Place-based Directory of Service Model
This Place-based model is an enabler for frontline workers, volunteers, service providers and statutory organisations from all sectors to participate in collecting, refining and maintaining accurate information about the services provided by local voluntary organisations, charities and statutory bodies.
It makes use of open data standards e.g. Open Referral UK and OpenActive Data Standards to enable information to be collected by different organisations and applications then aggregated into a single digital catalogue.
The model also moves away from the approach used by a traditional service directory where someone would search for what they wanted to find, to an approach where someone would present their needs and circumstances and appropriate services would be suggested to them through a data model.
The information details are specific and at a service level rather than an organisation level so that the client doesn’t need to search a website or speak to the provider to find out which and when is the right service for them.
The below conceptual diagram is the current model as of the end of 2021 but we expect to build on this further in 2022.
Service providers
The ultimate aim is that the providers of the services will maintain their own data. However, we need to make this really easy for them and ask them to do it only once. As we progress to place-based working then the 'ask once, share many' will become the 'business as usual' providing the most efficient use of tax payers money but at this moment in time we know we are not there yet. For this reason and along with the various degrees of digital maturity, we are implementing three roles of service providers and supporting them with an assurer as appropriate (see next element).
- Trusted Service Provider: Those service providers who are generally larger and have been commissioned by the public sector are happy and able to maintain their own data. We would advocate including this in any grant agreements. These are known as trusted service providers and will be accountable for their own information.
- Service Provider (Monitored): These are happy with updating their own data but may have limited resources and so an assurer will keep a small oversight and prompt them as necessary.
- Service Provider (by proxy): The third group are those that are likely to be hyper local and we want to help them as much as possible so they will have an assurer to add their information to the data-lake.
Assurers
We expect the role of assurers to diminish over time but they are essential to establish this new way of working. The main reason that silo based directory of services exist is because organisations do not trust each other to maintain accurate data. Ironically, the result is diluted resources in maintaining various pockets of data and everyone loses. The assurers accountability and facilitation of the federated approach to collection and maintenance provides the trust which enables frontline organisations to focus on their target audiences/context and simply use the data collected by others.
We would recommend that there are teams of assurers covering the geography of the ICS but one assurer (or team) should work with one service provider for all their services. A register of service provider to assurance teams will minimise duplication.
Frontline workers
The frontline workers are the main beneficiary of the model at this stage as they are reliant on their local knowledge and are frustrated at the lack of accurate information available to them. Employers would be able to move workers around and recruit more people without strong local knowledge as the frontline applications will empower them.
The frontline workers make for an army/crowd that can suppport the maintenance of the local data. They generally find out when services have ceased or changed. Providing a digital capability to allow frontline workers an easy way to report new services, changes and closures directly to the assurers is a great added value.
Volunteers
Volunteers are an extension of the frontline workers and so the same applies. Give them an easy ability to report new services, changes and closures directly to the assurers will add to the team work.
Collector Tools
The advantage of adopting an open data standard like Open Referral UK means that any application can collect data in the same format so that it can be published and aggregated. See Aggregation for further ideas about collections. However, Open Referral is open to some interpretation and so an ICS should develop an application profile to explain their interpretation including assurance rules and taxonomies.
Ideally the UK public sector would adopt the same application profile so that suppliers could resell to any public body. Also managers could understand a strategic view of local service supply and demand (see commissioner dashboard below) across the geographic and administrative boundaries and not be tied to different interpretations preventing this.
Website Widget
As stated earlier, the aim is for Service Providers to maintain their own data but it makes sense that we minimise this burden. Lancashire & South Cumbria have co-produced a Service Provider Website Widget which can easily be added to the providers website and will then show all their services and allow their users to print off the details. This saves the provider adding service details to their website.
Our intention is to make this technology available to infrastructure organisations so they can showcase the services that they support and have started to work with a couple of CVS to develop this.
This is also being extended to include an ability for the trusted provider to maintain their own data and push directly to the aggregation.
Commissioner Dashboard
One of the advantages of this model using open data across the place is that it can be aggregated and provide management information that can be grouped in different ways in any geography. It is simply to understand the sufficiency of local services especially if the same taxonomies are used. It is a simple query to understand all the support for struggling families as they will have all been tagged with the service-type 'Financial advice , support and free goods'. Decisions can then be made as to where to focus resources.
