Information for Data Providers

The Information Commissioner Office’s Anonymisation Code of Practice highlights conditions for processing personal data for research. The use of an independent NISRA-led Trusted Third Party (TTP) for ADR NI projects, as recommended in the ICO guidance, fulfils a vital role in ensuring the privacy of research subjects.  The TTP only receives limited personal information (e.g. name, date of birth and address) from which it creates an anonymous index that matches the IDs of the other datasets.

This index is then used by the data custodians and the independent Research Support Unit (RSU) team in NISRA to bring together de-identified datasets to produce an overall research dataset. The two teams in NISRA (TTP and RSU) are managed and run separately under separate Senior Civil Service direction to ensure that the TTP and RSU processes are independent.

Personal data which have undergone pseudonymisation, could still be attributed to a person by the use of additional information and should be considered to be information on an identifiable person.

While there is no potential identification of the data subject in anonymous information, there is a potential to identify a person in pseudonymised data.

Data requested for projects comes in two forms; personal data which goes to the TTP for matching and pseudonymised data which goes to the RSU for creating the research dataset. Both personal and pseudonymised data are protected by the GDPR.

Data providers require a legal gateway for sharing data with the ADR NI. The Digital Economy Act 2017 facilitates the linking and sharing of datasets for research and statistics purposes (see below for more information on accessing data for research purposes). There are further gateways under article 6 of the GDPR to use data solely for statistics or research purposes.

Privacy notices need to reflect that data are being shared for research purposes.

All data sharing in the ADR NI is underpinned by Data Sharing Agreements (DSAs) which are compliant with the GDPR and include the following headings:

• Role of data controller and data processor
• Legal basis for data sharing
• Information security
• Justification of the data being shared & documentation of the research project being pursued
• Data protection impact assessments and privacy notices
• Retention and disposal of data

All ADR NI projects are linked using the Trusted Third Party to match the personal data, ensuring safety of the data.

There is a code of practice under the DEA and NISRA will be accredited by the UK Statistics Authority for the TTP and RSU functions.

There is a separation of duties within NISRA by using a Trusted Third Party (TTP) to match the personal data and then a functionally independent Research Support Unit (RSU) with access to the attribute data. Neither functional unit can process the whole dataset for the research project.

ADR NI research datasets have no direct single variable identifiers.

When a research dataset is created, RSU carries out statistical disclosure checks to assess the risk of re-identification of a person within the dataset and implement controls to protect the data as required.

Data can only be accessed in NISRA's Secure Environment in Colby House. The Secure Environment is supervised by NISRA using the five safety principles below.

ADR NI projects are approved by an Approval Panel. The Approval Panel set up under the DEA include data providers, statisticians and academics to assess the public value of any research.

The researcher(s) and NISRA staff working in RSU and TTP receive training and are accredited under the DEA.

Data are pseudonymised by removing names, addresses and any other direct identifiers that would identify the individuals, before being made available for analysis.

There is a separation of duties within NISRA by using a Trusted Third Party (TTP) to match the personal data and a Research Support Unit (RSU) with access to the attribute data. No one person can see both personal and attribute data. 

Researchers are only given access to a bespoke extract of data that has been approved through a research approval panel.

NISRA carries out a disclosure assessment on the research datasets prior to them being accessed by the researcher.

Researchers have to apply to the Office for National Statistics (ONS) Approved Researcher scheme before they can access the data.

Researchers have to complete a training course on the Five Safes and disclosure and pass the assessment at the end; agree to their details being published; sign an agreement promising to protect the confidentiality of data and have an institutional guarantor.

NISRA staff processing the data are trained in data protection and have higher security clearance.  Researchers are also required to have basic security clearance.  

Researchers access the data through NISRA's Secure Environment based in Colby House and strictly controlled secure environment governed by policies, protocols and procedures to ensure data confidentiality. Within the Secure Environment, researchers are supervised by NISRA staff at all times.

The network which hosts the research dataset is air-gapped and has no connection to the internet.

To ensure that text, tables and charts produced by researchers cannot identify individuals, RSU checks all outputs to ensure that they meet confidentiality standards.

Output checkers are trained in statistical disclosure control.

Projects are only approved when the following conditions are met:

1. Data providers provide a letter of support for the project.
2. The project is approved through an approval panel. Projects are assessed against delivering a clear public benefit.
3. Ethical approval is obtained and there is an appropriate legal gateway.  
4. Researchers commit that they will publish their results.

If these conditions are not met, the application is rejected.