A compelling method for working in the space between addiction research and practice
It’s often said that in the addiction field, practice lags behind research–about 17 years behind. This issue is widely-recognized and (as you might imagine) considered a big problem.
On the other hand, it’s hard to argue with assertions that research can’t keep up with practice: study design, participant recruitment, and the four phases of a clinical trial takes years. In the meantime, practitioners keep treating people and encountering new sorts of questions.
What’s clear is that research and practice are not aligned. They are different beasts; they move at different paces and they generate different types of information. Research is good at asking if something works, but not when it should be used. Or, in real world settings, how it should be used.
As a public health student, I’m bombarded with questions and ideas about improving healthcare delivery by better aligning science, service and policy. I also have the good fortune of encountering these same complex, exciting issues in my work at IRETA. It’s a fun confluence.
One of the areas that IRETA works to narrow the research-practice gap is in medication assisted treatment (MAT) for substance use disorders (SUDs). And one of our solutions is the use of the RAND/UCLA Appropriateness Method (RAM), which I’ve come to see as an extremely cool approach to marrying research and practice.
Medication-Assisted Treatment for Substance Use Disorders
There’s a lot of work to do in MAT. Although MAT is an evidence-based practice, more and more questions are surfacing around how to deliver it in the real world to improve outcomes. Questions like:
- How can we offer MAT to a broader population?
- Which patients are best suited to which medications?
- Which patients are best suited to which treatment settings?
- What are the best guidelines to use around dose, duration, monitoring, and concurrent psychosocial treatment?
Research and practice in this area are both moving and changing quickly. Buprenorphine (Suboxone, Subutex) for opioid dependence has exploded onto the market (half a million people now use it) and has been joined by naltrexone for opioid and alcohol dependence. Other addiction medications are right on their tails. Human testing for a cocaine vaccine is expected to begin within a year.
Amid this flurry, in the last two years, IRETA has created treatment practice guidelines in two areas—how to use buprenorphine for opioid dependence treatment and how to manage patient benzodiazepine use in MAT. We’ve been able to do that using the RAM.
How the RAM Works
The RAM was developed in the mid 1980s to provide a standard methodology for determining when procedures in healthcare are over- or underutilized. Its purpose was to be an alternative to randomized clinical trials, which—although the “gold standard for evidence-based medicine—are generally either not available or cannot provide evidence at a level of detail sufficient to apply to the wide range of patients seen in everyday clinical practice.”
Right. So, because research doesn’t always directly apply to practice, the RAM was born.
Known as a “modified Delphi process” (after the ancient Greek Oracle at Delphi, which was believed to prophesize the future), the RAM combines the best available scientific evidence with the collective judgment of experts to yield recommendations about appropriate treatment in the real world. Here’s how it works:
- A panel of experts is assembled
- A review of available research literature is synthesized into hypothetical guideline statements
- These statements are sent to each panel member, who independently rates their “appropriateness” on a 9-point scale
- Statements that generate disagreement or reveal uncertainty are discussed in an extensive meeting among the experts, who explain their individual ratings and share clinical and research knowledge
- After the discussion, each expert independently re-rates statements based on insights from the meeting
- The final product is a list of statements classified as “appropriate,” “uncertain” and “inappropriate”
Why the RAM is Cool
Sure, not everyone is as enthused about this topic as I am, but hear me out. In public health, we talk about various ways to conduct group decision-making and I see clear advantages to this one.
No goal of consensus. It does not force consensus and is therefore less time-consuming and the results are less compromised than a consensus panel.
Allows for independent ratings. At the end of the day, each expert rates guidelines independently, although final ratings are made in light of discussion with other panelists. Independent ratings avoid the Groupthink pitfall and distinguish the RAM from consensus panels, nominal group processes, and other public methods of decision-making.
Creates shared perspectives and synergy. In the two projects IRETA has conducted, the composition of the expert panel itself has linked research and practice. Panelists have been researchers, practitioners, and researcher/practitioners. This direct communication has led to guidelines that represent and respect the perspectives of both sides.
In the thirty years it has been used, the RAM has been extensively tested for its internal consistency and external validity. While critics have rightly pointed out that rigorous guideline development using any methodology does not guarantee a product of excellent technical quality, many research studies have scrutinized the RAM and shown it to a) be as reliable and reproducible as commonly used diagnostic methods and b) have predictive validity.
And the RAM has been used to consider a staggering array of scenarios. Here’s a smattering:
“Who should get liver transplants?”
“Where should we locate a field hospital after an earthquake?”
“What is the best way to manage common bile duct stones?”
“Under which circumstances should we use diagnostic imaging (like CT scans)?”
“When should elderly patients undergo surgery?”
We use the RAM because there will never be randomized control trials or clear research evidence for every clinical scenario. Life is full of judgment calls. But medical care—and addiction treatment in particular—can certainly be improved and systematized. The issue of uneven quality in addiction treatment has, in fact, garnered popular attention recently, taken up by both David Sheff and Anne Fletcher.
As ONDCP Director Gil Kerlikowkse has pointed out, the U.S. has the world’s top addiction researchers. Given that, I’d say that linking research and practice to improve outcomes is a moral imperative, and one we can’t avoid as healthcare reform rolls out.
Reducing costs and improving care are twin goals of Obamacare that will affect behavioral and physical healthcare alike. In the fog of the ACA’s rollout, lots of abstract ideas are being tossed around: precision healthcare, patient-centered care, performance-based payment.
The RAM is a concrete way of thinking about how some of these goals can be achieved. It’s a strategy to draw from the breadth of research and clinical wisdom that we have available to us to improve our ability to address specific clinical situations, so we’re not just shooting fish in a barrel.
Information on IRETA’s buprenorphine project is available here
Information on IRETA’s benzodiazepines in MAT project is available here
Below, Dr. Peter Cohen discusses the process for generating treatment guidelines from existing research at our 2012 Management of Benzodiazepines in MAT conference
And look for us presenting both projects (with our outstanding partners, CCBH and CBH) at the AATOD conference in Philadelphia this November.
Free online guide to the RAM: RAND/UCLA Appropriateness Method Users Manual
Concise presentation on why we will always need methodologies like the RAM: Consensus-Based Documents are Reality when EBM doesn’t exist
Excellent primer on research methodology and translation into practice: Addiction Treatment Forum EBAM for Practitioners
Most recent report from NIDA on challenges connecting research and practice: Adoption of NIDA’s Evidence-Based Treatment in Real World Settings
Jessica Williams is a master of public health candidate at the University of Pittsburgh and a graduate of the University of Oregon with a Bachelor of Arts in history. She has served as a project manager for IRETA since February 2010. At IRETA, Ms. Williams oversees communications, including research, policy, and educational content on the IRETA website, social media, blog, and e-newsletters. She also writes and edits articles and grants.