A six-year descriptive study of hospital-associated infection in trauma patients: demographics, injury features, and infection patterns.

BACKGROUND AND PURPOSE: Hospital-associated infection (HAI) is of concern to surgeons providing care for traumatized patients, as such patients have a higher rate of infection than other patients. Infection surveillance programs often study trauma patients within other populations (e.g., intensive care unit [ICU], surgery), and important issues may be missed. Information identifying trauma patients at risk, most frequent infection sites, and pathogens is of importance for surveillance and infection control. Measurement is essential to improving care. METHODS: We evaluated the HAI rate, demographics, injury characteristics, and HAI patterns (microorganisms, sites, antibiotics) in trauma patients (1996-2001). We used two-tailed Mann-Whitney and Fisher exact tests for univariate analysis and a stepwise multivariable logistic regression model for association of multiple variables with the development of HAI. RESULTS: The incidence of HAI was 501 (9.1%) in 5,537 patients. Trauma patients with HAI were older (p < 0.001), more severely injured (p < 0.001), and more likely to have multi-system trauma (p = 0.027). Development of HAI was associated with all injury sites except the face. The most common pathogens were gram-positive cocci, and the most common infection sites were urinary and respiratory, with 157 of 501 (31%) being ventilator-associated pneumonia. The antibiotics most commonly used were cephalosporins and fluoroquinolones. Of 5,537 trauma patients, 19 (0.3%) had Staphylococcus aureus resistant to methicillin, which was higher (p < 0.001) than in the non-trauma patients (176 in 146,727 [0.1%]). CONCLUSIONS: Hospital-associated infections occur frequently in trauma patients. This paper identifies populations to target for surveillance and HAI control initiatives. With increased interest in adverse event prevention and continuing quality of care improvement, these data provide a benchmark for this institution and others.

Trauma patient hospital-associated infections: risks and outcomes.

BACKGROUND: Trauma patients with surgical procedures, acute lung injury (ALI), systemic inflammatory response syndrome (SIRS), and longer exposure to invasive devices may be at increased risk for hospital-associated infection (HAI). HAIs have been shown to affect outcome measures, but the extent is not well studied. METHODS: An infection control team identified HAIs in trauma patients from 1996 through 2001. The authors evaluated the relation of HAI to surgical procedures, ALI, SIRS, and device exposure time by comparing groups with and without HAI using Fisher's exact and Mann-Whitney tests. Using multiple linear and logistic regressions, the authors evaluated associations of HAI, age, and Injury Severity Score (ISS) with length of stay (LOS), cost of care, and mortality. They used Cox proportional hazard regression to further explore the relations of HAI, age, and ISS to LOS. RESULTS: In 501 of 5,537 trauma patients with HAI (9.1%), the percent having surgical procedures, ALI, and SIRS was significantly higher (p < 0.001). Exposure to all devices studied was significantly longer (p < 0.001) in HAI patients. When the population was controlled for age and ISS, HAI patients had longer lengths of stay (LOSs) and higher costs. Age had less effect than ISS on LOS, and the effect of increases in age was greater as ISS increased. ISS had a greater effect than HAIs on LOS. HAIs increased LOS more in patients less severely injured. When comparing patients with and without HAI, no difference in mortality rates was detected. CONCLUSION: In this study of trauma patients, ISS had the greatest effect on LOS, but increased age and presence of HAI did increase LOS and cost of care. HAI increased LOS more in the less severely injured patients.

Adverse drug events in trauma patients.

BACKGROUND: Adverse drug events (ADEs) are noxious and unintended results of drug therapy. ADEs have been shown to be a risk to hospitalized patients. The purpose of this study was to determine the rate and nature of ADEs in trauma patients and to characterize the population at risk. METHODS: An electronic medical record, a hospital wide computerized surveillance program, and a clinical pharmacist prospectively investigated ADEs in 4,320 trauma patients from 1996 through 1999. RESULTS: The rate of ADEs in trauma patients (98/4320, 2.3%) was twice that of non-trauma hospital patients (1,111/96,218, 1.2%, p < 0.001). Traumatized females had ADEs 1.5 times more often than traumatized males (2.7% versus 1.8%, p = 0.052). The medication class most often associated with ADEs was analgesics with 54% involving morphine and 20% involving meperidine. The most common ADEs were nausea, vomiting, and itching. Only one ADE was directly attributed to a medical error. CONCLUSIONS: Trauma patients are at double the risk for ADEs. Analgesics are particularly associated with ADEs and use should be carefully monitored.