EPA_2200015_01 1..2

Introduction: Population aging is an important global public health issue. Due to the lack of effective treatment for dementia, prevention, and early identification are essential. Objectives: This study aimed to explore predicting cognitive impairment early using a large population-based longitudinal survey of elderly Chinese people, and to identify reversible factors which may help slow the rate of decline in cognitive function over 3 years. Methods: We included 12,280 elderly people from four waves of the Chinese Longitudinal Healthy Longevity Survey (CLHLS). The Chinese version of the Mini-Mental State Examination (MMSE) was used to examine cognitive function. Five machine learning algorithms and a neural network model were trained on data split 2/3 for training and 1/3 testing using 3-fold cross-validation. The model performance was measured by area-under-curve (AUC), sensitivity, and specificity. Logistic regression was used to assess the association of life behavior and its change with cognitive impairment after 3 years. Results: The best-performing model used 9 features. XGBoost was found to be the best performing algorithm. The AUC was 0.823 with a sensitivity of 0.763 and a specificity of 0.725. Playing more Mahjong or cards (OR = 0.49, 95% CI: 0.38-0.64), doing more garden works (OR = 0.54, 95% CI: 0.43-0.68), watching TV or listening to the radio more (OR = 0.67, 95% CI: 0.59-0.77) were associated with decreased risk of cognitive impairment after 3 years. Conclusions: Risk-predictive algorithms may serve as a valuable tool to support assessing the risk of cognitive impairment in pre- ventive health care as well as identifying modifiable risk factors to slow the rate of cognitive decline. Disclosure: No significant relationships. Keywords: cognitive impairment; risk factor; machine learning; intervention Perinatal Mental Health O0104 The role of subclinical depressive symptomatology during the prenatal period in cortisol rhythm alterations and postpartum depression risk Á. Castro Quinta s 1 , 2 * , M. Daura-Corra l 1 , E. Eixarch 3 , F. Crisp i 3 , L. De La Fuente Toma s 4 , M. Rocavert Barranc o 1 , A. Miguel Valer o 1 , L. Marques Feix a 1 , 2 , H. Palma Gudie l 2 , M.P. Garcia-Portill a 2 , 4 and L. Fañana s 1 , 2 1 University of Barcelona, Evolutionary Biology, Ecology And Environmental Sciences, Barcelona, Spain; 2 Centre for Biomedical Research Network on Mental Health (CIBERSAM), Instituto De Salut Carlos Iii, Madrid, Spain; 3 Maternitat Hospital Clinic Barcelona, Bcnatal Fetal Medicine Research Center, Barcelona, Spain and 4 University of Oviedo, Department Of Psychiatry, Oviedo, Spain *Corresponding author. doi: Introduction: Cortisol, the hormonal endpoint of Hypothalamic Pituitary Adrenal (HPA) axis, coordinates the body response in front of daily stressful situations. Disturbances in cortisol circadian rhythm have been implicated in the pathophysiology of depression and neurodevelopment lasting consequences. Although pregnancy entails a progressively increase in cortisol levels, the consequences of subclinical depression traits during pregnancy in cortisol circa- dian rhythm remains unclear. Objectives: To analyze the impact of prenatal subclinical depressive symptomatology in cortisol circadian rhythm through pregnancy and its relevance for postpartum depression risk. Methods: A cohort of 112 healthy pregnant women (Mean age SD = 32.32 4.37) of the general population was followed throughout their first pregnancy and first two months of postpartum period. Diurnal salivary cortisol curve (four measures) was obtained for every trimester; the Area Under the Curve with respect to the ground (AUCg) and with respect to the increase (AUCi) were used as measures of basal HPA axis functioning. Depressive symptom- atology was assessed every pregnancy trimester and postpartum period following EPDS criteria. All the analyses were adjusted for maternal age, weight, ethnicity and socioeconomic status and sam- ple collection ’ s time. Results: Prenatal subclinical depressive symptomatology (EPDS>10) was associated with a blunted cortisol rhythm during first trimester (F = 3.913,p = .011) but not during second (F = 2.629, p = 056) or third trimesters (F = .411,p = .724). Furthermore, a logis- tic regressionmodel showed a positive association between Prenatal subclinical depressive symptomatology and the risk of postpartum depression ( χ 2 = 13.8, p<.001,OR = 9.6; 95%CI 2.5 – 35.5). Conclusions: Women with subclinical depressive symptomatology in early pregnancy had alterations in cortisol circadian rhythmicity and a higher risk of postpartum depression. Disclosure: No significant relationships. O0105 The impact of maternal SARS-COV-2 infection in early stages of newborn neurodevelopment: preliminary results in a multicenter Spanish study Á. Castro Quinta s 1 , 2 * , N. San Martí n 3 , I. De Las Cueva s 4 , E. Eixarc h 5 , M. Daura-Corra l 3 , M. Lope z 5 , L. De La Fuente Toma s 6 , M.P. Garcia-Portill a 1 , L. Fañana s 1 , 2 and R. Ayesa-Arriol a 1 , 7 1 Centre for Biomedical Research Network on Mental Health (CIBERSAM), Instituto De Salut Carlos Iii, Madrid, Spain; 2 University of Barcelona, Evolutionary Biology, Ecology And Environmental Sciences, Barcelona, Spain; 3 University of Barcelona, Department Of Evolutive Biology, Ecology And Ambiental Sciences, Barcelona, Spain; 4 Universitary Hospital Marques de Valdecilla, Neonatology, Santander, Spain; 5 Maternitat Hospital Clinic Barcelona, Bcnatal Fetal Medicine Research Center, Barcelona, Spain; 6 University of Oviedo, Department Of Psychiatry, Oviedo, Spain and 7 Valdecilla Biomedical Research Institute, Psychiatry, Santander, Spain *Corresponding author. doi: Introduction: The consequences for the COVID-19 pandemic in the newborns of affected mothers remains unknown. Previous clinical experiences with other infections during pregnancy lead to considered pregnant women and their offspring especially vul- nerable for SARS-COV-2. That is, the underlying physiopatholo- gical changes caused by the infection (e.g. storm of cytokines, micro-coagulation in placenta or vertical transmission) could clearly compromise fetal neurodevelopment. S110 Oral Communication EPA 2022 provisional catalogue of accepted congress abstracts