Foundations for progress: insights from the PrEP Impact Trial supporting change to PrEP delivery and implementation in England Authors: Dana Ogaz, Holly Mitchell, Chiara Chiavenna, Andrea Cartier, James Bell, Rainer Golombek, Andre Charlett, Sajjida Jaffer, Ross Harris, Victor Diamente, Hamish Mohammed, Chris James Ward, John McSorley, O Noel Gill, Ann Sullivan, John Saunders, Monica Desai Published: 2026 ABSTRACT AbstractObjectivesThe HIV pre-exposure prophylaxis (PrEP) Impact Trial demonstrated the feasibility and effectiveness of PrEP in England, providing critical evidence to inform national commissioning. Using trial data, we assess regional variation in delivery and examine how service provision differences impacted outcomes across the PrEP Prevention Care Continuum (PPCC) (ie, those at risk of HIV acquisition, those eligible for PrEP, PrEP uptake and coverage).MethodsWe assessed PPCC outcomes among HIV-negative men who have sex with men (MSM) attending trial sexual health services (SHS) from October 2017 to February 2020. Outcomes were stratified by SHS region (London, Outside London) and MSM throughput, defined as the mean annual number of MSM attendees, to approximate differences in service structure and provision based on attendee composition. HIV incidence per 100 person-years was calculated for SHS in and outside of London, restricted to those with ≥2 visits during the study period.ResultsAcross 157 trial SHS, 165 270 MSM attended during the study period, of whom 20 349 were enrolled in the trial. HIV incidence was calculated among 102 842 MSM, including 17 770 trial participants. PrEP uptake ranged from 42% to 92%, and coverage from 16% to 31%, varying by MSM throughput strata and consistently higher among London SHS. HIV incidence was significantly lower in trial participants (London: 0.07 (95% CI 0.04 to 0.14); Outside London: 0.22 (0.13 to 0.36)) versus non-trial attendees (London: 0.98 (0.88 to 1.10); Outside London: 0.93 (0.82 to 1.04)).ConclusionsThis analysis supports ongoing enhancements to PrEP delivery across England. Findings highlight the success of varied service models, including those not traditionally focused on MSM populations. High HIV seroconversions among individuals without clear markers of risk for HIV acquisition support the need for broader, less restrictive PrEP access, aligned with recent updates to national guidance. To evaluate the long-term impact of PrEP on HIV and sexually transmitted infection (STI) incidence, consistent, high-quality data reporting to national surveillance remains essential. WHAT IS ALREADY KNOWN ON THIS TOPIC Outcomes from the HIV pre-exposure prophylaxis (PrEP) Impact Trial, an implementation trial of PrEP across 157 sexual health services to inform routine PrEP commissioning in England, have demonstrated substantial national PrEP need. A routine PrEP programme across England has been in place since October 2020. Outcomes from the HIV pre-exposure prophylaxis (PrEP) Impact Trial, an implementation trial of PrEP across 157 sexual health services to inform routine PrEP commissioning in England, have demonstrated substantial national PrEP need. A routine PrEP programme across England has been in place since October 2020. WHAT THIS STUDY ADDS Using data collected from the PrEP Impact Trial, we describe previously unexamined regional and sexual health service-based breakdowns of the PrEP Prevention Care Continuum (PPCC), which includes PrEP eligibility, uptake and coverage. Understanding how PrEP access and outcomes differ across regions and service models is essential to evaluate impact, benchmark progress and inform targeted improvements as routine PrEP availability continues. Using data collected from the PrEP Impact Trial, we describe previously unexamined regional and sexual health service-based breakdowns of the PrEP Prevention Care Continuum (PPCC), which includes PrEP eligibility, uptake and coverage. Understanding how PrEP access and outcomes differ across regions and service models is essential to evaluate impact, benchmark progress and inform targeted improvements as routine PrEP availability continues. