Premium
O3‐02‐06: CREATING A DEMENTIA RESEARCH COHORT USING ROUTINELY COLLECTED HEALTHCARE DATA
Author(s) -
Wilkinson Tim,
Schnier Christian,
Gallacher John,
Lyons Ronan A.,
Sudlow Cathie
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.06.2781
Subject(s) - dementia , cohort , medicine , cohort study , population , diagnosis code , health care , family medicine , gerontology , pediatrics , disease , environmental health , economics , economic growth
Figure 2. Frequency of dementia subtypes within cohort Subtypes non–mutually exclusive. All-cause dementia includes unspecified dementia codes as well as subtypes codes. Mixed dementia includes participants who received both an Alzheimer’s disease and vascular dementia code at any point during follow-up. Background:The Secure Anonymised Information Linkage (SAIL) databank contains a wide range of regularly updated, anonymised, linked, routinely-collected healthcare data for the entire population of Wales, UK (3 million people). It is a valuable resource for dementia research owing to its size, long follow-up time and prospective collection of data during routine clinical care. We aimed to transform these large, coded, healthcare datasets into a flexible, ‘virtual cohort’ for dementia research. Methods:SAIL contains individual-level, linked primary care, hospital admissions, mortality and demographic data. We selected an initial observation period of 1 January 1990 to 31 March 2016, following participants from an inception date of 1 April 1997 (or first registration with a General Practitioner [GP] if that came later). We created >20 International Classification of Diseases version 10 and Read version 2 code lists for: dementia and its subtypes; comorbidities (e.g. stroke, rheuma-