Modernising PHE's Population Health Intelligence System

Tender description

Summary of the work To perform Discovery and potentially other phases in order to explore the needs of users to upgrade and/or replace population health analytics and surveillance systems for a single Population Health Intelligence System including non-communicable disease (NCD) risk factors and wider determinants of health. Expected Contract Length 24 months (Total duration of all phases of project) Latest start date Monday 29 October 2018 Budget Range £60,000 including VAT for Discovery. Additional phases are TBC dependent on the outcomes of Discovery and subsequent budgetary approvals. Why the Work is Being Done PHE has a complex and fragmented data landscape,and knows that there are efficiencies to be gained in how it gathers, manages and processes data. PHE calculates over 1000 public health indicators for profiles like the Public Health Outcomes Framework; and 13 of the 20 priorities listed in the 2018-19 remit letter (https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/693855/PHE_Remit_Letter_-_March_2018.pdf) are monitored through the Fingertips platform (https://fingertips.phe.org.uk/profile/public-health-outcomes-framework) using data from over 200 different datasets. However, PHE’s stakeholders continue to require and demand more granular and timely data than it currently provides. Problem to Be Solved • Develop a clear understanding of the main internal users’ needs, the context of their work, what they aim to achieve and how in the future. • Document an understanding of PHEs current data and processes. • Identify any gaps/future workforce needs: PHE may require additional data skills in order to deliver a new approach to population health intelligence. • Market research: To identify if ‘off the shelf solutions’ exist in the market which meet the needs identified or whether something more bespoke is required. • Identify any potential requirements for hosting the new Population Health Intelligence System (PHIS). Who Are the Users PHE data analysts: Develop public health indicators and reporting mechanisms by accessing and manipulating data sets / surveys. PHE data scientists and researchers who primarily work in Health Improvement, National Infection Services (NIS) and Field Epidemiology Service (FES) on both primary and secondary data sources. It will also be necessary to engage with academic colleagues, students and interns who support the work of PHE and staff within the PHE ICT Department. Early Market Engagement None Work Already Done PHE have the 'data lake' and an indicator database PHOLIO to manage some of the data, and have developed ways to visualise and present Data. The ‘data lake’ is a combination of different data sources but the assets are not consistently catalogued or labelled making combining data from multiple sources to answer questions difficult. PHE have been promoting an automation programme with the aim of improving the efficiency of data processing and freeing up analytical time Existing Team TBC Current Phase Discovery Skills & Experience • E1) Experience of delivering projects that model a proposed high-level system architecture • E2) Recent and demonstrable experience of designing systems to manage personally identifying information under the General Data Protection Regulation (GDPR). • E3) Recent and demonstrable experience in conducting the discovery phase relating to the use of digital solutions and technology in data analysis / MIS projects. • E4) Recent and demonstrable experience of applying the principles of user centred design to the development of digital solutions. • E5) Recent and demonstrable experience of digital systems that take input and subject this to both automated and manual validation. Nice to Haves • N1) Recent and demonstrable experience of working with healthcare organisations • N2) Experience of working with a range of users and stakeholders including scientific experts, technical and digital development teams • N3) Recent and demonstrable experience of working on Data Science / “big data” / machine learning projects. • N4) Recent and demonstrable experience of working with interdisciplinary and cross disciplinarily project teams including clinical and research science project members Work Location The work can be performed from any location but is likely to be co-ordinated by staff in PHE’s Offices Working Arrangments Work may be carried out from multiple locations including the Supplier’s offices. Attendance at PHE Offices in London and in other locations around England will be required for meetings. Supplier staff are expected to be on-site at PHE locations sufficient to manage relationships with senior stakeholders; the exemplar project team and users; and, stakeholders and users in other organisations so that the outcomes of the project may be met Security Clearance Not required Additional T&Cs None No. of Suppliers to Evaluate 5 Proposal Criteria • P1) Proposed approach to user research to assess the needs of users including those who come from academic/scientific backgrounds. • P2) Proposed approach to Discovery to determine and demonstrate what digital tools and technology can be used to meet the public health outcomes of this project. • P3) Proposed approach to balancing user needs with off the shelf solutions particularly in the area of data analytics. • P4) Supplier's team structure / experience Cultural Fit Criteria • The Presentation (to be held in October 2018, date TBC) should summarise the Proposal and also demonstrate the approach that will be used to: • C1) Work with the project team’s stakeholders who have varying experience of agile methods. • C2) Flexible working arrangements and knowledge sharing; • C3) Work within the constraints of existing services, budgets and technologies already used within the programme Payment Approach Capped time and materials Evaluation Weighting Technical competence 70% Cultural fit 10% Price 20% Questions from Suppliers 1. Would PHE expect the supplier's BA's and Consultants to have clinical experience and expertise of leveraging analytics within the specific context of the PHE's requirements (Specific industry vertical knowledge). Or are PHE looking for technology specialists in analytics with broad horizontal expertise that can be applied to the specific use case described and in which case, be supported by the PHE's own subject matter experts? PHE are looking for technology specialists in analytics with broad horizontal expertise that can be applied to the specific use case described and in which case will be supported by the PHE's own subject matter experts. 2. Can the authority provide details of the technology stack behind the "data lake", the PHOLIO database and the automation programme? SQL Cluster, R and Python -> potentially moving to OpenStack 3. What is the expected duration of the Discovery phase? 6 - 8 Weeks TBC 4. What is the expected ROI for this project and which factors were taken into account? Until Discovery has been completed and the outcomes assessed this will not be known. 5. What is(are) the uncertain event(s)/value(s)/thing(s) that you want/hope to predict? This will be determined through Discovery by the successful supplier 6. What decisions and actions will be taken e.g about how to treat people, based on the predictions made by the model(s)? This is about population health not individual care so more about influencing spend, policy and practice. 7. How confident are we that the data is representative of the target population? The project team have high confidence with regards to this. 8. What is the forecast horizon? Are the data labelled? E.g Do we know whether someone with specific biomarkers went on to develop a disease.symptoms after say 5 years? There is a lot of data, some is labelled, some is not, some could be, much is aggregate. We are more interested in population health markers than individual biomarkers. 9. The problem to be solved section describes tasks to be completed during the Discovery phase, however, the skills and experience are more architectural and solution based criteria. Is it expected that the bidders will include a solution design in the Discovery phase? It is unlikely that any solution design will be undertaken in Discovery. However, this may be dependent on what is established during Discovery by the successful supplier. 10. Is the buyer intending to use a NHS section 251 as a legal basis for aggregating of data with the implication that the data is being used for a purpose different to that that it was original captured for (DPIA)? We have a range of legal bases under s251 and GDPR 11. Is there an expectation on the bidder during the Discovery phase to identify all potential GDPR and DPA 2018 issues? Yes - to a degree. This will be confirmed with the successful supplier.