Appropriate usage of post-exposure prophylaxis-in-pocket for HIV prevention by individuals with low-frequency exposures

Abstract

PEP-In-Pocket (Post-Exposure Prophylaxis-In-Pocket, or “PIP”) is a biobehavioural HIV prevention strategy wherein patients are proactively identified and given a prescription for HIV post-exposure prophylaxis (PEP) medications to self-initiate in case of high-risk exposures. We evaluated this strategy in a prospective observational study at two hospital-based clinics in Toronto, Canada. HIV-negative adults using PIP underwent chart review and completed quarterly electronic questionnaires over 12 months. The primary objective was to quantify appropriate PIP initiation, defined as starting PIP within 72 h of a high-risk exposure. Secondary objectives were to quantify HIV seroconversions, changes in sexual risk behaviour, sexual satisfaction, and satisfaction with the PIP strategy. From 11/2017 to 02/2020, 43 participants enrolled and completed ≥1 questionnaire. PIP was self-initiated on 27 occasions by 15 participants, of which 24 uses (89%) were appropriate, 2 were unnecessary, and 1 was for an unknown exposure. Chart review identified no inappropriate non-use. Over 32 person-years of testing follow-up, we observed zero HIV seroconversions. Sexual risk declined modestly over follow-up, with a HIRI-MSM (HIV Incidence Risk Index for MSM) change of −0.39 (95% CI = −0.58, −0.21 per 3 months, p < .001). Sexual satisfaction was stable over time. At 12 months, 31 (72%) remained on PIP, 8 (19%) had transitioned to pre-exposure prophylaxis and 4 (9%) were lost-to-follow-up. Among participants who remained on PIP and completed questionnaires at 12 months, 24/25 (96%) strongly/somewhat agreed that PIP decreased their anxiety about contracting HIV and 25/25 (100%) strongly/somewhat agreed that they would recommend PIP to a friend. PIP is a feasible HIV prevention strategy in carefully selected individuals at modest HIV risk.

Background

Despite the development of new HIV prevention strategies, incident HIV infections persist in Canada, with gay, bisexual, and other men who have sex with men (GBM) remaining disproportionately affected. ¹ While daily oral pre-exposure prophylaxis (PrEP) is an effective prevention modality for those at increased risk of contracting HIV, ² it may result in unacceptable cost, ³ inconvenience due to need for chronic dosing, tolerability issues, or risk of renal/bone toxicity^4,5 in those at modest risk. On-demand oral PrEP (also known as “event driven” or “2-1-1” PrEP) is also highly effective but is supported by a smaller body of evidence,^6–8 particularly for those with very infrequent exposures. ⁹ Further, on-demand oral PrEP requires individuals to initiate medications prior to an anticipated exposure. Additional prevention methods are needed to ensure that a range of options is available to meet the needs of diverse individuals at risk, including those with less frequent exposures.

For individuals at sporadic risk of HIV infection, in whom exposures are infrequent or unanticipated, PEP-In-Pocket, or “PIP,” is an alternative strategy wherein patients are provided with a prescription for 28 days’ worth of post-exposure prophylaxis (PEP) medications to self-initiate in the event of an exposure. Evidence supporting the efficacy of PEP in preventing HIV infection comes from numerous observational ¹⁰ and non-human primate ¹¹ studies. The PIP strategy allows immediate access to PEP, and avoids the need to urgently seek care at an emergency department (ED) or walk-in clinic, potentially alleviating anxiety and providing autonomy over one’s care. In contrast to on-demand PrEP, PIP is a reactive strategy, wherein individuals initiate medications after an exposure (as soon as possible, and within 72 h). This approach may offer increased flexibility for individuals in whom exposures are infrequent or unanticipated (e.g. unplanned sexual encounter, condom breakage, etc.)

We have previously published our preliminary clinical experience with PIP,^12,13 wherein we found no HIV seroconversions over 97.3 patient-years of follow-up using this strategy. However, the retrospective nature of those reports precluded evaluation of whether PIP was being used appropriately. We therefore sought to prospectively evaluate how patients were using the PIP strategy in this report.

