October 5, 2022

1:30 pm / 3:00 pm

TIME: 1:30–3:00 PM


Theresa Lii (1)

Boris Heifets (1)

(1) Anesthesiology, Perioperative, and Pain Medicine

LOCATION: Conference Room X303, Medical School Office Building, 1265 Welch Road, Stanford, CA



The Data Studio Workshop brings together a biomedical investigator with a group of experts for an in-depth session to solicit advice about statistical and study design issues that arise while planning or conducting a research project. This week, the investigator(s) will discuss the following project with the group.

This is a single-site, double-blinded RCT to evaluate the antidepressant superiority of IV ketamine compared to placebo, when administered during surgery (while under anesthesia) to adult patients with symptomatic major depressive disorder.

Participants were randomly allocated in a 1:1 ratio to one of two groups: the intervention group (n=20) received a single administration of IV ketamine during surgery; the placebo group (n=20) received IV saline during surgery. The study drug was given after anesthetic induction to ensure participant blinding. Healthcare providers, investigators, and outcomes assessors were also blinded.

Our primary outcome measure is the Montgomery-Asberg Depression Rating Scale (MADRS), which is widely used in depression trials. Baseline MADRS scores were obtained during screening and immediately prior to surgery. MADRS scores were also collected on postoperative days 1, 2, 3, 5, 7, and 14. Secondary outcomes include the Hospital Anxiety and Depression Scale (HADS), pain scores, opioid use, and hospital length of stay.

Our pre-specified analysis method uses linear mixed modeling (specifically MMRM) using the following timepoints: day 0 (baseline), and postoperative days 1, 2 and 3.


This study has completed enrollment, randomization, and collection of all pre-specified outcomes.

  1. Are the linear mixed model parameters we chose appropriate for our data? (e.g. random intercepts, random slopes, interaction term)
  2. Is an unstructured covariance structure appropriate for our data?
  3. Does our interpretation of the linear mixed model results make sense?