iconstar paper   HIV Articles  
Back grey arrow rt.gif
Redefining HIV Preexposure Prophylaxis Failures - PrEP at Kaiser
  Download the PDF here
CID 29 July 2017
To the Editor-Failures of daily oral human immunodeficiency virus (HIV) preexposure prophylaxis (PrEP) with tenofovir disoproxil fumarate/emtricitabine are rare, with only 3 HIV seroconversions documented among patients adherent to PrEP at the time of infection [1-3]. However, PrEP failures could be defined more broadly as HIV infections that occur at any point along the PrEP continuum of care [4]. Here, we describe HIV infections among individuals who sought or were referred for PrEP within the Kaiser Permanente Northern California (KPNC) healthcare system.
KPNC provides comprehensive medical services to 4.1 million members; the KPNC PrEP program has been described previously [5]. In this analysis, we included KPNC members with patient- or provider-initiated referrals for PrEP from July 2012 through February 2017, as identified from outpatient encounter and referral data. Duration of PrEP use was estimated from first pharmacy fill to last day of PrEP in possession, regardless of gaps between fills. HIV infections were identified using the KPNC HIV registry. End of follow-up was the earliest of health plan disenrollment, HIV diagnosis in KPNC, or 28 February 2017. We identified 7124 individuals who sought or were referred for PrEP. Of those, 26 (0.4%) were diagnosed with HIV infection during assessment for PrEP eligibility. Of the remaining 7098 individuals, 4991 (70%) started PrEP and 2107 (30%) did not start PrEP. Of the 2107 who did not start PrEP, 22 were later diagnosed with HIV infection, corresponding with an incidence rate of 1.1 per 100 person-years (95% confidence interval [CI], 0.7-1.7). Of the 4991 who started PrEP, there were no HIV infections during 5104 person-years of PrEP use (mean duration of use, 12.4 months; upper limit of 1-sided 97.5% CI, 0.1). Of 1303 (26%) who no longer had PrEP in possession at the end of follow-up, 11 were diagnosed with HIV infection between the last supply of PrEP and the end of follow-up, corresponding with an incidence rate of 1.3 per 100 person-years (95% CI, 0.8-2.4).
We identified no HIV infections during more than 5000 person-years of PrEP use, consistent with the high adherence previously observed in this setting [5]. However, HIV infections were identified among individuals who were being assessed for PrEP eligibility (ie, late to access PrEP), who sought or were referred for PrEP but did not start (ie, failure to initiate PrEP), or who discontinued PrEP (ie, failure to be retained in PrEP care). Strategies are critically needed to ensure that patients start, restart, or continue PrEP during periods of risk for HIV acquisition.
Julia L. Marcus,1 Leo B. Hurley,2
Dong Phuong Nguyen,3
Michael J. Silverberg,2 and Jonathan E. Volk3
1Harvard Medical School and Harvard Pilgrim Health Care
Institute, Boston, Massachusetts; and 2Kaiser Permanente
Division of Research, Oakland, and 3Kaiser Permanente San
Francisco Medical Center, California

  iconpaperstack View Older Articles   Back to Top   www.natap.org