A UK study found that models designed to screen for those at high risk of developing dementia revealed that many who went on to have the disease were classified as “normal risk.”
A new study from the UK suggests that the models used to screen for potential future cases of dementia are not reliable predictors of who will eventually go on to develop the disease.
The study, which looked at patient data from over 30,000 people across three decades, found that only a small percentage of patients considered high risk for dementia went on to develop it, and that many who developed the disease had been classed as normal risk instead.
Dr. Sebastian Walsh, a co-author of the study, explained that it doesn’t mean that most people from normal-risk populations are more likely to develop dementia. “Rather, [it’s] that most cases of dementia will come from the normal-risk group. If the high-risk group are four times more likely to get dementia, but there are 1,000 times more people in the normal-risk group, then more cases will come from the normal-risk group, just by sheer weight of numbers.”
The study cross-referenced five dementia screening models against another long-term study containing data from 30,387 patients over 29 years. Each risk assessment model used different criteria to screen for dementia, including age, sex, education, stress, heart health, cholesterol, physical activity, alcohol intake and smoking, body mass index, and APOE4, the highest risk genetic variant for Alzheimer’s.
For every ten people who developed dementia across the nearly 30-year period, seven had been classed as “normal-risk.” Of the patients deemed at “high risk” for developing dementia, only two out of every ten did.
“This … doesn’t mean that the high-risk are not high-risk,” said Walsh. “Risk prediction can still be helpful for setting eligibility criteria for trials,” he explained. By relying on established risk factors to include people who are considered “higher than normal risk,” the study has a chance to examine people who eventually do develop dementia versus how many do not.
“Also, some people go to their physician and want to know their risk and how to lower it, so giving those people the best available evidence is perfectly fine.”
The study instead argues for a policy approach that supports the overall health of the general population as the best way to reduce everyone’s risk of developing the disease.
According to the study, large-scale, dementia-informed policy, also known as “population-level interventions” that target risk factors could supplement individualized care. As for what these policies could include, another study by the same authors suggests increasing prices on alcohol and tobacco products, taxing sugary drinks, creating mandates around how unhealthy food is marketed to children, and making “compact neighborhoods” with interconnected walking and biking trails the standard for urban planning.
“If we want to prevent as much future dementia as possible,” Walsh said, “we need to lower everyone’s risk, not just the group who are at higher risk.”
Lisa Power is a recent graduate of the Columbia University Graduate School of Journalism, and has written about health policy for VICE.
