While supplies last! Free 2022 Essentials of Interventional Radiology Coding book with every ICD10monitor webcast order. No code required. Order now >

Office visits, medical necessity and ICD-10-CM coding: How do these three things fit together? Of eminent importance, proving the medical necessity of physician services has become the newest and most relevant matter at the spearhead of various federal auditing efforts to recoup reimbursements made to unsuspecting providers.

Until recently, the Centers for Medicare & Medicaid Services (CMS) in many instances gingerly sidestepped direct issues regarding medical necessity. The crux of the matter lies in the fact that to question the medical necessity of a service performed by a physician is to unequivocally question the clinical judgment of that physician. To rid the medical necessity part of the equation of any gray areas open to clinical opinion, interpretation or cogent argument, CMS and its jurisdictional entities have developed an extensive library of black-and-white Local Coverage Determinations (LCDs) and National Coverage Determinations (NCDs).

Most of these coverage policies promulgate very specific ICD-9-CM codes that spell out the medical necessity of specific covered services and procedures. In these cases, navigating the medical necessity roadmap is quite clear:  the physicians document the diagnostic statements, the coders assign codes and the billers report them. The service either is or is not covered, per the LCDs or NCDs, period. There is no room for argument, no need to question anyone’s clinical judgment.

Recently, however, other areas of medical necessity challenge have surfaced that aren’t quite so clear-cut. One of these areas falls squarely within the most-reported service by any physician across all specialties and geographic areas: the typical office visit. A few jurisdictional Medicare MACs and Part B carriers have been – to give them the benefit of the doubt – quietly finessing the guidelines established for documenting and coding evaluation and management (E/M) services (or to be more dramatic, they may be reinterpreting and/or manipulating those guidelines). Specifically, the current 1995 and 1997 E/M Documentation Guidelines for established office visits state that the appropriate E/M level is assigned by using two of three required key components (history, physical examination and/or medical decision-making). In the past it has been left up to the clinical judgment of the provider or the professional judgment of the coder as to which two components would be used in the equation to meet or exceed requirements. However, a few select MACs and Part B carriers have been enforcing a new (but still largely unknown) requirement. With alarming frequency, these entities now require one of the two documented key components for established patients to be the medical decision-making (MDM) component.

What does this have to do with ICD-9-CM coding, soon to be ICD-10-CM coding?

A Real-Life Clinical Scenario

The three concepts of “office visits, medical necessity and ICD-10-CM coding” all were about to converge abruptly during a coding compliance (re-abstraction) and clinical documentation improvement audit I recently conducted as part of a clinic’s preparation for ICD-10-CM. This review exercise was being done for a multi-specialty clinic with numerous providers. First up was a longstanding provider in the group, the “matriarch,” if you will, of this particular clinic. Her medical record (MR) notes were consistently troublesome, lacking in many of the elements needed to establish medical necessity, to assign appropriate ICD-9-CM codes in compliance with guidelines, and to prepare for ICD-10-CM. The documentation rambled; it lacked clear statements; it was vague; it used non-committal terms such as “history of” and “appears to be” for observations and findings (which struck me as representing either confirmed and/or already documented diagnostic conditions established by numerous providers in the patient healthcare continuum).

In numerous instances diagnoses and conditions were validated further by the patients’ current medications. It was one of those examples of documentation we consultants tirelessly have been warning providers about, saying “your MR documentation must be concise, detailed and unambiguous down to a granular level if such documentation is clinically relevant to the patient encounter. This helps establish medical necessity and provides the necessary data for coders to abstract your records accurately. This level of detail will be needed for ICD-10-CM.”

In one particular case, I re-abstracted a rather rambling clinical scenario of an established 76-year-old patient (coded at a high level E/M visit for a returning patient) presenting with painful skin lesions, several of which were weeping, of a six-week duration on the distal portions of the legs. The patient was active and ambulatory. Various other chronic conditions were documented in the “past medical history” portion of the note using the preamble “history of,” with examples including diabetes mellitus, CAD, COPD, hypertension and even current tobacco smoking.

As a healthcare clinician, I did not construe these conditions as medical history data points, but took note that they were active, ongoing diagnoses under assessment, treatment and monitoring. There was also a litany of meds the provider needed to consider in the assessment and treatment of this patient, including daily insulin and anti-coagulant therapy.