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Findings from these analyses highlight the success of PrEP implementation across a range of service models including those that may not be tailored for men who have sex with men. High HIV incidence across non-trial attendees with and without markers of higher risk for HIV acquisition supports the need to re-examine the role of clinical assessment, including clinical history, in PrEP planning and delivery. Contemporary evaluation of delivery strategies across sexual health services and implementation progress against PPCC outcomes will need to be reassessed following major periods of disruption. Findings from these analyses highlight the success of PrEP implementation across a range of service models including those that may not be tailored for men who have sex with men. High HIV incidence across non-trial attendees with and without markers of higher risk for HIV acquisition supports the need to re-examine the role of clinical assessment, including clinical history, in PrEP planning and delivery. Contemporary evaluation of delivery strategies across sexual health services and implementation progress against PPCC outcomes will need to be reassessed following major periods of disruption. Introduction The HIV pre-exposure prophylaxis (PrEP) Impact Trial was a prospective, open-label, single-arm, multicentre trial carried out across 157 sexual health services (SHS) (ie, sexual health clinics) in England from October 2017 through July 2020.1 The trial demonstrated the feasibility and impact of large-scale PrEP provision in England and directly informed national commissioning, which commenced in October 2020 amidst COVID-19-related social restrictions.2 Over 24 000 participants, mostly men who have sex with men (MSM), were recruited to the trial through free and open access SHS in England.1 3 4 To assess regional variations in PrEP delivery during the trial and to understand how differences in service provision influenced outcomes, we conducted a descriptive analysis of the PrEP Prevention Care Continuum (PPCC), which examined individuals at risk of HIV acquisition, those eligible for PrEP and PrEP uptake and coverage. We present PPCC measures across SHS regions (London, Outside London) and by service measures based on SHS throughput of attending MSM (herein, ‘MSM throughput strata’). Here, strata serve as a proxy for differences in service structure and delivery models, where we assumed that services with higher MSM throughput were more likely to offer specialised PrEP provision, resulting in differing levels of uptake, compared with lower throughput services. These analyses complement previously published trial findings and provide a valuable benchmark for PPCC outcomes prior to the introduction of routine commissioning and the major disruptions brought about by the COVID-19 pandemic. Methods Full trial inclusion criteria, methods and primary outcomes measures have been previously described.1 We provide a brief overview below. Study methods and participants Trial participants included HIV-negative individuals aged ≥16 who were clinically assessed as eligible for PrEP (eg, no contraindications, willing to adhere to study procedures)5 attending participating SHS from 13 October 2017 to 12 July 2020. Non-trial attendees (NTA, those who never joined the trial) included attendees of participating SHS anytime from the start of respective trial recruitment through the study period. Trial data were linked, with informed consent, to national STI surveillance (GUMCAD STI Surveillance System),6 where SHS activity and outcomes were prospectively examined. NTA represented SHS attendees who were not recruited into the Impact Trial at any stage and continued to receive standard care through SHS. To minimise the impact of SHS access disruptions caused by COVID-19 on the assessment of study outcomes, we analysed trial data from 13 October 2017 through 29 February 2020 (online supplemental appendix I). All analyses were carried out in MSM trial participants and NTA as this group accounted for most PrEP uptake throughout the trial. For context, the English population based on 2021 census estimates was at nearly 56.5 million usual residents, of which 8.9 million were in the London region of England; about 1 million men in England and Wales identified as gay or bisexual.7 8 SHS region and MSM throughput We defined MSM throughput as the number of MSM attending a SHS. Mean annual MSM attendance at each SHS was used to group services into mutually exclusive strata assumed to reflect service models based on composition of SHS attendees. MSM throughput strata were derived from terciles of the mean yearly number of MSM attending each participating SHS (classified as high, medium or low), separately for services inside and outside of London during the study period. One London-based SHS, which was a clear outlier in terms of volume and composition of attendees, was designated as a separate ‘highest throughput’ stratum (see online supplemental appendices II and III). These strata broadly represented differences in the proportion of MSM SHS attendees across varying