To identify the determinants of influenza vaccination and the moderators of the intention-behavior relationship among healthcare workers (HCWs).

Prospective survey with 2-month follow-up.

Three university-affiliated public hospitals.

Random sample of 424 HCWs.

The intention of an HCW to get vaccinated against influenza was measured by means of a self-administered questionnaire based on an extended version of the theory of planned behavior. An objective measure of behavior was extracted 2 months later from the vaccination database of the hospitals.

Controlling for past behavior, we found that the determinants of influenza vaccination were intention (odds ratio [OR], 8.32 [95% confidence interval {CI}, 2.82–24.50]), moral norm (OR, 3.01 [95% CI, 1.17–7.76]), anticipated regret (OR, 2.33 [95% CI, 1.23–4.41]), and work status (ie, full time vs part time; OR, 1.99 [95% CI, 1.92–3.29]). Moral norm also interacted with intention as a significant moderator of the intention-behavior relationship (OR, 0.09 [95% CI, 0.03–0.30]). Again, apart from the influence of past behavior, intention to get vaccinated was predicted by use of the following variables: attitude (ß = .32; P < .001), professional norm (β = .18; P < .001), moral norm (β = .18; P < .001), subjective norm (ß = .09; P < .001), and self-efficacy (ß = .08; P < .001). This latter model explained 89% of the variance in HCWs' intentions to get vaccinated against influenza during the next vaccination campaign.

Our study suggests that influenza vaccination among HCWs is mainly a motivational issue. In this regard, it can be suggested to reinforce the idea that getting vaccinated can reduce worry and protect family members.

Mandatory active surveillance culturing of all patients admitted to Veterans Affairs (VA) hospitals carries substantial economic costs. Clinical prediction rules have been used elsewhere to identify patients at high risk of colonization with methicillin-resistant Staphylococcus aureus (MRSA) or vancomycin-resistant enterococci (VRE). We aimed to derive and evaluate the clinical efficacy of prediction rules for MRSA and VRE colonization in a VA hospital.

Prospective cohort of adult inpatients admitted to the medical and surgical wards of a 119-bed tertiary care VA hospital.

Within 48 hours after admission, patients gave consent, completed a 44-item risk factor questionnaire, and provided nasal culture samples for MRSA testing. A subset provided perirectal culture samples for VRE testing.

Of 598 patients enrolled from August 30, 2007, through October 30, 2009, 585 provided nares samples and 239 provided perirectal samples. The prevalence of MRSA was 10.4% (61 of 585) (15.0% in patients with and 5.6% in patients without electronic medical record (EMR)-documented antibiotic use during the past year; P < .01). The prevalence of VRE was 6.3% (15 of 239) (11.3% in patients with and 0.9% in patients without EMR-documented antibiotic use; P < .01 ). The use of EMR-documented antibiotic use during the past year as the predictive rule for screening identified 242.8 (84%) of 290.6 subsequent days of exposure to MRSA and 60.0 (98%) of 61.0 subsequent days of exposure to VRE, respectively. EMR documentation of antibiotic use during the past year identified 301 (51%) of 585 patients as high-risk patients for whom additional testing with active surveillance culturing would be appropriate.

EMR documentation of antibiotic use during the year prior to admission identifies most MRSA and nearly all VRE transmission risk with surveillance culture sampling of only 51% of patients. This approach has substantial cost savings compared with the practice of universal active surveillance.

USA300 is the most prevalent strain type of community-acquired methicillin-resistant Staphylococcus aureus (MRSA), but the proportion of patients colonized with USA300 strains and their risk for infection are largely unknown.

To determine the prevalence of USA300 strains among MRSA nasal colonizing isolates in our hospital, to identify risks for USA300 nasal colonization, and to determine risks of developing infection for nasally colonized patients.

Retrospective cohort study of patients found to be nasally colonized with MRSA through active surveillance at a tertiary care hospital from January 2005 through December 2007. MRSA isolates were identified as USA300 or non-USA300. Risks for colonization and infection were identified with logistic regression.

