Friday, 9 May 2025 — medConfidential
There isn’t a Nobel prize at the end of the researcher rainbow
The “five safes”* Trusted Research Environments model has been tested over decades, meaning “safe people” doing “safe projects” on “safe data” in “safe settings” to produce “safe outputs”, and while the precise meaning of the five safes evolves with the context and the datasets, half baked additions of “new safes” weaken the whole model for short term gain.
The Department of Health in England prefers the acronym “SDE” because Sometimes Data Escapes.
No model can be infallible; cheats and crooks will always try to game the system, and an organisation saying they follow the “five safes” can still catastrophically screw up, but understanding the model is necessary to start with and there’s one way that is often used to undermine it for gain.
The ‘Five Safes’ (Wikipedia)
* The Five Safes is a framework for helping make decisions about making effective use of data which is confidential or sensitive. It is mainly used to describe or design research access to statistical data held by government and health agencies, and by data archives such as the UK Data Service.[1]
Two of the Five Safes refer to statistical disclosure control, and so the Five Safes is usually used to contrast statistical and non-statistical controls when comparing data management options.
Concept
The Five Safes proposes that data management decisions be considered as solving problems in five ‘dimensions’: projects, people, settings, data and outputs. The combination of the controls leads to ‘safe use’. These are most commonly expressed as questions, for example:[2][3]
| Safe projects | Is this use of the data appropriate? |
| Safe people | Can the users be trusted to use it in an appropriate manner? |
| Safe settings | Does the access facility limit unauthorised use? |
| Safe data | Is there a disclosure risk in the data itself? |
| Safe outputs | Are the statistical results non-disclosive? |
The New Dark Age 
Leave a comment