Why is “vulnerability” important?

One of the most intransigent and most frequently identified issues during the Discovery workshops, was that agencies could sometimes be unwilling to share information; their interpretation of “secondary use legislation” was different and there was an anxiety that the purpose of the data sharing was not something that they felt met their own understanding of the necessary legislation. What has emerged through the project is that a clearly defined
(to the data sharing) that is expressed in terms of
@Vulnerability Attributes
, is likely to facilitate speedier and more law-abiding sharing of data. Thus, for example, if the purpose of the data sharing is to prevent homelessness, then a specific named data set should be defined with respect to the insight (or attribute) that the data brings to the process of identifying people at risk of homelessness. This approach will deliver a data sharing request and process that will neatly fall in line with the necessary legislation.

The role of the SAVVI programme, therefore, is not to offer a semantic definition of vulnerability, but to identify the data sets (and the related data standards) that indicate different attributes or circumstances about a particular individual or household. These attributes or circumstances are those that are more likely to be relevant to identifying people suffering or at risk of suffering different types of vulnerability (homelessness for example). Thus for example, the Government in 2018 documented a
; SAVVI is not concerned by the semantic definition of loneliness, but there are a number of sets of data or assessments that exist across Local Government and partners, that will indicate whether someone is living alone or is bereaved for example (which are both indicators that someone might be at higher risk of “being lonely”).

The role of data and data standards

Individual sets of data offer some insight into the circumstances about individuals and / or everyone living at that address - provided there is a legal case for using the information for this purpose. When brought together (again supported by an appropriate legal gateway), the data sets can offer more accurate and more nuanced insight - but only if the
are properly understood in terms of exactly what they represent, how recently they were checked (and how they were checked) and whether they apply for example to everyone or only one person at the household.

These data sets therefore offer indicate what is happening (and perhaps what has recently happened) at an address or to an individual (as long as the data is maintained and accurate). This information is very important in providing insight into the comparable level of need between different individuals and households; and hence in the context of any initiative like supporting people during COVID-19 lockdown, informing decisions about who to target first for support.

At the same time, there is an increasing interest in
@Dynamic Risk Modelling
to take these data sets and move beyond a static view of current need and offer a prediction about which
or individual is statistically at higher risk (or chance) or suffering a specific outcome in the future. The first example is the work that is being delivered by Huntingdonshire DC to build a
@Risk Model
@Risk Algorithm
to identify individuals who, based upon their current and historic
and events, are at higher risk of becoming homeless in the future. The most widely recognised and adopted risk modelling within the Public Sector is the use of a model to identify new claims for housing benefit that are most likely to be fraudulent; this allows councils to target more resource at verifying claims that are more likely to be fraudulent (and less effort where there is less risk of fraud). The approach is called Risk Based Verification

In relation to “risk modelling”, the purpose of the SAVVI programme is to identify and offer initial definition of the
for the data sets that are used in the Huntingdonshire risk model. Where these standards are adopted, then this will allow other Local Authorities to use the same model - by running the
@Risk Algorithm
against data sets that are captured, verified and maintained / reviewed to the same standard. Longer-term, it is easy to understand therefore how the SAVVI work can broaden to cover other data sets and be supportive of other sets of risk modelling.

What about the link to “systems / software”?

One of the strands of work within the SAVVI programme has been to engage with providers of software solutions that are used across the Local Government market. How data is held within software or a database is often defined by the software provider; the local authority simply adopts the underlying data standards that the provider has written into their system. The purpose of the engagement with the market, therefore, is to share these standards to encourage their adoption. This will ensure that different sets of software will define and hold data to the same standards - thus whenever data is being shared or used (for whatever purpose) there will be no need for an extensive investment of effort to learn and then allow for the stand-alone data standards in each system.

What data exists therefore that defines vulnerability?

To date, SAVVI has identified specific sets of data and has started to capture data standards for 20 different
@Vulnerability Attributes
. These are data sets that were identified through the initial Discovery workshops related to COVID-19 or homelessness. These are defined within the
. These 20 data sets represent a very small example of the totality of the range of indicators of vulnerability across the sector. The ambition remains to extend the number of data sets and data standards that are defined and hence can be adopted by software providers and promoted by councils and their partners.

describes these data sets, the data flows related to them and the data standards that should be adopted. This information may be of value in isolation, for example, where a Local Authority is setting out to procure a new system, to ensure that the specification sets out the data standards expected. It is more likely, that the information will be valuable where a council and / or partners is setting out on a defined initiative that is about identifying specific types of vulnerable people. The following pages, therefore, set out a headline end-to-end process; and identifies how and which
could be adopted and applied.

Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
) instead.