期刊名称：International Journal of Population Data Science
电子版ISSN：2399-4908
出版年度：2018
卷号：3
期号：2
页码：1-1
DOI：10.23889/ijpds.v3i2.572
出版社：Swansea University
摘要：BackgroundWelsh Government (WG) invests over £120m annually in housing related support to help prevent and tackle homelessness under the Supporting People (SP) Programme. A 2016 data-linkage Feasibility Study for two Local Authorities (LAs) indicated health-service utilisation reductions post-intervention, and led to a four-year ADRC-Wales project to create a national, all-Wales dataset to provide robust statistical results. ObjectivesEstablish data sharing agreements and acquire anonymised individual level data into the SAIL (Secure Anonymised Information Linkage) databank to create a national, all-Wales, longitudinal SP dataset. Create control groups and carry out statistical analysis to investigate how the Supporting People Programme may help service users and the effects on other public services. Methods (including data)Intervention data for Supporting People service users was collected from individual LAs and imported into the SAIL databank using a split file process to maintain anonymity. Anonymised service users data were linked at the individual level to demographic and health data records for in excess of 40,000 individuals, with outcome measures created based on emergency hospital admissions, accident and emergency visits and General Practice events. Several control group creation methods were examined: 1) Internal Intervention Programme Data; 2) Matched controls using the SAIL databank; 3) Healthcare-Utilisation Patterns; 4) External Data Sources (Housing Options and Substance Misuse datasets). FindingsData sharing agreements and data acquisition complete for nineteen of twenty-two LAs, discussions ongoing for remaining LAs. Health data-linkage complete, other public service data acquisition ongoing. Annual progress reports published, main statistical analysis underway and due for 2018 publication. ConclusionsWe discuss challenges and opportunities around acquiring multi-sourced administrative data, methods of creating meaningful control groups and planned statistical analysis.
