Background to the memorandum

While eligibility for free treatment by the NHS has never been absolute or universal, in recent years, the UK Government has increasingly enforced the principle that the service is only available to those ‘ordinarily resident’ in the United Kingdom. In April 2015, the Government changed how the NHS in England charges migrants, visitors and former residents of the United Kingdom for hospital treatment. First, an immigration health charge (or ‘surcharge’) payable by non-European Economic Area (EEA) nationals who apply for a visa to enter or remain in the United Kingdom for more than 6 months was introduced, and subsequently increased in cost. Second, the Government also increased the scale and scope of charging for NHS services for those not eligible for free-at-the-point-of-use care. Those who are not ‘ordinarily resident’ – that is, those who are not ‘broadly, living in the UK on a lawful, voluntary and properly settled basis for the time being’ (NHS, 2017a) – who have not paid the levy are to be charged for non-urgent secondary care, including non-emergency hospital services, mental health and community care. From October 2017, non-NHS providers of NHS care, such as charities, will be required to charge for such care, and a requirement of upfront charging will be introduced, unless this would delay urgent or immediately necessary care (NHS, 2017b).

As a result, NHS bodies are now obliged both to check the status of a patient presenting to their service against flags placed on the patient file indicating status, while obtaining from the patient information about their residency status to identify whether they should be charged and, if they incur debts, to pass this information back to the Home Office (Department of Health, 2012). An outstanding NHS debt of £500 is now a basis for refusal of right to remain, and potentially leading to removal from the UK according to amendments to Appendix FM to the Immigration Rules (UK Government, 2016; Home Office, Department of Health, NHS Digital, 2017). In this way, charging for NHS services is now linked to immigration enforcement. In addition, responsibility for surveillance of migrants has been dispersed throughout the public sector, and among landlords and banks, amongst others, who must determine residence status before providing services (UK Government, 2013).

In practice, however, numerous issues with departmental sharing of information for the purposes of immigration enforcement have been identified. First, the system for transferring information between various NHS bodies and the Home Office was dysfunctional, plagued by burdensome administrative processes. Prior to January 2017, NHS Digital only identified the area in which the individual was registered as a patient, while the Home Office had to contact local NHS organisations to obtain their contact information (Gordon, Reference Gordon2017). This required a series of actions by many different parties, using several systems.

The MoU sought to simplify this situation, allowing NHS Digital to trace the personal data of the individual and supply it directly to the Home Office, thereby cutting out the need to contact local NHS bodies for information (Gordon, Reference Gordon2017). The Government justified the MoU on the basis that there is a ‘substantial public interest in sharing this non-clinical information to support effective immigration enforcement’ and for protection of people who might be at risk (Gordon, Reference Gordon2017). The agreement came into force on 1 January 2017, without a public consultation ever taking place. The Government only undertook a limited stakeholder review, justifying the lack of consultation and review on the basis that the MoU was only for the ‘operationalisation of existing functions’ (Gordon, Reference Gordon2017).

These functions are grounded in Section 261 of the Health and Social Care Act 2012. Section 261(5)(e) allows disclosure in connection with investigation of a criminal offence. Section 261(5)(d) permits disclosure where it is necessary, or expedient, for the NHS to have information for the purpose of exercising its functions under or by virtue of any provision of any Act. Section 261(5)(c) allows disclosure where it is necessary or expedient for the purposes of protecting the welfare of the individual. Any disclosure of data is subject to consideration of individual circumstances, the person to whom the information is disclosed, and the purpose of the disclosure. NHS digital is required, by Section 261(5)(e), to take account of common law that would otherwise prohibit or restrict the disclosure, weighing the public interest in favour and/or against disclosure. It must do this on a case-by-case basis. The public interest justification offered for breaching confidentiality is for ‘maintaining effective immigration controls’, as ‘it is in the public interest that limited UK resources and public services (including the NHS, jobs, schools, housing) are protected from unnecessary financial and resource pressures’ (Home Office, 2017).

