A new analysis of US health data shows a sharp fall in LGBTQ+ identification among younger teenagers, particularly ninth-grade girls. The figures have prompted fierce political debate, but researchers caution that the reasons remain uncertain and that identity, disclosure and social attitudes can all shift over time.

Essential Takeaways

  • The biggest change: Ninth-grade girls identifying as lesbian or bisexual fell from 30% in 2021 to 17% in 2025.
  • Bisexual identification halved: The proportion reporting they were bisexual dropped from 26% to 13% over the same period.
  • Older teenagers were steadier: Identification among older pupils declined less dramatically than among younger students.
  • The data needs care: Survey responses reflect how teenagers describe themselves at one moment, not a permanent measure of sexuality.
  • Politics has filled the gap: Commentators on both sides are using the figures to support wider arguments about culture, education and media influence.

A striking drop among younger girls

The clearest finding is also the hardest to ignore. Analysis of the US Centers for Disease Control and Prevention's Youth Risk Behaviour Surveillance System data found that the share of ninth-grade girls identifying as lesbian or bisexual fell from nearly one in three in 2021 to fewer than one in five in 2025.

That change has been described as a "free fall" by Jean Twenge, the San Diego State University psychology professor who analysed the figures, according to the Washington Post. The fall was especially pronounced among pupils identifying as bisexual, whose proportion dropped from 26% to 13%.

Still, the numbers describe self-identification, not a clinical diagnosis or a definitive account of a teenager's future. As Them has noted in its coverage of the trend, young people may use different labels as they learn more about themselves, and some may be more or less comfortable answering a survey openly.

Why the figures are prompting political arguments

The timing has made the figures irresistible to political commentators. The sharp rise in LGBTQ+ identification among young people occurred during the pandemic era, followed by a decline in the years afterwards. That pattern has led conservative voices to claim that cultural messaging drove the earlier increase, while progressive commentators have questioned whether survey changes, social pressure or shifting language might explain the movement.

The original analysis doesn't establish that any government programme caused the rise or the fall. Nor does it show that ending foreign-aid initiatives would change how American teenagers understand their sexuality. Yet online debate has linked the figures to the Trump administration's approach to USAID, schools and LGBTQ+ policy, turning a complicated demographic trend into a neat partisan talking point.

That leap deserves caution. Correlation can be suggestive, but it isn't proof of causation, particularly when the period also included school closures, isolation, changing mental-health patterns and major arguments about gender and sexuality.

The age gap tells an important story

The decline isn't evenly spread across the teenage years. Younger pupils showed the steepest movement, while older teenagers' responses were comparatively stable. That may indicate that younger respondents are more open to changing labels, or that they are more influenced by the language and social atmosphere around them.

It may also reflect a cohort effect. A ninth-grader answering a survey in 2021 grew up amid unusually intense public discussion of identity, representation and inclusion. A similar pupil answering in 2025 encountered a more sceptical political climate and a sharper backlash against corporate and educational LGBTQ+ campaigns.

Them's reporting places the figures in that wider context, stressing that labels can carry different meanings for different respondents. For some teenagers, bisexual may have represented uncertainty, openness or experimentation rather than a settled identity. A change in the percentage choosing that word doesn't automatically mean an equivalent change in attraction.

What the data can, and can't, tell us

Large surveys are useful because they reveal patterns that individual anecdotes can't. But they have limits. The CDC's figures depend on students choosing whether to participate, understanding the questions in the same way and feeling safe enough to answer honestly.

There are also practical differences between saying "I'm bisexual", selecting a category on a form and discussing sexuality with friends or family. Social acceptance can make disclosure easier, while fear of judgement can make it less likely. That means a falling percentage could reflect changes in identity, language, confidence or privacy, rather than a single cultural cause.

For parents and teachers, the sensible response is to avoid treating either rise or fall as a victory. A teenager shouldn't feel pushed towards a label, but shouldn't feel pressured to hide one either. Quiet, respectful support is more useful than turning survey data into a culture-war scoreboard.

A trend worth watching, not overselling

The latest figures are significant because the change is large, concentrated and visible across several years. They deserve proper analysis, especially as further surveys may show whether the decline continues or settles.

But the political claims now attached to the numbers go well beyond what the evidence can prove. The data offers a snapshot of how American teenagers described themselves in 2025, not a verdict on USAID, schools, media or any political party.

Teen identity is rarely a straight line, and these figures are best read as a prompt for better questions rather than a ready-made answer.

A changing label is not the same thing as a solved mystery.

Source Reference Map

Story idea inspired by: [1]

Sources by paragraph: