Patient safety culture in surgical centers : nursing perspectives *

* Extracted from the dissertation: “Cultura de segurança em centro cirúrgico: perspectiva da equipe de enfermagem”, Programa de Pós-Graduação em Enfermagem, Universidade Federal do Piauí, 2017. 1 Universidade Federal do Piauí, Programa de Pós-Graduação em Enfermagem, Teresina, PI, Brazil. 2 Universidade Estadual do Ceará, Programa de Pós-Graduação em Cuidados Clínicos em Enfermagem e Saúde, Fortaleza, CE, Brazil. 3 Universidade Federal do Piauí, Programa de Pós-Graduação em Ciências e Saúde, Teresina, PI, Brazil. ABSTRACT Objective: To analyze patient safety culture in the different spheres of management in the perspective of the nursing team providing services in surgical centers. Method: Cross-sectional study with 200 nursing professionals, in three surgical centers of Piauí state – one municipal, one federal, and one state – from January to August 2016 through the application of the instrument Hospital Survey on Patient Safety Culture. Results: The first surgical center did not present a strengthened patient safety culture; the others presented the dimension “Organizational learning – continuous improvement” (80.6%/75.6%) and “Frequency of reported events” (76.2%) as strengthened areas. In the first, the safety score “average” was prevalent, whereas participants of the second and third judged patient safety as “very good”. Most participants of the three surgical centers (80.0%) reported no adverse event in the previous 12 months. Conclusion: The state and federal surgical centers obtained the best patient safety scores when compared to the municipal surgical center. Therefore, for a more effective and safe care, the strengthened dimensions in each type of management should be enhanced and the weakened ones should be improved.


INTRODUCTION
Patient safety issues became, over the last decades, a priority health issue worldwide, being characterized as one of the main goals of care-related institutions. However, even with the preoccupations around safety policies to improve quality of care, the risks and the occurrence of adverse events have been increasing substantially, mainly in the hospital environment.
Care-related adverse events are important causes of morbidity and mortality around the world and has considerable impact over the health area, leading to damages not only for the patient but also for professionals, who suffer ethical and moral damage, in addition to damage in the professionalpatient interaction. For institutions, these also lead to cost increase, reduced institution reliability, and moral and organizational damage (1) .
A study conducted in 3,720 Japanese hospitals over three years has estimated 1,326 to 1,433 deaths caused by adverse events over one year and a mortality rate ranging from 8.81 to 9.52 cases per 100,000 patients (2) . A study in Ireland has demonstrated that the prevalence of adverse events was 12.2% and that more than 70% of the events were considered avoidable (3) . In Brazil, data from the Hospital Information System of the Unified Health System have registered 31,774 incidents in 2015, with an estimate that approximately 93% of them took place in the hospital environment (4) .
Given this context, topics related to patient safety must be urgently disseminated, since they pose a public health issue. Therefore, to implement safety strategies, health organizations must adopt a model of "safety culture, defined as a product of values, attitudes, competencies, and individual and collective behavioral patterns, which determine commitment, style, and proficiency in the administration of a safe and healthy organization (5)(6) ".
The promotion of safety culture is one of the bases of the patient safety movement. With the objective of understanding the organizational factors which may influence and cause damage and adverse events to patients in health institutions, strategies for safety culture assessment have been stimulated, enabling these studies' results to help with improvement regarding the quality of processes on safety of care provided to the patients (7) .
The Surgical Center (SC) is, in this context, an environment demanding a working dynamic which differs from that of other services. In perioperative care, the possibility of errors and adverse events related to medication and transfusions, accidents with the retention of foreign bodies inside the patient, falls, burns due to inappropriate use of equipment, and mistakes in patient identification, safety complications associated to anesthesia and surgery in a wrong site or patient -a fact which demands more attention from the team involved with patient care in this sector -are emphasized (8) .
There are many professions which provide care in the surgical environment, and nursing is noteworthy in the development of a large share of actions related to population health care, since it is in constant contact with the patient. It is also directly related to the conduction of procedures which may be associated to the occurrence of healthcare practice errors (9)(10)(11)(12) .
The development of research related to this theme is relevant, since it closes knowledge gaps and sensibilizes professionals on the importance of patient safety in the SC, aiming at reducing risks and damage in care provided in this environment. To guide the investigation, the following research question was chosen: what is the patient safety culture in the perspective of the nursing team working at a SC? Therefore, the objective was to analyze patient safety culture at a SC in different spheres of management from the perspective of the nursing team.

