Reducing spinal fusions in the real world
Back in 2015, a team of researcher clinicians at Virginia Mason Medical
Center in Seattle set out to find out if an algorithm and multidisciplinary teams would move the needle on unnecessary spinal fusions.
Dr. Rajiv Sethi and other providers there eventually published a study showing that only about a third of patients seeking a second opinion at Virginia Mason actually needed the costly surgery. The conclusion: There was and still exists a big opportunity for providers to take a hard look at their practices. Sethi talked with MH about this ongoing work.
Why did you all embark on this experiment?
In a fee-for-service healthcare system, you’re often enfranchised and paid more to do interventions, the costliest of which are surgeries. I think that’s why you see a general overutilization of spinal fusion surgery in America, as has been reported by a number of different manuscripts. When patients underwent an isolated decision-making pattern with the surgeon only, oftentimes surgery was the first option, when it should, in fact, be the last option.
Can you describe what you did?
The algorithm that’s referenced in the paper uses all the stakeholders to choose and optimize patients for spinal fusion. It allowed the team to ensure that the patient has maximized conservative treatment, and that there was no other treatment that they can go through first.
Was it hard to implement in the beginning?
It was very difficult, because we had to get a lot of people on board with this idea: the surgeons, anesthesiologists, the physical medicine rehabilitation doctors and the C-suite. At Virginia Mason, there was a culture of collaboration and doing what’s right for the patient, but it definitely was challenging.
Are you still using this tool?
We’re using even more sophisticated algorithms, which includes computer decision-support tools. We’re now holding virtual conferences for people from multiple states to provide multidisciplinary virtual opinions. So we’re really using technology a lot more now.
What’s next?
In the last three years, we’ve added the technology piece in how we use virtual decision making and artificial intelligence. That’s where you’re really going to see a lot of speed in the next five years: how do we use softwarebased technology to help stratify risk in patients to provide a much more significant level of informed consent for patients going forward. You bring a patient to the office, and you put their characteristics into a computer algorithm. And you show them inperson what their risks are. It’s modeled to them. We’re also starting to see now that we can potentially dial the dose of surgery to a patient based on their risk.