What are these game-changing tools? 

GitHub: the hub of collaboration 

NHS Midlands and Lancashire’s (NHS ML) Business Intelligence team is no stranger to innovation. Among our various initiatives, we run Knowledge Sharing Networks (KSNs) aimed at fostering learning and best practice within the organisation.

One such KSN, focused on data science, has spearheaded the development of two technical tools that are essential to increase productivity: an organisational GitHub account and an R package specifically tailored for NHS ML. This blog post aims to introduce these collaborative tools, delve into their purpose, and discuss the strides we’ve made in their development. 

The first tool is the NHS ML GitHub account. GitHub is a free platform that serves as a repository for code, facilitating version control and collaboration. As Henry Ford aptly put it, “Coming together is a beginning, staying together is progress, and working together is success.” 

Given the vast geographical expanse that NHS ML covers and the diverse projects our Business Intelligence teams are engaged in, it’s crucial to have a centralised platform. GitHub allows us to share automated and reproducible procedures, aligning with NHS Digital’s guidelines on Reproducible Analytical Pipelines (RAP). 

We are not alone in this endeavour. In setting up an organisational GitHub account we are following in the footsteps of leading NHS data teams such as:

• NHS-R Community
• The Strategy Unit
• NHS Digital

We’ve already gained valuable insights from our colleagues at the Strategy Unit, which is part of NHS ML, further emphasising the power of collaboration. 

NHS ML R package: the efficiency multiplier 

The second tool in the pipeline is the NHS ML R package. If GitHub is the vehicle for collaboration, consider the R package as the fuel that powers it. 

Designed to streamline the workflow for R users within NHS ML, this package embodies the principle that “You should consider writing a function whenever you’ve copied and pasted a block of code more than twice,” as stated by Hadley Wickham & Garrett Grolemund in their book, “R for Data Science.” 

Currently, the NHS ML package includes a function that allows users to directly read data from the NHS ML Business Intelligence SQL server. While this might save only 30-60 seconds per use, the cumulative time-saving potential across multiple users is significant. 

Looking ahead: the journey continues 

We’re at the nascent stages of our journey into the world of RAP. These tools, although simple in their current form, represent stepping stones towards a more streamlined and collaborative future. The learning curve is steep, but as the saying goes, “small progress is better than none.” 

Get involved 

If you’re intrigued by our work and would like to contribute to the growing suite of collaborative tools developed by the NHS ML Business Intelligence team, please don’t hesitate to contact Scott Johnson at scott.johnson2@nhs.net. 

By Scott Johnson, Business Intelligence Lead at NHS ML.