Three years ago, Angelina Jolie announced in a New York Times op-ed that she'd had a preventive double mastectomy after testing positive for mutations in the BRCA1 gene, which put her at an increased risk of breast and ovarian cancers.
The article went hugely viral, and became a flashpoint in the debate about breast cancer risk and prevention. It also spurred a bunch of researchers to study what impact Jolie's decision might have on mastectomy rates and testing for the cancer-causing BRCA1 and 2 genetic mutations.
More from Vox:
A woman who called Michelle Obama an ape has her job back. This is part of a pattern.
Why Obamacare enrollees voted for Trump
Why anthropologists were so excited to discover a set of 3.66-million-year-old footprints
In the latest paper, published in the BMJ, researchers from Harvard looked at insurance data from nearly 10 million women before and after Jolie's May 2013 editorial.
In the two weeks following the article, they found that BRCA testing rates shot up by 65 percent. But mastectomy rates remained unchanged in the months after that. This suggests, they wrote, that "BRCA tests obtained as a result of the Jolie editorial did not yield additional BRCA positive mutations that might warrant preventive mastectomy."
These BRCA1 and 2 mutations are rare, and most experts agree only women with a very particular family history and risk profile should bother getting tested. Yet the study suggests women followed Jolie into medical screening they didn't need.
That's not to mention the cost to the health system: At about $3,000 per test, the researchers estimate this Jolie-inspired surge led to $14 million in health care spending in those two weeks alone.
And the "Jolie effect" seemed to persist long after the op-ed appeared. Average monthly test rates increased from 16 tests per 100,000 women between January and April 2013 to 21 tests per 100,000 women after the op-ed during May through December that year.
In sum, this looks like a case of celebrity-induced overtesting, the researchers wrote: "Celebrity announcements can reach a broad audience but may not effectively target the population that would benefit most from the test."
This is just one of many examples of where a celebrity says something about health and we follow like sheep. In 2000, Katie Couric's awareness campaign about colorectal screening led to an increase in colonoscopy use, and was dubbed the "Katie Couric effect." News of Kylie Minogue's announcement about a breast cancer diagnosis in 2005 led to an "unprecedented increase" in mammography bookings. After Charlie Sheen disclosed last year that he was HIV-positive, researchers sifted through Google search information and found that his announcement "corresponded with the greatest number of HIV-related Google searches ever recorded in the United States."
"The research — on the aggregate, at the population level — says it's clear: Celebrity culture, celebrity endorsements, have an impact," said Tim Caulfield, a researcher and author of the book Is Gwyneth Paltrow Wrong About Everything?. "Now we can debate whether that's good or bad."
Jolie's efforts to raise awareness about the BRCA genes were probably helpful for women who may also carry the mutations that can lead to cancer. But the rarity of the BRCA 1 and 2 mutations seemed to be lost on the public.
As far as celebrity health statements go, Jolie's op-ed — and another that followed — was actually quite measured in how she characterized risk. "My doctors estimated that I had an 87 percent risk of breast cancer and a 50 percent risk of ovarian cancer, although the risk is different in the case of each woman," she wrote. "Only a fraction of breast cancers result from an inherited gene mutation." Jolie clearly tried to convey the uniqueness of her situation and the fact that it may not apply to all women.
But that didn't translate to more public understanding. One study, in Genetics in Medicine, looked at the effect Jolie's op-ed had on public awareness. It found that she did not actually help to improve people's understanding of breast and ovarian cancer risk. "While three of four Americans were aware of Angelina Jolie's double mastectomy," the University of Maryland researchers wrote, "fewer than 10 percent of respondents had the information necessary to accurately interpret Ms. Jolie's risk of developing cancer relative to a woman unaffected by the BRCA gene mutation."
The news media, which often botches risk communication, is partly to blame for this. Another article, also published in Genetics in Medicine (and co-authored by Caulfield), found that journalists took an overwhelmingly positive slant on Jolie's preventive surgery, instead of discussing the relative rarity of her genetic mutation and the fact that most women would have many other options besides a double mastectomy.
Celebrity health endorsements, no matter how well-intentioned and carefully crafted, frequently lead to misunderstanding. But there are exceptions, Caulfield said: "Celebrities are most helpful in areas where the message is straightforward and simple — wear your seatbelt, don't smoke, eat fruits and vegetables." They are less instructive when the messages are complex, like communicating individual genetic risk information. So we should keep the Jolie case study in mind next time we hear a celebrity with a health message that's more nuanced than "eat your broccoli."