RESUMO Objetivo Avaliar a percepção dos profissionais de saúde sobre o clima e a cultura de segurança do paciente em Unidades de Terapia Intensiva (UTI) e a relação entre os instrumentos Hospital Survey on Patient Safety Culture (HSOPSC) e o Safety Attitudes Questionnaire (SAQ). Método Estudo transversal realizado em hospital de ensino no interior do estado de São Paulo, Brasil, em março/abril de 2014. Aplicaram-se o HSOPSC, o SAQ e um instrumento para levantamento das informações sociodemográficas e profissionais aos funcionários das UTI adulto, pediátrica e neonatal. A análise utilizou a estatística descritiva. Resultados As escalas apresentaram boa confiabilidade. Maiores fragilidades para a segurança do paciente foram observadas nos domínios “condições de trabalho” e “percepções da gerência” do SAQ e “resposta não punitiva aos erros” do HSOPSC. As fortalezas no SAQ foram o “clima de trabalho em equipe” e a “satisfação no trabalho” e para o HSOPSC “expectativas e ações de promoção de segurança supervisores/gerentes” e “aprendizado organizacional e melhoria mútua”. Na UTI Neonatal houve maior satisfação no trabalho do que nas demais UTI. A UTI Adulto apresentou menores pontuações para a maioria dos domínios do SAQ e HSOPSC. A correlação entre as escalas foi de força moderada (r=0,66). Conclusão Há diferenças de percepções quanto à segurança do paciente entre as UTI, o que corrobora com a existência de microculturas locais. O estudo não demonstra que o SAQ e o HSOPSC sejam equivalentes.

This study investigated the socio-demographic profile, clinical procedures and etiology of nosocomial infection associated with deaths in the Hospital Estadual Sumaré, state of São Paulo, Brazil, from 2007 to 2008. The retrospective study of medical records (n = 133) revealed an average of 35 days of hospitalization. Most patients (97%) underwent some invasive procedure associated with nosocomial infection (p ≤ 0.05), including: 90 (67.7%) pneumonia, 62 (46.6%), urinary infections and 97 (73%) septicemia. Infection was the leading cause of death in 75 (56.4%) cases, with defined etiology in 110 (82.7%); 34 (30.9%) because of microorganisms that were multidrug-resistant. The most common was Staphylococcus aureus (25%), related to pneumonia and blood stream infection. The monitoring of hospital infection contributed to intervention at risk situation and death.

Introduction: Patient safety has become a formal concern in several health systems in the world, in the last decades. In 2004 the World Health Organization (WHO) proposes the Alliance for patient safety and aims safety culture evaluation in healthcare institutions as one of the key aspects to this process. Method: Crosssectional quantitative research approach, performed in a teaching hospital in São Paulo State between the months of March and April 2014. The study population was composed of all the professional who were part of the work schedule of intensive care unit (ICU) adult, pediatric and neonatal and did not fit the exclusion criteria (less than six months in the unit). Two instruments for assessing the culture environment and patient safety, the Hospital Survey on Patient Safety (HSOPSC) the Safety Attitudes Questionnaire (SAQ), and an instrument for survey of demographic and professional information were applied. For data analysis, the test of reliability of the scales by Cronbach's alpha was used. The presence of associations of scales with study variables was checked by Pearson's chi-square test or Fisher's exact test in the qualitative variables, the ANOVA for quantitative variables. The presence of correlation between the SAQ and the HPSOPSC instruments was tested by Pearson correlation test. Result: sociodemographic data regarding gender and age and position were homogenous in the three ICUs. Professional of the Neonatal ICU had worked longer time in this unit and specialty when compared to other units. Both scales showed good reliability by Cronbach's alpha, 0.853 for SAQ and 0.889 for HSPOSC. In the analysis of the SAQ domains, it was observed score ≤ 62 for "Working Conditions and Perceptions of Management", while for HSPOSC dimension "Non-punitive Response to Error" had the lowest percentage of positive responses (29.6%), the dimension "Open Communication" and "Return of Communication and Information on the Error" a proportion of neutral responses more than 30%. The total score of patient safety by HSPOSC was 85% (summed up great and very good). Analyzing the behavior of ICUs through each scale, Neonatal ICU had higher job satisfaction than the other ICUs. Adult ICU had lower scores in each domain compared to other domains and for HSPOSC only the area "Open Communication" obtained the proportion of positive responses slightly superior to the other ICUs. The correlation between the scales through Pearson correlation was of moderate strength (Pearson correlation coefficient of 0.656). The open responses showed that changes in hospital as a result of accreditation processes, contributed to a better perception of professionals about patient safety. Conclusions: There are differences in perceptions of patient safety among ICUs within the same hospital, which corroborates the existence of local microcultures. Rating scales of climate/culture of patient safety seems to measure similar phenomena.