Enabling Social Prescribing

Project Deliverables

User Research

Technical Research

Project Management

Resources

Tooling Logins

Coda doc filters

Copy of Numbers and measures

Explore

Project Deliverables

Old Research Findings Report

Will be complete by 10am 23rd March.

1.0 Introduction

This document is an Executive Summary of the research and findings looking at potential strategies, and tooling or data standard refinements to OpenActive to try to position it as a data standard inside Social Prescribing. The work was commissioned by the Open Data Institute who are the custodians of the OpenActive data standard on behalf of Sport England. Digital Gaps Ltd were commissioned to deliver the between Mid-January to Mid-March 2021.

The research should help foster rich local ecosystems of prescribing opportunities by:

Exploring requirements for, and barriers to, data access and data sharing for social prescription at the local (council) level, including physical activity data

Developing strategies for surmounting identified barriers, and measuring the success of these

Undertake a gap analysis of OpenActive data standards and tooling with regard to these strategies

2.0 Background

Yet to be completed will be sorted by 22nd March

Pull wording from proposal

Social Prescribing is a big Government strategy

Sport is targeting SP- reference the Sport England Towards An Active Vision programme

Open Data and Digital Transformation

An unhealthy country

In theory Open Data provides ubiquitous access to information that can be searched and

The standard was built initially to help NGB to get q view of what is happening. The roll-out of OpenActive is going well. The question is whether there should be some amendments made to strategies, the data fields and the tooling to make it fit for purpose for social prescribing.

The research was delivered in Pennine Lancashire. We should reference TaAF as a partner to this work

3.0 User Research Headlines

The following bullet points give an easy read summary of the research findings. The section toward the end of this document outlines the research approach and provides summaries of each step with pointers to the research material. The full research is documented in Coda and provides a better experience if the reader signs up for free Coda account. Details are given at the very end of this document.

Less than 10% of social prescribing referrals are to physical activity

Social prescribers say physical activity is not at the forefront of their clients’ minds

Social prescribers say clients are more interested in local activity that is easy to get to, doesn’t cost a lot and has interest and social value rather than just be exercise.

There are very few activities found when checking within a couple of miles of a client’s postcode.

There is a general mistrust of any directory of activity information by Social Prescribers however there is a trust of their peers opinions

There is concern for their client attending an unsuitable activity, for their anxiety to attend on their own and if an issue occurs whether the activity provider will help

Local activity providers are not aware that they can advertise their sessions using OpenActive, once they know they are keen to do it

Managers and workers are not aware of the benefits that open data can provide and prefer to gather their own information. However, once explained then they do see the benefit but can only really impact the critical mass as one part of the whole

Application providers supporting social prescribers are frustrated that there is not a rich set of reliable data describing activities for them to consume.

Consider others after final review

4.0 Key issues & recommendations

The hypothesis or vision is that OpenActive as an Open Data standard can become a key part of Social Prescribing to underpin increased referrals into physical activity. The opportunity to realise this vision will rely upon a broad range of factors, some of which are within and others outside the direct influence of ODI.

The following section summarises these key factors under three headings:

4.1 Cultural and Process Key Issues and Opportunities

4.2 Standards and tooling enhancements

4.3 Further Considerations

4.1 Culture & Process Barriers

There are three sets of summary barriers that are holding back the adoption of Open Data generally and that specifically affect the role that OpenActive can play in Social Prescribing. Below we briefly summarise these barriers, and then identify a range of opportunities or recommendations that ODI can take forward to mitigate those barriers.

4.1.1 OpenActive is part of an Open Data Public Sector Strategy

OpenActive is playing a growing role within the sport and leisure sphere. Over the next 10 years, adoption of Open Data generally will inevitably increase; OpenActive will benefit in terms of take-up as this broad trend continues. Social Prescribing is a high-profile initiative and a busy marketplace with a number of national initiatives and an increasing number of private sector suppliers targeting the area with software, training and consultancy offers.

It is an inevitability that Open Data will become part of Social Prescribing, again perhaps not quickly, but certainly over time; it is less inevitable that OpenActive will become part of this eco-system as a stand-alone data standard. Whilst continuing to promote adoption into the sport and leisure world, the opportunity to position OpenActive into the hectic and but highly fragmented Social Prescribing market may benefit from some of the following potential approaches that seek to align and embed OpenActive within a broader Open Data strategy.

4.1.1 Recommendations

Ensure Open Data is a simple message and resolve alignment between Open Referral UK and OpenActive

For Open Data to become a reality within Social Prescribing processes requires alignment across the identified 5 stakeholder types. Furthermore, most of these stakeholders work in highly fragmented ways, for example

There are 1,250 PCNs nationally and most of the decisions and leadership about the process, the culture and the behaviour of Social Prescribers are taken at this level

Within any PCN area, there may be in excess of 100 hyper-local service offers that have value to a Social Prescribing process – and hence all 100 will need to adopt the standard for the data to be available

There will continue to be growth in the numbers of suppliers that enter the marketplace (for example Social Care providers who will create an SP module).