Design of stuDy
This is a cross-sectional study developed between February and August 2016 in three SC of reference public hospitals in the state of Piauí. Each SC is inserted into a hospital linked to different governmental spheres. These collection places have been chosen due to being reference hospitals in Piauí, Brazil, and belonging to different governmental spheres. The possibility of dominating aspects in each level of organization, complexity, and type of care structure, in addition to singularities regarding the flow of patients in these hospitals, has instigated the investigation in different contexts.
Surgical Center 1 (SC1) is in a municipal hospital, which provides urgency and emergency care to the regions and is estimated to perform around 50 surgeries daily. Surgical Center 2 (SC 2) is inserted into a federal institution which receives patients regulated according to their bed availability, with a mean of 14 surgeries per day in diverse specialties. Surgical Center 3 (SC 3) is in a general state base and teaching hospital, administered by the State Health Office, with circa 37 surgeries per day. The hospitals in which the SC are inserted are emphasized to perform the safe surgery checklist and are undergoing a process of accreditation.

PoPulation
The population comprised the members of the nursing team who performed activities in the three SC. The SC1 had 145 professionals (15 nurses and 130 technicians and assistants); out of these, 15 were on leave and 12 on vacation. The SC2 had 62 professionals (12 nurses and 50 technicians), out of which three were on leave and one on vacation. In SC3, 87 professionals (15 nurses and 72 technicians and assistants) and three of them were on leave and nine on vacation.

selection criteria
The inclusion criteria were being a nurse, technician or nursing assistant working for at least six months in the SC of the institution, with a minimum weekly workload of 20 hours. Professionals who were not part of the nursing team, team professionals not in the sector due to vacation, health leave, maternity leave, bonus leave, and/or medical certificate and with instruments less than half filled were excluded. The sample was selected by convenience. The professionals who accepted participation amounted to 203: 92 in SC1, 47 in SC2, and 64 in SC3. Out of these, three instruments were excluded due to inappropriate filling, totaling a sample of 200 professionals.

Data collection
To conduct this research, an instrument was used to collect sociodemographic data and Hospital Survey on Patient Safety Culture (HSOPSC), created in 2004 by the Agency for Healthcare Research and Quality (AHRQ) and broadly used worldwide, was opted for. The version used in this study was translated and culturally adapted to the Portuguese language (6) . The HSOPSC enables evaluating the patient culture from the perspective of professionals through 42 questions grouped into 12 dimensions. The first seven approach aspects within the unit; the other three, within the hospital environment; and the last two variables are result variables.
The instrument uses the Likert scale, with attributions which vary from "completely disagree" to "completely agree". The excluded instruments were those left completely blank, had only answers for the sociodemographic questions or had the same answer for all questions, since some items are negatively formulated and the same answer for all items indicates that the interviewee was probably not paying attention and the answers were invalid.
The collection was performed simultaneously in the three SC. The instrument was filled by the participants themselves supervised by the researcher to answer questions which might emerge during this step. The instrument was provided to the participating professional during work hours and its return was expected until the end of the corresponding shift. In case they were unavailable to fill the instrument during their shift, the return was scheduled, or, if necessary, they were approached the day after to reduce losses.
"The variables related to the level of patient safety culture were the global safety score provided by the interviewees and the percentage of positive answers to the questions related to each dimension of the patient safety culture. To estimate the influence of the type of management and the characteristics of the workers, the possible predictive variables of the patient safety culture level were position/function, years working in the hospital, age group, and type of hospital management (federal, state, or municipal)" (13) .

