Who's Coding and How in Physician Practices: a Survey of E/M Documentation and Coding Practices

by Susan Fenton, PhD, MBA, RHIA, and Larry D. Gamm, PhD

A survey of physician practices finds that most use multiple documentation and coding methods, few of them involving new technologies.

Evaluation and management (E/M) codes are primarily assigned by physicians for office services; that is, nonprocedural visits. More than half of the top 110 Medicare Part B codes (ranked by charges) were E/M codes in calendar year 2005. The charges topped $28 billion, more than one-quarter of total Medicare Part B charges.1

Given the magnitude of the services and the complexity of assigning the codes, E/M documentation and coding have been sources of concern, especially within the Department of Health and Human Services. There, the Office of the Inspector General reviews E/M code assignment as part of the Correct Coding Initiative, which promotes national correct coding practices and seeks to control improper coding that leads to inappropriate payment in Part B claims.

Despite the concern, little has been established about how physician practices document and code services. Important subjects such as the prevalence of different documentation methods along with the prevalence of the E/M coding methods have scarcely been addressed in research to date.

This article offers results of an exploratory survey determining the documentation and coding methods used by physician practices employing HIM professionals. The survey is the first step in a study that ultimately will examine code profile differences by documentation and coding method.

Methodology and full results of the study are available online in the FORE Library: HIM Body of Knowledge under the title “Evaluation and Management Documentation and Coding: Results from an Exploratory Study.” AHIMA members may access the Body of Knowledge by logging in to myAHIMA at www.ahima.org.

A Variety of Settings and Experience

In all, 442 AHIMA members working in physician practices, ambulatory care facilities, and rural health centers took part in the survey. They were asked about their settings from 2005, since the most current data available from the Centers for Medicare and Medicaid Services is from that year.

One-third of respondents identified themselves as coding professionals, and another 30 percent identified themselves as managers or directors. The largest share held an associate degree (38 percent). An additional 28 percent held a baccalaureate degree, and 9 percent a master’s degree.

Smaller percentages report being high school graduates, HIM certificate holders, or AHIMA independent study program graduates. Nearly one-half of respondents have held their position for one to four years, and more than one-quarter have held their positions for five to 10 years.

By far, the largest share of respondents worked in a private physician group (41 percent). The majority worked in multi-specialty practices (63 percent). Practice sizes varied overall. More than one-third of respondents represented practices of 10 or fewer physicians (37 percent). An equal share was relatively evenly distributed in practices ranging from 11 to 100 physicians. Approximately one-quarter were from practices of more than 100 (26 percent).

The largest group consisted of HIM professionals in practices with 10 or fewer physicians. This is a somewhat surprising finding. It was not previously known that many HIM professionals were employed in small physician practices.

Documentation Methods

How did the physicians in your practice document their regular office visits?





Computerized template


Hard-copy template


Free-form EHR




(Respondents were able to select more than one documentation method.)

Respondents were asked to indicate the documentation methods used by greater than 75 percent of their physicians in 2005. Slightly more than 58 percent reported using more than one documentation method. Only 41 percent reported using a single method. Almost 10 percent reported using four or more methods.

Slightly more than a quarter of the organizations (27 percent) had significantly changed their documentation practices in 2005. This was defined as projects such as implementing an EHR system or special physician education; that is, efforts beyond normal code validation or regular documentation improvement. Of those reporting changes, 29 percent said the changes were the result of EHR implementations.

Coding Methods

How were E/M codes assigned in your practice?

Coding Methods Pie chart

Clinician assigns


Coder assigns manually


Coder assigns with encoder


EHR suggests codes


Percentages may not total 100 due to rounding.


Many respondents reported that their practices assigned E/M codes in multiple ways in 2005. A number of respondents indicated that more than one coding method was used, and in several cases, that although the physicians handwrote or dictated their notes, handheld software was used to assist in assigning the code. Results were grouped into like documentation methods, choosing the most technologically advanced method for each.

In nearly half of practices, clinicians assigned the code (49 percent). Coders assigned codes manually in nearly one-third of practices surveyed (32 percent).

Validation Methods

How was code validation performed in your practice?

Coding Methods Pie chart

Coder/manager manually


Coder/manager with encoder


Code validation not done


Contract company


Percentages may not total 100 due to rounding.


It is important for organizations to check the accuracy of code assignment to minimize their risk of fraud and abuse charges. There are multiple methods for validating E/M coding. More than half of respondents internally check E/M coding manually. Surprisingly, 16 percent of respondents reported that their practices did not validate E/M coding in 2005.

Some practices use more than one method, and the multiple responses were grouped into single variables. Recoding assigned a respondent to the highest technology level of coding validation. If a respondent reported using a mix of contract company and internal validation, they were recoded to either coder/manager manually or coder/manager with software. If both manual validation and software validation was reported it was assigned to coder/manager with software since that is a higher level of technology.

