Dr. Jeremy Sharp (00:00)
What’s going on folks? Hey, welcome back to the testing psychologist podcast. Today I am talking about something that may be the bane of your existence. And the thing that I’m talking about is medical necessity. So you’ve been there, I think you spend hours with a complex patient and you carefully select your battery to figure out exactly what’s going on. You hit submit.
this prior authorization and then it comes back with those magic words denied. And often you don’t get a great reason. It’s usually just a single sentence that says something like does not meet interqual criteria or maybe you don’t even get that right. And at least in my case, this feels like you’re kind of shouting into the void a little bit. There’s just this
invisible judge that’s sitting in a dark room with a red stamp. And you have no idea what they’re using to determine medical necessity, what the rules of the game are. And so today I am going to hopefully turn the lights on a little bit and demystify interqual. So interqual is the most widely used clinical decision support tool in the world.
We’re going to talk about where it came from, the checkbox logic that it uses to determine medical necessity and judge your work. And most importantly, we’re going to talk about how to speak their language, quote unquote, to actually get your clients the care that they may need. you’ll take away a lot from this episode. mean, if you take insurance, obviously this is going to be pretty helpful. If you do not,
still good knowledge to have and you may learn a thing or two just about the history of our field and why we have to wrestle with this in the first place. So without further ado, let’s jump to a conversation about interqual and medical necessity.
Dr. Jeremy Sharp (01:52)
All right, everybody, we are back and we are talking about interqual and medical necessity. OK, so where did this come from? So like a lot of the content from the podcast, this is born from a personal story. So we are in network with a couple of insurance panels. And without naming names, I will say that one of our panels that makes up a pretty big majority of at least the insurance clients that we have.
recently implemented a prior authorization process. Now we are pretty lucky with the panels that we’re in network with such that we have not had to really deal with prior authorization for the vast majority of time that we’ve been in practice, which has been great. But over the last few months, one of our major panels or payers did implement this prior authorization process. And after
I’m not exaggerating here after submitting probably between 75 and 100 prior auth requests and having literally none of them approved like zero zero were approved. I got really curious about what’s happening here and decided to dig in a little bit and figure out where this criteria is coming from, how they were making these decisions and hopefully how to combat.
these decisions because, you know, I’m not going to sit here and say that like every single person that we test needs every single test that we administer. And there have likely been times when we have maybe over tested or tested in some cases where it wasn’t quote unquote medically necessary, but just playing the odds to have, you know, nearly a hundred prior authorization requests denied across a huge variety of
age ranges, presenting concerns, complexity of the case, etc. I got real interested in what’s going on here. And that’s what led me down this path. just to define some terms a little bit, we’re going to be talking primarily about interqual. So interqual is the decision making matrix, let’s say, that drives the vast majority of
decisions around medical necessity across payers, at least in the United States. So chances are if you are dealing with prior authorization requests, the person who is reviewing your request is using the interqual criteria to determine if services are indeed medically necessary. So a little bit of history here. I love history, and I want to know where this came from. So this didn’t just appear out of thin air.
Of course, Interqual was born back in the 70s. So back then, healthcare was a little bit like the Wild West. Okay. If you went to a hospital in New York, for example, for a procedure, you might stay for 10 days following the procedure while somebody in California might go in for the exact same procedure and stay for three days. All right. So there’s very little consistency. And this is where Interqual started to be developed.
It was developed to be the quote unquote measurable indicator to help define and advance the disciplines of utilization and care management. I’m you’ve heard that term utilization management. That’s the term for folks who decide how much of an insurance or payer services get to be utilized. So the goal was to standardize care so that medical necessity wasn’t just some kind of like subjective
vibe. It was based on theoretically evidence based decision support. All right. So fast forward to today almost well I guess it’s 50 years later almost exactly 50 years later and Interqual is a truly massive operation that is housed under the Optum and United Health Care Group ecosystem. When I say massive I mean
We’re talking like over 4,000 licenses to different entities out there. So that’s like 4,300 hospitals and more than 300 payers or insurance panels. And government agencies, in fact, are actually using Interqual to decide what gets paid for and what doesn’t. Now, there are competitors out there, but…
Interqual is certainly the main utilization management decision-making protocol. And so that’s why I’m tackling it. And it’s also the one that is the underpinning of our particular issue. The payer that we are dealing with uses Interqual criteria. So when we talk about pre-authorization, Interqual is basically like the engine under the hood. It’s the thing that is setting the rules.
that determines if your neuropsych testing is medically necessary or just kind of nice to have in the eyes of the insurance panel. So what does this actually mean? This was actually really illuminating and interesting to me. So the interqual criteria are available out there. You have to dig a little bit. You can’t typically get there just by Googling.
