Influence of referral bias on the clinical characteristics of patients with Gram-negative bloodstream infection

M. N. Al-Hasan, Jeanette E Eckel-Passow, L. M. Baddour

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Referral bias can influence the results of studies performed at tertiary-care centres. In this study, we evaluated demographic and microbiological factors that influenced referral of patients with Gram-negative bloodstream infection (BSI). We identified 2919 and 846 unique patients with Gram-negative BSI in a referral cohort of patients treated at Mayo Clinic Hospitals and a population-based cohort of Olmsted County, Minnesota, residents between 1 January 1998 and 31 December 2007, respectively. Multivariable logistic regression analysis was used to determine factors associated with referral. Elderly patients aged ≥80 years with Gram-negative BSI were less likely to be referred than younger patients [odds ratio (OR) 0·43, 95% confidence intervals (CI) 0·30 - 0·62] as were females (OR 0·63, 95% CI 0·53 - 0·74). After adjusting for age and gender, bloodstream isolates of Escherichia coli (OR 0·50, 95% CI 0·43 - 0·58) and Proteus mirabilis (OR 0·49, 95% CI 0·30 - 0·82) were underrepresented in the referral cohort; and Pseudomonas aeruginosa (OR 2·26, 95% CI 1·70 - 3·06), Enterobacter cloacae (OR 2·31, 95% CI 1·53 - 3·66), Serratia marcescens (OR 2·34, 95% CI 1·33 - 4·52) and Stenotrophomonas maltophilia (OR 17·94, 95% CI 3·98 - 314·43) were overrepresented in the referral cohort. We demonstrated that demographic and microbiological characteristics of patients with Gram-negative BSI had an influence on referral patterns. These factors should be considered when interpreting results of investigations performed at tertiary-care centres.

Original languageEnglish (US)
Pages (from-to)1750-1756
Number of pages7
JournalEpidemiology and Infection
Volume139
Issue number11
DOIs
StatePublished - Nov 2011

Fingerprint

Referral and Consultation
Odds Ratio
Confidence Intervals
Infection
Tertiary Care Centers
Demography
Stenotrophomonas maltophilia
Enterobacter cloacae
Proteus mirabilis
Serratia marcescens
Pseudomonas aeruginosa
Logistic Models
Regression Analysis
Escherichia coli
Population

Keywords

  • Bacteraemia
  • epidemiology
  • Gram-negative
  • population-based
  • Rochester Epidemiology Project
  • selection bias

ASJC Scopus subject areas

  • Infectious Diseases
  • Epidemiology

Cite this

Influence of referral bias on the clinical characteristics of patients with Gram-negative bloodstream infection. / Al-Hasan, M. N.; Eckel-Passow, Jeanette E; Baddour, L. M.

In: Epidemiology and Infection, Vol. 139, No. 11, 11.2011, p. 1750-1756.

Research output: Contribution to journalArticle

@article{a9dd453b757c42d69e0adca549d6fec2,
title = "Influence of referral bias on the clinical characteristics of patients with Gram-negative bloodstream infection",
abstract = "Referral bias can influence the results of studies performed at tertiary-care centres. In this study, we evaluated demographic and microbiological factors that influenced referral of patients with Gram-negative bloodstream infection (BSI). We identified 2919 and 846 unique patients with Gram-negative BSI in a referral cohort of patients treated at Mayo Clinic Hospitals and a population-based cohort of Olmsted County, Minnesota, residents between 1 January 1998 and 31 December 2007, respectively. Multivariable logistic regression analysis was used to determine factors associated with referral. Elderly patients aged ≥80 years with Gram-negative BSI were less likely to be referred than younger patients [odds ratio (OR) 0·43, 95{\%} confidence intervals (CI) 0·30 - 0·62] as were females (OR 0·63, 95{\%} CI 0·53 - 0·74). After adjusting for age and gender, bloodstream isolates of Escherichia coli (OR 0·50, 95{\%} CI 0·43 - 0·58) and Proteus mirabilis (OR 0·49, 95{\%} CI 0·30 - 0·82) were underrepresented in the referral cohort; and Pseudomonas aeruginosa (OR 2·26, 95{\%} CI 1·70 - 3·06), Enterobacter cloacae (OR 2·31, 95{\%} CI 1·53 - 3·66), Serratia marcescens (OR 2·34, 95{\%} CI 1·33 - 4·52) and Stenotrophomonas maltophilia (OR 17·94, 95{\%} CI 3·98 - 314·43) were overrepresented in the referral cohort. We demonstrated that demographic and microbiological characteristics of patients with Gram-negative BSI had an influence on referral patterns. These factors should be considered when interpreting results of investigations performed at tertiary-care centres.",
keywords = "Bacteraemia, epidemiology, Gram-negative, population-based, Rochester Epidemiology Project, selection bias",
author = "Al-Hasan, {M. N.} and Eckel-Passow, {Jeanette E} and Baddour, {L. M.}",
year = "2011",
month = "11",
doi = "10.1017/S095026881100001X",
language = "English (US)",
volume = "139",
pages = "1750--1756",
journal = "Epidemiology and Infection",
issn = "0950-2688",
publisher = "Cambridge University Press",
number = "11",

