Appendix 1: full assessment of impacts

The Equality Duty is a duty on public bodies and others carrying out public functions. It ensures that public bodies consider the needs of all individuals in their day to day work – in shaping policy, in delivering services, and in relation to their own employees. The public sector equality duty that is set out in the Equality Act 2010 requires public authorities, in the exercise of their functions, to have due regard to the need to:

1. Eliminate unlawful discrimination, harassment and victimisation and other conduct prohibited by the Act.
2. Advance equality of opportunity between people who share a protected characteristic and those who do not.
3. Foster good relations between people who share a protected characteristic and those who do not.

These are sometimes referred to as the three aims of the general equality duty. The Act further explains that having due regard for advancing equality involves:

1. Removing or minimising disadvantages suffered by people due to their protected characteristics.
2. Taking steps to meet the needs of people from protected groups where these are different from the needs of other people.
3. Encouraging people from protected groups to participate in public life or in other activities where their participation is disproportionately low.

How will the policy or piece of work contribute to achieving the Public Sector Equality Duty (PSED) and reducing health inequalities in access and outcomes?

The national data opt-out service is expected to have the most impact on the second of the aims of the PSED, by contributing to advancing equality of opportunity between people who share a protected characteristic and those who do not. The national data opt-out gives patients more control over how their confidential patient information is used and lets patients exercise their right to make an informed choice about whether their confidential personal information is shared for research and planning purposes. Informed data sharing enables disadvantaged groups to benefit from research / health service planning, and whilst some individuals may choose to opt-out, the national data opt-out has the benefit of increasing public trust in what is happening with patient data. This EIA therefore focuses on ensuring there is equality of opportunity to register an opt-out, as well as assessing any equality risks from differential opt-out rates, including whether there might be high rates of opt-outs within particular groups and any potential impact this could have so that mitigating actions can be applied where possible.

This assessment did consider the extent to which the national data opt-out service could have an impact on fostering good relations between different people with protected characteristics. However, it was concluded that the impact of the service is neutral in relation to this specific aim.

What are the intended outcomes of this work?

NHS Digital was directed by the Department of Health and Social Care (DHSC) to develop a new system to support the national data opt-out which gives patients more control over how their confidential patient information is used. The system lets patients exercise their right to make an informed choice about whether their confidential patient information is only used for their individual care and treatment or also shared for research and planning purposes.

Who will be affected by this project, programme or work?

Patients are the main group who will be affected by this work: particularly patients who decide they do not want to share their confidential patient information for planning and research purposes as they will be able to set their national data opt-out preference online or via a non-digital alternative.

Staff members working in health and care also need to be aware of the national data opt-out so that they can advise patients about the benefits of sharing data, the choices they can make and where they can find more information. This is so patients can make an informed decision about how their confidential patient information will be used.

Health and care professionals who send confidential patient information to other organisations for reasons other than a patient's individual care and treatment will also need to know how and when to uphold a patient's preference.

Which groups protected by the Equality Act 2010 and/ or groups that face health inequalities are very likely to be affected by this work?

A full analysis of the impact of this work on groups protected by the Equality Act 2010 and groups that commonly face health inequalities is presented below, with a list of impacts identified and potential mitigating actions. For each group considered, the following questions were held in consideration:

1. Does the equality group face discrimination in this work area?
2. Could the work tackle this discrimination and/or advance equality or good relations?
3. Could the work assist or undermine compliance with the Public Sector Equality Duty (PSED)?
4. Could the work contribute to reducing inequalities in health outcomes?
5. Does any action need to be taken to address any important adverse impact? If yes, what action should be taken?

Impacts identified on specific groups and actions taken to mitigate these.

Older people

Although old age does not automatically mean poor health, a growing older population will lead to great demand for health and care services. An estimated 4 million older people in the UK (36% of people aged 65-74 and 47% of those aged 75+) have a limiting longstanding illness. Overall life expectancy has risen further and faster than disability-free expectancy, meaning more people are living into older age with multiple long-term conditions, frailty, dementia and social care needs.

