a b s t r a c t

Objectives: 1) To measure frequency and yield of blood cultures obtained for Observation status adult patients with skin and soft tissue infection (SSTI), 2) describe how often blood cultures were performed according to In- fectious Diseases Society of America (IDSA) SSTI guideline indications, 3) identify proportion of patients meeting Center for Medicare Services (CMS) Sepsis criteria.

Design: Retrospective cohort.

Setting: Tertiary academic center.

Patients: Consecutive adult observation status patients hospitalized with SSTI between July 2017 and July 2018. Methods: We measured the proportion and results of blood cultures obtained among the study cohort and pro- portion of obtained cultures that satisfied IDSA indications.

Results: We identified 132 observation status patients with SSTI during the study period; 67 (50.8%) had blood cultures drawn. Only 14 (10.6%) patients met IDSA indications for culture; 51 (38.%) met Center for Medicare Ser- vices definition for sepsis. We identified two (3.0%) cases of bacteremia and two (3.0%) cases of skin bacteria con- tamination. In multivariable analysis, only temperature > 38 ?C (OR 3.84, 95%CI 1.09-13.60) and white race (OR 2.71, 95%CI 1.21-6.20) were associated with blood culture obtainment; neither meeting IDSA SSTI guideline in- dications nor meeting CMS sepsis criteria was associated with culture.

Conclusions: Among observation status patients with SSTI, over half had blood cultures drawn, though 10% satis- fied guideline indications for culture. The proportion of cultures with bacterial growth was low and yielded as many skin contaminants as cases of bacteremia. Our study highlights the need for further quality improvement efforts to reduce unnecessary blood cultures in routine SSTI cases.

(C) 2021 Published by Elsevier Inc.

1. Introduction

skin and soft tissue infections (SSTIs) are common, with 14.5 million cases leading to 650,000 hospitalizations annually in the United States [1]. The 2014 Infectious Diseases Society of America (IDSA) practice guide- lines for diagnosis and management of SSTIs recommend against rou- tine blood cultures for patients with SSTI, except in cases with immersion injury, animal bite, malignancy on chemotherapy, severe cellular immune deficiency or neutropenia [2]. Current Center for Med- icaid Services (CMS) guidelines define sepsis as evidence of infection plus two or more systemic inflammatory response (SIRS) criteria, with severe sepsis definition adding evidence of organ dysfunction [3].

* Corresponding author at: University of North Carolina, Department of Medicine, Division of Hospital Medicine, 101 Manning Drive, CB#7085, Chapel Hill, NC 27599- 77086, United States of America.

E-mail address: [email protected] (J.R. Stephens).

Practitioners evaluating patients with SSTI thus may find conflicting rec- ommendations in trying to satisfy CMS sepsis core measures while ad- hering to IDSA guidelines.

The rate of positive blood cultures in patients with uncomplicated SSTI ranged from 1 to 7% in prior studies [4-6]. Despite the low yield and the IDSA recommendations, the proportion of SSTI cases with blood cultures obtained was reported as 32% in a mixed inpatient/ob- servation population [4] and 35-73% in inpatient populations [5,6]. Pa- tients meeting observation status are expected to stay 24-48 h according to Medicare rules [7], are typically less acutely ill than those meeting inpatient criteria, and thus should have low clinical suspicion for bacteremia. We are unaware of prior studies of blood culture in SSTI in an observation status population. Given the overall lower acuity of observation status patients, we hypothesized that the frequency of obtaining blood cultures for SSTI cases in this population would be lower than that reported in prior studies involving inpatient cases. Knowing this information may be useful to providers and health care systems as they focus quality improvement efforts.

https://doi.org/10.1016/j.ajem.2021.02.026 0735-6757/(C) 2021 Published by Elsevier Inc.