Understanding demand is slightly more difficult and involves understanding what is being searched for and shared with clients. This is something that we will focus on later but initial thinking is that an agreement for an organisation to extract the assured data through an API could also include a requirement to provide their search data through an API.
Translator
We have found one issue that collecting of open data doesn't resolve and that is making it available in other languages. It obviously doesn't make sense to ask the source to offer other languages and it requires effort and some degree of 'wording' to get the right translation. It would be nice to get it right once and then share with everyone else. There are tools to convert languages out there and in use but we have adopted a place-based model for translation and co-produced a tool with Lancashire. Currently the tool only works with Service Finder (see below) but there will be an API for any application to consume the languages.
The tool allows a Service Finder user to see what languages are availble for a particular service. If the required language is not present then it can be requested or added by that person. Once added and approved then it is made available through Service Finder.
Orcha
Covid has seen an unprecedented rise in the uptake and provision of digital health solutions. A big issue remains as to which digital solutions a frontline worker can recommend. The place-based model allows all apps and online services to be added and it is the quality 'review' data that will give frontline workers the confidence to make recommendations.
We have worked with Orcha who review digital health applications. They provide information and reviews to allow an organisation to consider which apps should be recommended by their frontline workers. We have created a process that uses the Orcha API to publish the recommended apps using the quality review data set in open referral.
Aggregation
It is tempting to try to provide a single means for collecting open data so that there is no duplication of effort. However, it is highly unlikely that this will work in practice. Our current thinking is for many collectors to make use of open data standards like OpenActive and open referral such that their contributions can be aggregated together. This does require collecting applications to make their data available and it be pulled into an aggregated directory. This will require an aggregator application. This application will then provide a set of APIs to allow the data to be extracted. See frontline applications below.
We are currently co-producing a tool with Lancashire which will provide a simple dashboard to manage assurance and potential duplication. Trusted assurers/publishers can pass content straight through whereas monitor and proxy sources can have assurance applied. We are also working on a register of approved/preferred sources to minimise duplication.
Frontline workers and their applications
The key to this model is that many software applications and frontline teams will benefit from using the assured local support services data. They don't want to be burdened with having to collect and maintain their own data but rather that they can use trusted up to date data. The aggregator will make the data available to applications through a set of APIs offering various features such as proximity to a post code or for a particular circumstance or a type of service and can check whether a client is eligibility or not. The APIs also include an export as a CSV file (generally used as spreadsheets) if a compliant application is not available.
Service Finder
We have co-produced a Service Finder application for frontline workers working primarily with Social Prescribing Link Workers, Community Connectors and Mental health workers. The application is web-based and responsive to the size of the device being used. It allows different teams to configure it to focus on their particular client context and to make things quick and easy so it can be used with clients on site.
It consumes open referral and OpenActive data and makes suggestions to the user of appropriate local support for the needs and circumstances of their client.
Self-care: Citizen, Carer/Family/Friends and Charities
Our current focus is on empowering frontline workers but the model should provide a significant boost to self-care. The name self-care suggests that the patient/citizen will access the information but the majority case is that someone else will be supporting the person in need like their family or firends or carer or a local charity. The information provided to them will help them support the person in preventing issues and improving their quality of life.
Taxonomy model
One of the issues with the traditional Directory of Services was that they often asked service providers to categorise their own service. Inevitably this creates an inconsistency of categorisation for the same services in different geographical areas. It also usually meant that the service was over-categorised i.e. the service was deemed to add-value to a lot of categories rather than its key target.
As services can be useful for lots of different contexts then it might either have very general categories or lots of different categories for each context. No solution was very simple for the provider of information or consumer of information.
We believe a better model is for the service to simply indicate which type of service it is. It is then possible to map several front-end context criteria to the appropriate service-types. This makes it easy for a service provider, provides consistency across the services, provides for a variety of contexts and allows the updating of the model to be carried out easily.
We have also extended the OR standard to include a target audience which allows a service to specifically say if they are a service for a particular issue e.g. unemployed or learning disabilities or older people etc.