%) 55 (37.2%) 28 (18.9%)
17 (37.0%) 19 (41.3%) 10 (21.7%)
130 (87.8%) 18 (12.2%) 70.8 +/- 15.2 27.7 +/- 8.1 75.2 +/- 40.1 127 (85.8%)
38 (82.6%) 8 (17.4%) 70.5 +/- 18.6 28.5 +/- 7.8 80.8 +/- 54.6 42 (91.3%)
34 (23.0%) 111 (75.0%) 3 (2.0%) 39 (26.4%) 26 (17.6%)
7 (15.2%) 37 (80.4%) 2 (4.4%) 7 (15.2%) 4 (8.7%)
P value
.70 — — — .36 — — .90 .54 .46 .33 .39 — — — .12 .15
A total of 194 encounters required interventions, with an acceptance rate of 76% (n = 148) (Table 2). Antibiotic therapy was not indicated in 37.6% of encounters that required intervention. Interventions were more likely to be accepted when antibiotic therapy was indicated (68.9% vs 41.3%, P = .007). It should be noted that multiple interventions were made on some encounters; and that a total of 229 interventions were made. The majority of interventions (73.8%, n = 169) involved a change to the prescribed antibiotic regimen. Of the interventions with a change to the regimen, 45.6% (n = 77) were recommendations to change the antibiotic, 38.5% (n = 65) were recommendations for discontinuation of therapy, 9.5% (n = 16) were recommendations to add or initiate antibiotic therapy, 4.1% (n = 7) were dose changes, and 2.4% (n = 4) were duration changes. The encounters where an intervention was rejected all involved a change of the antibiotic regimen. Pharmacist intervention was associated with significantly decreased rates of 30-day treatment failure (5.4% vs 28.3%, P < .0001) and 30-day admission (0.7% vs 10.9%, p = .0005) when interventions were accepted rather than rejected. There was no difference in 30-day mortality (P = .38) (Table 3).
ARTICLE IN PRESS B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5
4
Table 2 Accepted vs rejected interventions: Antibiotic regimen, culture, and intervention data
Antibiotic class Beta-lactam Fluoroquinolone Sulfamethoxazole/trimethoprim Macrolide Miscellaneous Nitrofurantoin Antibiotic duration (d) Indication UTI Skin and soft tissue Epididymoorchitis or prostatitis Pharyngitis Status post procedure Abscess Pre-op urology Helicobacter pylori Clostridium difficile Pneumonia Urethral injury UTI prophylaxis Osteomyelitis Otitis Media Vaginitis Time to culture finalization (d) Culture results Gram negative Gram positive Mixed gram negative and positive Negative culture Sexually transmitted infection Other Drug-bug match Yes No Unknown Antibiotic indicated Yes No Intervention (change in regimen) Intervention (education/counseling) Intervention (culture/tests)
Accepted (n = 148, 76.3%)
Rejected (n = 46, 23.7%)
67 (45.3%) 45 (30.4%) 22 (14.9%) 2 (1.4%) 7 (4.7%) 5 (3.4%) 8.4 +/- 3.4
19 (41.3%) 13 (28.3%) 13 (28.3%) 0 0 1 (2.2%) 9.9 +/- 9.1
106 (71.6%) 14 (9.5%) 5 (3.4%) 5 (3.4%) 4 (2.7%) 4 (2.7%) 3 (2.0%) 2 (1.4%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 0 0 2.8 +/- 1.8
32 (69.6%) 4 (8.7%) 2 (4.4%) 0 2 (4.4%) 0 2 (4.4%) 0 0 0 0 2 (4.4%) 0 1 (2.1%) 1 (2.2%) 2.8 +/- 1.3
51 (34.5%) 44 (29.7%) 21 (14.2) 23 (15.5%) 2 (1.4%) 7 (4.7%)
9 (19.6%) 5 (10.9%) 15 (32.6%) 13 (28.3%) 1 (2.2%) 3 (6.5%)
38 (25.7%) 106 (71.6%) 4 (2.7%)
5 (10.9%) 35 (76.1%) 6 (13.0)
102 (68.9%) 46 (31.1%) 123 (83.1%) 28 (18.9%) 20 (13.5%)
19 (41.3%) 27 (58.7%) 46 (100%) 5 (10.9%) 7 (15.2%)
P value
.25 — — — — — — .13 .36 — — — — — — — — — — — — — — — .97 .004 — — — — — — .004 — — — .001 .003 .20 .77
the ED, primary care (on site, off site, and home-based primary care), and various outpatient clinics. Whereas most antibiotic prescriptions requiring intervention were from the ED (42.3%), a large portion was also from primary care (38.7%). This finding demonstrates the need for culture review services in a variety of outpatient settings. Although literature evaluating pharmacist led culture review services is limited, a few studies evaluate the impact such services have on patient outcomes. Only one study looked at re-presentation rates to the ED.6 Representation rates decreased from 19%-7% in patients treated in the pharmacist led culture review are, compared to a physician led culture