Among 1,306 MRSA nasal isolates, 307 (24%) were USA300 strains, and this proportion significantly increased over time (17.4% in 2005 and 26.7% in 2007; P = .003). African American race was an independent risk for USA300 nasal colonization (odds ratio [OR], 1.81 [95% confidence ratio {CI}, 1.38-2.38]). Older age (OR, 0.97 [95% CI, 0.96-0.98]) and female sex (OR, 0.74 [95% CI, 0.56-0.97]) decreased risk. Among nasally colonized patients, 238 (18.2%) developed infection. Increased length of stay (OR, 1.03 [95% CI, 1.01-1.06]) independently increased risk for infection among patients nasally colonized with USA300 MRSA, and female sex decreased risk (OR, 0.48 [95% CI, 0.24-0.95]). Increased length of stay (OR, 1.03 [95% CI, 1.02-1.04]) and treatment in an intensive care unit (OR, 1.64 [95% CI, 1.10-2.44]) independently increased risk for infection among patients nasally colonized with non-USA300 MRSA, and female sex decreased risk (OR, 0.67 [95% CI, 0.47-0.94]).

The proportion of MRSA nasally colonized patients with USA300 strains significantly increased during the study, and risks included African American race. Strain type had no significant effect on the proportion of patients who developed infection, and risk factors for infection were similar.

To evaluate the risk factors that influence time to acquisition of a laboratory-confirmed bloodstream infection (LCBI).

Prospective cohort study with an 18-month follow-up.

A 16-bed medical and/or surgical pediatric intensive care unit that cares for patients of the Brazilian Public Health System exclusively.

We included children from 0 to 18 years old who were represented by 875 consecutive admissions to the pediatric intensive care unit from January 1, 2005, through June 30, 2006. The children from all but 5 (0.6%) of the admissions were followed up until discharge or death. The majority (506 [58.2%]) were hospitalized for surgical pathology, and 254 (29.2%) underwent heart surgery.

We used a standardized questionnaire and data collection from daily charts. Information on risk factors was collected before the onset of first LCBI. Survival analysis was performed using the Kaplan-Meier method. The effect of the variables on the risk of LCBI each day was estimated through a Cox model fitting.

Fifty-seven children (6.6%) developed an LCBI, 54 (94.7%) of whom made use of a central venous catheter. LCBI incidence was 11.27 episodes/1,000 patient-days and 17.92 episodes/1,000 patient-days when associated with a central venous catheter. Factors associated with time to the first LCBI in the Cox model were age less than 2 years (hazard ratio [HR], 1.99; 95% confidence interval [CI], 1.02–3.89), malnutrition (HR, 1.74; 95% CI, 1.01–3.00), use of a central venous catheter (HR, 4.36; 95% CI, 1.30–14.64), use of antibiotics before admission (HR, 0.58; 95% CI, 0.33–0.98), and use of transfused blood products (HR, 0.40; 95% CI, 0.22–0.74).

Factors associated with time to acquisition of LCBI were age less than 2 years, weight-for-age z score less than −2, and the use of a central venous catheter. Therefore, intensification of LCBI prevention efforts in patients with these characteristics is fundamental.

Before the emergence of the pandemic (H1N1) 2009 virus, estimates of the stockpiles of facial protective equipment (FPE) and the impact that information had on personnel during a pandemic varied.

To describe the impact of H1N1 on FPE use and hospital employee absenteeism.

One tertiary care hospital and 2 community hospitals in the Vancouver Coastal Health (VCH) region, Vancouver, Canada.

All persons with influenza-like illness admitted to the 3 VCH facilities during the period from June 28 through December 19, 2009.

Data on patients and on FPE use were recorded prospectively. Data on salaried employee absenteeism were recorded during the period from August 1 through December 19, 2009.

