The Medical Deconfounder: Assessing Treatment Effects with Electronic Health Records

Linying Zhang, Yixin Wang, Anna Ostropolets, Jami J. Mulgrave, David M. Blei, George Hripcsak
; Proceedings of the 4th Machine Learning for Healthcare Conference, PMLR 106:490-512, 2019.

Abstract

The treatment effects of medications play a key role in guiding medical prescriptions. They are usually assessed with randomized controlled trials (RCTs), which are expensive. Recently, large-scale electronic health records (EHRs) have become available, opening up new opportunities for more cost-effective assessments. However, assessing a treatment effect from EHRs is challenging: it is biased by unobserved confounders , unmeasured variables that affect both patients’ medical prescription and their outcome, e.g. the patients’ social economic status. To adjust for unobserved confounders, we develop the medical deconfounder , a machine learning algorithm that unbiasedly estimates treatment effects from EHRs. The medical deconfounder first constructs a substitute confounder by modeling which medications were prescribed to each patient; this substitute confounder is guaranteed to capture all multi-medication confounders, observed or unobserved (Wang and Blei , 2018 ). It then uses this substitute confounder to adjust for the confounding bias in the analysis. We validate the medical deconfounder on two simulated and two real medical data sets. Compared to classical approaches, the medical deconfounder produces closer-to-truth treatment effect estimates; it also identifies effective medications that are more consistent with the findings in the medical literature.

Cite this Paper


BibTeX
@InProceedings{pmlr-v106-zhang19a, title = {The Medical Deconfounder: Assessing Treatment Effects with Electronic Health Records}, author = {Zhang, Linying and Wang, Yixin and Ostropolets, Anna and Mulgrave, Jami J. and Blei, David M. and Hripcsak, George}, booktitle = {Proceedings of the 4th Machine Learning for Healthcare Conference}, pages = {490--512}, year = {2019}, editor = {Finale Doshi-Velez and Jim Fackler and Ken Jung and David Kale and Rajesh Ranganath and Byron Wallace and Jenna Wiens}, volume = {106}, series = {Proceedings of Machine Learning Research}, address = {Ann Arbor, Michigan}, month = {09--10 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v106/zhang19a/zhang19a.pdf}, url = {http://proceedings.mlr.press/v106/zhang19a.html}, abstract = {The treatment effects of medications play a key role in guiding medical prescriptions. They are usually assessed with randomized controlled trials (RCTs), which are expensive. Recently, large-scale electronic health records (EHRs) have become available, opening up new opportunities for more cost-effective assessments. However, assessing a treatment effect from EHRs is challenging: it is biased by unobserved confounders , unmeasured variables that affect both patients’ medical prescription and their outcome, e.g. the patients’ social economic status. To adjust for unobserved confounders, we develop the medical deconfounder , a machine learning algorithm that unbiasedly estimates treatment effects from EHRs. The medical deconfounder first constructs a substitute confounder by modeling which medications were prescribed to each patient; this substitute confounder is guaranteed to capture all multi-medication confounders, observed or unobserved (Wang and Blei , 2018 ). It then uses this substitute confounder to adjust for the confounding bias in the analysis. We validate the medical deconfounder on two simulated and two real medical data sets. Compared to classical approaches, the medical deconfounder produces closer-to-truth treatment effect estimates; it also identifies effective medications that are more consistent with the findings in the medical literature.} }
Endnote
%0 Conference Paper %T The Medical Deconfounder: Assessing Treatment Effects with Electronic Health Records %A Linying Zhang %A Yixin Wang %A Anna Ostropolets %A Jami J. Mulgrave %A David M. Blei %A George Hripcsak %B Proceedings of the 4th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2019 %E Finale Doshi-Velez %E Jim Fackler %E Ken Jung %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v106-zhang19a %I PMLR %J Proceedings of Machine Learning Research %P 490--512 %U http://proceedings.mlr.press %V 106 %W PMLR %X The treatment effects of medications play a key role in guiding medical prescriptions. They are usually assessed with randomized controlled trials (RCTs), which are expensive. Recently, large-scale electronic health records (EHRs) have become available, opening up new opportunities for more cost-effective assessments. However, assessing a treatment effect from EHRs is challenging: it is biased by unobserved confounders , unmeasured variables that affect both patients’ medical prescription and their outcome, e.g. the patients’ social economic status. To adjust for unobserved confounders, we develop the medical deconfounder , a machine learning algorithm that unbiasedly estimates treatment effects from EHRs. The medical deconfounder first constructs a substitute confounder by modeling which medications were prescribed to each patient; this substitute confounder is guaranteed to capture all multi-medication confounders, observed or unobserved (Wang and Blei , 2018 ). It then uses this substitute confounder to adjust for the confounding bias in the analysis. We validate the medical deconfounder on two simulated and two real medical data sets. Compared to classical approaches, the medical deconfounder produces closer-to-truth treatment effect estimates; it also identifies effective medications that are more consistent with the findings in the medical literature.
APA
Zhang, L., Wang, Y., Ostropolets, A., Mulgrave, J.J., Blei, D.M. & Hripcsak, G.. (2019). The Medical Deconfounder: Assessing Treatment Effects with Electronic Health Records. Proceedings of the 4th Machine Learning for Healthcare Conference, in PMLR 106:490-512

Related Material