June 22, 2023
1:30pm – 3:00pm
TITLE: Buprenorphine Use in the Perioperative Period
INVESTIGATOR: Sesh Mudumbai (1, 2)
(1) Department of Anesthesiology, Perioperative, and Pain Medicine
(2) VA Palo Alto HCS
DATE: Wednesday, 15 March 2023
TIME: 1:30–3:00 PM
LOCATION: Conference Room X399, Medical School Office Building, 1265 Welch Road, Stanford, CA
WEBPAGE: https://dbds.stanford.edu/data-studio/
ABSTRACT
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.
Introduction
The prevalence of opioid use disorders (OUD) in surgical patients has increased over the past two decades. The treatment guidelines for OUD recommend Opioid Agonist Therapy (OAT) such as buprenorphine to decrease illicit opioid use and overdose. However, the use of buprenorphine in acute care, particularly surgery, has conflicting management approaches (continue vs. interrupt) due to varying emphasis on postoperative pain vs. OUD-related risks. Comparative safety and effectiveness data for perioperative management strategies for buprenorphine are limited, and no randomized controlled trials exist. This study aims to evaluate the comparative safety and effectiveness of perioperative buprenorphine utilizing state-of-the-art epidemiological methods and identify optimal strategies for individual management. At the end of this study, we will have substantial evidence to guide clinical practice guidelines regarding the continuation or interruption of buprenorphine prior to surgical procedures among COU.
Hypothesis & Aim
The overall goal of this study is to generate unbiased or minimally biased, population-level estimates of the comparative safety and effectiveness of perioperative buprenorphine and identify optimal strategies for individual management.. The study aims are to evaluate (1) the comparative safety of current perioperative regimens (interruption vs. continuation) for buprenorphine on OUD-related outcomes, (2) the comparative effectiveness of current perioperative regimens (interruption vs. continuation) and formulations for buprenorphine on pain-related outcomes, on opioid-related adverse drug events (ORADES) and healthcare utilization; and (3) identify subgroups and factors to identify the optimal strategy of interruption vs. continuation for individual level decision making.
Dataset
Our strategy involves four phases. First, we will utilize a national-level Veterans Health Affairs (VHA) cohort of VA surgical patients who received buprenorphine preoperatively. Then, the cohort will be enriched with Centers for Medicare & Medicaid Services (CMS) and Social Determinants of Health (SDoH) data to better track risk factors and outcomes. Second, we will evaluate the relationship between regimens and post-discharge safety outcomes, including opioid-related emergency visits and hospitalizations. Third, we will look at how prescribing patterns affect pain-related outcomes. Finally, we’ll evaluate the impact of these regiments on ORADES and healthcare utilization, taking into account a variety of SDoH. Using OMOP’s common data model of a few domains and concept IDs, we identified an initial population of 5,146 patients satisfying the study criteria.
Statistical models
The study will use a cohort design and a variety of statistical methods. For our first two aims, we will plan to use logistic regression, to evaluate the effect of buprenorphine on clinical outcomes, pain-related outcomes, and ORADES ,and healthcare utilization. The analysis will adjust for potential confounding variables and use propensity scores and inverse probability of treatment weighting to balance covariates. We will use high-dimensional propensity scores to account for proxies of unmeasured confounders, and a doubly robust estimator to mitigate the problems of estimating models with many variables. The potential confounders will include socio-economic descriptors, clinical measures, and SDoH variables. A variety of sensitivity analyses will address residual confounding, exposure misclassification, outcome misclassification, and selection bias. For our third aim, we will plan to use machine learning methods including clustering approaches k-means clustering and others.
- What is the optimal approach to set up doubly robust estimators and the impact on the Average Treatment Effect of the Treated (ATT) vs Average Treatment Effect (ATE)
- What are the set of sensitivity analyses? The role of negative control methods?
- What sensitivity analyses are needed?
- How should we validate and present our preliminary data and algorithms?
- What types of machine learning approaches should we use and why ? prediction of receipt of
ZOOM MEETING INFORMATION
Join from PC, Mac, Linux, iOS or Android: https://stanford.zoom.us/j/97196061848?pwd=ajY3MmJOUU9oYitMdFZXL3NQYmFEZz09
Password: 571460
Or iPhone one-tap (US Toll): +18333021536,,97196061848# or +16507249799,,97196061848#
Or Telephone:
Dial: +1 650 724 9799 (US, Canada, Caribbean Toll) or +1 833 302 1536 (US, Canada, Caribbean Toll Free)
Meeting ID: 971 9606 1848
Password: 571460
International numbers available: https://stanford.zoom.us/u/aeA1opIz3O
Meeting ID: 971 9606 1848
Password: 571460
Password: 571460TITLE: Data Studio Office Hour