Disparities in Uptake of HIV Preexposure Prophylaxis in a Large Integrated Health Care System

Although uptake of HIV preexposure prophylaxis (PrEP) is increasing,^1,2 lower uptake among key HIV risk groups could further increase existing disparities in the HIV epidemic.^3,4 However, it is unknown whether certain demographic subgroups are underrepresented among PrEP users compared with individuals who recently acquired HIV infection. Thus, we compared demographic characteristics of PrEP users with those of recent HIV seroconverters within the Kaiser Permanente Northern California (KPNC) health care system.

KPNC provides comprehensive medical services to 3.9 million members, comprising 46% of insured individuals in the KPNC service area (R. Fong, KPNC Market Strategy and Analysis, written communication, April 2016) and mirroring the demographics of the surrounding population. We included KPNC members who either initiated PrEP or acquired HIV infection. PrEP users were those who initiated PrEP after the US Food and Drug Administration (FDA) approval (i.e., July 2012‒December 2014). Because the demographic composition of those who acquired HIV infection may have changed as high-risk individuals have initiated PrEP, we included HIV seroconverters from the three-year period prior to FDA approval of PrEP (i.e., July 2009‒June 2012). Seroconversions were defined as a first lifetime HIV diagnosis at KPNC after a negative HIV test during the three-year pre-PrEP period. Seroconverters were identified using the KPNC HIV registry, and demographic data were collected from the electronic health record. Complete data were not available on transgender identity or HIV-transmission risk factor.

We identified 235 HIV seroconverters and 972 PrEP users (Table 1). Seroconverters were significantly younger than PrEP users (mean age = 34.9 vs 37.9 years; P < .001), with 37.9% of seroconverters younger than 30 years compared with 25.7% of PrEP users. Nearly all seroconverters and PrEP users were male (98.3% vs 97.9%, P = .73). There were differences between HIV seroconverters and PrEP users with respect to race/ethnicity (P < .001); compared with PrEP users, HIV seroconverters were more frequently Black (23.6% vs 4.3%) or Hispanic (19.7% vs 12.2%), and less frequently White (45.9% vs 69.6%).

TABLE 1—

Demographic Differences Between Those With Incident HIV Infection and Those Receiving PrEP: Kaiser Permanente, Northern California

	HIV Seroconverters (July 2009–June 2012)	PrEP Users (July 2012–December 2014)	P
No.	235	972
Age, y, mean (SD)	34.9 (10.6)	37.9 (10.1)	< .001
Age, y, %			.003
≤ 29	37.9	25.7
30–39	30.2	36.0
40–49	22.1	24.8
≥ 50	9.8	13.5
Sex, %			.73
Male	98.3	97.9
Female	1.7	2.1
Race/ethnicity, %			< .001
White	45.9	69.6
Hispanic	19.7	12.2
Asian/Pacific Islander	9.2	10.3
Black	23.6	4.3
Other	1.8	3.6

Note. PrEP = preexposure prophylaxis. P values obtained from χ² test for categorical variables and t-test for continuous variables.

Our results suggest that there may be substantial racial/ethnic and age differences between individuals initiating PrEP and the underlying population at risk for HIV infection, even in a setting with comprehensive access to health care. Strategies are critically needed to increase PrEP use in key HIV risk groups with lower PrEP uptake. Without intervention, the underrepresentation of Blacks, Hispanics, and younger individuals among PrEP users may exacerbate existing demographic disparities in the HIV epidemic.