However, on closer inspection, this statement is somewhat problematic. First, it ignores mounting evidence that problems facing many public services, particularly health and social care, reflect cuts in public funding under the guise of austerity (Loopstra et al., Reference Loopstra, McKee, Katikireddi, Taylor-Robinson, Barr and Stuckler2015, Reference Loopstra, Reeves, Taylor-Robinson, Barr, McKee and Stuckler2016; Hiam et al., Reference Hiam and McKee2017). In fact, although difficult to assess the estimated cost to the NHS of EEA and non-EEA visitors is ~1.8% of the total NHS England spend (Appleby, Reference Appleby2013). It also ignores any public interest in protecting public health, and having a health system that preserves confidentiality and is trusted.

The Government has acknowledged that ‘the increased transparency enabled by the MoU and Review may itself have an effect [of deterring immigrants seeking health services]’ (Gordon, Reference Gordon2017). This view is shared by many health care workers, who have refused to co-operate with data collection obligations that facilitate charging (Bulman, Reference Bulman2017). In response, the National Audit Office has advised on mechanisms to encourage enforcement (National Audit Office, 2016). In these circumstances, we ask whether the MoU could harm patients, the public at large, and health practitioners? If so, what should be done?

Issues for individual patients

The Government justifies the sharing of data with the Home Office on the basis that it ‘is potentially able to bring suspected immigration offenders back into contact, to either regularise their stay or work towards their removal’ (Gordon, Reference Gordon2017). It also states that it ‘is particularly important in respect of safeguarding responsibilities towards vulnerable children and adults who may be at risk’. However, as the Government has noted, there is a risk that migrants, fearful of the consequences of data sharing, may not seek health care (Department of Health, 2015). Public Health England has asserted that there is a lack of statistical evidence to suggest that this is happening, as there is a dearth of research. It has, however, conceded that:

Previous research supports these concerns. In the United States, fear of disclosure of undocumented status can both reduce or delay health-seeking behaviour for several diseases, including tuberculosis (Asch et al., Reference Asch, Leake and Gelberg1994; Martinez et al., Reference Martinez, Wu, Sandfort, Dodge, Carballo-Dieguez, Pinto, Rhodes, Moya and Chavez-Baray2015). In the UK a fear of deportation does influence testing for communicable diseases, including HIV (Dodds et al., Reference Dodds, Hickson, Weatherburn, Reid, Hammond, Jessup and Adegbite2008; Thomas et al., Reference Thomas, Aggleton and Anderson2010), although migrants do not carry a high burden of infectious disease. Victims of human trafficking have been shown to be particularly vulnerable to worries about disclosure, even when legal protections are offered to them (Kings College London, 2017). These findings are borne out by accounts not only from the scene of the Grenfell Tower fire, with reports of individuals assessed by volunteer doctors as requiring hospital treatment, for example, for head injuries, refusing it (Gentleman, Reference Gentleman2017a), but across the country, where many, including pregnant women, are too afraid to go to hospital, despite the clear medical need (Gentleman, Reference Gentleman2017a, Reference Gentleman2017b).

While such outcomes may on face appear clearly undesirable to the United Kingdom, it is critical to consider the context in which they have been implemented; the MoU is situated within a series of Government policies designed to create a ‘hostile environment’ for migrants (Steele et al., Reference Steele, Stuckler, McKee and Pollock2014). It was recently revealed that the Government had acquired data on distribution and needs of rough sleepers – data intended to allow for the allocation of services and to protect this vulnerable group – and then used it to target areas for immigration enforcement against homeless migrants, including specifically EEA nationals (Townsend, Reference Townsend2017). The fears that these actions generate clearly jeopardise the health of individual migrants and, as the next section shows, also the health of the population at large.

Issues for the public at large

Although most migrants arrive in good health, exhibiting the ‘healthy migrant effect’, and are at low risk of infectious diseases (European Centre for Disease Prevention and Control, 2015), undiagnosed and untreated infections, and lack of routine immunisations, can result in issues for the local population and communities into which migrants are received (Legido-Quigley et al., Reference Legido-Quigley, Otero, la Parra, Alvarez-Dardet, Martin-Moreno and McKee2013; Kentikelenis et al., Reference Kentikelenis, Karanikolos, Reeves, McKee and Stuckler2014). Failure to treat non-communicable diseases, such as diabetes and hypertension, may increase future health care needs and costs and burdens on health services, as well as impairing the ability of the individual to contribute to the economy (Bozorgmehr and Razum, Reference Bozorgmehr and Razum2015). Thus, any measurable impact on the health care-seeking behaviour of individuals risks impacting on public health more widely.

