Bayesian Models to Individualize Health Care

Hopkins inHealth researchers are developing statistical models that integrate patient data to improve diagnosis of and treatment for a range of diseases. In 2015, Hopkins inHealth received funding from the Patient-Centered Outcomes Research Institute* (PCORI) to build Bayesian hierarchical models using data from patients with prostate cancer, childhood pneumonia, and depression as case studies. The project is also developing statistical software that will allow researchers to apply these models. See below for more information on the work derived from this project.

Statistical Models

Publications

  • Coley RY, Fisher AJ, Mamawala M, Carter HB, Pienta KJ, Zeger SL. "A Bayseian Hierarchical Model for Prediction of Latent Health States from Multiple Data Sources with Application to Active Surveillance of Prostate Cancer." (2016). Link.
  • Ogburn E and Zeger SL. "Statistical reasoning and methods in epidemiology on the occasion of the 100th anniversary of the Johns Hopkins School of Public Health." (2016). American Journal of Epidemiology. Link.
  • Wu Z, Deloria-Knoll M, Hammitt LL, Zeger SZ and for the Pneumonia Etiology Research for Child Health Core Team. "Nested partially-latent class models for dependent binary data; Estimating disease etiology." (Under revision). Link.
  • Coley RY, Zeger SL, Mamawala M, Fisher AJ, Pienta KJ, Carter HB. "Prediction of the cancer state to inform a personalized management program for prostate cancer." (In preparation). 
  • Coley RY, Deng D, Du Y, Ji Z, Rao K, Wu Z, Zhu Y. "Predicting Prostate Cancer Survival; A Multiple-Imputation-Assisted Super Learning Approach." F1000 Research (DREAM Challenges Channel). (Under review).
  • Fisher AJ, Coley RY, Zeger SL. "Fast Out-of-Sample Predictions for Bayesian Hierarchical Models of Latent Health States." (2015). Link.
  • Wu Z and Zeger SL. "Bayesian Regression Analysis for Estimating Disease Etiology from Case-Control Data." (In preparation). 
  • Wu Z and Zeger SL. "Individualizing Health with Longitudinal Measurements and Feedback in Treatment Assignments." (In preparation).
  • Wu Z and Zeger SL. "Sparse Latent Class Regression for Multivariate Binary Data; A Bayesian Approach." (In preparation). 
  • Wu Z and Zeger SL. "baker: Bayesian Analytic Kit for Etiology Research." (In preparation). 

Presentations

  • R. Yates Coley. "Optimizing Surveillance of Low Risk Prostate Cancer." (2015). International Conference on Health Policy Statistics; Providence, RI. Link.
  • Scott Zeger. "A Statistical Framework for Individualizing Healthcare." (2015). Royal Statistical Society Invited Lecture, Essex, England.
  • Scott Zeger. "Individualized and Population Health: Two Sides of the Same Coin; Lessons from Jon W. Tukey in honor of his 100th Birthday." (2015). Princeton University.
  • Scott Zeger. "Hopkins inHealth." (2015). Johns Hopkins Applied Physics Laboratory, National Health Seminar Series.
  • R. Yates Coley. "Precision Medicine, Learning Health Systems, and Improving Surveillance of Low Risk Prostate Cancer." (2015). Data Science Affinity Group, Fred Hutchinson Cancer Research Center, Seattle, WA. Link.
  • R. Yates Coley. "Optimizing Surveillance of Low Risk Prostate Cancer." (2015). Pacific NW Specialized Program of Research Excellence (SPORE) in Prostate Cancer. Link.
  • Zhenke Wu. "Informative Bayes Models for Estimating Disease Etiology." (2015). Biostatistics Grand Rounds, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.
  • R. Yates Coley. "Optimizing Surveillance of Low Risk Prostate Cancer: An Application of Precision Medicine and Learning Health Systems at Johns Hopkins." (2015). Data Science Interest Group, Johns Hopkins Medicine, Baltimore, MD. Link.
  • R. Yates Coley. "Active Surveillance Modeling and Decision-Making at Johns Hopkins." (2015). Cancer Intervention and Surveillance Monitoring Network (CISNET) – Prostate, National Cancer Institute, Maryland. Link.
  • R. Yates Coley. "Precision Medicine, Learning Health Systems, and Improving Surveillance of Low Risk Prostate Cancer." (2016). Department of Healthcare Policy and Research, Weill Cornell Medicine, Ithaca, NY. Link.
  • R. Yates Coley. "Precision Medicine, Learning Health Systems, and Improving Surveillance of Low Risk Prostate Cancer." (2016). Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD. Link.
  • R. Yates Coley. "Optimizing Surveillance of Low Risk Prostate Cancer." (2016). High Value Research Symposium, Johns Hopkins Medicine, Baltimore, MD.
  • R. Yates Coley. "Precision Medicine, Learning Health Systems, and Improving Surveillance of Low Risk Prostate Cancer." (2016). Biostatistics Research Branch, National Institute of Allergy and Infectious Diseases, Bethesda, MD. Link.
  • R. Yates Coley. "Precision Medicine, Learning Health Systems, and Improving Surveillance of Low Risk Prostate Cancer." (2016). RAND Corporation, Santa Monica, CA. Link.
  • R. Yates Coley. "Precision Medicine, Learning Health Systems, and Improving Surveillance of Low Risk Prostate Cancer." (2016). Stanford University, School of Medicine, Biostatistics, Workshop in Biostatistics. Link.
  • Scott Zeger. "Biomedical Data Science: A Statistical Perspective." (2016). Stanford University, School of Medicine.

Meeting Minutes

  • Prostate Cancer Advisory Board Meeting, February 16, 2016. Minutes
  • Prostate Cancer Advisory Board Meeting, May 17, 2016. Minutes.
  • Prostate Cancer Advisory Board Meeting, November 15, 2016. Minutes.
  • OSLER inHealth Advisory Board Meeting, February 18, 2016. Minutes.
  • OSLER inHealth Advisory Board Meeting, August 16, 2016. Minutes.

Research

  • Glazer, K. "Multi-symptom trajectories and correlation with sucidal behavior." National Network of Depression Centers 2016 Annual Conference. Abstract. Link.

*Research reported on this page was partially funded through a Patient-Centered Outcomes Research Institute (PCORI) Award ME-1408-20318.

The statements published are solely the responsibility of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute (PCORI), its Board of Governors or Methodology Committee.

Join Our Mailing List:

Please enter the word you see in the image below: