Health as an independent predictor of the 2017 French presidential voting behaviour: a cross-sectional analysis.

BACKGROUND: It has been suggested that poor health has influenced vote for Brexit and the US presidential election. No such research has been published regarding the 2017 French presidential election. METHODS: We performed a cross-sectional analysis using a comprehensive set of socioeconomic and health indicators, to be compared with voting outcome at the first round of the 2017 French presidential election. The 95 French departments were selected as the unit of analysis. Data were obtained from publicly available sources. The linear model was used for both univariate and multivariate analysis to investigate the relation between voting patterns and predictors. Sensitivity analyses were done using the elastic-net regularisation. RESULTS: Emmanuel Macron and Marine Le Pen arrived ahead. When projected on the first factorial plane (~ 60% of the total inertia), Emmanuel Macron and Marine Le Pen tended to be in opposite directions regarding both socioeconomic and health factors. In the respective multivariate analyses of the two candidates, both socio-economic and health variables were significantly associated with voting patterns, with wealthier and healthier departments more likely to vote for Emmanuel Macron, and opposite departments more likely to vote for Marine Le Pen. Mortality (p = 0.03), severe chronic conditions (p = 0.014), and diabetes mellitus (p < 0.0001) were among the strongest predictors of voting pattern for Marine Le Pen. Sensitivity analyses did not substantially change those findings. CONCLUSIONS: We found that areas associated with poorer health status were significantly more likely to vote for the far-right candidate at the French presidential election, even after adjustment on socioeconomic criteria.

Readmissions After Surgery: A French Nationwide Cross-Sectional Study of 1,686,602 Procedures Performed in 2010.

BACKGROUND: Surgical readmissions have been extensively studied in North America, but very few data from other countries are available. We aimed to describe surgical readmissions in France and to assess their association with hospital status, surgical volume, and day surgery activity. METHODS: We performed a cross-sectional study encompassing all 1270 French hospitals, except for military hospitals and hospitals with very small volume. Data were retrieved from the national database regarding all patients undergoing surgery between January 1, 2010 and November 30, 2010. The main outcome measure was 30-day readmission rate. Association with hospital status, surgical volume, and the level of day surgery were assessed. Risk adjustment was performed based upon administrative categories. RESULTS: After exclusion of deaths and hospital transfers, there were 1,686,602 patients in the study cohort. Thirty-day readmission rate was 5.9 %. Distribution was skewed, with 21.5 % of procedures accounting for 33.5 % of all 30-day readmissions. Early readmissions (≤3 days) were associated with higher mortality as compared to those occurring later (>7 days) (3.2 vs. 2.6 %; p < 0.0001). After multivariate analysis, University hospitals were shown to be affected by a significantly greater risk of 30-day readmission as compared to private hospitals (odds ratio 1.46 [95 % CI 1.42-1.5]). Other independent factors were as follows: male gender, longer initial hospital stay, and comorbidities. CONCLUSIONS: Surgical 30-day readmission rate was low, with early readmissions being associated with higher mortality. Conversely to prior research, University hospitals were shown to be associated with significantly higher risk of 30-day readmissions, even after risk adjustment.