The Big Data Paradox in Clinical Practice

Msaouel, Pavlos

doi:10.1080/07357907.2022.2084621

Cited by 25 publications

(24 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If biobanks are not representative of the target population, they are vulnerable to selection bias, 3,[19][20][21][22] a naïve analysis is not expected to align with the population truth. [23][24][25] Handling selection bias is particularly challenging because the magnitude and direction of its impact on estimates are hard to determine, 26 its effect cannot be mitigated by increasing sample size, 27,28 and it can be coupled with other data imperfections. 3,[29][30][31][32] Moreover, contrary to previous arguments, 33,34 recent evidence suggests that even genetic association analyses with inherited germline susceptibility factors can also be prone to selection bias.…”

Section: Background and Significancementioning

confidence: 99%

To weight or not to weight? Studying the effect of selection bias in three large EHR-linked biobanks

Salvatore,

Kundu,

Shi

et al. 2024

Preprint

View full text Add to dashboard Cite

Objective: To explore the role of selection bias adjustment by weighting electronic health record (EHR)-linked biobank data for commonly performed analyses. Materials and methods: We mapped diagnosis (ICD code) data to standardized phecodes from three 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 be more representative of the US adult population. We used weights previously developed for UKB to represent the UKB-eligible population. We conducted four common descriptive and analytic tasks 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 PheWAS for colorectal cancer, the strongest associations remained unaltered and there was large 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 limited impact on dimensionality estimation and large-scale hypothesis testing but impacted prevalence and association estimation more. Results from untargeted association analyses should be followed by weighted analysis when effect size estimation is of interest for specific signals. 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.

show abstract

Section: Background and Significancementioning

confidence: 99%

To weight or not to weight? Studying the effect of selection bias in three large EHR-linked biobanks

Salvatore,

Kundu,

Shi

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, I coined the term "data confession" [26] to encourage more disclosures in research publications about defects in data conceptualization, collection or pre-processing, as another component in enhancing the replicability and ultimately the reliability of published scientific studies, since data quality matters far more than data quantity [23,4,27,3]. The retrospective introspection summarized above suggests a more general data science confession (DSC) in our publications, where we can benefit from each others' mistakes and lessons learned, especially how we reason with ourselves, where we can engage in a pure intellectual dialogue without being distracted by suspicions of impure motivations.…”

Section: Taking a Lead In Data (Science) Confessionmentioning

confidence: 99%

A Not-so-radical Rejoinder: Habituate Systems Thinking and Data (Science) Confession for Quality Enhancement

Meng¹

2023

The New England Journal of Statistics in Data Science

View full text Add to dashboard Cite

“…As more variables are added, a causal graph may become too complex to interpret usefully. An important practical goal thus may be to reduce an overly complex causal graph to a more coarse-grained causal description that can be used to deal with the problem at hand [ 28 , 29 , 30 , 31 ]. For example, the causal diagram in Figure 2 C may be appropriate if a drug that modifies EMT is being studied.…”

Section: Causal Diagramsmentioning

confidence: 99%

A Causal Framework for Making Individualized Treatment Decisions in Oncology

Msaouel

Lee

Karam

et al. 2022

Cancers

Self Cite

View full text Add to dashboard Cite

We discuss how causal diagrams can be used by clinicians to make better individualized treatment decisions. Causal diagrams can distinguish between settings where clinical decisions can rely on a conventional additive regression model fit to data from a historical randomized clinical trial (RCT) to estimate treatment effects and settings where a different approach is needed. This may be because a new patient does not meet the RCT’s entry criteria, or a treatment’s effect is modified by biomarkers or other variables that act as mediators between treatment and outcome. In some settings, the problem can be addressed simply by including treatment–covariate interaction terms in the statistical regression model used to analyze the RCT dataset. However, if the RCT entry criteria exclude a new patient seen in the clinic, it may be necessary to combine the RCT data with external data from other RCTs, single-arm trials, or preclinical experiments evaluating biological treatment effects. For example, external data may show that treatment effects differ between histological subgroups not recorded in an RCT. A causal diagram may be used to decide whether external observational or experimental data should be obtained and combined with RCT data to compute statistical estimates for making individualized treatment decisions. We use adjuvant treatment of renal cell carcinoma as our motivating example to illustrate how to construct causal diagrams and apply them to guide clinical decisions.

show abstract

The Big Data Paradox in Clinical Practice

Cited by 25 publications

References 43 publications

To weight or not to weight? Studying the effect of selection bias in three large EHR-linked biobanks

To weight or not to weight? Studying the effect of selection bias in three large EHR-linked biobanks

A Not-so-radical Rejoinder: Habituate Systems Thinking and Data (Science) Confession for Quality Enhancement

A Causal Framework for Making Individualized Treatment Decisions in Oncology

Contact Info

Product

Resources

About