}

TY - JOUR

T1 - Influence of referral bias on the clinical characteristics of patients with Gram-negative bloodstream infection

AU - Al-Hasan, M. N.

AU - Eckel-Passow, Jeanette E

AU - Baddour, L. M.

PY - 2011/11

Y1 - 2011/11

N2 - Referral bias can influence the results of studies performed at tertiary-care centres. In this study, we evaluated demographic and microbiological factors that influenced referral of patients with Gram-negative bloodstream infection (BSI). We identified 2919 and 846 unique patients with Gram-negative BSI in a referral cohort of patients treated at Mayo Clinic Hospitals and a population-based cohort of Olmsted County, Minnesota, residents between 1 January 1998 and 31 December 2007, respectively. Multivariable logistic regression analysis was used to determine factors associated with referral. Elderly patients aged ≥80 years with Gram-negative BSI were less likely to be referred than younger patients [odds ratio (OR) 0·43, 95% confidence intervals (CI) 0·30 - 0·62] as were females (OR 0·63, 95% CI 0·53 - 0·74). After adjusting for age and gender, bloodstream isolates of Escherichia coli (OR 0·50, 95% CI 0·43 - 0·58) and Proteus mirabilis (OR 0·49, 95% CI 0·30 - 0·82) were underrepresented in the referral cohort; and Pseudomonas aeruginosa (OR 2·26, 95% CI 1·70 - 3·06), Enterobacter cloacae (OR 2·31, 95% CI 1·53 - 3·66), Serratia marcescens (OR 2·34, 95% CI 1·33 - 4·52) and Stenotrophomonas maltophilia (OR 17·94, 95% CI 3·98 - 314·43) were overrepresented in the referral cohort. We demonstrated that demographic and microbiological characteristics of patients with Gram-negative BSI had an influence on referral patterns. These factors should be considered when interpreting results of investigations performed at tertiary-care centres.

AB - Referral bias can influence the results of studies performed at tertiary-care centres. In this study, we evaluated demographic and microbiological factors that influenced referral of patients with Gram-negative bloodstream infection (BSI). We identified 2919 and 846 unique patients with Gram-negative BSI in a referral cohort of patients treated at Mayo Clinic Hospitals and a population-based cohort of Olmsted County, Minnesota, residents between 1 January 1998 and 31 December 2007, respectively. Multivariable logistic regression analysis was used to determine factors associated with referral. Elderly patients aged ≥80 years with Gram-negative BSI were less likely to be referred than younger patients [odds ratio (OR) 0·43, 95% confidence intervals (CI) 0·30 - 0·62] as were females (OR 0·63, 95% CI 0·53 - 0·74). After adjusting for age and gender, bloodstream isolates of Escherichia coli (OR 0·50, 95% CI 0·43 - 0·58) and Proteus mirabilis (OR 0·49, 95% CI 0·30 - 0·82) were underrepresented in the referral cohort; and Pseudomonas aeruginosa (OR 2·26, 95% CI 1·70 - 3·06), Enterobacter cloacae (OR 2·31, 95% CI 1·53 - 3·66), Serratia marcescens (OR 2·34, 95% CI 1·33 - 4·52) and Stenotrophomonas maltophilia (OR 17·94, 95% CI 3·98 - 314·43) were overrepresented in the referral cohort. We demonstrated that demographic and microbiological characteristics of patients with Gram-negative BSI had an influence on referral patterns. These factors should be considered when interpreting results of investigations performed at tertiary-care centres.

KW - Bacteraemia

KW - epidemiology

KW - Gram-negative

KW - population-based

KW - Rochester Epidemiology Project

KW - selection bias

UR - http://www.scopus.com/inward/record.url?scp=80053567510&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053567510&partnerID=8YFLogxK

U2 - 10.1017/S095026881100001X

DO - 10.1017/S095026881100001X

M3 - Article

C2 - 21281552

AN - SCOPUS:80053567510

VL - 139

SP - 1750

EP - 1756

JO - Epidemiology and Infection

JF - Epidemiology and Infection

SN - 0950-2688

IS - 11

ER -