The incidence of Clostridium difficile infection (CDI) has increased among hospitalized patients and is a common complication of leukemia. We investigated the risks for and outcomes of CDI in hospitalized leukemia patients.

Adults with a primary diagnosis of leukemia were extracted from the United States Nationwide Inpatient Sample database, 2005–2011. The primary outcomes of interest were CDI incidence, CDI-associated mortality, length of stay (LOS), and charges. In a secondary analysis, we sought to identify independent risk factors for CDI in leukemia patients. Logistic regression was used to derive odds ratios (ORs) adjusted for potential confounders.

A total of 1,243,107 leukemia hospitalizations were identified. Overall CDI incidence was 3.4% and increased from 3.0% to 3.5% during the 7-year study period. Leukemia patients had 2.6-fold higher risk for CDI than non-leukemia patients, adjusted for LOS. CDI was associated with a 20% increase in mortality of leukemia patients, as well as 2.6 times prolonged LOS and higher hospital charges. Multivariate analysis revealed that age >65 years (OR, 1.13), male gender (OR, 1.14), prolonged LOS, admission to teaching hospital (OR, 1.16), complications of sepsis (OR, 1.83), neutropenia (OR, 1.35), renal failure (OR, 1.18), and bone marrow or stem cell transplantation (OR, 1.27) were significantly associated with CDI occurrence.

Hospitalized leukemia patients have greater than twice the risk of CDI than non-leukemia patients. The incidence of CDI in this population increased 16.7% from 2005 to 2011. Development of CDI in leukemia patients was associated with increased mortality, longer LOS, and higher hospital charges.

Infect Control Hosp Epidemiol 2015;36(7):794–801

Paramedics often are asked to care for patients at the end of life. To do this, they must communicate effectively with family and caregivers, understand their legal obligations, and know when to withhold unwanted interventions. The objectives of this study were to ascertain paramedics' attitudes toward end-of-life (EOL) situations and the frequency with which they encounter them; and to compare paramedics' preparation during training for a variety of EOL care skills.

A written survey was administered to a convenience sample of paramedics in two cities: Denver, Colorado and Los Angeles, California. Questions addressed: (1) attitudes toward EOL decision-making in prehospital settings; (2) experience (number of EOL situations experienced in the past two years); (3) importance of various EOL tasks in clinical practice (pronouncing and communicating death, ending resuscitation, honoring advance directives (ADs)); and (4) self-assessed preparation for these EOL tasks. For each task, importance and preparation were measured using a four-point Likert scale. Proportions were compared using McNemar chi-square statistics to identify areas of under or over-preparation.

Two hundred thirty-six paramedics completed the survey. The mean age was 39 years (range 22–59 years), and 222 (94%) were male. Twenty percent had >20 years of experience. Almost all participants (95%; 95% CI = 91–97%) agreed that prehospital providers should honor field ADs, and more than half (59%; 95% CI = 52–65%) felt that providers should honor verbal wishes to limit resuscitation at the scene. Ninety-eight percent of the participants (95% CI = 96–100%) had questioned whether specific life support interventions were appropriate for patients who appeared to have a terminal disease. Twenty-six percent (95% CI = 20–32%) reported to have used their own judgment during the past two years to withhold or end resuscitation in a patient who appeared to have a terminal disease. Significant discrepancies between the importance in practice and the level of preparation during training for the four EOL situations included: (1) understanding ADs (75% very important vs. 40% well prepared; difference 35%: 95% CI = 26–43%); (2) knowing when to honor written ADs (90% very important vs. 59% well-prepared; difference 31%: 95% CI = 23–38%); and (3) verbal ADs (75% very important vs. 54% well-prepared, difference 21%: 95% CI = 12–29%); and (4) communicating death to family or friends (79% very important vs. 48% well prepared, difference 31%: 95% CI = 23–39%). Paramedics' preparation in EOL skills was significantly lower than that for clinical skills such as endotracheal intubation or defibrillation.

There is a need to include more training in EOL care into prehospital training curricula, including how to verify and apply ADs, when to withhold treatments, and how to discuss death with victims' family or friends.