service sizes, and similar constructs have been used previously to inform trial place allocation, commissioning estimates3 and analyses of HIV diagnosis reductions in England.9 Strata were calculated separately for services in London and those Outside London to account for structural and attendee demographic differences. London has a larger population of MSM, higher density of SHS attendees and number of services which may differ substantially from regions elsewhere in England10 11 (note, this rationale also applies to regional stratification for PPCC and HIV incidence analyses below). PrEP uptake, coverage and the PPCC Those considered at risk of HIV acquisition included attendees recorded as PrEP eligible (had a record of PrEP eligibility, offer or prescription coding during the study period), or, to account for the under-report of PrEP eligibility coding, the record of clinical and behavioural markers associated with higher risk of acquiring HIV (based on coding in the GUMCAD STI Surveillance System) (markers of higher risk, MoHR).12 These included having a rectal bacterial STI diagnosis, use of HIV postexposure prophylaxis, being a sex worker, being in contact with someone diagnosed with HIV or syphilis (attending as a consequence of partner notification) in the 12 months prior to and during the analysis period. PPCC outcomes included the following ever reported during the study period: the number of HIV-negative attendees to participating trial SHS (with at least one visit during the trial period), attendees ever at risk of HIV acquisition (ever PrEP-eligible or with MoHR), attendees ever PrEP eligible, attendees enrolled to the trial, and the percentage of PrEP uptake and coverage. PrEP uptake is reported as the percentage of trial participants who took up PrEP among those considered PrEP eligible; PrEP coverage was reported as the percentage of trial participants who took up PrEP among those considered at risk of HIV acquisition. We report PPCC outcomes regionally and by throughput strata (online supplemental appendix III). HIV incidence HIV incidence was calculated among trial attendees with at least one post enrolment SHS visit, and among NTA, those with at least two visits to a participating SHS from respective recruitment start through the end of the study period (ie,. among all with at least two visits through the study period and not known to be HIV-positive at first visit) (online supplemental appendix III). Follow-up time was calculated from enrolment or first visit with follow-up ending on date of first positive HIV test. Attendees were censored at trial discontinuation (trial participants), or last SHS visit on or before 29 February 2020 (trial participants, NTA) in the absence of known HIV acquisition. We present incidence rates per 100 person-years and associated 95% CI in London and Outside London to minimise deductive disclosure.13 We conducted sensitivity analyses restricting to NTA ever at risk of HIV acquisition (see online supplemental appendix III, see definition above). Data were managed and analysed at the UK Health Security Agency (formerly Public Health England) using STATA V.13 or higher. The study is registered with ClinicalTrials.gov, NCT03253757. Results From 13 October 2017 to 29 February 2020, 165 270 MSM had at least one SHS visit among participating trial SHS; 20 349 MSM were enrolled to the trial (table 1, online supplemental appendix I). Sociodemographic and clinical information of trial and NTA has been previously described.1 Table 1The PrEP PPCC among MSM in participating trial SHS by region and throughput strata, October 2017–February 2020SHS attendees (row %)Ever at risk of HIV acquisition (row %)Ever eligible (row %)Enrolled (row %)UptakeCoverageAll MSM attendees*165 270 (100%)94 279 (51%)34 880 (21%)20 349 (12%)58%22%SHS region† London74 472 (100%)45 478 (61%)16 427 (22%)11 600 (16%)71%26% Outside London90 798 (100%)48 801 (54%)18 453 (20%)8749 (10%)47%18%SHS region†Outside London Midlands and East of England29 853 (100%)14 810 (50%)4949 (17%)2246 (8%)45%18% North of England32 797 (100%)18 028 (55%)6471 (20%)3280 (10%)51%18% South of England28 148 (100%)15 963 (57%)7033 (25%)3223 (11%)46%20%SHS throughput strata‡Highest throughput27 321 (100%)17 824 (65%)5889 (22%)5438 (20%)92%31%London High throughput tercile24 146 (100%)13 769 (57%)4886 (20%)2635 (11%)54%19% Medium throughput tercile14 657 (100%)8785 (60%)3192 (22%)2057 (14%)64%23% Low throughput tercile8348 (100%)5100 (61%)2460 (29%)1470 (18%)60%29%Outside London High throughput tercile38 370 (100%)21 188 (55%)8092 (21%)3403 (9%)42%16% Medium throughput tercile28 098 (100%)14 608 (52%)4720 (17%)2341 (8%)50%16% Low throughput tercile24 330 (100%)13 005 (53%)5641 (23%)3005 (12%)53%23%Percentage of SHS attendees. Ever at risk of HIV acquisition: PrEP eligibility, offer or prescription codes or MoHR from follow-up start to end of February 2020. Ever eligible: ever coded as eligible for PrEP. Uptake: enrolled/ever eligible. Coverage: enrolled/ever at risk of HIV acquisition.