review. Our study is unique in that it evaluates outcomes of a culture review service in general as no prior service was in place at our facility. Also unique to our study, we evaluated rates of 30-day admission to the hospital related to original presentation. This endpoint is especially important as it can significantly impact health care systems because this is where the most health care dollars can be saved. Outpatient stewardship literature is scant regarding the impact of interventions. In an article by Suda et al,11 they noted a lack of literature evaluating appropriate antibiotic selection and the impact of stewardship interventions in the outpatient setting. Patients had improved outcomes in our subset of patients where stewardship recommendations were accepted. Patients with accepted recommendations had lower rates of 30-day treatment failure (5% vs 28%) and 30-day admission (0.7% vs 11%). This study had several limitations. One limitation was that this study was based on chart review and therefore availability of information was limited to documentation in the electronic medical record. Another limitation was the outpatient nature of this study which limited follow up information. Compliance could not be assessed, and if patients presented to an outside facility for further care it was largely unknown. Additionally, because this study involved review of cultures, time to intervention was limited by the speed of microbiology lab work up of cultures. Last, external validity of this study may be limited considering this was conducted at a Veteran’s Affairs health care system in New York State, with a primarily older white male population.
UTI, urinary tract infection.
CONCLUSIONS Table 3 Accepted vs rejected interventions: Outcomes
30-d treatment failure 30-d-related admission 30-d representation 30-d all-cause mortality
Accepted (n = 148, 76.3%)
Rejected (n = 46, 23.7%)
P value
8 (5.4%) 1 (0.7%) 6 (4.1%) 1 (0.7%)
13 (28.3%) 5 (10.9%) 4 (8.7%) 1 (2.2%)
<.001 .001 .21 .38
DISCUSSION Approximately 20% of antibiotic orders with related cultures required intervention in this study. Rates of pharmacist intervention vary among the literature, ranging from 5.1%-49.3%.2-6 One possible reason for this variation is the range in sample sizes. Variation in local practices could also be responsible for the differences noted. One study by Dumkow et al2 was similar to our study with 22% of cultures requiring intervention.4 Another consisted of a study cohort of only 73 patients. Much of the literature evaluating culture review services takes place in EDs or urgent care facilities.2-6 Our study evaluated a culture review service in a variety of outpatient locations including
Pharmacist-led microbiology review and interventions positively impacted outcomes for patients in the outpatient setting. For those antibiotic orders that required intervention, pharmacist intervention was associated with significantly decreased rates of 30-day treatment failure and 30-day admission when interventions were accepted. Culture review services can improve patient care by closing a gap in follow up between various levels of care in the outpatient setting. Implementing a culture review service can decrease rates of re-presentation to the health care system, thereby decreasing health care utilization and saving health care dollars. References 1. Centers for Disease Control and Prevention. National action plan for combating antibiotic-resistant bacteria; 2015. Available from: https://www.cdc.gov/drugresistance/pdf/national_action_plan_for_combating_antibotic-resistant_bacteria. pdf. Accessed March 8, 2019. 2. Baker SN, Acquisto NM, Ashley ED, Fairbanks RJ, Beamish SE, Haas CE. Pharmacistmanaged antimicrobial stewardship program for patients discharged from the emergency department. J Pharm Pract 2012;25:190-4. 3. Davis LC, Covey RB, Weston JS, Hu BB, Laine GA. Pharmacist-driven antimicrobial optimization in the emergency department. Am J Health Syst Pharm 2016;73 (Suppl 1):49-56.
