Building a cancer risk and survival prediction model based on social determinants of health combined with machine learning: A NHANES 1999 to 2018 retrospective cohort study

Medicine, Год журнала: 2025, Номер 104(6), С. e41370 - e41370

Опубликована: Фев. 7, 2025

The occurrence and progression of cancer is a significant focus research worldwide, often accompanied by prolonged disease course. Concurrently, researchers have identified that social determinants health (SDOH) (employment status, family income poverty ratio, food security, education level, access to healthcare services, insurance, housing conditions, marital status) are associated with the many chronic diseases. However, there paucity examining influence SDOH on incidence risk survival survivors. aim this study was utilize as primary predictive factor, integrated machine learning models, forecast both prognostic survival. This grounded in data derived from National Health Nutrition Examination Survey dataset spanning 1999 2018. It employs methodologies including adaptive boosting, gradient boosting (GradientBoosting), random forest (RF), extreme light machine, support vector logistic regression develop models for predicting hyperparameters these models—specifically, number estimators (100–200), maximum tree depth (10), rate (0.01–0.2), regularization parameters—were optimized through grid search cross-validation, followed performance evaluation. Shapley Additive exPlanations plots were generated visualize each feature. RF best model (area under curve: 0.92, accuracy: 0.84). Age, non-Hispanic White, sex, status 4 most important characteristics model. gender, employment household income/poverty ratio features developed exhibited strong estimating time, identifying several factors significantly survival, thereby providing new evidence management. Despite promising findings, acknowledges certain limitations, omission survivor potential biases inherent dataset. Future warranted further validate using external datasets.

Язык: Английский

Assessing the Incremental Health Care Utilization and Expenditures Associated With Serious Psychological Distress by Living Arrangements Among Older Adults

Journal of Applied Gerontology, Год журнала: 2025, Номер unknown

Опубликована: Фев. 26, 2025

Older adults suffering from serious psychological distress (SPD) have higher health care utilization and expenditures. However, it is unknown whether living alone might further amplify differences in usage costs among older with SPD. Using pooled data 2007-2019 Medical Expenditure Panel Survey, we estimated the incremental expenditures associated SPD others. The results show that both arrangements, having increases emergency department visits, hospital admissions, prescription fill-ups, which leads to total of SPD-related between others were similar. Our findings call for targeted community programs go beyond eliminating physical loneliness, innovative coordination strategies prioritize mental preventative adults.

Язык: Английский

Acceptability of Vaginal Human Papillomavirus Self-Collection Among Colposcopy Clinic Attendees

Women s Health Reports, Год журнала: 2025, Номер 6(1), С. 421 - 430

Опубликована: Янв. 1, 2025

Язык: Английский