Sensitivity analysis for unmeasured confounders using an electronic spreadsheet

In studies assessing the effects of a given exposure variable and a specific outcome of interest, confusion may arise from the mistaken impression that the exposure variable is producing the outcome of interest, when in fact the observed effect is due to an existing confounder. However, quantitative techniques are rarely used to determine the potential influence of unmeasured confounders. Sensitivity analysis is a statistical technique that allows to quantitatively measuring the impact of an unmeasured confounding variable on the association of interest that is being assessed. The purpose of this study was to make it feasible to apply two sensitivity analysis methods available in the literature, developed by Rosenbaum and Greenland, using an electronic spreadsheet. Thus, it can be easier for researchers to include this quantitative tool in the set of procedures that have been commonly used in the stage of result validation.

Risk factors associated with the death of patients hospitalized for juvenile systemic lupus erythematosus

We assessed the risk factors associated with death in patients hospitalized for juvenile systemic lupus erythematosus (JSLE) and evaluated the autopsy reports. A total of 57,159 hospitalizations occurred in our institution from 1994 to 2003, 169 of them involving 71 patients with JSLE. The most recent hospitalization of these patients was evaluated. Patients were divided into two groups based on mortality during hospitalization: those who survived (N = 53) and those who died (N = 18). The main causes of hospitalization were JSLE activity associated with infection in 52% and isolated JSLE activity in 44%. Univariate analysis showed that a greater risk of death was due to severe sepsis (OR = 17.8, CI = 4.5-70.9), systemic lupus erythematosus disease activity index (SLEDAI) ³8 (OR = 7.6, CI = 1.1-53.8), general infections (OR = 6.1, CI = 1.5-25), fungal infections (OR = 5.4, CI = 3.2-9), acute renal failure (OR = 5.1, CI = 2.5-10.4), acute thrombocytopenia (OR = 3.9, CI = 1.9-8.4), and bacterial infections (OR = 2.3, CI = 1.2-7.5). Stratified analysis showed that severe sepsis and SLEDAI ³8 were not confounder variables. In the multivariate analysis, logistic regression showed that the only independent variable in death prediction was severe sepsis (OR = 98, CI = 16.3-586.2). Discordance between clinical diagnosis and autopsy was observed in 6/10 cases. Mortality of hospitalized JSLE patients was associated with severe sepsis. Autopsy was important to determine events not detected or doubtful in dead patients and should always be requested.