Hospital-Acquired Infections Increase Costs and Risks for Trauma Patients

Whereas hospital-acquired infections (HAIs) cost the US healthcare system tens of billions of dollars annually and cause >90,000 deaths, authors of a retrospective study report that the burden of HAIs in trauma patients is especially high [Arch Surg. 2011;146(7):794-801]. Individuals hospitalized for a traumatic injury were more likely to die, required longer stays, and incurred greater expense if they developed an HAI during hospitalization compared with trauma patients who did not acquire an infection. These findings underscore the need to identify and adopt methods for preventing HAI infections in this patient population. The authors analyzed administrative data from the 2005 and 2006 Nationwide Inpatient Sample, which is a large database of claims incurred at rural and urban hospitals of all sizes across the country. The study included 155,891 patients hospitalized for ≥3 days for a traumatic injury, indicated by an International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code of 800-959.9. Trauma patients who did not develop an HAI during hospitalization made up the control group (n=148,880); the remaining patients (n=7011) were stratified into 4 subgroups, according to whether claims contained a diagnosis code for sepsis (n=2955), pneumonia (n=2479), Staphylococcus infection (n=3803), or Clostridium difficile–associated disease (CDAD; n=768). The authors noted some patients had >1 HAI and therefore fell into multiple categories. Compared with controls, pneumonia occurred more often among patients hospitalized for motor vehicle accidents (35.2% vs 18.4%, respectively) or pedestrian trauma (15.6% vs 7.4%, respectively). Control group patients were older than HAI group patients, with a median age of 57 years compared with 53 years, respectively, and they were more likely to be female (48.0% vs 33.4%, respectively). Overall, blunt trauma was the most common cause of injury in each group, although the affected body region(s) varied. After adjusting for differences in baseline patient characteristics, mechanism of injury, and type of injury, data showed the subgroup of patients with sepsis had nearly 6 times the risk of death compared with control group patients (odds ratio, 5.78; 95% confidence interval, 5.03-6.64; P<.001). The multivariate analysis revealed a mortality risk 1.5- to 1.9-fold higher for the remaining HAI subgroups as compared with the control group (P<.005 for all comparisons). HAI was associated with significantly higher inpatient costs, with patients with an HAI accruing median costs that were 2.6 to 6 times greater than the median of $12,849 in costs for control group patients. Median costs were highest for the pneumonia subgroup, totaling $77,393. Costs were $60,398 for the sepsis subgroup, $47,908 for the Staphylococcus subgroup, and $33,294 for the CDAD subgroup. In addition, trauma patients treated for an HAI had nearly twice the length of stay (LOS) as patients who were not treated for an HAI (P<.001). The median overall LOS was 6 days in the control group, 16 days in the CDAD group, 17 days in the Staphylococcus arm, 24 days in the pneumonia group, and 20 days in the sepsis arm. The records for patients with an HAI did not specify the portion of their stay attributable to the infection, and the authors described this as a limitation of their findings, since each day a patient is hospitalized increases the opportunity for infection to occur. Although data failed to indicate whether a patient had one of these infections on admission, the authors believed preexisting infection was unlikely in this patient population. Private and public insurers are adopting payment strategies to pressure hospitals to do more to prevent HAIs. “In light of the preventability of many hospital-acquired conditions and the magnitude of the clinical and economic burden of HAIs, the current emphasis on implementing interventions aiming to decrease the incidence of HAIs may have a potentially large impact,” said the authors, who suggested additional studies assess the effect of Medicare payment changes on the incidence of HAIs.