Abstract

A population-based active-surveillance study of the Calgary Health Region (population, 929,656) was conducted from May 1999 to April 2000, to define the epidemiology of invasive Staphylococcus aureus (ISA) infections. The annual incidence was 28.4 cases/100,000 population; 46% were classified as nosocomial. Infection was most common in people at the extremes of the age spectrum and in males. Several conditions were associated with acquisition of ISA infection, and the highest risk was observed in persons undergoing hemodialysis or peritoneal dialysis and in persons infected with human immunodeficiency virus. Forty-six patients (19%) died. Significant independent risk factors for mortality included positive blood-culture result, respiratory focus, empirical antibiotic therapy, and older age. A higher systolic blood pressure at presentation was associated with reduced case-fatality rate. ISA infections are common, with several definable groups of patients at increased risk for acquisition and death from these infections. This study provides important data on the burden of ISA disease and identifies risk groups that may potentially benefit from preventive efforts

Staphylococcus aureus is one of the most important bacterial pathogens that causes human disease and death. This organism commonly causes community-acquired superficial infections of soft tissues but has the potential to cause severe infections, including primary bacteremia, endovascular infection, ventriculitis, deep tissue abscesses, and bone and joint infections [1]. S. aureus is an important cause of nosocomial infections and has particular impact on critically ill patients [2, 3]. Several factors that increase the risk of acquisition of invasive S. aureus (ISA) infection have been suggested, such as diabetes, alcohol abuse, immunosuppression, nasal colonization by S. aureus prolonged hospital or intensive care unit (ICU) admission, intravenous drug abuse, hemodialysis, human immunodeficiency virus (HIV) infection, older age, and use of intravenous cannulas [4–8]. S. aureus infections have recently become more severe, because of the appearance of strains with reduced susceptibility to conventional antibiotics, such as vancomycin [9, 10]

Despite their importance, the epidemiology of ISA infections and the risk factors for acquisition have not been defined by use of population-based study design. Other studies have been limited either by the inclusion of only selected patients with ISA or by the failure to include clinical information [11–14]. As a result, the general-population incidence of and risk factors for acquisition of these infections are not known. The objectives of this study are to define the incidence of and the risk factors for acquisition of ISA disease, in a large, well-defined population, to characterize their clinical spectrum, and to characterize the risk factors for mortality

Patients, Materials, and Methods

Patient populationThe Calgary Health Region (CHR) provides virtually all medical and surgical care to the residents of the cities of Calgary and Airdrie and ∼20 nearby smaller towns. Only patients requiring liver, heart, and/or lung transplantation are routinely referred out of the region. All persons who resided in the CHR and who had an ISA infection between 1 May 1999 and 30 April 2000 were included in the study

Study protocolA population-based active-surveillance cohort design, with chart review for clinical patient information, was used. Surveillance for ISA infection was done by Calgary Laboratory Services, a regional laboratory that has handled all routine bacterial specimens from CHR hospitals and the CHR community, since 1996 [15]. Trained research nurses obtained clinical information from patient charts by use of standardized forms. A study physician (K.B.L.) assigned a primary diagnosis relating to ISA infection and reviewed all information before it was entered into the database

DefinitionsISA infection was defined by the isolation of S. aureus from blood, cerebrospinal fluid (CSF), pleural or synovial fluid, or aseptically obtained deep-tissue aspirates or surgical-tissue samples. Community-acquired infections were those that were either present or incubating at the time of hospital admission or associated with the first positive culture result obtained within 48 h of admission [16]. A case was classified as nosocomial if the patient was a nursing-home resident or if the first S. aureus culture was achieved >48 h after admission to the hospital. In patients, infections related to receipt of outpatient health-care services but not to hospital or nursing-home admission were classified as community acquired. A primary diagnosis was assigned on the basis of the clinical, radiological, and microbiological information available by use of a priori-defined criteria. Clinical isolates were isolated, were confirmed as S. aureus and were tested for antimicrobial susceptibility, by standard techniques [17, 18]. Phenotypic methicillin-resistant S. aureus strains were confirmed to be mecA-positive by polymerase chain reaction assay [19]

