- Email: [email protected]
Impact of Antimicrobial Stewardship on Outcomes in Hospitalized Veterans With Pneumonia
Clinical Therapeutics/Volume 38, Number 7, 2016
Impact of Antimicrobial Stewardship on Outcomes in Hospitalized Veterans With Pneumonia Kari E. Kurtzhalts, PharmD1; John A. Sellick Jr, DO, MS1,2; Christine A. Ruh, PharmD, BCPS4; James F. Carbo, PharmD1,3; Michael C. Ott, PharmD, BCPS4; and Kari A. Mergenhagen, PharmD, BCPS AQ-ID1 1
Infectious Diseases Department, Veterans Affairs Western New York Healthcare System, Buffalo, New York; 2University at Buffalo School of Medicine, The State University of New York, Buffalo, New York; 3 Veterans Affairs Southern Nevada Healthcare System, North Las Vegas, Nevada; and 4Pharmacy Department, Erie County Medical Center, Buffalo, New York ABSTRACT Purpose: The purpose of this study was to evaluate the impact of an antimicrobial stewardship program (ASP) on outcomes for inpatients with pneumonia, including length of stay, treatment duration, and 30day readmission rates. Methods: A retrospective chart review comparing outcomes of veterans admitted with pneumonia before (2005–2006) and after (2013–2014) implementation of an ASP was conducted; pneumonia was defined according to International Classification of Diseases, Ninth Revision (ICD-9) codes. Infectious diseases physicians and pharmacist in the ASP provided appropriate recommendations to the primary medicine teams. Bivariate analysis of baseline characteristics and comorbid conditions were performed between the time frames. Least squares regression was used to analyze length of stay, time of IV to PO conversions, and duration of antibiotics. Multivariate logistic regressions were used to determine odds of 30-day readmission and odds of Clostridium difficile infections between time periods. Findings: There were 86 patients in the pre-ASP period and 88 patients in the ASP period. Mean length of stay decreased from 8.1 to 6.6 days (P ¼ 0.02), total duration of antibiotic therapy decreased from 12 to 8.5 days (P o 0.0001), and time of IV to PO antibiotic conversions decreased from 5.3 to 3.9 days (P ¼ 0.0003), before ASP and during ASP, respectively. The odds ratio of 30-day readmission before ASP was 2.78 and 0.36 during the ASP (P ¼ 0.05). The odds ratios of Clostridium difficile infections before ASP was 2.08 and 0.48 during the ASP (P ¼ 0.37).
1750
Implications: The ASP interventions were associated with shorter durations of therapy, shorter lengths of stay, and lower rates of readmission and Clostridium difficile infections within 30 days. Limitations of this study are retrospective cohort design, small study population, limited study population diversity, and non-concurrent cohort times periods. (Clin Ther. 2016;38:1750–1758) Published by Elsevier HS Journals, Inc. Key words: antimicrobial stewardship, patientcentered, pneumonia, readmission.
INTRODUCTION Pneumonia is one of the leading causes of death in the United States, accounting for 57,000 deaths in 2013.1,2 Worsening antibiotic resistance and multidrug resistant (MDR) organisms may be linked to death and poor treatment outcomes.3–9 Health care exposure, inappropriate use of antibiotics at institutions, and overprescribing of antibiotics in the community are some factors associated with development of MDR organisms and antibiotic resistance.10–14 Antibiotic resistance can result in other negative outcomes such as increased length of hospital stay (LOS), prolonged treatment duration, and development of Clostridium difficile infections.15–17 The use of antimicrobial stewardship programs (ASPs) in health care facilities has been encouraged Accepted for publication June 1, 2016. http://dx.doi.org/10.1016/j.clinthera.2016.06.004 0149-2918/$ - see front matter Published by Elsevier HS Journals, Inc.
Volume 38 Number 7
K.E. Kurtzhalts et al. as an important tool in improving and preserving antibiotic susceptibilities, along with improving patient care and health care outcomes.18 In 2007 the Infectious Disease Society of America (IDSA) and the Society of Healthcare Epidemiology of America published guidelines that laid foundations to aid institutions in developing ASPs.18 The IDSA updated these guidelines in 2016 with recommendations for the implementation and measurement of the success of these ASPs.19 These guidelines recommend that antimicrobial stewardship teams consist of collaboration between infectious disease (ID) physicians and ID-specialized clinical pharmacists. Since then, ASPs have emerged in facilities across the country and have improved antimicrobial utilization, decreased treatment durations, and reduced treatment costs.20–22 In 2009 the Veteran’s Affairs Healthcare System in Buffalo, New York, began their ASP team with an ID-specialized clinical pharmacist and a team of ID physicians. This ASP uses a patientcentered approach. Automated protocols are not used, and the ID pharmacist performs daily chart reviews for each patient on IV antibiotics. The goal of this study was to address the impact our ASP has had on the outcomes of pneumonia treatment, including LOS, duration of antibiotic treatment, time of IV to PO antibiotic conversion, 30-day hospital readmissions, and occurrence of C difficile infections. Interventions of the ASP were characterized as secondary objectives.
PATIENTS AND METHODS This single-centered retrospective cohort study took place at the Veterans Affairs Western New York Healthcare System (VAWNYHCS), which is a 150bed level 1b facility. The institutional review board approved this study. Patient information was collected for patients who were hospitalized and received antibiotics for pneumonia during a 1-year period before the facility’s ASP was initiated and during a 1-year period after the program had been well established. The pre-ASP group consisted of the period of time between June 1, 2005 and June 1, 2006, and between June 1, 2013, and June 1, 2014, for the ASP group. Patients must have had a primary diagnosis of pneumonia, based on International Classification of Diseases, Ninth Revision (ICD-9) codes (507, 486, 518, 381, 482, 481, 038.9) to be included in this study; both general ward and patients in the intensive
July 2016
care unit were included. Exclusion criteria included death on discharge after being placed on comfort care measures, o48 hours of inpatient admission or inpatient antibiotics, lung abscess, empyema, necrotizing pneumonia, having a thoracentesis, severe immunosuppression, concurrent extra-pulmonary source of infection (such as endocarditis, urinary tract infection, meningitis, etc), and if there was no documented duration of therapy (ie, lost to follow-up). The VAWNYHCS is a vertical health care system, providing both outpatient and inpatient care to patients, which allowed us to accurately follow treatment both while the patients were hospitalized and after discharge.
Definitions Patients were considered to be immunosuppressed if they had systemic chemotherapy within 28 days of admission, HIV infection/AIDS, solid organ or bone marrow transplant, or absolute neutrophil count of o1500 cells within 28 days of admission. Morbidity was calculated using the Charlson Comorbidity Score.23 Severity of illness was calculated by using the CURB-65 (confusion, uremia, elevated respiratory rate, low blood pressure, and age Z 65 years) score. The CURB-65 score was used to determine level of illness severity on admission for patients with community-acquired pneumonia (CAP) and for health care-associated pneumonia (HCAP).24,25 A patient was considered to have HCAP if he or she met the following criteria on admission: either acute hospitalization for 42 days or long-term care facility residence within 90 days of admission, or any of the following within 28 days of admission: chronic hemodialysis, antibiotics, wound care, tracheotomy, ventilator, or aspiration.26
Antimicrobial Stewardship The antimicrobial stewardship team consists of ID physicians and a clinical ID pharmacist with added qualifications in IDs. The ASP is patient centered, using individualized patient care by daily chart reviews. Select antibiotics were restricted, requiring approval by the ID team for use. These restrictions were in place before the ASP was established. Each morning a report is printed of all hospitalized patients on IV antibiotics, and the ASP then reviews each patient’s chart to assess appropriateness of antibiotic therapy. Interventions consist of prospective audit and feedback regarding antibiotic selection, microbiology,
1751
Clinical Therapeutics and duration of therapy. Care of each patient is individualized, based on prior history and response to treatment. In addition, brief monthly educational conferences on antimicrobial stewardship and local antimicrobial resistance patterns are provided to the house staff to underscore the importance of microbial cultures and to promote appropriate use of antimicrobial agents. Specialized antibiograms are created by the ASP and used for education.19 These antibiograms consist of a hospital-wide antibiogram, intensive carespecific antibiogram, and specific antibiograms for respiratory, urinary, and blood. In addition, a Pseudomonas aeruginosa cross-antibiogram is created annually. The ASP pharmacist has a scope of practice that includes the ability to change antibiotic dosing as needed. Antibiotic changes, including choice of antibiotic, are reviewed with the physician of record before implementation; dosing changes can be made independently. The pharmacist can also order other diagnostic and laboratory tests such as drug levels, cultures, complete metabolic panel, complete blood count, etc. Generally, urine, blood, and sputum cultures can be ordered by the pharmacist; however, the physician of record orders other invasive cultures, such as induced sputum. The pharmacist operates with ID physician support. There is 1 full-time equivalent (FTE) pharmacist, 1 FTE pharmacy resident, and 0.1 FTE physician dedicated to the ASP. The VA computerized electronic medical record (EMR) system enables information to be easily gathered. Notes, laboratory values, cultures, vital signs, and electrocardiograms are all readily available in this system. The computerized EMR system did not have electronic clinical decision support or automated protocols for pneumonia at the time of prescribing, but instead relied on real-time interventions by the ASP team. The VA provides ongoing inpatient and outpatient care for patients, thus enabling the ASP to use past cultures and antibiotic treatment to tailor a regimen specific for that patient. The ASP team visits the microbiology laboratory daily to obtain interim culture data, which may enable antimicrobial therapy optimization before cultures are finalized. The only rapid diagnostic test available is polymerase chain reaction (PCR) for C difficile, starting in 2011. Procalcitonin and other rapid diagnostics are not used at our institution.