Our study has several objectives for adult observation status patients with SSTI: 1) to measure the frequency and yield of blood cultures ob- tained, 2) to describe how often blood cultures were performed accord- ing to IDSA SSTI guideline indications, 3) to describe how often patients met CMS sepsis and severe sepsis criteria, and 4) to identify predictors of blood culture obtainment in the study population.

1. Methods
   1. Study setting

Our institution is a tertiary academic center with over 800 licensed beds and includes residency programs for internal medicine, Family Medicine and emergency medicine. The hospital’s Observation Unit is physically separated from the emergency department (ED) and is staffed by hospitalist internal medicine, family medicine and advanced practice providers along with trainees in internal medicine.

• 1. Study population

We performed a retrospective study of consecutive patients 18 years of age or older admitted to our institution’s Observation Unit from the emergency department (ED) with a diagnosis of SSTI between July 1, 2017 and July 1, 2018. Patients were prospectively identified by the unit medical director (CC) around the time of admission. As a separate means of identification, we also retrospectively identified all patients initially hospitalized in the unit with International Classification of Dis- eases, 10th Revision (ICD-10) codes for cellulitis or abscess as a primary, secondary or tertiary diagnosis. We manually reviewed all records using the institution’s electronic health record . Patients were included if cellulitis, soft tissue infection, abscess or erysipelas was documented as the one of the main clinical issues addressed at the time of arrival in the ED or within initial documentation by the admitting physician. For each patient, we collected demographic information (age, sex, race), presenting symptoms, location of SSTI, initial vital signs, diagnos- tic testing (including lactate measurements, white blood cell count, and blood cultures), number of Systemic Inflammatory Response Syndrome criteria and initial antibiotics (Table 1). We also identified pa- tients with diabetes mellitus and persons who inject drugs (PWID) via manual chart review. Study methods were approved by our institu- tional review board (IRB# 20-1243).

• 1. Outcomes

The primary outcome was the frequency with which blood cultures were obtained in the ED for adult patients admitted under observation status for SSTI. Secondary outcomes included: 1) the proportion of blood cultures positive for bacterial growth in this cohort and 2) the pro- portion of patients with SSTI who had blood cultures obtained according to IDSA guideline indications and 3) the proportion of patients meeting CMS sepsis and severe sepsis criteria. Patients with SSTI for whom IDSA guidelines recommends blood cultures include those with animal bites, malignancy on chemotherapy, immersion injuries, severe cellular im- mune deficiency and neutropenia. For two indications in which the IDSA guidelines were insufficiently specific, defining severe cellular im- mune deficiency and neutropenia, we referenced guidelines on use of live-attenuated zoster vaccine in immunocompromised hosts [8] and managing neutropenic fever [9], respectively (Appendix 1).

• 1. Data analysis

We calculated descriptive statistics for the study population. We used univariable and multivariable logistic regression to evaluate for associa- tions between demographic and clinical variables and the primary out- come of blood cultures obtainment for SSTI management. The independent variable of interest (whether the patient satisfied IDSA

Table 1

Characteristics of observation status adults with skin and soft tissue infection (N = 132)

Characteristics Value

Median age, years (range) 50.5 (20-96)

Male, N (%) 77 (58.3)

Race, N (%)

White 84 (63.6)

Black 32 (24.2)

Other 12 (14.3)

Asian 4 (3.0)

Hispanic ethnicity, N (%) 12 (9.1)

Hispanic non-white 5 (41.7)

no insurance, N (%) 36 (27.3)

Diabetes mellitus, N (%) 30 (22.7)

Persons who inject drugs, N (%) 3 (2.3)

WBC >= 12 K, N (%) 28 (21.2)

Fever (temperature >= 38.0 ?C), N (%) 19 (14.4)

SBP <= 90 mmHg, N (%) 0 (0%)

SIRS criteria, N (%)

Hyperthermia (>=38.0 ?C) or hypothermia (<= 36.0 ?C) 72 (55.4) WBC (leukocytosis >=12 K or leukopenia <=4 K) 31 (23.5)