Methods

Study design and participants

Consenting HIV-negative adults were eligible for inclusion in this 12-month prospective observational cohort study if they were prescribed PIP by a clinician at one of two hospital-based HIV clinics in Toronto, Canada. The decision to use PIP was made through shared decision-making on clinical grounds alone, and eligible participants were prospectively identified by trained research staff through chart review.

Our primary objective was to determine the proportion of participants who used the PIP strategy appropriately, meaning that PIP was either not used at all due to the absence of HIV risk behaviour, or was consistently initiated within 72 h of any high-risk sexual exposure. High-risk exposures were in turn defined as episodes of anal or vaginal sex in which a condom broke or was not used, with a partner whose HIV status was either unknown or known to be positive (except if known to have an undetectable HIV viral load), or if injection drug paraphernalia was shared, in accordance with clinical guidelines. ¹⁴

Secondary objectives were to quantify: (a) the incidence of new HIV infection among study participants, (b) changes in sexual risk-taking behaviour, (c) sexual satisfaction, and (d) satisfaction with the PIP strategy.

We obtained ethical approval for this study from the Research Ethics Boards of St. Michael’s Hospital and University Health Network. All study participants provided written informed consent prior to the initiation of any study activities.

Provision of PIP

Clinical decisions about the provision of PIP, including the specific regimen, were made through shared decision-making between participants and their clinical providers. Prior to deciding to use PIP, participants received information about all available HIV prevention options, including condoms, daily PrEP, on-demand PrEP, and routine PEP use, in accordance with the standard of care. Individuals who selected a PIP strategy were provided with a prescription for a standard of care PEP regimen (typically tenofovir disoproxil fumarate 300 mg coformulated with emtricitabine 200 mg daily, together with dolutegravir 50 mg daily) and instructed to begin PIP immediately (and no later than 72 h) following an eligible high-risk exposure, as defined above. Participants received an informational pamphlet outlining these clinical criteria for PIP initiation, and were instructed to immediately contact their clinic to arrange appropriate follow up and testing if they initiated PIP. If clinic visits were missed, study personnel contacted participants up to three times to remind them to reschedule. Participants could change their preferred HIV prevention modality at any time during this study. Clinical follow-up and screening for HIV and sexually transmitted infections was arranged by clinical providers as per their usual practices for patients receiving PIP.

Data collection

Participants were asked to complete an electronic questionnaire at enrollment and then at 3-monthly intervals over 12 months of follow up. The baseline questionnaire addressed demographics, current sexual risk behaviour, and sexual satisfaction. Because the majority of participants were anticipated to be GBM, sexual risk behaviour in GBM participants was measured using the HIRI-MSM (HIV Incidence Risk Index for MSM), ¹⁵ a validated score predictive of incident HIV in a cohort of HIV-seronegative GBM. Sexual satisfaction was measured according to a 20-item sexual satisfaction survey which has been validated in large American and Croatian cohorts, including a cohort of non-heterosexual men and women. ¹⁶

Follow-up electronic questionnaires assessed PIP usage, sexual risk behaviours, sexual satisfaction, and satisfaction with the patient pamphlet and the PIP strategy itself. Participants were given $10 Canadian dollars in compensation for the completion of each study questionnaire.

Trained study personnel reviewed participants’ medical charts quarterly, in order to identify when visits were missed (to trigger reminder calls) and to extract data on HIV testing. Participants who were lost to follow up, transitioned to another HIV prevention strategy, or did not complete follow up questionnaires were excluded from further analysis after completion of their final questionnaire.

Statistical considerations

In our prior clinical experience with PIP,^12,13 we were not aware of any inappropriate use or non-use of PEP, so we conservatively assumed that the true proportion of appropriate use over the planned follow up period of 12 months may lie between 0.80 and 1.0. Assuming a normal distribution and a desired confidence interval of 0.10, we targeted a sample size of 35 participants to allow reasonable estimation of the true proportion of appropriate versus inappropriate PIP usage.

