Sexual Orientation and Gender Identity Data and HIV Preexposure Prophylaxis Disparity Detection

US federal reporting changes in 2025 removed sexual orientation and gender identity (SOGI) measures from health-center reporting, raising concerns about equity surveillance.

To quantify how incomplete or absent SOGI data are associated with detection of disparities in documented HIV preexposure prophylaxis (PrEP) orders.

This cross-sectional study used electronic health record data from a large safety-net health system in Southern California. Data were from patients 18 years or older registered as of October 28, 2025, with documented PrEP orders from January 2, 2019, through October 27, 2025. Registration dates were unavailable, precluding common follow-up. Three surveillance scenarios were assessed: complete SOGI, sex assigned at birth (SAAB) only, and joint SOGI missingness from 0% to 90%. Adults with classifiable SAAB, sexual orientation, and gender identity were included.

Complete, unavailable, or partially missing SOGI data.

The primary outcome was any documented PrEP order. Logistic regression models adjusted for age and race and ethnicity, with interactions of sexual orientation or gender identity with SAAB. Marginally standardized probabilities and adjusted percentage-point differences were estimated within SAAB strata. Each nonzero missing level included 250 simulations to assess the magnitude, precision, and statistical detectability of each difference.

This study included 165 198 adults (mean [SD] age, 45.0 [16.4] years). Most patients (97 621 [59.1%]) were female SAAB; 5790 (3.5%) were lesbian, gay, bisexual, or another sexual minority (LGB+) and 303 (0.2%) were transgender or gender diverse (TGD). A total of 1406 patients (0.9%) had a documented PrEP order. Under the complete SOGI analytic scenario, the adjusted probability of a documented PrEP order among cisgender patients assigned male at birth was 26.7 percentage points (pp) (95% CI, 24.9-28.5 pp; P < .001) higher for LGB+ patients than for heterosexual patients. Among all patients assigned male at birth, the adjusted probability was 12.3 pp (95% CI, 7.0-17.6 pp; P < .001) higher for TGD patients than for cisgender patients. Neither difference was significant among patients assigned female at birth. Under SAAB-only surveillance, only a difference between male and female SAAB was measurable (1.4 pp; 95% CI, 1.3-1.5 pp; P < .001), whereas SOGI-specific differences were not estimable. Under joint missing-completely-at-random deletion, positive-direction detectability of the male-SAAB sexual orientation difference remained at 100% through 90% missingness; detectability of the gender identity difference declined from 100% at 40% missingness to 51.6% at 80% missingness and 19.2% at 90% missingness. Selective missingness attenuated estimates.

In this cross-sectional study, the complete SOGI analytic scenario enabled estimation of SOGI-stratified differences, whereas SAAB-only surveillance did not. Partial collection preserved estimability but reduced precision, and selective missingness attenuated estimates.