You know, so I had to go into a validity, which is kind of a payer portal. I’m guessing many of you have heard of it. I had to go into Availity and dig around in there and was able to find it through some of those resources available in Availity. But it was really interesting to look through the criteria and see what it actually looks like. So here’s what happens when you hit submit and send in prior authorization. So this is the part that
to me honestly felt like kind of a black box up until this point. But I’m trying to illuminate that for you a little bit. OK. So you hit submit and someone on the team utilization management opens your case. They unfortunately are not just sitting there reading your beautifully crafted pre-authorization request with this nuanced clinical narrative.
that you put so much time and energy into. What they are doing is they’re opening up an interactive software that’s designed to guide them through a structured Q &A kind of format. It’s essentially like a big decision tree that is driven by yes and no check boxes. Okay. With a bunch of different questions. So if you imagine like a procedures Q &A on a screen,
The reviewer is essentially, you know, they start at the top with relatively general questions. OK. And they just go through the decision tree and click yes or no to answer each question in the series based on the information that is contained in your pre-auth request. So they are definitely looking for specific keywords in your request so that they can click these checkboxes. OK. So some examples of these questions will be.
Like a top level question might be, has a clinical interview already been performed? Okay, they check yes. Then it leads them to maybe something like, has a case specific question been formulated? Yes. Then it leads to another question. Does the provider document what action will be taken or how the treatment plan will be affected by the test results? And then if, let’s just say they click no, then it stops, period. That’s it.
So if they click yes, then it goes to more questions. And I ran through a bunch of like common scenarios, I would say, for our practice in terms of referral questions and what we’re trying to get approved for pre-auth. And generally, you know, we’re able to go down like five, six, seven, eight levels in the questions. But…
It’s pretty restrictive. I’ll be honest. It is pretty restrictive, especially if we are like outside the. What would I call it like pure neuropsych realm? So once you kind of stray outside of, you know, documented TBI, stroke, epilepsy, tumor, you know, medical kind of brain based events, it gets real narrow in terms of what types of referrals are approved or can make it through this.
decision tree. like ADHD? No. Autism? Maybe. Based on certain criteria and conditions that might be met. So for those of you who are just doing like plain quote unquote ADHD or uncomplicated ADHD or even autism evaluations, this it’s going to be really really tough to get through this decision tree. So.
Lots of different questions and again, a decision tree. So the thing that you’ve got to keep in mind is again, they are not reading your case formulation in depth. They’re kind of like scanning and looking for keywords. And this is, know, these reviewers are essentially acting as like a first level screening tool. They can say yes all day long, you know, if the documentation is there, right? If you hit the benchmarks, but
The instant that they say no, the process stops and then it’s just rejected. OK. So if they can’t check enough boxes to meet that like medical necessity, medical necessity criteria met status, then the case has to be referred to maybe a medical director or a physician reviewer for a second level determination. And that’s where we get into things like peer to peer review, which in our case never go.
in our favor. So the thing to take away here is that it’s generally pretty tough for most of our general practices to meet medical necessity criteria. So you might be asking, gosh, that sounds really, really strict. Where did this come from? Where’s the evidence for this? And so
I want to take a little side trip just to talk about the approach that was used to even develop these criteria in the first place. Okay. So there is some science behind it theoretically. So Interqual identifies as an evidence-based medicine tool. Okay. So they have a clinical development team of like 55 specialists, know, physicians, RNs, and something called a medical librarian. don’t know what that is, but they follow us like a systematic review of
literature and, you know, determine what’s out there and what’s best practice. They use something called the GRADE approach. Some of you may have heard of this if you’re in the research realm. So it stands for grading of recommendations, assessment, development, and evaluation. So they rate the quality of evidence from high, where the research is very unlikely to change, confidence in the estimate of effect to very low, where the estimate is very uncertain.
