33 Physician and Nurse Perceptions of Non-UrgentCommunication in the Emergency Department

Morgan SR, Kawar M, Rahman S, Gatewood JS, Fairbanks III RJ/GeorgetownUniversity Hospital/Washington Hospital Center, Washington, DC; GeorgetownUniversity School of Medicine, Washington, DC; Mercer University School ofMedicine, Macon, GA; Washington Hospital Center, Washington, DC; NationalCenter for Human Factors Engineering in Healthcare, MedStar Institute forInnovation, Washington, DC

Study Objective: Accurate communication affects both the work environment inthe emergency department (ED) and patient safety. Although prior studies haveexamined communication and interruptions in the ED, few focus on non-urgentcommunication or compare nurse/physician perceptions.

Methods: Non-urgent communication was defined as that which did not relate totime-sensitive issues or unstable patients. Phase I was a qualitative analysis of 1-on-1interviews with 14 ED staff (4 attendings, 4 residents, and 6 nurses) at an urban,academic ED using a set of open-ended questions on experiences with non-urgentcommunication. Phase II involved an electronic survey of staff in the same ED, with5 demographic questions and 16 others on non-urgent communication strategies,interruptions, and possible improvements. Responses were collected anonymouslyand analyzed with descriptive statistics.

Results: 73/131 (55%) subjects responded: 21 nurses, 20 residents, 26 attendings(6 did not report role). A plurality of respondents (38%) felt the quality of non-urgent communication was average. The most common barrier reported was aninability to find others, with 40% stating this happens 5 to 10 times per shift.Opinions on other barriers varied by groups; physicians expressed difficulty withnurses’ names, and nurses often did not know a physician’s level of training.Forgetting to convey a message or giving up altogether occurred at least once a shiftaccording to 65% of residents, 58% of attendings, and 19% of nurses. Allrespondents stated that they thanked others more than they themselves were thankedand reported the inverse for negative comments. Frequency of perceived interruptionsper hour were reported as 6 for attendings, 1 to 3 for residents, and less than 1 forRNs. Half of respondents (51%) reported negative outcomes such as delays, near-misses, and errors on “some” shifts while 3% estimated this occurred more than 5times per shift. When asked for examples of inappropriate interruptions, the mostcommon response was during sign out or phone calls, and the least cited was duringdocumentation or nurses obtaining medicines. Respondents across all groupsidentified introductions at each shift change as offering the best potential benefit forcommunication. Among nurses, there was also support for discussing initial plans foreach patient, and physicians articulated interest in a staff directory with names andpictures.

Conclusions: Nurses and physicians share many similarities in their perception ofnon-urgent communication. All staff expressed difficulty finding others, while otherbarriers varied by group. Rates of interruptions also differed by group and wereconsistent with rates reported in prior studies. Overall, there is much informationthat staff feel they do not convey. The question arises whether this representsimportant information being lost or simply lower priority information that is nevercommunicated due to competing higher priority interests. Interestingly, interruptionsduring reports or phone calls were more commonly viewed as inappropriate thanthose while obtaining meds, communicating with patients, or performingdocumentation--activities likely to affect safety and patient satisfaction. Finally,nurses and physicians feel simple changes such as posting staff names and pictures orhaving formal introductions could improve non-urgent communication.

34 Lack of Sex Disparity in Cardiovascular Testing AfterCoronary CTA

Ginty CT, Chang A, Matsuura AC, Walsh KM, Le J, Decker C, Green M, HollanderJE/University of Pennsylvania, Philadelphia, PA

Study Objective: It is known that there is a sex disparity in cardiovascular care inthis country; it is unclear how much of this can be accounted for by differences inpresentation and risk. We assessed whether there was a sex disparity in testing ofpatient after knowledge of the coronary anatomy was determined with coronarycomputerized tomographic angiography (CTA) for ED patients with potential acutecoronary syndromes (acute coronary syndrome). In theory, once the coronaryanatomy has been determined by coronary CTA, any disparity in subsequent workupshould not be the result of differences in presentation. Hypothesis: No difference incardiac test utilization between sexes found to have similar amount of disease oncoronary CTA.

Methods: Study Design: Prospective cohort study. Setting: University hospital.Patients: ED patients with potential acute coronary syndrome who received a

coronary CTA. Data: Demographics, history, cardiac risk factors, follow-up testingand procedures. Follow-up: at 30 days by structured record review & telephone.Main Outcome: Cardiac tests (stress test or catheterization) within 30 days. Analysis:Patients stratified by sex and coronary CTA results (max stenosis: none, 1-24%, 25-49%, 50-69%, �69%). Outcomes reported as RR and 95% CI.

Results: During the study period, 1187 patients received a coronary CTA (meanage, 47.8 � 8.7 years, 55% female and 64% black), of which 1144 had % maxstenosis available. Overall, men were more likely to receive further testing (RR 1.51;95% CI, 1.14-1.99 %). When stratified by % stenosis, men were not more likely toreceive further testing 30 days after coronary CTA compared to women (aRR 1.14,95% CI 0.68-1.91).(see table)

Conclusion: Male patients with potential acute coronary syndrome who receive acoronary CTA as a part of their ED evaluation were no more likely than females toreceive follow-up testing at 30-day follow-up when stratified by coronary CTAresults.

35 Comparison of Cardiac Risk Scores in EmergencyDepartment Patients With Potential Acute CoronarySyndrome

Lee B, Chang A, Matsuura AC, Marcoon S, Hollander JE/University ofPennsylvania, Philadelphia, PA

Study Objective: The ability to risk stratify patients presenting to the ED withpotential acute coronary syndrome is critical. Several risk scores for patients withdefinite acute coronary syndrome have been developed, but only the TIMI risk scorehas been shown to risk stratify ED patients with potential acute coronary syndrome.We compared the prognostic value of the GRACE and PURSUIT risk scores to theTIMI score in the broader ED patient population presenting with potential acutecoronary syndrome.

Methods: We performed a secondary analysis of a prospective cohort study thatenrolled patients who presented to the ED with potential acute coronary syndrome.Demographics, history, and components of the TIMI, GRACE and PURSUIT scoreswere obtained. Follow-up was conducted by structured record review and phone. Themain outcome was a composite of 30-day death, nonfatal acute myocardial infarction,and revascularization. The GRACE Scores ranged from 0-330 and PURSUIT scoresranged from 0-18 and were subsequently divided into 8 equivalent strata tocorrespond to TIMI score range and facilitate comparison. For each of the 3 riskscores, receiver operating characteristic curves were used to compare prediction of30-day event rates.

Results: There were 4743 patients enrolled (mean age, 52.5 �/- 13.3; 56%female; 65% Black). By 30-days, there were 59 deaths, 172 acute myocardialinfarction, and 175 revascularizations. The AUC for TIMI was 0.757 (95% CI;0.728-0.785); GRACE 0.728 (95% CI; 0.701-0.755), and PURSUIT 0.691 (95%CI; 0.662-0.720).

Conclusion: In this large cohort of ED patients, the TIMI score had the bestdiscriminatory ability to predict 30-day cardiovascular events.

Research Forum Abstracts

Volume , . : October Annals of Emergency Medicine S189