Antibiotic Approval Program on Patients

1
Antibiotic Approval Program on
Patients Clinical Outcomes and
Antibiotic Expenditures
at Siriraj Hospital Pinyo Rattanaumpawan
MD Patama Sutha
MD Visanu Thamlikitkul MD Department of Medicine
Faculty of
Medicine Siriraj Hospital
Mahidol University, Bangkok, Thailand
2
Siriraj Hospital Profile
Faculty of Medicine Siriraj Hospital, Mahidol
University

• Tertiary Care University Hospital in Bangkok
• 2,300 Hospital Beds
• 2,600,000 Out-patients visits per year
• 100,000 In-patients per year
• 1,400 Physicians (700 residents clinical
  fellows)

3

• Background
• Thailand National List of Essential Medicines
• Drugs in category D should be approved by
• qualified specialists and DUE should be
• conducted to avoid inappropriate use and
  
• to promote appropriate use of such drugs

4

• Background
• Thailand National List of Essential Medicines
• Drugs in category D should be approved by
• qualified specialists and DUE should be
• conducted to avoid inappropriate use and
  
• to promote appropriate use of such drugs
• Many antibiotics are listed in category D
• including
• Piperacillin/Tazobactam
• Imipenem
• Meropenem

• very broad spectrum
• last line therapy for
• resistant bacteria
• very expensive

5

• Background
• Cost of Piperacillin/Tazobactam, Imipenem
• and Meropenem at Siriraj Hospital in 2006
• 4,000,000

6

• Background
• Cost of Piperacillin/Tazobactam, Imipenem
• and Meropenem at Siriraj Hospital in 2006
• 4,000,000
• DUE of these 3 drugs in 2007 revealed that
• Inappropriate use at prescription 10
• Inappropriate use at continued treatment
• (2-3 days after initial prescription) 50

7

• Background
• Cost of Piperacillin/Tazobactam, Imipenem
• and Meropenem at Siriraj Hospital in 2006
• 4,000,000
• DUE of these 3 drugs in 2007 revealed that
• Inappropriate use at prescription 10
• Inappropriate use at continued treatment
• (2-3 days after initial prescription) 50
• Hospital administrators were concerned on
• inappropriate use and cost of these 3 drugs

8

• Background
• Siriraj Hospital decided to have antibiotic
• approval measure and DUE of these 3 drugs
• according to recommendation of Thailand
• National List of Essential Medicines

9

• Background
• Siriraj Hospital decided to have antibiotic
• approval measure and DUE of these 3 drugs
• according to recommendation of Thailand
• National List of Essential Medicines
• Responsible physicians raised a concern on
• negative effect of this measure on clinical
• outcomes for patients whose antibiotics are
• changed or discontinued

10

• Background
• Siriraj Hospital decided to have antibiotic
• approval measure and DUE of these 3 drugs
• according to recommendation of Thailand
• National List of Essential Medicines
• Responsible physicians raised a concern on
• negative effect of this measure on clinical
• outcomes for patients whose antibiotics are
• changed or discontinued
• No solid evidence from literature to support
• such negative effect

11

• Background
• Siriraj Hospital decided to have antibiotic
• approval measure and DUE of these 3 drugs
• according to recommendation of Thailand
• National List of Essential Medicines
• Responsible physicians raised a concern on
• negative effect of this measure on clinical
• outcomes for patients whose antibiotics are
• changed or discontinued
• No solid evidence from literature to support
• such negative effect
• Siriraj Hospital endorsed a study to convince
• responsible physicians on patients outcomes

12

• Objective
• To determine effectiveness of antibiotic
• approval measure on patients clinical
• outcomes and antibiotic expenditures

13

• Methods
• Randomized controlled study in hospitalized
• adult patients who received piperacillin/
• tazobactam, imipenem and meropenem from
• August to November 2007

14

• Methods
• Randomized controlled study in hospitalized
• adult patients who received piperacillin/
• tazobactam, imipenem and meropenem from
• August to November 2007
• Once there was a prescription of any target
• antibiotics, Pharmacy Department notified
• study team
• Study team allocated the target antibiotic
• prescription to either control group or
• antibiotic approval group by simple
• randomization