Where these stakeholders are faced by options about adoption of either Open Referral UK or OpenActive, this may create confusion. In the high-pressured and fragmented world in which they work, stakeholders may then adopt pragmatic solutions and adopt neither standard. There will be plenty of private sector and local influences that will be pushing un-structured data capture of the information and a reliance on Google searches. This is a simple message as the competitor to Open Data and hence the Open Data proposition must be simple to compete. The technical recommendations detail the options for how to create this alignment.

Help create cross-sector National Open Data Leadership and Governance

As the custodian of OpenActive, the ODI, alongside Sport England, should play a role within the national governance and national leadership of Open Data. To take this forward, ODI and Sport England would benefit from aligning with the Open Community programme that is sponsored by MHCLG and ensuring that the Government Lead on Open Data, Sam Roberts, is brought into the discussion on bringing together that Governance. This role would focus on two main responsibilities perhaps – firstly, to make decisions on issues such as taxonomies and new data fields and secondly to create business case templates and case studies to help drive adoption of Open Data.

Without this national cross-sector governance, each individual agency will adopt their own data standard or policy related to overlapping considerations. The result will leave duplicated sets of data across these sectors and will hamper the market for innovation (because tools will only be able to work with a smaller set of data), thus undermining the very core of the business case for Open data in the first place.

For example, the following are all considerations that will need to be resolved, where single joined-up governance will underpin the cross-sector business for Open Data:

Defining where to capture and how to define intensity levels. ManchesterActive have set out to define their approach. Inevitably other Active and other social prescribing partnerships will be defining their own solutions.

Having a standard related to safeguarding checks. ManchesterActive have defined their approach and Westminster have defined theirs. Where an organisation delivers services in both localities they will be checked through different models. As other localities develop their own approaches, so the confusion will increase

Having standards related to accessibility of venues and also services.

Porism and Digital Gaps can facilitate some introductions to help move this forward.

Champion the message that Social Prescribing is a wider concept, than solely a GP referral mechanism

Social Prescribing is firmly set within the NHS domain. The summary document on

Numbers and Measures⁠

offers some insight into the broader set of potential referral processes that are about social prescription in their nature. The amendments to the data standards that are recommended, all recognise this slightly broader social prescribing role.

ODI and Sport England might consider funding some pilots that look to scope out how demand and supply of open data can be developed to support some of these wider examples. For example, Mental health as a pathway is an area where increased exercise is known to be beneficial. Working with an MH trust and a council in one small area would allow a pathway to be defined that would ensure, for example, that anyone who does not meet the threshold for support from a Community Mental health Team is immediately connected into a range of low-level exercise groups locally to where they live.

The moment when someone is told that they are not “unwell” enough to be given any help offers an opportunity to create some motivation to change. Signposting them to local groups as part of the process and connecting them to volunteers would be an excellent opportunity.

Creative Support is a walking football service in Pennine Lancashire that is delivered by qualified Mental Health professionals. They would be willing to develop some headline definition of the potential process, identify a range of local activity offers and help pilot how this model might work.

Similarly, through the interviews, we met with a Social worker from within the CMHT in Hammersmith and Fulham. They may be willing to consider documentation of similar example pathway.

4.1.2 Digital Tick Box Culture at a Local Level

Adoption of Open Data needs more focused co-ordination and leadership locally. Adoption is guilty of a tick-box approach where, for example, councils have signed up to the Digital Declaration, but the resolve to hold suppliers to account, and to drive changes into the organisation is lacking. Similarly the strategic Open Data ambitions of DCMS, NHSi, as well as Sport England are not matched in terms of local implementation. Without stronger messages, the tick-box culture will prevail, with local partners claiming to have implemented data standards, but without the required attention to detail and without the necessary promotion to internal partners and to local activity providers.

Successful adoption of Open Data relies on co-ordination of what happens cross-sector at a local level. It is perhaps not helpful to define “local” too prescriptively, as this will differ depending on local leadership behaviours and a range of factors. However, there are a range of actions and messages that ODI and Open Data Leaders should be championing into local stakeholders. These messages should focus on the benefits to those local partners, namely councils, NHS organisations (mainly Primary Care – CCGs tend to be very focused on Acute Hospital issues and hence wherever possible the focus should on the PCNs), Sport and Leisure, the Police, the third sector etc.

The messages and actions must focus on the benefits to stakeholders, most importantly access to accurate information and secondly the efficiency from capturing and maintaining information once only. Adoption of Open Data per se has little direct impact; it is what it enables that is the benefit. The following offers some options for how to help champion the right messages.

The following recommendations are some examples of case studies, materials and actions that can be taken to help shift the digital dial to mature.

4.1.2 Recommendations

Procurement

It is recognised that one of the issues for the success of OpenActive is that it requires an ecosystem where there is supply and demand. This is not always easy to put in place (see working as a place below). However, the public sector can use procurement to promote compliance to open standards. Currently the public sector is not doing enough to provide an infrastructure that complies with OpenActive data standards. One reason for this is the lack of a real understanding of the benefits of open data (see Open data working as a place below) and so any compliance is treated as a ‘ticking the box’ exercise rather than seeking to gain the benefits.

The lack of use of procurement and funding/grant power effectively makes the task of creating the ecosystem very difficult. It takes time and money for existing applications and processes to take advantage of the open data standard.

