 |
|
| |
|
| |
 |
| |
 |
| |
It's not too late! Your support of the SIIM Research & Education Fund through the 4th Annual "Ride to SIIM" will help fund the SIIM Grant Program and the Samuel J. Dwyer, III, PhD, FSIIM, Memorial Lecture.
Make a per-mile contribution to the SIIM Research & Education Fund today!
|
|
| |
|
| |
|
|
 |
 |
Why Are We Still Entering Data Manually?
Workflow in the Era of IHE |
| |
| Authors: |
Steven C. Horii, MD, FSIIM, University of Pennsylvania Medical Center; Kevin Rakszawski; Vandana Sood
|
| |
| Hypothesis: |
Sonographer time performing manual data entry is still significant, despite filmless operation and the use of DICOM Modality Worklist.
|
| |
| Introduction: |
Prior to the introduction of PACS in the Ultrasound Section, the author (SH) performed a time-and-motion study to determine how much time sonographers spent interacting with the RIS. He found that the time required for checking the radiology schedule, entering patient information into the ultrasound machine, and completing the examination when done was in the range of 3-5 minutes per sonographer per patient. Though the Section has been operating a PACS since 1996, he has noted that sonographers still spend much time using RIS workstations. With the aid of the co-authors, he again studied the time spent in several RIS-related tasks. The authors also studied the potential impact on revenue of the time spent completing examinations.
|
| |
| Methods: |
| Specific tasks - examination completion, scheduling a study, reviewing a previous exam report, and reviewing the daily schedule - were manually timed by the two co-authors (KR, VS). Descriptive statistics were computed in Excel (Microsoft Corporation, Redmond, WA). A review of the typical sonographer workflow was done and compared the availability of a replacement of the manual step with an automated one, based on the IHE Scheduled Workflow Profile. Using average labor rates for sonographers (in Philadelphia), the cost of the time spent completing examinations was determined. The potential revenue gain, if this time could be eliminated, was also evaluated. |
| |
| Results: |
The following are the mean (std deviation) times observed
Complete examination: 59.3 (19.4) seconds
Schedule examination: 344.2 (292.3) seconds
Review prior exam: 67.2 (51.5) seconds
Review schedule: 48.4 (24.2) seconds
For an examination volume of 80 studies/day, the total time spent completing examinations is 4,740 seconds or about 79 minutes.
|
| |
| Discussion: |
The reason for focusing on exam completion time is that the other tasks are either not normally performed by sonographers (scheduling examinations) or are not amenable to automation (reviewing a prior exam and reviewing the schedule). There is also a mechanism provided for in DICOM, that would allow for automation; the Modality Performed Procedure Step (MPPS). This is also a part of the IHE Scheduled Workflow Profile.
Use of the DICOM Modality Worklist feature has been widely implemented and has, for the majority of cases, eliminated manual entry of patient demographic data by technologists. Many ultrasound machines support MPPS, so why is it not used?
The problem with MPPS is that it is difficult to implement without creating other types of problems. For example, it is not unusual in ultrasound to take additional images after the sonographer has done his or her part of the study. In some ultrasound systems, either the whole examination, or the last image taken, will not be sent to the PACS unless the examination is “ended” at the machine. It is the process of ending the examination that creates the MPPS message. While the concept of “procedure step” means that it is possible to end one of the examination steps and end another after the radiologist checks the study, many ultrasound examinations are a single procedural step. For systems that allow the sending of examinations in progress, it usually is only the last image that is not sent until the study is ended. One work-around for this has been to capture the last image twice. Storing one extra image per study seems like a trivial additional storage “cost,” but with large examination volumes, this can amount to the storage equivalent of another examination per day.
While the potential time savings by automating examination completion is about eight minutes per sonographer per day (in our Department), that is enough time to perform only a very short ultrasound examination. So, could the recovered eight minutes be put to practical use? In the aggregate, the time savings per day would amount to enough time to perform an additional examination per day, but since scanning is on a per sonographer basis, the time cost should be viewed as a per sonographer time. One scenario that might allow for using this time is practiced about 10% of the time. That is to have one sonographer start an examination and another continue, or complete, it. That would allow for aggregating time savings from two sonographers, or about 16 minutes per day - enough time to perform an additional study per day.
The extra time could also be used on a per sonographer basis in supporting the teaching role the sonographers perform. An additional minute per study per day is enough to allow the sonographer to expand teaching without affecting overall productivity.
The cost of this time, based on the average hourly labor rate for our sonographers ($47.90) is an additional $63.07 per sonographer per day.
|
| |
| Conclusion: |
| While the amount of time spent by sonographers using the RIS has decreased from the previous 3-5 minutes per sonographer per patient, the lack of an automated, or simplified, method of completing an examination means that sonographer interaction with the RIS persists. Though the recovered time is genrally too little for a single sonographer to perform an additional study, the sharing of examination tasks and the incorporation of teaching during the study are two situations in which the recovered time could be used in a productive manner. |
| |
| References: |
| Horii SC, Feingold E, Coleman B, et al. Use of a miniPACS for printing in ultrasound: potential for improved productivity. PACS: Design and Evaluation. Proc. SPIE. 1995;2435:257-262. |
| |
|
|
| |
| |
|
|