Data analysis
After collection, data were organized in Excel and, after validation, transferred to the software IBM Statistical Package for the Social Sciences (SPSS) for Windows 20,0, in which a descriptive statistics study was performed.
A significance level p ≤ 0.05 was adopted for a 95% confidence interval among the data.
For data analysis and interpretation, the methodology of the original instrument AHRQ was proposed; it has both questions formulated positively and negatively, which were recodified. Assessment of each dimension is performed from the percentage of positive answers obtained by the calculation of the combination of two higher answer categories. Higher percentages indicate positive attitudes toward safety culture (14) .
For questions formulated positively, percentages of positive answers under 50% represent weak spots, problematic areas, or critical aspects. The results between 50% and 75% were considered in this study as non-problematic points, and over 75%, as strong points (5) .

ethical asPects
This study was approved in 2015 by the Research Ethics Committee of Universidade Federal do Piauí, in opinion. 1,238,858, and by the three hospitals in which the research was performed, meeting the requirements by the National Health Council concerning the conduction of research involving human beings, according to Resolution No. 466/12.

RESULTS
The study participants were 200 nursing professionals, including nurses, technicians, and nursing assistants; SC1 had the participation of 63.4% of the 145 nursing professionals of this SC, the SC2 has amounted to 74.2% of the 62 participants, and SC3 had 71.3% of the 87 participants of this SC.
Regarding sociodemographic characteristics, females were predominant, with a mean of 90% in the three SC. In SC1 and SC2, the age ranged from 31 and 40 years and in SC3, from 24 to 71 years old, with a mean and standard deviation of 39.5 ± 11.3 years and confidence interval of 95% of 37.9-41.1 years. Concerning education, complete secondary education was predominant in SC1, with 43.5%, and SC3, with 51.6%; in SC2, most had completed higher education and post-graduation (41.3%).
Concerning position/function, nursing technicians were the majority in the three SC, with a general mean of 71.5%, followed by nurses (17.5%) and nursing assistants (11.0%). A significant statistical difference was observed among the variables age group, education (p < 0.001) and position/function (p = 0.016) and type of SC. Concerning time working at the hospital, in SC1 and SC2, most professionals worked for five years in these institutions, with 51.1% and 91.3%, respectively. The SC3 has presented a percentage of 30.6% of professionals working at the hospital for over 15 years. Table 1 presents the consolidated positive, neutral, and negative answers of the dimensions of HSOPSC of the nursing professionals per hospital, characterizing the SC, presenting the weak and strong points of each dimension. Regarding the overall perception of the patient safety score, 7.5% of the participants of the three SC considered it as "excellent". In SC1, the predominant safety score was "average", with 48.9% of the total. The SC2 and SC3 assessed patient safety as "very good", with 47.8% and 48.3% of the opinions, respectively (Figure 1).  Most frequent mention of adverse events notification was observed among nurses, mainly in SC1 and SC2, with 66.7% in SC1, 66.7 in SC2, and 33.4% in SC3. The nursing assistants of SC1 have not notified adverse events, whereas notification by nursing technicians was low in all the SC: 8.6% in the SC1, 14.7% in the SC2, and 12.8% in SC3.