We characterized the study population using frequencies and proportions for categorical variables, and measures of central tendency and dispersion for continuous variables. Appropriate versus inappropriate use of PIP was ascertained from questionnaire responses to determine the primary outcome. HIV incidence was calculated from the number of infections observed divided by total accrued person-time during follow up. Mean HIRI-MSM Risk Index scores were calculated for participants at baseline and at 3, 6, 9, and 12 months. Sexual satisfaction scores were calculated at baseline, 6 and 12 months. To test for changes in HIRI-MSM and sexual satisfaction scores over time, we used linear mixed-effects models (STATA’s mixed command), with individual identifier as the group variable, and using maximum likelihood and an independent correlation structure. Normality of residuals and homoscedasticity were assessed by visual inspection. We compared models with random intercept only versus random intercept and random slopes with the likelihood ratio test. Statistical analysis was performed using STATA Version 16.1.

Results

Between November 2017 and February 2020, 43 participants enrolled and completed at least 1 questionnaire. Median age was 36 (IQR, interquartile range, 28, 48) years and all but one was GBM (Table 1). Half (52%) had used PEP prior to enrollment in the study, and the majority (84%) had either public or private drug insurance, such that PrEP medications would have been readily accessible to them as an alternative if desired.

Table 1.

Participant characteristics at baseline (n = 43).

Characteristic	n (%)
Median age in years (interquartile range)	36 (28, 48)
Gender
Cisgender man	42 (98%)
Transgender woman	1 (2%)
Sexual orientation (n = 42)
Gay	37 (88%)
Bisexual	4 (10%)
Pansexual	1 (2%)
Education (n = 42)
High school	4 (10%)
College or undergraduate degree	17 (41%)
Graduate or professional degree	21 (50%)
Ethnicity (n = 41)
White European	18 (44%)
Mixed heritage	2 (5%)
Asian East	9 (20%)
Asian South East	2 (5%)
Black Caribbean	1 (2%)
Indian Caribbean	1 (2%)
Latin American	4 (10%)
Middle Eastern	4 (10%)
Yearly income, CAD (n = 42)
$0–29,999	9 (21%)
$30,000–59,999	11 (26%
$60,000–70,999	7 (17%)
$80,000–99,000	7 (17%)
$100,000+	8 (19%)
Diagnosed with an STI in the past 6 months (n = 42)	3 (7%)
Prior PEP use (n = 42)
Never	20 (48%)
Once	20 (48%)
Twice or more	2 (5%)
Medication insurance
Private	29 (67%)
Public	7 (16%)
None/out of pocket	7 (16%)

Over 32 patient-years of follow-up, 15 (35%) participants initiated PIP on 27 occasions, including five participants who used PIP twice, two who used it three times, and one who used it four times. Of all 27 PIP initiations, 24 (89%) were deemed appropriate (15 episodes of receptive anal sex, either condomless or when a condom broke, 9 episodes of insertive anal sex, either condomless or when a condom broke), 2 were unnecessary (receptive anal sex with a condom; pre-exposure use when an anticipated encounter did not ultimately happen), and 1 was for an unknown exposure. There was no inappropriate non-use identified based on chart review. Of the 43 participants, 40 (93%, 95% CI, confidence interval = 81%, 99%) exhibited only appropriate use or non-use of PIP during follow-up.

Over 32 person-years of testing follow up, we observed zero HIV seroconversions.

HIRI-MSM (HIV Incidence Risk Index for men who have sex with men) scores declined from a median of 15^10,21 at baseline to 8^6,15 at 12 months, giving an average change per 3-months interval of −0.39 (95% CI: −0.58 to −0.21, p < .001). Sexual satisfaction scores (whose possible values range from 20 to 100) did not significantly change over the study period, with a median score of 65 (55–73) at baseline, and 61.5 (44–72) at 12 months, giving an average change per 6 months of −0.55 (95% CI: −1.11 to 0.01, p = .052).

