Suicidal behavior is relatively well studied in younger populations but remains understudied among older adults despite significant risks linked to loneliness, physical ailments, and depression. This study addresses the gap by analysing suicide among older adults in Sweden using advanced statistical and machine learning techniques. Leveraging both static and longitudinal data – including suicide attempts, psychopharmaceutical prescriptions, medical conditions, and sociodemographic factors – we tackle the challenges associated with temporal dynamics and data censoring. The proposed approach integrates survival models such as Cox Proportional Hazards (CoxPH), Random Survival Forest (RSF), and Gradient Boosting Survival Analysis (GBSA) with sequential machine learning methods, specifically Long Short-Term Memory (LSTM) networks. Our results demonstrate that combining survival analysis with sequential models significantly enhances prediction accuracy by effectively capturing time-dependent patterns and censored observations.