OBJECTIVE: To measure the prevalence of bacteremia and any correlation between signs and symptoms, risk factors, and laboratory data in elderly patients. DESIGN: Prospective analysis. All patients were contacted by the study nurse at 48 hours and 7 days after study entry. SETTING: Adult tertiary care hospital with an emergency department managing 48,000 visits yearly in a metropolitan area of 250,000. PARTICIPANTS: Members of the study population referred to the emergency department for medical or surgical problems. Of 113 nursing home patients, blood culture results were available for 111. MAIN OUTCOME MEASURES: Blood cultures were obtained by standard protocol. Demographic and medical information was collected from the medical record. Three groups of patients were compared with respect to symptoms, risk factors, laboratory data, and outcome. RESULTS: Group 1 (n = 86) had two sets of negative blood cultures. Group 2 (n = 10) had true-positive cultures. Group 3 (n = 15) had false-positive cultures of Staphylococcus epidermidis. The prevalence of bacteremia was 9.8% in the study population. No risk factors were predictive of bacteremia. Great variation in signs and symptoms were noted in all three groups, none correlating with bacteremia. Although seven of the 10 patients with positive cultures were febrile, this association did not reach statistical significance. All groups had high admission (> 50%) and mortality (20% to 37%) rates. CONCLUSIONS: The prevalence of bacteremia in the nursing home population presenting to the emergency department was 9.8%. The symptoms and signs of bacteremia in this population were variable and nonspecific. The high rate of false-positive cultures in this setting is of concern.
OBJECTIVE: To measure the prevalence of bacteremia and any correlation between signs and symptoms, risk factors, and laboratory data in elderly patients. DESIGN: Prospective analysis. All patients were contacted by the study nurse at 48 hours and 7 days after study entry. SETTING: Adult tertiary care hospital with an emergency department managing 48,000 visits yearly in a metropolitan area of 250,000. PARTICIPANTS: Members of the study population referred to the emergency department for medical or surgical problems. Of 113 nursing home patients, blood culture results were available for 111. MAIN OUTCOME MEASURES: Blood cultures were obtained by standard protocol. Demographic and medical information was collected from the medical record. Three groups of patients were compared with respect to symptoms, risk factors, laboratory data, and outcome. RESULTS: Group 1 (n = 86) had two sets of negative blood cultures. Group 2 (n = 10) had true-positive cultures. Group 3 (n = 15) had false-positive cultures of Staphylococcus epidermidis. The prevalence of bacteremia was 9.8% in the study population. No risk factors were predictive of bacteremia. Great variation in signs and symptoms were noted in all three groups, none correlating with bacteremia. Although seven of the 10 patients with positive cultures were febrile, this association did not reach statistical significance. All groups had high admission (> 50%) and mortality (20% to 37%) rates. CONCLUSIONS: The prevalence of bacteremia in the nursing home population presenting to the emergency department was 9.8%. The symptoms and signs of bacteremia in this population were variable and nonspecific. The high rate of false-positive cultures in this setting is of concern.