Consequently, older people tend to be the expert patients who experience the issues/frustrations of data not being shared, but also may have security fears over the use of personal data. For example, the elderly often worry about insurance companies having access to their data as it is hard for older people to get insurance and this can make life difficult.  Maximising informed data sharing across the system will bring multiple benefits to elderly people as heavy users of healthcare, whilst providing a means of opting out of data sharing will allow greater public trust in data sharing.

There is conflicting evidence about whether older people are more likely to opt out than the general population, with different studies showing different results. Whilst analysis of type 2 opt-out data shows a peak in those aged 50-60 years, research carried out by the Behavioural Insight Team has shown that opt-out rates are lowest among those aged 65+. Discussions with stakeholders have also suggested that patients’ perception of how the data is to be used is crucial in determining whether someone will opt-out.

Digital exclusion has been identified as an issue for this group. Internet usage is increasing among older people (and has increased substantially over the last 5-6 years), and it should not be assumed that older people would not use the digital route, as many do use online systems – particularly with assistance. However, many are still non-users. Even if elderly people use the internet, usage may be less frequent and elderly users may be less digitally-skilled. Less frequent use of the internet can also mean they can have less familiarity with scams, for example, than regular internet users. Internet usage by the elderly is also less ‘transactional’ when compared to other age groups: just 29% of users in this age bracket undertook transactions weekly compared to 50% of all internet users in all age groups. (Later life in the United Kingdom - Age UK.)

Elderly people may not have mobile phones or an email address, one of which is required for verification of ID in the online system. Ofcom figures show that fewer people aged 65 or above own mobile phones compared the figure for all adults (89%). However, the drop is even more marked in those that are even older – there is a fall in mobile phone ownership in those aged 65 and over (78%) and an even more marked fall for those aged 75+ (52%).

Impacts identified on specific groups and actions taken to mitigate these.

People with learning disabilities

There are around 11.9 million disabled people in the UK. People with a disability may have a number of issues that mean they can potentially be disadvantaged. For example, there are those who have a learning difficulty who may have low literacy or may not be able to read a letter. Disabled people have lower literacy levels than the UK average - accessible information is needed to ensure these groups can make informed decisions.

Those who are socially excluded, e.g. people with disabilities, are often the most digitally excluded – for example, 25% of disabled adults have never used the internet, compared to 10.2% of the entire UK population.

Digital exclusion: Disabled people are significantly less likely than non-disabled people to have internet access (65% compared to 88%), and disproportionately low internet access affects disabled people across all age groups. Moreover, ONS statistics show that the proportion of adults who were recent internet users was lower for those that were disabled, compared with those that were not.

Actions taken

The service has provided accessible materials, including an Easy Read version of the handout, to help people make an informed choice. This has been promoted online and through the voluntary sector, who have helped us with our approach to accessibility.

The service meets web accessibility standards and user research has also ensured that the digital system is accessible. The service has been undergoing a Government Digital Service (GDS) assessment which has a high focus on accessibility.

The digitally-assisted and non-digital routes ensure that digitally-excluded people are assisted to set a preference online or are able to set a preference via the non-digital route.

People with sensory impairments

An online system is beneficial for some people with sensory impairments (e.g. in allowing use of assistive technologies), and our approach offers benefits over the GP-only opt-out system for previous programmes. However, the online system needs to be designed with accessibility in mind.

British Sign Language is the first language of many deaf people, rather than English.

Sense raised that the digitally-assisted route requires contact centre staff who can be aware of type-talk / video relay systems, e.g. skype.

Sense noted that many people with sensory impairments often have rare conditions so may be identifiable even if anonymised. They, therefore, may be more likely to opt out of data-sharing than the general population. However, the policy decision that people can consent to specific research studies may be important for this group, as many people with sensory impairments choose to be linked into particular research programmes.

An online system offers benefits over a GP system as it does not require travel, which can be a burden on this group. However, some patients may visit their GP in order to update email address/mobile phone/address details on PDS so that they can set a preference online.

Actions taken

The service meets web accessibility standards and user research has also ensured that the digital system is accessible. The service has been undergoing a Government Digital Service (GDS) assessment which has a high focus on accessibility.