At 12 months, 31 (72%) participants remained on PIP, 8 (19%) had transitioned to PrEP, and 4 (9%) were lost-to-follow-up. Transitions to PrEP were on the basis of patient-reported increases in frequency of risk behaviours or patient preferences, such that PIP was no longer an appropriate strategy. Among participants who remained on PIP and completed questionnaires at 12 months, 24/25 (96%) strongly/somewhat agreed that PIP decreased their anxiety about contracting HIV and 25/25 (100%) strongly/somewhat agreed that they would recommend PIP to a friend. When asked how PIP compared to other HIV prevention strategies, 11/25 (44%) agreed with the statement, “I prefer PIP to PrEP because I don’t have to remember to take a pill every day”, 10/25 (40%) indicated “I prefer PIP to PrEP because it saves me money to only take the medication when I need it” and 16/25 (64%) indicated “I prefer PIP to PEP because if I need it, I don’t have to wait in the Emergency Department”. None of these 25 participants reported preferring a strategy other than PIP.

Discussion

This prospective study of PEP-In-Pocket highlights its feasibility as an additional HIV prevention approach in carefully selected individuals. In the majority of cases, our participants were able to successfully determine whether an exposure was high risk and initiate PIP appropriately. The rare episodes of inappropriate PIP use that we observed involved being overly conservative (i.e. involving unnecessary use of medication) rather than inappropriately failing to self-initiate the intervention when clinically indicated. In accordance with this overall appropriate medication use, we observed no HIV seroconversions among any study participants during follow up. Importantly, PIP was acceptable to participants, the majority of whom reported satisfaction with the strategy and stable levels of sexual satisfaction during follow-up.

There was a fair rate of retention among participants using the PIP strategy, with 72% of patients continuing on PIP at 1 year of follow up. This is a higher rate of retention than reported in some longitudinal studies of individuals receiving PrEP.^17–19 During our study period, 19% of participants transitioned to PrEP, usually on the basis of increased risk behaviour, highlighting the flexibility of this strategy to adapt to the dynamic nature of HIV risk over time. Among participants retained on the PIP strategy, we did not observe risk compensation (increase in frequency of high risk behaviours); in fact, there was a statistically significant decline in HIRI-MSM risk scores over time. The explanation for this observed decline is unclear but may be attributable to participants with greater frequencies of high risk behaviours preferentially transitioning to PrEP. Social distancing directives issued in the context of the COVID-19 pandemic may also have contributed, ²⁰ given that the final questionnaire responses were collected in October 2020.

PIP expands the range of prevention options available to persons at risk of HIV. While daily and on-demand oral PrEP are effective HIV prevention strategies, they can be associated with risks of kidney dysfunction, ⁴ decreased bone mineral density, ⁵ gastrointestinal intolerance, inconvenience from high pill burden, and significant cost. ³ Compared to traditional models of providing PEP, PIP has been associated with a shorter time interval between potential HIV exposure and ingestion of the first dose of medication, an important predictor of HIV prevention success. ²¹ For individuals at lower-to-modest risk of HIV infection due to infrequent high-risk exposures, PIP provides an acceptable and feasible alternative which mitigates these risks. Given the considerable number of participants who did not need to self-initiate PEP at all, PIP may be associated with considerable medication cost savings, particularly when compared to daily PrEP.

Strengths of our study include the prospective collection of sexual behaviour data to contextualize the appropriate use of PIP in our sample, and the use of a pamphlet to standardized information about indications for PIP to facilitate patient decision-making.

There are also limitations that warrant mention. First, our sample size was modest, limiting the precision of our estimates. Second, although the majority of participants were retained on either PIP or PrEP at 12 months, 9% were lost-to-follow-up, and it is unclear whether an alternative strategy may have been associated with better persistence. Finally, we may have overestimated the appropriateness of PIP use if our participants under-reported their risk behaviours. However, participants underwent regular assessments by experienced infectious diseases physicians in order to select the most clinically appropriate prevention strategy for each patient over time.

Overall, these findings suggest that PIP is a feasible option for appropriately selected individuals at modest risk of HIV infection. Future work should continue to evaluate its appropriateness, acceptability, HIV prevention impact and other implementation outcomes in larger samples and more diverse populations.

ORCID iD

Darrell HS Tan https://orcid.org/0000-0002-3069-2875