Article Text

Download PDFPDF

Putting the NHS England on trial: uncertainty-as-power, evidence and the controversy of PrEP in England
  1. Maurice Nagington1,
  2. Tony Sandset2
  1. 1 School of Health Sciences, Faculty of Medicine Biology and Health, University of Manchester, Manchester, United Kingdom
  2. 2 Institute for Interdisciplinary Health Sciences, Faculty of Medicine, University of Oslo, Oslo, Norway
  1. Correspondence to Dr Maurice Nagington, Health Sciences, Faculty of Medicine Biology and Health, University of Manchester, Manchester M13 9PL, UK; maurice.nagington{at}


Pre-exposure prophylaxis (PrEP) (Truvada) is a medication which if taken correctly is almost entirely effective in preventing HIV infection. In regions and countries where it has been widely taken up, HIV seroconversion rates have significantly decreased. Alongside testing and treatment, it offers the very real prospect of ending HIV infections. However, in England, commissioning it has (and still is) a controversial process, where NHS England has repeatedly raised supposed ‘uncertainties’, first legal and then scientific. The same has not happened in Scotland, where PrEP was commissioned to anyone who needed it in April 2017. This article presents a close reading of the IMPACT trial protocol, which we conclude cannot answer the questions it sets out to answer. We then suggest that the uncertainties the trial claims to address are in fact a tool of power which is deployed to strategically ration healthcare; introduce uncertainty about commissioning PrEP; and shift the boundary between individual responsibilities and state responsibilities for public health and HIV prevention. We conclude that all the above constitute an unethical use of clinical trial rhetoric, systematically discriminate against minority and vulnerable groups, and ration healthcare for those who most need it. As such, we call on all academics, clinicians and activists to resist further unethical misuses of clinical trial rhetoric.

  • sexual medicine
  • public health
  • health policy
  • medical ethics/bioethics

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


In this article we will argue that uncertainty, rather than helping protect the public and develop research and knowledge, is instead being constructed and deployed in such a way as to become a highly controversial and unethical way of rationing access to the HIV prevention strategy pre-exposure prophylaxis (PrEP). PrEP (also known as Truvada, which is a single combination pill of emtricitabine and tenofovir) is one of the WHO’s essential medicines targeted at populations with a high risk of HIV seroconversion.1 When taken prophylactically as directed, it reduces HIV seroconversion risk by up to 99%.2 In England, activists, healthcare professionals and non-governmental organisations have argued it should be commissioned, but NHS England has repeatedly declined to commission PrEP, raising uncertainty about cost-effectiveness, responsibility to commission and implementation.

We will examine how evidence and uncertainty are mobilised, and how stakeholders frame arguments for or against National Health Service (NHS) commissioning. As Russell et al suggest, if ‘we wish to study policy-making ‘as is’ rather than as the naïve rationalists believe it should be, then we require a framework of ideas that address that role of language, argument, and discourse’.3 As such, medical humanities can claim a space for policy analysis by drawing on humanities theories. First, we will use a close reading strategy, and second we will use rhetorical analysis of policy papers and trial protocols. In doing so, we follow Russell et al 3 when they argue for a return of an analysis of health policy which is informed by Aristotelian and humanistic scholarships. In doing this, we argue that the medical humanities can uncover (1) struggles over ideas in health policy; (2) uncertainty in health policy; and (3) how rhetorical arguments are often aimed at specific audiences.

Theoretically our work draws on the inseparability of ‘power/knowledge’, and this serves as a critical point of departure for our analysis. For Foucault the power/knowledge coupling is manifested in the dual operation of power being based on, and making use of, knowledge. Conversely, knowledge is reinforced by power and power reproduces knowledge, shaping it according to the ways in which localised power mechanism shapes epistemic and ontological claims.4 However, we wish to extend this thesis and suggest that the generation of uncertainty (ie, a perceived lack of knowledge) is also an important optic for understanding the ways in which controversies are staged and rhetorically orchestrated, and how uncertainty-as-power is exerted to ration healthcare and shift responsibilities onto individuals.