Data sourcesLocal data sources that account for all patients in the CHR were used to either precisely ascertain or estimate the study-midyear prevalence of potentially predisposing conditions for ISA infection. The study-midyear population was 929,656, according to the Alberta Health Population Registry. The number of HIV-seropositive persons in the CHR was estimated by use of statistics from the Southern Alberta Clinic [20]. The midyear number of patients who had undergone kidney transplantation or who required chronic dialysis was obtained from the regional database [21]. The number of patients who had undergone liver or heart/lung transplantation was ascertained from the Provincial Liver Transplant Program and Cardiothoracic Surgery Research Office, both of which are at the University of Alberta. The numbers of patients who had had major trauma or had been admitted to ICUs were obtained from regional trauma and ICU databases, respectively [22]. For other potentially predisposing conditions, numbers of patients were estimated by use of Canadian or international data. The prevalences of chronic medical conditions, intravenous drug abuse, and alcoholism were estimated by use of data from the 1996 (Canadian) National Population Health Survey [23]. The prevalence of a past or present cancer diagnosis was based on calculated Canadian rates [24], and the prevalence of systemic lupus erythematosus and rheumatoid arthritis was based on American rates [25, 26]

Data management and analysisData were analyzed by use of the Statistica (version 5.0; Statsoft) and Stata (version 7.0; Stata) software programs. Means with SDs were used to describe normally distributed variables, and medians with interquartile ranges (IQRs) were used to describe nonnormally distributed variables. Differences in proportions were compared by Fisher's exact test, means were compared by Student's t test, and medians were compared by Wilcoxon&amp;rank sum test. The relative risks (RRs) and 95% confidence intervals (CIs), for acquisition of ISA associated with predisposing conditions, were calculated as described elsewhere [27]. A logistic-regression model was developed to assess risk factors for mortality from ISA infection. Factors found to be associated with death in univariate analyses (P⩽.1) and in previous studies, and clinically suspected variables were initially included, and backward stepwise variable elimination was used to develop the most efficient model. Interactions were tested among the final variables in the model, in a pairwise fashion. Results were reported as odds ratios (ORs) with 95% CIs. A 2-tailed P<.05 was considered significant for all comparisons, and adjustments for multiple testing were not made

Results

During the 1-year study, 264 CHR residents had ⩾1 ISA infection; 7 patients had a second discrete episode of illness, but only first episodes were included in the analysis. Demographic data were available for all 264 patients who had fulfilled inclusion criteria, and clinical information was available for 247 (94%) patients: 25 (9%) patients were residents of nursing homes, and 218 (83%) patients had been admitted to a CHR acute-care hospital

Incidence and risk factors for acquisition The annual incidence of ISA infection among residents of the CHR was 28.4 cases/100,000 population. The monthly occurrence of ISA disease is shown in figure 1. One hundred twenty-one (46%) patients had nosocomial ISA infections; 6 (24%) patients had health care-related infections (all catheter-associated infections, not associated with a prior hospital admission) that were classified as community acquired; otherwise, all community-acquired infections were unrelated to prior health care. Of the 25 patients who had infections classified, on the basis of nursing-home residency, as nosocomially related, 8 (32%) developed their ISA infections >48 h after admission to an acute-care hospital

Figure 1

Occurrence of invasive Staphylococcus aureus infections in Calgary, Canada (May 1999 to April 2000). Data are no. of patients with nosocomial and community-acquired infections

Figure 1

Occurrence of invasive Staphylococcus aureus infections in Calgary, Canada (May 1999 to April 2000). Data are no. of patients with nosocomial and community-acquired infections

The median age of the patients was 58.8 years (range, 7 days to 97 years), and the incidence of infection was highest among persons ⩾65 years old (figure 2). The overall rate of infection was higher in males than in females (35.4 vs. 21.5/100,000 population; RR, 1.6; 95% CI, 1.3–2.1), and this was most pronounced among those patients ⩾50 years old (RR, 2.1; 95% CI, 1.5–3.0). Several groups were identified as being at significantly higher risk for acquisition of ISA infection (table 1), with patients undergoing dialysis and patients with HIV infection at the highest levels of risk. Of 61 patients who required admission to ICUs, 20 (33%) had community-acquired ISA infection. Among all hospital admissions (n=94,407), both ICU admission (RR, 12.2; 95% CI, 8.8–16.8) and multiple trauma (RR, 14.6; 95% CI, 7.3–26.5) were associated with higher rates of ISA infection