1752
Outcomes The primary objective for the study was to evaluate the impact of an ASP on pneumonia treatment outcomes in hospitalized patients. Primary outcomes were LOS, duration of treatment, time from IV to PO antibiotic conversion, 30-day readmission rates, and C difficile infections occurring within 30 days of antibiotic initiation. Secondary outcomes included the number of antibiotic therapies narrowed, number of blood and sputum cultures obtained, and number of therapy changes made during treatment. The patientcentric ASP was characterized by using interventions such as dose optimization, ordering cultures, ordering diagnostic laboratory tests, therapy escalation and deescalation, IV to PO antibiotic changes, and adjustments in antibiotics due to QTc prolongation.
Statistical Analysis Bivariate analysis was performed on baseline characteristics between the pre-ASP and ASP groups and for secondary outcomes, number of antibiotics streamlined, number of blood and sputum cultures obtained, and number of therapy changes made during treatment. Ordinary least squares regression was used to analyze the following primary outcomes: LOS, duration of antibiotics, and time to convert from IV to PO therapy. Pre-ASP and ASP groups were included in the model a priori. In the least squares regression variables with significance (P o 0.05) were built into the model. Odds ratios (OR) and 95% CIs were calculated for 30-day readmissions and C difficile infections. Statistical tests were two tailed with statistical significance achieved with a P value r 0.05. ASP interventions were analyzed as percentages. Statistical analysis was performed with JMP software (Version 10; SAS, Cary, North Carolina).
RESULTS A total of 303 patient charts were reviewed using the VAWNYHCS EMR; 174 patients were evaluated and 129 patients were excluded (Figure 1). There were 86 patients in the pre-ASP group and 88 patients in the ASP group. No statistical significant differences were found in baseline characteristics between the 2 groups; the median ages were 75.04 years, most were white men, and with a median Charlson Comorbidity Score of 3.53 (Table I). In the ASP group more patients were treated for HCAP than CAP compared with the
Volume 38 Number 7
K.E. Kurtzhalts et al.
303 patient charts reviewed
174 patients included
86 Pre-ASP group
129 patients excluded
88 ASP group
Pneumonia not primary dx - 52 < 48 hrs of admission - 8 Death - 17 < 48 hrs of inpatient abx - 9 Empyema - 1 Thoracentesis performed- 8 Chemo w/in 20 days of admission - 11 HIV/AIDS - 2 Solid organ/bone marrow transplant - 3 Concurrent infection - 5 No documented therapy duration - 8 Aspiration/post - obstructive pna - 5
Figure 1. Patient enrollment. Abx ¼ antibiotics; ASP ¼ antimicrobial stewardship program; dx ¼ diagnosis; pna ¼ pneumonia.
pre-ASP group. Patients in the ASP group had a higher severity of illness than the pre-ASP group as measured by the CURB-65 score (P ¼ 0.032). Results of the multivariate analysis of LOS, duration of treatment, and time of IV to PO conversion are listed in Table II. With the use of a least squares regression model to adjust for CURB-65 score and type of pneumonia, LOS was decreased from 8.07 ⫾ 0.44 days (pre-ASP) to 6.68 ⫾ 0.40 days (ASP) (P ¼ 0.018). In this model LOS varied significantly with CURB-65 (P ¼ 0.018) and type of pneumonia (P o 0.0001). The mean LOS for a patient with CAP was 6.0 days compared with 8.87 days in patients with HCAP. In a second least squares regression model adjusted for CURB-65 score and type of pneumonia, duration of treatment was decreased from 11.93 ⫾ 0.45 days (pre-ASP) to 8.45 ⫾ 0.41 days (ASP) (P o 0.0001). In this model total duration of treatment did not vary significantly with CURB-65 score (P ¼ 0.056) or type of pneumonia (P ¼ 0.089). In a third
July 2016
least squares regression model adjusted for CURB-65 score and type of pneumonia, time of IV to PO conversion varied significantly between pre-ASP (5.32 ⫾ 0.3 days) and ASP (2.93 ⫾ 0.26 days) cohorts (P ¼ 0.0003). In this model time of IV to PO conversion was significantly affected by CURB-65 score (P ¼ 0.0085) but not by type of pneumonia (P ¼ 0.4). The OR of 30-day readmission in the pre-ASP group was 2.78 (95% CI, 0.99–8.29) compared with 0.36 (95% CI, 0.12–0.99) in the ASP group (P ¼ 0.05) (Table III), in this multivariate logistic regression. In this model, CURB-65 was not a significant predictor of readmission (P ¼ 0.98); however, patients with HCAP were more likely to be readmitted, (OR ¼ 7.97; 95% CI, 2.72–25.90) than patients with CAP (OR ¼ 0.13; 95% CI, 0.039–0.37). The OR of C difficile occurrence 30 days after antibiotic discontinuation was 2.08 (95% CI, 0.41–11.76) and 0.48 (95% CI, 0.085–2.42), pre-ASP and ASP groups,
1753
Clinical Therapeutics
Table I. Bivariate analysis of baseline characteristics. Cohort (n ¼ 174)
Pre-ASP (n ¼ 86)
ASP (n ¼ 88)
75.40 (11.26)
74.74 (11.58)
75.03 (10.97)
83.91 12.64 3.45 97.13 68.98 85.72 27.86 1.52 57.21 11.53 32.76 3.53 43.68 27.01 5.17 3.45 27.01 18.97 5.17 22.99 5.75 1.15 1.73 2.87 48.84 1.93
82.56 16.28 1.16 96.51 69.16 83.19 26.90 1.34 58.06 11.42 31.40 3.29 41.86 26.74 4.65 4.65 24.42 16.28 2.33 19.77 5.81 1.16 0 3.49 44.71 1.76
85.23 9.09 5.68 97.73 68.80 88.20 28.79 1.70 56.37 11.64 34.09 3.76 45.45 27.27 5.68 2.27 29.55 21.59 7.95 26.14 5.68 1.14 3.41 2.27 52.87 2.09
Variable Age, mean (SD) years Race, % (no.) White Black Other Sex, male, % (no.) Height, mean (SD), inches Weight, mean (SD), kg BMI, mean (SD), kg/m2 SCr, mean (SD), g/dL CrCl, mean (SD), mL/min WBC, mean (SD), 103/μL Diabetes, % (no.) Charlson score, mean (SD) COPD, % (no.) Heart failure, % (no.) Cirrhosis, % (no.) Anatomic abnormalities, % (no.) Acute hospitalization w/in 90 days prior, % (no.) Long-term care w/in 90 days, % (no.) Dialysis, % (no.) Antibiotics w/in 28 days, % (no.) Wound care, % (no.) Tracheostomy, % (no.) Ventilator, % (no.) Aspiration, % (no.) Risk, % (no.) CURB-65, mean (SD) Type of pneumonia, % (no.) CAP HCAP Service, % (no.) ICU Medical
(146) (22) (6) (169) (3.15) (26.59) (8.10) (1.67) (25.56) (5.59) (57) (2.16) (76) (47) (9) (6) (47) (33) (9) (40) (10) (2) (3) (5) (84) (1.03)
(71) (14) (1) (83) (3.09) (21.27) (6.41) (1.05) (24.52) (5.52) (27) (2.25) (36) (23) (4) (4) (21) (14) (2) (17) (5) (1) (0) (3) (38) (1.07)
(75) (8) (5) (86) (3.21) (30.84) (9.41) (2.10) (26.66) (5.69) (30) (2.06) (40) (24) (5) (2) (26) (19) (7) (23) (5) (1) (3) (2) (46) (0.97)
68.97 (120) 31.03 (54)
81.40 (70) 18.60 (16)
56.82 (50) 43.18 (38)
9.20 (16) 90.80 (158)
5.81 (5) 94.19 (81)
12.50 (11) 87.50 (77)
P 0.45 0.11
0.63 0.46 0.21 0.12 0.14 0.66 0.79 0.71 0.071 0.63 0.94 0.76 0.39 0.45 0.37 0.094 0.32 0.97 0.99 0.084 0.63 0.28 0.032 0.0005
0.13
ASP ¼ antimicrobial stewardship program; BMI ¼ body mass index; CAP ¼ community acquired pneumonia; COPD, chronic obstructive pulmonary disease; CrCl ¼ creatinine clearance; CURB-65 ¼ confusion, uremia, elevated respiratory rate, low blood pressure, and age Z 65 years; HCAP ¼ health care-associated pneumonia; ICU ¼ intensive care unit; SCr ¼ serum creatinine; WBC ¼ white blood cell; w/in ¼ within;
respectively (P ¼ 0.37) (Table III). CURB-65 score and type of pneumonia were not significant (P ¼ 0.58 and 0.16, respectively).
