Skip to main content
. 2013 Oct 3;8(10):e76225. doi: 10.1371/journal.pone.0076225

Table 1. Characteristics of Patients by Healthcare Exposure Status.

Total Cohort (N =578)a Community-acquired (n=217)a Healthcare-associated (n=320)a Hospital-acquired (n=41)a P valuee
Patient Characteristics Age, years, median (range) 72 (18-97) 69 (18-97) 73.5 (19-97) 73 (36-94) 0.01
Male Sex 257 (44) 85 (39) 146 (46) 26 (63) 0.01
Non-white Race 266 (47) 87 (40) 164 (52) 15 (34) 0.02
Medicare/Medicaid 463 (81) 158 (74) 267 (85) 38 (83) <.001
No insurance 31 (5) 18 (8) 12 (4) 1 (2) 0.06
BMI, median (range) 25 (13-55) 27 (13-55) 24 (13-55) 24 (18-54) 0.004
Charlson score,bmedian (range) 3 (0-11) 1 (0-11) 3 (0-10) 3 (0-9) <.001
APACHE II,cmedian (range) 14 (4-30) 13 (4-28) 15 (4-30) 14 (7-29) 0.006
MCCabe score b <.001
Rapidly fatal <2 weeks 105 (19) 35 (17) 56 (18) 14 (35)
Ultimately fatal <5 years 302 (52) 82 (39) 199 (62) 21 (51)
Nonfatal 159 (28) 95 (45) 59 (18) 5 (12)
Dependent in ≥1 ADL b 323 (56) 91 (42) 208 (65) 24 (59) <.001
Dependent in ≥3 ADL b 136 (24) 24 (11) 107 (34) 5 (11) <.001
Admission source b Home 440 (75) 211 (100) 201 (64) 28 (61) d
Skilled nursing facility 121 (21) 0 (0) 114 (36) 7 (15)
Other hospital 6 (1) 0 (0) 0 (0) 6 (13)
Prior hospitalization in last 1 year b 251 (43) 0 (0) 232 (73) 19 (41) d
Comorbidities b Diabetes 220 (38) 70 (32) 147 (46) 8 (17) <.001
MI 118 (20) 40 (18) 63 (20) 15 (33) 0.03
CHF 109 (19) 29 (13) 68 (21) 12 (26) 0.02
PVD 77 (13) 26 (12) 44 (14) 7 (15) 0.6
Stroke 112 (19) 21 (10) 86 (27) 5 (5) <.001
Dementia 102 (18) 21 (10) 78 (24) 3 (7) <.001
COPD 100 (17) 31 (14) 58 (18) 11 (24) 0.1
Liver disease 36 (6) 15 (7) 17 (5) 4 (9) 0.6
Dialysis 35 (6) 0 (0) 32 (10) 3 (7) d
Active Malignancy 127 (22) 36 (17) 80 (25) 11 (24) 0.05
Metastatic Malignancy 26 (4) 6 (3) 16 (5) 4 (9) 0.09
Decubitus ulcer 79 (14) 11 (5) 62 (19) 6 (13) <.001
Immunosuppression 56 (10) 21 (10) 31 (10) 4 (9) 0.99
Devices b Central IV catheter 61 (10) 0 (0) 47 (15) 14 (30) d
Foley catheter 64 (11) 0 (0) 61 (19) 3 (7) d
PEG 28 (5) 0 (0) 23 (7) 5 (11) d
Infection Characteristics Polymicrobial 48 (8) 10 (5) 30 (9) 8 (17) 0.004
Multiple GN pathogens 18 (3) 4 (2) 10 (3) 4 (9) 0.1
Multidrug Resistance 84 (15) 20 (9) 58 (18) 6 (13) 0.02
Primary Source 0.008
Urine 96 (17) 42 (19) 48 (15) 6 (15)
Wound 4 (1) 2 (1) 2 (1) 0 (0)
Lower respiratory tract 8 (1) 2 (1) 3 (1) 3 (7)
Other  9 (2) 3 (1) 2 (1) 4 (10)
No culture or unknown 406 (70) 168 (77) 223 (70) 15 (37)
IV catheter related 55 (10) 0 (0) 42 (13) 13 (32) <.001
Outcomes
  • Time to appropriate antibiotic therapy, hours, median


  • (95% CI)

26.9 25.9 27.9 27.3 0.02
(26.2, 28.3) (25.4, 27.7) (26.6, 28.9) (25.3, 48.0)
  • Time to any antibiotic therapy, hours, median


  • (95% CI)

26.4 25.8 27.2 26.4 0.3
(25.8, 27.5) (25.2, 26.8) (26.1 28.2) (25.0, 36.8)
Disposition Death or hospice 82 (14) 17 (8) 53 (17) 12 (29) <.001
Discharge to skilled nursing facility 193 (34) 18 (8) 102 (32) 9 (22) <.001
Discharge home with home health services 42 (7) 10 (5) 28 (9) 4 (10) 0.2
ICU admission 151 (26) 65 (30) 70 (22) 16 (40) 0.02
Intubation 55 (10) 19 (9) 27 (9) 9 (22) 0.02
Vasopressors 66 (12) 26 (12) 32 (10) 8 (21) 0.3
Readmission in 90 days 143 (25) 42 (20) 93 (29) 8 (20) <.001
Length of stay post BSI, days, median (range) 6 (0-158) 5 (0-158) 6 (0-143) 7 (0-52) 0.3

Note: Data are number (%) of bloodstream infection events unless otherwise indicated.

BMI = body mass index; APACHE II = Acute Physiology and Chronic Health Evaluation II score; ADL = activity of daily living; MI = myocardial infarction; CHF = congestive heart failure; PVD = peripheral vascular disease; COPD = chronic obstructive pulmonary disease; IV = intravenous; PEG = percutaneous endoscopic gastrostomy; GN = gram negative; CI = confidence interval; ICU = intensive care unit; BSI = bloodstream infection.

a

Sums may not equal total N due to small numbers of missing data (not shown) which were less than 10% for any single variable and distributed evenly among groups of healthcare exposure.

b

Measured on admission.

c

Measured on day of bloodstream infection.

d

Significance testing not performed on variables directly used in definitions of healthcare exposure groups.

e

P values indicate chi-squared or Fisher exact tests for categorical variables, or ANOVA or Kruskall-Wallis tests for continuous variables, as appropriate. All tests were two-sided, using two degrees of freedom. Time to appropriate therapy and time to first therapy were assessed using log-rank tests.