The Accessibility Workshop ensured input from groups representing people with sensory impairments. Accessible materials, including a handout in easy-read, large print, audio, braille and British Sign Language have been developed and are available on request or online, and have been promoted through Local Healthwatch and specialist voluntary sector organisations. GP practices have also been made aware so they can signpost their patients with different reading preferences. Meanwhile, contact centre staff have been briefed on the use of text-relay services to ensure that people with sensory impairments are supported.

The provision of an opt-out helps to re-assure people about flows of their confidential patient information; whilst in relation to anonymised data ICO guidance requires that the risks of re-identification are reduced such that they are remote.

Voluntary sector involvement and advocacy through the public engagement strategy has helped to ensure that messages are being sent out to ensure greater awareness of what the opt-out will cover, whilst sitting within a wider data-sharing communications strategy has helped to ensure that people can make an informed choice, and that people understand the benefits of data-sharing before making their choice.

The non-digital option supports those patients who do not have a telephone number or an email address registered on the Personal Demographics Service (PDS).

People living with HIV

These are particular sensitivities around data for people living with HIV and historically rules have required records relating to sexual health to be held separately from other health information.  Additionally, some sexual health clinics may not use NHS numbers.

However, with proper reassurance, people with HIV are often highly aware of the value of research/sharing with relevant safeguards.

There would be concerns about a national data opt-out preventing data flowing to the HIV and Aids Reporting System (HARS) dataset.

55% of people living with diagnosed HIV are migrants - some may speak English well, but others may not.

Actions taken

The provision of an opt-out helps to re-assure people about flows of their confidential patient information.

The scope of data sharing and potential uses is being communicated clearly as part of privacy/transparency information.

The provision of HIV clinic data to Public Health England for the HARS dataset would be out of scope for the national data opt-out – due to this falling under regulation 3 of The Health Service (Control of Patient Information) Regulations 2002.

Accessible materials, including easy-read, are available on request or online. Materials have also been produced in the commonest languages amongst refugees and asylum seekers in England and a telephone interpreting service is in place for users who do not speak English

No particular impact was found on this group. It should also be noted that the public sector equality duty applies to marriage and civil partnership only in respect of the requirement to have due regard to the need to eliminate discrimination

High opt-out rates would have an impact on research into pregnancy-related conditions/treatments, but type 2 opt-out analysis shows opt-outs from individuals diagnosed with conditions related to pregnancy and childbirth are similar to the opt-out rates across all conditions.

Stakeholders expressed concerns that some vulnerable migrant women may be less likely to engage with maternity services due to fears of data being shared with the Home Office. However, it should be noted that since these conversations with stakeholders, it has been announced that there will be a significant narrowing of the tracing service. As the existing MOU between NHS Digital, the Home Office and the Department of Health and Social Care is no longer required, it has been withdrawn.

There may be links between a mother’s record and her baby’s record, e.g. baby records which include identifiable information about the mother.  

Actions taken

The scope of data sharing and uses is being communicated clearly as part of the privacy/transparency information.

The service has engaged with the voluntary sector including groups that deal with vulnerable women through the public engagement strategy.

For records containing details about multiple individuals a national data opt-opt applies to the entire record, irrespective of whether an opt-out is identified for the individual who is the subject of the record or for a 3rd party whose confidential information is contained within the record. This would only apply where the NHS number is present for the third party.

Impacts identified on specific groups and actions taken to mitigate these.

Ethnic minorities

Evidence from several studies suggests that people from ethnic minorities may be less likely to want to share their data than the general population (Hussain-Gambles et al, Knies et al, Garrison et al).

Most Asian groups express lower levels of satisfaction and less positive experiences of NHS General Practice services than other ethnic groups (NHS Outcomes Framework / Race Disparity Audit) – this may make opt-outs more likely (although this is not reflected in current type-2 opt-out statistics). In recent years, Bangladeshi, Pakistani, Indian and Chinese patients were among the least likely to report a positive experience of getting a GP appointment or of seeing a GP.

Analysis of type-2 opt-outs show that opt-out rates are higher amongst people identifying themselves as Black or Black British, and the BIT research points to higher opt-out rates across ethnic minorities as compared to white groups.

