Five Safes: designing data access for research
Tanvi Desai, Felix Ritchie, Richard Welpton · RePEc: Research Papers in Economics · 2016
What is the best way of managing access to sensitive data? This is not a straightforward question, as it involves the interaction of legal, technical, statistical and, above all, human components to produce a solution. This paper introduces a modelling tool designed to simplify and structure such decision-making.The Five Safes model is a popular framework for designing, describing and evaluating access systems for data, used by data providers, data users, and regulators. The model integrates analysis of opportunities, constraints, costs and benefits of different approaches, taking account of the level of data anonymisation, the likely users, the scope for training, the environment through which data are accessed, and the statistical outputs derived from data use.Up to now this model has largely been described indirectly in other papers which have used it as a framing device. This paper focuses specifically on the framework, discusses usage, and demonstrates where it sits with other data and risk management tools. The aim is to provide a practical guide to the effective planning and management of access to research data.