Skip to Content

Genetics Doesn’t Help Predict Diabetes Risk

A comparative study shows that lifestyle factors are more effective.
January 21, 2010

A couple of years ago, spurred by a strong family history of diabetes and journalistic curiosity, I took one of the newly available genetic tests to asses my risk for the disease. The test assessed whether I carried a variant called TCF7L2, which was discovered by scientists at deCode in 2005–almost 20 percent of people with type 2 diabetes carry two copies of the high-risk version of the gene.

Since then, scientists have discovered a number of other genetic risk factors for type 2 diabetes, and tests to detect them are available directly to consumers via the internet, along with an analysis of that person’s lifetime risk. But a new study confirms what some have said all along–that these analyses add little to our ability to predict who will get the disease. So-called phenotypic factors, such as age, body mass index, waist circumference, and cholesterol levels are much more accurate predictors.

Risk of developing type 2 diabetes is highly dependent on environmental factors, notably diet and obesity. But it also has a strong genetic component–someone with a sibling with the disease is much more likely to get it than someone without, regardless of diet. Companies including 23andMe and Navigenics have marketed genetic tests to assess risk of type 2 diabetes and other diseases with the idea that this information can help people try to prevent diseases for which they are genetically at risk. But the new study calls that into question.

According to GenomeWeb:

Researchers from the University College London followed thousands of individuals in London’s Whitehall district over roughly 20 years. When they looked specifically at factors affecting type 2 diabetes risk, they found that two non-genetic risk models were better predictors of diabetes risk than a genetic model based on nearly two dozen risk alleles.

…When the researchers assessed the so-called Cambridge and Framingham type 2 diabetes risk models, which are based on non-genetic factors such as age, sex, family history, waist circumference, body mass index, smoking behavior, cholesterol levels and so on, they found that both predicted risk of the disease better than a genetic risk model based on 20 common, independently inherited risk SNPs.

The Cambridge model had 19.7 percent sensitivity for detecting type 2 diabetes cases in the Whitehall cohort based on a five percent false positive rate, while the Framingham model had 30.6 percent sensitivity. The gene count score, meanwhile, detected 6.5 percent of cases at a five percent false positive rate and 9.9 percent of cases at a 10 percent false positive rate. In addition, the team noted, adding genetic risk information did not significantly improve the ability to identify individuals at risk of type 2 diabetes over either non-genetic risk model alone.

The researchers who conducted the study didn’t discount studies searching for genetic risk factors for diabetes. They said that the most useful outcome of those studies will be a better understanding of the biological basis of the disease, potentially pointing toward new drug targets.

Direct-to-consumer genomics companies say it’s too soon to disregard the value to the individual. It’s not yet clear whether people will be more motivated by a genetic risk than an environmental one. Or whether the ability to intervene very early on, as genetic risk factors would allow, is more effective than later interventions.

Here’s 23andMe’s response, via GenomeWeb:

Joanna Mountain, senior director of research for direct-to-consumer genetics firm 23andMe, agreed that non-genetic factors are important contributors to type 2 diabetes risk. But she says there is still a place for genetic testing–particularly for a subset of individuals with a relatively high genetic risk for the disease.

“One point that the authors don’t make is that a small fraction of individuals learn from genetic data that their type 2 diabetes risk is very high,” Mountain told GWDN by e-mail. “The results for these individuals don’t influence the statistic that is used heavily in this study, but for those individuals, the genetic information can be very valuable.”

Mountain also noted that genetic testing allows for earlier risk assessment than non-genetic factors. And, she said, in contrast to weight, body mass index, and other non-genetic factors, genetic profiles do not change over time.

23andMe currently evaluates eight of the 20 alleles tested in the study. Mountain said the company will continue adding genetic markers for type 2 diabetes risk as these alleles meet their scientific criteria.

Keep Reading

Most Popular

This new data poisoning tool lets artists fight back against generative AI

The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. 

Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist

An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.

The Biggest Questions: What is death?

New neuroscience is challenging our understanding of the dying process—bringing opportunities for the living.

Driving companywide efficiencies with AI

Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.