The NHS is currently considering what can be done to scale Social Prescribing. The HSSF procurement framework is likely to insist on Social Prescribing applications integrating with Open Referral and OpenActive APIs. Also note that they will ask for other details which may be different to other sectors and so put a burden on activity providers - see Cross Sector and Place Governance above).

We recommend that the ODI, perhaps with the cross-sector and place governance team, work with Crown Commercial Services to document appropriate paragraphs that can be included into relevant specifications. This will ensure that only software providers that are compliant with OpenActive and other open data standards are procured.

Funding / commissioning

The above recommendation is to embed the data standards into procurement specifications. At the same time, the ODI, alongside the Cross Sector Open Data Working party, should also approach the Sport NGBs and the key third sector funding bodies as well as. They should be requested / ordered to mandate that wherever funding is awarded, then the definition of the activity must be defined in adherence to OpenActive.

Furthermore, to help tighten adherence to the promises made through the MHCLG Digital Declaration, funding should be prioritised at areas that signed up to this commitment. Ideally, this would go a step further and require some evidence of the progress that councils have made in progressing their declarations. Placing this pressure on the council as local place-leaders will not only increase adoption of the standard, but it will also help local sports organisations strengthen their relationship with the council which will be important to assure the creation of a strong demand and supply eco-system.

Similarly the local public sector commission significant local third sector delivery and should ensure that this adoption of Open Data is similarly cascaded out to the local third sector. This will dovetail with messages that come nationally from the national third sector funders.

Blueprint

We recommend that the ODI work alongside a pilot local area to develop a “blueprint” that will set out a headline plan with supporting tools and guidance on how to implement an OpenActive ecosystem to realise the benefits across a place. This should include ‘understanding the benefits of open data’ for managers and workers and ensuring they have the right applications, skills and equipment.

The best examples are perhaps

Manchester Active⁠

Westminster Active⁠

and

London Sport⁠

. The opportunity exists to unpick what they have achieved and re-build the templates and the key decisions they have taken. The interface to Social Prescribing exists as an ambition and plan in both Westminster and Manchester Active; as they start to plan the detail of the implementation it may be beneficial to capture the live learning through that process, which will provide a fantastic set of materials for others to adopt. Pennine Lancashire could provide some “testing” of how these materials are put into practice as they follow behind with the plans related to their Together an Active Future programme.

Place-based example business case

One of the challenges to Open Data adoption is that the business case only really stacks up when it is widely adopted across a place. There could be as many as 20 local referral pathways (Social Prescribing being one example) that may benefit from access to accurate information about local community-based activities and groups.

If all 20 pathways adopt their own data standard and define this separately then the reality is that there will be no business case. Where each maintains a separate record of local offers the duplication and the lack of accuracy and reliability occurs.

The benefit that the demand-side (Social Prescribers and other local referral pathways) are seeking is reliable accurate data. The benefit that the suppliers are seeking is aggregation of demand (all of the pathways together), alongside the opportunity to capture information once for all of the pathways (rather than offering separately for each separate pathway).

Open Data can be as inaccurate as non-structured data. The messages should focus on delivering accurate data, and not open data. The messages should focus on efficiency of collecting once and using numerous times. Leading with the data standard, turns off the leaders.

Focus on aggregating demand and the value of aligned taxonomies

The market for adoption of open data lacks demand. The messages should be demand-led, with a focus on the actual search process. Taxonomies are key, key, key.

Working as a place (creating the ecosystem)

Our research has identified that five main stakeholders are involved in the end-to-end process that surfaces activities through OpenActive open data. (See

stakeholders⁠

in the Research section at the bottom of this document) One of the stakeholders is an application provider but they could provide an application to collect & publish the data and an application which will consume the data. All the stakeholders need to work together to form the end to end process. There is little point an activity provider loading up their session data if there is not a social prescribing application provider being used by a frontline worker looking for an activity session for their service user. Equally there is no point a social prescriber looking for a local session near to their client’s address so that they can walk to it because they don’t have time or money to go anywhere else and find that there are very few activities available.

The ODI has given the market the chance and it has become clear that the ecosystem will not emerge over time. Playwaze have had some success xxxxx but xxxx

Manchester Active, Westminster Active and London Sport all seem to be progressing with a place-based ecosystem approach. They are generally a single sector implementation i.e. Leisure but with links to Councils and CCGs. The ideal solution would be if this could be implemented by the 42 new Integrated Care Systems which are cross-sector by nature. These may be too large an area and it may be that starting with an ICP and scaling across ICPs to become an ICS might be a more pragmatic approach.

A place-based ecosystem will need to prepare the stakeholders:

The implementation will depend on managers at various levels and in different organisations. They will need to be on board with the benefits of open data or they could find ways to just ‘tick boxes’ and end up frustrating the ecosystem.

Frontline Workers will need to be confident that they have the means to identify suitable sessions

Application Provider software will need to be compliant with OpenActive

Activity Provider will need to have a means to easily maintain their session data

Our research interviewed managers involved in Social Prescribing and found a mix of views, approaches and frustrations. Some understood the benefits of open data but didn’t believe that it would ever reach critical mass. Some did not understand open data but were happy to comply as long as it didn’t detract from their day job. Our interview of SPLWs identified a lack of training and lack of digital devices.