During the study period, 865 patients with influenza-like illness were admitted to the 3 VCH facilities. Of these patients, 149 (17.2%) had laboratory-confirmed H1N1 influenza infection. The mean duration of hospital stay for these patients was 8.9 days, and the mean duration of intensive care unit stay was 9.2 days. A total of 134,281 masks and 173,145 N95 respirators (hereafter referred to as respirators) were used during the 24-week epidemic, double the weekly use of both items, compared with the previous influenza season. A ratio of 3 masks to 4 respirators was observed. Use of disposable eyewear doubled. Absenteeism mirrored the community epidemiologic curve, with a 260% increase in sick calls at the epidemic peak, compared with the nadir.

Overall, FPE use more than doubled, compared with the previous influenza season, with respirator use exceeding literature estimates. A significant proportion of FPE resources were used while managing suspected cases. Planners should prepare for at least a doubling in mask and respirator use, and a 3.6-fold increase in staff sick calls.

Catheter-related bloodstream infections (CRBSIs) account for the majority of hemodialysis-related infections. There are no published data on the efficacy of the chlorhexidine-impregnated foam dressing at reducing the rate of CRBSI among patients undergoing hemodialysis.

A prospective, nonblinded, crossover intervention trial to determine the efficacy of a chlorhexidine-impregnated foam dressing to reduce the rate of CRBSI among patients undergoing hemodialysis.

Two outpatient dialysis centers.

A total of 121 patients who underwent dialysis through tunneled central venous catheters received the intervention during the trial.

The primary outcome of interest was the incidence of CRBSI. A nested cohort study of all patients who received the chlorhexidine-impregnated foam dressing was also conducted. Backward stepwise logistic regression analysis was used to determine independent risk factors for development of CRBSI.

Thirty-seven CRBSIs occurred in the intervention group, for an incidence of 6.3 CRBSIs per 1,000 dialysis sessions, and 30 CRBSIs occurred in the control group, an incidence of 5.2 CRBSIs per 1,000 dialysis sessions (risk ratio, 1.22 [95% confidence interval {CI}, 0.75-1.97]; P = .46). The chlorhexidine-impregnated foam dressing was well tolerated, with only 2 patients (<2%) experiencing dermatitis that led to its discontinuation. The only independent risk factor for development of CRBSI was dialysis treatment at one dialysis center (adjusted odds ratio, 4.4 [95% CI, 1.77-13.65]; P = .002). Age of at least 60 years (adjusted odds ratio, 0.28 [95% CI, 0.09-0.82]; P = .02) was associated with lower risk of CRBSI.

The use of a chlorhexidine-impregnated foam dressing did not decrease the incidence of CRBSI among patients with tunneled central venous catheters who were undergoing hemodialysis.

To compare 3 measures of hand hygiene adherence—direct observation, product usage, and electronic counting devices—in an intensive care unit.

A 12-week observational study.

A 40-bed medical-surgical intensive care unit at a private tertiary care hospital.

Over a 12-week period, we assessed hand hygiene adherence by 3 different methods: direct observation of practice, collection of data from electronic counters for dispensers of alcohol-based hand rub, and measurement of the amount of product used (alcohol-based hand rub and chlorhexidine).

There were 2,249 opportunities for hand hygiene observed, and the overall rate of hand hygiene adherence was 62.3% (representing 1,402 cleansing episodes). A total of 76,389 dispensing episodes were recorded by the electronic devices. The mean number of dispensing episodes per patient-day was 53.8. There was 64.1 mL of alcohol-based hand rub used per patient-day (representing 65.5% of total product used) and 33.8 mL of Chlorhexidine used per patient-day (representing 34.5%). There was no significant correlation between observed hand hygiene adherence and total product used per patient-day (r = 0.18; P = .59).

Direct observation cannot be considered the gold standard for assessing hand hygiene, because there was no relationship between the observed adherence and the number of dispensing episodes or the volume of product used. Other means to monitor hand hygiene adherence, such as electronic devices and measurement of product usage, should be considered.