15

• Methods
• Control Group
• Patients received antibiotics according to
• their physicians decision
• Study team observed patients clinical
• courses and outcomes until the patients left
• the hospital

16

• Methods
• Antibiotic Approval Group
• Patients were assessed by infectious disease
• specialist to determine if target antibiotics
• were indicated according to predetermined
• criteria

17

• Predetermined Criteria
• Piperacillin/Tazo., Imipenem, Meropenem
• Confirmed or suspected P.aeruginosa infection
• Infection due to pathogen resistant to
• cephalosporins, aminoglycosides, quinolones
• Empiric therapy for neutropenic fever
• Infection due to pathogen susceptible to other
  antibiotics but the patient is unable to receive
  such antibiotics

18

• Predetermined Criteria
• Additional criteria for imipenem meropenem
• Therapy of nosocomial infection which is not
  responding to other antibiotics
• Infection due to pathogen which is resistant to
  beta-lactam beta-lactamase inhibitor
• Severe infection due to ESBLve pathogen

19

• Methods
• Antibiotic Approval Group
• Patients were assessed by infectious disease
• specialist to determine if target antibiotics
• were indicated according to predetermined
• criteria
• If target antibiotic was indicated, it was
• continued
• If target antibiotic was not indicated, study
• team suggested stopping or modifying it

20

• Methods
• Antibiotic Approval Group
• Patients were assessed by infectious disease
• specialist to determine if target antibiotics
• were indicated according to predetermined
• criteria
• If target antibiotic was indicated, it was
• continued
• If target antibiotic was not indicated, study
• team suggested stopping or modifying it
• Study team observed patients clinical
• courses and outcomes until the patients left
• the hospital

21

• Clinical Outcomes
• Favorable outcomes Cure Improvement
• Eradication of causative pathogens
• Length of hospital stay
• Duration of fever
• Side effects of antibiotics
• Death due to infection
• Status at discharge

22

• Antibiotic Consumption Expenditure
• Duration of all antibiotics
• Duration of target antibiotics
• Total cost of all antibiotics
• Total cost of target antibiotics
• Defined Daily Dose (DDD) of all antibiotics
  
• DDD of target antibiotics

23

• Methods
• Sample Size
• Non-inferiority study
• Favorable outcome in control gr. 70
• Favorable outcome in intervention group
• was lt 5 of control group
• Type I error 5 and Type II error 20
• 500 prescriptions in each group

24

• Results
• 1,028 prescriptions of imipenem, meropenem,
• piperacillin/tazobactam in 948 patients
• 516 prescriptions (486 patients) - Control
• 512 prescriptions (462 patients) - Intervention

25

• Characteristics of Patients
• Characteristic Control Intervention
  p (N516) (N512)
  
• Males 52.9 53.0 0.9
• Mean Age 62.1 y. 63.5 y.
  0.2
• Location of Patients
  
• Medical Wards 68.9 70.8
  0.7
• Surgery Wards 25.1 24.5
• Type of Infections
  
• CAI 18.0
  20.9 0.2
• HAI 82.0 79.1

26

• Characteristics of Patients
• Characteristic Control Intervention
  p (N516) (N512)
• Purpose of Antibiotic Usage
• Treatment 98.6 98.6
  1
• Prophylaxis 1.4 1.4
• Type of Target Antibiotic
  
• Pip./Tazobac. 46.9 44.5
  0.6
• Imipenem 35.4 35.7
• Meropenem 17.6 19.7

27

• Inappropriate Prescriptions
• 42 of target antibiotic prescriptions were
• considered inappropriate

28

• Compliance to Suggestions
• 42 of target antibiotic prescriptions were
• considered inappropriate
• 46 of inappropriate prescriptions were changed
  to more appropriate antibiotics
• according to suggestions

29

• Compliance to Suggestions
• 42 of target antibiotic prescriptions were
• considered inappropriate
• 46 of inappropriate prescriptions were changed
  to more appropriate antibiotics
• according to suggestions
• 24 of inappropriate prescriptions were
  discontinued according to suggestions
• (antibiotics were not needed)

