Sunday, March 4, 2012

"The land of 10,000 90862s"

The title of this post is an inside joke for psychiatrists.  90862 is the billing code associated with a psychiatric visit that is commonly referred to as the "med check."  It is an example of what can happen to a profession when government bureaucrats and businesses run amok in determining what they think you do or what you should be doing when you provide patient care.
I first became aware of the political importance of this system in the 1990s, when I had to attend mandatory billing and coding seminars at my place of employment.   In those seminars I learned that the politicians and insurance companies were so desperate to use this arbitrary system that they told us we could go to federal prison for a long time if we submitted a "fraudulent" billing document.  The "fraudulent" document they were talking about was any bill connected to the document of a patient encounter that did not have enough bullet points to qualify for that level of billing.
That is an important concept so let me say it another way.  After every patient encounter, the physician needs to document a note about what happened and indicate a level of billing for that encounter.  When I first started training the note could be as little as one or two lines.  For example, at one point in my training I covered an entire surgical service with a team of doctors.  We could round on 25-30 patients with very complicated problems and write all of the documentation in about 2 hours.  The documentation was "Pain is well controlled, surgical site looks good, vital signs are stable."  We did not have to bother with any billing documents because a hospital billing specialist came by and confirmed that we had seen the patient and submitted the bill.
Somewhere  in the 1990s, a government initiative changed all of that.  The government decided that they needed a way to control the global budget for physician salaries and they decided to develop a system of codes for patient encounters that they assign relative values to and then multiply that by a certain number to set reimbursement for that code.  The entire system rests on the assumption that somebody can look at the description of a patient encounter as written in a note and audit the associated billing document.  It turns out that when this assumption was tested several years later - it was determined to be false, but that did not deter the federal government or the health insurance industry (see reference).
The 90862 is probably the most abused billing code in the psychiatric profession.  The interpretation of what constitutes an encounter that qualifies for this code varies from practice to practice and between organizations.  Patient experience varies from literally talking to a psychiatrist for 5 minutes with the goal of getting a prescription refill to a much richer encounter that includes a discussion of other current problems, additional medical diagnostic discussions and psychological advice.  In some cases, acute medical problems requiring emergency care have been identified in these sessions.  There is no doubt that a considerable amount of gaming occurs on the part of some clinicians and most insurance companies and government payers.
The only gaming possible by the clinician occurs at two levels.  The first is total time spent with the patient.  The folklore is that these are all 15 minute encounters.  Some clinicians insist on seeing patients in half hour blocks and others see 3 - 4 people per hour.  The second is total documentation.  You can literally do a few lines or you can write several paragraphs and stay after work just to do the documentation.  A lot depends on whether you think you will be audited and somebody will be making an arbitrary decision about whether your note qualifies for the charge that you assign to it.
There are myriad ways that a managed care company can game the system.  First of all, they can assign any level of reimbursement to any billing code that they want.  I quoted a New York Times article in another post as saying that a psychiatrist could see three patients for medications and get reimbursed at $50 per session, but the actual reimbursement can be less than half of that.  That same managed care company can also take any bills submitted for patient encounters with higher reimbursement levels and say: "we are only paying you for a 90862 no matter what you do."  If you happen to be working in an institutional setting, a managed care company can negotiate a per diem rate with your employer and not pay the 90862 billing at all.
Stated another way, a psychiatrist can see a patient with complex medical and psychiatric problems and get reimbursed at a level that might lead to them break even - to getting no reimbursement at all depending on the insurance company and contracting arrangements.  Within organizations the relative values for these codes are the basic way that physicians are manipulated to see more patients.  It is referred to as their "productivity" even though producing work for little or no reimbursement is not really productive activity.  The physician managers can demand that they see more and more patients to compensate for the poor or nonexistent rate of reimbursement by managed care companies.
Another artifact of this system is that procedures like surgeries, endoscopies, and angioplasties are reimbursed at a higher rates than a doctor talking with you and discussing the diagnosis and treatment.  That lead to a movement to reimburse the cognitive or nonproceduralist specialties at higher rates.  But given the amount of government payer and insurance company leverage it is impossible to make that happen.
Is there a solution to this problem that in effect makes physicians work impossibly harder to earn a professional salary?  The solution is as easy as considering how I pay my attorney, accountant, mechanic, plumber, electrician, and chimney sweep.  I pay them all by the hour.  In some cases there is an agreed fixed amount, but it is generally many times more than what I would get reimbursed for the lowest 90862 reimbursement.
Getting back to the title of this post, when I looked at the lowest current reimbursement for a 90862 and calculated how many of those bills would need to be submitted to make a professional wage, it came out to about 10,000 patient encounters per year.  Working 50 weeks per year that would mean seeing 40 established patients per day.  The only clinic where I have ever observed those numbers had three nurses rooming the patients and doing all of the documentation before they were briefly seen by a psychiatrist.
As I contemplated all of this I had the thought: "I am living in the 'Land of 10,000 lakes' - maybe we should just change that to the 'Land of 10,000 90862s.' "

George Dawson, MD
King MS, Lipsky MS, Sharp L. Expert agreement in Current Procedural Terminology evaluation and management coding. Arch Intern Med. 2002 Feb 11;162(3):316-20.

90862 Redux? An Update.


  1. The "15 minute med check" is destroying psychiatry. The psychiatrist as drug dealer. I've learned over the years that you have to go in there with an agenda, knowing what you want, or agree to what the psychiatrist suggests- there often isn't much time for discussion. This can be good for the empowered patient, who can do her own research and figure things out herself- but occasionally I'm too depressed for that. And not everyone is medically literate. I have a very good psychiatrist right now, because I pay out of pocket- but sometimes I wish I could just have a prescription pad, and skip the middle man. The really important questions, such as when to medicate, and when not to, and whether the research supports the use of maintenance meds- there just isn't much time for that.

  2. One reason forensic psychiatry appeals to me is that I do bill by the hour, starting with a basic rate for record review, time spent in the examination, travel, etc, and higher rates for deposition and time on the witness stand.

    But there a problems with this approach in clinical billing: The patient will want to spend less time to save money, perhaps by avoiding bringing up critical information. Compensation for a 5' encounter may be far too little to cover fixed costs, including overhead, such charting (unless we bill by the minute for that too), malpractice insurance, et al. The cost to the patient of an encounter becomes less predictable, too. The patient who pays a fixed fee for 90862 knows how much a visit will cost when they walk in the door so they can budget and pay at the time of service. (You'll never get 3d party payers to pay by time. Fugetaboutit.)

    To avoid charges of fraudulent billing, start by refusing to treat Medicare/Medicaid beneficiaries. Then start dropping contracts with inscos. And don't take money from them. Even if the don't accuse you of fraud they can demand repayment of funds issued "by mistake" for years after paying you. Get your $ from the patient.