Current guidelines for control of Clostridium difficile infection (CDI) suggest that contact precautions be discontinued after diarrhea resolves. However, limited information is available regarding the frequency of skin contamination and environmental shedding of C. difficile during and after treatment.

We conducted a 9-month prospective, observational study involving 52 patients receiving therapy for CDI. Stool samples, skin (chest and abdomen) samples, and samples from environmental sites were cultured for C. difficile before, during, and after treatment. Polymerase chain reaction ribotyping was performed to determine the relatedness of stool, skin, and environmental isolates.

Fifty-two patients with CDI were studied. C. difficile was suppressed to undetectable levels in stool samples from most patients during treatment; however, 1-4 weeks after treatment, 56% of patients who had samples tested were asymptomatic carriers of C. difficile. The frequencies of skin contamination and environmental shedding remained high at the time of resolution of diarrhea (60% and 37%, respectively), were lower at the end of treatment (32% and 14%, respectively), and again increased 1-4 weeks after treatment (58% and 50%, respectively). Skin and environmental contamination after treatment was associated with use of antibiotics for non-CDI indications. Ninety-four percent of skin isolates and 82% of environmental isolates were genetically identical to concurrent stool isolates.

Skin contamination and environmental shedding of C. difficile often persist at the time of resolution of diarrhea, and recurrent shedding is common 1-4 weeks after therapy. These results provide support for the recommendation that contact precautions be continued until hospital discharge if rates of CDI remain high despite implementation of standard infection-control measures.

Healthcare-associated infection (HAI) is costly and causes substantial morbidity. We sought to understand why some hospitals were engaged in HAI prevention activities while others were not. Because preliminary data indicated that hospital leadership played an important role, we sought better to understand which behaviors are exhibited by leaders who are successful at implementing HAI prevention practices in US hospitals.

We report phases 2 and 3 of a 3-phase study. In phase 2, 14 purposefully sampled US hospitals were selected from among the 72% of 700 invited hospitals whose lead infection preventionist had completed a quantitative survey on HAI prevention during phase 1. Qualitative data were collected during 38 semistructured phone interviews with key personnel at the 14 hospitals. During phase 3, we conducted 48 interviews during 6 in-person site visits to identify recurrent and unifying themes that characterize behaviors of successful leaders.

We found that successful leaders (1) cultivated a culture of clinical excellence and effectively communicated it to staff; (2) focused on overcoming barriers and dealt directly with resistant staff or process issues that impeded prevention of HAI; (3) inspired their employees; and (4) thought strategically while acting locally, which involved politicking before crucial committee votes, leveraging personal prestige to move initiatives forward, and forming partnerships across disciplines. Hospital epidemiologists and infection preventionists often played more important leadership roles in their hospital's patient safety activities than did senior executives.

Leadership plays an important role in infection prevention activities. The behaviors of successful leaders could be adopted by others who seek to prevent HAI.

Data on the resources and staff compensation of hospital epidemiology and infection control (HEIC) departments are limited and do not reflect current roles and responsibilities, including the public reporting of healthcare-associated infections. This study aimed to obtain information to assist HEIC professionals in negotiating resources.

A 28-question electronic survey was sent via e-mail to all Society for Healthcare Epidemiology of America (SHEA) members in October 2006 with the use of enterprise feedback management solution software. The survey responses were analyzed using Microsoft Excel.

Responses were received from 526 (42%) of 1,255 SHEA members. Of the respondents, 84% were doctors of medicine (MDs) or doctors of osteopathy (DOs), 6% were registered nurses, and 21% had a master of public health or master of science degree. Sixty-two percent were male (median age range, 50-59 years). Their practice locations varied across the United States and internationally. Two-thirds of respondents practiced in a hospital setting, and 63% were the primary or associate hospital epidemiologist. Although 91% provided HEIC services, only 65% were specifically compensated. In cases of antimicrobial management, patient safety, employee health, and emergency preparedness, 75%-80% of respondents provided expertise but were compensated in less than 25% of cases. Of the US-based MD and DO respondents, the median range of earnings was $151,000-$200,000, regardless of their region (respondents selected salary ranges instead of specifying their exact salaries). Staffing levels varied: the median number of physician full-time equivalents (FTEs) was 1.0 (range, 1-5); only about 25% of respondents had 3 or more infection control practitioner FTEs.