Page Break

4.1.3 Adopt a holistic approach aligned to health conditions and use the language of pathways

It is recommended that the ODI looks to identify some strong examples or case studies of holistic approaches where exercise classes / sessions (that are defined using OpenActive) can be combined alongside other non-exercise related recommendations to help drive good health outcomes. The development of these should be led through some small cross-sector pilots, where they can consider key target cohorts and design some simple sets of recommendations (such as attending a social walking group, undertaking a one-week dietary analysis and joining an online (non-Zoom – as video often turns people off) stretching class to music.

This holistic approach is recommended for four reasons

Firstly, the strongest evidence for good health outcomes is where exercise forms part of a wider set of behavioural changes. There is good evidence base for the benefit of exercise on certain conditions (in particular MH for example), but creating a holistic set of interventions allows exercise to be hidden within a broader set of interventions

Secondly, it will soften the hard sell required to some Social Prescribers who will certainly remain ardently against motivating people to take up exercise. All Social Prescribers will be able to sell the beneficial set of recommendations that include exercise, hence easing some of the backlash experienced during the interviews

Thirdly, it is certainly true that many of the patients being supported by link workers provide infertile ground for driving take up of activity

Finally, this is also an easier sell to the 1,250 PCNs. Health has a very academic approach to evaluation and evidence and a broader focus on doing groups and trying to persuade people that exercise is the answer may be a tougher sale than selling a vision that there are identified pathways that might work well together that perhaps balances exercise with diet and social activities.

One manager commented that the focus should be on getting people moving in whatever activity they are interested in e.g. Gardening can get people exercising without them noticing, rather than getting them to do physical activity. Gardening does seem to be on the OpenActive activity list but it seems on its own as a broader physical activity. This sort of thinking comes up when considering the alignment of OA and OR (See Technical Issues section below).

Focusing on physical activity for now, another issue for Social Prescribers is that OpenActive basically provides a choice of physical activity sessions to someone interested in finding a session. This generally is an activity finder using a keyword search. This is great for those that want to book a tennis court but the majority of Social Prescribing clients and, anecdotally, the in-active or under-active cohort do not have a specific interest.

Again this is not directly in scope of the ODI but we think some work with Social Prescribers could identify some pathways or similar ideas such that a Social Prescriber can carry out a simple assessment or select particular defined pathways to then be provided with some ideas that might suit that particular person.

If these models were made available to application providers then they could implement them to enhance their activity finders. Certainly, the

market leaders in Social Prescribing⁠

software have different means to find the right services and no doubt they will support finding the right activities. However, the means to implementing pathways and similar models will be dependent in some ways on the taxonomies (See Taxonomies in Standards & Tooling key issues below). The NHS are advocating

Snomed terms⁠

for Social Prescribing and we have not as yet seen any examples of how that will work.

4.1.3 Recommendation

The holistic approach does not provide a clear recommendation for ODI but is an important understanding to provide context for the use of OpenActive by Social Prescribers. For pathway ideas, we recommend that the ODI carries out a project to define a mapping between broader holistic/life based terms that Social Prescribers can select to have OpenActive activities suggested. The pathway idea will link with taxonomies and the recommendations from that will need to be borne in mind.

For information

Digital Gaps⁠

is looking to start working with two GP practices, namely Norwood surgery in Southport and Central Waltham Forest in London with a very simple example of using a pathway mapping to physical activity in the pre-diabetic context. The work is under-resourced and would benefit from some additional funding to deliver a more robust outcome.

4.2 Standards & Tooling Enhancements

This section provides the conclusions formed by

Digital Gaps⁠

and

Porism⁠

in carrying out the user and technical research with some recommendations to ODI.  In many ways this represents the real heart of the work. There is a separate

Technical Review⁠

document but this section attempts to summarise for a non-technical audience.

There are, we believe, two keys barriers that are being addressed here:

Firstly, looking to create a clearer delineation between the core public sector-funded infrastructure and the opportunity for innovators to deliver into the resulting market.  At the same time, the public-sector funded infrastructure must help deliver a more transparent and more attractive market that entices more innovation

Secondly, recommending some additional fields within the standard and some

tooling ideas to provide the additional information and support that is required to tailor OpenActive for Social Prescribers to look to create more demand through the system.

4.2.1 Clearer delineation of the commercial opportunities through open definition of standards

Means to access the data: API v Feed

The biggest technical barrier to application providers looking to consume OpenActive data is the complexity and cost in consuming the data from the RDPE feed. The more accepted model preferred by developers is to use REST API. Our understanding is that the model of adopting a standard API previously has been due to it being easier for publishers (those who collect OpenActive data and make it available) to use the feed and that there is a risk of Distributed Denial of Service (DDOS) in other words crashing the publisher’s system. . The technical review report details and explains how an API can be implemented without DDOS risk. The thinking is that the benefits to those consuming the published data will be greater than the risk of having an API.

