One of the many things I learned when I became a resident was how seemingly little there was to know about coding and documentation. I was taught simply to code on the basis of time spent with the patient or according to the level of complexity of the visit. The only guideline for complexity that I was aware of went something like this: “Most office visits are a level three. If you do more during a particular encounter, then code a level-four visit. If you do less, code a level two.” As for documentation, I was told simply to “document everything.” Sound familiar?

Although these “guidelines” had the advantage of being easy to remember, I grew increasingly uncomfortable with their application. I felt I was documenting too little to support my coding decisions — or documenting too much to be efficient. I began to wonder: How much documentation was enough?

During my second year of residency, I thought I had found the answer. It was 1994, and the “Documentation Guidelines for Evaluation and Management Services” developed by HCFA and the AMA were released. I read and reread the guidelines. I pored over countless articles on the subject. I even tried to use a pocket guide. In the end, despite my genuine desire to learn, I was more confused than ever.

Convinced there had to be a better way to learn the guidelines and put them in practice, I developed the charts that accompany this article. I've since found them to be an effective teaching tool and a boon to my busy practice. A number of family physicians I've shared the charts with have found them useful, and I'm publishing them in hopes that you will benefit from them too.

I recently updated the charts to reflect the revised documentation guidelines for multisystem exams that take effect July 1. If you've read the recent articles in this journal, you know how complicated the guidelines have become.¹ As you'll see, the charts that I've developed make the guidelines more manageable.

Orientation to the charts

Let's take a look at the charts. You can tear them out along the perforation and refer to them as we go. The charts are designed to be used either to work out the code for a visit from the documentation or to serve as a quick reference against which you can check the code produced by the “seat-of-the-pants coding” that we all probably use most of the time. I use them as a reference as I dictate the documentation, code the visit and fill out the superbill — and it takes me no more time than it takes each patient to get dressed and ready to go.

You'll see that many sections of the charts include check-boxes. You don't need to mark the boxes to use them, though; if you count items mentally rather than with check marks, you can use the charts indefinitely. If you do prefer to make check marks, consider having the charts laminated so that you can use write-on, wipe-off markers.

Because the coding and documentation requirements for established patient visits and new patient visits are different, I created a chart for each. CPT defines a new patient as one who has not received any professional services from you or another family physician who belongs to your practice within the past three years. Each chart is divided into three sections for the three key components of E/M services, with history and examination on the front and medical decision making on the back. One column is devoted to each level of visit.

The exam criteria included in the charts are for the general multisystem exam. The history and decision making sections are each divided into three parts by horizontal rules. History documentation must meet criteria in all three sections. Decision making must meet criteria in two of the three sections.

The charts are organized with both the H&P (History & Physical) and the SOAP (Subjective/Objective/Assessment/Plan) formats in mind so that they can be easily incorporated into any documentation process.

As you compare the two charts, you'll notice one important difference. The documentation guidelines for new patient visit codes require that criteria for all three key components be met, while the documentation guidelines for the established patient visit codes require that criteria be met for two of the three key components. (Of course this doesn't mean that for established patient visits you need to document only two components. It simply means that the code selection depends on the highest level that two of the components have in common.) A sentence at the bottom of each chart serves as a reminder.

The charts also differ in that except for at level five, the levels of history and exam associated with the new patient visit codes are higher than the levels of history and exam associated with the corresponding established patient visit codes.

Practice with the charts

The best way to understand the charts is to put them to work. Let's apply them to two case presentations, one for an established patient and the other for a new patient.

Case 1. This represents a common type of Medicare visit: the routine follow-up for a patient with a chronic problem. The patient in this case is your partner's patient, a gentleman you have not seen before. You are seeing him today because your partner is at a CME course and the patient does not want to wait a week until your partner returns.

If a code leaps to mind just from your reading of the note, fine. Let's use the established patient chart to determine a code for the visit; then you can see how good your first impression was.

History. As we review the note, the chart reminds us that the overall level of history depends on the levels of the HPI, ROS and PFSH documented in the note. The history touches on a couple of HPI elements, which qualifies it for an expanded problem-focused history. The ROS covers four systems: Cardiovascular, Respiratory, Gastrointestinal and Neurologic, which meets the requirements for a detailed history. By noting the patient's past history of CAD and CHF, as well as his social history, the PFSH is thorough enough to qualify for a comprehensive history.

The overall level of history will be equal to that of the lowest-level component. So although the ROS and PFSH are more involved, this documentation reaches no higher than expanded problem focused, the level associated with the HPI. By the way, a shortcut often useful for determining the level of the history is to check these two points:

• Is there any ROS? If not, the history is problem focused, and you can ignore the rest of the history for the purposes of coding. If even one system is reviewed, then ask yourself:

• Does the HPI cover four elements or the status of three chronic diseases? If not, the history is expanded problem focused and you can ignore the rest in determining the code. If so, then you have to go the whole route, counting ROS and PFSH elements to determine the level of history.

Exam. The exam documentation stacks up against the revised documentation guidelines as follows:

• Constitutional: Any three vital signs (Vital signs: “BP 140/80 sitting; 135/85 standing; P 80; R 14; T 98.7; Wt. 185#”);

• Constitutional: General appearance of patient (General: “Well-developed, well-nourished white male, pleasant and cooperative, in no acute distress”);

• Eyes: Conjunctivae & lids (HEENT: “Conjunctivae: nonicteric”);

• ENT: Oropharynx (HEENT: “Oropharynx: moist mucous membranes”);

• Neck: Neck (Neck: “No JVD”);

• Resp: Respiratory effort (Lungs: “Breathing unlabored”);

• Resp: Auscultation (Lungs: “clear to auscultation bilaterally”);

• CV: Carotids (Neck: “no bruits”);

• CV: Auscultation (Heart: “Regular rate without murmur or S₃”);

• CV: Extremities for edema and/or varicosities (Extremities: “No cyanosis, clubbing or edema”);

• GI: Masses, tenderness (Abdomen: “Nontender, nondistended”);

• GI: Liver & spleen (Abdomen: “No hepatosplenomegaly”);

• Psych: Mood & affect (General: “Mood is somewhat anxious”).