Most professionals working in HEIC have had additional training and provide a wide, growing range of services. In general, only traditional HEIC work is compensated and at levels much less than the time dedicated to those services. Most HEIC departments are understaffed. These data are essential to advocate for needed funding and resources as the roles of HEIC departments expand.

Standard precautions are advocated for reducing the number of injuries caused by needles and sharp medical devices (“sharps injuries”), but the effectiveness of gloves in preventing such injuries has not been established. We evaluated factors associated with gloving practices and identified associations between gloving practices and sharps-injury risk.

Usual-frequency case-crossover study.

Thirteen medical centers in the United States and Canada.

Six hundred thirty-six healthcare workers who presented to employee health clinics after sharps injury.

Structured telephone questionnaires were administered to assess usual behaviors and circumstances at the time of injury.

Of 636 injured healthcare workers, 195 were scrubbed in an operating room or procedure suite when injured, and 441 were injured elsewhere. Nonscrubbed individuals were more commonly gloved when treating patients who were perceived to have a high risk of human immunodeficiency virus, hepatitis B virus, or hepatitis C virus infection than when treating other patients (adjusted odds ratio [aOR], 2.53 [95% confidence interval {CI}, 1.30-4.91]). Nurses (aOR, 0.11 [95% CI, 0.04-0.32]) and other employees (aOR, 0.24 [95% CI, 0.07-0.77]) were less commonly gloved at injury than were physicians and physician trainees. Gloves reduced injury risk in case-crossover analyses (incidence rate ratio [IRR], 0.33 [95% CI, 0.22-0.50]). In scrubbed individuals, involvement in an orthopedic procedure was associated with double gloving at injury (aOR, 13.7 [95% CI, 4.55-41.3]); this gloving practice was associated with decreased injury risk (IRR, 0.20 [95% CI, 0.10-0.42]).

Although the use of gloves reduces the risk of sharps injuries in health care, use among healthcare workers is inconsistent and may be influenced by risk perception and healthcare culture. Glove use should be emphasized as a key element of multimodal sharps-injury reduction programs.

To gather baseline data on hand hygiene (HH) practices in an African hospital with a newly established infection prevention and control team.

Cross-sectional, observational study.

Komfo Anokye Teaching Hospital, a large teaching hospital in Ghana with approximately 1,000 beds.

All hospital staff with patient contact were eligible for assessment of HH practices.

HH observations were conducted using a standardized data collection tool and method based on the World Health Organization's “5 Moments of Hand Hygiene.” Small-group interviews were conducted to gather additional information on perceptions of HH and barriers to its use. HH resource needs were also assessed using a standardized tool.

HH was attempted in 12% of the opportunities and was performed appropriately in 4% of the opportunities. Most main wards (89%) had at least 1 functional HH station. The most commonly identified barriers to HH were limited resources and lack of knowledge on appropriate times to perform HH.

We developed and applied tools to evaluate HH resources and practices in a large African hospital. These assessments were undertaken to guide future efforts to improve HH at this hospital but can also serve as a model of the way to perform a systematic assessment at acute care hospitals in developing countries.

Carbapenem-resistant Klebsiella pneumoniae (CRKP) is resistant to almost all antimicrobial agents, and CRKP infections are associated with substantial morbidity and mortality.

To describe an outbreak of CRKP in Puerto Rico, determine risk factors for CRKP acquisition, and detail the successful measures taken to control the outbreak.

Two case-control studies.