Figure 2

Age-specific incidence of invasive Staphylococcus aureus infections in Calgary, Canada (May 1999 to April 2000). Data are annual incidence per 100,000 population

Figure 2

Age-specific incidence of invasive Staphylococcus aureus infections in Calgary, Canada (May 1999 to April 2000). Data are annual incidence per 100,000 population

Table 1

Risk of invasive Staphylococcus aureus infection, associated with selected underlying conditions, in adults ⩾20 years old

Table 1

Risk of invasive Staphylococcus aureus infection, associated with selected underlying conditions, in adults ⩾20 years old

Primary diagnosesThe most common primary diagnoses were soft tissue, respiratory, and bone and joint infections (table 2). Among patients classified with a respiratory-infection diagnosis, 41 had pneumonia alone and 4 had pneumonia with concomitant empyema. Another 9 patients with empyema likely had associated pneumonia, but no infiltrates were described by the reporting radiologist, and thus they did not meet our prespecified diagnostic criteria, which mandated a chest radiographic reporting of an infiltrate consistent with pneumonia. Endovascular infection was diagnosed as endocarditis (10 patients, 4 of whom had metastatic bone or joint infection), septic thrombophlebitis (3 patients), and infected axillofemoral arterial graft (1 patient). Bone and joint infections were diagnosed as septic arthritis (22 patients), osteomyelitis (16 patients), and both diseases (5 patients). Intraabdominal or pelvic infections were diagnosed as pelvic abscesses (5 patients: 2 as psoas, 2 as endometrium, and 1 as periprostatic), abdominal abscesses (3 patients), septic prostatitis (2 patients), and diffuse peritonitis (3 patients). Of 60 soft tissue infections, 10 (17%) were associated with postoperative wounds and 38 (63%) were community acquired. The central nervous system infections were diagnosed as meningitis (2 patients) and, in neurosurgical patients, ventriculitis (4 patients). No cases fulfilled criteria for toxic shock syndrome [28]

Table 2

Clinical diagnoses associated with invasive Staphylococcus aureus infections in 247 patients, in Calgary, Canada (May 1999–April 2000)

Table 2

Clinical diagnoses associated with invasive Staphylococcus aureus infections in 247 patients, in Calgary, Canada (May 1999–April 2000)

MicrobiologyA total of 550 invasive isolates of S. aureus were obtained, from different sites or on different days, from the 264 patients in the study; 123 patients (47%) had 1, 79 (30%) had 2, and 62 (23%) had ⩾3 positive culture results; 166 patients (63%) had ⩾1 positive blood-culture result, and 93 (35%) had a positive result on ⩾1 deep-tissue sample. Culture of synovial fluid yielded positive results for 26 patients (10%), pleural fluid for 13 (5%), CSF for 6 (2%), and peritoneal fluid for 4 (2%). Twenty patients (8%) had a concomitant isolate from a sterile site that yielded another organism (typical blood-culture contaminants excluded). These organisms were gram-negative bacilli (7 patients), Enterococcus species (6 patients), Streptococcus species (5 patients), Clostridium butyricum (1 patient), and Candida albicans (1 patient). Seven patients (3%) had isolates of methicillin-resistant S. aureus, and 5 of those patients had nosocomial infections. Of 29 patients who had S. aureus bacteremia without focus, 26 had ⩾1 of the following: multiple positive cultures, fever, or elevated white blood cell counts. Of the remaining 3 patients, 2 were treated with antibiotics for 10 days, and 1 was undergoing dialysis—we could not obtain enough information regarding the reason the culture samples were drawn from the patient undergoing dialysis to conclude whether this was a possible contaminant

TreatmentOf the patients for which we had sufficient data regarding treatments, 44% (104/238) were receiving antibiotic therapy at the time that samples were drawn, and 74% (174/235) received adequate empirical antibiotic therapy within 8 h of the time first samples were drawn. For patients who survived infection, the median treatment duration was 19 days (IQR, 13–42.5 days; n=177), with a median of 13 days (IQR, 8–30 days; n=187) of intravenous treatment. Patients had an overall median length of stay in the hospital of 19 days (IQR, 9–43 days) and a median length of stay in the hospital after diagnosis of 14 days (IQR, 8–32 days). Patients with community-acquired infections who survived had a much shorter median length of stay in the hospital after diagnosis than did patients with nosocomial ISA (10 vs. 26 days; n=169; P<.001). For patients with nosocomial ISA, the median length of stay in the hospital, from admission to first culture positivity, was 7 days (IQR, 0.5–18.5 days)