ARTICLE IN PRESS B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5 4. Dumkow LE, Beuschel TS, Brandt KL. Expanding antimicrobial stewardship to urgent care centers through a pharmacist-led culture follow-up program. Infect Dis Ther 2017;6:453-9. 5. Santiago RD, Bazan JA, Brown NV, Adkins EJ, Shirk MB. Evaluation of pharmacist impact on culture review process for patients discharged from the emergency department. Hosp Pharm 2016;51:738-43. 6. Randolph TC, Parker A, Meyer L, Zeina R. Effect of a pharmacist-managed culture review process on antimicrobial therapy in an emergency department. Am J Health Syst Pharm 2011;68:916-9. 7. Dellit TH, Owens RC, McGowan JE Jr., Gerding DN, Weinstein RA, Burke JP, et al. 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;44:159-77.
5
8. Fleming-Dutra KE, Hersh AL, Shapiro DJ, Bartoces M, Enns EA, File TM Jr., et al. Prevalence of inappropriate antibiotic prescriptions among US ambulatory care visits, 2010-2011. JAMA 2016;315:1864-73. 9. Centers for Disease Control and Prevention. Outpatient antibiotic prescriptions — United States, 2016. Available from: https://www.cdc.gov/antibiotic-use/community/programs-measurement/state-local-activities/outpatient-antibiotic-prescriptions-US-2016.html. Accessed March 8, 2019. 10. Russo T, Johnson J. Harrison’s principles of internal medicine. Vol 2. 19th ed. New York (NY): McGraw Hill; 2015. 11. Suda KJ, Livorsi DJ, Goto M, Forrest GN, Jones MM, Neuhauser MM, et al. Research Agenda for antimicrobial stewardship in the Veterans Health Administration. Infect Control Hosp Epidemiol 2018;39:196-201.
Contents lists available at ScienceDirect
American Journal of Infection Control journal homepage: www.ajicjournal.org
Major Article
Outpatient antimicrobial stewardship: Optimizing patient care via pharmacist led microbiology review Bethany A. Wattengel PharmD, BCPS a, John A. Sellick DO, MS b, Kari A. Mergenhagen PharmD, BCPS, BCIDP a,* a b
Department of Pharmacy, Veteran’s Affairs Western New York Healthcare System, Buffalo, NY Department of Infectious Diseases, Veteran’s Affairs Western New York Healthcare System, Buffalo, NY
Key Words: Culture Antibiotic
Background: Cultures are often taken in the outpatient setting but results are not acted upon, leading to unnecessary re-presentations to the health care setting. Methods: This study was a prospective study with interventions made between January 1, 2018, and January 1, 2019. Cultures were reviewed to ensure appropriate antimicrobial coverage. The objective was to compare outcomes with accepted versus rejected interventions. Descriptive statistics were used to summarize data. Results: A total of 7,360 antibiotic orders were reviewed by the infectious diseases pharmacists. Pharmacists intervened on 20.1% (n = 194) of encounters with related cultures. Interventions were most frequent in the emergency department (42%). Ciprofloxacin required the most interventions (26%), followed by third-generation cephalosporins (22%). The intervention acceptance rate was 76%, which was associated with decreased rates of 30-day treatment failure (5% vs 28%, P < .001) and 30-day admission (0.7% vs 11%, P = .001), when interventions were accepted rather than rejected. Discussion: Approximately 20% of patients required intervention. Culture review services may be beneficial in a variety of outpatient settings. Outpatient stewardship literature is limited, and our study found a decrease in admission and treatment failure. Conclusions: Microbiology review and intervention positively impacted care for outpatients. Intervention was associated with significantly decreased rates of treatment failure and admission when interventions were accepted. Published by Elsevier Inc. on behalf of Association for Professionals in Infection Control and Epidemiology, Inc.