1754
Compared with the pre-ASP group, the ASP group had significantly more sputum cultures and blood cultures obtained (P ¼ 0.023). The number of antibiotic
Volume 38 Number 7
K.E. Kurtzhalts et al.
Table II. Least squares regression of primary outcomes. Variable Length of stay, days Duration of antibiotics, days IV to PO conversion, days*
Pre-ASP (n ¼ 86)
ASP (n ¼ 88)
P
8.07 (0.44) 11.93 (0.45) 5.32 ⫾ 0.30
6.68 (0.40) 8.45 (0.41) 3.92 ⫾ 0.26
0.018 o0.0001 0.0003
Values are least squares mean (SE). Outcome variables were adjusted for type of pneumonia (community acquired or healthcare associated) and CURB-65 (confusion, uremia, elevated respiratory rate, low blood pressure, and age Z 65 years) score. ASP ¼ antimicrobial stewardship program. * Pre-ASP: n ¼ 70; ASP: n ¼ 76.
therapies changed (P ¼ 0.009) and streamlined (P ¼ 0.004) were also significantly greater in the ASP group than in the non-ASP group (Table IV). These results were achieved through 130 recommendations made by the ASP team and accepted by patients’ primary teams, 1.5 recommendations per patient. Accepted recommendations were distributed among 7 categories; optimization of antibiotic dosing (39%), de-escalation of therapy (20%), obtain cultures (blood or sputum) (12%), IV to PO therapy (8.5%), order various diagnostic laboratory tests (4.6%), escalation of therapy (3.8%), and antibiotic change due to QTc prolongation (3%).
DISCUSSION Our study found that ASPs may have a significant impact on treatment outcomes in pneumonia. Our ASP was able to significantly decrease time to convert from IV to PO therapy, duration of treatment, and LOS while also decreasing 30-day readmissions. These outcomes have significant impacts on
improvement in patient care, by decreasing length of hospitalization and subsequent readmissions and also on the burdens of the health care system. Decreasing treatment duration can help prevent adverse drug events that may occur with excessive use of antibiotics. Earlier conversion from IV to PO antibiotics decreases risk of line infections, decreases pharmacy costs, and also eliminates a barrier to discharge.27 Decreasing LOS reduces the amount of health care resources needed per admission, increases bed availability within the facility, and improves patient outcomes by decreasing a patient’s exposure time to hospital-associated pathogens. Previous studies analyzing the impact of ASPs have reported similar results.21,22 Avdic et al22 studied 127 patients being treated for CAP and found that antimicrobial stewardship helped to reduce duration of therapy from 10 to 7 days; however, they found no difference in LOS between their pre-ASP and intervention groups. Vettese et al21 found that a thrice-weekly ASP decreased total antibiotic expendi-
Table III. Odds ratios of 30-day readmission and C difficile infections. Variable 30-day readmission* C difficile†
Pre-ASP
ASP
P
2.78 (0.99–8.29) 2.08 (0.41–11.76)
0.36 (0.12–1.00) 0.48 (0.085–2.42)
0.05 0.37
Values are odds ratio (95% CI). ASP ¼ antimicrobial stewardship program. * Pre-ASP: n ¼ 12, ASP: n ¼ 9 † Pre-ASP: n ¼ 4, ASP: n ¼ 3
July 2016
1755
Clinical Therapeutics
Table IV. Bivariate analysis of treatment outcomes. Variable Sputum cultures, % (no.) Blood cultures, % (no.) Therapy change, % (no.) Antibiotics narrowed, % (no.)
Cohort (n ¼ 174) 56.32 95.40 59.77 54.02
(98) (166) (104) (94)
Pre-ASP (n ¼ 86) 47.67 91.86 50.00 43.02
(41) (79) (43) (37)
ASP (n ¼ 88) 64.77 98.86 69.32 64.77
(57) (87) (61) (57)
P 0.023 0.027 0.009 0.004
ASP ¼ antimicrobial stewardship program.
ture by 6.4% in a 253-bed hospital. They were able to find a 37% cost reduction per quarter with decreased antibiotic use. Yu et al28 compared 2 sites with ASPs with 3 sites without ASPs and similarly found a significant antibiotic cost reduction (17.3%) at the sites with ASPs. Notably, 72% of the ASP interventions focused on narrowing or discontinuing antibiotic therapy. Studies have also reported that early switching from IV to PO antibiotic therapy can decrease LOS without increase readmissions.29 Our ASP was able to facilitate outcome improvement by aiding in the care of the patients throughout the treatment duration. On a daily basis our ASP monitors all IV antibiotics in the facility. The EMR allows for quick ascertainment of data. Therapy is altered based on prior cultures, and a working relationship with the microbiology laboratory allows therapy to be optimized based on real-time data. This contributes to the patient-centered approach to daily care of the patient and encourages better monitoring of patient improvement and antibiotic susceptibilities. The ASP was able to significantly increase the number of cultures obtained, which then allowed for earlier streamlining of antibiotics before susceptibilities. Earlier antibiotic streamlining often leads to quicker conversion from IV to PO, decreased duration of treatment, and reduced LOS. These interventions illustrate how the ASP influenced more appropriate management of patients with pneumonia. The prevention of C difficile infections is one goal of ASPs because these infections increase health care costs by increasing antibiotic use, increasing duration of treatment, and prolonging hospital stay.30 Previous studies on ASPs have found both significant and nonsignificant decrease in C difficile infections.22,28,31,32 Our study did not find a
1756
statistically significant change in C difficile infection rates. However, in 2011 our facility switched from using immunochromatographic assays to using PCR to detect C difficile, which could have led to higher numbers of C difficile detection as previously discussed by Dubberke et al.33 A previous ASP study by Nowak et al31 looked at the outcomes of pneumonia treatment and failed to find a difference in 30-day readmissions before and after ASP implementation. Analysis of 30-day readmission between the 2 groups in our study found that readmission in the ASP group (OR ¼ 0.36) was less likely than in the pre-ASP group (OR ¼ 2.78; P ¼ 0.05). Therefore, despite treating a sicker population, decreasing treatment duration, decreasing time of IV to PO conversion, and decreasing LOS, our ASP did not sacrifice patient care and improved overall patient outcomes by preventing 30-day readmissions. Our study has limitations. Because of the retrospective cohort design, the study is subject to biases, including selection, information, and misclassification biases. External validity can be limited by the study being conducted at a single center with limited population diversity. This study was conducted at a VA hospital where the pharmacist has an expanded scope of practice. This approach to antimicrobial stewardship may not be possible at a facility without an EMR in which data, including cultures, from past hospitalizations and outpatient visits are available to make therapeutic decisions using specific patient data. Small sample sizes increase risk of difference between groups being due to chance. Non-concurrent cohort time periods could have confound results. The percentage of HCAP was higher in the ASP time frame. The pre-ASP time frame corresponded to after the publication of the IDSA guidelines on HCAP to
Volume 38 Number 7
K.E. Kurtzhalts et al. exclude the bias from definition change.26 Changes independent of the ASP such as provider turnover and change in facility prescribing patterns could have occurred between the 2 cohort time periods. ID physicians, however, did not change between the 2 cohort time periods.