So when they built the criteria for neuropsych testing, they didn’t just guess, thankfully. They did look at, you know, tens of thousands of citations. They monitor peer-reviewed journals and use an automated kind of surveillance system to scan, you know, thousands of sites for new literature. They look for systematic reviews, meta-analyses, randomized controlled trials, all the things that seemingly mark good research. But for the neuropsych criteria specifically, they focus on brain-based
pathologies like I said so things like TBI stroke tumors CNS infections that kind of thing and the research selected kind of focuses on whether the testing will actually change patient management and improve outcomes which is what they call clinical utility. Now on the surface I mean this sounds like a very reasonable question right. Like it’s kind of asking that question of is testing quote unquote an evidence based practice and
I’m fine with that question, but when you dig into the citations that are provided for the neuropsych interqual criteria specifically, it’s honestly pretty tough. I was not overwhelmingly convinced, at least in the work that I did to look through the citations and dig into the research, I was not convinced of the relevance, honestly, to testing.
There was a lot on telehealth, which is interesting. The efficacy of telehealth. There were some studies that looked at some of these brain-based pathologies and treatment modalities primarily, but there were very little, if any, studies that specifically addressed, is neuropsych testing helpful? It was more proving the case that like,
standard of care is to just jump to kind of a medical treatment or intervention and then see what happens versus like utilizing testing to make a treatment plan ahead of time. So theoretically it’s evidence-based but I don’t know I was was iffy on that when I reviewed it. Of course I am NOT an expert and I’m sure there is a fantastic rationale I just can’t figure it out. So through all of this
there were some benchmarks that were developed. So let’s talk about hours and approvals if you get anything approved. Many of you have probably run into this benchmark of eight hours for testing. If you get a pre-auth approved, it’s typically for eight hours. So in the integral world, a full evaluation is typically approved for up to eight hours total. And that is everything included. It’s test administration.
It is face-to-face time, it’s scoring, data integration, and report writing. So you, like myself, may ask, where did that come from? Well, Interqual leans heavily on this concept of testing fatigue, which is the idea that excessive testing hours can lead to invalid results because the patient simply loses cognitive stamina in the process. Question, of course, is like, is this born out of research or is it
just a guess essentially. And as best I can tell, it’s a little bit of both. The research used to support these benchmarks did include some cross-sectional studies that looked at quote unquote surrogate outcomes like diagnostic accuracy versus clinical workup. Fatigue is a recognized clinical phenomenon in neurology, but there are some critics, specifically the APA.
that argued that applying a rigid eight hour cap is kind of a cookie cutter approach that fails to account for complex comorbidities or clients with severe processing speed deficits who need more time to produce a valid profile. And I think this is where we really get into the problem with something like Interqual. It is a very cookie cutter approach. There’s very little in the decision tree as far as questions to ask that address
clinical complexity in any amount of depth. and it certainly doesn’t take into account something like, social determinants of health, for example. Right. So this is where I think the system really starts to break down and where I have trouble with it being implemented so widely. So the standardized criteria are great for like typical patients, right? But, patients are not algorithms and this social determinants of health or SDO age,
It’s kind of a massive gap in the interqual criteria. So it focuses primarily, like I said, on medical indications like TBI or dementia or tumors and things like that. It very often neglects any environmental or social factors that might be influencing someone’s functioning, like housing instability or transportation barriers that might change how or where a client receives care.
