April 2, 1998
Celeste Kirschner, director
AMA Department of Coding and Nomenclature
MSMS reimbursement ombudsman
Re: Evaluation and Management Documentation Coding Guidelines
Dear Ms. Kirschner:
Practicing physicians have been asked for comments concerning the above guidelines by April 3, 1998. This is submitted in response and we appreciate the opportunity. All references are to the Documentation Guidelines for Evaluation and Management Services from November 1997.
The rationales within the “Introduction” and “General Principles of Medical Record Documentation” are reasonable and will serve as the strategic objectives for the remainder of this paper. An outline of the objectives is important as a filter to judge various tactics. If a tactic enhances the objectives it is used. If a tactic detracts from the objectives it needs to be modified or removed. It is within that context the following was generated.
The section “Documentation of History” is confusing for two reasons.
1. Nomenclature: Two different comparisons of magnitudes are made. The first is the type of history with the divisions of “Problem Focused, Expanded Problem Focused, Detailed and Comprehensive”. The second is a comparison of elements by “Brief, Pertinent, Problem Pertinent, Extended, and Complete”. These words are ambiguous because they have indistinct boundaries of meaning. What is needed is to replace these words with say “small, medium, and large”. When more than three divisions are desired, subdivisions should be made with the root words. For example, split medium into medium-small and medium-large. This way everyone understands the concept the other person is trying to convey and easily remembers the format.
2. Organization: First there is a comparison of magnitude (type), then a division of function (elements) then a second comparison of magnitude. The divisions should be “functional then magnitude” or “magnitude then functional” and not “ magnitude then functional then magnitude” which is difficult to follow. What it appears you are trying to say is that a small history has small elements.
That section is much clearer as (page 5):
HPI ROS PFSH Type of History
small small/0 small/0 small
medium medium medium medium
large large large large
The “Chief Complaint” is fine.
The “History of Present Illness” (page 7) is easier to
“what, where, when, how, why, while, how much and how long”
Comments relative to “ROS”, “PFSH”, “General Multi-System Examination” and “Documentation of the Complexity of Medical Decision Making” are as follows.
The medical record documentation is in effect a model of the health care delivered and while no model is an exact representation of the real events there are different ways of looking at the same thing. Prior to 1850 objects were studied in terms of their “magnitude”. Since that time it has become apparent a deeper understanding of an object is in terms of its “structure”. The structure in turn depends on the functions it contains. Everyone intrinsically understands this process by viewing an object more in terms of what it does (function) than in terms of its appearance (form). Comparisons of magnitudes are important but they not as revealing as classification by function.
The structure of the “Documentation Guidelines” is one of magnitude instead of function. There are no intrinsic limits to magnitudes and boundaries are arbitrary. As a result conflicts arise over territorial considerations rather than a study of the processes involved. The section on medical decision making alludes to the functional changes each physician makes to the patient over time but then confuses complexities and amounts.
As an example, it is natural for a trained practitioner to see a patient and generate many “bullets” and then discard most of them as not pertinent. You perceive this process, usually interpreted as intelligence, as fraud and abuse. The immediate options are either not to charge for the expertise or go back and reconstruct all the irrelevancies. The approach of capturing relevant data by capturing all data is inefficient. Worse, it encourages stockpiling of ammunition not for patient care, but to shoot for maximal reimbursement.
A better solution is to change the structure. Functions have the fundamental constraint of a singular output. This allows a great deal of utility, understanding and basis for study. Since the prevention, making, modifying and removing diagnoses are the important functions, the documentation should indicate the necessary and sufficient conditions needed by the individual practitioner for those purposes.
The next question then of course becomes the definition of those conditions. No two people use the same inputs to generate a given output. This fact has interesting consequences. The nihilistic approach of denying care does not produce the appropriate output while at the same time the least complex inputs to reach an output are the most efficient. Since there are multiple outputs, the situation is not quite so simple but the pattern is clear. Thus, what is appropriate, is to reward the physicians who generate the fewest “bullets” of exams, labs, treatments etc for the desired outputs, not the other way around as you have done.
June 25, 2002
The May meeting of the HHS Advisory Committee on Regulatory Reform has recommended the E&M guidelines be abolished without suggesting an alternative. The above letter has earlier musings but while criticisms are easy, solutions are difficult.
Consider an analogy of medical positions and moves to chess positions and moves. For any position, there are many possible moves, while each move generates a new position. Chess is played using a relative value of positions. Consequently, the evaluation of a move becomes the difference between the beginning and subsequent positions.
The objective in chess is to win by moving to the best available position. The idea here is to define the objective of medicine as a move to a better position of health care.
The E&M guidelines, CPT and ICD-9 codes are all positions. Moves are not defined and thus no objective can be reached. The lack of moves is the fundamental problem with the present system of reimbursement and attempts with modification. Reconstructing positions with the addition of moves is suggested.
The present positions would be replaced by a virtual patient. Positions would be assigned a relative value of health. Conversion factors would then be applied to a change in positions to determine remuneration.
The difficulty of this approach will be in the number of positions. This proposal is ultimately a complete model of medical care representing a large, difficult project. Nonetheless, the key to intelligent health care is to compensate providers if they improve the care of the patient. There are interesting consequences to this logic but the point is to let physicians decide how and put the incentive on if.
James J. Rice M.D.
Muskegon Surgical Associates back index