A 328-bed tertiary care teaching hospital.

Twenty-six CRKP case patients identified during the outbreak period of February through September 2008, 26 randomly selected uninfected control patients, and 26 randomly selected control patients with carbapenem-susceptible K. pneumoniae (CSKP) hospitalized during the same period.

We performed active case finding, including retrospective review of the hospital's microbiology database and prospective perirectal surveillance culture sampling in high-risk units. Case patients were compared with each control group while controlling for time at risk. We sequenced the blaKPC gene with polymerase chain reaction for 7 outbreak isolates and subtyped these isolates with pulsed-field gel electrophoresis.

In matched, multivariable analysis, the presence of wounds (hazard ratio, 19.0 [95% confidence interval {CI}, 2.5-142.0]) was associated with CRKP compared with no K. pneumoniae. Transfer between units (adjusted odds ratio [OR], 7.5 [95% CI, 1.8-31.1]), surgery (adjusted OR, 4.0 [95% CI, 1.0-15.7]), and wounds (adjusted OR, 4.9 [95% CI, 1.1-21.8]) were independent risk factors for CRKP compared to CSKP. A novel K. pneumoniae carbapenemase variant (KPC-8) was present in 5 isolates. Implementation of active surveillance for CRKP colonization and cohorting of CRKP patients rapidly controlled the outbreak.

Enhanced surveillance for CRKP colonization and intensified infection control measures that include limiting the physical distribution of patients can reduce CRKP transmission during an outbreak.

To determine whether the use of dual antimicrobial therapy based on the results of a combination antibiotic susceptibility report (antibiogram) increases the likelihood of selecting adequate empirical coverage in critically ill patients with infection due to potentially resistant gram-negative pathogens.

Retrospective data analysis.

Urban academic medical center.

An analysis of culture results and susceptibility data from intensive care unit patients determined by the clinical microbiology laboratory was performed. The proportion of 5 common gram-negative pathogens susceptible to monotherapy with 1 of 3 antipseudomonal antibiotics (piperacillin-tazobactam, ceftazidime, or imipenem) was compared with the proportion susceptible to each of these 3 “backbone” agents plus 1 of 4 additional antimicrobial agents used in combination.

More than 5,000 clinical isolates were examined. When all isolates recovered during the entire study period were included, the addition of any of the second antibiotics studied to each of the 3 backbone agents significantly increased the likelihood of covering the causative pathogen (P<.01 for each). The benefit of combination therapy was variable when results for each of the 5 organisms were examined individually. When temporal trends in susceptibility were examined, the decrease in the proportion of organisms susceptible to monotherapy was statistically significant for both imipenem and ceftazidime (P<.01).

Reporting antibiotic susceptibility data in the form of a combination antibiogram may be useful to clinicians who are considering empirical antimicrobial therapy in the intensive care unit.

To examine the mechanical integrity of hemodialysis catheters after exposure to the combination of N-acetylcysteine, ti-gecycline, and heparin, compared with that of catheters exposed to heparin alone.

We used 3 types of hemodialysis catheters: polyurethane, silicone, and carbothane catheters. Catheter segments were incubated in vitro for various time intervals of up to 2 weeks either in a novel catheter lock solution (novel CLS), which consisted of N-acetylcysteine, tigecycline, and heparin, or in heparin alone (as a control). At the time of testing, each segment was rinsed and cut longitudinally into 2 sections. All catheter sections were scanned using an optical dissecting microscope to check for surface abnormalities and to measure wall thickness. We also carried out tensile strength testing of another set of catheters using a universal testing machine. Tested parameters included stress at yield, strain at yield, stress at break, strain at break, modulus of elasticity, and force at break.

The surfaces of catheters in both groups appeared similar by microscopy. The mean thickness of the catheter wall was not significantly different for the catheters exposed to the novel CLS and the catheters exposed to heparin (P >.05). Results for most of the tensile strength parameters tested were similar in the 2 groups of catheters at the end of 2 weeks of incubation. In particular, the force-at-break value of all tested catheters remained much greater than that recommended by industry standards.

