Researchers have successfully used artificial intelligence to forecast the outcomes of hospitalization for elderly individuals who have dementia on either the first or second day of admission. Early outcome evaluation allows for more rapid interventions, improved care coordination, prudent resource allocation, focused care management, and early treatment for more susceptible patients at a higher risk.
Because older patients with dementia have more extended hospital stays and incur more significant health care expenditures than other patients, the team attempted to find a solution to this problem by identifying modifiable risk factors and constructing an artificial intelligence model that improves patient outcomes, enhances their quality of life, and decreases their hospital readmission risk, as well as reduces hospitalization expenses after the model is put into practice. This model will improve patient outcomes and enhance their quality of life.
This study aimed to identify risk factors for poor outcomes among subgroups of patients with various types of dementia caused by diseases such as Alzheimer’s, Parkinson’s, vascular dementia, and Huntington’s, among others. The records of 8,407 geriatric patients with dementia were reviewed over ten years at Houston Methodist’s eight hospitals. The researchers used this information to construct a machine learning model to rapidly identify predicted risk variables and their prioritized relevance for unfavorable hospitalization outcomes early on during these patients’ hospital stays.
Their approach beat all other popular risk assessment methods for these various forms of dementia, with an accuracy of 95.6%. According to the researchers, none of the other currently available methods have applied A.I. to comprehensively predict hospitalization outcomes of elderly patients with dementia in this manner. Furthermore, these other methods do not identify specific risk factors that additional clinical procedures or precautions can modify to reduce the risks.
“The study revealed that if we can identify senior patients with dementia as soon as they are hospitalized and detect the key risk factors, then we may apply some effective therapies right away,” said Eugene C. Lai, M.D., Ph.D., the Robert W. Hervey Distinguished Endowed Chair for Parkinson’s Research and Treatment in the Stanley H. Appel Department of Neurology. “By reducing and correcting the modifiable risk factors for unfavorable outcomes immediately
Lai, a neurologist, has spent many years working with these patients. She wanted to investigate ways to understand better how they are handled and their behavior when hospitalized so that physicians may enhance the treatment and quality of life for people with these conditions. Because of their prior work together, he knew that Stephen T.C. Wong, Ph.D., P.E., an expert in bioinformatics and the Director of the T. T. and W. F. Chao Center for BRAIN at Houston Methodist, would be the best person to approach with this idea. He also knew that Wong’s team had access to a large clinical data warehouse containing information on Houston Methodist patients and the ability to use artificial intelligence to analyze large amounts of data.
It has been determined which risk factors are associated with each form of dementia, including those accessible to treatments. The most significant identified hospitalization outcome risk factors included encephalopathy, the number of medical problems present at admission, pressure ulcers, urinary tract infections, falls, admission source, age, race, and anemia, with several overlapping factors found in multi-dementia groups.
Ultimately, the researchers aim to develop mitigating factors that can direct therapeutic actions to decrease these adverse effects. According to Wong, the unique emerging approach, which is to use sophisticated A.I. predictions to trigger the deployment of “smart” clinical courses in hospitals, would not only enhance clinical results and the patient experience but also cut the expenses associated with hospitalization.
Wong, the paper’s corresponding author and the John S. Dunn Presidential Distinguished Chair in Biomedical Engineering with the Houston Methodist Research Institute, stated that “Our next steps will be to implement the validated A.I. model into a mobile app for the ICU and main hospital staff to alert them to geriatric patients with dementia who are at high risk of poor hospitalization outcomes and to guide them on interventional steps to reduce such risks.” “Our next steps will be implementing the validated A.I. model into a mobile app. We will collaborate with the hospital’s information technology department to ensure that this application is integrated smoothly into EPIC as part of a system-wide rollout for normal clinical usage.
He stated that this would follow the same intelligent clinical pathway strategy that they have been working on to integrate two other novels A.I. apps that his team developed into the EPIC system for routine clinical use to guide interventions that reduce the risk of patient falls with injuries and better assess breast cancer risk to reduce unnecessary biopsies and overdiagnosis. He also stated that this would follow the same intelligent clinical pathway strategy that they have been working on to integrate two other novels A.I. apps that his team developed into the EPIC system for routine clinical use to guide interventions.