After meandering through a very disjointed but thorough history and a problem-focused physical examination, I then waded into the MDM portion of the note. The “assessment” simply stated “lesions on legs, several of which have ruptured” and the “plan” gave direction for direct care of the ruptured lesions only (i.e., “cover the lesions with dry sterile bandages.”) No correlation between the lesions and the diabetes, or with any of the other potentially complicating comorbidities, was drawn. In fact, none of those conditions originally were coded – only “skin lesions.”

Sensing the provider had to have considered the influencing chronic conditions as well as the patient’s daily medications in this particular assessment and treatment, I called on her to explain the premise of a physician query. She confirmed my inclinations. She had, in fact, considered cause-and-effect relationships between the diabetes and the peripheral lesions, as well as any possible adverse effects of long-term medications.

“Yes, of course there is correlation. There were many correlations, in fact,” she said.  “I thought about writing more but did not have time. However, I know that patient quite well and remember all I need to know. Why the 22 questions? It’s just an office visit.”

Well, not to be glib, but … hello? If it’s not documented, it’s not done. By this point even my 9-year-old nephew knows that adage. Extend that trite saying to “if it’s not documented, you don’t get paid” and we’re more on target, or go even further in this audit-aggressive climate: “if it’s not documented, you might have to pay it back!”

Lessons Learned

Based on the above clinical scenario, the findings of which surfaced repeatedly in numerous other cases throughout the audit, several opportunities for improvement were established, presenting opportunities to correct MR documentation quality, improve the establishment of medical necessity, solidify ICD-9-CM coding and begin to move toward accommodating ICD10-CM code assignments. Aside from specific code-correction recommendations and focused CDI efforts, among those general opportunities were:

  1. MR documentation must contain all clinically relevant cognitive efforts made by the provider so they can be used in the MDM area of leveling an E/M service, as well as in establishing the medical necessity of the service;
  2. Follow the architecture of the SOAP note format or other preferred format to improve documentation prompting, flow and clinical data integrity;
  3. All current confirmed diagnoses should be listed clearly, and not prefaced with “history of” or “appears to be” unless such qualifying language is appropriate to the patient encounter for past or as-yet-unconfirmed diagnoses;
  4. Coders must comb through the MR documentation for scenarios in which a clinical correlation may exist but was missed or omitted in the documentation by the provider, and physician queries should be generated as needed;
  5. Always remember: “If it’s not documented …” (you know the rest).


It has not gone unnoticed that there is a trend of increasing levels of E/M services being reported nationwide; payments for these services likewise have spiked during the last few years.  This also predicts a spike in targeted audits. The various differences between E/M services, for example 99214 versus 99212, can mean huge differences in reimbursement revenue for medical practices as well as enormous costs to the payers, such as the Medicare program. Thus the scrutiny, thus the intense interest in medical necessity, and of course, thus the need for providers to close gaps in MR documentation and for coders to query providers when they fail to do so.

Concise, accurate and fully thorough documentation of patient encounters will support solid ICD-9-CM coding – and in preparation for the ICD-10-CM shift, it will mitigate ambiguous or incomplete MR documentation, creating the granular information necessary to accommodate the expanded code descriptions under ICD-10-CM.

Ensuring that the progress notes indicate medical necessity of E/M services, some of the most common services in healthcare, also will do wonders for preparation. These efforts are on parallel tracts. In particular, fleshing out the medical decision-making component and drawing clinically appropriate correlations between the history and PE elements certainly will aid in the effort to solidify MR documentation in anticipation of ICD-10-CM. This also will help ensure appropriate payment now under ICD-9-CM coding by deflecting questions of medical necessity.

By following all of these parameters, the new adage could be “if we documented it, it most certainly was done!”

About the Author

Michael G. Calahan, PA, MBA, is vice president of hospital and physician compliance at Healthcare Consulting Solutions (HCS).

Contact the Author



To comment on this article please go to editor@icd10monitor.com


Michael Calahan, PA, MBA, AHIMA-Approved ICD-10 CM/PCS Trainer

As Vice President of Hospital and Physician Compliance at HealthCare Consulting Solutions (HCS), Michael G. Calahan, PA, MBA, provides subject matter expertise and consulting services. He has been a speaker for MedLearn for nearly 10 years and has provided webinars on device credit reporting since 2013.

You May Also Like

Medlearn Media NPOS Non-patient outcome spending

The PHE Update

The renewal of the PHE indicates that a public health emergency (PHE) exists. As expected, the COVID-19 PHE (Public Health Emergency) was extended for the

Read More

Leave a Reply

Your Name(Required)
Your Email(Required)