Criteria for detection of married or cohabiting couples in the UK Biobank
Kristen M Kelly1, Tanya B Horwitz2, Matthew C Keller3
1Institute for Behavioral Genetics, University of Colorado Boulder, 1480 30th St, Boulder, CO, United States of America, 80303.
Introduction:
The UK Biobank (UKB) dataset contains a large number of spouse/partner pairs (hereafter, "couples"), making it a valuable resource for research on human mating. UKB data does not report relationships between participants, and UKB's preliminary participant colocation data (based on participants sharing the same street address) is not available for new researchers to request. This has led to different criteria for identification of spouse/partner pairs and highly discrepant sample sizes across papers, potentially contributing to heterogeneity in results. To address this, we developed and validated a standardized method for identifying couples that maximizes sample size while minimizing selection bias.
Methods:
We evaluated combinations of geographically-derived variables for identifying colocated UKB participants and selected six variables that performed well when compared to with the original UKB colocation data. These variables, such as "distance to coast," do not reveal participants' precise locations but are unlikely to match for non-colocated individuals. We then established additional criteria to confirm that colocated participants shared a household and were likely to be couples. These criteria were designed to be compatible with either geographic colocation or the original UKB colocation data.
Results:
The geographically-derived variables identified 92,510 putative couples, compared to 89,278 detected using the UKB colocation data. Further analyses suggested that the additional pairs identified in our geographic sample were likely valid pairs not captured by the UKB colocation data. We also assessed certain criteria used in previous studies-such as age, income, and duration of residence-and demonstrated how they could result in biased or non-representative samples.
Discussion:
Our approach produced a large, robust sample of couples while minimizing false positives. The criteria are flexible and can be applied using either geographic or UKB colocation data. To facilitate further research, we have made the R code for implementing these criteria publicly available at https://github.com/kkellysci/UKBCouples/.
More Related Videos
11:09Multiplex Detection of Bacteria in Complex Clinical and Environmental Samples using Oligonucleotide-coupled Fluorescent Microspheres
Published on: October 23, 2011
05:49Use of Magnetic Resonance Imaging and Biopsy Data to Guide Sampling Procedures for Prostate Cancer Biobanking
Published on: October 10, 2019
Related Concept Videos
Equivalent Couples
Two couples are considered to be equivalent if they produce the same rotational effect on a rigid body. In other words, the two couples have the same magnitude and act in the same direction, causing the same angular displacement or acceleration in the body.
For instance, consider two couples lying in the plane of the page, with one having a pair of equal...
Sign Test for Matched Pairs
To conduct the sign test, we first calculate the differences in...
Cross-Sectional Research
Data Collection by Observations
An astronomer viewing the motion and brightness of stars in the sky and recording the data is an example of observational data collection. A botanist recording...
Bias in Epidemiological Studies
Pedigree Analysis
