The challenge of opt-outs from NHS data: a small-area perspective.

Abstract

One of the founding principles of the NHS is that it offers comprehensive, universal and free public health services at the point of delivery. As a result, NHS data provide a huge and invaluable resource of routinely collected primary (e.g. visits to GP practices) and secondary (e.g. hospital admissions, outpatient appointments, A&E attendances) healthcare data covering near-100% of the population of England. NHS Digital has the responsibility for collecting and publishing data and information from across the health and social care system in England and controls the dissemination of these data. Detailed analysis of NHS data by public health and research institutions has the potential to considerably improve health and social care in England. Data privacy and confidentiality of identifiable health data are a legal responsibility of the NHS, and a source of concern for patients. As illustrated by recent controversies and cyber-attacks,1,2 ensuring data security has become a growing technological challenge for the NHS, for example, with the move towards electronic health records. As per the NHS Constitution,3 patients have the right to request that their confidential information is not used beyond their own care and treatment. This right was formalized in January 2014 through an opt-out system, following a recommendation made by the National Data Guardian (NDG) for Health and Care, Dame Fiona Caldicott, in her 2013 report.4 A National Data Opt-out Programme is scheduled to replace the current two types of opt-outs from March 2018, following new recommendations made by the NDG in her Review of data security, consent and opt-outs published in 2016.5 Public health research, including disease surveillance, can contribute vital information for the direct care and treatment of patients,6,7 as well as broader issues related to population health,8 but many studies are no longer granted access to the full national health datasets held by NHS Digital. Our aim here is to comment, ahead of the launch of the new National Data Opt-out Programme, on the potential consequences of these non-random gaps in national health databases. This is particularly a concern for research that is currently unable to access complete health data, i.e. where opt-outs are upheld, such as, for example, small-area health studies of non-communicable diseases.9