There are

51

⁠

existing feeds and the hope would be that they were prepared to provide the standard API which will allow application developers of activity finder products to consume the data more easily. At the same time it is recognised that it is easier for the application providers to provide a feed rather than an API. The hope comes from an increased demand for data which may create the case to shift to the API but if they don’t then this will continue to provide space in the market for innovators to provide feed to API tools. 

The real benefit comes from a ‘standard’ API which will mean that application developers can invest in products knowing that they can sell them to all of the ‘places’ (e.g. the 33 UKActive areas of the country) consume data from any source collecting OpenActive data.

As demand for the data continues to grow, this will stimulate the market to produce various applications building on the benefits that OpenActive delivers.  

Recommendation: 

It is recommended that ODI develop a definition of a REST API standard for accessing

@Session Series⁠

data from a publisher (what does publisher mean? Is this clear? Are we recommending that there is a single publisher or is published any software provider?). Through documentation of the standard, this will more clearly delineate what is available and will allow the market to potentially evolve as follows

It will create an opportunity for other developers to deliver more targeted APIs which will drive competition. These would adopt the standards, but perhaps only focus on particular localities or sub-sets of the data where there is an existing and stronger business case.

There will be a market for developing Feed to API products, again that could cover the breadth of the growing data-sets or focus on more profitable subsets

This may also create a marketplace for example for more NGBs to subsidise the development of products that target their sports where that market needs some stimulation

More details are given in the technical requirements document

here⁠

⁠

OA and OR convertors⁠

The alignment of OpenActive with Open Referral has been mentioned by a number of research stakeholders. Physical activity is a subset of the services and activities that a

@SPLW⁠

will be likely to make referrals into. Based on the types of people currently being supported and the increasing focus on

@SPLW⁠

throughput, it is inevitable that referral into physical activity will always be marginal compared to the wider set of support services sought.

It is obvious that an alignment between the two standards needs to be made clear to all interested parties and this is referenced as part of the Cultural/Process issues within this document. Application providers, subject to there being a return on any investment, would be happy to implement integration with OR and OA if that is what was required. The technical review focused on whether it would be possible for the two data standards to co-exist and provide converters between formats. 

Recommendation: 

It is possible to convert both ways but each would have

fields missing⁠

. It makes more sense for OR to consume OA. It is recommended to trial some of the conversions to be consumed by Open Referral service finders.

ODI should engage with the custodians of Open Referral to explore the mutual benefits and the way forward. The big picture will hopefully prevail – that there is a focus on them being stronger together rather than a focus on their relative position which may result in a decision to “compete” and leave data consumers, activity providers and software providers (both activity / service finders and activity application providers) to choose (one, the other or both).

4.2.2 Enhancing the standard and the tools to address the reasonable requirements of Social Prescribing

Data richness and quality 

The

@SPLW⁠

needs to be confident that the data is up to date. There is a modified date included in the feed but this should be brought into the data set so it can be consumed by the activity finder (or the social prescribing software) and can provide that confidence and evidence.

Additionally, there is a new field to show an assurance date so that an activity provider can confirm that their data is up to date. Data assurance wasn’t an issue for

@SPLW⁠

but this is probably because they are currently checking that all information is correct themselves. However, as productivity in social prescribing becomes more of an issue then data assurance will be a concern. The best and cheapest form of data assurance would be that carried out by the activity provider themselves but this may need further support for small providers as open referral has discovered. 

The OpenActive data should provide data that a

@SPLW⁠

needs in order to identify a suitable activity for their client. Our research has identified a set of fields that have been requested by

@SPLW⁠

. We have labelled these as SP ready fields indicating that these are the fields required by a

@SPLW⁠

and they can be viewed in

Coda here - Social Prescribing Data Requirements.⁠

⁠

However, having data attributes in the data standard will not guarantee that it will be populated.  Essentially this is down to the Activity Provider but if they want to reach an

@SPLW⁠

audience then they will make the effort and some tooling could help. Using the SP ready fields as a profile then a completeness checker would review the data entered and could return messages to the provider, alerting them to any missing fields or attributes.

Obviously not everyone will be concerned about meeting Social Prescribing needs so it is recommended there that is another tool for the frontend (in the activity finder or social prescribers software tool) that would check the richness of the data and allow a

@SPLW⁠

the option to filter out activities that do not meet the SP requirements. This then removes the risk of viewing multiple activity offers that don’t offer the key support that they are seeking for their clients (

@facilitator_type ⁠

for example). This will improve the experience for the end-user and retain their trust and faith in the process and the Open data

A further issue raised by the research was that often descriptions and definitions don’t offer the real understanding of a potential activity option. “A picture paints a thousand words”; the recommendation is that including the necessary fields and standards for photos to be uploaded may drive increased confidence, understanding & trust, and could also reduce the ‘anxiety of the unknown’ to

@SPLW⁠

clients who have never attended this particular activity. 

Recommendation: 

The technical requirements document and the data requirements document include the proposed fields and it is recommended that the OpenActive technical team review these.

It is recommended that the ODI trial photos of activity provider sessions with

@SPLW⁠

in Pennine Lancashire to understand if this will make a difference. Most activity providers already have websites and facebook pages with photos and so this should not be a big burden on them. 

