Innovative use of administrative data to improve local service design and delivery

In summary

Professor Les Mayhew and Gill Harper have been developing a system for the exploitation of administrative data and estimating and profiling populations in a commercial capacity since 2000, under the name ‘Neighbourhood Knowledge Management’ (nkm). Gill Harper even earned her PhD from helping to develop the system.

To date over 60 projects have been completed for more than 20 local authorities, healthcare organisations, central government and the third sector following a wide range of commissions.

Users have gained access to more accurate, detailed, and relevant data which have helped them make better evidence-based policy and planning decisions and to save money.

What did we explore and how?

Administrative data refers to information collected primarily for administrative (not research) purposes. These are collected for the purposes of registration, transaction and record keeping, usually during the delivery of a service and can include that collected for welfare, tax and health purposes and via educational record systems.

The potential for these data for social science research is increasingly recognised, although it has not yet been fully exploited.

Professor Mayhew and co-author Gill Harper started developing a unique system for the exploitation of administrative data and measuring populations, this involves combining different data sources with different population coverage according to defined and therefore replicable set of rules by linking data at address level.

Algorithms are used to remove duplicate data and screen out data that has been superseded. The result is current and geographically more flexible data which are more cost-effective to produce than a survey-based census.

The research also demonstrated that data sets produced are more accurate than traditional statistics as they use current data sources, are more economical to produce and contain richer demographic and socio-economic information unavailable elsewhere.

The innovative methodological features of their approach, approved by the Information Commission, include secure systems for data handling, legally binding data sharing protocols, information governance certification, algorithms for cleaning and linking data at the person and address levels, algorithms for combining data to produce population estimates and innovative analytical and modelling tools for processing and reporting data.

Administrative data sources cover many domains – mostly local in origin such as Council Tax records, housing benefit data, and information about local services such as social services, library membership, and housing conditions.

A majority are managed by the local authority but some sources such as the GP register require different levels of permission to access and use.

The GP register is the most strategic source of information as measured by its scope and coverage. It contains details of everyone in the UK registered with a doctor including variables such as sex, date of birth and address. Other useful registers include those for births and deaths.

Benefits and influence of this research

Professor Mayhew’s research has had considerable influence and impact at both local and national levels.

1. On the local level in his extensive collaboration with the London borough of Tower Hamlets (LBTH) but also other local authorities mostly in England
2. In the charity sector such as his work for children’s charity Buttle UK
3. At the national level via his expert testimony in Parliament and continued involvement in the development and planning of the Census for the future.

Supporting Councils & Local Authorities

In a recent study Mayhew, via nkm, was invited to employ his methodology in “Whole Systems Data Project” (WSDP) for the London borough of Tower Hamlets. The project targets health inequalities within Tower Hamlets and is the first of its kind to be launched in the field. The scope included not only the data sets mentioned above but also individual medical records.

The work is ongoing and results are being used by LBTH and clinical commissioning groups to evaluate, plan and deliver local services. The results can pinpoint for example links between social deprivation, health conditions and the uptake of medical services including hospital stays. An illustration of the impact the WSDP is having can be found in the nkm report on health inequalities and health care consumption (2019).

This report allowed the council to identify previously overlooked health and social issues such as the high correlation (>90%) between deprivation and health costs, the fact that older people in deprived areas have more long-term health conditions on average, the increased secondary care costs for persons living in social housing (£100/year) or benefits (£300/year), and other issues such as mental health, smoking behaviour and obesity.

These findings feed into the Joint Strategic Needs Assessments and inform the council’s decisions and actions in tackling these issues.

LBTH and Professor Mayhew have collaborated extensively in other successful initiatives, chief among them was the launch of the Selective Licensing Scheme, the decision for which was based on Prof Mayhew’s research and report, that highlighted the link between private rented properties and anti-social behaviour (ASB).

Selective licensing is part of a larger Government initiative to make landlords accountable for the upkeep of their properties as part of a drive to improve housing conditions in the private rented sector.

The scheme was launched in 2016 on a five-year cycle, is covering three wards and an estimated 5,900 properties. According to the council’s own consultations, the majority of tenants/business was supportive of the scheme, indicating that Prof Mayhew’s advice resonates not only with the local authorities but also with the people directly affected.

His contribution is considered of such critical value that at the end of the first 5-year cycle (2021) he is expected to review and report on the outcomes.

Around 20 local authorities have commissioned Professor Mayhew to conduct similar reporting of housing, ASB and private licensing schemes.

Contributions and Impact in Charities

Beyond the substantial work with local authorities, Professor Mayhew has contributed his expertise to charities and private businesses.

A prominent example of this kind of impact is the Crisis Points report for Buttle UK. The report was hailed by Buttle for helping the charity to raise awareness of the poor circumstances in which some families and young people live.

Furthermore, it provided the charity with its most comprehensive set of data, at that point, allowing them to focus their efforts on areas of unexpected crisis.

The impact achieved by Mayhew’s work is also evident in the influence it has exerted on Buttle’s attempts to identify and tackle child poverty via the ‘Chances for Children’ campaign, which was shaped to a large extent by nkm’s reporting.

The campaign aims to raise £20 million, which translates into £133 million of public expenditure saved over the next fifteen years.

Mayhew’s research has also contributed to other areas of larger social benefit such as healthcare, which has provided him with access to many, hitherto unexploited, data sets. For example, Mayhew worked with Specsavers Audiologists to prepare the “Listen Up Report”, which called attention to hearing loss issues that affect an ageing population.

It also underlined the enormous costs associated with the current situation by calculating that hearing loss is costing the UK economy more than £1 billion a year due to NHS costs, disability benefits costs, and lost output.

This helped to strengthen the overall call to action on the issue by revealing the unrecognised consequences of a particular aspect of ageing.

Professor Mayhew’s research has also achieved impact at a national level, specifically in the use of administrative data in the decennial Census of Population, last undertaken in 2011.

The Office of National Statistics (ONS) has approached Mayhew and his associate Gill Harper to further discuss development of methodologies and has publicly acknowledged the importance of their research and the fresh perspective it offers.

Mayhew’s contributions in shaping the debate culminated in an invitation to provide expert testimony and written evidence to the Public Administration Select Committee (PASC) at its Fifteenth Report of Session 2013–14, entitled “Too Soon to Scrap the Census”. The committee’s recommendations mirrored Mayhew’s advice.

Mayhew’s persistent championing of the value of using associative matching of administrative data has culminated in the adoption of these methodologies, via the recommendations of the National Statistician and the ONS as is evident in the “Help Shape our Future” White paper.

This paper lays out the direction of the production of official statistics in the future, with administrative data at the heart of the system.

This adoption is expected to produce a wide range of benefits for multiple users, as laid out in the white paper, from service planning and policy-making to market analysis and academic research.