The US Food and Drug Administration approved PrEP in 2012, and the UK waited for the results of the English PROUD (Pre-exposure Option for reducing HIV in the UK: immediate or Deferred) study,5 the interim analysis of which showed an 86% risk reduction in seroconversion. As it would have been unethical to continue, the trial was ended. In light of the building weight of research confirming the effectiveness of PrEP, many expected NHS England would commission PrEP. However, on 21 March 2016, they announced they would not.6 They instead opted to extend support and availability to the individuals who were part of the PROUD trial and ‘build on the excellent work to date and…[make] available up to £2 m over the next two years to run a number of early implementer test sites’.7 They would also

seek to answer the remaining questions around how PrEP could be commissioned in the most cost-effective and integrated way to reduce HIV and sexually transmitted infections in those at highest risk…[and] test the ‘real life’ cost-effectiveness and affordability of PrEP as part of an integrated HIV and STI prevention service.8

In reply, professionals and activists argued PrEP was cost-effective vis-a-vis HIV treatment; a desired preventive tool by those at risk of HIV infection; and had already been successfully implemented in many countries.9 10 Indeed, activist groups such as iwantPrEPnow and Prepster had already been facilitating access to generics for several years, which many attributed the dropping HIV seroconversion rates to.11 Nevertheless, NHS England raised a legal uncertainty about commissioning PrEP, which was challenged in court by the National AIDS Trust. NHS England lost their case, twice.12 However, rather than commissioning PrEP, they announced a nationwide PrEP trial with 10 000 participants over 4.5 years to address the remaining uncertainties about PrEP.13

Just as NHS England had delayed because of uncertainties that they constructed about the legal status of commissioning PrEP, uncertainty was again constructed and used as a rhetorical device for delaying the commissioning of PrEP by instigating a further clinical trial to answer ‘key questions’.14 We will demonstrate that the questions addressed were in fact not answerable via a trial methodology and that therefore power is being produced not through a form of knowledge that is based on the power of evidence, but rather the power to introduce notions of uncertainty. Uncertainty is therefore nuancing the Foucauldian notion that knowledge is always power, and instead shows that uncertainty-as-power is a rhetorical device used primarily by NHS England in order to negate the applicability and value of clinical trials that have been conducted outside of the UK; obfuscate the evidence-based arguments around the economics of commissioning PrEP; and shift the boundary of where individual responsibilities and state responsibilities lie for HIV prevention, thus constituting a form of healthcare rationing.

The IMPACT trial

The PrEP IMPACT trial protocol (latterly referred to as ‘IMPACT’) responds to and plays on the uncertainties introduced by NHS England. For example, it states it will ‘measure PrEP-eligability, PrEP-uptake…duration of PrEP-eligability…[and] duration of use’.15 First, this is problematic because eligibility cannot be measured by trial methodologies. A trial must have predetermined eligibility criteria to recruit, in this case: over 16, HIV-negative, able to consent, willing to take PrEP and have follow-ups. Participants must also be in one of the three high-risk groups: men who have sex with men and transgender women who have had and intend to have condomless (excluding oral) sex with men; people with an HIV-positive partner who is not virally suppressed and with whom they anticipate condomless sex; and people who are ‘assessed by a clinician as at a similar risk’.15 Paradoxically, the trial seems to only be able to conclude that those who fit in the eligibility criteria for IMPACT did in fact join as per the eligibility criteria—a clear tautology. The trial is therefore constructing an uncertainty about eligibility (which it paradoxically cannot answer) in order to justify holding back on commissioning PrEP.

Additionally, the vague eligibility criteria are disadvantaging certain groups. For example, recruitment for IMPACT has been disproportionately (white) gay men,16 possibly because the first criterion allows clinicians to easily identify this group. The same could be applied to partners of people who know that a sexual partner has a detectable viral load with whom they may not use condoms. However, the final category ‘of similar high risk’ is simply not presented with conceptual rigour. How is risk being measured? What level of similarity is required? There is no mention of other demographic groups known to be at high risk, such as sex workers, trans* women, the black-British community and certain migrant communities,17 thereby rendering at-risk communities invisible, under-recruited and without the benefits of PrEP. In other words, uncertainty about how the trial criteria apply is also leading to the concurrent disadvantaging of certain groups from accessing PrEP via the trial.

Next, IMPACT claims to examine uptake of PrEP; however, there are similar issues with this argument. Clinical trials require sample sizes to be predetermined so that meaningful statistical data can be produced. Placing a cap on the number of people who can uptake PrEP automatically invalidates the potential for a trial to measure uptake. No further data are being collected, hence the question cannot be answered. Retrospective audit of uncapped PrEP prescribing might be better suited to this task. Again, uncertainty is being used as a rhetorical device to limit numbers without any meaningful answers being produced to justify the ethical implications of limiting access to a highly effective HIV prevention strategy. This constitutes covert healthcare rationing.

IMPACT also aims to ‘determine whether or not incident HIV infections in trial participants are due to non-adherence or biological failure’.18, 19 The latter, being so rare that it may not be detected, leaves non-adherence as the catch-all explanation measured by ‘routinely collected data on number of pills dispensed and number of days between clinic visits’.20 While this approach works well for medications that must be taken at fixed and regular intervals, IMPACT allows participants to follow one of two regimens: daily PrEP or on-demand PrEP. When one considers that no data are planned to be collected about participants’ sexual habits, and given the fact that participants may change regimens, assessing adherence via pill counts offers contextless data which again cannot answer the initial question. Again, therefore, uncertainty about adherence and biological failure is used as a rhetorical device to justify the existence of a trial despite the proposed trial being unable to answer the constructed uncertainty.

Ideologically this assumes an individual’s adherence is the only explanation for non-taking of medication, and this fails to account for other structural conditions under which the trial is being undertaken. First, clinics receive no funding to conduct IMPACT, an extremely unusual (likely unprecedented) decision for such a large trial where full economic costings are standard, instead only drug costs are covered. In addition, IMPACT is occurring in the context of significant funding cuts to sexual health services in England (mean cuts are between 15% and 20%, with some services cutting 55%).21 This has led to 63% of clinics turning away patients each week and 19% turning away more than 50 patients per week.22 Furthermore, a Freedom of Information request to Manchester Foundation Trust in England, which runs six clinics, reveals that between December 2017 and November 2018 there was a mean of 17.4% presenting patients, with up to 24.3% turned away from their walk-in service. Given that IMPACT is running via the walk-in services at these clinics, it is reasonable to assume that at least some of these people were attending as trial participants. These types of structural conditions introduce significant confounding variables which are not mentioned in the trial protocol. Yet, as the trial protocol stands, anyone who is non-adherent and/or seroconverts will be assumed to do so because of their own individual failings. Finally, in the same Trust (in the Hathersage Clinic), as of February 2019, 799 patients were eligible for a place on a trial, of whom only 263 were accessing PrEP via the trial (43 could not be traced in clinic records), equating to 32% of the eligible patient population. The average waiting time for these people was 192 days, ranging from 45 to 576 days. This has had major consequences, with nine people HIV-seroconverting while waiting.23 Again the rhetorical claim that there is uncertainty about why HIV infections may occur is being used to justify the running of a trial, which is leading to people seroconverting when they cannot access the trial.

Following the above analysis, it would seem as though the IMPACT trial cannot answer many of the questions it sets out to answer. Some of this can be attributed to some parts being poorly conceptualised. Second of all, the trial seems underfunded in such a way that it cannot run as a clinical trial should be expected to run—that is, to see the people who are participating in it without undue difficulty, and therefore provide valid and reliable data that do not have significant confounding variables. Finally, the production of uncertainty is being used to justify a trial which is having perverse ethical implications.

Healthcare rationing and structural bias

As we have argued above, the creation of uncertainty about uptake can be read as a form of healthcare rationing. Clearly, balancing cost with need is always necessary in any healthcare system. However, this must be done in just and transparent ways using equitable procedures. Case in point (as of June 2019), only 147 clinics out of a potential 232 sexual health clinics in England are giving out PrEP via IMPACT.24 We therefore suggest scientific rhetoric is being misappropriated to ration something predominantly used by sex workers, certain migrant communities, gay, black-British and trans* people. These people are therefore systematically disadvantaged from accessing the healthcare that they need via NHS England. This leads to the logical conclusion that the trial is implicitly perpetuating structural inequalities.25, 26 As we note above, activists have been encouraging individuals to buy generic PrEP online, and while we fully support the approach of these activists for getting drugs into bodies, we remain concerned about the ideology it is occurring under, namely that NHS England approaches this activism by proffering uncertainty that results in individuals rather than the society providing for HIV prevention. HIV prevention for all is just as much the responsibility of a government as treating childhood cancer.


Scotland commissioned PrEP in April 2017 and it became available via sexual health clinics on 1 July 2017; the generic Truvada was secured in November 2017.27 There were no court cases or trials. Instead, uncertainties about how PrEP would function were conceptualised as something to monitor retrospectively (not operationalise as tools of power), and where necessary develop knowledge via research which can actually answer the questions it addresses. For example, initial eligibility criteria within NHS Scotland remained vague with regard to people who have equivalent high risk of HIV seroconversion. This has resulted in 99% of the uptake being men. This led to Scotland arguing after their first end-of-year audit that further research and engagement were needed to establish how to improve access for trans* people, black-British communities, certain migrant communities and sex workers. In other words, uncertainty in Scotland led to innovation, while in England it led to trials which preclude it.

As in England, Scottish services are under funding pressures, meaning they have been unable to ‘contact individuals who were late for, or had missed follow up appointments’.28 Yet this suggests NHS Scotland takes the ideological position that there is at least some responsibility that clinics are failing to address. Either way, the uncertainties and difficulties are being used to develop eHealth models to ‘increase access to PrEP and reduce the burden on sexual health services’.29 These may be imperfect efforts, but it at least directs uncertainties towards developing answerable questions that may improve, not stifle and ration, healthcare and recognises society-level responsibility for HIV prevention.


Activists have had success in improving access to PrEP in the UK; however, NHS England has frustrated these efforts by restricting access to PrEP by claiming that uncertainties exist, only to then answer them via methods which produce unethical consequences. Additionally, an ideological shift has occurred where a lucky few must adhere to underfunded and inadequate healthcare systems, while the unlucky assume the full responsibility of purchasing HIV-preventing medication privately. This constitutes an unethical form of healthcare rationing where commissioning is held back in the name of addressing uncertainties. In NHS Scotland, while the same uncertainties potentially existed, they are being addressed via audit and research which contain no such potentials for rationing care, in fact they aim to broaden access. We therefore call on academics, clinicians and activists to ensure trials do not engage in the unethical practice of producing uncertainties which ration healthcare and force individuals to care for themselves; in society we all should have access to the healthcare we need. Furthermore, now that these issues have been made clear, doing otherwise would be a potential breach of professional standards as well as an abuse of the power that medical discourse contains.


1. Robert M. Grant et al. (2016), “What People Want from Sex and Preexposure Prophylaxis,” Current Opinion in HIV and AIDS 11, no. 1: 3–9.

2. World Health Organization, “20th Essential Medicines List,” accessed June 19, 2019,

3. Jill Russell et al. (2008), “Recognizing Rhetoric in Health Care Policy Analysis,” Journal of Health Services Research & Policy 13, no. 1: 41.

4. M Foucault (1977), Discipline and Punish: The Birth of the Prison, trans. Alan Sherridan (London: Penguin), 27–8.

5. Sheena McCormack et al. (2016), “Pre-Exposure Prophylaxis to Prevent the Acquisition of HIV-1 Infection (PROUD): Effectiveness Results from the Pilot Phase of a Pragmatic Open-Label Randomised Trial,” The Lancet 387, no. 11013: 53–60.

6. The Lancet HIV (2015), “PrEP: Why Are We Waiting?” The Lancet HIV, 3, no. 1: e401.

7. NHS England, “Update on Commissioning and Provision of Pre-Exposure Prophylaxis (PrEP) for HIV Prevention,” accessed June 20, 2019,

8. NHS England, “Update on Commissioning.”

9. The Lancet HIV (2016), “Time for NHS England to Accept Responsibility for PrEP,” The Lancet HIV 3, no. 9: e397.

10. The Lancet HIV (2017), “Better Late Than Never: PrEP in England,” The Lancet HIV 4, no. 1: e1.

11. Nneka Nwokolo et al. (2017), “Rapidly Declining HIV Infection in MSM in Central London,” The Lancet HIV 4 no. 11: e482.

12. The Lancet HIV (2015), “PrEP: Why Are We Waiting?”, e401.

13. The Lancet HIV (2016), “Time for NHS England to Accept Responsibility for PrEP,” e397.

14. PrEP Impact Trial, “Frequently Asked Questions,” accessed June 20, 2019,

15. PrEP Impact Trial, “Frequently Asked Questions.”

16. NHS England, “PrEP Trial Updates,” accessed July 31, 2019,

17. Public Health England (2015), HIV: Surveillance, Data and Management (London: Public Health England).

18. A Sullivan, “PrEP IMPACT Trial: Trial Protocol Version 3,” accessed April 9, 2019,

19. A Sullivan, “PrEP IMPACT Trial.”

20. A. Sullivan, “PrEP IMPACT Trial,” 19.

21. R Robertson, “Sexual Health Services and the Importance of Prevention,” accessed May 20, 2019,

22. British Association for Sexual Health and HIV, “Government Funding Cuts Leave Sexual Health and HIV Care at 'Breaking Point,’” accessed May 20, 2019,

23. R Pebody, “Nine HIV Diagnoses in Men on PrEP Waiting List, Manchester,” accessed August 8, 2019,

24. NHS England, “PrEP Trial Updates.”

25. J.Y. Nazroo et al. (2009), “Ethnic Inequalities in Access to and Outcomes of Healthcare: Analysis of the Health Survey for England,” Journal of Epidemiology & Community Health 63, no. 12: 1026.

26. J Fish and K Karbana (2015), Lesbian, Gay, Bisexual and Trans Health Inequalities: International Perspectives in Social Work (Bristol: Policy Press).

27. NHS Scotland, “Implementation of HIV PrEP in Scotland: First Year Report,” accessed March 26, 2019,

28. NHS Scotland, “Implementation of HIV PrEP in Scotland,” 11.

29. Ibid., 41.



  • Twitter @NagingtonUoM, @@TweetingTony

  • Contributors Both authors have contributed to the original ideas of this article. No one else has provided any significant input.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.