Activity & Provider Quality Assurance

Having rich and up to date data does not remove the risk of the activity quality or whether the provider can be trusted. This was a core underlying concern for Social Prescribers through the interview and most wanted to actually visit and experience something before they would ever make a referral. Quest (

https://questnbs.org/active-communities/introduction-to-active-communities⁠

) is an example of a Service quality kitemark that would give assurance of safe-guarding or other policy requirements. This is supported by Sport England but we do not know how useful that would be to

@SPLW⁠

nor do we know the burden impact on the activity provider. It is noted that MCRActive will be carrying out their own provider checks but it is not yet known how they will communicate this to

@SPLW⁠

Establishing the trust of the activity provider should not create an insurmountable barrier to very small activity providers and it should always be a case by case risk based decision as to whether a

@SPLW⁠

client wishes to attend a session hosted by a provider. It could be the provider is the client’s local church and in this instance it may well be that the need for a kitemark is not considered necessary.

Currently

@SPLW⁠

tend to either check out the activity and provider themselves or many commented that they would accept a recommendation from a trusted colleague. Consideration has been given to some form of tooling to allow an SPLW to ‘look up’ a professional review of an activity or provider.

Recommendation: 

The technical review recommends attributes similar to Open Referral UK so that any kitemark can be added by the activity provider and verified by the

@SPLW⁠

It is recommended to carry out further understanding of Quest and MCRActive or other similar kitemark systems to understand whether they might be a solution to the assurance of an activity provider. 

It is also recommended to encourage innovators to provide a professional rating of OpenActive activities/providers i.e. a form of Trust Pilot for professionals in order to encourage

@SPLW⁠

to be more confident in making referrals.

Booking

The booking capability was not considered an issue by

@SPLW⁠

but it is likely that this is because of the trust issues with the data, the nervousness of ensuring suitability of activity that the SPLW had the time and energy to make the booking manually using a telephone or email or to follow a booking link. However as Social Prescribing matures, the focus on productivity continues and demand increases then we believe the need for activity data and the ability to make an immediate booking will become more critical.

In addition, a wider consideration is self-prescribing, self-care, self-finding and is more likely to have a higher expectation and requirement for live bookings.

Recommendation: 

It is recommended to discuss the booking capability with

@SPLW⁠

managers as they are becoming concerned about the productivity of

@SPLW⁠

. Having access to trusted data and an ability to make a booking is a productivity issue and an easy one for them to encourage without adversely impacting the

@SPLW⁠

clients.

It is also recommended that ODI works on a small pilot to consider the self-prescribing, find and book use case, targeting perhaps a high-profile activity such as gym access. There is a very large affinity gym-booking programme that may offer an interesting case study to establish demand.

Taxonomies across sectors

OpenActive is making use of three taxonomies. It has a robust activity-type taxonomy but this lacks broader terms to make it accessible to someone who doesn’t really know what activity they would like to participate in. This issue is addressed in the Culture & Process Issues in this document. There is a conditions taxonomy which is not fully established which attempts to link activity to certain conditions. Thirdly, there is an accessibility taxonomy which is currently being developed so that activities can indicate the assistance that they provide. 

These taxonomies have to be understood by activity providers and be a minimal burden on data entry. However, the bigger issue here is that

@SPLW⁠

have to use

@SNOMED⁠

codes which means there needs to be a mapping of

Snomed terms⁠

to the three taxonomies so that appropriate activities can be extracted from the data. This is also an issue for Open Referral. Our research has concluded that central cross-sector taxonomies are required to protect the activity (and service) providers from too many ‘mapping to category’ demands from the various sectors. See diagram and explanation in the

Technical Review⁠

Recommendation: 

It is recommended to carry out a short piece of co-production work with Pennine Lancashire to create some broader terms to group the activity-types.

It is recommended to consider the Cultural/Process recommendations on cross sector Governance to support the adoption of an accessibility taxonomy and a simple conditions taxonomy rather than continue to develop anything specific for OpenActive. The idea of a small set of central cross-sector taxonomies should also be explored with this governance.

The risk to the broadest Open data strategy, is that a lack of alignment across sectors will entirely undermine the core business case for open data through resulting in the following

Inaccurate data entry where providers have to understand different taxonomies across multiple sectors

Inaccurate searching by social prescribers where the mapping of one set of terms to another is inaccurate.

4.3 Further Considerations

Further considerations

This section provides some wider observations for consideration. These are not recommendations, but offer some wider comments on the Social Prescribing market which may help OpenActive in their thinking in doing further work with Social Prescribing.

Social Prescribing remains an immature, highly fragmented and potentially volatile programme

Decisions and processes tend to be defined at a PCN level. Influencing change across this breadth of 1,250 sets of different local leaders will be complex. In the last 12 months, there have been probably in excess of 1,000 new Social Prescribers recruited nationally, almost all on short-term contracts. It is a high profile programme right now; although PCNs can make decisions from April 2022 to move away from SPLW should they wish.

SP is only a small part of the referring community

As identified already in this report and the Numbers and Measures, there is an opportunity to consider the widest scale of community-based social prescribing that happens. The total scale will be ten-times the volume of pure GP-triggered social prescribing cases. People in the core SPLW pathway are probably fairly complex and others referral pathways may provide a richer vein for exercise referrals, including family and friend and self-care considerations.