There are differing rates of digital exclusion among ethnic groups, e.g. ONS data shows that 10% of White, Pakistani and Bangladeshi ethnicities have never used the internet, compared to just 2.6% of people of Chinese ethnicity. The NHS BME Network also expressed the view that digital exclusion is high among elderly people of ethnic minority groups.

Actions taken 

Issues are often ethnic or cultural - not directly related to faith. However, religion interacts with other characteristics – e.g. an LGBT individual from a faith community going to GP may be less likely to want information disclosed, or women of particular faith groups may not want information regarding abortion / contraception disclosed.

Actions taken

Wider data-sharing communications strategy, public engagement strategy and Understanding Patient Data work are helping to ensure that people can make an informed choice, and that people understand the benefits of data-sharing before making their choice.

Type 2 opt-out analysis shows that opt-out rates in male and female populations are broadly similar, although women are slightly more likely to opt-out and there is some variation by age between different genders. Research provided by the Behavioural Insight Team showed that males are slightly more likely to opt-out than females.

Engagement suggests that women may be more likely to opt-out regarding particularly sensitive sexual health issues, e.g. contraception / abortion / miscarriage / hysterectomy.

Actions taken

Wider data-sharing communications strategy, public engagement strategy and Understanding Patient Data work are helping to ensure that people can make an informed choice, and that people understand the benefits of data-sharing before making their choice.

There is a substantial body of evidence demonstrating that lesbian, gay, bisexual and trans (LGB&T) people experience significant health inequalities, which impact both on their health outcomes and their experiences of the healthcare system.  The relationship between sexual orientation and gender identity and health has often been overlooked by the healthcare system, and there is a lack of sexual orientation and gender identity monitoring in service provision and population level research.

However, the use of sexual orientation data (e.g. in Sexual Orientation Information Standard) may result in more evidence-based approaches and may aid health and social care organisations in providing equitable access for lesbian, gay and bisexual individuals, and may increase the understanding of the impact of inequalities on health outcomes for different populations in England, similar to monitoring of ethnicity and gender.

Some individuals may fear the potential for data on sexual orientation to fall into the wrong hands and be misused, particularly in the event of it being “de-anonymised”, rendering individuals potentially identifiable.

Data sharing can present a double-edged sword for stigmatised groups, with some seeing beneficial opportunities and others having privacy concerns (see Stablein, Timothy et al). However, some studies have found no particular spike in fear of data sharing around the LGB community. Stakeholders reported that as monitoring becomes more widespread, the benefits are high as there is a greater potential to explore health concerns that exist because these are not widely monitored.

Actions taken

Wider data-sharing communications strategy, public engagement strategy and Understanding Patient Data work are helping to ensure that people can make an informed choice, and that people understand the benefits of data-sharing before making their choice.

Strengthened criminal penalties for deliberate or negligent re-identification of individuals.

Voluntary sector involvement through public engagement strategy, particularly with regard to high opt-out rates.

Some individuals within the Trans community have historical distrust of health system, e.g. due to perceived barriers to receiving transitional healthcare or lengthy referrals into trans-system regardless of symptoms.

Engagement has suggested that a minority within this community will have a “knee-jerk response” and immediately want to opt-out.

There are potentially high benefits to data sharing for this group – for example, there is limited published research into trans-health issues outside of gender reassignment pathways of care. There is also limited research into the long-term impact of hormonal treatment, although there is evidence of increased incidence of metabolic syndrome in male to female trans individuals using hormones.

There are conditions that are more likely to affect this particular demographic group, so data sharing can enable the targeting of prevention work.

However, the service will need to be clear about what data is being shared, as gender reassignment certificate is protected.

This community is likely to have particular concerns about how their data is shared, i.e. birth gender vs gender they are transitioning to. 

Actions taken

Wider data-sharing communications strategy, public engagement strategy and Understanding Patient Data work are helping to ensure that people can make an informed choice, and that people understand the benefits of data-sharing before making their choice.

Voluntary sector involvement through public engagement strategy, particularly with regard to high opt-out rates. The scope of data sharing and potential uses are being communicated clearly as part of privacy/transparency information.

