Exploiting the potential of routine data to better understand the disease burden posed by allergic disorders


Aziz Sheikh, Division of Community Health Sciences: GP Section, University of Edinburgh, 20 West Richmond St, Edinburgh EH8 9DX, UK. E-mail: aziz.sheikh@ed.ac.uk


The Department of Health and Scottish Executive are currently undertaking independent reviews of allergy services in England (and Wales) and Scotland. Each review will assess the disease burden posed by allergic problems, involving secondary analyses of routine National Health Service (NHS) datasets. Major suggestions for re-structuring and/or re-focusing the NHS efforts to better deal with allergic disease are anticipated. The UK has some of the best datasets of routine health data in the world, but despite their strengths, they have important limitations. These include gaps in data collection, particularly in relation to monitoring of Accident & Emergency and out-patient consultations, and in-patient prescribing, thereby resulting in considerable under-estimates of hospital workload. The current gaps in service monitoring are likely to under-estimate the burden and workload associated with allergic problems, particularly in secondary care. One major limitation of existing data sources is the general inability to link individual patient level data between different datasets. By unlocking this potential there are very considerable potential gains to be made. Data linkage techniques currently being developed in the UK offer exciting new possibilities of looking across the primary-, secondary- and tertiary-care interfaces and also assessing short-and long-term social and educational outcomes in relation to allergic disorders. The current reviews of allergy services being undertaken need to be cognisant of these inherent limitations of existing data sources and would do well to recommend strategic initiatives that could enhance the availability, accessibility and quality of these datasets. Ideally, this should include investment in central data repositories staffed by teams with the necessary technical and statistical expertise, which would also take responsibility for progressing data linkage capabilities.