Tachycardia (HR >= 90) 54 (40.9)

Tachypnea (RR >= 20) 8 (6.1)

CMS Sepsis criteria, N (%)^? 51 (38.6)

CMS Severe sepsis criteria, N (%)^?? 9 (6.8)

SSTI location, N (%)

Trunk and Limbs 98 (74.2)

Head/Neck 22 (16.7)

Groin/Perineum 6 (4.5)

Abscess associated with cellulitis (any location) 6 (4.5)

Lactate ordered, N (%) 52 (39.4)

Lactate level >= 2.0 mmol/L, N (%) 11 (8.3)

Satisfied IDSA SSTI Criteria for Blood Culture, N (%) 14 (10.6)

Sets of Blood Cultures Obtained in ED, N (%) 67 (50.8)

1 of 2 positive 4 (6.0)

Streptococcus agalactiae^? 1 (1.5)

Methicillin-susceptible Staphylococcus aureus?	1 (1.5)
Staphylococcus capitis??	1 (1.5)
Staphylococcus hominis??	1 (1.5)
2 of 2 positive	0 (0.0)
Patients administered antibiotics in ED, N (%)	124 (93.9)
Intravenous antibiotics ordered, N (%) Antibiotic agents, N (%)+	118 (95.2)
Vancomycin	48 (36.4)
Clindamycin	42 (31.8)
Multiple antibiotics	34 (25.8)
Piperacillin/tazobactam	20 (15.2)
Ampicillin/sulbactam	13 (9.8)
Cefepime	5 (3.9)
Linezolid	5 (3.9)
Daptomycin	5 (3.9)
Trimethoprim-sulfamethoxazole	4 (3.0)
Cefazolin	3 (2.3)
Doxycycline	1 (0.8)
Other agents++	4 (3.0)

Abbreviations: N - number; WBC - white blood cell count; BP - systolic blood pressure; SIRS - systemic inflammatory response syndrome; CMS - Center for Medicare Services; SSTI - skin and soft tissue infections; L - liter; mmol - millimoles; IDSA - Infectious Dis- eases Society of America; ED - Emergency Department.

⁺ antibiotics not mutually exclusive, so patients were often administered more than one antibiotic

⁺⁺ ertapenem (1), metronidazole (1), amoxicillin/clavulanate (1), cephalexin (1)

^? true positive blood culture; ^??presumed skin contaminant; ^? >= 2 SIRS criteria;

^?? >=2 SIRS criteria plus evidence of end organ damage (i.e., Lactate >=2.0);

SSTI guidelines indications for blood cultures) and all covariates with p-values <0.2 in the univariable analysis were included in the multivari- able analysis. All analyses were performed using Stata/SE 16.1 (StataCorp College Station, TX).

1. Results

We identified 132 patients with a primary diagnosis of SSTI during the study period (Table 1). About half, 67/132 (50.8%), had blood

Table 2

Number of SIRS criteria and proportions with ordered intravenous antibiotics and blood cultures (N = 132)

Number of SIRS Criteria	N (%)	% Receiving IV antibiotics	% with Blood Culture Ordered
0	27 (20.5)	81.5%	48.2%
1	54 (40.9)	87.0%	51.9%
2	42 (31.8)	97.6%	53.4%
3	9 (6.8)	88.9%	44.4%
4	0 (0.0)	-	-

Abbreviations: SIRS - systemic inflammatory response syndrome; IV - intravenous.

cultures drawn. Of the blood cultures obtained, no cases occurred in which both sets were positive for bacterial growth; 4/67 (6.0%) had one of two blood culture sets positive for bacterial growth. Of the four positive cultures, two (3.0%) grew species that were believed to be true pathogens. One patient with breast cellulitis had Streptococcus agalactiae bacteremia and another patient with forearm cellulitis at a peripheral intravenous (IV) line site from a hospitalization the week prior to admission had bacteremia with methicillin-sensitive Staphylo- coccus aureus. Two cultures (3.0%) grew species presumed to be skin contaminants, S. hominis and S. capitis.

