Concurrence of water and food insecurities, 25 low- and middle-income countries.

Objective: To investigate how water and food insecurity were associated in nationally representative samples of individuals from 25 low- and middle-income countries. Methods: We used data from the 2020 World Gallup Poll in which the Individual Water Insecurity Experiences Scale and the Food Insecurity Experience Scale had been administered to 31 755 respondents. These scales measure insecurity experiences in the previous 12 months. We classified individuals as water insecure if their score was ≥ 12 and food insecure if the Rasch probability parameter was ≥ 0.5. For estimating the proportions, we used projection weights. We estimated the relationships between binary and continuous measures of water insecurity and food insecurity for individuals within each country and region using multivariable logistic and linear regression models, adjusting for key socioeconomic characteristics including income, gender, age and education. Findings: Among the 18.3% of respondents who experienced water insecurity, 66.8% also experienced food insecurity. The likelihood of experiencing moderate-to-severe food insecurity was higher among respondents also experiencing water insecurity (adjusted odds ratio, aOR: 2.69; 95% confidence interval, CI: 2.43 to 2.98). Similar odds were found in Asia (aOR: 2.95; 95% CI: 2.04 to 4.25), Latin America (aOR: 2.17; 95% CI: 1.62 to 2.89), North Africa (aOR: 2.92; 95% CI: 2.17 to 3.93) and sub-Saharan Africa (aOR: 2.71; 95% CI: 2.40 to 3.06). Conclusion: Our results suggest that water insecurity should be considered when developing food and nutrition policies and interventions. However, more research is needed to understand the paths between these insecurities.

Longitudinal quantile regression in the presence of informative dropout through longitudinal-survival joint modeling.

We propose a joint model for a time-to-event outcome and a quantile of a continuous response repeatedly measured over time. The quantile and survival processes are associated via shared latent and manifest variables. Our joint model provides a flexible approach to handle informative dropout in quantile regression. A Monte Carlo expectation maximization strategy based on importance sampling is proposed, which is directly applicable under any distributional assumption for the longitudinal outcome and random effects. We consider both parametric and nonparametric assumptions for the baseline hazard. We illustrate through a simulation study and an application to an original data set about dilated cardiomyopathies.

Robust estimation for the Cox regression model based on trimming.

We propose a robust Cox regression model with outliers. The model is fit by trimming the smallest contributions to the partial likelihood. To do so, we implement a Metropolis-type maximization routine, and show its convergence to a global optimum. We discuss global robustness properties of the approach, which is illustrated and compared through simulations. We finally fit the model on an original and on a benchmark data set.