Studies conducted during the COVID-19 pandemic found high occurrence of suicidal thoughts and behaviours (STBs) among healthcare workers (HCWs). The current study aimed to (1) develop a machine learning-based prediction model for future STBs using data from a large prospective cohort of Spanish HCWs and (2) identify the most important variables in terms of contribution to the model’s predictive accuracy.

This is a prospective, multicentre cohort study of Spanish HCWs active during the COVID-19 pandemic. A total of 8,996 HCWs participated in the web-based baseline survey (May–July 2020) and 4,809 in the 4-month follow-up survey. A total of 219 predictor variables were derived from the baseline survey. The outcome variable was any STB at the 4-month follow-up. Variable selection was done using an L1 regularized linear Support Vector Classifier (SVC). A random forest model with 5-fold cross-validation was developed, in which the Synthetic Minority Oversampling Technique (SMOTE) and undersampling of the majority class balancing techniques were tested. The model was evaluated by the area under the Receiver Operating Characteristic (AUROC) curve and the area under the precision–recall curve. Shapley’s additive explanatory values (SHAP values) were used to evaluate the overall contribution of each variable to the prediction of future STBs. Results were obtained separately by gender.

The prevalence of STBs in HCWs at the 4-month follow-up was 7.9% (women = 7.8%, men = 8.2%). Thirty-four variables were selected by the L1 regularized linear SVC. The best results were obtained without data balancing techniques: AUROC = 0.87 (0.86 for women and 0.87 for men) and area under the precision–recall curve = 0.50 (0.55 for women and 0.45 for men). Based on SHAP values, the most important baseline predictors for any STB at the 4-month follow-up were the presence of passive suicidal ideation, the number of days in the past 30 days with passive or active suicidal ideation, the number of days in the past 30 days with binge eating episodes, the number of panic attacks (women only) and the frequency of intrusive thoughts (men only).

Machine learning-based prediction models for STBs in HCWs during the COVID-19 pandemic trained on web-based survey data present high discrimination and classification capacity. Future clinical implementations of this model could enable the early detection of HCWs at the highest risk for developing adverse mental health outcomes.

NCT04556565

Primary health care (PHC) supported long-term care facilities (LTCFs) in attending COVID-19 patients. The aim of this study is to describe the role of PHC in LTCFs in Europe during the early phase of the pandemic.

Retrospective descriptive study from 30 European countries using data from September 2020 collected with an ad hoc semi-structured questionnaire. Related variables are SARS-CoV-2 testing, contact tracing, follow-up, additional testing, and patient care.

Twenty-six out of the 30 European countries had PHC involvement in LTCFs during the COVID-19 pandemic. PHC participated in initial medical care in 22 countries, while, in 15, PHC was responsible for SARS-CoV-2 test along with other institutions. Supervision of individuals in isolation was carried out mostly by LTCF staff, but physical examination or symptom’s follow-up was performed mainly by PHC.

PHC has participated in COVID-19 pandemic assistance in LTCFs in coordination with LTCF staff, public health officers, and hospitals.

Evidence suggests that the rate of glucose release following consumption of carbohydrate-containing foods, defined as the glycaemic index (GI), is inversely associated with cognitive function. To date, most of the evidence stems from either single-meal studies or highly heterogeneous cohort studies. We aimed to study the prospective associations of diet GI at age 53 years with outcomes of verbal memory and letter search tests at age 69 years and rate of decline between 53 and 69 years.

Longitudinal population-based birth cohort study.

MRC National Survey for Health and Development.

Cohort members (n 1252).

Using multivariable linear and logistic regression, adjusted for potential confounders, associations of higher-GI diet with lower verbal memory, lower letter search speed and lower number of hits in a letter search test were attenuated after adjustments for cognitive ability at age 15 years, educational attainment, further training and occupational social class. No association was observed between diet GI at 53 years and letter search accuracy or speed–accuracy trade-off at 69 years, or between diet GI at 53 years and rate of decline between 53 and 69 years in any cognitive measure.

Diet GI does not appear to predict cognitive function or decline, which was mainly explained by childhood cognitive ability, education and occupational social class. Our findings confirm the need for further research on the association between diet and cognition from a life-course perspective.

To assess the prevalence and sociodemographic determinants of breast-feeding (BF) and exclusive breast-feeding (EBF) in Cyprus up to the sixth month.

Cross-sectional and longitudinal descriptive study. BF and EBF were estimated based on mothers’ self-reported BF status in line with Step 7 of the WHO/UNICEF Baby-Friendly Hospital Initiative questionnaire and based on 24 h recall.

Maternity wards in all public hospitals and twenty-nine (of thirty-five) private maternity clinics nationwide.

Consecutive sample of 586 mothers recruited within 48 h from birth, followed up by telephone interview at the first, fourth and sixth month.

Although 84·3 % of mothers initiated BF before discharge, prevalence of BF at the sixth month was 32·4 %, with the highest reduction observed between the first and fourth months. Prevalence of EBF at 48 h was 18·8 % and fell gradually to 5·0 % at the sixth month. Mothers with higher educational attainment or higher family income were more likely to breast-feed until the sixth month. In terms of EBF, an association was observed only with education, which persisted until the sixth month. Other than social gradient, mode of delivery was the strongest determinant of BF initiation, exclusivity and continuation. Mothers who gave birth vaginally were three to four times more likely to initiate BF (OR=3·1; 95 % CI 1·7, 5·4) and EBF (OR=4·3; 95 % CI 2·7, 6·8).

The low prevalence of BF and EBF in Cyprus, together with the fact that caesarean section rates are currently among the highest in Europe, suggest the need for further research to understand this multidimensional phenomenon and for interdisciplinary policy action to protect, promote and support BF.