Antimicrobial stewardship programs are well established in the inpatient setting; however, progress has lagged in the outpatient setting. Recent development of outpatient programs can partially be attributed to the publishing of the White House’s national action plan for combating antibiotic-resistant bacteria, which called for a 50% decrease of inappropriate antibiotic use in the outpatient setting by 2020.1 Although no longer supported by the administration in the United States, this plan had spurred the development of outpatient antimicrobial stewardship programs including services, which review microbiology cultures. Outpatient use of antibiotics is often overlooked due to relatively low drug costs; however, it is an important cause of antimicrobial usage. Outpatient stewardship services are in their infancy. A targeted approach to outpatient antimicrobial stewardship includes review * Address correspondence to Kari A. Mergenhagen, PharmD, BCPS, BCIDP, Veteran’s Affairs Western New York Healthcare System, Pharmacy Department, 119, 3495 Bailey Ave, Buffalo, NY 14215 E-mail address: [email protected] (K.A. Mergenhagen). Conflicts of interest: None to report.
of microbiologic cultures to ensure appropriate follow through. Although studies are limited, most of the available literature pertains to pharmacist-led culture review services.2-6 According to the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA), pharmacists play a crucial role due to their knowledge of appropriate antimicrobial use, dosing, duration of therapy, and drug interactions.7 With approximately 270.2 million outpatient antibiotic prescriptions prescribed in the United States in 2016, and at least 30% judged to be inappropriate, outpatient stewardship is key, and pharmacists are poised to play an integral role.8,9 Most of the literature available regarding culture review services takes place in the emergency departments (EDs) of hospitals or urgent care centers.2-6 Such literature focuses on outcomes such as time to culture review and/or time to provider notification.2,5 Studies also evaluated the quantity of interventions made by pharmacists and compared it with that of other health care professionals such as nurses and physicians.5,6 Limited information, however, is available on the impact of such interventions on patient outcomes.6
https://doi.org/10.1016/j.ajic.2019.07.018 0196-6553/Published by Elsevier Inc. on behalf of Association for Professionals in Infection Control and Epidemiology, Inc.
ARTICLE IN PRESS 2
B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5
With a growing need for outpatient stewardship, data are needed to guide the development of new services to improve patient care. Often cultures are obtained in the outpatient setting but results are not acted upon, leading to unnecessary re-presentations to the health care setting. This study evaluates the utility of a pharmacist-led culture review service in the outpatient setting. Rates of pharmacist intervention will be assessed, as well as the impact of such interventions, including re-presentation to an outpatient location (eg, ED, primary care, and specialty clinic), subsequent admission to the hospital, and treatment failure requiring additional antibiotic therapy. METHODS Study design This study was a prospective chart review at the Veterans Affairs Western New York Healthcare System via the computerized patient record system with interventions made as needed. The study period was January 1, 2018, through January 1, 2019. The infectious diseases pharmacists received alerts when oral antibiotics for outpatient use were ordered. Special attention was paid to those antibiotics ordered in relation to cultures, thus forming the culture review service. Orders were reviewed no later than 12 hours post time of antibiotic order. Cultures were reviewed daily to ensure appropriate antimicrobial coverage and timely interventions. All cultures were taken before the antibiotic was dispensed. Antibiotic regimens were also evaluated for appropriateness in terms of dose, duration, antibiotic allergies, drug interactions, and QTc interval. Interventions therefore included changes to antibiotic regimens (eg, antibiotic choice, dose, and duration), diagnostic recommendations (cultures and tests), and patient counseling. Interventions were made via direct contact with providers (eg, phone, instant message, and electronic medical record notes with provider attachment for acknowledgment) and documented via notes in the electronic medical record. Providers entered their own orders based on recommendations from the stewardship team. The infectious diseases physician was available to review patients as needed. This study was approved as a quality assurance/ quality improvement project. Patient population Patients were included if they were 18 years of age and older, were prescribed an antibiotic in an outpatient setting, and had a related culture ordered. Patients were excluded if they were under the age of 18 or did not have a culture obtained in relation to an outpatient antibiotic order. Patients were also excluded if they were admitted to the hospital immediately following outpatient antibiotic order or culture. “Outpatient setting” included ED, primary care (on site, off site, and home based primary care), and outpatient clinics. Outcomes The primary objective of this study was to compare outcomes in patients with accepted interventions vs rejected interventions: 30-day all-cause mortality, 30-day representation rates, 30-day admission rates, and 30-day treatment failure. Thirty-day treatment failure was a composite outcome which included patient who represented to an outpatient location or were admitted within 30 days for additional antibiotic therapy. This study also evaluated rates of appropriate antibiotic based on culture results, location of patient presentation, type of culture, and number and type of interventions made.
