Risk assessment and standardized nosocomial infection rate in critically ill children.

OBJECTIVES: To develop and validate a pediatric nosocomial infection risk (PNIR) assessment model, and to compare the daily trends in risk factors between patients with nosocomial infection (cases) and without nosocomial infection (controls) in the pediatric intensive care unit (ICU). DESIGN: Prospective cohort. SETTING: A 16-bed pediatric ICU in an urban, university-affiliated, multidisciplinary, regional referral center. PATIENTS: Patients available for study included consecutive admissions to the unit between May 1, 1992, and April 30, 1993, and between May 9, 1995, and December 11, 1995. Patients from both data collection periods were pooled and randomly divided into training (70%) and validation (30%) samples. MEASUREMENTS AND MAIN RESULTS: In the logistic regression analysis using admission day data, three factors were shown to remain as independent risk factors. Invasive device use, parenteral nutrition, and the interaction between severity of illness-modified Pediatric Risk of Mortality III-24 score and postoperative care were associated with 2, 6, and 1.5 times the risk of developing nosocomial infection, respectively. This PNIR model performed well in both the training and validation samples as indicated by the goodness-of-fit test, which evaluated standardized nosocomial infection rates (observed vs. predicted nosocomial infection rates). The internal validity of the PNIR model was good. In trend analysis, severity of illness and invasive device use appear to have similar trend patterns, during the first week of pediatric ICU stay. There was no difference in any of these risk factors between cases and controls after 7 days of pediatric ICU stay. CONCLUSIONS: The PNIR assessment model incorporates intrinsic factors, such as patient severity of illness, and extrinsic factors contributing to the development of nosocomial infection in this high-risk population. The methodology using intrinsic and extrinsic factors to adjust for nosocomial infections should be taken into consideration when evaluating interhospital comparison of nosocomial infection rates, quality assessment, intervention strategies, and use of treatment modalities.

Prognostication and certainty in the pediatric intensive care unit.

OBJECTIVE: Prognostication is central to developing treatment plans and relaying information to patients, family members, and other health care providers. The degree of confidence or certainty that a health care provider has in his or her mortality risk assessment is also important, because a provider may deliver care differently depending on their assuredness in the assessment. We assessed the performance of nurse and physician mortality risk estimates with and without weighting the estimates with their respective degrees of certainty. METHODS: Subjective mortality risk estimates from critical care attendings (n = 5), critical care fellows (n = 9), pediatric residents (n = 34), and nurses (n = 52) were prospectively collected on at least 94% of 642 eligible, consecutive admissions to a tertiary pediatric intensive care unit (PICU). A measure of certainty (continuous scale from 0 to 5) accompanied each mortality estimate. Estimates were evaluated with 2 x 2 outcome probabilities, the kappa statistic, the area under the receiver operating characteristics curve, and the Hosmer and Lemeshow goodness-of-fit chi(2) statistic. The estimates were then reevaluated after weighting predictions by their respective degree of certainty. RESULTS: Overall, there was a significant difference in the predictive accuracy between groups. The mean mortality predictions from the attendings (6.09%) more closely approximated the true mortality rate (36 deaths, 5.61%) whereas fellows (7.87%), residents (10.00%), and nurses (16.29%) overestimated the mean overall PICU mortality. Attendings were more certain of their predictions (4.27) than the fellows (4.01), nurses (3.79), and residents (3.75). All groups discriminated well (area under receiver operating characteristics curve range, 0.86-0.93). Only PICU attendings and fellows did not significantly differ from ideal calibration (chi(2)). When mortality predictions were weighted with their respective certainties, their performance improved. CONCLUSIONS: The level of medical training correlated with the provider's ability to predict mortality risk. The higher the level of certainty associated with the mortality prediction, the more accurate the prediction; however, high levels of certainty did not guarantee accurate predictions. Measures of certainty should be considered when assessing the performance of mortality risk estimates or other subjective outcome predictions.

Risk factors for nosocomial infection in critically ill children: a prospective cohort study.

OBJECTIVE: To identify factors in pediatric intensive care unit (ICU) patients that are associated with an increased risk of nosocomial infections. DESIGN: A prospective, 1-yr cohort study. SETTING: A 16-bed pediatric ICU in a multidisciplinary, regional referral center. SUBJECTS: All patients admitted to the pediatric ICU. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The primary outcome variable was the development of nosocomial infection. Out of 945 consecutive admissions, 75 patients developed 96 nosocomial infections. The most frequent infection sites were the lower respiratory tract (35%), the bloodstream (21%), and the urinary tract (21%). The most common organisms isolated were Gram-negative bacteria (53%, Gram-positive bacteria (27%), and fungi (9%). Variables significantly associated with the development of nosocomial infections included age, weight, Pediatric Risk of Mortality (PRISM) score, device utilization ratio, antimicrobial therapy, histamine-2 (H2) receptor blocker use, immune status, parenteral nutrition, and length of stay. When combined in a multivariate logistic regression model, the significant variables were operative status, PRISM score, device utilization ratio, antimicrobial therapy, parenteral nutrition, and length of stay before the onset of infection. The area under the receiver operating characteristic curve was 0.868. At a probability of 0.15, the sensitivity was 66.67%, and the specificity was 87.82%. CONCLUSIONS: Patients at risk for developing nosocomial infection can be identified using a multivariate logistic regression model with a high degree of sensitivity and specificity. These data indicate that institutional nosocomial rates need to be adjusted for risk factors. This model could help target patients at high risk for developing nosocomial infections for preventive strategies.