An evaluation of provider-chosen antibiotic indications as a targeted antimicrobial stewardship intervention

Design

Retrospective analysis 

N= 154 antibiotic orders

Objective

To determine whether the provider-chosen indication for selected antibiotics matched the patient’s diagnosis documented in the electronic medical record (EMR) and whether it was clinically appropriate

Study Groups

Vancomycin (n= 81)

Cefepime (n= 27)

Fluoroquinolones (n= 46)

Inclusion Criteria

Random samples of eligible antibiotic orders via the electronic health system

Exclusion Criteria

Patients younger than age 18 years; antibiotics ordered via an order set or entered by a pharmacist

Methods

At the time of ordering 1 of these medications, providers were required (via hard stop) to either pick a prepopulated indication or choose “other” and type in an indication for use.

Orders were determined to match if the diagnosis in the note(s) mirrored the indication in the order. Orders mismatched were further characterized into 2 categories: complete mismatch and failure of indication. A complete mismatch was defined as the diagnosis documented in the EMR was a different type of infection from what was indicated in the order. A failure of indication was defined as the indication in the order was the closest available indication choice to what was documented in the EMR (ie, the indication chosen was “abscess with risk for MRSA,” but there was no indication choice for “cellulitis with risk for MRSA,” and the EMR stated “cellulitis with risk for MRSA”). To characterize the “other” indication selection used for each antibiotic, the team read through the comments entered by the ordering provider and summarized the indication by the syndrome.

Duration

From April 1, 2016, to April 30, 2016

Outcome Measures

Matching and clinical appropriateness of targeted antibiotic agents

Baseline Characteristics

Vancomycin (n= 81)

Cefepime (n= 27)

Age, years

Female

Type of therapy

Empiric

Targeted

70 (86%)

11 (14%) 

25 (93%)

2 (7%)

39 (86%)

7 (14%)

Location

ICU

Floor

23 (28%)

58 (72%)

8 (29%)

19 (71%)

1 (2.5%)

45 (97.5%)

Results

Endpoint

Total

Matching

Odds Ratio (OR) (selected vs. other); p-Value

Vancomycin

Selected

Other

81

55

26

65 (80.0%)

41 (74.5%)

24 (92.3%)

76 (94%)

52 (94.5%)

24 (92.3%)

Cefepime

Selected

Other

27

21

6

21 (77.7%)

15 (71.4%)

6 (100%)

27 (100%)

21 (100%)

6 (100%)

Fluoroquinolones

Selected

Other

46

15

31

35 (74.0%)

8 (53.3%)

27 (87.1%)

30 (68.0%)

13 (86.7%)

17 (54.8%)

When providers chose indications from the list as opposed to choosing “other” and entering free text, antibiotic orders were significantly more likely to be appropriate (OR 5.8; p= 0.001) but also less likely to match clinical documentation (OR 0.25; p= 0.0043).

Adverse Events

N/A

Study Author Conclusions

Selecting an indication from an evidence-based list as opposed to free-text indications increases the odds that antibiotic agents will be used appropriately.

InpharmD Researcher Critique

This retrospective analysis is for institution-specific antimicrobial stewardship intervention and thus may limit generalizability.

A mandatory indication-registration tool in hospital electronic medical records enabling systematic evaluation and benchmarking of the quality of antimicrobial use: a feasibility study

Design

Multicenter, feasibility study

N= 3 hospitals; 3,021 antibiotic prescriptions for respiratory tract infections (RTI) or urinary tract infections (UTI)

Objective

To investigate the real-life feasibility of mandatory documentation of the indication for all antibiotic prescriptions, for the purpose of systematic evaluation of not only the extent but also the appropriateness of antimicrobial use

Study Groups

Hospital A (n= 1,910 antibiotic prescriptions for RTI or UTI)

Hospital B (n= 130 antibiotic prescriptions for RTI or UTI)

Hospital C (n= 981 antibiotic prescriptions for RTI or UTI)

Inclusion Criteria

Antibiotic prescriptions for all hospitalized patients aged 18 years and older, admitted to any general ward (admitted to the ward for at least 12 hours) and receiving empiric antibiotic treatment (prescribed antibiotic [combination] therapy within 24 hours of hospitalization or the last prescribed antibiotic therapy at the time of discharge in patients who were hospitalized for 12 to 24 hours) for an RTI or UTI

Exclusion Criteria

Intensive care unit patients, readmissions (defined as an admission within 30 days after the initial hospital discharge), prescriptions of patients with both RTI and UTI, and erroneous prescriptions

Methods

A standardized prescription format was implemented in each hospital's electronic medical record (EMR) that required providers to select an indication for the prescription from a predefined list whenever a systemic antimicrobial was prescribed. The possible indications that physicians could select were empiric therapy, targeted therapy, or prophylaxis. The physicians also had to select the main focus of the infection, first on tract level, followed by a further specification.