Definitions A drug-bug mismatch was defined as discordance between prescribed antibiotic therapy and culture susceptibility report. A drugbug mismatch also included cases of bacteria with known resistance mechanisms to prescribed antibiotic therapy. Such cases included patients with cultures positive for SPICE organisms (Serratia, Providencia, indole positive Proteus, Citrobacter, and Enterobacter) who were prescribed a cephalosporin when viable alternatives existed due to concern for inducible resistance via the AmpC gene.10 Last, if a culture was negative for growth and antibiotic therapy was prescribed, this was considered a drug-bug mismatch. A drug-bug match was defined as concordance between prescribed antibiotic therapy and culture susceptibility report. In cases where susceptibilities were not reported, cases were considered to be a drug-bug match if the organism report is predictably susceptible to prescribed antibiotic therapy. Accepted intervention occurred when recommendations to change a component of the antibiotic regimen (antibiotic choice, dose, or duration), discontinue an antibiotic, add an antibiotic, or order diagnostic tests or cultures were acted upon in accordance with the stewardship team’s recommendations. Rejected intervention occurred when recommendations to change a component of the antibiotic regimen (antibiotic choice, dose, or duration), discontinue an antibiotic, add an antibiotic, or order diagnostic tests or cultures were declined by providers. Statistical analysis Descriptive statistics were generated using JMP Pro software (version 14; SAS Institute Inc, Cary, NC). Data were presented as a mean and SD when normally distributed. A bivariate analysis was used to compare outcomes in patients with accepted interventions versus rejected interventions. RESULTS A total of 7,360 antibiotic orders were reviewed by the infectious diseases clinical pharmacists, including 965 encounters with related cultures (Fig 1). Time spent reviewing electronic medical records and antibiotic orders varied. Pharmacist review took as little as 5 minutes for some orders without cultures to as many as 30 minutes for those with cultures that required intervention. Pharmacists intervened in 20.1% (n = 194) of patient encounters with related cultures. Results of this study specifically focus on the 194 patient encounters that required pharmacist intervention with regard to the culture review service. The majority of antibiotic prescriptions that required intervention were from the ED (42.3%) and primary care (38.7%), with the remaining 19% being from various outpatient specialty clinics (Table 1). Most patients were older white males, with an average age of approximately 70 years. Antibiotic use varied, however, the most common antibiotics prescribed for patients requiring intervention were ciprofloxacin (26.3%), third-generation cephalosporins (22.2%), and sulfamethoxazole/trimethoprim (18%) (Table 2). The most frequent indication for use was urinary tract infection (71.1%) and therefore urine was the most common type of culture reviewed. An average of 3 days was required for the microbiology laboratory to finalize cultures. Thirtyone percent of cultures grew gram-negative organisms, 25.3% gram positive, 18.6% were mixed with both gram negatives and gram positives, and 18.6% of cultures were negative for growth. Patients were treated with appropriate antibiotic therapy in accordance with culture results in 22.2% of cases; there was drug-bug mismatch in 72.7% of cases. Approximately 5% (n = 10) of the cultures contained results
ARTICLE IN PRESS B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5
3
Fig 1. Inclusion and exclusion criteria.
