Background: Benzodiazepine (BZD) use is a significant and persistent concern among older Veterans given the association of BZDs with adverse outcomes including falls, fractures, motor vehicle accidents, impaired cognition, increased risk of dementia, and mortality. In FY2015, over 82,000 Veterans ?75 years of age were prescribed a BZD despite abundant evidence of harms from both acute and chronic use and recommendations from professional organizations to limit use. Beginning in FY2013, the VA's Psychotropic Drug Safety Initiative (PDSI) began monitoring a variety of psychotropic quality measures, including BZD use among Veterans ?75 years of age (hereafter referred to as ?BZD?75?). Among the 52 facilities (out of 140) that selected BZD?75 as a priority, each developed its own local strategy to improve prescribing. In order for the VA to build upon local successes at addressing the seemingly intractable problem of BZD prescribing, the aim of this proposal is to understand the combinations of local facility strategies and context that generate these outcomes. In addition, it is critical to understand the patient experience of these strategies, as Veterans may experience some ?successful? strategies as extremely distressful. This work may directly influence the clinical care of Veterans and is of both great interest and immediate relevance to our operational partners, including PDSI, Pharmacy Benefits Management, and the Academic Detailing Service. Objectives:
Aim 1 : Identify high- and low-performing facilities on acute and chronic BZD prescribing among facilities that prioritized the BZD?75 measure.
Aim 2 : Assess facility-level strategies and associated barriers and facilitators to addressing the BZD?75 goal along with the acceptability of those strategies to older Veterans.
Aim 3 : Identify and pilot test context-sensitive strategies for facilities to successfully reduce both acute and chronic BZD use among older Veterans. Methods: We will use interrupted time-series analyses of national VA administrative data to examine BZD prescribing across BZD?75 priority facilities (n=52). We will specifically measure change in: (1) rate of acute use (new BZD initiation) and (2) intensity of chronic use (average daily dose [mg/day]). As an exploratory sub- aim, we will measure the rate of fall-related injury by facility to determine whether change in BZD prescribing is associated with change in associated adverse outcomes.
In Aim 2 a we will conduct semi-structured telephone interviews with up to 16 PDSI local facility champions and additional staff from top- and bottom-performing facilities (identified in Aim 1), followed by on-site, semi-structured interviews with key stakeholders at up to 6 sites. We will gather information on facility-level strategies, barriers, and facilitators that may help or hinder facility efforts. Finally, we will conduct telephone interviews with older Veterans prescribed chronic BZDs that attempted a taper to learn whether patients experienced these facility strategies as distressful. An expert panel of clinicians, researchers, and administrators will review the strategies, local context, associated BZD prescribing outcomes, and patient acceptability to develop a toolkit of context-sensitive best practices facilities can implement to address BZD use, which we will then pilot test.

Public Health Relevance

Use of benzodiazepines (BZDs) is associated with health problems for older adults, including falls, fractures, automobile accidents, difficulty thinking, and even death. Despite this, over 82,000 Veterans 75 years of age or older were prescribed this class of medication in 2015. As a result of the VA's Psychotropic Drug Safety Initiative, 52 facilities have focused on reducing BZD prescribing, though success across these facilities has varied. While success is partially related to the specific strategy pursued, it is also related to the characteristics of that particular VA facility. In addition, while some facilities may have reduced BZD use, Veterans may have experienced their strategies as especially distressing. Ultimately, continuing to reduce BZD prescribing across the VA requires understanding the combination of strategies and context that can lead to successful outcomes, as well as how to minimize distress to Veterans.

National Institute of Health (NIH)
Veterans Affairs (VA)
Non-HHS Research Projects (I01)
Project #
Application #
Study Section
Blank (HSR6)
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Veterans Health Administration
Ann Arbor
United States
Zip Code