*HIV negative attendees accessing a participating SHS at least once after recruitment at that SHS had begun and before the end of February 2020, with a recorded HIV test and no evidence of obtaining PrEP from another source; excludes 44 trial participants with no record of receiving a PrEP prescription during the analysis period, and 10 trial participants who likely acquired HIV prior to enrolment and use of trial PrEP.†SHS region based on location postcode.‡SHS throughput strata based on terciles (high, medium, low) of mean yearly SHS throughput by MSM within participating SHS in and outside of London during the study period; highest throughput is highest national mean MSM throughput.MoHR, markers of higher risk; MSM, men who have sex with men; PPCC, PrEP Prevention Care Continuum; PrEP, pre-exposure prophylaxis; SHS, sexual health service. The PrEP PPCC among MSM in participating trial SHS by region and throughput strata, October 2017–February 2020 Percentage of SHS attendees. Ever at risk of HIV acquisition: PrEP eligibility, offer or prescription codes or MoHR from follow-up start to end of February 2020. Ever eligible: ever coded as eligible for PrEP. Uptake: enrolled/ever eligible. Coverage: enrolled/ever at risk of HIV acquisition. HIV negative attendees accessing a participating SHS at least once after recruitment at that SHS had begun and before the end of February 2020, with a recorded HIV test and no evidence of obtaining PrEP from another source; excludes 44 trial participants with no record of receiving a PrEP prescription during the analysis period, and 10 trial participants who likely acquired HIV prior to enrolment and use of trial PrEP. SHS region based on location postcode. SHS throughput strata based on terciles (high, medium, low) of mean yearly SHS throughput by MSM within participating SHS in and outside of London during the study period; highest throughput is highest national mean MSM throughput. MoHR, markers of higher risk; MSM, men who have sex with men; PPCC, PrEP Prevention Care Continuum; PrEP, pre-exposure prophylaxis; SHS, sexual health service. PrEP Prevention Care Continuum PPCC outcomes were variable across SHS strata (table 1). Uptake and coverage among participants to London SHS (71% (11 600/16 427), 26% (11 600/45 478), respectively) were higher compared with those Outside London (47% (8749/18 453), 18% (8749/48 801)). The proportion of SHS attendees ever at risk of HIV acquisition was higher in London SHS relative to Outside London (61% (45 478/74 472) vs 54% (48 801/90 798)). Higher MSM throughput SHS generally had lower measures of uptake and coverage compared with lower throughput SHS (see table 1) (eg, 54% and 19% in London high throughput, 64% and 23% in London middle throughput, 60% and 29% in London low throughput), with the exception of the highest national stratum. Here, PrEP uptake was 92% (5438/5889) and coverage was 31% (5438/17 824). HIV incidence There were 102 842 MSM included in HIV incidence analyses (17 770 trial participants (19 075 person-years); 85 072 NTA (61 605 person-years)). Total follow-up time was higher in trial participants enrolled to London SHS (12 124 person-years; median time per participant: 434 person-days) than those Outside London (6951 person-years; median time per participant: 273 person-days). There were comparable levels of follow-up time among NTA (London: 30 060 person-years; median time per participant: 241 person-days; Outside London: 31 545 person-years; median time per participant: 183 person-days). Among MSM attending London SHS, there were 304 HIV seroconversions (9 impact participants, 295 NTA) and 307 HIV seroconversions among those Outside London (15 impact participants, 292 NTA). HIV incidence rates per 100 person-years were lower among trial participants relative to NTA (London: 0.07 (95% CI: 0.04 to 0.14) vs 0.98 (0.88 to 1.10); Outside London: 0.22 (0.13 to0.36) vs 0.93 (0.82 to 1.04)) (table 2A); HIV incidence among NTA in either region was broadly similar. We calculated a proportionate HIV reduction of 92.4% (85.4%–96.6%) among SHS in London and 76.7% (60.9%–87.1%) Outside London. Table 2HIV incidence rate in MSM impact trial participants and all MSM non-trial attendees(A) All trial