Real-time computer-generated alerts to select ...

Real-time computer-generated alerts to select interventions and optimize an antimicrobial stewardship program

IA01

P. DOLCE1, L. BERGERON2, E. LABRECQUE2, A. BOIVIN3, and H. BERNATCHEZ1. 1Department of Medical Microbiology and Infectious Diseases, CSSS Rimouski-Neigette (CSSSRN), 2Department of Pharmacy, CSSSRN, 3Department of Biology, Université du Québec à Rimouski,. ABSTRACT

PATIENTS AND METHODS

Objectives: Implementation of antimicrobial stewardship program (ASP) is a challenge, with limited physician and pharmacist time. A multi-disciplinary approach (clinical pharmacists, medical microbiologists, infectious diseases, information technology) was undertaken to integrate information technology in our ASP, with alerts to tract potential interventions.

•CSSSRN is a 230-beds acute-care regional hospital.

•A total of 1004 alerts were generated by the software, in 686 patients.

•Antibiokos (Nosotech) algorithms were designed and alerts included these core elements

•In comparison to previous year, Antimicrobials consumption (DDD) decreased by 14%, and costs by 34% (Table 1) • A total of 184 interventions were performed in 135 patients (Tables 2, 3).

•overlapping spectra

•Most interventions targeted Quinolones 23%, Pip-Tazo 21%, Cephalosporins 17%, Carbapenems 9%, Macrolides 5%, Antifungals 4%, Antivirals 3%.

•switch from intravenous to oral

•drug optimization according to culture results

•ASP interventions are shown in Table 4. One third of interventions resulted in replacement or discontinuation of therapy (Table 5)

•formulary restriction •pharmacokinetics (PK) •Alerts could be turned off automatically if the antimicrobial changed. Also, one drug could trigger more then one alert. •Interventions were recorded by the ASP team 2 to 4 times weekly.

Vancomycine Valaciclovir

14000

50

Trimethoprim/Sulfamethoxazole Trimethoprim Tobramycin

•All inpatients with oral or parenteral antimicrobials, from December 12, 2015 (P10) to March 31, 2015 (P13) were eligible for ASP intervention targeted with Antibiokos.

45

Tigecycline Rifampicin

43

12000

Posaconazole Piperacillin/Tazobactam

40

Penicillin V oral (PenVee) Penicillin G Parenteral

•Comparison of antimicrobials utilization with previous year was performed, including metrics (DDD, DOT, costs). Voriconazole

Table 1. Antimicrobial consumption per period P10-P13 8545 DDD 118404$ 19609JP

2500

Vancomycin

P10-P13 7348 DDD (-14%) 76548$ (-35%) 19609JP

Antibiokos Implementation

ASP Introduction

Trimethoprim Tobramycin Tetracycline

10000

Nystatin

33

Moxifloxacin Metronidazole

30 8000 25 23

23

6000

21

20

Meropenem

4000

Gentamicin Fluconazole

Daptomycin

15

14

Colistin Cloxacillin Clindamycin

10

Posaconazole

Clarithromycin Ciprofloxacin

2000

Cephalexin

5

Cefuroxime

5

Ceftriaxone

Nitrofurantoin

Ceftazidime

2

Moxifloxacin

1500

Imipenem

Doxycycline

Oseltamivir

2000

Levofloxacin

Ertapenem

Rifampicin

Phenoxymethylpenicillin (PenV)

Linezolid

Erythromycin

20

Trimethoprim/Sulfamethoxazole

Piperacillin/Tazobactam

Defined Daily Doses

Oseltamivir

35

0

0

Cefprozil

0

0

Metronidazole

Cefoxitin

Meropenem

Cefotaxime

Linezolid

Cefazolin

Levofloxacin

Caspofungin

Itraconazole

Azithromycin

Isoniazid

Ampicillin

Imipenem

Amphotericin B

Gentamicin

Amoxicillin/Clavulinic Acid

Fluconazole

Amoxicillin

Ethambutol

Aciclovir

Erythromycin

Interv

Ertapenem Doxycycline Daptomycin

1000

Table 4. ASP Interventions

Cloxacillin Clindamycin Clarithromycin Ciprofloxacin Cefuroxime Ceftriaxone Ceftazidime Cefprozil

INTRODUCTION

Voriconazole

Table 3. Antimicrobial Utilization and Interventions per ward

Minocycline

Conclusions: Realtime alerts provided a powerful tool to our ASP, with significant reduction of antimicrobial consumption and cost reduction

•The software provided significant time-reduction with estimates : >95% to get the metrics (DDD or DOT), and >50% to target and perform ASP interventions.