30

• Compliance to Suggestions
• 42 of target antibiotic prescriptions were
• considered inappropriate
• 46 of inappropriate prescriptions were changed
  to more appropriate antibiotics
• according to suggestions
• 24 of inappropriate prescriptions were
  discontinued according to suggestions
• 22 of inappropriate prescriptions were changed
  to other inappropriate antibiotics not suggested
  by infectious disease specialist
  (non-compliance rate squeezing the balloon)

31
Outcomes of Patients Clinical Outcome
Control Intervention p (N516)
(N512) Favorable Outcome 60.5
68.9 lt0.01 Eradicate Pathogens 50.2
65.4 lt0.01 Mean LOS
30.7 d. 30.4 d. 0.8 Mean Length
Fever 11.0 d. 7.5 d. lt0.01
Antibiotic Allergy 1.4 0.4
0.1 AAD 3.5
4.9 0.2 Death from Infection
35.4 29.4 0.05 Mean d. all ATB
16.4 d. 12.7 d. lt0.01 Mean d.
target ATB 9.3 d. 7.5 d. lt0.01
Alive at Discharge 57.4 55.6
0.6
32
Outcomes of Patients Clinical Outcome
Control Intervention p (N516)
(N512) Favorable Outcome 60.5
68.9 lt0.01 Eradicate Pathogens 50.2
65.4 lt0.01 Mean LOS
30.7 d. 30.4 d. 0.8 Mean Length
Fever 11.0 d. 7.5 d. lt0.01
Antibiotic Allergy 1.4 0.4
0.1 AAD 3.5
4.9 0.2 Death from Infection
35.4 29.4 0.05 Mean d. all ATB
16.4 d. 12.7 d. lt0.01 Mean d.
target ATB 9.3 d. 7.5 d. lt0.01
Alive at Discharge 57.4 55.6
0.6
33
Antibiotic Consumption Expenditure
Control
Intervention (N516)
(N512) Total Cost all ATB (m.) 0.71
0.54 Total Cost target ATB (m.) 0.34
0.28 All ATB/Episode ()
1,238 967 Target ATB/Episode ()
661 538 DDD all ATB
13,528 10,738 DDD target ATB
3,696 2,972 DDD all
ATB/Episode 26.2 21.0 DDD
target ATB/Episode 7.2 5.8
34
Antibiotic Consumption Expenditure
Control
Intervention (N516)
(N512) Total Cost all ATB (m.) 0.71
0.54 Total Cost target ATB (m.) 0.34
0.28 All ATB/Episode ()
1,238 967 Target ATB/Episode ()
661 538 DDD all ATB
13,528 10,738 DDD target ATB
3,696 2,972 DDD all
ATB/Episode 26.2 21.0 DDD
target ATB/Episode 7.2 5.8
Minimal Squeezing the Balloon Effect
35

• Antibiotic Consumption Expenditure
• Actual difference in antibiotic expenditures
  between antibiotic approval group and control
  group was 143,793
• Annual antibiotic cost savings from antibiotic
  approval program would be 862,704 if all
  prescriptions of target antibiotics were reviewed
  by infectious disease physicians

36
Conclusion Antibiotic approval program for
imipenem, meropenem, piperacillin/tazobactam at
Siriraj Hospital is an effective measure
in reducing antibiotic consumption expenditure
without compromising patients clinical outcomes
37

• Cost-Benefit of Antibiotic Approval Program
• If all prescriptions of target antibiotics are
  reviewed by infectious diseases physicians,
  hospital will have annual saving US 862,704
• Extra costs of DUE and antibiotic approval
  processes are not taken into account when
  aforementioned amount of cost savings is
  estimated
• Extra costs generated by DUE and antibiotic
  approval (personnel costs and additional costs
  for microbiologic investigations) are estimated
  as up to US 50,000 per year
• DUE antibiotic approval are cost-benefit

38

• Prerequisite of Antibiotic Approval Program
• Antibiotic approval can be applied only to
  institutions where there are qualified infectious
  diseases physicians who are willing to take
  responsibility in approving restricted
  antibiotics and are able to continue providing
  suggestions regarding antibiotic prescription and
  management of infections

