New method predicts Alzheimer's timetables
A research team from Columbia University Medical Center has validated a new algorithm for predicting the length of time until full-time care, nursing home residency, and death for Alzheimer’s patients.
Pinpointing the stages and progression of Alzheimer’s is getting more precise. A research team from Columbia University Medical Center has validated a new algorithm for predicting the length of time until full-time care, nursing home residency, and death for Alzheimer’s patients.
Led by Dr. Yaakov Stern, professor of neuropsychology at CUMC, researchers followed two sets of Alzheimer’s patients for 10 years. Together, they developed a Longitudinal Grade of Membership (L-GoM) model, which looks at 16 variables in patients.
“The benefit of the L-GoM model is that it takes into account the complexity of Alzheimer’s disease,” Stern said. “Patients don’t typically fall neatly into mild, moderate, or severe disease categories. Our method is flexible enough to handle missing data.”
The model includes variable like ability to participate in routine, everyday activities, motor skills, duration of tremors, and mental status.
Two 68-year-old Alzheimer’s patients demonstrated the same mental status, but different psychiatric symptoms and levels of dependency. The method accurately predicted that the more dependent patient would die within three years, while the other would survive more than 10 years.
Now, the team is working to create a computer program that will allow doctors to input patient variables and receive a report.
Study co-author Dr. Nikolas Scarmeas says this information will be valuable for physicians and families of Alzheimer’s patients. “This method is more practical for routine use,” he says.
Results can be viewed as the expected amount of time to a particular outcome, allowing patients and families to prepare for what’s to come.
The entire study is available in the Journal of Alzheimer’s Disease.