The purpose of this study was to associate the anti-inflammatory dietary diversity and depressive symptoms among a nationwide sample of Chinese older adults. We used data from the 2018 wave of Chinese longitudinal healthy longevity survey (CLHLS). We assessed depressive symptoms using the 10 items of the center for epidemiologic studies depression scale (CES-D-10). Based on the dietary diversity index (DDI) generated by previous studies, we construct two novel indicators: the protein-enriched dietary diversity index diet (PEDDI) and the anti-inflammatory dietary diversity index diet (AIDDI). We used multivariate logistic models to evaluate the associations of DDI, PEDDI, and AIDDI with depressive symptoms, statistically adjusted for a range of potential confounders. A total of 12,192 participants (mean age 83.6 years) were included in the analysis. We found that participants with a higher score of DDI (OR = 0.91, 95% CI: 0.89–0.92) and PEDDI (OR = 0.91, 95% CI: 0.88–0.93) showed lower odds of having depressive symptoms, while the association between AIDDI and depressive symptoms was more marked (OR = 0.80, 95% CI: 0.78–0.83). The associations remained in subgroup analyses and sensitivity analyses. The results indicate that intaking diversified diet, particularly anti-inflammatory foods, may be associated with a lower risk of depressive symptoms. The findings of this study, if confirmed as causal, provide evidence that an intervention of adopting an anti-inflammatory diversified diet may reduce the burden of depression among older adults.

We aim to develop and validate an interpretable machine-learning model that can provide critical information for the clinical treatment of critically ill patients with acute myocardial infarction (AMI). All data was extracted from the multi-centre database (training and internal validation cohorts: MIMIC-III/-IV, external validation cohort: eICU). After comparing different machine-learning models and several unbalanced data processing methods, the model with the best performance was selected. Lasso regression was used to build a compact model. Seven evaluation methods, PR, and ROC curves were used to assess the model. The SHapley Additive exPlanations (SHAP) values were calculated to evaluate the feature's importance. The SHAP plots were adopted to explain and interpret the results. A web-based tool was developed to help application. A total of 12,170 critically ill patients with AMI were included. The balance random forest (BRF) model had the best performance in predicting in-hospital mortality. The compact model did not differ from the full variable model in performance (AUC: 0.891 vs 0.885, P = 0.06). The external validation results also demonstrated the stability of the model (AUC: 0.784). All SHAP plots have shown the contribution ranking of all variables in the model, the relationship trend between variables and outcomes, and the interaction mode between variables. A web-based tool is constructed that can provide individualized risk stratification probabilities (https://github.com/huangtao36/BRF-web-tool) . We built the BRF model and the web-based tool by the model algorithm. The model effect has been verified externally. The tool can help clinical decision-making.