Emergency Radiology

, Volume 12, Issue 4, pp 158–159

Effect of digital radiography on emergency department radiographic examinations


    • Michigan State University–Grand Rapids Medical Education and Research Center, Radiology Residency
  • Joseph Junewick
    • Michigan State University–Grand Rapids Medical Education and Research Center, Radiology Residency
    • Department of Radiology, Spectrum Health
    • Advanced Radiology Services, P.C.
  • Charles Luttenton
    • Michigan State University–Grand Rapids Medical Education and Research Center, Radiology Residency
    • Department of Radiology, Spectrum Health
    • Advanced Radiology Services, P.C.
Original Article

DOI: 10.1007/s10140-006-0465-0

Cite this article as:
Lee, B., Junewick, J. & Luttenton, C. Emerg Radiol (2006) 12: 158. doi:10.1007/s10140-006-0465-0


The purpose of this paper is to determine the time-savings realized as a result of conversion from film-based radiography to digital radiography (DR) in emergency department patients receiving multiple radiographic examinations. We retrospectively reviewed the computer records of 69 patients from the preconversion group (January 2004) and 62 patients from the postconversion group (January 2005), calculated the mean examination time per image from each group, and used a one-tailed t test to compare the two means. The preconversion (a mix of screen-film and computed radiography) group had a mean examination time of 3.27±1.50 min per image. The postconversion (DR) group had a mean examination time of 2.51±1.16 min. The p value between the two means is 0.000822 with a standard error of 0.330 min. DR reduces radiographic examination time by 23% when compared to conventional screen-film radiography.


Digital radiographyEmergency department patients


Radiography is currently going through a digital revolution. In conventional screen-film radiography, obtaining an image requires exposing and developing films before the images can be reviewed for technical adequacy by the technologists and interpreted by the radiologists. The speed of this process is limited by the speed of the film processing and takes approximately 2 min for each film. In digital radiography (DR), electronic detectors capture radiographic images—either directly with amorphous silicon thin-film transistors or indirectly through a combination of intensifying screen and photodetectors—and can transmit the results for viewing by technologists and radiologists in seconds [1]. DR has the additional benefit over traditional screen-film radiography in that images produced by DR can easily be incorporated into a picture archival and communication system. In such an environment, the current study, prior comparison studies, and clinical information can be available to both treating physicians and interpreting radiologists [2, 3], which in turn allows quicker interpretation and timely implementation of treatment plans in conditions such as occult fractures, abdominal diseases, and pneumonia.

Our institution’s emergency department radiography equipment was converted from screen-film to DR equipment (Kodak Direct View 9000, Eastman Kodak, Rochester, NY, USA) in October 2004. Decreasing the time patients spend in radiology examination rooms was a driving force behind the conversion to DR. In this study, we investigate if, and by how much, DR reduces radiographic examination time in complex emergency room patients undergoing three or more radiographic examinations.

Materials and methods

This retrospective case-review study was approved by our institutional review board. Based on billing records, we randomly selected patients with three or more radiographic examinations performed on the same date in the Emergency Department of our institution, and we looked at their radiographic examination start and end times as recorded by technologists in the hospital information system (Cerner Millennium, Cerner, Kansas City, MO, USA). Sixty-nine patients from January 2004 were selected and placed in the preconversion group, which underwent a mixture of conventional screen-film and computed radiography (CR) examinations. Computer records for seven patients in the group were invalid (same start and end times for the exam), yielding a total of 62 entries.

Based on the same billing record criteria, another 126 patients from January 2005 were selected and placed in the postconversion group. Of these, the entries for 58 patients were invalid because these entries were CR examinations or had the same start and end exam times, yielding a total of 68 entries. For each entry, the average time spent per image was derived by dividing the total time (end time minus start time) by the number of images. We used a one-tailed t test to compare the means of the average time per image of the two groups.


Figure 1 is a boxplot of both groups displayed together. In the preconversion group, the average time per image (both film and CR) is 3.27 min with a standard deviation of 1.50 min. In the postconversion group, the average time per image for DR is 2.51 min with a standard deviation of 1.16 min. The p value for the two groups is 0.000822. The reduction in time from pre- to postconversion is 0.76 min (standard error 0.330 min) or 23%.
Fig. 1

Boxplot of time per image in the preconversion and postconversion groups


We chose patients with three or more radiographic studies based on the assumption that their conditions may be more complicated and thus have more to gain by timely implementation of their treatment plans. By using data from the same time period of the year, we tried to minimize any difference in the types of radiographic examinations ordered, e.g., chest radiographs for pneumonia or extremity radiographs for similar types of trauma (slipping on ice, car accidents).

We chose January 2005, 3 months after the installation of the equipment, as our study month in an attempt to minimize the effect of any operational errors by technologists due to unfamiliarity with the new equipment. However, based on anecdotal information from the technologists, the learning curve for the new DR equipment was steeper than we anticipated. The computer records did not include how many DR images were repeated for individual exams. The only information available to us was the total repeated images for the whole month of January. Our internal audit showed that the repeat rate for DR was 1,507 out of 12,124 images in January 2005, or 12.4%. In comparison, the repeat rate for CR was 537 out of 7,753 images, or 6.9%, during the same month. The implication of these data is twofold: the learning curve remained steep at 3 months after installation of the new DR equipment and that, despite the higher repeat rate, DR still showed a 23% reduction in time. If the repeat rate was lower, examination time may have been further reduced.

The technologists record the start and end times of each examination. Therefore, examination time depends on the accuracy of their records. We realize that the technologists may not record the start and end times precisely, but we assumed that the bias would be systematic and remain the same for CR/screen-film and DR examinations.

The reduction of time required to perform radiographic examinations in the emergency radiology department translates into decreased time for interpretation, quicker implementation of treatment, and a decreased patient length of stay. The overall result is improved patient care. The reduction in examination time with DR may also have a significant impact on the planning of conversion from screen-film-based equipment. The reduction of time translates into increased patient throughput. This in turn means a reduction in the number of examination rooms needed for the same output. In our experience, four general diagnostic screen-film radiography units could be replaced by three DR units in the emergency department setting.


DR reduces radiographic examination time by 23% when compared to conventional screen-film radiography.

Copyright information

© Am Soc Emergency Radiol 2006