From Podcast, ,

DHHS recently unveiled the first major overhaul of its documentation guidelines for the E&M codes often used in outpatient psychiatry. We show you how to make those rules work for your patients.

CME: Podcast CME Post-Tests are available using this subscription. If you have already enrolled in that program, please log in.

Published On: 12/27/2021

Duration: 24 minutes, 16 seconds

Chris Aiken, MD, and Kellie Newsome, PMHNP, have disclosed no relevant financial or other interests in any commercial companies pertaining to this educational activity.

Note bloat. Alert Fatigue. Clinician burnout. Can new guidelines from the department of health and human services reduce these problems? Sometimes help comes in the most unlikely of places.

Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief of the Carlat Report. And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.

Last year, DHHS overhauled the billing and documentation guidelines. Today, we’ll show you how to make them work for your patients and trace a little history back to the dawn of the medical record. But first, a preview of the CME test you can take to earn credits for this episode.

The physician who developed the first medical record in America is better known for taking care of which founding father on his deathbed?

A. Benjamin Franklin

B. Benjamin Rush

C. George Washington

D. Alexander Hamilton

KELLIE NEWSOME: 25 years ago the department of health and human services (DHHS) released the evaluation and management or E&M code: 99212, 99213, 99214, and 99215, each representing an increased level of complexity and reimbursement. The idea was to reward clinicians for doing more complex work, but like a lot of good ideas it backfired. The E&M scorecard calculated complexity in large part by how many signs and symptoms you assessed and how many words you wrote in the EMR. In response, clinicians started packing their notes with cut and pasted text – spilling over with endless symptom checklists that had nothing to do with the patient’s problem. 

CHRIS AIKEN: All of the sudden my dermatologist started checking my blood pressure, “It’s part of our insurance requirements,” her nurse told me.

KELLIE NEWSOME: Soon it was the medical record that had a disease, and the disease had a name: Note bloat. Clinicians were no longer able to find the information they needed to help the patient in these rambling notes. Many lost trust in the accuracy of what they were reading. Errors spread through EMRs like a virus as they were cut and pasted from one specialty to the next. And to generate this tower of Babel the clinicians turned their back to the patient, their faces fixed to the blue light of the computer monitor.

Many of you lived through this, and some of you may have noticed a disturbing trend. As EMRs spread in the past decade – bringing with them the technology to supersize the medical note – so did the number of talks and articles on clinician burnout. And that is no coincidence. Research from the Mayo clinic has directly linked burnout to these ungainly EMRs. Here’s a quote from their 2020 study:

“The usability of current EHR systems received a grade of F by physician users… and a strong dose-response relationship between EHR usability and the odds of burnout was observed.”

CHRIS AIKEN: For a while, psychiatry was relatively immune to all this, as the code we used for medication management – 90862 – did not have many documentation requirements. But in 2013 DHHS put 90862 out to pasture, replacing it with the 9921X E&M codes that required us to document a lengthier symptom checklist, and then turned us into bean-counters as we tried to tally our notes with the unwieldy bullet system that DHHS devised. 

KELLIE NEWSOME: So when DHHS rolled out the first major overhaul of the E&M documentation requirements last year, we weren’t too excited. And when we read the Byzantine guidelines on how to pick a billing code we wanted to crawl back into our psychopharm textbooks and keep our Carlat reporting to more clinical topics. But we found two points of light in that labyrinth to inspire you.

1. The new guidelines reward clinicians for what they do, not what they write
2. They reward providers for taking on complex patients, rather than writing complex notes.

CHRIS AIKEN: Like the old CPT guidelines, the new ones have a point system, but instead of rewarding points for each symptom you document they are rewarded based on the complex and time-consuming tasks you do to help patients. Do you ever call the family to get their perspective on a patient with mania or psychosis? That counts. Do you ever coordinate with the primary care doctor? Add a point. Do you ever draw labs, read hospital records, review neuropsychological testing, or go out of your way to help patients with major psychosocial barriers to health – like they can’t afford their medications? You’re on your way to a high score.

KELLIE NEWSOME: I like that analogy. Let’s gamify this – make it fun.

CHRIS AIKEN: The guidelines have 3 categories, each with 4 levels of complexity. The higher the level, the greater your score (ie, the higher your reimbursement). But if you fail in one category and score high in the other two, you still win, because your final score is based on the highest level achieved in 2 out of 3 categories. Those categories are:

1. The number and complexity of problems (or diagnoses)
2. The number of extra steps you took in gathering information – that’s where all interviews with family members, coordination with PCPs, and review of hospital records and labs comes in.
3. The overall risk of the treatments and the patient’s condition. Are they suicidal? Do they need extra support to keep them out of the hospital? Are you monitoring a high-risk medication like lithium or clozapine? Any of those will move you up to gold-start status in this category.