5.
6.
CONCLUSION The development of patient-centered ASPs is now a necessary component in improving treatment outcomes in health care facilities. Our study found that significant improvement in duration of antibiotic use and LOS in pneumonia treatment can be achieved with the addition of an ASP. A well-designed ASP is able to achieve these types of outcomes through a patient-centered approach, monitoring patients daily and recommending therapy changes based on individual patient progress and pathogen susceptibilities. Our data indicate that an ASP is able to aid in decreasing treatment duration, decreasing LOS, and increasing antibiotic streamlining while also decreasing 30-day readmissions.
CONFLICTS OF INTEREST This material is the result of work supported with resources and the use of facilities at the Veterans Affairs Western New York Healthcare System. The contents of this manuscript are not intended to represent the views of the Department of Veterans Affairs or the United States Government. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
7.
8.
9.
10.
11.
12.
13.
14.
ACKNOWLEDGMENTS All authors contributed equally. 15.
REFERENCES 1. Kochanek KD, Murphy SL, Xu J, et al. Mortality in the United States, 2013. NCHS Data Brief. 2014;178:1–8. 2. Murphy SL, Xu J, Kochanek KD. Deaths: final data for 2010. Natl Vital Stat Rep. 2013;61:1–117. 3. Feikin DR, Schuchat A, Kolczak M, et al. Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 1995-1997. Am J Public Health. 2000;90: 223–229. 4. Butler JC, Hofmann J, Cetron MS, et al. The continued emergence of drug-resistant Streptococcus pneumoniae in the United States: an update from the Centers for Disease
July 2016
16.
17.
18.
Control and Prevention's Pneumococcal Sentinel Surveillance System. J Infect Dis. 1996;174:986–993. Nuorti JP, Butler JC, Crutcher JM, et al. An outbreak of multidrug-resistant pneumococcal pneumonia and bacteremia among unvaccinated nursing home residents. N Engl J Med. 1998;338:1861–1868. Low DE. What is the relevance of antimicrobial resistance on the outcome of community-acquired pneumonia caused by Streptococcus pneumoniae? (should macrolide monotherapy be used for mild pneumonia?). Infect Dis Clin North Am. 2013;27:87–97. Perez-Trallero E, Marimon JM, Iglesias L, et al. Fluoroquinolone and macrolide treatment failure in pneumococcal pneumonia and selection of multidrug-resistant isolates. Emerg Infect Dis. 2003;9:1159–1162. Sanchez GV, Master RN, Karlowsky JA, et al. In vitro antimicrobial resistance of urinary Escherichia coli isolates among U.S. outpatients from 2000 to 2010. Antimicrob Agents Chemother. 2012;56:2181–2183. Vasoo S, Barreto JN, Tosh PK. Emerging issues in gramnegative bacterial resistance: an update for the practicing clinician. Mayo Clin Proc. 2015;90:395–403. Aliberti S, Di Pasquale M, Zanaboni AM, et al. Stratifying risk factors for multidrug-resistant pathogens in hospitalized patients coming from the community with pneumonia. Clin Infect Dis. 2012;54:470–478. Gonzales R, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for treatment of acute respiratory tract infections in adults: background, specific aims, and methods. Ann Emerg Med. 2001;37:690–697. Gonzales R, Malone DC, Maselli JH, et al. Excessive antibiotic use for acute respiratory infections in the United States. Clin Infect Dis. 2001;33:757–762. Arias CA, Murray BE. Antibiotic-resistant bugs in the 21st century–a clinical super-challenge. N Engl J Med. 2009;360: 439–443. Hicks LA, Chien Y-W, Taylor TH, et al. Outpatient antibiotic prescribing and nonsusceptible Streptococcus pneumoniae in the United States, 1996-2003. Clin Infect Dis. 2011;53:631–639. Davidson R, Cavalcanti R, Brunton JL, et al. Resistance to levofloxacin and failure of treatment of pneumococcal pneumonia. N Engl J Med. 2002;346:747–750. Hirsch EB, Tam VH. Impact of multidrug-resistant Pseudomonas aeruginosa infection on patient outcomes. Expert Rev Pharmacoecon Outcomes Res. 2010;10:441–451. Aloush V, Navon-Venezia S, Seigman-Igra Y, et al. Multidrug-resistant Pseudomonas aeruginosa: risk factors and clinical impact. Antimicrob Agents Chemother. 2006;50: 43–48. Dellit TH, Owens RC, McGowan JE Jr., et al. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing
1757
Clinical Therapeutics
19.
20.
21.
22.
23.
24.
25.
26.
an institutional program to enhance antimicrobial stewardship. Clin Infect Dis. 2007;44:159–177. Barlam TF, Cosgrove SE, Abbo LM, et al. Implementing an Antibiotic Stewardship Program: Guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis. 2016; 62:e51–e77. Kaki R, Elligsen M, Walker S, et al. Impact of antimicrobial stewardship in critical care: a systematic review. J Antimicrob Chemother. 2011; 66:1223–1230. Vettese N, Hendershot J, Irvine M, et al. Outcomes associated with a thrice-weekly antimicrobial stewardship programme in a 253-bed community hospital. J Clin Pharm Ther. 2013;38:401–404. Avdic E, Cushinotto LA, Hughes AH, et al. Impact of an antimicrobial stewardship intervention on shortening the duration of therapy for community-acquired pneumonia. Clin Infect Dis. 2012; 54:1581–1587. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40: 373–383. Jeong BH, Koh WJ, Yoo H, et al. Performances of prognostic scoring systems in patients with healthcare-associated pneumonia. Clin Infect Dis. 2013;56:625–632. Fang WF, Yang KY, Wu CL, et al. Application and comparison of scoring indices to predict outcomes in patients with healthcareassociated pneumonia. Crit Care. 2011;15:R32. American Thoracic Society, Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med. 2005;171:388–416.
1758
27. Cyriac JM, James E. Switch over from intravenous to oral therapy: A concise overview. J Pharmacol Pharmacother. 2014;5:83–87. 28. Yu K, Rho J, Morcos M, et al. Evaluation of dedicated infectious diseases pharmacists on antimicrobial stewardship teams. Am J Health Syst Pharm. 2014;71: 1019–1028. 29. Athanassa Z, Makris G, Dimopoulos G, et al. Early switch to oral treatment in patients with moderate to severe community-acquired pneumonia: a meta-analysis. Drugs. 2008;68:2469–2481. 30. Kyne L, Hamel MB, Polavaram R, et al. Health care costs and mortality associated with nosocomial diarrhea due to Clostridium
difficile. Clin Infect Dis. 2002;34: 346–353. 31. Nowak MA, Nelson RE, Breidenbach JL, et al. Clinical and economic outcomes of a prospective antimicrobial stewardship program. Am J Health Syst Pharm. 2012; 69:1500–1508. 32. Lew KY, Ng TM, Tan M, et al. Safety and clinical outcomes of carbapenem de-escalation as part of an antimicrobial stewardship programme in an ESBL-endemic setting. J Antimicrob Chemother. 2015; 70:1219–1225. 33. Dubberke ER, Burnham CA. Diagnosis of Clostridium difficile Infection: Treat the Patient, Not the Test. JAMA Intern Med. 2015;175: 1801–1802.
