1359. Developing a Regional Antibiogram for Community Hospitals
Session: Poster Abstract Session: HAI: Epidemiologic Methods
Friday, October 28, 2016
Room: Poster Hall

Background: Community hospitals may not isolate sufficient numbers of individual organisms to develop an accurate annual antibiogram. We aimed to develop a clinically useful regional antibiogram by combining the microbiologic data for numerous community hospitals within a single geographic region.

Methods: Calendar year 2012 antibiogram data from 20 DICON hospitals located in central and eastern NC and south-central VA were combined for 13 clinically relevant combinations of bacterial organisms and antibiotics.  The mean (standard deviation, SD) susceptibility for the region was calculated for each combination. Modified forest plots were then developed to depict each hospital’s relationship to the regional mean.

Results: 69,778 tested isolates across the 13 clinically relevant pathogen/test combinations (median for each combination 1100, range 174-27,428) were used to develop a regional antibiogram.  Depending on the pathogen/test combination, 50-82% of individual hospitals’ antibiogram susceptibility fell within one SD of the regional susceptibility, while 94-100% of individual hospitals’ antibiogram susceptibility fell within two SD of the regional susceptibility (Table 1).  For pathogens with particularly low numbers of isolates (<500 for the region), all hospitals’ reported susceptibilities fell within two SD of the regional susceptibility.

Conclusion: Regional antibiograms may provide useful adjunctive information for small hospitals with low numbers of isolates.

Table 1. Regional susceptibilities among 20 DICON hospitals.

Pathogen

Test

Regional Percent Susceptible (SD)

N (%) Hospitals within 1 SD

N (%) Hospitals within 2 SD

P. aeruginosa

FQ

70 (7.57)

13 (68)

18 (95)

P. aeruginosa

Pip/Tazo

89 (8.80)

13 (76)

16 (94)

S. maltophilia

T/S

93 (5.68)

4 (57)

7 (100)

S. maltophilia

Levofloxacin

81 (8.44)

5 (71)

7 (100)

A. baumannii

Amp/Sul

80 (12.79)

4 (57)

7 (100)

A. baumannii

Carb

69 (21.88)

4 (50)

8 (100)

E. cloacae

FQ

88 (7.99)

11 (61)

17 (94)

E. cloacae

Carb

98 (2.20)

14 (82)

17 (100)

P. mirabilis

FQ

66 (10.68)

13 (65)

19 (95)

E. coli

FQ

71 (5.76)

14 (70)

19 (95)

E. coli

T/S

70 (5.56)

15 (75)

19 (95)

C. freundii

FQ

85 (7.66)

8 (67)

12 (100)

M. morganii

FQ

61 (9.96)

4 (67)

6 (100)

Figure 1. Hospital versus Regional Susceptibilities for E. coli and fluoroquinolones.

Title: Modified forest plot of susceptibility of E.coli to Fluoroquinolones

Christopher Hostler, MD, MPH1, Rebekah W. Moehring, MD, MPH1,2,3, Elizabeth Dodds Ashley, PharmD, MHS, FCCP, BCPS1,3,4, Melissa Johnson, PharmD, MHS1,3, Angelina Davis, PharmD, MS1,3, Sarah S. Lewis, MD MPH1,2, Daniel J. Sexton, MD, FIDSA, FSHEA1,2,3 and Deverick Anderson, MD, MPH, FIDSA, FSHEA1,2,3, (1)Duke University Medical Center, Durham, NC, (2)Duke Infection Control Outreach Network (DICON), Durham, NC, (3)Duke Antimicrobial Stewardship Outreach Network (DASON), Durham, NC, (4)Duke Antimicrobial Stewardship Outreach Network, Durham, NC

Disclosures:

C. Hostler, None

R. W. Moehring, None

E. Dodds Ashley, None

M. Johnson, None

A. Davis, None

S. S. Lewis, None

D. J. Sexton, None

D. Anderson, None

Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.