Sexual orientation has been linked to certain health conditions, and yet quantifying sexual orientation in longitudinal studies is challenging. This study examined different methods of accounting for sexual orientation in a cohort study of 300 homeless and unstably housed women followed every 6 months over 3 years. Altogether, 39.7% (n = 119) could be considered sexual minority at one or more time points based on identity and/or behavior; 16.3% (n = 49) reported shifts in sexual identity. Only 24.0% (n = 72) were identified as sexual minority through a single measure of sexual identity, 27.0% (n = 81) were identified with a single measure of identity and behavior, 33.0% (n = 99) were identified through annual measures of identity and behavior, and 22.0%-22.3% (n = 66-67) were identified through latent class analysis including all identity/behavior measures. This study found that sexual fluidity is common in unstably housed women, and many sexual minority women would be missed in longitudinal studies with different methods of accounting for sexual orientation.
Keywords: Sexual orientation; bisexual; fluctuation; homelessness; lesbian; sexual behavior; sexual identity; women.