that could not confirm whether prescribed therapy provided appropriate coverage due to lack of complete microbiological work up. Of the patients with confirmed drug-bug match or mismatch (n = 184), the highest degree of drug-bug mismatch occurred in gram negative cultures (29.1%). When comparing drug-bug mismatch to drug-bug match, mismatch occurred more often in mixed cultures with both gram negative and gram positive organisms (20.6% mismatch vs 14.0% match) and negative cultures without bacterial growth that did not require antibiotic therapy (25.5% mismatch vs 0% match) (P = .0014). A total of 36 cultures were mixed,
containing both gram positives and negatives. Patients with mixed culture results were prescribed an antibiotic that did not provide adequate coverage of culture results 80.6% (n = 29) of the time; however, only 58.3% of recommended interventions were accepted. In the majority of these cases, neither bacterium (gram positive or negative) was covered appropriately by the prescribed antibiotic. Patients with mixed cultures were more likely to experience 30-day treatment failure when interventions were rejected rather than accepted (75.0% vs 25.0%, P = .030). Accepted vs rejected interventions
Table 1 Accepted vs rejected interventions: demographic information
Location ED Primary care Specialty Location Main Rural Age Body mass index Creatinine clearance Male Race African American White Other Antibiotic allergy Beta-lactam allergy ED, emergency department.
Accepted (n = 148, 76.3%)
Rejected (n = 46, 23.7%)
65 (43.9%) 55 (37.2%) 28 (18.9%)
17 (37.0%) 19 (41.3%) 10 (21.7%)
130 (87.8%) 18 (12.2%) 70.8 +/- 15.2 27.7 +/- 8.1 75.2 +/- 40.1 127 (85.8%)
38 (82.6%) 8 (17.4%) 70.5 +/- 18.6 28.5 +/- 7.8 80.8 +/- 54.6 42 (91.3%)
34 (23.0%) 111 (75.0%) 3 (2.0%) 39 (26.4%) 26 (17.6%)
7 (15.2%) 37 (80.4%) 2 (4.4%) 7 (15.2%) 4 (8.7%)
P value
.70 — — — .36 — — .90 .54 .46 .33 .39 — — — .12 .15
A total of 194 encounters required interventions, with an acceptance rate of 76% (n = 148) (Table 2). Antibiotic therapy was not indicated in 37.6% of encounters that required intervention. Interventions were more likely to be accepted when antibiotic therapy was indicated (68.9% vs 41.3%, P = .007). It should be noted that multiple interventions were made on some encounters; and that a total of 229 interventions were made. The majority of interventions (73.8%, n = 169) involved a change to the prescribed antibiotic regimen. Of the interventions with a change to the regimen, 45.6% (n = 77) were recommendations to change the antibiotic, 38.5% (n = 65) were recommendations for discontinuation of therapy, 9.5% (n = 16) were recommendations to add or initiate antibiotic therapy, 4.1% (n = 7) were dose changes, and 2.4% (n = 4) were duration changes. The encounters where an intervention was rejected all involved a change of the antibiotic regimen. Pharmacist intervention was associated with significantly decreased rates of 30-day treatment failure (5.4% vs 28.3%, P < .0001) and 30-day admission (0.7% vs 10.9%, p = .0005) when interventions were accepted rather than rejected. There was no difference in 30-day mortality (P = .38) (Table 3).
