Electronic documentation vs. dictation: how do they compare? (Medical Records).

Dictation has long been the standard for making notes to document patient encounters. It is convenient and reasonably fast compared to handwritten notes. However, there are several problems with dictation:

* Transcribing the dictated note, awaiting review and signature by the physician and filing it in the patient record adds a significant delay to the process.

* Previous studies documented that free text documents have high rates of missing information.

* There is little or no standardization of the content of free text and it cannot be queried consistently or systematically for data retrieval.

* A significant cost is associated with transcribing dictated notes and tracking their course to the medical record.

Structured documentation (SD) using electronic notes has several advantages over dictated and transcribed notes. Electronic notes:

* Store coded data for subsequent retrieval

* Provide consistently complete and accurate data

* Are immediately available

* Decrease transcription costs

* Potentially promote best practices

Despite these advantages, physicians have not embraced SD. Some believe that in order to make SD acceptable to physicians SD must:

* Provide sufficient expressive power to describe relevant information

* Correspond intuitively to the physician's usual method of working

* Be flexible

* Present data in a predictable order

* Require less time, or barely little more time, than current reporting methods (2-4)

To evaluate differences between the dictation/transcription and similar notes created electronically with an SD tool called PowerNote, a small study was conducted recently at University of Missouri Health Care (UMHC). The study involved three clinical procedures. *

* Central venous catheter placement (CVC)

* Chest tube insertion (CT)

* Arterial line placement (AL)

Dictation standards for these and other procedures appear in the UMHG General Surgery Procedure Dictation Forms booklet. CVC notes contain 21 essential content elements. Al notes, 20, and CT, 25.

Using the forms as a guide, electronic PowerNotes were developed for each procedure.

PowerNote, produced by Cerner Corporation in Kansas City, Mo., is UMHC's electronic medical record (EMR). Users simply point and click to select appropriate choices from lists of possible terms and phrases. Free text entry is also an option. The end result is a document that closely resembles a transcribed procedure note.

Physicians, the medical records committee, the compliance office and legal counsel reviewed the notes to ensure the electronically generated notes met all medical, legal and administrative requirements.

Two general/trauma surgeons were recruited and given 15-minute training on how to use PowerNote. Then, the surgeons documented 34 CVC, CT, and Al procedures. The electronic procedure notes were compared to 34 dictated and transcribed notes created by a separate group of surgeons and residents during the same period.

The notes were measured for:

* Time necessary to complete dictation or PowerNote creation (to the nearest half minute)

* Time until available to the medical record

* Number of essential content elements captured on each note

The mean time needed to dictate notes was two minutes. The comparative mean time to create a PowerNote was 2.44 minutes after deducting the times for the first five as a learning process.

Not surprisingly, the mean time from dictation to chart availability for the 34 dictated notes was greater than 16.85 days. On the other hand, the procedure notes created in PowerNote became automatically available to the EMR as soon as they were created.

The dictated notes captured 86.7 percent of essential content elements for the group as a whole. PowerNote captured 95.2 percent of essential content elements for the group as a whole.

More different content elements were omitted by dictated notes than by PowerNote. Out of 21 essential content elements for CVC, 14 different elements were omitted at least once; for AL, 9 of 20; and for CT, 9 of 25.

For PowerNotes only 5 of 21 CVC elements, 4 of 20 AL elements, and 2 of 25 CT elements were omitted.

Based on this limited sample, notes created in PowerNote appear to meet all of the requirements necessary for physician acceptance of SD. They have sufficient expressive power to allow clinically relevant information to be described in detail. They are flexible, giving the physician the opportunity to include or exclude any portion of the template in the final document.

Creating a note in PowerNote is quite similar to dictating a note and corresponds closely to the routine employed by physicians to document a procedure. The data entered is in a predictable form. Finally, the time needed to create a note using PowerNote is only slightly longer than dictating the same document.

Saving time

Time is probably the most critical of these requirements. Physicians will not use a tool if it involves more time than they feel is acceptable (2-4) and will ignore other benefits of the method in the process.

