Limitations and conclusions

While this analysis provides a rich overview of the Impacts realised from DARS data, there are several limitations.

Benefit statements are not currently structured or coded and and they represent the views of the applicants themselves. The quality of statements can be variable and the benefits listed generally did not follow a strict benefits methodology, so they have been described as Impacts, to avoid confusion. This approach could have the advantage of picking up Impacts that, although important, are not easily quantified and therefore generally under-reported. It does however make it impossible to assess their scale, without further work.

This was a snapshot. Impacts may vary over time. The categories were identified through a grounded approach, that is, they emerged from the data, rather than being based on an existing framework. This meant that the terminology used did not always align with the description of benefits in the literature or strategic policy documents.

DARS data does not include the statistical publications and other sources of open data that NHS Digital publish, so this analysis does not provide a complete picture of the impacts of NHS Digital data.

This report identifies what we believe to be an almost complete set of Impacts generated by our data. It does not quantify the scale of these Impacts, but it does suggest the distribution of Impacts across different types of organisations that make use of our data. It also provides a breakdown of the category of Impact by organisation type.

This information provides increased transparency around our work. The impacts identified span research, policy-making, driving efficiency and ultimately, improving health and care for patients and citizens.

This work can help to standardise how we record Impacts in future. It would be possible to ask applicants to self-code their benefits and yielded benefits statements as part of the DARS process, while still allowing free text input. This would enable an automated statement of Impacts to be produced at regular intervals. It would improve transparency around how we use patient data, the greater number of coded statements would allow us to drill down and analyse more granular Impacts by type of applicant and even by data products involved. Changes in Impacts over time could be monitored and used as part of a data utility improvement process, within NHS Digital or to evaluate the success of changes in our systems. 

This information could be used to show individual patients how their data has contributed to Impacts and to show applicants how the Impacts of their work compare to other projects. This could help to spread innovative methodologies.

Finally, gaps in the Impacts produced by our data, compared to other sources of data in the UK and around the world, could point to improvements that we could make, to enable the realisation of new types of Impact.

A broad multidisciplinary team provided input to this project at various stages, including an impact grouping workshop. The team included:

• Mohammed Basser
• Phil Cooke
• Karen Hillier
• James Hussey
• Joan Pons Laplana 
• Dominic Povey
• Brian Roberts
• Tahmina Rokib
• Bharat Sharma
• Kimberley Watson
• NHS Digital design team
• NHS Digital web team