Reality check

Very few people within the SPLW pathway will become sustainably physically active just because the SPLW is more aware of activities around them. However it will help and does remove a barrier; it may even simply be that it creates more belief for the patient in the entire SPLW process and hence valuable in its own right.

Measuring the outcome of a physical activity intervention in an unscientific way is without value

The research evidenced an industry of reporting that seems to be measuring things that cannot be proven by the data collected and the process used. Where there is a real interest to evidence the value of certain activities on improving outcomes, this should be subject to the right statistical and scientific control. The core function of SPLW is making links. Measurement of success should focus on that objective, rather than trying to evidence whether attending a class on ad hoc basis is beneficial.

Where an individual takes on personal ownership for monitoring their own sense of well-being; this can have significant impact for their well-being. This must be their data for their use and consideration; only they know what else is happening in their lives that may have greater impact on their health, and indeed whether they actually attended and committed to the activity they were referred to. Exercise on referral is recognised as a failed model for the stubbornly inactive; ownership of behaviour change is the key to sustainability, not externally-driven evaluation.

Adoption of OpenActive is not a complex technical consideration

The OpenActive standards and tooling is actually reasonably good. Yes there are improvements that can be made and indeed we have made some suggestions but generally OpenActive is fit for purpose. The reason why this has not taken off is because success depends on an ecosystem with supply and demand and that is difficult to establish across different sectors and stakeholders.

There isn’t a business case

There is no business case for adopting OpenActive solely to try to increase referrals into exercise through GP-referred Social Prescribing. There is a business case for place-based adoption of Open Data, but not for the individual components of the ecosystem. They are all interdependent and need a critical mass.

Application providers are commercial

The same message was given by all the application providers we spoke to. There must be a commercial reason for any investment. Generally they will provide whatever the paying customer would like. They were not put off by the complexities of implementing an open data standard and would implement whatever was either demanded through procurement or if they thought it would increase their chance of generating income.

Clear on opportunities for innovation

The ODI need to be clear what will be provided as the standard and what is left to innovation. Greater transparency of this will not only drive innovation, but will also lay bare where additional incentives are required.

Open Data as a national infrastructure

The UK nationally needs to take open data more seriously and treat it as a digital infrastructure. It is perhaps not as obviously required as broadband but if this does not become the fabric of our digital services then the UK is going to miss out on the benefits of joining up disparate information.

5.0 Research

The following outlines the steps taken in carrying out the user research and gives a summary where useful. The full research is recorded in Coda and is available

here⁠

. Coda provides the reader interaction with hoverable reference links and dynamic tables & graphs. To better experience the full interaction of this Coda doc the reader can request access. An explanation of Coda and the structure is given toward the end of this document.

5.1 Stakeholders

The end to end process of a social prescriber referring physical activity was mapped out and the stakeholders identified as below:

● Managers: responsible for strategic decisions on public health, budget setting & commissioning

● Frontline workers: responsible for ‘prescribing’ activities to meet personal need

● Activity providers: responsible for providing activities to meet personal needs

● Service users: responsible for deciding which activities to participate in

● Application providers: responsible for developing ‘tooling’ to support the above ecosystem roles

Working with Pennine Lancashire, we sought to interview at least 6 people within each stakeholder group as to what they believed were the barriers to social prescribing referring people into physical activity.

5.1.1 Stakeholders interview numbers

5.1.1 Stakeholders interview numbers

Activity providers

Frontline workers

Managers

OpenActive application providers

Other stakeholders

Social prescribing application provider

⁠

5.2 Personas

From the interviews we developed generic personas to help form an understanding of the stakeholder context and issues.

5.2.1 User Research in Numbers

User research participants interviewed

48

⁠

Activity providers interviewed 6

Application providers interviewed 11

Frontline workers interviewed 9

Managers interviewed 12

Service Users interviewed 5

Wider Research conducted 4

5.2.2 Personas summary

5.2.2 Personas summary

Persona

Avatar

Name

Role

Description

Page

Activity providers

Charlotte

Activity/Class facilitator

Charlotte is passionate about older people’s wellbeing and believes that appropriate exercise can have lots of benefits – falls prevention, mental health, loneliness prevention etc

⁠

Activity Provider - Persona

⁠

Frontline workers

Lorraine

Social Prescriber

Lorraine works for a GP practice and aims to prevent people needing to see a GP when they may benefit more from social/physical activities than medicine/drugs. Lorraine will work with frequent GP practice visitors, rehabilitations, GP referrals (medicine can’t help) or when a patient is happy to have a chat to see if they may benefit from local social/physical activities.

⁠

Frontline worker - Persona

⁠

Service users

David

Retired

David is a widower that is heading towards 80. He has always been fit but has not got out of the house as much lately and notices that things are a bit more of a struggle these days. His family are not too far away and do visit reasonably regularly but his friend died last year and he doesn’t have his dog anymore.