Duration

Data review period:

Hospital A: January 1, 2017, until December 31, 2017

Hospital B: June 1, 2019, until October 31, 2019

Hospital C: May 14, 2019, until June 9, 2020

Outcome Measures

Primary: the accuracy of the dataset, defined as the percentage agreement between the selected indication for the prescription and the documented indication in the EMR

Secondary: the percentage of antibiotic prescriptions in each hospital that was prescribed according to the national guidelines

Results

Endpoint

Hospital A
(n= 300 prescriptions)

Hospital B
(n= 243 prescriptions)

Selected indication did not match documented diagnosis^*

Hospital A
(n= 1,900 prescriptions)

Guideline adherence rate

Overall community-acquired pneumonia (CAP)

Mild to moderate-severe CAP

Severe CAP

Complicated UTI

Cystitis

49.5%

--

--

67.1%

45.3%

--

33.4%

35.4%

40.0%

5.6%

--

38.5%

53.0%

56.6%

28.1%

^*Indication selection errors were mostly due to inaccurate sub-indications

Study Author Conclusions

The real-life feasibility of mandatory documenting the indication of all antibiotics prescribed in EMR using ChipSoft HIX or EPIC software for quality evaluation purposes has been demonstrated. It enables a reliable and time-efficient method for the systematic registration of the extent and appropriateness of empiric antimicrobial use. However, initial local validation and, if necessary, optimization of the datasets are required to assure the accuracy of the extracted data.

InpharmD Researcher Critique

Mandatory documentation of the indication of antibiotics was shown to be feasible and can be beneficial for antimicrobial stewardship efforts. Programs have to be specific to hospitals though, as even hospitals using the same EMR systems will have local variations in the details of its use and workflow process.

Antibiotic prescribing without documented indication in ambulatory care clinics: national cross sectional study

Design

Cross-sectional study 

N= 130,502,788 antibiotics prescribed during visit 

Objective

To identify the frequency with which antibiotics are prescribed in the absence of a documented indication in the ambulatory care setting, to quantify the potential effect on assessments of appropriateness of antibiotics, and to understand patient, provider, and visit level characteristics associated with antibiotic prescribing without a documented indication

Study Groups

Appropriate prescriptions (n= 74,220,933)

Inappropriate prescriptions (n= 32,553,713)

No documented indications (appropriateness unknown; n= 23,728,142)

Inclusion Criteria

Patients with at least one antibiotic prescription in the analyses

Exclusion Criteria

Not specified 

Methods

Encounter diagnosis codes (international classification of diseases, 9th revision, clinical modification [ICD-9-CM]) were reviewed to identify bacterial infections or other conditions for which antibiotics are frequently prescribed. Researchers considered indications to be “appropriate” if any bacterial infection or other condition for which antibiotics are always or sometimes indicated was documented, “inappropriate” if only a condition for which antibiotics are not indicated but are commonly prescribed (for example, upper respiratory tract infection) was documented, or “no documented indication” if neither of the preceding categories as applicable.

A multivariable survey-weighted logistic regression model was used to identify risk factors associated with antibiotic prescriptions with no documented indication. 

Duration

The National Ambulatory Medical Care Survey (NAMCS) in 2015

Outcome Measures

Baseline Characteristics

Inappropriate indication
(n= 32,553,713)

Age, years*

< 18

18-64

≥ 65

78%

56%

40%

13%

24%

38%

8%

20%

22%

Female

White

Non-Hispanic black

56%

50%

25%

26%

18%

24%

Provider type*

Primary care 

Common prescribers†

All other specialties 

69%

43%

35%

18%

32%

36%

12%

24%

29%

Culture taken*

71%

24%

5%

Antibiotic classes* 

Cephalosporins

Macrolides 

Penicillin 

Quinolones

Sulfonamides 

Tetracyclines 

Urinary anti-infectives 

64%

67%

72%

42%

32%

61%

33%

16%

19%

18%

38%

16%

30%

33%

20%

14%

11%

20%

52%

18%

34%

Main reason for visit*

New problem

Chronic problem, routine 

Chronic problem, flare-up

Pre-surgery 

Post-surgery 

Preventative care 

71%

45%

50%

3%

43%

31%

20%

29%

32%

68%

37%

21%

9%

26%

18%

29%

21%

49%

*p< 0.001; †Includes specialists in gynecology, urology, dermatology, and otolaryngology

Results

Adjusted odds ratios for association between significant independent predictors and antibiotic prescription without documented indication

Predictors 

Age and sex

Males: ≥ 18 vs < 18

Females: ≥ 18 vs < 18

2.3 (1.02 to 5.3)

1.1 (0.6 to 2.2)

Specialty

Primary care

All other specialists 

Common prescribers

Reference

2.1 (1.2 to 3.7)

1.9 (1.1 to 3.3)

Longer visit: ≥ 17 min vs < 17 min

Culture taken 

Antibiotic class

Penicillin

Sulfonamides

Urinary anti-infectives

Reference 

4.9 (1.5 to 15.7)

3.1 (1.3 to 7.6)  

Study Author Conclusions

This nationally representative study of ambulatory visits identified a large number of prescriptions for antibiotics without a documented indication. Antibiotic prescribing in the absence of a documented indication may severely bias national estimates of appropriate antibiotic use in this setting. This study identified a wide range of factors associated with antibiotic prescribing without a documented indication, which may be useful in directing initiatives aimed at supporting better documentation.

InpharmD Researcher Critique