Where a new NHS number is assigned to an individual following gender reassignment, any national data opt-out will not be automatically applied to the new record, although the patient should be informed of their choice as part of the change process.

Impacts identified on specific groups and actions taken to mitigate these.

Homeless / rough sleepers

This group may have a lack of awareness of opt-out – it is therefore difficult to ensure this group can make an informed choice.

Digital exclusion through low level of computer accessibility / web skills.

Studies show high usage of digital technology in the form of mobile or smart phones, but regular and frequent internet access can be problematic due to dependence on library/homeless services, unreliable internet connections, phones being lost/stolen, or loss of feeling in fingertips leading to difficulty using a smartphone.

This group may face some difficulty in verifying their identity via the non-digital route if they are not able to provide an NHS number, due to lack of address and identification documents.

In terms of updating details on PDS, though GP registration can be quite high among the homeless, there are definite barriers to registration e.g. some being refused access to a GP because they do not have proof of address. They may also have trouble accessing GPs even though they may be registered – many access services through A&E.

Actions taken

Voluntary sector involvement through public engagement strategy, particularly with regard to high opt-out rates and verification of identity, e.g. staff in homelessness organisations / Homeless Health Peer Advocacy.

Digitally assisted/Non-digital routes for setting the opt-out, e.g. via contact centre.

If a homeless person is GP-registered or ever has been then they will have a record on the Personal Demographics Service (PDS) along with an NHS Number. Patients may also get a record added to PDS (and an NHS Number) when presenting at another care setting (e.g. A&E, walk in centre, community services).

Those living with mental health issues

Stakeholders reported that there could be particular concern about data sharing due to stigma attached to mental health problems, e.g. those who have been sectioned who may not want this shared or widely known.

Greater digital exclusion - 49% of those with above average mental health have access to the internet compared with 38% of those without internet access.

Actions taken

Voluntary sector involvement through public engagement strategy.

Non-digital route for setting the opt-out allows those without phone / internet access to set a preference.

People in a secure mental health setting / detained under the Mental Health Act

There is likely to be a lack of awareness of opt-out and communication is difficult through mainstream methods.

People in secure mental health settings have a limited ability to set preference online - people in secure mental health facilities (particularly High Security) are likely to have reduced access to the online system. However, they should be supported to decide a preference where possible.

Actions taken

Voluntary sector involvement through public engagement strategy, particularly with regard to verification of identity.

The service is looking at providing non-standard ways of setting a preference in these settings; however, the non-digital route for setting the opt-out allows those without phone / internet access to set a preference.

Information has been shared with providers enabling them to make their patients aware. Information has also been shared with the National Survivor User Network.

People with autism

Autistic people face particular health inequalities and struggle to access health services, e.g. primary care. Some evidence suggests that autistic people may be at greater risk of dying prematurely (though there is a lack of data in the UK). Improved data recording / sharing may be beneficial as there is so little data collected about people with autism. GPs are now encouraged to record someone’s autism diagnosis and many people with autism are happy for it to be recorded on their health records; however, there is also a nervousness about data that could potentially be used prejudicially and concern about how diagnosis information would be used.

People who are on the autism spectrum may not be able to make an informed choice without support, and may have particular information and communication requirements. All information provided must be clear, concise and non-metaphorical, with time being given to process information. For some, a conversation is required in order to make a choice and not all people are able to use computers.

Actions taken

The scope of data sharing and potential uses are being communicated clearly as part of privacy/transparency information.

Accessible materials, including easy-read, are available on request or online.

Digitally-assisted/Non-digital routes for setting the opt-out, e.g. via contact centre, offers support for conversations about the opt-out or an alternative means of setting a preference without the use of computers.

People in custody

A high proportion of the prison population have low literacy or learning disabilities, which may mean that extra care needs to be taken to ensure they are able to make an informed choice.

Within prisons there is often no access, or highly restricted access, to online systems / mobile phones. This makes it more difficult for people in custody to set an opt-out preference. Additionally, they may be unable to access identity documents required for the non-digital process, without the help of family members/friends.

Prisoners/those in the justice system often don't want their data shared especially with their GP for fear of being removed from the register.

Actions taken

Accessible materials, including easy-read and large print, are available on request or online.