DISCUSSION
The teamwork dimension at the unit, compared to the other dimensions, has received one of the best evaluations in SC2 and SC3. Teamwork in these units has not achieved 75%, but was perceived by the professionals as potentially positive, which contributes to safe care, since its members work efficiently together. A study conducted in China with 2,230 participants, comparing the safety culture in surgical units and other areas of the hospital, has obtained 88.9% of positive answers for this dimension at the SC, a percentage which is higher than that of this study, denoting respect, cooperation, and coordination between health teams in this country (15) .
The SC3 was noteworthy regarding expectations and actions of the supervisor for promotion of patient safety, showing that most professionals mention that the manager puts effort to promote safety and takes into account the team's suggestions. In a study conducted in Rio Grande do Norte state, Brazil, with 215 professionals, in three hospitals, this dimension has achieved 66.7% of positive answers (13) .
Thus, the role of the supervisor/head is essential in the support to the promotion of patient safety within the work environment, since they spend a considerable amount of time leading people and making decisions; therefore, they are expected to have management skills to transform and translate knowledge, skills, and attitudes into positive results for institutionalizing a patient and worker safety culture (16) .
The dimension "Organizational learning -continuous improvement" has obtained the highest percentage among the three SC, being considered a strengthened patient safety culture, showing that nursing professionals perceive that in this institution there is a philosophy of continuous improvement that enables worker development. These data are in agreement with a study conducted in Southern Brazil (17) and in China (15) , which have obtained 81.8% and 79.1%, respectively, of positive answers to this dimension. In this perspective, professionals and hospital managers must be committed to guarantee a safe care to patients and health professionals, promoting organizational learning and continuous improvement of care and management practices in the hospital organizations (18) .
The SC2 and SC3 have obtained percentages considered as not problematic regarding the return of information and error communication. These data coincide with results found in a study conducted in São Paulo with 197 professionals, which obtained 54.1% of positive answers (19) . Therefore, there must be the return of information and communication of errors within the work sector, so as to alert professionals to discuss the prevention of possible further mistakes.
Communication is fundamental for the development of the nurse's work with the team so as to study errors from a nonpunitive approach and event notifiers receive feedback on the generated information (20) .
Most SC had low percentages of positive answers, showing that professionals do not feel at ease regarding opening for communication. Corroborating this finding, a study conducted in an intensive care unit (ICU) in Southern Brazil has presented a percentage of 30.9% of positive answers in this same dimension (21) . A study conducted in China, in surgical units, has shown that the opening for communication has a worse performance than the other sectors and that this dimension and feedback and error communication are correlated to a higher degree of safety and more reported events; therefore, if there is a deficit in these dimensions, the safety degree is deficient and the reported events are minimized (15) .
Concerning worker adequation, most SC had a smaller number of positive answers, i.e., the amount of personnel was considered a weak spot. A study conducted in a public hospital in Fortaleza, Ceará state, has pointed personnel sizing and workload as one of the main intervening factors for patient safety (22) . Following this logic, a proper sizing of nursing personnel in the institutions is necessary, since they represent a considerable part of the health team. An appropriate number of nursing personnel is crucial for the establishment of a safety culture.
Punitive response to errors was the dimension with the lowest overall percentage in the three SC, showing that, in the professionals' opinion, communicated errors may be used against them and be maintained in their professional record. These factors may prevent professionals from reporting errors, since periodical assessments of their performance are conducted; in federal institutions, this is usually more effective. In the ICU of hospitals in Southern Brazil, this dimension was also among the smallest indexes of positive response, with 17.5% (21) . In an exploratory study performed in many units of a public hospital in São Paulo, the smallest scores correspond to the dimension "Non-punitive response to error" (29.6%) (19) .
The SC1 was the one receiving the least emphasis regarding support of hospital manage to patient safety, reflecting a fragile area of culture and the need for improvements by hospital managers, since, according to participants, there is little commitment and support by the management regarding patient safety. A study conducted in teaching hospitals in Iran has concluded that, to improve patient safety, an important step must be the obtention of support by the hospital administration, which should assume a non-punitive attitude towards the professionals who have made mistakes, improving thus patient safety culture (23) . The administration should understand that the errors and adverse events are systemic and that professionals are prone to commit them when processes are poorly planned or complex (24) .