Address correspondence to: Kari A. Mergenhagen, PharmD, BCPS AQID, Infectious Diseases Department, Veteran Affairs Western New York Healthcare System, 3495 Bailey Avenue, Buffalo, NY 14215. E-mail: kari. [email protected]
Volume 38 Number 7
Impact of Antimicrobial Stewardship on Outcomes in Hospitalized Veterans With Pneumonia Kari E. Kurtzhalts, PharmD1; John A. Sellick Jr, DO, MS1,2; Christine A. Ruh, PharmD, BCPS4; James F. Carbo, PharmD1,3; Michael C. Ott, PharmD, BCPS4; and Kari A. Mergenhagen, PharmD, BCPS AQ-ID1 1
Infectious Diseases Department, Veterans Affairs Western New York Healthcare System, Buffalo, New York; 2University at Buffalo School of Medicine, The State University of New York, Buffalo, New York; 3 Veterans Affairs Southern Nevada Healthcare System, North Las Vegas, Nevada; and 4Pharmacy Department, Erie County Medical Center, Buffalo, New York ABSTRACT Purpose: The purpose of this study was to evaluate the impact of an antimicrobial stewardship program (ASP) on outcomes for inpatients with pneumonia, including length of stay, treatment duration, and 30day readmission rates. Methods: A retrospective chart review comparing outcomes of veterans admitted with pneumonia before (2005–2006) and after (2013–2014) implementation of an ASP was conducted; pneumonia was defined according to International Classification of Diseases, Ninth Revision (ICD-9) codes. Infectious diseases physicians and pharmacist in the ASP provided appropriate recommendations to the primary medicine teams. Bivariate analysis of baseline characteristics and comorbid conditions were performed between the time frames. Least squares regression was used to analyze length of stay, time of IV to PO conversions, and duration of antibiotics. Multivariate logistic regressions were used to determine odds of 30-day readmission and odds of Clostridium difficile infections between time periods. Findings: There were 86 patients in the pre-ASP period and 88 patients in the ASP period. Mean length of stay decreased from 8.1 to 6.6 days (P ¼ 0.02), total duration of antibiotic therapy decreased from 12 to 8.5 days (P o 0.0001), and time of IV to PO antibiotic conversions decreased from 5.3 to 3.9 days (P ¼ 0.0003), before ASP and during ASP, respectively. The odds ratio of 30-day readmission before ASP was 2.78 and 0.36 during the ASP (P ¼ 0.05). The odds ratios of Clostridium difficile infections before ASP was 2.08 and 0.48 during the ASP (P ¼ 0.37).
1750
Implications: The ASP interventions were associated with shorter durations of therapy, shorter lengths of stay, and lower rates of readmission and Clostridium difficile infections within 30 days. Limitations of this study are retrospective cohort design, small study population, limited study population diversity, and non-concurrent cohort times periods. (Clin Ther. 2016;38:1750–1758) Published by Elsevier HS Journals, Inc. Key words: antimicrobial stewardship, patientcentered, pneumonia, readmission.
INTRODUCTION Pneumonia is one of the leading causes of death in the United States, accounting for 57,000 deaths in 2013.1,2 Worsening antibiotic resistance and multidrug resistant (MDR) organisms may be linked to death and poor treatment outcomes.3–9 Health care exposure, inappropriate use of antibiotics at institutions, and overprescribing of antibiotics in the community are some factors associated with development of MDR organisms and antibiotic resistance.10–14 Antibiotic resistance can result in other negative outcomes such as increased length of hospital stay (LOS), prolonged treatment duration, and development of Clostridium difficile infections.15–17 The use of antimicrobial stewardship programs (ASPs) in health care facilities has been encouraged Accepted for publication June 1, 2016. http://dx.doi.org/10.1016/j.clinthera.2016.06.004 0149-2918/$ - see front matter Published by Elsevier HS Journals, Inc.
Volume 38 Number 7
K.E. Kurtzhalts et al. as an important tool in improving and preserving antibiotic susceptibilities, along with improving patient care and health care outcomes.18 In 2007 the Infectious Disease Society of America (IDSA) and the Society of Healthcare Epidemiology of America published guidelines that laid foundations to aid institutions in developing ASPs.18 The IDSA updated these guidelines in 2016 with recommendations for the implementation and measurement of the success of these ASPs.19 These guidelines recommend that antimicrobial stewardship teams consist of collaboration between infectious disease (ID) physicians and ID-specialized clinical pharmacists. Since then, ASPs have emerged in facilities across the country and have improved antimicrobial utilization, decreased treatment durations, and reduced treatment costs.20–22 In 2009 the Veteran’s Affairs Healthcare System in Buffalo, New York, began their ASP team with an ID-specialized clinical pharmacist and a team of ID physicians. This ASP uses a patientcentered approach. Automated protocols are not used, and the ID pharmacist performs daily chart reviews for each patient on IV antibiotics. The goal of this study was to address the impact our ASP has had on the outcomes of pneumonia treatment, including LOS, duration of antibiotic treatment, time of IV to PO antibiotic conversion, 30-day hospital readmissions, and occurrence of C difficile infections. Interventions of the ASP were characterized as secondary objectives.
PATIENTS AND METHODS This single-centered retrospective cohort study took place at the Veterans Affairs Western New York Healthcare System (VAWNYHCS), which is a 150bed level 1b facility. The institutional review board approved this study. Patient information was collected for patients who were hospitalized and received antibiotics for pneumonia during a 1-year period before the facility’s ASP was initiated and during a 1-year period after the program had been well established. The pre-ASP group consisted of the period of time between June 1, 2005 and June 1, 2006, and between June 1, 2013, and June 1, 2014, for the ASP group. Patients must have had a primary diagnosis of pneumonia, based on International Classification of Diseases, Ninth Revision (ICD-9) codes (507, 486, 518, 381, 482, 481, 038.9) to be included in this study; both general ward and patients in the intensive
July 2016
care unit were included. Exclusion criteria included death on discharge after being placed on comfort care measures, o48 hours of inpatient admission or inpatient antibiotics, lung abscess, empyema, necrotizing pneumonia, having a thoracentesis, severe immunosuppression, concurrent extra-pulmonary source of infection (such as endocarditis, urinary tract infection, meningitis, etc), and if there was no documented duration of therapy (ie, lost to follow-up). The VAWNYHCS is a vertical health care system, providing both outpatient and inpatient care to patients, which allowed us to accurately follow treatment both while the patients were hospitalized and after discharge.
Definitions Patients were considered to be immunosuppressed if they had systemic chemotherapy within 28 days of admission, HIV infection/AIDS, solid organ or bone marrow transplant, or absolute neutrophil count of o1500 cells within 28 days of admission. Morbidity was calculated using the Charlson Comorbidity Score.23 Severity of illness was calculated by using the CURB-65 (confusion, uremia, elevated respiratory rate, low blood pressure, and age Z 65 years) score. The CURB-65 score was used to determine level of illness severity on admission for patients with community-acquired pneumonia (CAP) and for health care-associated pneumonia (HCAP).24,25 A patient was considered to have HCAP if he or she met the following criteria on admission: either acute hospitalization for 42 days or long-term care facility residence within 90 days of admission, or any of the following within 28 days of admission: chronic hemodialysis, antibiotics, wound care, tracheotomy, ventilator, or aspiration.26
Antimicrobial Stewardship The antimicrobial stewardship team consists of ID physicians and a clinical ID pharmacist with added qualifications in IDs. The ASP is patient centered, using individualized patient care by daily chart reviews. Select antibiotics were restricted, requiring approval by the ID team for use. These restrictions were in place before the ASP was established. Each morning a report is printed of all hospitalized patients on IV antibiotics, and the ASP then reviews each patient’s chart to assess appropriateness of antibiotic therapy. Interventions consist of prospective audit and feedback regarding antibiotic selection, microbiology,
1751
Clinical Therapeutics and duration of therapy. Care of each patient is individualized, based on prior history and response to treatment. In addition, brief monthly educational conferences on antimicrobial stewardship and local antimicrobial resistance patterns are provided to the house staff to underscore the importance of microbial cultures and to promote appropriate use of antimicrobial agents. Specialized antibiograms are created by the ASP and used for education.19 These antibiograms consist of a hospital-wide antibiogram, intensive carespecific antibiogram, and specific antibiograms for respiratory, urinary, and blood. In addition, a Pseudomonas aeruginosa cross-antibiogram is created annually. The ASP pharmacist has a scope of practice that includes the ability to change antibiotic dosing as needed. Antibiotic changes, including choice of antibiotic, are reviewed with the physician of record before implementation; dosing changes can be made independently. The pharmacist can also order other diagnostic and laboratory tests such as drug levels, cultures, complete metabolic panel, complete blood count, etc. Generally, urine, blood, and sputum cultures can be ordered by the pharmacist; however, the physician of record orders other invasive cultures, such as induced sputum. The pharmacist operates with ID physician support. There is 1 full-time equivalent (FTE) pharmacist, 1 FTE pharmacy resident, and 0.1 FTE physician dedicated to the ASP. The VA computerized electronic medical record (EMR) system enables information to be easily gathered. Notes, laboratory values, cultures, vital signs, and electrocardiograms are all readily available in this system. The computerized EMR system did not have electronic clinical decision support or automated protocols for pneumonia at the time of prescribing, but instead relied on real-time interventions by the ASP team. The VA provides ongoing inpatient and outpatient care for patients, thus enabling the ASP to use past cultures and antibiotic treatment to tailor a regimen specific for that patient. The ASP team visits the microbiology laboratory daily to obtain interim culture data, which may enable antimicrobial therapy optimization before cultures are finalized. The only rapid diagnostic test available is polymerase chain reaction (PCR) for C difficile, starting in 2011. Procalcitonin and other rapid diagnostics are not used at our institution.
