Stopping and restarting PrEP and loss to follow-up among PrEP-taking men who have sex with men and transgender women at risk of HIV-1 participating in a prospective cohort study in Kenya

Abstract

Objective: To assess frequency and predictors of switching between being on and off PrEP and being lost to follow-up (LTFU) among men who have sex with men (MSM) and transgender women (TGW) with access to PrEP services in Sub-Saharan Africa.

Methods: This was a prospective cohort study of MSM and TGW from coastal Kenya who initiated daily oral PrEP from June 2017 to June 2019. Participants were followed monthly for HIV-1 testing, PrEP refill, risk assessment and risk reduction counselling. Follow-up was censored at the last visit before 30 June 2019, or the last HIV-1-negative visit (for those with HIV-1 seroconversion), whichever occurred first. We estimated transition intensities (TI) and predictors of switching: (i) between being off and on PrEP; and (ii) from either PrEP state and being LTFU (i.e. not returning to the clinic for > 90 days) using a multi-state Markov model.

Results: In all, 134 participants starting PrEP were followed for a median of 20.3 months [interquartile range (IQR): 7.7-22.1]. A total of 49 (36.6%) people stopped PrEP 73 times [TI = 0.6/person-year (PY), 95% confidence interval (CI): 0.5-0.7] and, of these, 25 (51.0%) restarted PrEP 38 times (TI = 1.2/PY, 95% CI: 0.9-1.7). In multivariable analysis, stopping PrEP was related to anal sex ≤ 3 months, substance-use disorder and travelling. Restarting PrEP was related to non-Christian or non-Muslim religion and travelling. A total of 54 participants were LTFU: on PrEP (n = 47, TI = 0.3/PY, 95% CI: 0.3-0.5) and off PrEP (n = 7, TI = 0.2/PY, 95% CI: 0.1-0.4). In multivariable analysis, becoming LTFU while on PrEP was associated with secondary education or higher, living in the area for ≤ 1 year, residence outside the immediate clinic area and alcohol-use disorder.

Conclusions: Switching between being on and off PrEP or becoming LTFU while on PrEP was frequent among individuals at risk of HIV-1 acquisition. Alternative PrEP options (e.g. event-driven PrEP) may need to be considered for MSM and TGW with PrEP-taking challenges, while improved engagement with care is needed for all MSM and TGW regardless of PrEP regimen.

Keywords: HIV-1; Kenya; MSM; PrEP; lost to follow-up; predictors.

FIGURE 1

Transitions between pre‐exposure prophylaxis (PrEP) states, and from either pre‐exposure prophylaxis state to being lost to follow‐up (LTFU) among 134 HIV‐1‐negative participants who started PrEP in Kilifi, Kenya, between June 2017 and June 2019. Adm‐Censor, administrative censoring

FIGURE 2

(a, b) Kaplan–Meier failure curve for stopping pre‐exposure prophylaxis (PrEP) for the first time (a) and being lost to follow‐up (LTFU) from either PrEP state (on or off PrEP) (b) among 134 HIV‐1‐negative participants who started PrEP in Kilifi, Kenya between June 2017 and June 2019

FIGURE 3

Transitions between being off pre‐exposure prophylaxis (PrEP) and on PrEP and from each PrEP state to being lost to follow‐up (LTFU) among 134 HIV‐1‐negative participants who started PrEP in Kilifi, Kenya between June 2017 and June 2019. n, number of follow‐up segments between two visits. Black arrows depict transitions from one state to another occurring between follow‐up visits. Non‐black arrows depict individuals remaining in the same state between follow‐up visits. CI, confidence interval; TI, transition intensity; PY, person‐years

FIGURE 4

(a–c) Proportion of any receptive anal intercourse (RAI) (a), insertive anal intercourse (IAI) (b), and < 100% condom use for anal sex (CAS) (c) in the past 3 months among 134 HIV‐1‐negative participants in Kilifi, Kenya, between June 2017 and June 2019. PrEP, pre‐exposure prophylaxis.