ARTICLE IN PRESS B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5
4
Table 2 Accepted vs rejected interventions: Antibiotic regimen, culture, and intervention data
Antibiotic class Beta-lactam Fluoroquinolone Sulfamethoxazole/trimethoprim Macrolide Miscellaneous Nitrofurantoin Antibiotic duration (d) Indication UTI Skin and soft tissue Epididymoorchitis or prostatitis Pharyngitis Status post procedure Abscess Pre-op urology Helicobacter pylori Clostridium difficile Pneumonia Urethral injury UTI prophylaxis Osteomyelitis Otitis Media Vaginitis Time to culture finalization (d) Culture results Gram negative Gram positive Mixed gram negative and positive Negative culture Sexually transmitted infection Other Drug-bug match Yes No Unknown Antibiotic indicated Yes No Intervention (change in regimen) Intervention (education/counseling) Intervention (culture/tests)
Accepted (n = 148, 76.3%)
Rejected (n = 46, 23.7%)
67 (45.3%) 45 (30.4%) 22 (14.9%) 2 (1.4%) 7 (4.7%) 5 (3.4%) 8.4 +/- 3.4
19 (41.3%) 13 (28.3%) 13 (28.3%) 0 0 1 (2.2%) 9.9 +/- 9.1
106 (71.6%) 14 (9.5%) 5 (3.4%) 5 (3.4%) 4 (2.7%) 4 (2.7%) 3 (2.0%) 2 (1.4%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 0 0 2.8 +/- 1.8
32 (69.6%) 4 (8.7%) 2 (4.4%) 0 2 (4.4%) 0 2 (4.4%) 0 0 0 0 2 (4.4%) 0 1 (2.1%) 1 (2.2%) 2.8 +/- 1.3
51 (34.5%) 44 (29.7%) 21 (14.2) 23 (15.5%) 2 (1.4%) 7 (4.7%)
9 (19.6%) 5 (10.9%) 15 (32.6%) 13 (28.3%) 1 (2.2%) 3 (6.5%)
38 (25.7%) 106 (71.6%) 4 (2.7%)
5 (10.9%) 35 (76.1%) 6 (13.0)
102 (68.9%) 46 (31.1%) 123 (83.1%) 28 (18.9%) 20 (13.5%)
19 (41.3%) 27 (58.7%) 46 (100%) 5 (10.9%) 7 (15.2%)
P value
.25 — — — — — — .13 .36 — — — — — — — — — — — — — — — .97 .004 — — — — — — .004 — — — .001 .003 .20 .77
the ED, primary care (on site, off site, and home-based primary care), and various outpatient clinics. Whereas most antibiotic prescriptions requiring intervention were from the ED (42.3%), a large portion was also from primary care (38.7%). This finding demonstrates the need for culture review services in a variety of outpatient settings. Although literature evaluating pharmacist led culture review services is limited, a few studies evaluate the impact such services have on patient outcomes. Only one study looked at re-presentation rates to the ED.6 Representation rates decreased from 19%-7% in patients treated in the pharmacist led culture review are, compared to a physician led culture review. Our study is unique in that it evaluates outcomes of a culture review service in general as no prior service was in place at our facility. Also unique to our study, we evaluated rates of 30-day admission to the hospital related to original presentation. This endpoint is especially important as it can significantly impact health care systems because this is where the most health care dollars can be saved. Outpatient stewardship literature is scant regarding the impact of interventions. In an article by Suda et al,11 they noted a lack of literature evaluating appropriate antibiotic selection and the impact of stewardship interventions in the outpatient setting. Patients had improved outcomes in our subset of patients where stewardship recommendations were accepted. Patients with accepted recommendations had lower rates of 30-day treatment failure (5% vs 28%) and 30-day admission (0.7% vs 11%). This study had several limitations. One limitation was that this study was based on chart review and therefore availability of information was limited to documentation in the electronic medical record. Another limitation was the outpatient nature of this study which limited follow up information. Compliance could not be assessed, and if patients presented to an outside facility for further care it was largely unknown. Additionally, because this study involved review of cultures, time to intervention was limited by the speed of microbiology lab work up of cultures. Last, external validity of this study may be limited considering this was conducted at a Veteran’s Affairs health care system in New York State, with a primarily older white male population.
UTI, urinary tract infection.
