IJCS | Volume 33, Nº2, March / April 2020

153 Figure 1 - Performance of the groups evaluated for the median number of correct answers in the questionnaire. Group A corresponds to the doctors who had taken an ACLS course; Group B corresponds to those who had not (p = 0.06). Number of right questions Catolino & Knofholz Medical behavior before and after advanced cardiac life support (ACLS) course Int J Cardiovasc Sci. 2020; 33(2):151-157 Original Article performance (OR = 6.75, 95% CI, 1.1 < OR < 41.0, p = 0.04) (Table 1). Of the 20 physicians who were part of Group A, only 2 scored less than 3 questions, so 90.00% of the sample answered correctly the majority of the questionnaire. In comparison, of the 14 professionals fromGroup B, 6 (42.86%) scored less than half of the test, while only 57.14% made 3 or more questions. To investigate whether the time since the ACLS course to the date of the questionnaire could affect the performance of the volunteers, GroupAwas subdivided into two parts, according to the year of the course: (i) volunteers who took theACLS course in the year of 2016; (ii) volunteers who took the ACLS course prior to the year of 2016 (Figure 2). None of the volunteers took the course during the year of 2017. No significant difference was noticed between the groups (p = 0.52). In each group, only 1 individual who took the course scored less than 3 questions. Thus, it could be observed that even physicians who were trained with outdated guidelines were able to achieve similar performance to that of recently trained professionals, and better than group B (p = 0.04). Discussion A questionnaire was distributed to physicians on duty at emergency units of hospitals in Curitiba, where 34 professionals were volunteers. Of these, 20 physicians had taken an ACLS course (group A), and 14 had not (group B). Regarding the number of correct answers in the questionnaire, it was observed that group A obtained median score greater than Group B. However, this difference was not significant (p = 0.06). It is believed that this tendency would be confirmed by an increase in the sample number, which was not possible in this study, since it was difficult to recruit volunteers who fulfilled the inclusion criteria (physicians on duty in emergency departments of hospitals in the city of Curitiba during the data collection period). The researchers chose to look for volunteer physicians working at emergency care units to participate in the study because they believed that in that environment, cardiac arrest situations were found more frequently than in other medical settings, such as outpatient clinics, for example. Therefore, it would be indispensable for the professional to be knowledgeable about the management of CPR for quality emergency care. However, one limitation to sample size was the typical agitation of emergency room units, as well as physicians’ allegations of insufficient time to answer the questionnaire questions with the utmost attention. Still, to avoid consultation or communication between professionals, they were required to answer the questionnaire in the presence of a researcher in charge, which also reduced the sample.