Voluntary sector involvement through public engagement strategy, particularly with regard to verification of identity.

During the public beta phase, the programme looked at providing alternative ways of setting a preference where people were unable to access via any standard route. A pilot scheme was run to help ensure healthcare professionals can verify the identity of people in detained estates through a secure alternative email channel. This is being rolled out across other Detained Estates as the service becomes fully operational.

Sex workers

Updating PDS records in order to set a preference online is impacted by differential access to GPs – e.g. difficulties attending an appointment, especially those working at night / street sex workers.

A distrust of services has also been reported, and some sex workers experience stigma from those running services for being involved in sex work, which may make them less likely to attend GP practices in order to update their PDS records.

A certain proportion may also be migrant sex workers working off-street who may face language difficulties, limited knowledge about how to access services, irregular and undocumented legal status.

Actions taken

Non-digital route for setting the opt-out allows people to set a preference without up-to-date PDS mobile/email details.

Voluntary sector involvement through public engagement strategy, particularly with regard to high opt-out rates.

Accessible materials, including easy-read, are available on request or online. Materials have also been produced in the commonest languages amongst refugees and asylum seekers in England and a telephone interpreting service is in place for users who do not speak English.

People lacking capacity for informed consent

People who lack capacity for informed consent often have greater health needs, so could benefit from researchers/care planners accessing health information.

Accessible information is also needed to ensure this group can make informed decisions.

Actions taken

Voluntary sector involvement and advocacy through public engagement strategy.

Outside of formal proxy relationships, someone with a signed declaration from the data subject themselves is able to act on behalf of them. In addition, formal proxy relationships are also supported, so that people who have a lasting a Lasting Power of Attorney (LPA) – either for health and welfare or for property and financial affairs – are able to set a preference.  Court appointed deputies are also able to register an opt-out.

Accessible materials, including easy-read and large print, are available on request or online.

Those who have experienced Female Genital Mutilation (FGM)

The NDG review clarifies that health professionals are required to report known cases of female genital mutilation to the police – the opt-out will not affect this.

Those who have experienced human trafficking or modern slavery

Victims of trafficking may fear being deported, and access to healthcare can be controlled by trafficking gangs. The traffickers have an interest in not allowing their people to be known to the NHS and may see the opt out as a way to achieve this.

Actions taken

The scope of data sharing and potential uses is being communicated clearly as part of fair processing information.

A risk assessment has been undertaken of the impact of an individual maliciously registering an opt-out on behalf of another without their knowledge, and the impact is seen as minimal since essential flows are protected.

Veterans

Research initiatives to understand trends in both physical and mental health are important, e.g. hearing / mental health issues to determine whether there are differences between those who have served and in general population. However, stakeholders suggested that an impact would only be likely to be felt if there were high opt-out levels.

Actions taken

Wider data-sharing communications strategy and Understanding Patient Data work.

Mobile communities, e.g. students / service personnel

Difficulty in maintaining up-to-date information in PDS means that a variety of mobile communities not covered in the groups above are less likely to be able to set a preference online – although if their information is up-to-date they are likely to find an online system easier to use than, for example, visiting GP practices.

The GP system that is used by Defence Medical Services (DMS) for the primary care of services personnel does not have a direct link to the PDS and therefore cannot update mobile phone and email details on PDS. Mobile phone and email address details may be present on PDS for services personnel from other interactions with the health and care services (for example, previous civilian GP registration or accident and emergency visits). However, these details are less likely to be current than would be the case for the general civilian population. This makes it less likely they will be able to set a preference online.

Actions taken

Non-digital route allows people to set a preference.

Specifically, the non-digital route for setting the opt-out should not present any barriers to setting an opt-out choice for service personnel stationed abroad, as all post will be handled under British Forces Post Office (BFPO).

During the public beta period, arrangements have been made for key updates to PDS to ensure service personnel can access the online service. It was established that DMS patients (including uniformed service personnel and their dependants) have so far not experienced any issues in setting their national data opt-out choice and both the online and contact centre/bureau service are meeting their needs.

Though some way off, a replacement system for DMS patients being developed should improve the capability for maintaining demographics data on PDS.