Valentin Rousson

Executive function (EF) impairments are prevalent in survivors of neonatal critical illness such as children born very preterm (VPT) or with complex congenital heart disease (cCHD). This paper aimed to describe EF profiles in school-aged children born VPT or with cCHD and in typically developing peers, to identify child-specific and family-environmental factors associated with these profiles and to explore links to everyday-life outcomes. Data from eight EF tests assessing working memory, inhibition, cognitive flexibility, switching, and planning in n = 529 children aged between 7 and 16 years was subjected into a latent profile analysis. Three EF profiles were identified: The "favorable" profile was defined by mean scores in the normal range (z-scores ≤ -0.5 below the norm; n = 263, 49.7%). The "at-risk" profile's mean scores were 0.5 to 1 SD below the norm (n = 236, 44.6%). The "impaired" profile's mean scores were >1 SD below the norm (n = 30, 5.8%). Children of the two clinical groups were more likely to fall into the at-risk or impaired profile. Irrespective of clinical group, lower socioeconomic status, slower processing speed and poorer fine motor skills were associated with a more impaired profile. In turn, falling into the at-risk or impaired profile was associated with a higher need for educational support, poorer everyday-life EFs and more behavioral problems. Children in the impaired profile reported lower psychosocial quality of life. This study provides evidence for an increased risk of survivors of neonatal critical illness to present with an impaired EF profile that translates into everyday-life difficulties. Long-term monitoring is needed to promote optimal outcome.

The concept of (potential) years of life lost is a measure of premature mortality that can be used to compare the impacts of different specific causes of death. However, interpreting a given number of years of life lost at face value is more problematic because of the lack of a sensible reference value. In this paper, we propose three denominators to divide an excess years of life lost, thus obtaining three indicators, called average life lost , increase of life lost , and proportion of life lost , which should facilitate interpretation and comparisons. We study the links between these three indicators and classical mortality indicators, such as life expectancy and standardized mortality rate, introduce the concept of weighted standardized mortality rate , and calculate them in 30 countries to assess the impact of COVID‐19 on mortality in the year 2020. Using any of the three indicators, a significant excess loss is found for both genders in 18 of the 30 countries.

Pearson's correlation coefficient is a popular statistical measure to summarize the strength of association between two continuous variables. It is usually interpreted via its square as percentage of variance of one variable predicted by the other in a linear regression model. It can be generalized for multiple regression via the coefficient of determination, which is not straightforward to interpret in terms of prediction accuracy. In this paper, we propose to assess the prediction accuracy of a linear model via the prediction interval reduction (PIR) by comparing the width of the prediction interval derived from this model with the width of the prediction interval obtained without this model. At the population level, PIR is one-to-one related to the correlation and the coefficient of determination. In particular, a correlation of 0.5 corresponds to a PIR of only 13%. It is also the one's complement of the coefficient of alienation introduced at the beginning of last century. We argue that PIR is easily interpretable and useful to keep in mind how difficult it is to make accurate individual predictions, an important message in the era of precision medicine and artificial intelligence. Different estimates of PIR are compared in the context of a linear model and an extension of the PIR concept to non-linear models is outlined.

Purpose/Background: Trazodone is indicated for the treatment of major depressive disorder, but more frequently prescribed off-label at lower doses for insomnia in women of childbearing age. The aim of this study was to assess the risks linked to trazodone exposure during pregnancy for which limited safety data are available. Methods/Procedures: This multicenter, observational prospective cohort study compared pregnancy outcomes in women exposed to trazodone in early pregnancy against those in a reference group of women exposed to a selective serotonin reuptake inhibitors (SSRIs) between 1996 and 2021. Findings/Results: The sample included 221 trazodone and 869 SSRI-exposed pregnancies. Exposure to trazodone in the first trimester was not associated with a significant difference in the risk of major congenital anomalies (trazodone [1/169, 0.6%]; SSRI [19/730, 2.6%]; adjusted odds ratio, 0.2; 95% confidence interval, 0.03–1.77). The cumulative incidences of live birth were 61% and 73% in the trazodone and reference group, respectively (25% vs 18% for pregnancy loss and 14% vs 10% for pregnancy termination). Trazodone exposure was not associated with a significantly increased risk of pregnancy termination and pregnancy loss. The rate of small for gestational age infants did not differ between the groups. Implications/Conclusions: This study did not reveal a significant difference in the risk of major congenital anomalies after first trimester exposure to trazodone, compared with SSRI exposure. Although this study is the largest so far, these results call for confirmation through further studies.

