Privacy-protecting methods

TIDE’s portfolio of privacy-protecting analytic and data-sharing methods research, led by Sengwee Darren Toh, ScD, focuses on developing and applying cutting-edge analytic and data-sharing methods that integrate epidemiology, biostatistics, data science, and informatics to improve the validity, feasibility, and efficiency of multi-center studies, particularly in distributed data network studies that leverage routinely collected electronic health data.

Many of these methods support robust multivariable-adjusted statistical analysis without the need to share individual-level data, thereby providing better protection for patient privacy and confidentiality in multi-database studies.

Our guiding principles are:

  • Sending analysis to the data
  • Getting more (info) by asking for less (data)

Click here to learn more about our approaches to data sharing.

Click here to learn more about our ongoing projects.

Click here to learn more about our completed projects.

Related Publications

Her Q, Malenfant J, Zhang Z, Vilk Y, Young J, Tabano D, Hamilton J, Johnson R, Raebel M, Boudreau D, Toh S. A distributed regression analysis application in large distributed data networks: Analysis of precision and operational performance. JMIR Med Inform 2020;8(6):e15073

Toh S, Rifas-Shiman SL, Bailey LC, Forrest CB, Horgan CE, Lin PD, Lunsford D, Moyneur E, Sturtevant JL, Young JG, Block JP, on behalf of the PCORnet Antibiotics and Childhood Growth Study Group. Privacy-protecting multivariable-adjusted distributed regression analysis for multi-center pediatric study. Pediatr Res 2020;87(6):1086-1092

Shu D, Yoshida K, Fireman BH, Toh S. Inverse probability weighted Cox model in multi-site studies without sharing individual-level data. Stat Methods Med Res 2020;29(6):1668-1681

Huang TY, Welch EC, Shinde MU, Platt RW, Filion KB, Azoulay L, Maro JC, Platt R, Toh S. Reproducing protocol-based studies using parameterizable tools – Comparison of analytic approaches used by two national medical product surveillance networks. Clin Pharmacol Ther doi:10.1002/cpt.1698

Shu D, Toh S. ppmHR: A Privacy-Protecting Tool to Fit Inverse Probability Weighted Cox Models in Multi-Site Studies. Epidemiology. 2020;Publish Ahead of Print. doi:10.1097/EDE.0000000000001300

Related Links

Principal Investigator

Sengwee Darren Toh, ScD