Issues for health care workers

When taking the Hippocratic Oath, new doctors swear that ‘whatsoever I shall see or hear in the course of my profession … I will never divulge, holding such things to be holy secrets’ (Miles, Reference Miles2005). As a result, patient confidentiality remains a core obligation for all in the medical profession. The General Medical Council (GMC) provides guidance on how to operationalise this requirement in daily practice, and, in Good Medical Practice, it makes it clear that ‘patients have a right to expect that their personal information will be held in confidence by their doctors’ (GMC, 2013). In April 2017, the GMC re-emphasised that ‘[a]sking for a patient’s consent to disclose information shows respect, and is part of good communication …’ (GMC, 2017).

As detailed above, the expectation of confidentiality is not absolute. Disclosure of a patients’ personal information may be necessary to support the direct care of the individual, for the protection of patients and others, or for other purposes covered by statutory obligations designed to protect the public interest. In all these cases, the objective is to prevent harm to the patient or others. Doctors who fail to follow GMC’s guidance on confidentiality and disclosure and who exhibit serious or persistent failures put their registration at risk.

Balancing confidentiality with disclosure of information involves challenges for practitioners. The GMC states that, for a disclosure to take place, it must be clear that ‘the benefits to an individual or to society of the disclosure must outweigh both the patient’s and the public interest in keeping the information confidential’. Considerations to be taken into account are listed in Box 1.

Source: General Medical Council.

Looking at these in turn, it is self-evident that disclosure will bring harm or distress to the patient and, as noted above, there is now evidence from several countries that disclosure of immigration status will impede the future engagement with treatment of some people in some circumstances. Second, there is also clear evidence from the events following Grenfell Tower that such measures can erode trust in doctors. In both of these cases, the offer of an amnesty by the government can be considered an implicit acceptance of these arguments. Third, where disclosure concerns risks of communicable disease transmission or terrorism, the risk of harm from non-disclosure are relatively clear, assuming that the information on which disclosure is based is accurate and non-discriminatory. However, this is far from clear in relation to immigration, where any theoretical benefits to society are very far from certain.

Where the government does have a case is the Home Office contention that this involves disclosure that is at or close to the level of minimum intrusion. In this they are supported by a ruling in the Court of Appeal in which four non-UK residents challenged the transmission of information identifying them from the Department of Health to the Home Office (Court of Appeal, 2015), in an appeal that was joined by the British Medical Association. The Court held that such information was ‘less intrusive’ than disclosure of detailed information about an individual’s medical condition and treatment. While noting that this did not mean that it was not intrusive at all, or that the information was not inherently private, it was held that it meant that it would be easier to justify disclosure. However, that judgement was on issues of data protection and did not take into account the other issues included in the GMC criteria.

Taking these criteria as a whole, it is unsurprising that many British doctors have asked whether their professional ethics conflict with the policy of their Government, in this case a policy promulgated by an MoU and not even by primary legislation that has undergone parliamentary scrutiny. Of course, this is a quandary that is faced by doctors in many countries now and in the past, sometimes even involving judicial and extra-judicial attacks on doctors treating individuals deemed unworthy by governments (Arie, Reference Arie2014; Vlassov, Reference Vlassov2014).

Also, it must be remembered that many health care staff are themselves migrants. The data from the GMC reveals that 30,472 doctors come from the EU and other countries in the EEA, while 71,139 were trained elsewhere in the world, outside the UK. Data collected by the House of Commons library show that, in December 2016, 22,081 nurses working in NHS hospital or community services had an EU nationality, accounting for 7% of the total (Baker, Reference Baker2017). Creating an environment where these health care workers are reminded daily that migrants are seen as a burden on the United Kingdom, and current migration levels are undesirable, may have significant impacts on morale and their on-going contribution to the NHS. Troublingly, a haemorrhaging of foreign staff from the NHS was noted after Brexit, with EU staff leaving in large numbers, while the numbers of registrations simultaneously dropped dramatically (Boffey, Reference Boffey2017). A critical decline in foreign staff and overseas-trained doctors in the NHS will in fact affect the sustainability of the service, threatening the quality of health care for the population at large. Accompanied by a reduction in applications for training places in shortage in fields like nursing and midwifery, there are real reasons for the public to be concerned about the staffing of the NHS, and therefore wait times, as well as safe and excellent care.