November 7, 2023

Increasing opioid use disorder treatment in primary care


A trial led by KPWHRI researchers found that adding nurse care managers helped more people get needed treatment

Most people with opioid use disorder, which is defined as an addiction to opioids, never receive medication treatment, even though it is effective, lifesaving, and can be provided in primary care. New findings from the PRimary Care Opioid Use Disorder treatment (PROUD) trial, published in JAMA Internal Medicine, show that an intervention tested in 12 primary care clinics across 6 health care systems significantly increased medication treatment for opioid use disorder, although the benefit was uneven across sites and patient populations.

“We are in the middle of an opioid overdose epidemic, but most primary care providers do not treat opioid use disorder even though effective medications can be prescribed in primary care,” said Paige Wartko, PhD, MPH, a collaborative scientist at Kaiser Permanente Washington Health Research Institute (KPWHRI) and co-lead author of the study. “Moreover, about 75% of people with opioid use disorder receive no medication treatment at all.”

The study is the first randomized controlled trial of an intervention that successfully increased medication treatment for opioid use disorder in primary care. The researchers tested the Massachusetts Model, which adds a nurse care manager to primary care teams. The nurses provide easy access to care and support ongoing high-quality treatment for opioid use disorder. The model was previously effective in publicly financed health systems in Massachusetts.

Clinics randomly selected to implement the intervention received funding for a full-time nurse care manager. They also were required to have at least 3 primary care providers who agreed to prescribe buprenorphine, a medication treatment for opioid use disorder approved by the Food and Drug Administration. Expert nurses at Boston Medical Center, where the model was developed, provided training and weekly videoconference support for all nurse care managers.

On average, the primary care clinics where the intervention was tested provided more opioid use disorder treatment during the 2 years of the trial than those providing usual care. The increase seen in the intervention clinics was equivalent to each clinic treating about 8 more patients for a year for every 10,000 primary care patients, as compared to usual care. By another measure, intervention clinics increased treatment for patients with recognized opioid use disorder by 21%, compared to 2.8% in usual care clinics. The intervention also increased the number of patients who remained in treatment longer than 6 months.

“We tested a practical model that showed effectiveness in Massachusetts to see if it could increase treatment and improve outcomes in diverse health systems across the country,” said Kathy Bradley, MD, MPH, a senior investigator at KPWHRI and co-lead author on the study. “What we found is that it increased treatment of opioid use disorder, which is a very important outcome, but that it was more effective in some health systems than others.”

The researchers found that the majority of the benefit was seen in 2 out of the 6 health systems testing the intervention, and primarily among patients who were new to treatment for opioid use disorder. Over the 2-year study, the most successful intervention clinic saw an increase that was equivalent to around 15 more patients being treated for a year for every 10,000 primary care patients, compared with usual care. On the other hand, some clinics saw no benefit. The intervention did not decrease use of urgent care, emergency department visits, and hospitalization among patients with documented opioid use disorder before the trial began. The study team plans to publish more in-depth analyses looking at use of urgent care, emergency care services, and hospitalization.

The investigators identified several key factors associated with the success of the intervention. These included broad support from health system leaders, full financial coverage for opioid use disorder treatment, and straightforward ways for patients to access the nurse care manager, such as a direct phone line.

The clinics were located in New York, Florida, Michigan, Texas, and Washington, and served about 20,000 patients each, with varying race and ethnicity, insurance coverage, and socioeconomic status. The health systems included 2 academically affiliated safety net systems (health care systems that don’t turn away patients without insurance), 1 community health system, 2 integrated health systems, and 1 university system. 

The researchers plan to publish another paper reporting intervention outcomes at 3 years, as well as a breakdown of how the intervention performed across race, ethnicity, and sex.  

“This evaluation was early — nurses had only been in the clinics 9 to 20 months — so we want to see if the intervention continued to increase treatment over time, and also look more closely at patients for whom the intervention increased opioid use disorder treatment,” Bradley said.

KPWHRI coauthors on the study are Jennifer Bobb, PhD; Onchee Yu, MS; Noorie Hyun, PhD; Abisola Idu, MPH, MS; Joe Glass, PhD, MSW; Rebecca Phillips, MA; Mary Shea, MA; Gwen Lapham, PhD, MPH, MSW; and Megan Addis, BA, as well as collaborative authors Leah Hamilton, PhD, and Rachael Burganowski, MS. Research discussed in this story was supported by the Clinical Trials Network of the National Institute On Drug Abuse of the National Institutes of Health under Award Number UG1DA040314 (CTN-0074). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

By Amelia Apfel

healthy findings blog


Partnering to advance lifesaving care

Amy Lee describes how the LHS Program is helping KP Washington track and improve treatment for opioid use disorder.

Healthy Findings Blog


Understanding the needs of people seeking care for substance use

Joe Glass explores how to engage patients with treatment for alcohol or drug use.

New findings


Better tools for assessing substance use disorder

A simple checklist developed at KPWHRI does well at measuring symptoms of substance use disorder.