KELLIE NEWSOME: 3 categories, each with 4 levels, and each level has 2-4 bullets that let you earn points. It’s a pretty complicated board game. And – like most DHHS initiatives – it wasn’t written with psychiatry in mind – so we’ve translated it for psychiatric use in the paper journal. For example, in DHHS-speak there’s a difference between a lab and a test. A lab is a simple blood draw, but a test is more complex, involving more specialized skills to interpret, and usually has its own billing code to run it, like a sleep study, neuropsych testing, an MRI or EEG of the brain. Another important distinction is in the word “stable.” We use that all the time in psychiatry to describe patients who are still functionally impaired by chronic, unchanging symptoms, but who aren’t in an active episode of – say – psychosis, depression, or mania.  Well you should stop doing that, because in DHHS-speak “stable” means full remission. And this matters, because the new system is going to pay at a higher rate if your patient has active symptoms – even if they are chronic and unchanging (what we used to call “stable”) – than if they are in full remission.

CHRIS AIKEN: In the end, what it all boils down to is whether you are going to bill 99212, 99213, 99214, or the high-scoring 99215. But – to gamify it a bit more – with high stakes come high risks, and if you bill too much 99215 you are setting yourself up for an audit. Insurers compare your billing patterns to those of your psychiatric colleagues, and if you are an outlier they may decide to audit you. But don’t let that scare you. If you run a clozapine clinic, you probably should bill more 99215s. Years ago I attended a workshop on the old CPT codes aimed at general medical providers, and they suggested that billing more than 10% as 99215 could trigger an audit, but those were different times. In the old system, it was almost impossible for a psychiatrist to bill 99215 because you’d have to do some cognitive tests and check blood pressure and temperature, along with dozens of other symptoms, AND justify the medical necessity of doing so. Perhaps delirium could qualify. But now a psychiatrist can reach 99215 by managing a severe rash on lamotrigine or working out a safety plan with a suicidal patient to avoid hospitalization.

KELLIE NEWSOME: And that is what is most inspiring about all this. I don’t want to offend any listeners, but it’s hard to be politic about this. Doctors – including psychiatrists – have a long history of avoiding seriously ill patients. Some refuse to accept hospital discharges, others don’t treat schizophrenia. A scathing article from 2014 in JAMA Psych found that only about half of US psychiatrists insurance – compared to about 10 percent in other specialties – and treating private pay patients is a sure way to limit your practice to the higher functioning set.

CHRIS AIKEN: When I was in training one attending shared a story about his private practice days that left a bad taste in my mouth. He said he screened out low functioning patients by making his office difficult to find. Later I read a book on the business of psychiatric practice that detailed the various ways that “comfort seeking, middle-aged psychiatrists seek to avoid treating patients with borderline personality disorder.” Now, in fairness, sometimes all these maneuverings are done out of a sincere belief that complex patients are better managed in a group practice – perhaps with a DBT program or an ACT team – but we can’t ignore the financial incentives that have long been in place for doctors to take care of the worried well.

KELLIE NEWSOME: And thanks to DHHS we can now say those days are over! People with overdoses, hospitalizations, suicide attempts, and frequent psychosis should no longer have to search far and wide to find a clinician to take care of them, because reimbursements are going to be much higher for those patients than they are for primary insomnia or generalized anxiety disorder.

CHRIS AIKEN: And you no longer have to write 5 page notes to get that reimbursement. The new system is very simple. At its core, all you need to do is document a problem list. The more problems you are treating, and the more severe, chronic, or complex they are, the higher the points. 

KELLIE NEWSOME: The rest of the note is up to you. It’s still expected that you’ll document according to professional standards in your field, but DHHS no longer wants to be in the business of dictating what those standards are for all professionals. And that makes sense. The old system – which required the same level of documentation for a psychiatrist as an ophthalmologist – wasn’t very realistic.

CHRIS AIKEN: Again, look online to personalize this to your practice, but we’ll end with a simple way that will work for most psychiatric clinicians. Remember, you only need a high score in 2 out of 3 categories. In one of those – the risk category – simply prescribing a medication brings you to a level 4 out of 5 – 99214. You don’t need to change the med, and the patient doesn’t need to accept your recommendation – so if you recommend an antipsychotic and the patient declines you’ll still get there. OK, you’ve got 1 out of 3, so all you need is one more.