1752
Outcomes The primary objective for the study was to evaluate the impact of an ASP on pneumonia treatment outcomes in hospitalized patients. Primary outcomes were LOS, duration of treatment, time from IV to PO antibiotic conversion, 30-day readmission rates, and C difficile infections occurring within 30 days of antibiotic initiation. Secondary outcomes included the number of antibiotic therapies narrowed, number of blood and sputum cultures obtained, and number of therapy changes made during treatment. The patientcentric ASP was characterized by using interventions such as dose optimization, ordering cultures, ordering diagnostic laboratory tests, therapy escalation and deescalation, IV to PO antibiotic changes, and adjustments in antibiotics due to QTc prolongation.
Statistical Analysis Bivariate analysis was performed on baseline characteristics between the pre-ASP and ASP groups and for secondary outcomes, number of antibiotics streamlined, number of blood and sputum cultures obtained, and number of therapy changes made during treatment. Ordinary least squares regression was used to analyze the following primary outcomes: LOS, duration of antibiotics, and time to convert from IV to PO therapy. Pre-ASP and ASP groups were included in the model a priori. In the least squares regression variables with significance (P o 0.05) were built into the model. Odds ratios (OR) and 95% CIs were calculated for 30-day readmissions and C difficile infections. Statistical tests were two tailed with statistical significance achieved with a P value r 0.05. ASP interventions were analyzed as percentages. Statistical analysis was performed with JMP software (Version 10; SAS, Cary, North Carolina).
RESULTS A total of 303 patient charts were reviewed using the VAWNYHCS EMR; 174 patients were evaluated and 129 patients were excluded (Figure 1). There were 86 patients in the pre-ASP group and 88 patients in the ASP group. No statistical significant differences were found in baseline characteristics between the 2 groups; the median ages were 75.04 years, most were white men, and with a median Charlson Comorbidity Score of 3.53 (Table I). In the ASP group more patients were treated for HCAP than CAP compared with the
Volume 38 Number 7
K.E. Kurtzhalts et al.
303 patient charts reviewed
174 patients included
86 Pre-ASP group
129 patients excluded
88 ASP group
Pneumonia not primary dx - 52 < 48 hrs of admission - 8 Death - 17 < 48 hrs of inpatient abx - 9 Empyema - 1 Thoracentesis performed- 8 Chemo w/in 20 days of admission - 11 HIV/AIDS - 2 Solid organ/bone marrow transplant - 3 Concurrent infection - 5 No documented therapy duration - 8 Aspiration/post - obstructive pna - 5
Figure 1. Patient enrollment. Abx ¼ antibiotics; ASP ¼ antimicrobial stewardship program; dx ¼ diagnosis; pna ¼ pneumonia.
pre-ASP group. Patients in the ASP group had a higher severity of illness than the pre-ASP group as measured by the CURB-65 score (P ¼ 0.032). Results of the multivariate analysis of LOS, duration of treatment, and time of IV to PO conversion are listed in Table II. With the use of a least squares regression model to adjust for CURB-65 score and type of pneumonia, LOS was decreased from 8.07 ⫾ 0.44 days (pre-ASP) to 6.68 ⫾ 0.40 days (ASP) (P ¼ 0.018). In this model LOS varied significantly with CURB-65 (P ¼ 0.018) and type of pneumonia (P o 0.0001). The mean LOS for a patient with CAP was 6.0 days compared with 8.87 days in patients with HCAP. In a second least squares regression model adjusted for CURB-65 score and type of pneumonia, duration of treatment was decreased from 11.93 ⫾ 0.45 days (pre-ASP) to 8.45 ⫾ 0.41 days (ASP) (P o 0.0001). In this model total duration of treatment did not vary significantly with CURB-65 score (P ¼ 0.056) or type of pneumonia (P ¼ 0.089). In a third
July 2016
least squares regression model adjusted for CURB-65 score and type of pneumonia, time of IV to PO conversion varied significantly between pre-ASP (5.32 ⫾ 0.3 days) and ASP (2.93 ⫾ 0.26 days) cohorts (P ¼ 0.0003). In this model time of IV to PO conversion was significantly affected by CURB-65 score (P ¼ 0.0085) but not by type of pneumonia (P ¼ 0.4). The OR of 30-day readmission in the pre-ASP group was 2.78 (95% CI, 0.99–8.29) compared with 0.36 (95% CI, 0.12–0.99) in the ASP group (P ¼ 0.05) (Table III), in this multivariate logistic regression. In this model, CURB-65 was not a significant predictor of readmission (P ¼ 0.98); however, patients with HCAP were more likely to be readmitted, (OR ¼ 7.97; 95% CI, 2.72–25.90) than patients with CAP (OR ¼ 0.13; 95% CI, 0.039–0.37). The OR of C difficile occurrence 30 days after antibiotic discontinuation was 2.08 (95% CI, 0.41–11.76) and 0.48 (95% CI, 0.085–2.42), pre-ASP and ASP groups,
1753
Clinical Therapeutics
Table I. Bivariate analysis of baseline characteristics. Cohort (n ¼ 174)
Pre-ASP (n ¼ 86)
ASP (n ¼ 88)
75.40 (11.26)
74.74 (11.58)
75.03 (10.97)
83.91 12.64 3.45 97.13 68.98 85.72 27.86 1.52 57.21 11.53 32.76 3.53 43.68 27.01 5.17 3.45 27.01 18.97 5.17 22.99 5.75 1.15 1.73 2.87 48.84 1.93
82.56 16.28 1.16 96.51 69.16 83.19 26.90 1.34 58.06 11.42 31.40 3.29 41.86 26.74 4.65 4.65 24.42 16.28 2.33 19.77 5.81 1.16 0 3.49 44.71 1.76
85.23 9.09 5.68 97.73 68.80 88.20 28.79 1.70 56.37 11.64 34.09 3.76 45.45 27.27 5.68 2.27 29.55 21.59 7.95 26.14 5.68 1.14 3.41 2.27 52.87 2.09
Variable Age, mean (SD) years Race, % (no.) White Black Other Sex, male, % (no.) Height, mean (SD), inches Weight, mean (SD), kg BMI, mean (SD), kg/m2 SCr, mean (SD), g/dL CrCl, mean (SD), mL/min WBC, mean (SD), 103/μL Diabetes, % (no.) Charlson score, mean (SD) COPD, % (no.) Heart failure, % (no.) Cirrhosis, % (no.) Anatomic abnormalities, % (no.) Acute hospitalization w/in 90 days prior, % (no.) Long-term care w/in 90 days, % (no.) Dialysis, % (no.) Antibiotics w/in 28 days, % (no.) Wound care, % (no.) Tracheostomy, % (no.) Ventilator, % (no.) Aspiration, % (no.) Risk, % (no.) CURB-65, mean (SD) Type of pneumonia, % (no.) CAP HCAP Service, % (no.) ICU Medical
(146) (22) (6) (169) (3.15) (26.59) (8.10) (1.67) (25.56) (5.59) (57) (2.16) (76) (47) (9) (6) (47) (33) (9) (40) (10) (2) (3) (5) (84) (1.03)
(71) (14) (1) (83) (3.09) (21.27) (6.41) (1.05) (24.52) (5.52) (27) (2.25) (36) (23) (4) (4) (21) (14) (2) (17) (5) (1) (0) (3) (38) (1.07)
(75) (8) (5) (86) (3.21) (30.84) (9.41) (2.10) (26.66) (5.69) (30) (2.06) (40) (24) (5) (2) (26) (19) (7) (23) (5) (1) (3) (2) (46) (0.97)
68.97 (120) 31.03 (54)
81.40 (70) 18.60 (16)
56.82 (50) 43.18 (38)
9.20 (16) 90.80 (158)
5.81 (5) 94.19 (81)
12.50 (11) 87.50 (77)
P 0.45 0.11
0.63 0.46 0.21 0.12 0.14 0.66 0.79 0.71 0.071 0.63 0.94 0.76 0.39 0.45 0.37 0.094 0.32 0.97 0.99 0.084 0.63 0.28 0.032 0.0005
0.13
ASP ¼ antimicrobial stewardship program; BMI ¼ body mass index; CAP ¼ community acquired pneumonia; COPD, chronic obstructive pulmonary disease; CrCl ¼ creatinine clearance; CURB-65 ¼ confusion, uremia, elevated respiratory rate, low blood pressure, and age Z 65 years; HCAP ¼ health care-associated pneumonia; ICU ¼ intensive care unit; SCr ¼ serum creatinine; WBC ¼ white blood cell; w/in ¼ within;
respectively (P ¼ 0.37) (Table III). CURB-65 score and type of pneumonia were not significant (P ¼ 0.58 and 0.16, respectively).