Number of Full-Time Coders

How many coders does your practice employ?

Full-Time Coders Bar

The survey took in a wide range of practice sizes, from single practitioner to practices of more than 100 physicians. Accordingly, the number of full-time coders employed in the practices varied widely, also. Answers ranged from 0 to 200. The mean was 8.5; the median was 3.

Credentials and Certifications
How many coders in your practice are credentialed or certified?
Credential/Certification Percent
Registered health information administrator and registered health information technician 58%
Certified coding specialist–physician-based 56%
Certified professional coder 51%
Certified coding specialist 37%
Certified coding associate 12%
Other 10%
Certified medical coder 8%

(Respondents were able to select more than one credential/certification.)

Coder credentialing and certification was developed in the 1990s as correct coding became essential in all healthcare settings. Only 14 percent of respondents reported that no credentialed or certified coders were employed at their practice in 2005. An additional 42 percent indicated some credentialed or certified coders, and 44 percent reported that all of their coders were credentialed or certified. A similar number reported that their practices required credentialing or certification in 2005, whether coders were on staff or contracted (46 percent).

A majority of the organizations had no formal, in-house education for their coders in 2005 (60 percent).

Documentation and Coding Improvement

How would you improve your current documentation and coding method if you could?

Coding Methods Pie chart

Physician education and behavior change


Computer-assisted coding


Non-EHR documentation


EHR implementation




Percentages may not total 100 due to rounding.


Asked to select their most desired method for improving current documentation and coding methods, the majority of respondents indicated they would choose clinician education and behavior change. Far fewer chose implementing an EHR. This suggests that technology will not be adequate to address the documentation and coding problems found in the current system.

Long on Methods, Short on IT

The survey found that more physician practices (51 percent) use traditional documentation methods of handwriting and dictation either singly or in combination than any other type of documentation method. The actual percentage using these traditional technologies is probably even higher because, for purposes of the study, a practice using dictation in combination with an electronic health record (EHR) was assigned to the EHR.

Increasing interest in health IT might suggest that EHRs and computer-assisted coding are commonplace, especially given the increased discussion of reimbursement incentives for physician practices that adopt technology. Health IT initiatives are being sponsored from the highest levels of the federal government.

However, the survey results illustrate a different picture. Technology, either in the form of an encoder or computer-assisted coding, EHR software suggesting the codes, is used for E/M coding in less than 20 percent of total practices. This is less than one-half of the practices that reported using an EHR. The reasons for this surprising finding cannot be deduced from this survey. Additional research is necessary.

This low use of technology has implications for documentation and coding, because effective IT implementation has been found to improve most processes. There is evidence to suggest that technology will benefit providers by ensuring proper reimbursement.2-4 At the same time, research indicates that computer-assisted coding properly designed and deployed has the potential to reduce healthcare fraud.5 The low levels of technology adoption, especially for coding, indicate that forces other than technology are involved.

The results from this exploratory study support recent initiatives promoting research in the HIM profession. It is vital that HIM professionals, as self-professed health information experts, understand the patterns, benefits, and barriers in the use of new and emerging documentation and coding technologies (among other topics).

The survey results presented here are only the first phase of this research study.  Other analyses examining the relationships between documentation and coding methods, practice characteristics, regulatory guidelines, and geographical and environmental characteristics are being conducted.  The intended final analysis for this study is to determine whether different documentation and coding methods result in different code profiles. It is essential the profession understand the impact of these different forces on the production of health information and coded data.


  1. Centers for Medicare and Medicaid Services. “Medicare Leading Part B CPT Procedure Codes Based on Allowed Charges.” Available online at www.cms.hhs.gov/MedicareFeeforSvcPartsAB/04_MedicareUtilizationforPartB.asp.
  2. Terry, Ken. “Blending: The New Downcoding.” Medical Economics, May 5, 2006: 1–4.
  3. Terry, Ken. “The Catch-22 in EHRs.” Medical Economics, July 21, 2006: 37.
  4. Bates, David W. “Physicians and Ambulatory Electronic Health Records.” Health Affairs 24, no. 5 (Sept./Oct. 2005): 1180–89.
  5. Foundation of Research and Education. “Automated Coding Software: Development and Use to Enhance Anti-Fraud Activities.” 2005. Available online at www.ahima.org.

Susan Fenton (susan.fenton@ahima.org) is director of research at AHIMA’s Foundation of Research and Education. Larry D. Gamm is MHA program director and director of the Southwest Rural Health Research Center School of Rural Public Health at Texas A&M University System Health Science Center.

Article citation:
Fenton, Susan H.; Gamm, Larry D.. "Who's Coding and How in Physician Practices: a Survey of E/M Documentation and Coding Practices" Journal of AHIMA 78, no.7 (July 2007): 52-55.