How youth experience and define their sexuality encompasses multiple dimensions including the identity labels they use, their sexual attractions, and their sexual partners. [1] Prior research has found that change over time in one or more of these dimensions—sexual orientation identity (SOI), attractions, and/or gender of sexual partners (behavior)—is common among youth. [2], [3], [4], [5] Prevalence of change has been found to vary by age (more common among adolescents compared to young adults), sexual orientation (more common among sexual minority compared to heterosexual individuals), and gender (more common among transgender and nonbinary individuals compared to cisgender individuals). [3], [4], [5], [6], [7], [8] Experiencing change in one or more dimensions of sexuality—sometimes called sexual fluidity—may exacerbate stigma and prejudice experienced by sexual minority youth and has been linked to a number of adverse health outcomes through minority stress pathways. [9], [10], [11], [12] As such, accurate measurement of sexual fluidity is critical to advancing youth health.

Sexual fluidity, changes in one or more dimensions of sexual orientation (SOI, attractions, behavior), may occur due to a range of individual, interpersonal, and societal factors. This includes shifts due to: changes in awareness of one’s attractions over the life course, disclosure of SOI to others (e.g., “coming out”), and discovery of identity labels (e.g., queer) that feel like a better description of one’s own sexual orientation than other labels (e.g., lesbian or gay). [13] Changes in attractions, labels, and/or behaviors may also occur in response to experiences of prejudice. For example, an individual identifying with a plurisexual identity (e.g., bisexual, pansexual) who experiences anti-plurisexual prejudice may choose to use a monosexual identity (e.g., lesbian, heterosexual) to prevent further experiences of prejudice, even if that identity does not accurately reflect their sexual orientation. [14]

Research documenting sexual minority health and health-related inequities has been limited by a belief that sexual orientation is a static characteristic, as well as inadequate methods to measure changes in sexual orientation over time. First, surveys often assess fluidity across only one dimension of sexual orientation in isolation. Specifically, studies typically focus on changes in SOI or attractions, and less commonly sexual behavior. [6] Prior work has shown these dimensions do not always align, especially among sexual minority individuals. [2] In addition, sexual orientation research often focuses on presumed cisgender individuals and either overtly excludes transgender and nonbinary individuals, due to sample size concerns, or does not assess the sexual orientations of individuals with transgender or nonbinary gender identities. Finally, cross-sectional study designs may be subject to recall bias. While more recent studies have assessed prospective change in sexual orientation across multiple timepoints, [3], [4], [5], [12] none to our knowledge have compared change as measured via prospective report with retrospective recall. Since longitudinal analyses are not always feasible, research is needed to understand whether retrospective measures capture changes in sexual orientation dimensions similarly to prospective assessment.

The purpose of the current study was to use a prospective cohort of youth from across the US to compare retrospective and prospective reports of change across two dimensions of sexual orientation (SOI and attractions). The first aim was to compare retrospective recall of past 4-month SOI change with prospective changes in self-reported SOI across 4-month follow-up. The second aim was to compare retrospective recall of past 4-month attraction change with prospective changes in self-reported attractions across 4-month follow-up. Given retrospective measures are subject to recall bias and findings from prior work assessing change in sexual orientation prospectively, [3], [4], [5], [12] we hypothesized that prospective assessment would capture more change experiences than retrospective recall across both dimensions, which would support the value of prospective assessment of change in both research and clinical care.