A minority of cases, 14/132 (10.6%), satisfied IDSA SSTI criteria for obtaining blood cultures: six animal bites, five patients with severe cellu- lar immune deficiency (two on etanercept, one each on chronic predni- sone >20 mg daily, infliximab, and natalizumab) and three immersion injuries in non-sterile water. Of those meeting IDSA criteria, 9/14 (64.3%) had cultures obtained. All blood cultures obtained that satisfied IDSA criteria were negative for bacterial growth.

Justover onethird ofpatients (51/132, 38.6%) met CMS sepsis criteria, defined as evidence of infection with two or more SIRS criteria present. About half of these patients (26/51, 51.0%) had blood cultures obtained. Theproportion of patientswith blood culture obtainedwassimilar across subsets with increasing number of SIRS criteria (Table 2). Nine patients (6.8%) met CMS criteria for severe sepsis with two or more SIRS criteria and lactate >2.0 mmol/L (no patients had SBP <90 mmHg). Of these pa- tients, 8/9 (88.9%) had blood cultures obtained.

In multivariable analyses (Table 3), white race (OR 2.78, 95%CI 1.23-6.30) and fever (OR 3.87, 95%CI 1.10-13.67) were significantly as- sociated with increased likelihood of having blood cultures obtained in the ED. Neither satisfying IDSA SSTI guideline indications nor meeting CMS sepsis or severe sepsis criteria were associated with blood culture obtainment in the multivariable analysis.

1. Discusssion

In our study of consecutive observation status patients with SSTI, blood cultures were ordered in over half of patients, though the large majority did not satisfy IDSA guideline indications for cultures. The yield of blood cultures was low and resulted in as many skin contami- nants as true positive cultures. First line empirical agents per Guideline recommendations would have appropriately covered the two cases of bacteremia [2]. Fever and white race were associated with blood cul- tures being drawn, while meeting IDSA guideline criteria for culture was not associated. Conversely, of those patients meeting IDSA criteria for cultures, 5 of 14 (35.7%) did not have them obtained, indicating both overuse and underuse of blood cultures in this setting.

The high proportion of SSTI encounters with blood cultures obtained despite few meeting guideline indications for culture suggests inade- quate knowledge or use of current IDSA SSTI guideline recommenda- tions at our institution. This is reinforced by our finding that meeting guideline criteria was not predictive of blood culture obtainment.

Table 3

Univariable and Multivariable Associations for Blood Cultures Ordered in Skin and Soft Tissue Infection Encounters (N = 132)

Characteristics	Univariable Analysis			Multivariable Analysis
	OR (95%CI)	P		OR (95%CI)	P
Age >= 50 years	0.50 (0.25-1.01)	0.05		0.54 (0.25-1.17)	0.12
Male	1.44 (0.72-2.89)	0.30		-	-
White race	2.33 (1.13-4.83)	0.02		2.78 (1.23-6.30)	0.01
Hispanic ethnicity	2.07 (0.59-7.24)	0.23
No insurance	1.30 (0.60-2.81)	0.50		-	-
Diabetes mellitus	0.81 (0.36-1.83)	0.64		-	-
Persons who inject drugs, %?	1.97 (0.17-22.26)	0.58		-	-
Satisfied IDSA SSTI Criteria for Blood Cultures	1.86 (0.59-5.89)	0.29		2.07 (0.60-7.21)	0.25
WBC >= 12 K	1.15 (0.50-2.66)	0.74		-	-
Fever (>=38.0 ?C)	4.40 (1.37-14.08)	0.01		3.87 (1.10-13.67)	0.04
CMS Sepsis criteria?	1.01 (0.50-2.04)	0.97		-	-
CMS Severe sepsis criteria??	8.68 (1.05-71.49)	0.04		7.12 (0.80-63.35)	0.08
Number of SIRS criteria	1.01 (0.68-1.51)	0.96		-	-
Hyperthermia (>=38.0 ?C) or hypothermia (<= 36.0 ?C)	1.00 (0.50-2.00)	1.00		-	-
WBC (leukocytosis >=12 K or leukopenia <=4 K)	0.88 (0.39-1.98)	0.76		-	-
Tachycardia (HR >= 90)	1.08 (0.54-2.16)	0.83		-	-
Tachypnea (RR >= 20)	1.67 (0.38-7.28)	0.50		-	-
Abscess present	5.16 (0.59-45.44)	0.14		8.49 (0.85-84.52)	0.07
SSTI location
General?	1.22 (0.56-2.67)	0.62		-	-
Head/Neck	0.49 (0.19-1.27)	0.14		0.60 (0.22-1.65)	0.32
Groin/Perineum	0.97 (0.19-4.98)	0.97		-	-

Abbreviations: N - number; OR - odds ratio, CI - confidence interval, IDSA - Infectious Diseases Society of America; CMS - Center for Medicare Services; SIRS - systemic inflammatory re- sponse syndrome; C - Celsius, WBC - white blood cell count; L - liter; mmol - millimoles; SSTI - skin and soft tissue infections.

^{? ?}lower or upper extremity, chest, abdomen, back; ^?only three persons who inject drugs were identified; ^? >= 2 SIRS criteria;

^?? >=2 SIRS criteria plus evidence of end organ damage (i.e., Lactate >=2.0).

Additionally, other aspects of our population showed incomplete adher- ence to IDSA guidelines. For example, while the guidelines recommend IV antibiotics only for moderate or severe infections, defined as having “systemic signs of infection,” (essentially SIRS criteria), 20% of our pa- tients had no SIRS criteria, suggesting a number of unnecessary admis- sions. Almost half of these patients had blood cultures ordered and over 80% were administered IV antibiotics. We should note that current guidelines are not specific in certain areas, particularly in defining “se- vere cellular immune deficiency,” which may complicate guideline application.

Fever (temperature >= 38 ?C) was associated with culturing in our study, which could reflect provider assessment of clinical severity or re- sidual practice influenced by earlier IDSA guidelines, which recom- mended cultures in patients with “signs and symptoms of systemic toxicity.” [10] Another possible explanation, however, is sepsis, rather than SSTI, as motivation for obtaining some of the cultures. This is sup- ported by the finding that eight of the nine patients meeting CMS defi- nition for severe sepsis had cultures obtained. As we noted in the introduction, practitioners evaluating patients with SSTI may find them- selves navigating conflicting recommendations, where IDSA guidelines would suggest not getting blood culture, while CMS severe sepsis core measures support culture. However, 38% of our patients met the CMS sepsis definition and half of these patients (similar to the entire cohort) had cultures obtained. Additionally, neither increasing number of SIRS criteria nor meeting CMS severe sepsis definition was associated with blood culture obtainment in our study, though severe sepsis almost met statistical significance. Sepsis is thus at best an incomplete explana- tion of the high proportion of blood cultures obtained in our population. It is unclear why white race was associated with blood culture in our study. Possible explanations include provider awareness of higher rates of Injection drug use in this demographic [11,12], although we identi- fied few PWID in our cohort. Effects of implicit bias [13] or other uniden-

tified confounders are other possible explanations for this finding.

The proportion of blood cultures drawn for SSTI cases in our study (52.1%) iswithin the 32-73% range reported in prior studies [4-6]. The pro- portions of SSTI encounters with true pathogens (3.2%) and contaminants (3.2%) weidentified were also similar tothose of prior studies, 1.7-7% [4-6] and 3.6% [6] respectively. It should be noted that the populations included in prior studies were somewhat distinct from ours. Perl et al. [6] and Torres et al. [5] appear to have included inpatient cases, though both are simply described as “admitted” cases of SSTI. Ko et al. studied a mix of inpatient and observation patients [4]. Additionally, the study reporting the highest proportion of true positive blood cultures (7%) included a population with high numbers of PWID [5]. We are unaware of prior studies limited to observation status patients. The high frequency of blood culture in this low-acuity population with SSTI suggests the need for broad quality im- provement efforts, including ED practice areas.

