Council leaders in the borough of Barking and Dagenham in the UK see the mission of local government as having the duty to develop communities, develop economic prosperity and look after the most vulnerable people in society.
On March 16th those leaders heard the instruction that the nationwide lockdown in the UK was highly likely and swung into action to protect as many residents as possible from the worse effects of the virus while being constrained by physical distancing.
Fortuitously the London council had digital and data tools already in place to help achieve that undertaking.
Council leaders had more reason than most to be highly concerned about its residents as London Borough of Barking and Dagenham, nine miles east of central London, has high levels of unemployment, income deprivation and child poverty. However, there is a great sense of community spirit and neighbourliness. The council leaders advised of the possible lockdown in the morning of March 16th and by the afternoon 40 community support groups had convened in the council chambers to offer their help.
That help had to be targeted. Pye Nyunt, Head of Insight and Innovation at Barking and Dagenham, spent that morning with his team trawling through data in the council's systems, especially social care data.
To quantify the scale of the challenge. We made a prediction that Monday before lockdown, looking through [and] using the risk factors from the World Health Organisation and applying that to our population, that there would be 11,753 residents or approximately 5.5% of the population that would be clinically very vulnerable to Covid.
Nyunt and his team also produced maps to show the location of those vulnerable residents, so the 40 community support groups, which within a week had formed a partnership called Barking and Dagenham Citizens Alliance Network, knew exactly where to focus their efforts.
They then set about supporting those residents with things like food parcels, shopping and prescriptions for people who needed it and were told to shield. And by the first week, we helped well over 300 people in the community. So this was a bit of a punt, but based on the level of statistical robustness that we had at the time.
Four weeks later central government provided each local authority with a list of the vulnerable people in their area. Nyunt said he and his team had correctly predicted where most deaths were likely to take place and keep the mortality rate comparatively low compared to the rest of the country because of their "proactive response". The government had given Barking and Dagenham a shielding list of 8,100 people and 4,000 residents self-reported as clinically vulnerable, totalling 12,100, meaning that Nyunt's estimate of 11,753 was 97% accurate.
A single resident view
Being a highly digitised organisation also assisted the council when it had to make direct contact with residents. Nyunt and his team were able to repurpose an existing tool to have a single customer view, or in this case, a single resident view called One View. With many agents from the borough making contact or attempting to make contact with a particular resident duplication can occur or communications can be missed. Of the platform, Nyunt said during a presentation in May:
We were immediately able to identify 1,114 households, who were not currently engaged with any council services but have significant debt of over £1,000, are over 65 and live alone. In the first week of using the case management system, 58 identified households were referred to a service area for further support. One View identified 1 or more risk factors for 99.6% of the individuals who self-referred to the NHS prior to receiving confirmed data from the NHS.
Other risk factors that were also used included having a respiratory condition, possessing a disabled parking badge or receiving free school meals. He added that the Xantura information governance platform allowed for "rapid, ethical and proportionate processing" while Claire Symonds, Chief Operating Officer of the Barking and Dagenham, credited Informatica with assisting in overcoming the challenge of removing data siloes between housing, social services and finance departments.
Forecasting future demand
One View also has within it a natural language generator which helps to read the various case notes present on a particular resident. Social care is the service area that local councils spend the most money and as a result, generates a lot of interactions and case notes. The natural language generator can search through those case notes and create executive summaries for each vulnerable resident, which is then used by a frontline worker to address the resident's case and proactively make contact with them.
This has helped the council's insight team to very accurately forecast the future demand for services in both the short and medium-term. They are now planning to expand to test and trace data management.
Barking and Dagenham isn't the only local authority planning to do so. Amid the confusion and disjointedness of the central NHS Track and Trace system three councils in West Yorkshire, Calderdale, Kirklees and Bradford, announced plans to develop their own systems.
The identification of specific groups was facilitated by Barking and Dagenham's creation of its own census. Nyunt told an audience at innovation foundation Nesta that the council was gathering up-to-date information from datasets including the school census, the electoral roll, council tax, registrations, benefits data. This allows for an accurate view of the different demographics of the residents in the council that is much more current than the national census which is now nine years out of date. Nyunt said:
By matching its own datasets, Barking and Dagenham created a very accurate demographic profile of the people that they serve. Not everybody pays council tax but sure enough, someone is going to get ill and the data that we can get from our GPs tell us that.
As the months become colder and colds and sniffles become more common, Barking and Dagenham is now preparing for a "twin-demic" of seasonal flu and the second wave of Covid-19. Nyunt and his team will be using their powers of hyper-local analysis to identify the best places to deliver the winter flu vaccine. These may potentially be churches, mosques, libraries or colleges. They will also be working with a software provider to deploy a patient-led SMS booking system so that residents can dictate the time, date and place at which they want to have their jab.
At the Nesta event, Nyunt urged listeners to rethink data through the lens of opportunity to be able to redefine public services for a better society. The rethinking of data, in this case, proved to be an investment which enabled council chiefs to be self-sufficient in the identification of most in-need population during the biggest public health crisis of our lifetimes. Adopting a digital strategy with a strong emphasis on data to understand, interact with and help residents, during emergencies and during less turbulent times is a path that other local authorities may want to follow.
Pye Nyunt was speaking The Role of Digital in Public Sector Resilience and Performance hosted by the Embassy of Denmark in the UK.