participants, all non-trial attendeesSHS regionAll MSMLondonOutside LondonImpact participantNon-trial attendeeImpact participantNon-trial attendeeImpact participantNon-trial attendeePerson-years follow-up time19 07561 60512 12430 060695131 545Median follow-up per individual (IQR)(person-days)368 (146–643)205 (56–433)434 (168–684)241 (77–488)273 (113–587)183 (43–390)No. HIV seroconversions24587929515292Incidence rate per 100 person-years (95% CI)0.12 (0.08–0.19)0.95 (0.88–1.03)0.07 (0.04–0.14)0.98 (0.88–1.10)0.22 (0.13–0.36)0.93 (0.82–1.04)Rate difference per 100 person-years (95% CI)0.83 (0.73–0.92)0.91 (0.79–1.03)0.71 (0.56–0.86)Proportionate reduction (95% CI)86.8% (83.9%–91.6%)92.4% (85.4%–96.6%)76.7% (60.9%–87.1%) (B) All trial participants, non-trial attendees ever at risk of HIV acquisition (sensitivity analysis)SHS regionAll MSMLondonOutside LondonImpact participantNon-trial attendeeImpact participantNon-trial attendeeImpact participantNon-trial attendeePerson-years follow-up time19 07558 83312 12426 292695126 541No. HIV seroconversions24389921415175Incidence rate per 100 person-years (95% CI)0.12 (0.08–0.19)0.74 (0.67–0.81)0.07 (0.04–0.14)0.81 (0.71–0.93)0.22 (0.13–0.36)0.66 (0.57–0.76)Rate difference per 100 person-years (95% CI)0.61 (0.52–0.70)0.74 (0.62–0.86)0.44 (0.30–0.59)Proportionate reduction (95% CI)82.9% (74.2%–89.2%)90.9% (82.4%–95.9%)67.3% (44.5%–82.1%)Incidence estimates based on HIV-negative attendees accessing a participating SHS with at least one follow-up attendance after enrolment (impact participants) or at least two visits after recruitment at that SHS had started (non-trial attendees) to the end of February 2020, with a recorded HIV test and no evidence of obtaining PrEP from another source; excludes 44 trial participants with no record of receiving a PrEP prescription during the analysis period, 10 trial participants who likely acquired HIV prior to enrolment and use of trial PrEP and 62 428 individuals with only one SHS attendance (2579 trial participants; 59 849 non-trial attendees). Ever at risk of HIV acquisition: PrEP eligibility, offer or prescription codes or MoHR from follow-up start to end of February 2020.MoHR, markers of higher risk; MSM, men who have sex with men ; PrEP, pre-exposure prophylaxis; SHS, sexual health service. HIV incidence rate in MSM impact trial participants and all MSM non-trial attendees Incidence estimates based on HIV-negative attendees accessing a participating SHS with at least one follow-up attendance after enrolment (impact participants) or at least two visits after recruitment at that SHS had started (non-trial attendees) to the end of February 2020, with a recorded HIV test and no evidence of obtaining PrEP from another source; excludes 44 trial participants with no record of receiving a PrEP prescription during the analysis period, 10 trial participants who likely acquired HIV prior to enrolment and use of trial PrEP and 62 428 individuals with only one SHS attendance (2579 trial participants; 59 849 non-trial attendees). Ever at risk of HIV acquisition: PrEP eligibility, offer or prescription codes or MoHR from follow-up start to end of February 2020. MoHR, markers of higher risk; MSM, men who have sex with men ; PrEP, pre-exposure prophylaxis; SHS, sexual health service. In sensitivity analyses, there were 389 seroconversions among all NTA (table 2B) with HIV incidence per 100 person-years more variable 0.81 (0.71–0.93) in London and 0.66 (0.57–0.76) Outside London (table 2B). Discussion This analysis provides evidence to inform and support changes in PrEP delivery and implementation across England. PPCC examinations by MSM throughput, which may reflect differing service structures and provision, found higher levels of PrEP uptake and coverage in SHS with lower MSM throughput. Findings suggest that a broad range of service models, including those not traditionally MSM-focused, can support robust PrEP delivery. The high number of HIV seroconversions among NTA, with and without MoHR, supports recommendations for less restrictive access for individuals who could benefit from PrEP, which has been proposed in contemporary PrEP guidance14 and service models. During the trial, differential PrEP uptake and coverage was likely influenced by a range of factors. Fluctuating trial place availability at the beginning of trial recruitment and early pauses in many high MSM throughput SHS due to high trial place demand15 may have displaced enrolment to lower MSM throughput SHS. We used this measure to stratify by potential differences in service delivery, using an assumption that higher MSM throughput SHS likely operated tailored