⁠

Service User - Older Person - Persona

⁠

Service users

Tony

Unemployed

Tony is unemployed and has not worked since a mental health issue approx. 20 years ago. He is separated from his wife with whom he has two kids each of whom have given him a grandchild. DWP have allowed him not to search for jobs but every two years they stress him out with another interview to see if he could work again. He lives on his own paying rent (benefits) in the house he’s been in for over 30 years. He grow Bonsai trees.

⁠

Service User - Mental Health - Persona

⁠

Managers

Gareth

‎Health Partnerships & Commercial Manager

Gareth has a strategic responsibility for public health. He is keen to prevent people needing GP or A&E. He’d like to engage a significant number of people to become more active. He thinks social prescribing could help with this but recognises that there are a number of barriers to overcome.

⁠

Managers - Persona

⁠

Application Provider

Greg

Product Manager

Greg is a product manager for SPActive He is trying to provide a product to help frontline workers identify the best local services/activities that will help support their client list. They want to focus on their UI and features to enable the frontline workers to be more productive and help more people. They want a feed/API to access local service information that is rich and correct.

⁠

Application Provider - Persona

⁠

There are no rows in this table

⁠

5.2.3 User Journeys

The personas and interviews were developed into user journeys to highlight the barriers. Below is just the frontline worker table and the full table is available under User Research.

MARCUS TO ADD USER JOURNEY SAMPLE TABLE - FRONTLINE WORKERS

Frontline worker user journey - barriers

Frontline worker user journey - barriers

1.Referral

2.Consider options

3.Shared decision

4.Attend

5.Participate

Frontline workers

Don’t see referral into exercise as their role

Difficult to know what activity sessions are availableNeed to understand exactly what the activity/service is offering to ensure client can copeClient isn’t interested in doing physical activity, too many other problems

Not enough local activity sessions can be found

Don’t trust information that exists Client lacks confidence, has anxiety, embarrassed so needs hand-holding to attend

Need to know if the providers can cope with needs of the cohortsNeed to know if activity appropriate to client needs Frontline workers lack confidence and skill set to motivate client to be active

⁠

5.3 Brainstorming & gap analysis

The barriers identified from the personas were consolidated into 14 key barriers. These were considered by the project team to brainstorm what might be possible to mitigate these. The team also carried out an initial tooling and standards review bearing the barriers in mind.

⁠

Research workshop - Barrier to strategies (1).jpg

⁠

5.4 Workshops

Three online workshops used Miro in different ways to identify and prioritise ideas as to how to mitigate the barriers identified in the interviews.

⁠

Barriers to social prescribing - FW - Picture - exec summary (1).jpg

⁠

5.4.3 Questionnaire

All the ideas/approaches were added into Coda tables and linked together to barriers through stakeholder groups and benefits.

Theses were then presented to anyone interesting in feedback with an opportunity to rate and comment on a barrier or idea/strategy/approach.

The questionnaire is available here

interactive questionnaire⁠

⁠

Example barriers feedback table.

⁠

⁠

5.6 Technical review of tooling and standards

The technical review progressed weekly bearing in mind the user research. A workshop in the last week of the project gave the application providers a chance to feedback before the report was delivered to ODI. Unfortunately key players couldn’t make it and so individual feedback has been sought through the questionnaire and a couple of meetings.

⁠

5.7 Data Requirements & taxonomies

The stakeholder groups had little knowledge of taxonomies so the pilot team considered the potential barriers this might present and agree that there was a risk of many sectors demanding the use of different taxonomies on the activity providers. The conclusion was to identify central taxonomies that made sends to the activity providers that would then allow a sector taxonomy to map to it or allow direct use of a central taxonomy to retrieve an activity.

⁠

⁠

5.7.1 Potential Cross-sector Taxonomy Solution

5.7.1 Potential Cross-sector Taxonomy Solution

Cross-sector Taxonomy Method

4 central taxonomies, agreed by all sectors, that make sense to activity (and service) providers.

Activity (and services) should be mapped to the central taxonomies by activity (and service) providers

Sector led taxonomies e.g. Snomed can map their terms to the central taxonomies

This allows a user to use a sector taxonomy to identify appropriate activity (or services)

Alternatively a user can filter on the central taxonomies to identify appropriate activity (or services)

Any change, therefore, to a sector taxonomy doesn’t impact on any other sector

There are no rows in this table

⁠

5.7.2 Possible linked Taxonomies

5.7.2 Possible linked Taxonomies

Name

Sector specific

URL

SNOMED

NHS

⁠

https://termbrowser.nhs.uk/?

⁠

Needs/Circumstances

Social Care

https://standards.esd.org.uk/?

Fire Service

Unkown

Education

Unkown

Leisure

Unkown

Police

Unkown

Housing

Unkown

Poverty

Unkown

There are no rows in this table

⁠

5.8 Ecosystem

The following table describes the stakeholders that were interviewed and made up the ecosystem. The user journey shows the barriers that were identified for each of the stakeholders.

ADD ECOSYSTEM TABLE

ADD USER JOURNEY TABLE

5.8.1 Data-flow

The data-flow was drafted based on the end to end process and the stakeholders. This was reviewed for potential improvements. There was discussion around working with RDPE and APIs. General conclusion was that there wasn’t a lot of barriers due to the data-flow but take up of the standard by Consumer Application Providers was critical to the success of OpenActive.

⁠