Here’s what we learned from two data collaboration initiatives during the Coronavirus crisis.
The London Office of Technology and Innovation (LOTI) has been working in partnership with the Greater London Authority (GLA) since the beginning of April 2020, to support London boroughs in accessing the necessary data, to model the changing nature of vulnerability. Here, I share some reflections about the journey we’ve been on over the past couple of weeks.
As the Coronavirus pandemic started to unfold, the priorities of local authorities shifted rapidly. They went from business as usual to crisis response mode, overnight! Significant effort and time was invested in setting up and mobilising resilience / support services for local people. We’ve observed three things, in particular, have contributed to making this operationally possible:
- Partnerships with local community organisations, including the Voluntary Community Sector
- Technology solutions / platforms matching volunteer support to local needs and
- Data, data and more data
In this post, I’ll focus on the third initiative in particular — the data powering the delivery of these targeted resilience services.
Data is crucial to identifying needs
This topic is significant enough to warrant an article in its own right, but here’s a brief note.
It is now well established that data is important, and much has been written about its role and impact in the public and private sectors. What I’d like to emphasise, is that data is more important than ever in informing the delivery of new, temporary, and entirely bespoke services capable of addressing a global pandemic, at a local level.
Establishing new teams, processes and ways of working are, of course, important elements, pivotal to the successful delivery of these services, even in unprecedented times. Access to good information / data, however, provides invaluable insight and empowers local authorities to identify where such services are most needed.
Whilst central government has supported local authorities with identifying clinically extremely vulnerable (shielding) individuals, local authorities are responsible for supporting communities with a broader range of needs and vulnerabilities, which often include but go beyond health concerns.
In engaging with London boroughs, we learned about the ways in which they are using existing data to identify vulnerable residents, model the changing needs of their communities, and develop the necessary insights required to better target resources to those most adversely impacted. Hackney council is a great example of the use of existing data to model these changing needs. Despite these great examples of local authorities using data to respond to the crisis more effectively, they continue to experience challenges, given the changing nature of the vulnerabilities experienced by the communities they support.
Whilst inherently a health crisis, the Coronavirus pandemic has had a ripple effect in many aspects of our society, including our economy. As a result, completely new cohorts of individuals (previously unknown to services typically delivered by a local authority), are finding themselves in a worsening set of circumstances. Identifying these cohorts and anticipating their needs, is particularly challenging for local authorities if, for example, the individuals’ interactions with their local authorities have been limited to date.
Other challenges relate to gaps in data resulting from the complex nature and relationship between central and local government departments / bodies. During the course of our engagement with boroughs, we’ve heard about gaps in (i) health; (ii) business; and (iii) education-related data sets.
LOTI and GLA have sought to ameliorate these deficiencies by removing barriers and facilitating data collaboration.
The data collaboration
We set out to support boroughs with: (i) free school meals data; and (ii) data relating to the deceased.
Free school meals data
Boroughs reported challenges in identifying vulnerable children, attending school outside of their borough of residence. By using eligibility for free school meals as a vulnerability indicator, we were able to identify: (i) how many children from each borough attend school in another borough (with thanks to Marta Lapsley from GLA’s data team) (ii) create a draft Information Sharing Agreement (ISA) in just 24 hours (with thanks to Alexandra West, Information Governance Manager for LB Redbridge); and (iii) secure endorsement from the Information Governance Group for London (IGfL) chaired by Brent’s Ranisha Dhamu. What would have otherwise taken at least 6 months, only took three days. Yes, just three days!
Our methodology for sharing data, heavily relying on individual borough to borough exchanges of detailed children’s information (via secure means, of course) brings me to the first learning point — that sharing data pan-London is harder than one might think.
Sharing data at a pan-London level is hard
There were two particular challenges we encountered: (i) obtaining the contact details of the relevant officers in each borough; this was mainly owed to the complex nature of the different roles involved in service delivery during the Coronavirus response and (ii) managing the sharing of data between 33 boroughs. The latter point is slightly more complex, here’s why.
One can imagine sharing data between two boroughs may well be straightforward when the right sharing agreements are in place. Sharing data between 33 boroughs is completely different, when for example, children living in any one borough may attend school in any one of the other 32 boroughs.
The number of combinations and therefore data sharing exchanges between boroughs are likely to be significantly high. And you’d be perfectly right in thinking about the increased risks associated with such a high number of exchanges — even when using secure methods for sharing. The main reason for undertaking this project is to protect the most vulnerable in our society and not expose them to additional, unnecessary risks.
What this process has highlighted is the importance of tools for pan-London data sharing. In fact, one such tool — the London Data Store — already exists, but more needs to be done to ensure that it is the platform of choice for initiatives such as this one. Over the next few months, we’ll be working with the GLA Data Team to further develop the London Data Store to ensure it is a secure platform, suitable for pan-London data sharing.
In addition to this, LOTI has also initiated work on a number of other areas, to address the more wider barriers to data collaboration. These include working closely with (i) IGfL colleagues in improving the Information Governance processes for data collaboration; (ii) the London Fire Brigade to support boroughs in adopting the Information Sharing Gateway for creating data sharing agreements; (iii) supporting boroughs in getting better procurement outcomes (for data) by ensuring that all future contracts and tenders are specific and prescriptive about data ownership and system interoperability and (iv) Greater Manchester Combined Authority and CC2i in developing a digital Data Protection Impact Assessment tool. We’ll also be developing a standards framework for all smart city deployments, to ensure consistency in London and avoid the age-old issue of data silos.
Despite the progress we’ve made this year, we recognise that more needs to be done to resolve these thorny, decade old challenges.
More details on all the LOTI initiatives mentioned above can be found in this document.
Data on deceased
The other data sets boroughs reported issues with was receiving timely data on the deceased. This was having an impact in supporting vulnerable families left behind. In England, all deaths must be registered with local registration services, which are regulated by the General Register Office (GRO). Whilst the Digital Economy Act 2017 provides for the use of data on deceased individuals for a public purpose, many local authorities are unaware of the opportunity to request this data directly from local registration offices. Even where these agreements exist data is often shared in inaccessible pdf documents.
We are working with Social Finance, a social consultancy, to support London boroughs in obtaining this data more easily and will share more details on our learnings as we progress.
Two more general reflections
Challenges with obtaining the right data due to fragmentation
In conducting the projects described in this post, we’ve noticed fragmentation on at least two different levels (i) within a local authority and (ii) with central government. Within a local authority, this is primarily a direct result of data held and managed in siloed legacy systems. Think about this as each council service having a piece of a much bigger jigsaw puzzle. In practice, this makes it incredibly hard to identify / cross reference residents who may be interacting with multiple services.
Within the wider ecosystem, which local government is part of, we’ve noticed that central government departments in some specific cases also hold crucial pieces of this puzzle. And it’s the complete puzzle that creates the powerful insight - an essential element for designing targeted interventions.
The Power of collaboration — almost anything is possible…if people are willing
The initiatives mentioned above are a product of the power of people coming together under unprecedented circumstances, willing to share and work together, in order to support vulnerable communities in a time of crisis.
Although it has not been an easy ride, this has demonstrated that local authorities can break organisational barriers and be resilient in a time of crisis when supported by passionate teams of experts willing to support those most adversely affected by the Coronavirus crisis. Creating an Information Sharing Agreement in just three days, is only one example of this.
We’ve just begun and a lot more needs to be done. Fixing the plumbing for data collaboration, within local government at least, is now more important than ever — to help improve the outcomes of the people we’re here to serve.
Look out for further progress updates on this and other LOTI projects on Medium and Twitter.