Interventions

Outcomes: During the first 4 months, a total of 7348 DDD were observed among inpatients, and 1004 electronic alerts were generated in 686 patients. The alerts were time-sensitive stop orders 55%, restriction 12%, PK 10%, optimization according to culture 6%, switch IV to oral 6%, overlapping spectra 6%, other 4%. A total of 184 interventions were done in 135 patients, mainly on Quinolones 23%, Pip-Tazo 21%, Cephalosporins 17%, Carbapenems 9%, Macrolides 5%, Antifungals 4%, Antivirals 3%. Suggestions were: no change 65%, replacement 23% and stop 12%. High acceptance rate of suggestions (95%) was observed. In comparison to the previous year, DDD were reduced by 14% and costs by 34%. The software provided significant time-reduction with estimates : >95% to get the metrics (DDD or DOT), and >50% to target and perform ASP interventions.

• time-sensitive stop orders (3 days for IV, 7 days for oral)

DDD/1000 patient-days

Methods: CSSSRN is a tertiary regional hospital with 230 acute-care beds. A software “Antibiokos” was implemented in November 2014 to enable our ASP team to tract interventions, using real-time data from interfaces with pharmacy, microbiology laboratory and admission-discharge-transfer. Electronic algorithms included core elements of ASP such as: time-sensitive stop orders (3 days for IV, 7 days for oral), overlapping spectra, switch from intravenous to oral, drug optimization according to culture results, formulary restriction and pharmacokinetics (PK). Real-time metrics (DDD, DOT, Costs) were provided.

RESULTS

Cefoxitin

500

Cefotaxime Cefixime Cefazolin Cefalexin

Pharmacokinetics 16 9%

Switch IV to oral 30 18%

Formulary restriction 9 5%

Overlapping spectra 7 4% Physician request 1 1%

Table 5. Intervention Results

Replacement 43 23%

Stop 22 12%

Caspofungin Azithromycin Ampicillin

0 1

2

3

4

5

6

7

8

9

10

11

12

13

1

2

3

4

5

6

7

8

9

10

11

12

Amoxicillin/Clavulinic Acid

Periods

Amoxicillin

2013-2014

2014-2015

•However, measurements of metrics was cumbersome, and it was difficult to record interventions.

Aciclovir

45

1000

Emergency Room

30 600

DDD

25 500

20 17

400

300

3

2 0

0

0

1

0

2 0

1

3

2 0

0

0

2

3

4

5 2

0

0

0

0

0

0

0

1

0

Trimethoprim

Tobramycin

Tigecycline

Rifampicin

Posaconazole

Piperacillin/Tazobactam

Penicillin V oral (PenVee)

Penicillin G Parenteral

Oseltamivir

Nystatin

Moxifloxacin

Metronidazole

Meropenem

Linezolid

Levofloxacin

Imipenem

Gentamicin

Fluconazole

Erythromycin

Ertapenem

Doxycycline

Daptomycin

Colistin

Cloxacillin

Clindamycin

Clarithromycin

Ciprofloxacin

Cephalexin

Cefuroxime

Ceftriaxone

Ceftazidime

Cefprozil

Cefoxitin

Cefotaxime

Cefazolin

Caspofungin

Azithromycin

Ampicillin

Amphotericin B

Amoxicillin/Clavulinic Acid

Amoxicillin

Aciclovir

0

Other authors : none

5C Surgery 5A Psychiatry 4C Orthopedics

3C Oncology

6

2

2

2C Medicine - Geriatrics

• Information technology greatly improved the efficiency of our ASP team, with friendly-user software. • Realtime alerts provided a powerful tool to our ASP, with significant reduction of human resources, antimicrobial consumption and cost reduction.

2A Cardiology 0

0 Voriconazole

4

5

Vancomycine

1

5

10

Valaciclovir

5

4

3

9

8

Trimethoprim/Sulfamethoxaz…

9

6C Pediatrics

4A OBG

11

DISCUSSION AND CONCLUSIONS

NICU

27

200

Conflict of interest : P Dolce is President of Nosotech Inc.

ICU

35

700

100

No change 119 65%

40

15

•This retrospective study presents the results of the Pilot Study.

Time-sensitive stop orders 96 57%

40

900

800

•Algorithms were designed to tract interventions opportunities.

Optimization according to cultures 8 5%

Optimization according to guidelines 1 1%

Table 2. Utilization and Interventions per antimicrobial

•Information technology with artificial intelligence has the potential to provide useful realtime assistance to ASP teams. •A pilot study was undertaken with the introduction of Antibiokos software (with realtime data from interfaces with Admission-Discharge-Transfer, Microbiology Lab and Pharmacy systems) in November 2014.

Amphotericin B

13

Interventions

•An ASP program was implemented in our facility with a pharmacist 12 hours/week and Infectious Disease specialist 3 hours/week in September 2014.

Benzylpenicillin (Pen G)

1C Psychiatry Interventions

Correspondence, reprints or requests to: Dr Patrick Dolcé MD FRCPC Department of Medical Microbiology and Infectious Diseases CSSS Rimouski-Neigette 150, Rouleau Rimouski, Québec Canada, G5L 5T1 Telephone (418) 724-3000 #8336 Fax (418) 724-8619 e-mail: [email protected]