StopCOVIDNI App

The StopCOVIDNI app enables smartphone users to be alerted to being in close proximity to another person who has subsequently tested positive for COVID-19. It can also alert others if the user of the app has subsequently tested positive for the virus.  This means that people are alerted to the risk of transmission of the virus and can begin to take steps including self-isolation to help prevent its further spread. It is important to note that the StopCOVIDNI App.

Is designed to respect an individual’s privacy and only exchanges anonymous information.

Background

Digital Health and Care Northern Ireland (DHCNI)

The DHCNI partnership sets the vision and strategy for Digital Transformation in Health and Social Care, directs and oversees the central budget for Digital investments and works in partnership with, and commissions projects and services from the Business Services Organisation which is the HSC’s shared services provider. DHCNI also aims to bring together the disparate Digital functions across the six Trusts: collaborating to develop a region wide strategy; facilitating shared leadership and decision making; setting policies and standards for the Trust teams to follow; and providing services and support.On taking up post in June 2019, the new Chief Digital Information Officer in the Department of Health, Dan West, reviewed the structures for delivery of digital services across the Department of Health and the HSC in Northern Ireland. This resulted in the formation of the Digital Health and Care Northern Ireland (DHCNI) partnership. It is led by the Department of Health and brought together the Health and Social Care Board’s eHealth and Care Directorate and the Public Health Agency’s Centre for Connected Health and Social Care.

Digital Co-ordination Cell

In April 2020, as part of the wider Departmental and HSC efforts to address the impact of the pandemic, the Department’s Chief Digital Information Officer (CDIO) convened a ”Digital Co-ordination Cell” (DCC) to oversee the Digital response to COVID-19 in Northern Ireland. The development of the StopCOVIDNI app was one of the products of the Cell.

Purpose

The purpose of the DCC was to co-ordinate the digital support and enablement of the DoH & HSC response to prepare for the additional burden that COVID-19 was anticipated would place upon the system. Many of these were related to public health and operational considerations, however there was also a sub-set of the preparations that related to issues such as Data collation, storage and access (including governance) and use of appropriate technology. There were also a number of questions that required system-wide resolution and / or decision making as well as funding approvals at the overarching leadership level.

Objectives

• Understand, articulate, and where required, coordinate the Digital activities underway across the HSC.
• Identify duplication in the activities being undertaken or requested, bringing together the disparate parts of the system that are attempting to solve the same problem.
• Clarify expectations and define DHCNI strategy where required.
• Support DHCNI teams in provision of resources and authority; through coordination of funding and approvals, and directing the unblocking of issues.

Reporting structure

The group reported to the Department of Health strategic decision making team, “Gold Command” and worked in close liaison with other operational Covid cells in the Department and with HSC organisations.

Digital interventions identified by the Digital Cell

Given that the COVID-19 pandemic was a rapidly developing and changing situation, the group identified a number of immediate priorities to help to cope with the impact on public health and the HSC. These included:

• enabling home working for DoH & HSC staff members
• the development of a COVID-19 web presence for healthcare advice
• the development of a telephone helpline
• development of a capability to carry out remote consultations

As the situation evolved and it became clear that further rapid innovative interventions would be needed to help lessen the impact of the virus, the group identified and commenced development of a number of other initiatives which were subsequently deployed. These included:

• The CovidCare App (symptom checker app providing immediate advice and links to trusted information on Covid 19).
• StopCOVIDNI (an app to alert users if they have been in close contact with other users who have tested positive for COVID-19).
• Digital Self Trace (an automated system to enhance the contact tracing process operated by the PHA telephone Contact Tracing Centre).
• Vaccinations Management System (digital processes to help manage the delivery of the COVID-19 vaccines).
• Data analytics platforms such as the Covid Dashboard (digital systems to assist in the analysis of data to produce information which could help the HSC and other public bodies to respond more effectively to the spread of the virus). 

Background to the development of a proximity app

In early May 2020 there was a recognition of the potential benefits of a COVID-19 proximity app to augment the manual contact tracing process operated by the Public Health Agency in Northern Ireland. The proximity app was designed to allow members of the public to use their mobile phones to anonymously communicate their proximity to other app users, and inform each other when one of the users has had a confirmed positive test result. The proximity app enhances the effectiveness of the overall contact tracing process as an individual’s recall of their recent whereabouts can only ever provide a partial contact history.

The Chief Medical Officer and Chief Scientific Adviser were in agreement that the App would, if downloaded and used extensively, make a significant contribution to reducing the spread of infection through breaking the chains of transmission.

Potential use of NHS England app

When the development of an app was first considered, consideration was given to the adoption of the NHS England app which was scheduled to be launched on 18 May subject to testing on the Isle of Wight. Potentially an app could have been deployed quickly with minimal development costs and delay.  