Objective: To quantify the (direct and indirect) impacts of the COVID-19 pandemic on mortality for actual populations of persons living in 12 European countries in 2020.Method: Based on demographic and mortality data, as well as remaining life expectancies found in the Human Mortality Database, we calculated a “population life loss” in 2020 for men and women living in Belgium, Croatia, Denmark, Finland, Hungary, Lithuania, Luxembourg, Norway, Portugal, Spain, Sweden, and Switzerland. This quantity was obtained by dividing the total number of years lost in 2020 (estimated from all-cause mortality data and attributed directly or indirectly to COVID-19) by the size of the population.Results: A significant population life loss was found in 8 countries in 2020, with men losing an average of 8.7, 5.0, 4.4, 4.0, 3.7, 3.4, 3.1, and 2.7 days in Lithuania, Spain, Belgium, Hungary, Croatia, Portugal, Switzerland, and Sweden, respectively. For women, this loss was 5.5, 4.3, 3.7, 3.7, 3.1, 2.4, 1.6, and 1.4 days, respectively. No significant losses were found in Finland, Luxembourg, Denmark and Norway. Life loss was highly dependent on age, reaching 40 days at the age of 90 in some countries, while only a few significant losses occurred under the age of 60. Even in countries with a significant population life loss in 2020, it was on average about 30 times lower than in 1918, at the time of the Spanish flu.Conclusions: Our results based on the concept of population life loss were consistent with those based on the classical concept of life expectancy, confirming the significant impact of COVID-19 on mortality in 8 European countries in 2020. However, while life expectancy losses were typically counted in months or years, population life losses could be counted in days, a potentially useful piece of information from a public health perspective.

Background: This meta-analysis summarized longitudinal findings pertaining to exhaustion’s predictors. In so doing, our aim was ultimately to identify target factors for the prevention of burnout. Methods: We searched for studies that (a) examined predictors of exhaustion longitudinally and (b) reported correlation coefficients as an effect estimate. We conducted our literature search in three databases: MEDLINE, PsycINFO, and Embase. We focused on studies published between January 1990 and November 2020. Predictors were grouped into families, subfamilies, and subgroups. A meta-analysis of z-transformed correlation coefficients (rho) was performed. The results were scrutinized in relation to studies’ follow-up length. Results: We included 65 studies assessing 242 predictors of different types captured across different occupations. Our findings highlighted mostly weak associations (rho < 0.30). For six predictors—Job control, Job resources, Interactions at work, Communication and leadership, Job attitudes, and Work-family interface—longer length of follow-up involved weaker associations with exhaustion. The quality of the evidence available was generally low. Conclusions: The evidence available does not point to clear target factors for preventing burnout. The decrease in associations as the follow-up length increases may suggest a relatively short latency period, followed by recovery. Higher-quality cohorts should be conducted to better understand the etiology and course of burnout.

The 5A score predicts in-hospital mortality of patients suffering from accidental hypothermia, including those not in cardiac arrest. The HOPE score was specifically developed to predict survival for the subgroup of hypothermic patients in cardiac considered for extracorporeal life support rewarming. The C-statistic in the external validation study of the HOPE score was 0.825 (95% CI: 0.753–0.897), confirming its excellent discrimination. In addition, its good calibration allows for a reliable interpretation of the corresponding survival probability after rewarming. The HOPE score should be used for predicting outcome and selecting hypothermic patients in cardiac arrest for rewarming.