ABSTRACT CONTEXT AND OBJECTIVE: Choosing a medical specialty and making decisions concerning a career are difficult processes for medical students and newly graduated physicians.This exploratory study aimed to investigate the influence of role models on the choice of surgery as a career, and to determine the most influential model characteristics. DESIGN AND SETTING: Qualitative analysis on responses to a self-administered questionnaire, in different teaching-learning settings. METHODS: Residents from all years of various surgical subspecialties in a university hospital were included in a survey about the factors that determined their choice of surgery. The questions included items on whether a role model had influenced them in choosing surgery, and the personal or professional characteristics of the models that had been most influential. The responses were subjected to qualitative content analysis. RESULTS: Sixty-four out of 96 medical residents participated. Fifty-three residents (82.8%) acknowledged the influence of role models. Sixteen model characteristics were indicated as important, with 136 mentions. Characteristics classified as technical skills (55%), such as"medical knowledge"and "manual dexterity" predominated over humanistic characteristics (35%), such as "patient-physician relationships" and "ethical behavior". However, this difference was not statistically significant (Fisher test, P = 0.11).There were no age differences regarding the proportions mentioning "technical" and "non-technical" attributes, but female residents mentioned significantly more technical skills than their male colleagues did. CONCLUSIONS: The influence of role models seems to be an important factor determining the choice of surgery as a career. The influential characteristics of the models include not only technical but also humanistic qualities.

The objective of this study was to evaluate clinical characteristics, etiology, and resistance to antimicrobial agents, among patients with ventilator-associated pneumonia (VAP). A case study vs. patients control under mechanical ventilation and hospitalized into clinical-surgical adults ICU of HC-UFU was performed from March/2005 to March/2006. Patients under ventilation for over 48 h were included in the study including 84 with diagnosis of VAP, and 191 without VAP (control group). Laboratory diagnosis was carried out through quantitative microbiological evaluation of tracheal aspirate. The identification of pathogens was performed by classical microbiological tests, and the antibiotics sensitivity spectrum was determined through the CLSI technique. VAP incidence rate over 1,000 days of ventilation was 24.59. The mean (± SD) duration of mechanical ventilation prior to VAP diagnosis was 23.2 ± 17.2 days. By multivariate analysis the risk factors predisposing for VAP were: mechanical ventilation time and mechanical ventilation > seven days, tracheostomy and use of > three antibiotics. Mortality rate was high (32.1 %) but lower than that of the control group (46.5%). Major pathogens were identified in most of patients (95.2%) and included: Pseudomonas aeruginosa (29%), Staphylococcus aureus (26%), Enterobacter/Klebsiella/Serratia (19%) and Acinetobacter spp. (18%), with expressive frequencies of P. aeruginosa (52%), S. aureus (65.4%) and Enterobacteriaceae (43.7%) resistant to imipenem, oxacillin and 3/4 generation cephalosporins, respectively. In conclusion, our observation showed VAPs caused by multiresistant microorganisms, the prescription of > three antibiotics, and mortality with unacceptably high rates. The practice of de-escalation therapy appears to be urgently needed in order to improve the situation.