CONCLUSIONS Table 3 Accepted vs rejected interventions: Outcomes
30-d treatment failure 30-d-related admission 30-d representation 30-d all-cause mortality
Accepted (n = 148, 76.3%)
Rejected (n = 46, 23.7%)
P value
8 (5.4%) 1 (0.7%) 6 (4.1%) 1 (0.7%)
13 (28.3%) 5 (10.9%) 4 (8.7%) 1 (2.2%)
<.001 .001 .21 .38
DISCUSSION Approximately 20% of antibiotic orders with related cultures required intervention in this study. Rates of pharmacist intervention vary among the literature, ranging from 5.1%-49.3%.2-6 One possible reason for this variation is the range in sample sizes. Variation in local practices could also be responsible for the differences noted. One study by Dumkow et al2 was similar to our study with 22% of cultures requiring intervention.4 Another consisted of a study cohort of only 73 patients. Much of the literature evaluating culture review services takes place in EDs or urgent care facilities.2-6 Our study evaluated a culture review service in a variety of outpatient locations including
Pharmacist-led microbiology review and interventions positively impacted outcomes for patients in the outpatient setting. For those antibiotic orders that required intervention, pharmacist intervention was associated with significantly decreased rates of 30-day treatment failure and 30-day admission when interventions were accepted. Culture review services can improve patient care by closing a gap in follow up between various levels of care in the outpatient setting. Implementing a culture review service can decrease rates of re-presentation to the health care system, thereby decreasing health care utilization and saving health care dollars. References 1. Centers for Disease Control and Prevention. National action plan for combating antibiotic-resistant bacteria; 2015. Available from: https://www.cdc.gov/drugresistance/pdf/national_action_plan_for_combating_antibotic-resistant_bacteria. pdf. Accessed March 8, 2019. 2. Baker SN, Acquisto NM, Ashley ED, Fairbanks RJ, Beamish SE, Haas CE. Pharmacistmanaged antimicrobial stewardship program for patients discharged from the emergency department. J Pharm Pract 2012;25:190-4. 3. Davis LC, Covey RB, Weston JS, Hu BB, Laine GA. Pharmacist-driven antimicrobial optimization in the emergency department. Am J Health Syst Pharm 2016;73 (Suppl 1):49-56.
ARTICLE IN PRESS B.A. Wattengel et al. / American Journal of Infection Control 00 (2019) 1−5 4. Dumkow LE, Beuschel TS, Brandt KL. Expanding antimicrobial stewardship to urgent care centers through a pharmacist-led culture follow-up program. Infect Dis Ther 2017;6:453-9. 5. Santiago RD, Bazan JA, Brown NV, Adkins EJ, Shirk MB. Evaluation of pharmacist impact on culture review process for patients discharged from the emergency department. Hosp Pharm 2016;51:738-43. 6. Randolph TC, Parker A, Meyer L, Zeina R. Effect of a pharmacist-managed culture review process on antimicrobial therapy in an emergency department. Am J Health Syst Pharm 2011;68:916-9. 7. Dellit TH, Owens RC, McGowan JE Jr., Gerding DN, Weinstein RA, Burke JP, et al. 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;44:159-77.
5
8. Fleming-Dutra KE, Hersh AL, Shapiro DJ, Bartoces M, Enns EA, File TM Jr., et al. Prevalence of inappropriate antibiotic prescriptions among US ambulatory care visits, 2010-2011. JAMA 2016;315:1864-73. 9. Centers for Disease Control and Prevention. Outpatient antibiotic prescriptions — United States, 2016. Available from: https://www.cdc.gov/antibiotic-use/community/programs-measurement/state-local-activities/outpatient-antibiotic-prescriptions-US-2016.html. Accessed March 8, 2019. 10. Russo T, Johnson J. Harrison’s principles of internal medicine. Vol 2. 19th ed. New York (NY): McGraw Hill; 2015. 11. Suda KJ, Livorsi DJ, Goto M, Forrest GN, Jones MM, Neuhauser MM, et al. Research Agenda for antimicrobial stewardship in the Veterans Health Administration. Infect Control Hosp Epidemiol 2018;39:196-201.