跳到正文
原文
MedPage Today·· 11 天前AI 评分45

自闭症风险误解:四分之一成人认为疫苗或孕期使用泰诺会致自闭症

Autism Risk Myths: 1 in 4 Adults Cite Vaccines or Tylenol

AI 导读

在一项针对30,000名美国成年人的调查中,约四分之一的人认为疫苗或孕期使用泰诺会增加自闭症风险。共和党支持者比民主党支持者更倾向于这种错误信念,且年龄在35至44岁的人群中这种信念最为普遍。频繁使用社交媒体与较低的错误信息接受率相关。

正文

— Beliefs tied to political affiliation, survey of 30,000 Americans showed

by Judy George, Deputy Managing Editor, MedPage Today

November 9, 2026 • 3 min read

  • About one in four adults endorsed vaccines or prenatal acetaminophen (Tylenol) as autism risks in a large U.S. survey.
  • Strong Republicans had higher odds of endorsing vaccine-autism beliefs than strong Democrats.
  • Frequent social media use correlated with lower misinformation endorsement.

About one in four U.S. adults endorsed childhood vaccinations or acetaminophen (Tylenol) use during pregnancy as an autism risk factor, a cross-sectional study showed.

In a nationwide survey of 30,000 adults, 26.6% endorsed the vaccination belief and 23.1% endorsed the acetaminophen belief, reported Roy Perlis, MD, MSc, of Massachusetts General Hospital in Boston, and co-authors in JAMA Health Forum.

Genetic causes, an evidence-based risk factor, was the most widely endorsed, at 66.7%.

Political affiliation was strongly associated with vaccination beliefs. In adjusted models, adults who identified as strong Republicans had 3.83 (95% CI 3.47-4.24) times the odds of endorsing the vaccine-autism belief compared with strong Democrats.

Vaccine-autism risk beliefs were highest among adults ages 35 to 44 years (adjusted OR 1.64 compared with those ages 18 to 24 years), and lowest among people 65 and older (adjusted OR 0.64).

Greater trust in scientific and medical institutions was associated with lower rates of endorsement, as was more frequent social media use.

"Our results add to abundant evidence of the growing politicization of beliefs about science and medicine," Perlis and co-authors stated. "This politicization of science preceded the COVID-19 pandemic, but variability in responses to the pandemic by state and federal leadership likely accelerated this trend."

Prior survey results have documented a decline in trust in healthcare and science during the pandemic, they noted. "Our results may reflect this diminished trust in traditional sources of public health information," Perlis and colleagues observed.

"This study documents the corrosive effect of the epidemic of disinformation that has been promoted by a range of government officials and others to foster false beliefs about vaccines," said Georges Benjamin, MD, CEO of the American Public Health Association in Washington, D.C., who wasn't involved with the research.

"It shows we have a lot a work to do as an evidence-based field to turn this around," Benjamin told MedPage Today.

"What is especially concerning is the preponderance of these false beliefs among people of different political parties," he pointed out. "Science is supposed to be apolitical, but this shows the negative impact of the politicization of health information."

In 1998, a case series published in The Lancet suggested vaccination was associated with autism risk; the paper was retracted after numerous flaws were identified. Since then, large population-based studies have consistently showed no link between vaccines and autism.

"There has been no subsequent credible evidence supporting a link, whereas numerous studies have demonstrated no such association," Perlis and co-authors wrote. However, both President Trump and Health Secretary Robert F. Kennedy Jr. have repeatedly suggested vaccines could play a role in rising autism rates.

Maternal acetaminophen use also has been suggested to increase autism risk despite strong evidence against a link, the researchers noted. In September 2025, Trump tied prenatal acetaminophen use to autism, warning pregnant women of possible risks.

To assess the potential effect of this type of information on the public, Perlis and colleagues evaluated data from an internet-based survey of adults ages 18 or older in all 50 U.S. states and the District of Columbia. The survey was administered between December 18, 2025, and January 27, 2026.

The primary study outcome was the endorsement of individual risk factors, scored on a 5-point Likert scale from definitely false to definitely true. Among the 30,288 survey respondents in the analysis, the mean age was 46.6 years. Most (54.1%) identified as women, 45.2% as men, and 0.6% as nonbinary.

Endorsement of vaccine-autism risk beliefs was associated with men (adjusted OR 1.26), and African American or Black (adjusted OR 1.25) or Hispanic (adjusted OR 1.13) respondents compared with white respondents.

Adults who used social media multiple times a day had lower odds of endorsing misinformation (adjusted OR 0.72 compared with those who used it less than weekly).

"Although social media use has been blamed for accelerating dissemination of misinformation, we did not identify a positive association between social media use and belief in non-evidence-based risk factors," Perlis and co-authors noted. "One possibility is that social media use parallels information seeking more generally -- i.e., those with lower levels of use are also those with less access to sources of reliable health information."

The analysis was cross-sectional and causality cannot be determined, the researchers noted. Social media use was self-reported and may not have been accurate.

来源:MedPage Today · medpagetoday.com