Our study should be interpreted in light of several limitations. Its single-center design may limit generalizability to other populations. Given the retrospective study design, it is possible we missed relevant cases that could have affected our study results and our modest sample size may limit our ability to identify all differences in our multivariate analysis. As we only assessed the initial set of ED vital signs for our anal- ysis, we may have missed clinical factors (i.e., fever later in the ED course) that influenced provider decision-making. Although the CMS sepsis definition is commonly used, it is not without controversy. Multi- ple definitions for sepsis exist [3], and utilizing an alternative definition could potentially lead to different results. Lastly, the factors we identi- fied as associated with having blood cultures drawn are potentially sub- ject to confounding with our study design.

In conclusion, although blood cultures were frequently obtained among observation patients with SSTI, they were not collected in accor- dance with IDSA SSTI guidelines and were of very low yield, with the rate of contaminants equal to that of true pathogens. The two cases with true blood culture pathogens in this study would have been

covered by Empirical antibiotics recommended as first line agents for SSTI per current guidelines, though the detection of bacteremia is likely to change the duration of IV therapy. These finding suggest that follow- ing specified criteria, such as those in the IDSA guidelines, for obtaining blood cultures in SSTI is valuable to decrease unnecessary testing. Given patient charges of $226 per set of blood cultures at our institution, the cultures drawn against guideline recommendations resulted in a mini- mum of $26,216 (based on 58 cultures obtained outside of guidelines) in unnecessary costs, not factoring in further potential downstream costs, including adverse consequences from contaminant cultures [14]. Our study highlights the need for continued broad quality improvement efforts in patients with SSTI to reduce unnecessary testing and provide the highest value care.

Author credit statement

Emily K. Sturkie, MD helped draft the manuscript, revised subse- quent manuscript versions and approve final draft.

Carlton R. Moore, MD helped conceptualize study, analyzed data,

assisted with draft revisions and approved final draft.

Christopher A. Caulfield, MD helped conceptualize study, obtained initial data, revised all drafts and approved final draft.

Erin Schmid, ANP obtained study data, reviewed and approved drafts and final manuscript.

Anne M. Lachiewicz, MD provided subject matter expertise, edits of

manuscript drafts and approved final draft.

John R. Stephens, MD helped conceptualize study, acquire data, draft initial and final versions of manuscript.

Funding

There were no sources of funding for this work.

Presentations: Preliminary data from this manuscript were pre- sented in abstract form at the Society of Hospital Medicine Annual Meeting in April 2020.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influ- ence the work reported in this paper. AML is a consultant for Shionogi and MicroGenDx and receives research support from Cidara Therapeu- tics and Novartis.

Appendix 1. Definitions Used for IDSA Guideline Indications for Blood Cultures for Skin and Soft Tissue Infection

Severe cellular immune deficiency² Lymphoma²

Lymphocytic leukemia² Solid organ transplant²

Anti-tumor necrosis factor agents or “certain monoclonal antibodies”²

>20 milligrams per day of chronic prednisone or prednisone equivalent⁷

>0.4 milligram per kilogram per week of methotrexate[8]

>3 milligram per kilogram per day of azathioprine[8]

>1.5 milligram per kilogram per day of 6-mercaptopurine[8] Neutropenia²

Absolute Neutrophil count <1000 cells/microliter⁸ Immersion injury²

Open wound submerged in non-sterile water Malignancy on chemotherapy²

Animal bite²

²Stevens et al; ⁷Harpaz et al; ⁸Taplitz et al.

