Innovation – D1

Sammir Radha

Alder Hey Children’s NHS Foundation Trust

North West, Integrated

 

 

“I’d like a “how to” guide for this innovation as i think it would be a fantastic way forward for finance teams to utilise.”

Peer Reviewer

“The advantages of this is that it can be incorporated into a datatable based working paper without having a separate form to act as an interface for the changes to the data warehouse.”

Peer Reviewer

“I’m really excited about this innovation as being able to “write back” to a data warehouse is something that could be applied to numerous finance tasks.”

Peer Reviewer

The Problem

Any team that consumes, analyses or comments on data, whether that be financial, clinical or other non-clinical data, needs to be able to use the systems they use to write back information that would enhance the quality and usefulness of the data so that it can be drawn upon for future use by the individuals enhancing the data, their colleagues and managers and for the systems themselves to start working out how to enhance the data on their own in future because otherwise the work to enhance this data is still done but in a way that means that the job often needs to be repeated or is not visible to the wider organisation.

The Challenge

How might we carry out our data enhancing tasks such as commenting on variances, grouping staff into teams or correcting errors in our theatre records so that the work is persisted into the fabric of our data systems so that work can be shared so it doesn’t need repeating and that we can leverage machine learning techniques to use the data to figure out how to enhance the data on its own in the future?

The Outputs

  • Audit history of changes to COVID-19 Reporting categories allows for identifying what was reported at a certain point in time
  • A transactional tagging system capable of enhancing data quality and using that knowledge to learn how to do the enhancements itself
  • Ability to perfectly match any NHSI return or costing return in time with source, detailed COVID-19 data
  • Single source of truth for COVID-19 reporting and analysis in order to identify what should/shouldn’t be there
  • Once the transactions are tagged, they are saved to the database which other reports use so people can see the mapped COVID-19 schemes in any reports that have the COVID-19 cost centres on them

The Outcomes

  • COVID-19 Reporting time required down 15 hours during month-end to 1 hour as a result of:
    • Suggestions being made that are 99% accurate for 80-85% of COVID-19 transactions (so effectively 80% of the job being done automatically)
    • No time wasted in identifying what transactions are new and adding to existing mappings to COVID-19 categories as all schemes are persisted to the database where the transaction report is being drawn down from, one click of a refresh button and you get the latest transactions with all your previous work not lost
    • Bulk Insert process uploads data extremely quickly
  • Senior leaders knowing exactly what makes up the COVID-19 expenditure in order to make decisions on how to reduce/fund it
  • Simple to understand procedures for COVID-19 reporting, albeit very complex workings in the background, so it can be passed on for others to pick up
  • Caveat to that is that any SQL required to interrogate the audit history and use elements beyond the transactional tagging spreadsheet will continue to require the specialist skills to do so