Pragmatic healthcare-integrated randomized trials require randomization for the right patients or clinicians at the right moment of time without interrrupting clinical workflows. Pseudo-randomization (e.g., using odd vs. even medical record numbers) or simple randomization without stratification or blocking is not hard to program, but these methods can yield imbalanced randomization arms and/or bias from predictability of individual assignments.
What It Does
- The Point-Of-Care Randomization (POCR) engine executes stratified, blocked randomization in real time at the point of care.
Features
- Programmable triggering of randomization events in real time
- Stratification and blocking with randomly varying block sizes
- Flexible unit of randomization (e.g., patients, encounters, clinicians)
- Flexible number of arms and randomization ratios
- Randomization assignments are immediately available for use by EHR system (e.g., filter rules in Epic) and saved for subsequent use as needed
- Snapshot of strata and other context variables saved along with randomization assignments for later use
Versions
- Original version programmed directly in Epic (full-featured, requires special installation and permission from Epic)
- CDS Hooks Standard (lightweight install, uses the CDS Hooks standard)
Get Started
UCSF can provide information, code, and implementation guidance.
Email us at [email protected]
The UCSF Clinical and Translational Science Institute (CTSI) is part of the Clinical and Translational Science Award (CTSA) program funded by the National Center for Advancing Translational Sciences (NCATS) at the National Institutes of Health (NIH) (Grant Number UL1 TR001872). The content of this website is solely the responsibility of CTSI and does not necessarily represent the official views of the NIH.