References

1. Raff AB, Kroshinsky D. Cellulitis: a review. JAMA. 2016;316(3):325-37. https://doi. org/10.1001/jama.2016.8825.
2. Stevens DL, Bisno AL, Chambers HF, Dellinger EP, Goldstein EJ, Gorbach SL, et al. Practice guidelines for the diagnosis and management of skin and soft tissue infec- tions: 2014 update by the infectious diseases society of America. Clin Infect Dis. 2014;59(2):147-59. https://doi.org/10.1093/cid/ciu296.
3. Kalantari A, Mallemat H, Weingart SD. Sepsis definitions: the search for gold and what CMS got wrong. West J Emerg Med. 2017;18(5):951-6. https://doi.org/10. 5811/westjem.2017.4.32795.
4. Ko LN, Garza-Mayers AC, St John J, Strazzula L, Vedak P, Dobry AS, et al. Clinical use- fulness of imaging and blood cultures in cellulitis evaluation. JAMA Intern Med. 2018;178(7):994-6. https://doi.org/10.1001/jamainternmed.2018.0625.
5. Torres J, Avalos N, Echols L, Mongelluzzo J, Rodriguez RM. Low yield of blood and wound cultures in patients with skin and soft-tissue infections. Am J Emerg Med. 2017;35(8):1159-61. https://doi.org/10.1016/j.ajem.2017.05.039.
6. Perl B, Gottehrer NP, Raveh D, Schlesinger Y, Rudensky B, Yinnon AM. Cost-effective- ness of blood cultures for adult patients with cellulitis. Clin Infect Dis. 1999;29(6): 1483-8. https://doi.org/10.1086/313525.
7. Medicare Benefit Policy Manual. https://www.cms.gov/Regulations-and-Guidance/ Guidance/Manuals/downloads/bp102c06.pdf.
8. Harpaz R, Ortega-Sanchez IR, Seward JF, Advisory Committee on Immunization Practices Centers for Disease C. Prevention. Prevention of herpes zoster: recommen- dations of the Advisory Committee on Immunization Practices (ACIP). MMWR

Recomm Rep. 2008;57(RR-5):1-30 quiz CE2-4 https://www.ncbi.nlm.nih.gov/ pubmed/18528318.

1. Taplitz RA, Kennedy EB, Bow EJ, Crews J, Gleason C, Hawley DK, et al. outpatient management of fever and neutropenia in adults treated for malignancy: American Society of Clinical Oncology and Infectious Diseases Society of America Clinical Prac- tice Guideline Update. J Clin Oncol. 2018;36(14):1443-53. https://doi.org/10.1200/ JCO.2017.77.6211.
2. Stevens DL, Bisno AL, Chambers HF, Everett ED, Dellinger P, Goldstein EJ, et al. Prac- tice guidelines for the diagnosis and management of skin and soft-tissue infections. Clin Infect Dis. 2005;41(10):1373-406. https://doi.org/10.1086/497143.
3. Cicero TJ, Kuehn BM. Driven by prescription drug abuse, Heroin use increases among suburban and rural whites. JAMA. 2014;312(2):118-9. https://doi.org/10.1001/ jama.2014.7404.
4. Cicero TJ, Ellis MS, Surratt HL, Kurtz SP. The changing face of heroin use in the United States: a retrospective analysis of the past 50 years. JAMA Psychiat. 2014;71(7): 821-6. https://doi.org/10.1001/jamapsychiatry.2014.366.
5. Dehon E, Weiss N, Jones J, Faulconer W, Hinton E, Sterling S. A systematic review of the impact of physician implicit Racial bias on clinical decision making. Acad Emerg Med. 2017;24(8):895-904. https://doi.org/10.1111/acem.13214.
6. Murofushi Y, Furuichi M, Shoji K, Kubota M, Ishiguro A, Uematsu S, et al. Adverse economic impact associated with blood culture contamination in a pediatric emer- gency department. Pediatr Infect Dis J. 2018;37(8):755-8. https://doi.org/10.1097/ INF.0000000000001898.