PowerNote allows users to save a pre-completed note, a generic note with the most commonly used terms already saved and in place. The user brings up the pre-completed note and adds data that changes from one procedure to the next. This technique significantly speeded up the process of creating the notes in this study and was a major factor in gaining physician acceptance in the use of SD.

One of the major advantages of PowerNote is the time needed to make the document available to the medical record. Since PowerNotes are created within the EMR, they are available to the medical record immediately upon completion. Users can review and sign their notes in real-time, eliminating the need for medical records personnel to track the progress of notes through the system.

This is most advantageous in a teaching environment where a resident's note needs an attending addendum and signature. SD is clearly superior to dictation/transcription in making information available to the medical record in a timely fashion.

Despite the fact that approved dictation models existed for each procedure, the dictated notes varied widely in the information that was included in the notes. Several of the dictated notes did not even follow the accepted format outlined by the templates.

SD provided more complete information in each note than dictated notes. Similar findings have been found in previous work. (5) Not only were the electronically generated notes more complete, they were also more consistent in capturing essential content elements.

Because PowerNote shows all possible descriptors of a concept, completeness and consistency is promoted by a reminder effect. The PowerNote templates cue physicians to enter information necessary to fully document each procedure. This phenomenon has been found in other applications of SD. (2.5-7)

Since the PowerNote template guides the user in a consistent manner, the format of notes created in PowerNote was unfailingly consistent.

Saving money

SD has the potential for significant cost savings. Clinical services at UMHC currently pay 12.5 cents per line for transcription. The number of lines in a document is determined by transcribers, but it is easy to see that a considerable savings accrue if more documents are created using SD.

Additional savings might also come from decreased need for medical records personnel to track and file all of the paper documents. Since a note is immediately available, the bill for the procedure can be sent out as soon as the procedure is done rather than waiting until a copy of a transcribed note is ready. Timely billing has been shown to result in improved reimbursement.

A commonly overlooked benefit of SD is the contribution toward best practices that accompanies standardization in the way a procedure is performed. In order to create a document template, physicians must first agree to a standardized technique for performing the procedure. This had already been done for the three procedures in this study.

In creating subsequent templates, it is interesting to observe physicians discussing the various ways that they do a given procedure and arriving at consensus about what they judge to be the optimal technique. By agreeing to standardize their approach, the possibility that errors will occur is reduced. The standardized approach is reinforced by the consistent manner in which it is documented by PowerNote. In this way, utilizing SD may enhance patient safety.

Although not evaluated in this study, one of SD's purported advantages over free text is the ability to retrieve data from stored sources. Data that is stored in a consistent and structured fashion can be queried and retrieved for analysis. Free text is not suitable for information management and decision support. (8)

The terms used by PowerNote are coded and stored a structured manner. The functionality of querying and retrieving this data is being developed, but does not currently exist. When this development is complete, users will have the ability to retrieve data that is found in clinical notes from the database, making clinical research on these documents easier than presently possible from free text documents. (8)

Because this study demonstrated potential benefits of adopting SD, more extensive PowerNote development is being supported at UMHC. In February 2001, the Newborn Intensive Care Unit went paperless adopting PowerNotes exclusively for admission, progress, discharge and selected procedure notes.

As of October 2002, over 161 users in more than 15 departments created more than 9,000 notes on patients. We feel there is a bright future for SD as the accepted way of documenting all manner of clinical encounters.

Acknowledgements

This research was supported by National Library of Medicine Medical Informatics Research Training Program Grant T15-LM07089. The following individuals were of great assistance conducting this study. Cheryl McGowan, Donna Wemboff and Lynn Haidar of Information Technology Services created the templates for the PowerNotes used in the study. Wade Davis provided statistical analysis. We are grateful for their efforts.

