Daniel Durand MD, Abstract Co-Author: Nothing to Disclose

Mahadevappa Mahesh MS, PhD, Abstract Co-Author: Nothing to Disclose

Daniel J. Brotman MD, Abstract Co-Author: Nothing to Disclose

John A. Carrino MD, MPH, Presenter: Research Grant, Siemens AG Research Grant, Carestream Health, Inc Research Consultant, General Electric Company

Leonard Feldman MD, Abstract Co-Author: Nothing to Disclose

Radiation dose is increasingly available to non-radiology providers, but there is no standard method for communicating this information. We surveyed medicine and pediatric residents to determine if expressing the same quantities in different terms resulted in consistent under- or overestimation of dose.

Survey: Questions 1 and 2 dealt with PGY year and radiation training. In questions 3 through 8, different quantities of radiation dose were expressed in different terms and respondents were forced to either rank the quantities from least to greatest or indicate all were equal. Statistics: "Rank discrepancy" was defined as the difference between the correct rank and the rank given. The order of terms was randomized and balanced so that the maximum mean discrepancy was +2 and the minimum -2. 95% confidence intervals for mean rank discrepancy were calculated and expressed in terms of % maximum. In addition, the frequency of correct rankings for each term was calculated and compared to the others by the Chi-square test with Yates correction.

51 medicine residents and 45 pediatric residents responded (44% response rate). Respondents were PGY-1 (38%), 2 (29%) or 3 (33%). Most (80%) reported <10 hours radiation training, 14% reported no training and 6% reported 11 to 50 hours. Confidence intervals for mean rank discrepancy (Figure 1) show the degree to which each term was over- or underestimated relative to the others. When analyzed in terms of frequency of correct ranking, the ACR Relative Radiation Level (ACR RRL) performed best (57.4%) followed by liftetime cancer risk (44.9%), chest x-rays (36.7%), cross country flights (30.1%) and natural background radiation (27.5%). Among these, only ACR RRL (p<0.0001) and lifetime cancer risk (p=0.0128) were significantly more likely to be ranked correctly.

Both groups significantly underestimated dose quantities when expressed in terms of natural background, minimally overestimated them when expressed in terms of chest x-rays or ACR RRL and were neutral towards cancer risk and cross country flights. At the same time, dose quantities were correctly ranked most often when expressed in terms of ACR RRL or cancer risk.

The language used to communicate radiation dose impacts physician perception of dose quantity, highlighting the need for consistency and standards in communicating this information.

Durand, D, Mahesh, M, Brotman, D, Carrino, J, Feldman, L, Word Choice Impacts Dose Quantity Perception Amongst Referring Physicians.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12027525.html