IJCS | Volume 33, Nº2, March / April 2020

169 Novaes Neto & Freitas Factors associated to knowledge of cardiac arrest Int J Cardiovasc Sci. 2020; 33(2):167-174 Original Article the descriptive measures of centrality and dispersion were calculated. In bivariate analysis, the prevalence ratios with their respective confidence intervals, at the 95% level, were estimated for the crossing between level of knowledge and the other variables related to sociodemographic characteristics, occupational characteristics and health professional’s training characteristics on CA. To perform these two analyses, the application software Statistical Package for the Social Sciences version 22.0 was used. Pearson’s Chi-square test (X²) or Fisher’s exact test was used in the selection of variables for this multivariate analysis. The level of significance for entry in the multivariate model was p ≤ 0.10. To verify the factors associated with health professionals’ knowledge, the multivariate logistic regression model was used, one of the most important tools in the statistical analysis of data when one intends to predict the occurrence of a given event as a function of a set of variables. Applying the logistic regression model is a form of statistical modeling suitable for categorical and usually dichotomous response variables. 7 The dependent or response variablewas the knowledge of professionals and the independent variables were sociodemographic, occupational and training variables that were included in themodel. For purposes of analysis, the continuous variables that entered the model were recategorized based on the theoretical framework. After defining the final multivariate model, the results obtained in Odds Ratio were converted to Prevalence Ratio using Poisson’s Regression. After adjustment of the model with the “maximum likelihood” estimator, the significance of the model variables was evaluated through Wald’s Test. In order to evaluate the accuracy and goodness-of-fit of the logistic model, the Hosmer- Lemeshow 7 test was used. The level of statistical significance adopted was 5%. For this analysis, the application software Data Analysis and Statistical Software version 12.0 was used. Results Construction and evaluation of content validity of the KCACPRQ The questionnaire submitted for evaluation of the judges presented 22 questions on Cardiac Arrest and Cardiopulmonary Resuscitation. The judges pointed out that eight of these questions lacked clarity, which required changes to achieve comprehensibility. The content was considered adequate in 21 questions, and only one itemwas considered inadequate. The pertinence of the items in their respective categories was considered adequate by the judges. The analysis of the feedback of the questions resulted in seven items with divergent templates, and changes were made to meet this criterion. The judges pointed out two items with confusing formulations and with content contemplated in other items, and as such, they were withdrawn from the instrument in order not to harm the knowledge determination of the participating professionals. Ten items had changed the wording of the statement and the organization of the alternatives. Thus, the final version of the questionnaire has 20 questions about cardiac arrest and cardiopulmonary resuscitation (Chart 1). The pre-test showed that the professionals answered the instrument without difficulties and considered the instrument adequate. Regarding the practicality of the instrument, Table 1 presents its main results. Profile of health professionals The study included 100 health professionals, of whom 50%were nursing technicians, 32% as nurses and 18% as doctors. Among the professionals, it was observed that 78%were women. Approximately half, 50.5%, were over 35 years of age. The most prevalent self-reported skin color was black or brown (72%). With regard to schooling, 34% had a graduate degree. The most mentionedmarital status was not having a partner (52%). More than half (57%) reported having children. Regarding occupational characteristics, the majority (83%) of professionals worked in closed units. The most prevalent work time was over 05 years (76.8%). A large proportion (61.6%) of the professionals worked full time. About 56.6% of the participants reported a different employment relationship. The most prevalent work shift was the on-call duty (68%). With regard to the working day, 37.9% worked with a weekly workload between 30 and 40 hours. Regarding training characteristics, the majority (59%) of professionals had up to 10 years of training. Regarding training in CA/CPR, 80.8% of the professionals reported having received it. Training in CA/CPR at work was reported by 57.6% of participants. The time interval that received training in the most frequent CA/PCR was 06 months ago (44.9%). Regarding the time of contact with CA/CPR, the majority (69.2%) reported it as being