The next category is data, but honestly we usually skip over that one because unless you’re ordering labs, reading hospital records, talking with the family, and consulting with the PCP, you’re not going to get a high score with psychiatric data. And, frankly, if you have to do all that stuff, you’re probably dealing with a complex, risky case and have already reached a high score in the other categories. 

So onto the third category – which is the problem list. If the patient has at least one chronic disorder with active symptoms, or at least 2 chronic disorders in full remission, you’ve made it to level 4 here as well. “Chronic” is based on the disorder, not necessarily the patient. So a patient who is in their first episode of mania has a chronic disorder – because bipolar is a chronic disorder. The bottom line is that most psychiatric disorders are considered chronic including ADHD. We’ve listed a few in the online issue that are not. So if your patient has Bipolar disorder and panic disorder, you’ve made it to level 4 in this category, even if both are in remission.

KELLIE NEWSOME: Bipolar disorder actually has a higher rate of comorbidities than other conditions, including panic, PTSD, ADHD, addictions, eating disorders, personality disorders, and OCD. We’ve featured a few articles recently on managing these comorbidities – like a 2 part series last fall on bipolar and ADHD, and in this month’s report there’s a new study on treating bipolar with OCD.

CHRIS AIKEN: These new guidelines are part of a long history of trying to find some meaningful use in medical records – that dates back to their origins in the early 1800’s. That’s when the US physicians first began keeping notes on their patients. The movement was driven by doctors, and the goal was not to fend off malpractice suits or bill insurers. The first medical notes were developed for research and teaching – doctors figured if they gathered all their patient histories together they might see trends that could guide future medical innovations. These notes were shared with students and often kept in the lobby for relatives of patients to see – privacy was not a concern. And the physician who started it all also has a cameo role in a popular Broadway play.

KELLIE NEWSOME: On August 17, 1804, a New York physician sat down to document the case of a 49 year old man who had died a month earlier from a gunshot wound. The patient, Alexander Hamilton, was the creator of the US financial system and his doctor, 35 year old David Hosack, was about to create the country’s first medical record.

Dr. Hosack believed that if physicians recorded the details of their cases they could amass a collection of medical data that would be useful to students and researchers. He proposed this to the board of New York Hospital in 1805, and so began the medical record. It wasn’t until the 1890’s that malpractice attorneys started asking for medical records, giving doctors yet another reason to document their work. 

By the 1960’s the medical record had become bloated with labs, tests, order, and consultation notes, and doctors were having a hard time finding the information they needed in the unwieldy 3-ring binders. That’s when Larry Reed developed the problem oriented note – a simple solution that gained widespread attention after he published it in a 1968 issue of the New England Journal of Medicine. 

Dr. Reed called on physicians to organize all their treatments and data around problems, so that each note ended with a numerical list of symptoms or disorders they were attending to in their patient’s care. He died in 2017 at the age of 93, but he would have been pleased to see DHHS return to his problem oriented system. Reed was no bean-counter. He was a creative and dynamic physician who drew inspiration from the far reaches of art and philosophy as he put together this new method of medical documentation. Here is Reed speaking at medical grand rounds at Emory University in 1971:

And now for the word of the day…. Alert fatigue

KELLIE NEWSOME: Alert fatigue describes how clinicians become desensitized to safety alerts. The most common example is drug interaction alerts in electronic medical records. Studies have found that clinicians override nearly all of these – including the ones labeled “critical.” The longer they use the system, the more likely they are to ignore them, and the fact that most of the alerts put out by these systems are inconsequential only compounds the issue.

We have a solution. Drs. Aiken, Feder, and Carlat spent the last 3 years researching drug interactions to boil it down to the ones you really need to pay attention to. It’s all in their new book, Prescribing Psychotropics: From Drug Interactions to Genetics. The book doesn’t just tell you if an interaction exists – it tells you what to do about it. Here’s an example, let’s say you’re adding an antipsychotic to carbamazepine for bipolar mania. Your drug interaction software says that carbamazepine may lower that antipsychotic – but by how much? Turn to page X, and you’ll see what to do. If it’s risperidone, you’ll need to raise the antipsychotic 2-3 fold to adjust for the carbamazepine interaction. If it’s aripiprazole, you should raise it even more, 3-5 fold. And if it’s quetiapine, you probably shouldn’t use it at all because that interaction is going to create an active metabolite – norquetiapine – that has antidepressant qualities and can worsen mania. Since I’ve started using it I look forward to those interaction alerts – I know when to ignore them, and when they can help me fine-tune the dose to better suit my patient.

Start earning CME credits for this podcast through the link in the show notes. You’ll help us stay free from pharmaceutical support, so we can keep bringing you unbiased information you can trust.

Got feedback? Take the podcast survey.