References

(1.) Kuhn, K, Swobodnik, W, . Johannes, RS, Zemmler, T., Stange, EF, Ditschuneit, H. and others. "The Quality of Gastroenterological Reports based on Free Text Dictation: An Evaluation in Endoscopy and Ultrasonography." Endoscopy. 1991, 23(5): 262-264.

(2.) Moorman, PW., van Ginneken, AM., van Der, LJ., van Bemmel JH,.. "A Model for Structured Data Entry Based on Explicit Descriptional Knowledge." Methods Inf Med. 1994, 33(5): 454-463.

(3.) Campbell, KE., Wieckert, K., Fagan, LM., Musen, MA.. "A Computer-based Tool for Generation of Progress Notes." Proc Annu Symp Comput Appl Med Care. 1993, 284-288.

(4.) Young, DW. "What Makes Doctors Use Computers?" J R Soc Med. 1984, 77(8): 663-667.

(5.) Kent, DL., Shortliffe, EH., Carlson, RW., Bischoff, MB., Jacobs, CD.. "Improvements in Data Collection Through Physician use of a Computer-based Chemotherapy Treatment Consultant." J Clin Oncol 1985. 3(10): 1409-1417.

(6.) Gouveia-Oliveira, A., Raposo, VD., Salgado, NC., Almeida, I., Nobre-Leitao, C., de Melo, FG.. "Longitudinal Comparative study on the Influence of Computers on Reporting of Clinical Data." Endoscopy. 1991, 23(6): 334-337.

(7.) Brown, SH., Miller, RA., Camp, HN., Guise, DA., Walker, HK.. "Empirical Derivation of an Electronic Clinically Useful Problem Statement System." Ann Intern Med. 1999, 131(2): 117-126.

(8.) Rector, AL., Nowlan, WA., Kay, S.. "Foundations for an Electronic Medical Record." Inf Med. 1991, 30(3): 179-186.

RELATED ARTICLE: Appearance of completed note created in PowerNote

Final report

Arterial Line Placement

Patient:

Age: 36 Years Sex: Male

DOB: 09/02/1964

Author: Kamath, Ravishanker K

Procedure

UMHC Procedure Note 07/17/2001 6:30:00 PM Service

Surgery.

Attending Physician

Dr. Kamath, Ravishanker K.

Procedure Performed By

Resident Physician: Dr. A. Steven.

Name of Procedure and Codes

Arterial Line Placement: Percutaneous CPT #36620.

Pre Op Diagnosis and Codes

Multitrauma.

Post Op Diagnosis and Codes

Same.

Location of Procedure

SICU 

Indication for Procedure

Frequent Arterial Blood Sampling.

Monitoring During Procedure

Blood Pressure Monitoring.

Cardiac Monitoring.

Continuous Pulse Oximetry.

Prep and Technique

Operator Prep: Cap, Mask, Sterile gloves, Sterile gown.

Patient Prep: Prepped in sterile manner, Draped in sterile manner. A

Anesthesia

Local.

Position of Patient

Supine.

Site

Radial: Right

Catheter Size

20 gauge.

Description of Procedure

Collateral Circulation: Assured by Allen's test.

Positioning: Wrist positioned in extension and held in place on an arm board with adhesive tape, volar surface of the wrist.

Localization of Artery: By palpation.

Arterial Puncture: Without difficulty.

Catheter Advancement: needle and guide wire withdrawn after arterial bleeding noted from outer end of the catheter.

Transfer Location

None.

Condition

Critical.

Procedure Tolerated

Well.

Estimated Blood Loss

2 cc.

Complications

None.

Attending Present During Procedure Authorization

This document is approved: and electronically authorized by Dr. Ranvishanker K. Kamath.

William M. Sangster, MD, is a physician executive at Cerner Corporation in Kansas City, Mo. He can be reached by phone at [phone] or by e-mail at bsangster@cerner. Com.

Robert Hodge, MD, FACPE, CPE, is a professor of clinical medicine at University of Missouri School of Medicine in Columbia, Mo. He can be reached by phone at [phone] or by email at [e-mail].