Financial behavior changes up to 6 years before dementia diagnosis

Late bill payments, falling credit scores among early warning signs

Researchers use 20 years of credit data to build dementia-risk AI

Early detection hopes tempered by privacy and discrimination concerns

An illustration depicting dementia. (Yonhap)
An illustration depicting dementia. (Yonhap)

What if a parent who once managed money with meticulous care suddenly starts missing utility payment deadlines? Or a lifelong cautious investor begins making impulsive large withdrawals and asking the same financial questions over and over? In the United States, researchers are drawing attention to the idea that such shifts in financial behavior may be early warning signs of dementia — and technology is now being developed to let banks use AI to analyze customers' transaction records and flag the risk before a diagnosis is ever made.

Bloomberg reported Nov. 13 (local time) that American researchers are studying how financial data could be used to detect dementia risk at an early stage, based on the observation that changes in how a person manages money can appear years before a formal diagnosis.

Sandy Adams, a financial planner who has worked in Michigan for 30 years, regularly spots signs of cognitive decline in her clients' financial behavior — sudden requests for large cash withdrawals, repeated questions, unusual anxiety about the stock market, and forgetting the location of scheduled meetings.

About 75 percent of her clients are between 65 and 95 years old. "If you're not paying close attention, you might miss these changes," Adams said. "More and more people are experiencing cognitive impairment, so I'm trying to prepare in advance."

Research is also confirming the link between financial behavior and dementia. According to a study by Lauren Hersch Nicholas, a health economics professor at the University of Colorado Anschutz Medical Campus, people diagnosed with Alzheimer's disease or related dementias may begin experiencing financial difficulties — such as late bill payments and falling credit scores — up to six years before their diagnosis.

A separate study found that the gap in average household assets between those later diagnosed with dementia and those who were not widened to nearly $100,000 starting about six years before clinical onset. Researchers said the gap was more likely tied to a decline in financial judgment than to rising medical costs or falling income.

"Every domain related to financial management overlaps with the areas dementia affects first," Nicholas said. "That includes the ability to pay bills on time, evaluate investments and avoid fraud."

As a growing body of research shows that warning signs can appear in financial activity before a dementia diagnosis, efforts are underway in the United States to develop early-detection technology using credit data.

Carol Roan Gresenz, dean of the McCourt School of Public Policy at Georgetown University, is conducting research on cognitive decline using credit data in collaboration with the Federal Reserve Bank of New York. The research team is analyzing thousands of variables in anonymized credit records to identify patterns in financial behavior associated with dementia risk.

Nicholas also launched a study in September using AI to analyze 20 years of accumulated credit data. The core aim is to determine what connections exist between changes in financial activity — such as late payments and credit score fluctuations — and dementia diagnoses.

Created using Gemini
Created using Gemini

Researchers are also exploring ways to eventually quantify dementia risk using financial data, developing indicators that doctors could consult when deciding whether a patient needs cognitive testing. For now, however, the work remains at the research stage and is not yet ready for clinical application.

The fact that financial institutions already operate systems to detect unusual transactions and fraud is also drawing notice. AARP, the US advocacy group for older adults, has since 2019 run a program called BankSafe that trains financial institution employees to identify abnormal financial behavior in elderly customers.

According to AARP, more than 1,500 financial institutions have joined the program to date, and it is estimated to have prevented roughly $550 million in consumer losses. Those figures, however, reflect prevention of financial fraud and exploitation — not direct detection of dementia.

Significant hurdles remain before dementia-risk prediction technology using financial data can be commercialized. Using individuals' spending records and credit information to infer health status raises serious privacy concerns.

Particularly worrying is the possibility that AI classifications of certain customers as high-risk for dementia could be used in loan assessments or to restrict access to financial services — meaning technology developed to protect customers could end up enabling financial discrimination.

Matt Haslinger, head of anti-money laundering at M&T Bank, acknowledged the need for research into financial vulnerability and the potential for exploitation, but stressed that "banks should not be in the business of diagnosing customers with cognitive impairment or dementia."

Researchers themselves maintain that financial data alone cannot determine whether someone has dementia. Because late utility payments and falling credit scores can also result from job loss or reduced income, they say financial data analysis should serve as a supplement to clinical diagnosis rather than a substitute.

Bloomberg also highlighted a real-life case in which changes in financial management ability appeared as early signs of dementia. Jan DeCateret, a Michigan resident, said she began noticing her husband Jerry's declining ability to calculate figures about 10 years before he died of Alzheimer's disease.

Jerry, a retired electrical engineer, had taken on the task of tallying scores at the couple's golf group after leaving work. Calculations that once took him a few hours gradually began consuming an entire day. Changes in his memory and behavior followed, and Jerry ultimately died of Alzheimer's disease in 2023.

After recognizing her husband's cognitive decline, Jan took over the household finances herself. She switched utility payments to automatic transfers, learned how to read financial statements and took the lead in meetings with their financial planner. Those steps allowed the couple to avoid serious financial losses from the disease.

"The most important thing is to have someone you trust by your side — like a reliable financial planner," Jan said.


sjy@heraldcorp.com