Skip to main content

Controlling for Confounding in Complex Survey Machine Learning Models to Assess Drug Safety and Risk

  • Chapter
  • First Online:
Machine Learning and Deep Learning in Computational Toxicology

Part of the book series: Computational Methods in Engineering & the Sciences ((CMES))

Abstract

Propensity scores are used to adjust for systemic confounding in conventional observational studies. Currently, methods for applying propensity scores to complex survey data remain virtually unknown, which discourages analysis on survey data more complex than that of prevalence. The National Health and Nutrition Examination Survey (NHANES) is a long-running, complex survey designed to assess the health and well-being of adults and children in the United States. At its core, NHANES is a cross-sectional observational survey subject to confounding issues like any other observational study. In addition, statistical methods must adequately account for the NHANES survey design and incorporate the sampling weights to make inferences about the larger US population. This chapter demonstrates the application of propensity scores to NHANES complex survey data through adjustments of the survey’s sampling weights. A viable approach for controlling confounding in complex observational surveys could open a new frontier for machine learning models and analysis in toxicological and medication studies with NHANES and other complex survey data.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 129.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 169.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
Hardcover Book
USD 169.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

References

Download references

Acknowledgements

The author thanks Joanne Berger, FDA Library, for manuscript editing assistance.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Paul Rogers .

Editor information

Editors and Affiliations

Additional information

Disclaimer: This chapter reflects the views of the author and does not necessarily reflect those of the U.S. Food and Drug Administration. Any mention of commercial products is for clarification only and is not intended as approval, endorsement, or recommendation.

Rights and permissions

Reprints and permissions

Copyright information

© 2023 This is a U.S. government work and not under copyright protection in the U.S.; foreign copyright protection may apply

About this chapter

Check for updates. Verify currency and authenticity via CrossMark

Cite this chapter

Rogers, P. (2023). Controlling for Confounding in Complex Survey Machine Learning Models to Assess Drug Safety and Risk. In: Hong, H. (eds) Machine Learning and Deep Learning in Computational Toxicology. Computational Methods in Engineering & the Sciences. Springer, Cham. https://doi.org/10.1007/978-3-031-20730-3_14

Download citation

Publish with us

Policies and ethics