However, there were a series of technical difficulties and in addition, concerns were raised by human rights and civil liberties groups over the proposed use of personal data held centrally by the NHS England app system (v1).

Privacy and civil liberties concerns

A range of groups including the NI Human Rights Commission raised concerns that the app might compromise privacy, by, for example, asking users to share location data. These arose from comments about the design of the NHSx app. The concerns prompted an app design that gave potential users reassurance that their privacy and their anonymity would be protected. These actions minimised the risk that such concerns would reduce the level of uptake.

Information on data use and access to the eventual source code was provided on the dedicated StopCOVIDNI App site - https://covid-19.hscni.net/stop-covid-ni-mobile-app/

Interoperability

To ensure the required interoperability with the RoI, NHS England and Scottish proximity apps, and that NI citizens would need only install and use one app, it was agreed that the NI app should use the Google/ Apple design model. This was announced on 18 June 2020.

Procurement decisions

Given the impact of the pandemic on all aspects of society, the pace at which public health interventions could be deployed meant that the procurement approach needed to be open, collaborative, flexible, agile and rapid.  Decisions, information and designs needed to pass between the top-tier of management to the delivery teams quickly and the ‘traditional’ commissioning and delivery paradigm was not appropriate. During the “Discovery process” a number of the priorities were identified on the effective deployment of a proximity app to augment the contact tracing process. These included:

• Timescale 
• Public confidence in the app to maximise the potential uptake in NI
• Interoperability of the app with RoI, other UK and European apps

Tendering Process

In light of the requirement for speedy deployment as described above, a decision was taken to compare the suitability of two potential suppliers, Civica and Nearform, experienced in the supply of such applications.

Background

Civica built the existing Northern Ireland COVID-19 symptom checker app (CovidCare NI) while NearForm built the RoI de-centralised symptom checker and proximity app.

Assessment framework

To assess the best supplier with which to contract, a scoring framework was constructed for each of their proposals using a range of criteria considered crucial to the efficacy of a proximity app.

These were:

1. Privacy concerns;
2. Human Rights concerns;
3. Effectiveness in Public Health;
4. Delivery pace;
5. Delivery and support price;
6. Delivery and integration risk and
7. Likelihood of adoption

Note: two options (proposals for a single integrated app and two separate apps) were considered in respect of Civica since they had already built the CovidCare NI app.

Summary of scoring

The overall difference in the total scores for each proposal was small with each proposal having its own merits and drawbacks.

The Civica proposal scored highly on Effectiveness in Public Health reflecting that they had successfully developed the COVIDCareNI app. The NearForm proposal received a lower score on this criterion reflecting that its app had only been relatively recently deployed in RoI and its efficacy in terms of public health was as yet unproven.

Nearform scored highly on the criteria of Privacy concerns and Human Rights concerns on the basis of its ability to address these issues prior to the launch of its proximity app in RoI. The Civica proposal for a single integrated app scored low on these criteria as the COVIDCareNI app which was already in use was not designed to protect individuals’ anonymity and privacy to the extent that a proximity app was required to. Civica achieved a higher score on these criteria in relation to their proposal for two separate apps. 

NearForm also scored highly on the criterion of delivery pace given that they had already launched their proximity app in RoI.

Cost comparison

The total build cost for the NearForm proposal was less than the Civica proposal while the “whole life operating cost” of the NearForm proposal for an 18 month operating period was marginally higher than the Civica proposal.

Decision making factors in choosing NearForm

1. Cross border interoperability more easily achievable;
2. Delivery time likely to be shorter as testing of the app was already done;
3. Separation of architectures would ensure privacy (as compared to an app integrated with the existing COVIDCareNI app);   
4. Given the anticipated short lifespan of any app, the whole life cost comparison was considered less important;
5. Likelihood of the app being interoperable with European apps in future and
6. Increased likelihood of adoption of this app as a result of NearForm working across Ireland.

Conclusion

After careful consideration of all of the above factors, a decision was made to approve the Nearform proposal as it offered the best chances of delivering a rapid introduction a proximity app, interoperable with RoI and other countries and designed to minimise privacy concerns and therefore maximise potential uptake.

As the company was not at that time on government procurement frameworks it was necessary to use the Direct Award contract mechanism to secure their services.

App development and maintenance costs 

Note 1. For ongoing staff costs

Note 2. For use of computing facilities

Note 3. A number of mobile phone handsets were procured for testing purposes using an existing SE Trust contract.

The Revenue costs listed are for one year in the first instance.

Local universities had no significant involvement in development of the StopCOVIDNI proximity app.

Advertising and promotion costs

The total cost of advertising for the StopCOVIDNI app across a variety of communications channels was £220K (phase one adult campaign £150K and under 18s campaign £70K). A more detailed breakdown by media type is not available.