Deriving household composition using population-scale electronic health record data—A reproducible methodology
Open Access
- 29 March 2021
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (3) , e0248195
Abstract
Background: Physical housing and household composition have an important role in the lives of individuals and drive health and social outcomes, and inequalities. Most methods to understand housing composition are based on survey or census data, and there is currently no reproducible methodology for creating population-level household composition measures using linked administrative data. Methods: Using existing, and more recent enhancements to the address-data linkage methods in the SAIL Databank using Residential Anonymised Linking Fields we linked individuals to properties using the anonymised Welsh Demographic Service data in the SAIL Databank. We defined households, household size, and household composition measures based on adult to child relationships, and age differences between residents to create relative age measures. Results: Two relative age-based algorithms were developed and returned similar results when applied to population and household-level data, describing household composition for 3.1 million individuals within 1.2 million households in Wales. Developed methods describe binary, and count level generational household composition measures. Conclusions: Improved residential anonymised linkage field methods in SAIL have led to improved property-level data linkage, allowing the design and application of household composition measures that assign individuals to shared residences and allow the description of household composition across Wales. The reproducible methods create longitudinal, household-level composition measures at a population-level using linked administrative data. Such measures are important to help understand more detail about an individual’s home and area environment and how that may affect the health and wellbeing of the individual, other residents, and potentially into the wider community.Keywords
Funding Information
- Medical Research Council (MR/V028367/1)
- Health Data Research UK (HDR-9006)
- Economic and Social Research Council (ES/S007393/1)
- Administrative Data Research UK
- Nuffield FJO (FJO/43766)
This publication has 32 references indexed in Scilit:
- A typology of kinship foster families: Latent class and exploratory analyses of kinship family structure and household compositionChildren and Youth Services Review, 2010
- The SAIL Databank: building a national architecture for e-health research and evaluationBMC Health Services Research, 2009
- Residential Anonymous Linking Fields (RALFs): a novel information infrastructure to study the interaction between the environment and individuals' healthJournal of Public Health, 2009
- The SAIL databank: linking multiple health and social care datasetsBMC Medical Informatics and Decision Making, 2009
- Who cares and how much: exploring the determinants of co-residential informal careReview of Economics of the Household, 2008
- Household structure and infectious disease transmissionEpidemiology and Infection, 2008
- Household Composition and Fatal Unintentional Injuries Related to Child MaltreatmentJournal of Nursing Scholarship, 2008
- Approximate disease dynamics in household-structured populationsJournal of The Royal Society Interface, 2007
- A Bayesian MCMC approach to study transmission of influenza: application to household longitudinal dataStatistics in Medicine, 2004
- New Estimates of Intergenerational Time Intervals for the Calculation of Age and Origins of MutationsThe American Journal of Human Genetics, 2000