39

• Prerequisite of Antibiotic Approval Program
• Antibiotic approval can be applied only to
  institutions where there are qualified infectious
  diseases physicians who are willing to take
  responsibility in approving restricted
  antibiotics and are able to continue providing
  suggestions regarding antibiotic prescription and
  management of infections
• Well trained clinical pharmacists should be able
  to perform approval

40

• Implementations
• Study results were presented to hospital
  administrators in April 2008
• Siriraj Hospital informed all responsible
  physicians regarding the study results
• Antibiotic approval measure has been continued
  since May 2008

41

• More Information
• Poster Number 291
• American Journal of Infection Control
• 2010 38 38 - 43.

42
Antibiotic Approval Program on
Patients Clinical Outcomes and
Antibiotic Expenditures
at Siriraj Hospital Pinyo Rattanaumpawan
MD Patama Sutha
MD Visanu Thamlikitkul MD Department of Medicine
Faculty of
Medicine Siriraj Hospital
Mahidol University, Bangkok, Thailand
43
Characteristics of Patients Characteristic
Control Intervention p
(N516) (N512) Underlying Conditions of
Patients Cardio-pulmonary 38.7
54.1 lt0.01 Malignancy 34.0
42.0 0.01 Liver Diseases
6.6 11.7 0.01 Renal
Diseases 16.3 22.3
0.02 Neurological Dis. 15.4
22.9 0.01
44

• Clinical Outcomes of Patients
• Different characteristics between 2 groups
• Multivariate analysis was performed
• Factors associated with unfavorable outcome
• OR (95CI) p
• Control group patient 1.35 (1.03-1.76) 0.03
• Pneumonia 2.36 (1.78-3.15)
  lt0.01

45

• Impact of Antibiotic Approval Program on
  Antimicrobial Resistance
• Impact on antibiotic resistance of pathogens was
  not measured because patients allocated to both
  groups stayed in the same wards, and duration of
  study was only 4 months
• Therefore, it would be unlikely to observe any
  changes in antibiotic resistance patterns
• of isolated pathogens either from patients or
• from environment
• There are many factors contributing to AMR

46
Subgroup Analysis in Microbiological Documented
Infections Clinical Outcome Control Intervention
p (N290) (N323) Favorable Outcome
58.9 65.6 0.08 Eradicate
Pathogen 50.2 65.4 lt0.01 Death
(Infection) 37.9 31.9 0.1
47

• Main Reasons for Non-Compliance
• Responsible physicians did not believe that
  patient was free of active infection according to
  infectious disease physicians opinion
• Some surgeons believed that antibiotic
  prophylaxis should be continued until
  endotracheal tube was removed
• Many responsible physicians were not certain
  that narrow spectrum antibiotics could replace
  target antibiotics that they initially gave to
  their patients, especially for the patients who
  were improving after receiving the target
  antibiotics

48

• Different Rates of Comorbidity
• Infectious diseases physicians did not involve
  in care processes in control group whereas
• infectious diseases physicians continued
  supervising management of patients who received
  antibiotic approval
• Additional medical information was recorded, and
  additional microbiologic
• investigations for diagnosis of infections and
  for assessment of microbiologic outcomes were
  made for antibiotic approval group

49

• Different Rates of Comorbidity
• Differences in characteristics of patients would
  not favor outcomes of patients who received
  antibiotic approval because higher rates of
  having comorbidity that compromised clinical
  outcomes were observed more often in patients who
  received antibiotic approval

50

• Inappropriate Use of Target Antibiotics
• Prevalence of inappropriate use of target
  antibiotics in antibiotic approval group was
  41.6 which was modestly less than 50 observed
  from a preliminary DUE period
• This might be due to each prescription of target
  antibiotics during study period was required to
  be accompanied with antibiotic order sheet that
  was filled in by responsible physician
• Therefore, some prescriptions of target
  antibiotics were cancelled earlier if responsible
  physicians thought that their prescriptions were
  obviously inappropriate