1754
Compared with the pre-ASP group, the ASP group had significantly more sputum cultures and blood cultures obtained (P ¼ 0.023). The number of antibiotic
Volume 38 Number 7
K.E. Kurtzhalts et al.
Table II. Least squares regression of primary outcomes. Variable Length of stay, days Duration of antibiotics, days IV to PO conversion, days*
Pre-ASP (n ¼ 86)
ASP (n ¼ 88)
P
8.07 (0.44) 11.93 (0.45) 5.32 ⫾ 0.30
6.68 (0.40) 8.45 (0.41) 3.92 ⫾ 0.26
0.018 o0.0001 0.0003
Values are least squares mean (SE). Outcome variables were adjusted for type of pneumonia (community acquired or healthcare associated) and CURB-65 (confusion, uremia, elevated respiratory rate, low blood pressure, and age Z 65 years) score. ASP ¼ antimicrobial stewardship program. * Pre-ASP: n ¼ 70; ASP: n ¼ 76.
therapies changed (P ¼ 0.009) and streamlined (P ¼ 0.004) were also significantly greater in the ASP group than in the non-ASP group (Table IV). These results were achieved through 130 recommendations made by the ASP team and accepted by patients’ primary teams, 1.5 recommendations per patient. Accepted recommendations were distributed among 7 categories; optimization of antibiotic dosing (39%), de-escalation of therapy (20%), obtain cultures (blood or sputum) (12%), IV to PO therapy (8.5%), order various diagnostic laboratory tests (4.6%), escalation of therapy (3.8%), and antibiotic change due to QTc prolongation (3%).
DISCUSSION Our study found that ASPs may have a significant impact on treatment outcomes in pneumonia. Our ASP was able to significantly decrease time to convert from IV to PO therapy, duration of treatment, and LOS while also decreasing 30-day readmissions. These outcomes have significant impacts on
improvement in patient care, by decreasing length of hospitalization and subsequent readmissions and also on the burdens of the health care system. Decreasing treatment duration can help prevent adverse drug events that may occur with excessive use of antibiotics. Earlier conversion from IV to PO antibiotics decreases risk of line infections, decreases pharmacy costs, and also eliminates a barrier to discharge.27 Decreasing LOS reduces the amount of health care resources needed per admission, increases bed availability within the facility, and improves patient outcomes by decreasing a patient’s exposure time to hospital-associated pathogens. Previous studies analyzing the impact of ASPs have reported similar results.21,22 Avdic et al22 studied 127 patients being treated for CAP and found that antimicrobial stewardship helped to reduce duration of therapy from 10 to 7 days; however, they found no difference in LOS between their pre-ASP and intervention groups. Vettese et al21 found that a thrice-weekly ASP decreased total antibiotic expendi-
Table III. Odds ratios of 30-day readmission and C difficile infections. Variable 30-day readmission* C difficile†
Pre-ASP
ASP
P
2.78 (0.99–8.29) 2.08 (0.41–11.76)
0.36 (0.12–1.00) 0.48 (0.085–2.42)
0.05 0.37
Values are odds ratio (95% CI). ASP ¼ antimicrobial stewardship program. * Pre-ASP: n ¼ 12, ASP: n ¼ 9 † Pre-ASP: n ¼ 4, ASP: n ¼ 3
July 2016
1755
Clinical Therapeutics
Table IV. Bivariate analysis of treatment outcomes. Variable Sputum cultures, % (no.) Blood cultures, % (no.) Therapy change, % (no.) Antibiotics narrowed, % (no.)
Cohort (n ¼ 174) 56.32 95.40 59.77 54.02
(98) (166) (104) (94)
Pre-ASP (n ¼ 86) 47.67 91.86 50.00 43.02
(41) (79) (43) (37)
ASP (n ¼ 88) 64.77 98.86 69.32 64.77
(57) (87) (61) (57)
P 0.023 0.027 0.009 0.004
ASP ¼ antimicrobial stewardship program.
ture by 6.4% in a 253-bed hospital. They were able to find a 37% cost reduction per quarter with decreased antibiotic use. Yu et al28 compared 2 sites with ASPs with 3 sites without ASPs and similarly found a significant antibiotic cost reduction (17.3%) at the sites with ASPs. Notably, 72% of the ASP interventions focused on narrowing or discontinuing antibiotic therapy. Studies have also reported that early switching from IV to PO antibiotic therapy can decrease LOS without increase readmissions.29 Our ASP was able to facilitate outcome improvement by aiding in the care of the patients throughout the treatment duration. On a daily basis our ASP monitors all IV antibiotics in the facility. The EMR allows for quick ascertainment of data. Therapy is altered based on prior cultures, and a working relationship with the microbiology laboratory allows therapy to be optimized based on real-time data. This contributes to the patient-centered approach to daily care of the patient and encourages better monitoring of patient improvement and antibiotic susceptibilities. The ASP was able to significantly increase the number of cultures obtained, which then allowed for earlier streamlining of antibiotics before susceptibilities. Earlier antibiotic streamlining often leads to quicker conversion from IV to PO, decreased duration of treatment, and reduced LOS. These interventions illustrate how the ASP influenced more appropriate management of patients with pneumonia. The prevention of C difficile infections is one goal of ASPs because these infections increase health care costs by increasing antibiotic use, increasing duration of treatment, and prolonging hospital stay.30 Previous studies on ASPs have found both significant and nonsignificant decrease in C difficile infections.22,28,31,32 Our study did not find a
1756
statistically significant change in C difficile infection rates. However, in 2011 our facility switched from using immunochromatographic assays to using PCR to detect C difficile, which could have led to higher numbers of C difficile detection as previously discussed by Dubberke et al.33 A previous ASP study by Nowak et al31 looked at the outcomes of pneumonia treatment and failed to find a difference in 30-day readmissions before and after ASP implementation. Analysis of 30-day readmission between the 2 groups in our study found that readmission in the ASP group (OR ¼ 0.36) was less likely than in the pre-ASP group (OR ¼ 2.78; P ¼ 0.05). Therefore, despite treating a sicker population, decreasing treatment duration, decreasing time of IV to PO conversion, and decreasing LOS, our ASP did not sacrifice patient care and improved overall patient outcomes by preventing 30-day readmissions. Our study has limitations. Because of the retrospective cohort design, the study is subject to biases, including selection, information, and misclassification biases. External validity can be limited by the study being conducted at a single center with limited population diversity. This study was conducted at a VA hospital where the pharmacist has an expanded scope of practice. This approach to antimicrobial stewardship may not be possible at a facility without an EMR in which data, including cultures, from past hospitalizations and outpatient visits are available to make therapeutic decisions using specific patient data. Small sample sizes increase risk of difference between groups being due to chance. Non-concurrent cohort time periods could have confound results. The percentage of HCAP was higher in the ASP time frame. The pre-ASP time frame corresponded to after the publication of the IDSA guidelines on HCAP to
Volume 38 Number 7
K.E. Kurtzhalts et al. exclude the bias from definition change.26 Changes independent of the ASP such as provider turnover and change in facility prescribing patterns could have occurred between the 2 cohort time periods. ID physicians, however, did not change between the 2 cohort time periods.