MortalityForty-six patients (19%) died before being discharged from the hospital, for a mortality rate of 4.9 deaths/100,000 population, per year. Factors found to be significant (P⩽.1) categorical predictors of case fatality in univariate analysis are shown in table 3. The mean age of patients who died was significantly higher than that of those who survived (mean±SD, 67.5±18.3 vs. 53.1±22.9 years; P<.001). Although no difference in case-fatality rate was observed between those with nosocomial infection and those with community-acquired infection (26/119 vs. 20/128; P=.3), the median prediagnosis length of stay in the hospital for those who died (3.0 days; IQR, 0–23.0 days) was significantly higher than the median for those who survived (0 days; IQR, 0–7.0 days) (P=.04). Patients who died had somewhat lower mean systolic blood pressure at presentation than did patients who survived (mean±SD, 118.8±22.4 vs. 128.0±26.5 mmHg; P=.05). No other baseline vital signs or laboratory tests were significantly different, at presentation, between patients who survived and patients who died. Multivariable logistic regression techniques were used to assess risk factors for death in patients with ISA infection, and the final model (n=198) included age, sex, blood-culture positivity, empirical antibiotic therapy within 8 h of presentation, presenting systolic blood pressure, and diagnosis with respiratory focus. The model variables, with associated ORs, are shown in table 4

Table 3

Univariate analysis of significant (P⩽.1) predictive factors for death among patients with invasive Staphylococcus aureus infections

Table 3

Univariate analysis of significant (P⩽.1) predictive factors for death among patients with invasive Staphylococcus aureus infections

Table 4

Logistic-regression modeling of risk factors for mortality in patients with invasive Staphylococcus aureus infections

Table 4

Logistic-regression modeling of risk factors for mortality in patients with invasive Staphylococcus aureus infections

Discussion

This study documents the importance of S. aureus as an invasive pathogen and provides baseline data to use if new risk factors or preventive methods evolve in the future. The incidence of ISA disease is comparable to that of invasive pneumococcal disease and is 5–20-fold higher than that of invasive group A streptococcal or group B streptococcal infections, in nonpregnant adults, as is reported in similarly designed studies of other regions [27, 29, 30]. Hospital-based studies suggest that the rate of S. aureus bacteremia has been increasing during the last 2 decades [4, 31]. However, these rates are dependent on hospitalization, and it is possible that the increase may reflect, in part, differences in referral patterns or admission rates rather than a higher infection rate. Population-based studies in which all persons in a defined population are identified by fulfilling a case definition minimize or eliminate this important bias and provide more valid rates for comparison over time

The population-based design facilitates comparison of infection rates among different regions. In this study, 166 patients had bacteremia, 26 patients had positive synovial fluid-culture results, and 6 patients had positive CSF-culture results, for annual rates of 17.9, 2.4, and 0.6/100,000 population, respectively. Other studies of ISA either have evaluated only bacteremic infections or have been hospital-based studies of tertiary care centers that are highly subject to selection bias [11, 12, 32–34]. We observed a rate for septic arthritis due to S. aureus similar to that observed in Amsterdam (2.5/100,000 population), by Kaandorp et al. [13], and a rate for combined bacteremia and CSF infection similar to that observed in Wales (17.9/100,000 population), by Morgan et al. [14]. The Amsterdam study was a study of incidence and sources of bacterial arthritis, not specifically due to S. aureus whereas the Wales study was designed to establish incidences, without evaluation of risk factors. In a case-control study limited to S. aureus bacteremia acquired in hospitals, Jensen et al. [12] estimate an incidence of bacteremia in Copenhagen of 30/100,000 population, a rate nearly twice as high as that in our study. Rates for ISA infections may vary among population-based studies for a number of reasons, including physician thresholds for or practices of ordering cultures, regional differences in use of empirical antibiotic therapies, different demographics of at-risk populations, and different laboratory practices. In addition, if surveillance is done in the setting of an outbreak, reported rates of infection might overestimate the baseline occurrence of disease in that population. To our knowledge, the present study was not conducted during an outbreak of S. aureus infection. This is supported by our observation of a relatively stable monthly incidence (figure 1). The overall annual incidence of 28.4 cases/100,000 population likely reflects a true baseline annual incidence of ISA disease in the Calgary region

