Aging leads to significant changes in the heart and respiratory system, increasing the risk of heart disease and respiratory issues like cough and dyspnea, especially in individuals 65 and older. Symptoms of heart disease and deteriorating sleep quality due to circadian rhythm changes are common in individuals from this age group. As global populations age due to longer life expectancy, a growing demand for effective healthcare solutions for older adults has been observed, with a focus on timely detection of health emergencies. This research focuses on the development of a wearable device for real-time health monitoring of older adults, which provides insights about key health parameters like heart rate, blood oxygen levels, sleep quality, stress, anxiety, cough intensity, and physical activity. The sensors employed in this device are MAX30100 for heart rate and blood oxygen level, MPU6050 for estimating intensity of physical activity and amount of sleep, a galvanic skin response sensor for estimating the stress and anxiety levels, and audio recording module combined with a CNN model for detecting user’s coughing severity. The CNN classifier was trained on spectrograms of audio samples containing coughing at different levels of severity. Random forest classifiers were trained on the heart rate, blood oxygen level, electrodermal activity, physical activity and sleep data to make predictions about all this data received in real time and then provide health related feedback to the user. The device uses WiFi capabilities of the ESP32 for wireless communication of data between the device and the computer where machine learning models and neural networks make predictions with the data about the user’s health. The values of the health parameters along with the predictions made by the trained classifiers are displayed on the OLED screen of the device and a user interface on the computer. The device effectively monitored key health parameters and provided real-time feedback. RF and CNN classifiers, both exhibited superior performance with CNN achieving 95.83% accuracy in classifying coughing intensity as normal, moderate, or intense. These results displayed on both the computer and the device's OLED screen, aiding users in managing their symptoms and deciding whether to seek medical assistance or not. Multiple trials with volunteers validated the benefits of the device for real-time health management of older adults. Volunteers found the feedback helpful in managing symptoms and praised the device for its potential to aid in cardiovascular and respiratory health monitoring.