PROJECT SUMMARY
DAT18-06 The personal, social, and public health burdens of substance use disorders (SUDs) in the United
States have been compounded by the opioid epidemic and legalization of recreational and medical cannabis
use. Traditionally, healthcare systems have referred patients with SUDs to specialists. Although SUD treatment
is associated with improved outcomes, most people with SUDs remain unidentified and untreated. The
magnitude of unmet need has led experts to recommend incorporating screening and assessment for SUDs
into routine primary care (PC), to increase rates of diagnosis and treatment of SUDs. Despite extensive
observational research, to our knowledge no randomized controlled trial has evaluated the impact of
PC-based programs of screening and assessment on new SUD treatment. The proposed study takes
advantage of a unique and fortuitous opportunity to evaluate the impact of a PC program of screening and
assessment for SUDs on new diagnosis and treatment of SUDs. The Sustained Patient-centered Alcohol
Related Care (SPARC) trial, completed in July 2018, was a rigorously-designed randomized implementation
trial testing a multifaceted approach to improving care for unhealthy alcohol use. However, at the request of
clinical leaders at Kaiser Permanente Washington (KPWA) where the SPARC program was implemented,
routine screening and assessment for opioid, cannabis, and other drug use disorders (CUD, OUD, and DUD,
respectively) were added before the trial was launched in 22 KPWA PC clinics. As a result, the SPARC trial's
implementation program—consisting of practice coaching, electronic health record (EHR) support,
performance monitoring and feedback, and provider training—also targeted SUDs. Data from health system
monitoring reports indicate the trial achieved high, sustained rates of screening for SUDs, exceeding the target
of 80% of patients with PC visits. However, the SPARC trial was only funded to evaluate alcohol-related
outcomes. The proposed study will rigorously evaluate the SPARC program's impact on the clinically
meaningful primary outcome of new SUD treatment within 90 days of a primary care visit (Aim 1). To
illuminate potential continued gaps in the quality of SUD care (Aim 2), we will use natural language processing
(NLP) to extract and quantify reasons why: 1) patients reporting =4 DSM-5 symptoms of SUDs do not receive
formal coded diagnoses of OUD, CUD, or DUD, and 2) why patients with newly diagnosed OUDs, CUDs, and
DUDs do not received treatment. Most people with OUDs, CUDs, and other DUDs are never treated. An
estimated 2.5-3.9% of US adults have past-year SUDs, and nearly 90% report no SUD treatment. Yet >84%
of US adults have outpatient visits annually, and over half of these visits are in PC. If the proposed innovative
study demonstrates that screening and assessment in PC increase SUD treatment, it would provide support for
a widely endorsed yet previously untested approach to addressing a critically important gap in the quality of
care for SUDs.