This is the first study to quantify risk factors for acquisition of ISA infections in the general population and to identify risk groups that may benefit from preventive therapies. The risks that were determined for acquisition of ISA—for dialysis, transplant, and HIV-positive patients—should be accurate, because the numbers of persons in the region with these conditions are well defined. However, the findings for risks associated with other underlying conditions should be interpreted cautiously. This is because the population rates for these conditions were estimated on the basis of Canadian or American survey data and were compared with data obtained by chart review. As a result, the error associated with the estimates for the risks in table 1 is likely to be higher than calculated. However, we performed a sensitivity analysis of the risk factors in table 1 and found that each remains significant, even if the estimate of the underlying population prevalence is doubled (data not shown). Although it seems plausible that several risk factors, such as diabetes or immunosuppression, associated with transplantation or with HIV positivity inherently increase the risk for ISA infection, the reasons that heart disease, stroke, or chronic obstructive pulmonary disease increase the risk are less apparent. The relative risks reported in this study reflect overall risk in the entire population and are likely confounded, at least in part, by the fact that many of these patients may have a particular disease and therefore may be at increased risk for ISA infection by virtue of their increased rate of hospitalization or treatment with instruments—such as catheters or endotracheal tubes—which are risk factors for nosocomial infection. With the study design that we used, multivariable analysis could not be done for the risk factors associated with ISA or univariable analysis, because only patients with ISA were included. Irrespective of the explanation of why certain underlying factors increase risk for acquisition of ISA disease, this study identifies those persons for which preventive efforts—such as vaccination, which has undergone recent advances—may be focused [35]

ISA disease is associated with substantial morbidity and mortality. Patients who survived infection required hospitalization and intravenous therapy for an average of 2 weeks after presentation with ISA infection. The duration of infection-related stays in hospital beds in our region is important, because a regionwide home parenteral-antibiotic program is in place that rapidly facilitates discharge of patients who continue to receive intravenous therapy once they are clinically stable [36]. The number of days a patient stays in a hospital bed, because of ISA infection, is likely to be even higher in other regions that do not have similar home antibiotic programs. In our region, nearly 1 in 5 patients hospitalized with ISA infections died before being discharged from the hospital. Although patients' deaths may have ultimately been a result of noninfectious causes, S. aureus was likely a contributing or inciting event in the majority of cases, because most infections were community acquired and usually were the reason for hospital admission. The present study is the first study to document the annual population rate of mortality due to ISA (4.9/100,000 population, per year). It provides important information on the burden of ISA disease, information that may be useful for setting patient care and research funding priorities

A number of conditions were identified as risk factors for death in patients hospitalized with ISA infection (table 4). Previous hospital-based studies have identified hypotension, older age, renal failure, diabetes mellitus, and lung focus as independent risk factors for death, in patients with S. aureus bacteremia [32–34, 37, 38]. However, the present study is the first population-based study to identify that a positive blood culture result is an independent predictor of death in patients with ISA infection. In contrast to other studies, the presence, at presentation, of renal failure, nosocomial acquisition (rather than community acquisition), or diabetes mellitus did not increase the risk of death in the present study [33, 34, 38]. A number of studies of severe infections, such as S. aureus bacteremia, have identified inadequate empirical therapy as a risk factor for death. It is therefore curious that patients in our study who received empirical antibiotics within 8 h of presentation had a higher risk of death than did those who did not receive antibiotics within 8 h of presentation [37, 39, 40]. The most likely explanation for this finding is that patients with poorer prognoses presented with more severe clinical findings and were more likely to be treated empirically. Although the finding is controversial, a number of studies have suggested that methicillin resistance is a risk factor for mortality in patients with S. aureus infection [41, 42]. Because methicillin-resistant S. aureus infection was rarely observed in this study, a meaningful comparison of the outcomes of infections due to methicillin-susceptible and methicillin-resistant S. aureus was not possible. Despite a high level of acuity among patients treated in the CHR, it is not clear why such low rates of antimicrobial resistance were observed, given the high rates that were seen in many other North American health-care centers [22, 43]