Thirteen elements are documented and more than two systems/areas are covered, so the exam is detailed and meets the exam requirement for a level-four established patient visit.

Decision making: Decision making is made up of three components: diagnoses and/or management options, amount and/or complexity of data to be reviewed, and risk of significant complications, morbidity and/or mortality. On the chart, these appear as “diagnosis,” “data” and “risk.” The overall level of decision making will be determined by the highest level that two of the components have in common.

According to score sheets issued by HCFA in connection with the documentation guidelines, the diagnosis and management of an established problem that is stable counts for one point. The note earns one point for data because lab tests were studied. The risk-related documentation includes the prescription drugs that are part of the plan of care, and that's enough to make the level of risk moderate.

Because the overall level of decision making is determined by the highest level that two components have in common, the decision making in this case is straightforward, determined by diagnosis and management options and by risk. Straightforward decision making is associated with a level-two established patient visit. (By the way, here's another useful shortcut: If two elements of decision making have the same level, that's the overall level of the decision making; if all three have different levels, the middle one determines the overall level.)

Code selection: As the note at the bottom of the chart reminds us, the code for an established patient visit is determined by the highest level that two of the three key components have in common. With an expanded problem-focused history, a detailed exam and straightforward decision making, the visit documented in our sample note qualifies as level three. Therefore, this documentation supports code 99213.

Case 2. Let's look at a new patient visit now.

Following the chart for new patient visits, what level of visit should we code?

History: Three chronic problems were addressed — NIDDM, HTN and hypercholesterolemia. This HPI is associated with a detailed or comprehensive history. The ROS covered two systems, eyes and neurologic, which puts it in line with a detailed history. The PFSH noted the patient's past medical history, family history and social history, making it comprehensive. Since the overall level of history is determined by the highest level that the HPI, ROS and PFSH have in common, the result in this case is a detailed history. The HPI and PFSH reach the comprehensive level, but the detailed ROS is the determining factor. A detailed history is associated with a level-three new patient visit.

Exam: Here's how the exam documentation and the guidelines compare:

• Constitutional: Any three vital signs ... (Vital signs: “BP 138/80; P 78; R 14; T 98.5; Wt. 140#”);

• Eyes: Conjunctivae & lids (Eyes: “Conjunctivae clr”);

• Eyes: Optic discs (Eyes: “Funduscopic exam: benign”);

• ENT: Lips, teeth, gums (Oral: “No lesions”);

• Neck: Neck (Neck: “no JVD”);

• Resp: Auscultation (Lungs: “Clear to auscultation bilaterally”);

• CV: Carotids (Neck: “No bruits”);

• CV: Auscultation (Heart: “Regular rate without murmur”);

• CV: Extremities for edema and/or varicosities (Extremities: “Lower extremities: no pedal edema noted”);

• Skin: Inspection of skin & subcutaneous tissue (Skin: “Skin intact without ulceration”);

• Neuro: Cranial nerves (Neurologic: “Cranial nerves II–IX intact”).

This case demonstrates just how demanding the revised exam guidelines are. What seemed like a very thorough exam is, according to the guidelines, expanded problem focused. Eleven elements are documented and more than one system/area is covered, so the exam meets the exam requirement for a level-two new patient visit. Note that if one more element of the exam had been documented — even a note about the patient's general appearance — the exam would have qualified as detailed. We'll see later how this affects the code for the visit.

Decision making: Remember that decision making is made up of three components — diagnosis, data and risk — two of which will determine the overall level of decision making.

The diagnosis and management of three stable, established problems counts for three points. The documentation earns two points for data because labs and old records were ordered. The documentation listed the prescription drugs that the patient was instructed to continue taking, making the risk level moderate. In total, then, the diagnosis and management options score of three amounts to moderate complexity; the data score of two amounts to low complexity; and the risk level amounts to moderate complexity. Since the overall level of decision making is determined by the highest level that two components have in common, the medical decision making in this case is of moderate complexity, which is associated with a level-four new patient visit.

Code selection: As the note at the bottom of the chart indicates, the code for a new patient visit is determined by the highest level attained by all three key components. The detailed history, expanded problem-focused exam and decision making of moderate complexity qualify the visit as a level two and support code 99202. In this case, the expanded problem-focused exam is the determining factor. If a 12th exam element had been documented in the note, the exam would have been detailed, and the code for the visit would have been 99203.

What next?

The guidelines take some getting used to. They are undeniably and, as many have argued, unnecessarily complicated. But for now they're all we've got. (See “The Documentation Guidelines: Read ‘Em and Weep.”) HCFA, through its regional Medicare carriers, recently began conducting random prepayment reviews of E/M claims using the guidelines. Given that we have to sign our names to Medicare claims, subjecting ourselves to the possibility of an audit as a result, I'm glad we know the auditors’ rules — no matter how excessive they seem.

The guidelines are difficult but not impossible to use, and the charts I've developed make the task easier. If you have the opportunity right now, I'd encourage you to apply the charts to your notes from a few recent visits. The more you use them, the easier the process will get. Good luck!

Editor's note: While this article represents our best efforts to provide accurate information and useful advice, we cannot guarantee that third-party payers will accept the coding recommended.