51

• Clinical Outcomes in Antibiotic Approval
• Patients in antibiotic approval group had better
  outcomes
• Infectious diseases physicians not only approved
  the use of target antibiotics, but they also
  recommended appropriate microbiologic
  investigations so that the site of infections and
  the causative agents became known
• Antibiotics given to patients in antibiotic
  approval group could either be discontinued with
  certainty in patients with no infections or
  tailored to more specific antibiotics according
  to microbiologic results

52

• Clinical Outcomes in Antibiotic Approval
• Patients in antibiotic approval group had better
  outcomes
• Many patients in control group who had no
  infections or had colonization or had infections
  with resistant pathogens, (e.g. MRSA) continued
  to receive target antibiotics leading to
  super-infections with organisms (e.g. MRSA,
  pandrug-resistant Acinetobacter spp, or Candida
  spp.) which were resistant to target antibiotics,
  resulting in unfavorable outcomes and mortality

53

• Clinical Outcomes in Antibiotic Approval
• Although some baseline characteristics of
  patients in both groups were significantly
  different, the results of multivariate analysis
  showed that being a patient in control group and
  having pneumonia was significantly
• associated with unfavorable outcome
• This observation confirmed that a difference in
  favorable clinical outcomes was associated with
  antibiotic approval program because
  hospital-acquired pneumonia was a well-known
  factor associated with poor clinical
• outcome of hospitalized patients

54

• Clinical Outcomes in Antibiotic Approval
• Length of hospital stay and patients survival
  at hospital discharge were not significantly
  different in either group because most of
  patients in both groups had chronic underlying
  diseases, and they were hospitalized because of
  these medical problems and died of such
  comorbidities

55

• Squeezing the Balloon Effect
• Consumption of antibiotics other than target
  antibiotics were also collected because
  physicians might shift from target antibiotics to
  other antibiotics if target antibiotics were not
  approved
• 22 of prescriptions in antibiotic approval
  group were changed to other antibiotics which
  were not recommended by infectious diseases
  physicians
• Nonetheless, DDD of the target antibiotics and
  all antibiotics as well as DDD of target
  antibiotics and all antibiotics per
  episode of infection in antibiotic approval group
  were much lower than those in control group
• Therefore, squeezing the balloon effect
  found in our study was minimal

56

• Squeezing the Balloon Effect
• Consumption of antibiotics other than target
  antibiotics were also collected because
  physicians might shift from target antibiotics to
  other antibiotics if target antibiotics were not
  approved
• 21.6 of prescriptions in antibiotic approval
  group were changed to other antibiotics, such as
  cefepime, which were not recommended by
  infectious diseases physicians

57

• Squeezing the Balloon Effect
• Nonetheless, DDD of the target antibiotics and
  all antibiotics as well as DDD of target
  antibiotics and all antibiotics per episode of
  infection in antibiotic approval group were much
  lower than those in control group
• Therefore, the squeezing the balloon effect
  found in our study was minimal

58

• Defined Daily Dose (DDD)
• DDD is a measure of assumed average
• maintenance dose per day for a drug used as
• a main indication in adults

59
Defined Daily Dose (DDD) DDD (Amoxicillin 250
mg) dose form of strength (250mg)Quant. of
amoxi. dispensed WHO assigned DDD for
Amoxicillin (1000 mg) 25020 / 1000 5 DDD
(Erythromycin 250 mg) dose form of strength
(250mg)Quant. of erythro. dispensed
WHO assigned DDD for erythromycin (500 mg)
25020 / 500 10
60
Antibiotic Consumption Expenditure Antibiotic
Approval Program Antibiotic Cost Saving Control
- Intervention 4.8 mB for 4 months Annual
Antibiotic Cost Saving 14.4 mB Antibiotic Cost
Saving 4.5 mB for 4 months if 516 prescriptions
in the control group were reviewed by infectious
diseases specialist
61
(No Transcript)
62
(No Transcript)
63
(No Transcript)
64

• Methods
• Sample Size
• Non-inferiority study
• Favorable outcome in control gr. 70
• Favorable outcome in intervention group
• was lt 5 of control group
• Type I error 5 and Type II error 20
• 500 prescriptions in each group
• Statistics
• Descriptive statistics, Chi-square/ Fishers
• exact test, Student t-test/ Mann-Whitney
  
• U-test, multivariate analysis