Pier Alda Ferrari

ICU patients must be kept conscious, calm, and cooperative even during the critical phases of illness. Enteral administration of sedative drugs might avoid over sedation, and would be as adequate as intravenous administration in patients who are awake, with fewer side effects and lower costs. This study compares two sedation strategies, for early achievement and maintenance of the target light sedation. This was a multicenter, single-blind, randomized and controlled trial carried out in 12 Italian ICUs, involving patients with expected mechanical ventilation duration > 72 h at ICU admission and predicted mortality > 12% (Simplified Acute Physiology Score II > 32 points) during the first 24 h on ICU. Patients were randomly assigned to receive intravenous (midazolam, propofol) or enteral (hydroxyzine, lorazepam, and melatonin) sedation. The primary outcome was percentage of work shifts with the patient having an observed Richmond Agitation-Sedation Scale (RASS) = target RASS ±1. Secondary outcomes were feasibility, delirium-free and coma-free days, costs of drugs, length of ICU and hospital stay, and ICU, hospital, and one-year mortality. There were 348 patients enrolled. There were no differences in the primary outcome: enteral 89.8% (74.1–100), intravenous 94.4% (78–100), p = 0.20. Enteral-treated patients had more protocol violations: n = 81 (46.6%) vs 7 (4.2%), p < 0.01; more self-extubations: n = 14 (8.1%) vs 4 (2.4%), p = 0.03; a lighter sedative target (RASS = 0): 93% (71–100) vs 83% (61–100), p < 0.01; and lower total drug costs: 2.39 (0.75–9.78) vs 4.15 (1.20–20.19) €/day with mechanical ventilation (p = 0.01). Although enteral sedation of critically ill patients is cheaper and permits a lighter sedation target, it is not superior to intravenous sedation for reaching the RASS target. ClinicalTrials.gov, NCT01360346 . Registered on 25 March 2011.

A package for the stochastic simulation of discrete variables with assigned marginal distributions and correlation matrix is presented and discussed. The simulating mechanism relies upon the Gaussian copula, linking the discrete distributions together, and an iterative scheme recovering the correlation matrix for the copula that ensures the desired correlations among the discrete variables. Examples of its use are provided as well as three possible applications (related to probability, sampling, and inference), which illustrate the utility of the package as an efficient and easy-to-use tool both in statistical research and for didactic purposes.

This article focuses on a statistical tool for dependence analysis in scientific research. Starting from a recent index of concordance for a multiple linear regression model, a coefficient suitable in catching any monotonic dependence relationship between a dependent variable and an independent variable is derived and discussed. Given its interpretation in terms of monotonic dependence, it is called monotonic dependence coefficient (MDC). It is appropriate to all contexts where the dependent variable is quantitative (continuous or discrete) and the independent variable is at least of ordinal nature; tied data are also allowed. MDC's adequacy is validated through Monte Carlo simulations led by taking into account different scenarios of dependence. Finally, an application to real data is provided to stress MDC's capability of detecting dependence relationships between two variables, even if some pieces of information about original data are lost.