To weight or not to weight? The effect of selection bias in 3 large electronic health record-linked biobanks and recommendations for practice

Maxwell Salvatore; Ritoban Kundu; Xu Shi; Christopher R Friese; Seunggeun Lee; Lars G Fritsche; Alison M Mondul; David Hanauer; Celeste Leigh Pearce; Bhramar Mukherjee

doi:10.1093/jamia/ocae098

To weight or not to weight? The effect of selection bias in 3 large electronic health record-linked biobanks and recommendations for practice

J Am Med Inform Assoc. 2024 May 14:ocae098. doi: 10.1093/jamia/ocae098. Online ahead of print.

Authors

Maxwell Salvatore^{1

2}, Ritoban Kundu^{2

3}, Xu Shi³, Christopher R Friese^{4

5

6}, Seunggeun Lee^{3

7}, Lars G Fritsche^{2

3

4}, Alison M Mondul^{1

4}, David Hanauer⁸, Celeste Leigh Pearce^{1

4}, Bhramar Mukherjee^{1

2

3}

Affiliations

¹ Department of Epidemiology, University of Michigan, Ann Arbor, MI 48109-2029, United States.
² Center for Precision Health Data Science, Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109-2029, United States.
³ Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109-2029, United States.
⁴ Rogel Cancer Center, Michigan Medicine, University of Michigan, Ann Arbor, MI 48109-2029, United States.
⁵ Center for Improving Patient and Population Health, School of Nursing, University of Michigan, Ann Arbor, MI 48109-2029, United States.
⁶ Department of Health Management and Policy, University of Michigan, Ann Arbor, MI 48109-2029, United States.
⁷ Graduate School of Data Science, Seoul National University, Gwanak-gu, Seoul, Republic of Korea.
⁸ Department of Learning Health Sciences, University of Michigan Medical School, Ann Arbor, MI 48109-2054, United States.

PMID: 38742457
DOI: 10.1093/jamia/ocae098

Abstract

Objectives: To develop recommendations regarding the use of weights to reduce selection bias for commonly performed analyses using electronic health record (EHR)-linked biobank data.

Materials and methods: We mapped diagnosis (ICD code) data to standardized phecodes from 3 EHR-linked biobanks with varying recruitment strategies: All of Us (AOU; n = 244 071), Michigan Genomics Initiative (MGI; n = 81 243), and UK Biobank (UKB; n = 401 167). Using 2019 National Health Interview Survey data, we constructed selection weights for AOU and MGI to represent the US adult population more. We used weights previously developed for UKB to represent the UKB-eligible population. We conducted 4 common analyses comparing unweighted and weighted results.

Results: For AOU and MGI, estimated phecode prevalences decreased after weighting (weighted-unweighted median phecode prevalence ratio [MPR]: 0.82 and 0.61), while UKB estimates increased (MPR: 1.06). Weighting minimally impacted latent phenome dimensionality estimation. Comparing weighted versus unweighted phenome-wide association study for colorectal cancer, the strongest associations remained unaltered, with considerable overlap in significant hits. Weighting affected the estimated log-odds ratio for sex and colorectal cancer to align more closely with national registry-based estimates.

Discussion: Weighting had a limited impact on dimensionality estimation and large-scale hypothesis testing but impacted prevalence and association estimation. When interested in estimating effect size, specific signals from untargeted association analyses should be followed up by weighted analysis.

Conclusion: EHR-linked biobanks should report recruitment and selection mechanisms and provide selection weights with defined target populations. Researchers should consider their intended estimands, specify source and target populations, and weight EHR-linked biobank analyses accordingly.

Keywords: ICD codes; biobank; electronic health records; phenome; selection bias.

Grants and funding

P30CA046592/CA/NCI NIH HHS/United States