The epidemiology of an infectious disease is optimally defined by population-based assessment. We have provided important information on the incidence of ISA infection in a general population and have defined several groups at increased risk for acquisition of and death from these infections. This study documents the substantial burden of ISA disease and identifies risk groups that may potentially benefit from future preventive efforts

Acknowledgments

We thank Kathy Lister and Bryan Haggerty, for reviewing charts; Stephanie Hui and Doreen Ma, for preparing and entering the database; and Heather Semeniuk at Calgary Laboratory Services, for preparing microbiology reports. We thank Stafford Dean, Steven Edworthy, John Gill, John Kortbeek, Dennis Modry, Garth Mortis, and Brendan Murphy for providing data on local underlying-condition population-prevalence rates

References

1.
Lowy
F
Staphylococcus aureus infections
N Engl J Med
 , 
1998
, vol. 
339
 (pg. 
520
-
32
)
2.
Richards
M
Edwards
J
Culver
D
Gaynes
R
Nosocomial infections in medical intensive care units in the United States. National Nosocomial Infections Surveillance System
Crit Care Med
 , 
1999
, vol. 
27
 (pg. 
887
-
92
)
3.
Vincent
JL
Bihari
D
Suter
P
, et al.  . 
The prevalence of nosocomial infection in intensive care units in Europe: results of the European Prevalence of Infection in Intensive Care (EPIC) study
JAMA
 , 
1995
, vol. 
274
 (pg. 
639
-
44
)
4.
Steinberg
J
Clark
C
Hackman
B
Nosocomial and community‐acquired Staphylococcus aureus bacteremias from 1980 to 1993: impact of intravascular devices and methicillin resistance
Clin Infect Dis
 , 
1996
, vol. 
23
 (pg. 
255
-
9
)
5.
O’Kane
G
Gottlieb
T
Bradbury
R
Staphylococcal bacteraemia: the hospital or home? A review of Staphylococcus aureus bacteraemia at Concord Hospital in 1993
Aust N Z J Med
 , 
1998
, vol. 
28
 (pg. 
23
-
7
)
6.
Pujol
M
Pena
C
Pallares
R
, et al.  . 
Nosocomial Staphylococcus aureus bacteremia among nasal carriers of methicillin‐resistant and methicillin‐susceptible strains
Am J Med
 , 
1996
, vol. 
100
 (pg. 
509
-
16
)
7.
Weinke
T
Schiller
R
Fehrenbach
F
Pohle
H
Association between Staphylococcus aureus nasopharyngeal colonization and septicemia in patients infected with the human immunodeficiency virus
Eur J Clin Microbiol Infect Dis
 , 
1992
, vol. 
11
 (pg. 
985
-
9
)
8.
von Eiff
C
Becker
K
Machka
K
Stammer
H
Peters
G
Nasal carriage as a source of Staphylococcus aureus bacteremia
N Engl J Med
 , 
2001
, vol. 
344
 (pg. 
11
-
6
)
9.
Smith
T
Pearson
M
Wilcox
K
, et al.  . 
Emergence of vancomycin resistance in Staphylococcus aureus
N Engl J Med
 , 
1999
, vol. 
340
 (pg. 
493
-
501
)
10.
Sieradzki
K
Roberts
R
Haber
S
Tomasz
A
The development of vancomycin resistance in a patient with methicillin‐resistant Staphylococcus aureus infection
N Engl J Med
 , 
1999
, vol. 
340
 (pg. 
517
-
23
)
11.
Morin
C
Hadler
J
Population‐based incidence and characteristics of community‐onset Staphylococcus aureus infections with bacteremia in 4 metropolitan Connecticut areas, 1998
J Infect Dis
 , 
2001
, vol. 
184
 (pg. 
1029
-
34
)
12.
Jensen
A
Wachmann
C
Poulsen
K
, et al.  . 
Risk factors for hospital‐acquired Staphylococcus aureus bacteremia
Arch Intern Med
 , 
1999
, vol. 
159
 (pg. 
1437
-
44
)
13.
Kaandorp
C
Dinant
H
vandeLaar
M
Bernelot Moens
H
Prins
A
Dijkmans
B
Incidence and sources of native and prosthetic joint infection: a community based prospective survey
Ann Rheum Dis
 , 
1997
, vol. 
56
 (pg. 
470
-
5
)
14.
Morgan
M
Salmon
R
Keppie
N
Evans‐Williams
D
Hosein
I
Looker
D
All Wales surveillance of methicillin‐resistant Staphylococcus aureus (MRSA): the first year’s results
J Hosp Infect
 , 
1999
, vol. 
41
 (pg. 
173
-
9
)
15.
Church
D
Hall
P
Centralization of a regional clinical microbiology service: the Calgary experience
Can J Infect Dis
 , 
1999
, vol. 
10
 (pg. 
393
-
402
)
16.
Garner
J
Jarvis
W
Emori
T
Horan
T
Hughes
J
CDC definitions for nosocomial infections, 1988
Am J Infect Control
 , 
1988
, vol. 
16
 (pg. 
128
-
40
)
17.
National Committee for Clinical Laboratory Standards
Performance standards for antimicrobial disk susceptibility tests: approved standard. 7th ed. NCCLS document M2‐A7, vol. 20 no. 1
 , 
2000
Villanova, PA
National Committee for Clinical Laboratory Standards
18.
National Committee for Clinical Laboratory Standards
Performance standards for antimicrobial susceptibility testing: eighth informational supplement. NCCLS document M100‐S8, vol. 18 no. 1
 , 
1998
Villanova, PA
National Committee for Clinical Laboratory Standards
19.
Pfaller
M
Isenberg
H
Detection of methicillin resistance in staphylococci by PCR (section 10.5.3)
Essential procedures for clinical microbiology
 , 
1998
Washington, DC
American Society for Microbiology
(pg. 
666
-
8
)
20.
Southern Alberta Clinic
  
