Genetic Testing and Polygenic Risk Scores for Longevity

What this page covers

Genetic testing is sometimes sold as a way to learn how long you will live or which age-related diseases you may face. This page explains the main kinds of tests, what the published evidence says about their accuracy and usefulness, and the privacy limits. It is information only. It does not interpret anyone's results and does not suggest ordering, skipping or acting on a test. Evidence labels used here: established (well supported), promising but limited, insufficient evidence, and marketing claim.

Three different things called "genetic testing"

The FDA notes that direct-to-consumer tests have varying levels of evidence behind their claims, and that companies do not all test the same variants, so they can give different results for the same condition. [2]

APOE: a risk shift, not a diagnosis

The APOE gene comes in several versions. MedlinePlus Genetics explains that people who carry the e4 version inherit an increased risk of developing Alzheimer's disease, not the disease itself. [3] The US Preventive Services Task Force lists the e4 version as a reported risk factor for Alzheimer disease in white and Asian populations. [4]

In a randomized trial of 162 adults without symptoms who had a parent with Alzheimer's disease, learning one's APOE result did not cause significant short-term anxiety, depression or test-related distress on average. [5] People who were already highly distressed before testing were more likely to struggle afterward. [5] Label: established for the risk association; insufficient evidence from the sources read that knowing the result changes health outcomes.

Polygenic risk scores

A polygenic risk score adds up the effects of many small variants. The National Human Genome Research Institute explains that it can only describe relative risk, meaning how one person compares with others, and that it only shows correlations, not causes. [1] Scores will always be probabilities, not certainties. [1]

What the research shows so far:

• A 2020 review found that clinical usefulness has yet to be established and that the ability to tell people apart in the general population is low. [6]
• In a Chinese cohort of 41,271 people followed for an average of 13 years, people in the top fifth of a coronary artery disease score had about three times the risk of those in the bottom fifth, but adding the score to a standard clinical risk score improved prediction only modestly (1% on one measure). [7]
• Accuracy may be best in people of European ancestry because most studies used them. [1] For other groups there may not be enough data to calculate a reliable score. [1]
• Health professionals do not yet routinely use these scores because there are no practice guidelines. [1]

Label: promising but limited for research and for refining risk in people already under care; marketing claim when a score is presented as a prediction of lifespan.

Regulation and privacy

In general, the FDA does not review direct-to-consumer tests for non-medical, general wellness or low-risk purposes before they are offered. [2] Laws change and this is not legal advice.

In the United States, the Genetic Information Nondiscrimination Act (GINA) protects against discrimination based on genetic information in health insurance and employment. [8] It does not cover long-term care, life or disability insurance, although some states add protections. [8] Think about this before sharing results.

Who may want to be extra careful

The FDA suggests that consumers consider discussing results with a genetic counselor or a qualified health care provider to understand what a result means for their own health. [2] This matters most for anyone with a strong family history or who expects a result to be upsetting.

How to judge a genetic test claim

• Which condition is it about, and in which ancestry group was it studied?
• Does it give a relative ranking or a real chance over a stated time?
• What would you do differently with the result, and has that action been shown to help?
• Who can see the data, and is it shared or sold?

Related guides

Read Biological Age vs Chronological Age, Interpreting Test Results Without Overinterpreting and Cognitive Testing and Brain Health Screening.