Representation, propagation, and decision issues in risk analysis under incomplete probabilistic information.

This article tries to clarify the potential role to be played by uncertainty theories such as imprecise probabilities, random sets, and possibility theory in the risk analysis process. Instead of opposing an objective bounding analysis, where only statistically founded probability distributions are taken into account, to the full-fledged probabilistic approach, exploiting expert subjective judgment, we advocate the idea that both analyses are useful and should be articulated with one another. Moreover, the idea that risk analysis under incomplete information is purely objective is misconceived. The use of uncertainty theories cannot be reduced to a choice between probability distributions and intervals. Indeed, they offer representation tools that are more expressive than each of the latter approaches and can capture expert judgments while being faithful to their limited precision. Consequences of this thesis are examined for uncertainty elicitation, propagation, and at the decision-making step.

Joint propagation of variability and imprecision in assessing the risk of groundwater contamination.

Estimating risks of groundwater contamination often require schemes for representing and propagating uncertainties relative to model input parameters. The most popular method is the Monte Carlo method whereby cumulative probability distributions are randomly sampled in an iterative fashion. The shortcoming of the approach, however, arises when probability distributions are arbitrarily selected in situations where available information is incomplete or imprecise. In such situations, alternative modes of information representation can be used, for example the nested intervals known as "possibility distributions". In practical situations of groundwater risk assessment, it is common that certain model parameters may be represented by single probability distributions (representing variability) because there are data to justify these distributions, while others are more faithfully represented by possibility distributions (representing imprecision) due to the partial nature of available information. This paper applies two recent methods, designed for the joint-propagation of variability and imprecision, to a groundwater contamination risk assessment. Results of the joint-propagation methods are compared to those obtained using both interval analysis and the Monte Carlo method with a hypothesis of stochastic independence between model parameters. The two joint-propagation methods provide results in the form of families of cumulative distributions of the probability of exceeding a certain value of groundwater concentration. These families are delimited by an upper cumulative distribution and a lower distribution respectively called Plausibility and Belief after evidence theory. Slight differences between the results of the two joint-propagation methods are explained by the different assumptions regarding parameter dependencies. Results highlight the point that non-conservative results may be obtained if single cumulative probability distributions are arbitrarily selected for model parameters in the face of imprecise information and the Monte Carlo method is used under the assumption of stochastic independence. The proposed joint-propagation methods provide upper and lower bounds for the probability of exceeding a tolerance threshold. As this may seem impractical in a risk-management context, it is proposed to introduce "a-posteriori subjectivity" (as opposed to the "a-priori subjectivity" introduced by the arbitrary selection of single probability distributions) by defining a single indicator of evidence as a weighted average of Plausibility and Belief, with weights to be defined according to the specific context.

Effect of gingival and dental plaque antiseptic decontamination on nosocomial infections acquired in the intensive care unit: a double-blind placebo-controlled multicenter study.

OBJECTIVE: To document the effect of gingival and dental plaque antiseptic decontamination on the rate of nosocomial bacteremias and respiratory infections acquired in the intensive care unit (ICU). DESIGN: Prospective, multicenter, double-blind, placebo-controlled efficacy study. SETTING: Six ICUs: three in university hospitals and three in general hospitals. PATIENTS: A total of 228 nonedentulous patients requiring endotracheal intubation and mechanical ventilation, with an anticipated length of stay > or =5 days. INTERVENTIONS: Antiseptic decontamination of gingival and dental plaque with a 0.2% chlorhexidine gel or a placebo gel, three times a day, during the entire ICU stay. MEASUREMENTS AND MAIN RESULTS: Demographic and clinical characteristics, organ function data (Logistic Organ Dysfunction score), severity of condition (Simplified Acute Physiologic Score), and dental plaque status were assessed at baseline and until 28 days. Bacteriologic sampling of dental plaque and saliva was done every 5 days, and blood, tracheal aspirate, and bronchoalveolar lavage cultures were performed when appropriate. The primary efficacy end point was the incidence of bacteremia, bronchitis, and ventilator-associated pneumonia, expressed as a percentage and per 1000 ICU days. All baseline characteristics were similar between the treated and the placebo groups. The incidence of nosocomial infections was 17.5% (13.2 per 1000 ICU days) in the placebo group and 18.4% (13.3 per 1000 ICU days) in the plaque antiseptic decontamination group (not significant). No difference was observed in the incidence of ventilator-associated pneumonia per ventilator or intubation days, mortality, length of stay, and care loads (secondary end points). On day 10, the number of positive dental plaque cultures was significantly lower in the treated group (29% vs. 66%; p < .05). Highly resistant Pseudomonas, Acinetobacter, and Enterobacter species identified in late-onset ventilator-associated pneumonia and previously cultured from dental plaque were not eradicated by the antiseptic decontamination. No side effect was reported. CONCLUSIONS: Gingival and dental plaque antiseptic decontamination significantly decreased the oropharyngeal colonization by aerobic pathogens in ventilated patients. However, its efficacy was insufficient to reduce the incidence of respiratory infections due to multiresistant bacteria.

Family participation in care to the critically ill: opinions of families and staff.

BACKGROUND: Allowing family members to participate in the care of patients in intensive care units (ICUs) may improve the quality of their experience. No previous study has investigated opinions about family participation in ICUs. METHODS: Prospective multicenter survey in 78 ICUs (1,184 beds) in France involving 2,754 ICU caregivers and 544 family members of 357 consecutive patients. We determined opinions and experience about family participation in care; comprehension (of diagnosis, prognosis, and treatment) and satisfaction (Critical Care Family Needs Inventory) scores to assess the effectiveness of information to families and the Hospital Anxiety and Depression score for family members. RESULTS: Among caregivers 88.2% felt that participation in care should be offered to families. Only 33.4% of family members wanted to participate in care. Independent predictors of this desire fell into three groups: patient-related (SAPS II at ICU admission, OR 0.984); ICU stay length, OR 1.021), family-related (family member age, OR 0.97/year); family not of European descent, OR 0.294); previous ICU experience in the family, OR 1.59), and those related to emotional burden and effectiveness of information provided to family members (symptoms of depression in family members, OR 1.58); more time wanted for information, OR 1.06). CONCLUSIONS: Most ICU caregivers are willing to invite family members to participate in patient care, but most family members would decline.