21.
Manns
B
Mortis
G
Taub
K
McLaughlin
K
Donaldson
C
Ghali
W
The Southern Alberta Renal Program database: a prototype for patient management and research initiatives
Clin Invest Med
 , 
2001
, vol. 
24
 (pg. 
164
-
70
)
22.
Laupland
K
Zygun
D
Davies
H
Church
D
Louie
T
Doig
C
Incidence and risk factors for acquiring nosocomial urinary tract infection in the critically ill
J Crit Care
 , 
2002
, vol. 
17
 (pg. 
50
-
7
)
23.
National Population Health Survey
  
24.
Mao
Y
Morrsion
H
Semenciw
R
Robson
D
Wigle
D
The prevalence of cancer in Canada
Can J Public Health
 , 
1991
, vol. 
82
 (pg. 
61
-
2
)
25.
Gabriel
S
Crowson
C
O’Fallon
W
The epidemiology of rheumatoid arthritis in Rochester, Minnesota, 1955–1985
Arthritis Rheum
 , 
1999
, vol. 
42
 (pg. 
415
-
20
)
26.
Hochberg
M
Perlmutter
D
Medsger
T
, et al.  . 
Prevalence of self‐reported physician‐diagnosed systemic lupus erythematosus in the USA
Lupus
 , 
1995
, vol. 
4
 (pg. 
454
-
6
)
27.
Davies
H
McGeer
A
Schwartz
B
, et al.  . 
Invasive group A streptococcal infections in Ontario, Canada
N Engl J Med
 , 
1996
, vol. 
335
 (pg. 
547
-
54
)
28.
Centers for Disease Control and Prevention
Case definitions for public health surveillance
MMWR Morb Mortal Wkly Rep
 , 
1990
, vol. 
39
 