service models facilitating PrEP uptake for MSM.16 However, higher uptake and coverage across lower MSM throughput strata (excluding the highest national throughput stratum) demonstrate a success in implementation across a range of service models. Lower throughput SHS may have benefited from consistent trial place availability and clinician engagement, suggesting that flexible and sustained resourcing is critical for equitable PrEP access. Since the start of routine commissioning, PrEP has been acknowledged as a key component of HIV combination prevention.17 18 However, individual-level and structural-level barriers continue to contribute to the complexity of PrEP uptake and delivery.19 Implementation progress will need to be reassessed following major periods of disruption caused by COVID-19 and STI surveillance enhancements, which include integration of behavioural data.20 This is of urgent need, especially as consistent declines in new HIV diagnoses in national surveillance have stalled through 2024,11 despite routine availability of PrEP. Trial findings reflect the success of embedding PrEP provision within SHS, where, since commissioning, over 80 000 individuals, primarily MSM, have accessed PrEP. Service delivery models used through the trial and into commissioning have facilitated PrEP uptake in settings designed to minimise stigma, create opportunities to integrate health promotion and uptake of other preventative interventions, and have enabled robust monitoring and evaluation through national surveillance systems. Going forward, contemporary and regionally disaggregated assessments of PPCC are needed to evaluate the impact of routine commissioning and to inform the optimisation of equitable delivery at a local level. Strengths and limitations This analysis represents one of the largest assessments of PrEP implementation, and the first to examine Impact Trial outcomes at a service level in England. Although the trial concluded in 2020, these findings reflect PrEP implementation prior to significant disruptions caused by the COVID-19 pandemic and provide valuable regional insights to inform HIV elimination progress in England. However, this study does have further limitations.1 Measures of uptake and coverage are highly dependent on clinic-level data quality which varied across trial SHS and may have affected service-level comparability. To address potential under-reporting of eligibility, markers of sexual risk were used to supplement estimates of PrEP eligibility and need. However, as markers were based on prior clinical history, they were subject to data quality and could potentially overestimate coverage estimates in clinics with poorer data quality or where participants newly attended. Measures of risk of HIV acquisition were based on coding within national surveillance, which, as a whole, could have underestimated PrEP eligibility given lack of behavioural data at the time of the trial.20 Although there were few indications of trial SHS transfer among participants (<10%), we may have underestimated trial participant movement between services and misclassified individuals without markers of risk (which may have had higher impacts on lower throughput SHS). Additional bias may have resulted from the inability to track NTA across SHS. The small number of HIV seroconversions through the study period limited disaggregate examinations of incidence; however, seroconversions across English regions were proportionately reflective of new HIV diagnoses reported in national surveillance from 2017 to 2020.21 Most seroconversions reported in the Impact Trial were linked to low self-reported adherence, with only one unconfirmed potential breakthrough HIV infection identified within the trial.22 Conclusions Our findings reinforce the need to adapt PrEP implementation strategies to ensure more equitable and efficient delivery across England. Successes in lower MSM throughput SHS show that PrEP can be effectively delivered through diverse service models. Consistent and high-quality data reporting as part of routine monitoring and evaluation23 (ie, consistently applied clinical coding, timely submissions of surveillance data) is integral to assessing the long-term impacts of PrEP on STI and HIV incidence. As England advances towards HIV elimination, optimising PPCC outcomes across all SHS, especially in higher throughput SHS, will require flexible, locally responsive care models. While PrEP Impact Trial findings lay a foundation for continued progress, future success depends on reducing regional disparities, ensuring access and sustaining effective delivery. Supplementary material