The use of the novel catheter lock solution did not impair the mechanical integrity or increase the propensity for fracture of hemodialysis catheters.

To develop and validate an algorithm to identify and classify noninvasive infections due to Staphylococcus aureus by using positive clinical culture results and administrative data.

Retrospective cohort study.

Veterans Affairs Maryland Health Care System.

Data were collected retrospectively on all S. aureus clinical culture results from samples obtained from nonsterile body sites during October 1998 through September 2008 and associated administrative claims records. An algorithm was developed to identify noninvasive infections on the basis of a unique S. aureus-positive culture result from a nonsterile site sample with a matching International Classification of Diseases, Ninth Revision (ICD-9-CM), code for infection at time of sampling. Medical records of a subset of cases were reviewed to find the proportion of true noninvasive infections (cases that met the Centers for Disease Control and Prevention National Healthcare Safety Network [NHSN] definition of infection). Positive predictive value (PPV) and negative predictive value (NPV) were calculated for all infections and according to body site of infection.

We identified 4,621 unique S. aureus-positive culture results, of which 2,816 (60.9%) results met our algorithm definition of noninvasive S. aureus infection and 1,805 (39.1%) results lacked a matching ICD-9-CM code. Among 96 cases that met our algorithm criteria for noninvasive S. aureus infection, 76 also met the NHSN criteria (PPV, 79.2% [95% confidence interval, 70.0%–86.1%]). Among 98 cases that failed to meet the algorithm criteria, 80 did not meet the NHSN criteria (NPV, 81.6% [95% confidence interval, 72.8%–88.0%]). The PPV of all culture results was 55.4%. The algorithm was most predictive for skin and soft-tissue infections and bone and joint infections.

When culture-based surveillance methods are used, the addition of administrative ICD-9-CM codes for infection can increase the PPV of true noninvasive S. aureus infection over the use of positive culture results alone.

Vancomycin-resistant Enterococcus (VRE) bloodstream infections (BSIs) are associated with increased morbidity and mortality.

To determine the hospital costs and length of stay attributable to VRE BSI and vancomycin-sensitive Enterococcus (VSE) BSI and the independent effect of vancomycin resistance on hospital costs.

A retrospective cohort study was conducted of 21,154 nonsurgical patients admitted to an academic medical center during the period from 2002 through 2003. Using administrative data, attributable hospital costs (adjusted for inflation to 2007 US dollars) and length of stay were estimated with multivariate generalized least-squares (GLS) models and propensity score-matched pairs.

The cohort included 94 patients with VRE BSI and 182 patients with VSE BSI. After adjustment for demographics, comorbidities, procedures, nonenterococcal BSI, and early mortality, the costs attributable to VRE BSI were $4,479 (95% confidence interval [CI], $3,500-$5,732) in the standard GLS model and $4,036 (95% CI, $3,170-$5,140) in the propensity score-weighted GLS model, and the costs attributable to VSE BSI were $2,250 (95% CI, $l,758-$2,880) in the standard GLS model and $2,023 (95% CI, $1,588-$2,575) in the propensity score-weighted GLS model. The median values of the difference in costs between matched pairs were $9,949 (95% CI, $1,579-$24,693) for VRE BSI and $5,282 (95% CI, $2,042-$8,043) for VSE BSI. The costs attributable to vancomycin resistance were $1,713 (95% CI, $1,338-$2,192) in the standard GLS model and $1,546 (95% CI, $1,214-$1,968) in the propensity score-weighted GLS model. Depending on the statistical method used, attributable length of stay estimates ranged from 2.2 to 3.5 days for patients with VRE BSI and from 1.1 to 2.2 days for patients with VSE BSI.

VRE BSI and VSE BSI were independently associated with increased hospital costs and increased length of stay. Vancomycin resistance was associated with increased costs.