RR‐13
(pg. 
33
-
9
)
29.
Whitney
C
Farley
M
Hadler
J
, et al.  . 
Increasing prevalence of multidrug‐resistant Streptococcus pneumoniae in the United States
N Engl J Med
 , 
2000
, vol. 
343
 (pg. 
1917
-
24
)
30.
Farley
M
Harvey
R
Stull
T
, et al.  . 
A population‐based assessment of invasive disease due to group B streptococcus in nonpregnant adults
N Engl J Med
 , 
1993
, vol. 
328
 (pg. 
1807
-
11
)
31.
Banerjee
S
Emori
T
Culver
D
, et al.  . 
Secular trends in nosocomial bloodstream infections in the United States, 1980–1989
Am J Med
 , 
1991
, vol. 
91
 
suppl 3B
(pg. 
86S
-
9S
)
32.
McClelland
R
Fowler
V
Sanders
L
, et al.  . 
Staphylococcus aureus bacteremia among elderly vs. younger adult patients
Arch Intern Med
 , 
1999
, vol. 
159
 (pg. 
1244
-
7
)
33.
Mylotte
J
Tayara
A
Staphylococcus aureus bacteremia: predictors of 30‐day mortality in a large cohort
Clin Infect Dis
 , 
2000
, vol. 
31
 (pg. 
1170
-
4
)
34.
Willcox
P
Rayner
B
Whitelaw
D
Community‐acquired Staphylococcus aureus bacteraemia in patients who do not abuse intravenous drugs
QJM
 , 
1998
, vol. 
91
 (pg. 
41
-
7
)
35.
Shinefield
H
Black
S
Fattom
A
, et al.  . 
Use of a Staphylococcus aureus conjugate vaccine in patients receiving hemodialysis
N Engl J Med
 , 
2002
, vol. 
346
 (pg. 
491
-
6
)
36.
Laupland
KB
Davies
HD
Calgary HPTP Study Group
Clinical and microbiologic features and outcome of olecranon septic bursitis managed in an ambulatory setting
Clin Invest Med
 , 
2001
, vol. 
24
 (pg. 
171
-
8
)
37.
Jensen
A
Wachmann
C
Espersen
F
Scheibel
J
Skinhoj
P
Frimodt‐Moller
N
Treatment and outcome of Staphylococcus aureus bacteremia
Arch Intern Med
 , 
2002
, vol. 
162
 (pg. 
25
-
32
)
38.
Hill
PC
Birch
M
Chambers
S
, et al.  . 
Prospective study of 424 cases of Staphylococcus aureus bacteremia: determination of factors affecting incidence and mortality
Intern Med J
 , 
2001
, vol. 
31
 (pg. 
97
-
103
)
39.
Ibrahim
E
Sherman
G
Ward
S
Fraser
V
Kollef
M
The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting
Chest
 , 
2000
, vol. 
118
 (pg. 
146
-
55
)
40.
Valles
J
Leon
C
Alvarez‐Lerma
F
Spanish Collaborative Group for infections in intensive care units of Sociedad Espanolade Medecina Intensivay Unidades Coronarias (SEMIUC)
Nosocomial bacteremia in critically ill patients: a multicenter study evaluating epidemiology and prognosis
Clin Infect Dis
 , 
1997
, vol. 
24
 (pg. 
387
-
95
)
41.
Conterno
L
Wey
S
Castelo
A
Risk factors for mortality in Staphylococcus aureus bacteremia
Infect Control Hosp Epidemiol
 , 
1998
, vol. 
19
 (pg. 
32
-
7
)
42.
Romero‐Vivas
J
Fernandez
C
Picazo
J
Mortality associated with nosocomial bacteremia due to methicillin‐resistant Staphylococcus aureus
Clin Infect Dis
 , 
1995
, vol. 
21
 (pg. 
1417
-
23
)
43.
Laupland
K
Zygun
D
Davies
H
Church
D
Louie
T
Doig
C
Population‐based assessment of intensive care unit–acquired bloodstream infections in adults: incidence, risk factors, and associated mortality rate
Crit Care Med
 , 
2002
, vol. 
30
 (pg. 
2462
-
7
)
This study was approved by the Conjoint Health Research Ethics Board, University of Calgary, and the Calgary Health Region, before commencement
Financial support: 2000 Bayer Healthcare/Canadian Infectious Diseases Society/Canadian Institutes of Health Research Infectious Diseases Research Fellowship (to K.B.L.); Alberta Heritage Foundation for Medical Research (AHFMR) Clinical Fellowship Award (to K.B.L.); Center for Advancement of Health (grant 15274/10413), Calgary Health Region. H.D.D. is a Medical Scholar of the AHFMR