5.
6.
CONCLUSION The development of patient-centered ASPs is now a necessary component in improving treatment outcomes in health care facilities. Our study found that significant improvement in duration of antibiotic use and LOS in pneumonia treatment can be achieved with the addition of an ASP. A well-designed ASP is able to achieve these types of outcomes through a patient-centered approach, monitoring patients daily and recommending therapy changes based on individual patient progress and pathogen susceptibilities. Our data indicate that an ASP is able to aid in decreasing treatment duration, decreasing LOS, and increasing antibiotic streamlining while also decreasing 30-day readmissions.
CONFLICTS OF INTEREST This material is the result of work supported with resources and the use of facilities at the Veterans Affairs Western New York Healthcare System. The contents of this manuscript are not intended to represent the views of the Department of Veterans Affairs or the United States Government. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
7.
8.
9.
10.
11.
12.
13.
14.
ACKNOWLEDGMENTS All authors contributed equally. 15.
REFERENCES 1. Kochanek KD, Murphy SL, Xu J, et al. Mortality in the United States, 2013. NCHS Data Brief. 2014;178:1–8. 2. Murphy SL, Xu J, Kochanek KD. Deaths: final data for 2010. Natl Vital Stat Rep. 2013;61:1–117. 3. Feikin DR, Schuchat A, Kolczak M, et al. Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 1995-1997. Am J Public Health. 2000;90: 223–229. 4. Butler JC, Hofmann J, Cetron MS, et al. The continued emergence of drug-resistant Streptococcus pneumoniae in the United States: an update from the Centers for Disease
July 2016
16.
17.
18.
Control and Prevention's Pneumococcal Sentinel Surveillance System. J Infect Dis. 1996;174:986–993. Nuorti JP, Butler JC, Crutcher JM, et al. An outbreak of multidrug-resistant pneumococcal pneumonia and bacteremia among unvaccinated nursing home residents. N Engl J Med. 1998;338:1861–1868. Low DE. What is the relevance of antimicrobial resistance on the outcome of community-acquired pneumonia caused by Streptococcus pneumoniae? (should macrolide monotherapy be used for mild pneumonia?). Infect Dis Clin North Am. 2013;27:87–97. Perez-Trallero E, Marimon JM, Iglesias L, et al. Fluoroquinolone and macrolide treatment failure in pneumococcal pneumonia and selection of multidrug-resistant isolates. Emerg Infect Dis. 2003;9:1159–1162. Sanchez GV, Master RN, Karlowsky JA, et al. In vitro antimicrobial resistance of urinary Escherichia coli isolates among U.S. outpatients from 2000 to 2010. Antimicrob Agents Chemother. 2012;56:2181–2183. Vasoo S, Barreto JN, Tosh PK. Emerging issues in gramnegative bacterial resistance: an update for the practicing clinician. Mayo Clin Proc. 2015;90:395–403. Aliberti S, Di Pasquale M, Zanaboni AM, et al. Stratifying risk factors for multidrug-resistant pathogens in hospitalized patients coming from the community with pneumonia. Clin Infect Dis. 2012;54:470–478. Gonzales R, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for treatment of acute respiratory tract infections in adults: background, specific aims, and methods. Ann Emerg Med. 2001;37:690–697. Gonzales R, Malone DC, Maselli JH, et al. Excessive antibiotic use for acute respiratory infections in the United States. Clin Infect Dis. 2001;33:757–762. Arias CA, Murray BE. Antibiotic-resistant bugs in the 21st century–a clinical super-challenge. N Engl J Med. 2009;360: 439–443. Hicks LA, Chien Y-W, Taylor TH, et al. Outpatient antibiotic prescribing and nonsusceptible Streptococcus pneumoniae in the United States, 1996-2003. Clin Infect Dis. 2011;53:631–639. Davidson R, Cavalcanti R, Brunton JL, et al. Resistance to levofloxacin and failure of treatment of pneumococcal pneumonia. N Engl J Med. 2002;346:747–750. Hirsch EB, Tam VH. Impact of multidrug-resistant Pseudomonas aeruginosa infection on patient outcomes. Expert Rev Pharmacoecon Outcomes Res. 2010;10:441–451. Aloush V, Navon-Venezia S, Seigman-Igra Y, et al. Multidrug-resistant Pseudomonas aeruginosa: risk factors and clinical impact. Antimicrob Agents Chemother. 2006;50: 43–48. Dellit TH, Owens RC, McGowan JE Jr., et al. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing
1757
Clinical Therapeutics
19.
20.
21.
22.
23.
24.
25.
26.
an institutional program to enhance antimicrobial stewardship. Clin Infect Dis. 2007;44:159–177. Barlam TF, Cosgrove SE, Abbo LM, et al. Implementing an Antibiotic Stewardship Program: Guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis. 2016; 62:e51–e77. Kaki R, Elligsen M, Walker S, et al. Impact of antimicrobial stewardship in critical care: a systematic review. J Antimicrob Chemother. 2011; 66:1223–1230. Vettese N, Hendershot J, Irvine M, et al. Outcomes associated with a thrice-weekly antimicrobial stewardship programme in a 253-bed community hospital. J Clin Pharm Ther. 2013;38:401–404. Avdic E, Cushinotto LA, Hughes AH, et al. Impact of an antimicrobial stewardship intervention on shortening the duration of therapy for community-acquired pneumonia. Clin Infect Dis. 2012; 54:1581–1587. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40: 373–383. Jeong BH, Koh WJ, Yoo H, et al. Performances of prognostic scoring systems in patients with healthcare-associated pneumonia. Clin Infect Dis. 2013;56:625–632. Fang WF, Yang KY, Wu CL, et al. Application and comparison of scoring indices to predict outcomes in patients with healthcareassociated pneumonia. Crit Care. 2011;15:R32. American Thoracic Society, Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med. 2005;171:388–416.
1758
27. Cyriac JM, James E. Switch over from intravenous to oral therapy: A concise overview. J Pharmacol Pharmacother. 2014;5:83–87. 28. Yu K, Rho J, Morcos M, et al. Evaluation of dedicated infectious diseases pharmacists on antimicrobial stewardship teams. Am J Health Syst Pharm. 2014;71: 1019–1028. 29. Athanassa Z, Makris G, Dimopoulos G, et al. Early switch to oral treatment in patients with moderate to severe community-acquired pneumonia: a meta-analysis. Drugs. 2008;68:2469–2481. 30. Kyne L, Hamel MB, Polavaram R, et al. Health care costs and mortality associated with nosocomial diarrhea due to Clostridium
difficile. Clin Infect Dis. 2002;34: 346–353. 31. Nowak MA, Nelson RE, Breidenbach JL, et al. Clinical and economic outcomes of a prospective antimicrobial stewardship program. Am J Health Syst Pharm. 2012; 69:1500–1508. 32. Lew KY, Ng TM, Tan M, et al. Safety and clinical outcomes of carbapenem de-escalation as part of an antimicrobial stewardship programme in an ESBL-endemic setting. J Antimicrob Chemother. 2015; 70:1219–1225. 33. Dubberke ER, Burnham CA. Diagnosis of Clostridium difficile Infection: Treat the Patient, Not the Test. JAMA Intern Med. 2015;175: 1801–1802.
Address correspondence to: Kari A. Mergenhagen, PharmD, BCPS AQID, Infectious Diseases Department, Veteran Affairs Western New York Healthcare System, 3495 Bailey Avenue, Buffalo, NY 14215. E-mail: kari. [email protected]
Volume 38 Number 7