Macro Medicine

1 Comment

When the Office of Inspector General (OIG) released its 2011 Workplan – a formal announcement of the areas it plans to investigate – it boldly brought physicians under increased scrutiny for fraud, based solely on what may be an unintended consequence of the evolution of documentation technology.

From Embarrassing Mistakes to OIG Fraud Investigations

“To err is human, but to really screw up takes a computer.”

When the Office of Inspector General (OIG) released its 2011 Workplan – a formal announcement of the areas it plans to investigate – it boldly brought physicians under increased scrutiny for fraud, based solely on what may be an unintended consequence of the evolution of documentation technology. Specifically, in the section titled, “Payments for Evaluation and Management Services,” the OIG wrote that “Medicare contractors have noted an increased frequency of medical records with identical documentation across services. We will also review multiple E&M services for the same providers and beneficiaries to identify electronic health records (EHR) documentation practices associated with potentially improper payments.” The implications of this statement for emergency physicians are broad and should be taken seriously. (download the Workplan at

Let’s take a look at exactly why the OIG has taken an interest in the phenomenon of “identical documentation.” How many ways are there to dictate a normal cardiac exam? After dictating the normal exam a few thousand times, the words flow out reflexively: “Heart: regular rate and rhythm without murmurs or gallops.” How about the HPI of a patient with a sprained ankle? Even histories become pretty rote: “Pt complains of right ankle pain after inversion injury.” Documentation systems looking for efficiencies have taken advantage of dictated or menu driven “macros.”  So, when the physician instructs, “use my normal exam macro,” the transcriptionist pops up the pre-formatted text, saving both the doc and the transcriptionist time and effort. Templated systems take advantage of the repetitive nature of much of what we do, reducing much of the required documentation to a series of check boxes.

Now we have electronic medical records, which combine both the best and worst of rote documentation, macros, and check boxes. Some of these EMRs require physicians to click on boxes or menus; the EMR then spits out a chart of pieced-together pre-formatted text (like the dictation “macros”). Although these are referred to as computer-generated narrative charts, the narrative is all too often obscured by a plethora of time stamps, user ID initials and the awkward juxtaposition of preformatted sentences or phrases. As a result, the end product is a chart which takes even longer for a reader to figure out than for the doc to create.

Macros, whether dictated or from an EMR, may be quickly, inadvertently, and inappropriately invoked. The result can be more than just embarrassing. For example, a female patient who sustained some minor injuries from a fall obtained a copy of her ED medical record and wrote a letter complaining about multiple discrepancies between the documentation and what actually happened in the ED.  The chart read as follows:

“KNEE EXAMINATION: The patient has full range of motion of the knee. There is no local swelling or tenderness. There is no instability with the medial or collateral lateral ligament stress. There is normal Lachman test with no instability. McMurray’s test is done with no locking or clicking & no significant pain. There are no effusions.”

In response to this, the patient wrote, “Not done at all. Had one been done, it would have been evident that I had multiple knee surgeries including a total knee replacement from the scars on my right knee.” The chart went on to say,

“There is normal external female genitalia & normal vaginal introitus & mucosa. There is no cervical motion tenderness or lesions. The uterus & adnexa have no masses or tenderness. There is no purulent vaginal discharge.”

The patient responded with, “Not done at all, nor would I expect it due to a fall landing on my shoulder. Also, I had a total hysterectomy in 1982. Therefore, I do not have a cervix, uterus, or ovaries. I am understandably upset about the inaccurate information concerning my physical exam and would not like to see my insurance company billed for things that were not done. Nor, do I want inaccurate information in my health record.”

A quick check with the doc confirmed a bad case of “documentation with brain-disengaged” syndrome. His correcting addendum included the following: “My normal template components were transcribed.” Fortunately for the doc, the patient directed her concerns to the EP group, rather than bringing what appeared to be a case of fraud to the attention of her insurer or federal authorities.

The bottom line is that when we use macros, unavoidable uniformity from one record to the next can raise the suspicion that the physician isn’t really doing the evaluation that has been documented. If the history and physical exam are all the same in the last twenty sprained ankles (or chest pain patients, etc…), the people who have to pay for the work you do may get the bright idea that you’re bending the truth. They may think your documentation is just as fudged as the documentation of the complete GU exam in the patient who’d had a total hysterectomy.

It would appear that the OIG wants to find out if physicians are merely duplicating documentation from one patient to the next, billing for services which may not actually have been provided. In other words, how often is “identical documentation” the result of “my normal template components were transcribed”? It should come as no surprise to anyone that with the scramble to adopt EMRs to obtain HITECH incentive monies, more and more ED records are generated by EMR systems and more and more of them appear to have “identical documentation across services.”

The irony is that with the right hand, the federal government is urging the rapid adoption of EMRs, and with the left, the OIG and other fraud-fighting agencies may be preparing to pounce on a natural by-product of those records as fraud. Does the right hand know what the left is doing? Maybe not, but just recall that under the misnamed Patient Protection and Affordable Care Act, an additional $350M (that’s on top of the annual budgetary allocation for fraud enforcement) was allocated to identify and prosecute fraud, and it’s a safe bet that the intent is to get a whole lot more back in recoveries and penalties than the newly allocated $350M.

Note that the OIG investigation is just that, an investigation, aimed at determining if the “increased frequency of medical records with identical documentation” is a fraud problem, or just an unintended and innocent consequence of the evolution of medical records technology. Until the OIG announces the results and conclusions of its investigation, we won’t really know how much liability exposure you may have from documentation systems which create records which utilize pre-formatted verbiage.

So, what should you do to minimize your risk of fraud liability?

First, and most obviously, document what you do and don’t document what you don’t do. That means using macros only when they accurately reflect the evaluation that was actually done. Patients are more frequently taking advantage of their right to obtain copies of their medical records, and you should expect that if your documentation doesn’t reflect what was actually done, sooner or later, you’re going to hear about it from a patient. That is, if you’re lucky. If you’re not so lucky, you’ll be hearing about it from a Medicare auditor, or perhaps a federal prosecutor.

Second, whenever your documenta
tion allows it, make sure there is some portion of your documentation of the patient encounter which is individualized for that patient, so that any subsequent reader (or auditor) would realize that the documentation (a) is “nonidentical” and (b) could only have been created by someone who actually evaluated the patient.

Third, if your hospital is planning to implement a new EMR, try to get involved in the planning process so that some kind of customization capability (ideally add-on dictation, typically electronic voice recognition) will be included in the final package. That way, you’ll at least have a system that allows quick and easy generation of customized and individualized reports, even if only for a brief portion of your medical record documentation.

Sensible documentation accurately reflecting the care provided is essential for compliance. Taking the short cut, in the form of macros, may seem to be a viable option to improve efficiency, but not at the expense of inaccuracies, fraudulent documentation and a painful investigation from the OIG.

Dr. Frank is a physician and attorney who serves as General Counsel, Director of Risk Management, and Compliance Officer for Emergency Medicine Physicians, based in Canton, Ohio.


1 Comment

  1. Errors increase because of forced choice computer screens. the computer forces an option to be checked, and you get a complete set of unreliable answers.

Leave A Reply