And I mean, essentially, just means that the criteria might inadvertently penalize the very populations that they’re intended to serve by kind of ignoring the fairness issue. And there’s plenty to say on the just clinical complexity front as well. I mean, we’ve all had cases where, yes, the referral question is ADHD, but there is also trauma. There is a
you know, anxiety, there’s depression, there’s maybe a question of autism, there’s a learning disorder question, there are behavioral concerns, a parent died recently. I mean, there’s all kinds of issues that can influence the case and, you know, make an argument for complexity that just do not get addressed in these interqual criteria. And at least in our case, the line that we get fed
with the denials is a psychiatrist can make this diagnosis during a comprehensive interview. And, you know, I don’t know about y’all, but that’s really, really challenging and that’s really hard to swallow for me personally. So there’s another paradox in this whole thing. Of course, I looked into some peer-reviewed validation of the criteria and it’s actually very sparse.
So in some patient cohorts, research has found that the criteria lack the sensitivity required for precise clinical categorization. Hey, surprise. And if you take it a step further, when you go down the path and look at folks who actually appealed some of these decisions, when these cases go to appeal, the criteria themselves, like Interqual, rarely wins the appeal. Success in the appeal usually relies on
the quote unquote totality of the record, which is high quality clinical documentation that proves why that case needs more than what a checkbox system can allow. OK. So all that said, feel like up to this point, I’m essentially complaining and just listing all the reasons. It’s not great. But I’m biased, of course. So let’s talk about how to play the game a little bit.
So if you are navigating this pre-authorization maze, there are some ways that you can maximize your chance of first level approval. But you have to kind of feed that reviewer some keywords that they need to put into the software. OK, so what does this mean? Jumping straight to answering like second order questions is really important. So first order questions that I would say are like symptoms, essentially. You don’t want to just list the symptoms.
You want to answer some of these second order questions like how will these specific results change the treatment plan? Okay. And an example there is, you know, to use language that is like very direct results will directly influence the formulation of a targeted neurological rehabilitation plan versus home-based care. Now that’s like an adult example, of course, but that’s one way that you can do it. You answer the second order questions very, very specifically.
Second thing is you can document that you have conducted what they will call like a gatekeeper exam. And this is like the neurobehavioral status exam or a 96116, you know, if you’re thinking in terms of CPT codes. So Interqual often uses this exam to guide the selection of a focus battery and make sure that you aren’t performing redundant tests. Okay.
You can also use some strategic keywords like I mentioned. instead of saying that something is a complex case, that’s not a keyword at all. You could say something like, diagnostic uncertainty remains following a comprehensive clinical interview. Say that again, diagnostic uncertainty remains even following a comprehensive clinical interview. Another example of specific language is, you know, instead of saying quote unquote needs more time,
You could use, you know, I’m requesting X additional units due to documented complicating factors such as and then list them language barriers, severe cognitive processing deficits, et cetera.
And then the last thing, one of the things that you want to explicitly state if it’s relevant is addressing the duplication issue. So if testing has not occurred in the last 12 months, make sure and say that. If testing has occurred in the last 12 months, you want to document that as well.
They do have to, you know, typically check the box for, know, whether previous testing has been conducted or not. So these are just a few ideas. You know, there’s a lot to say about this. I would encourage you to do your own your own research, of course, and try to determine how you can best word your pre-auths to to meet some of these guidelines. But at the end of the day,
Interqual, I think, is a tool for standardization, but it is not a replacement for your clinical judgment. It’s a hurdle, but if you know the height and the distance, hopefully you can clear that hurdle. All right. Your goal is to fill those clinical gaps with documentation so robust that even if the software says no, that it gets kicked up to a peer review and that human reviewer who actually looks at it says, yes, it’s tricky. It is tricky. Like I said, it is.
You know, the criteria are quite strict, especially for, you know, non-complex evaluations. And so a big part of this is just great clinical documentation. You absolutely want to make sure that you document if a previous treatment has been tried and failed. It certainly helps to have your client, you know, have already seen a psychiatric provider, preferably a psychiatrist.
if they tried medication and it didn’t work, you want to document that. there are lots of components here. I think we’re all working on this together. And I know that we’re all trying to jump through this hoop of prior authorization. So the hope is that this has been helpful. There will be some resources in the show notes, of course. But I think the most helpful thing that you could do is probably go and dig up those interqual criteria for neuropsych testing just to see what we are.
working with. So with that, will leave you keep fighting for your patience and for your practice and I’ll see you next time.
Click here to listen to the podcast instead.