The SC1 also had a low index of positive responses to the item duty/shift change and transferences, showing need for improvement. The difficulty of interaction and the little communication among the teams of the origin and destination of the patient contribute to a significant increase of complications in transportation. The constant training and improvement of professionals, as well as the standardization of the actions and the necessary equipment for patient clinical monitoring, must be available for prevention or minimization of adverse events, so as to promote the client's safety (24) .
Most SC had low percentages of positive answers regarding the overall perception of patient safety. Strategies for the improvement of quality of care are necessary. Concerning the frequency of communicated events, in SC2, the obtained percentage was considered a strengthened area. Patient safety culture proposes that incidents must be reported, enabling analysis and the institution's adoption of preventive and educational measures related to these occurrences. However, for errors to be notified, the method for their punishment must be changed and culture of guilty must not be maintained; the latter blames the subjects who make mistakes, and not the problem leading to the error, since the fear of repression, sanctions, dismissal, warnings, ethical processes, and shame favors under-reporting (25) .
Out of the three, SC1 has presented the most fragile safety culture, requiring the promotion of strategies to improve quality of care, associated to mechanisms of control and monitoring of actions with commitment, transparency, and collective responsibility to subsidize risk management and achieve a better perception of patient safety by the professionals (13) . A study in public Palestine hospitals, which compared their results to those of studies conducted in similar regions (Lebanon and Saudi Arabia) and in the United States has shown that the latter had more positive answers, obtaining more "very good" and "excellent" scores than the other ones, showing that more developed countries have better conditions of providing appropriate patient safety (26) .
The patient safety score may be related to the profile of the three hospitals and the type of management in which the studied SC are inserted. The SC1, in the municipal sphere which received the safety score "average", is an "open doors" emergency unit, providing care to patients which are forwarded to urgency and emergency surgeries. These places may pose more risks to the conduction of safe care due to some specificities, such as overcrowding, high workload, lack of material and human resources, unsatisfied professionals and users, and delays. These factors may compromise quality of care, increasing the risk of errors or adverse events in those places (27) .
perception of safety score attributed by nursing professionals. Thus, interventions to improve patient safety must consider the type of administrative management, according to this research's data. Professionals experiencing situations involving human beings and their well-being, in addition to demands of development of specific skills and competences to adjust to the reality of safe work for the patient, require constant updates (24) . Continuous education is essential, since, if conducted as a permanent process, enables the development of professional competence, aiming at the acquisition of knowledge, skills, and attitudes, in addition to helping reduce problems caused by educational gaps (28) .
Concerning the number of adverse events notified in the previous 12 months, low adherence to error communication may also be related to conducts directed at the professionals, mainly regarding the punitive approach to errors, as confirmed by research conducted in Southern China, in which "non-punitive response to error" had 48.1% of positive responses, not being considered a strengthened area for patient safety culture (29) . These data were confirmed by a Brazilian study, which also had a high percentage of non-reporting (87.8%) (12) . Thus, a punitive organizational culture, based on blame, may lead to omission of adverse events, hindering the construction of an institutional culture aimed at patient safety (18) .
This study has a limitation due to comprehending the specific perspective of the nursing team. Despite this, it had the participation of a large share of these professionals, which correspond to the highest amount of care-providing workers. Considering that the study has focused only on SC, this research is suggested to be reapplied in other units of these hospitals, so as to identify frailties and potentialities in other areas, aiming at planning change based on professional assessment, in order to sensibilize them for safe care.
This research has provided a broad analysis of patient safety culture in SC of different administrative spheres. This study's results are believed to contribute to a stronger basis on this theme, in addition to sensibilizing professionals and managers for the importance of encouraging a change of attitude towards fragile areas of each dimension, in each type of management.

CONCLUSION
Patient safety culture analysis, in the perception of the nursing teams of three reference SC, with different types of management, has shown that one of the centers did not have a strengthened patient safety area; two presented the dimensions "Organizational learning -continuous improvement" and "Frequency of reported events" as strengthened areas. In the municipal SC, an "average" safety score prevailed; in the state and the federal centers, patient safety obtained a "very good" score. Therefore, the sensibilization of managers in each type of administrative sphere and health professionals for promoting safety culture in the SC, as well as the education of these professionals, is understood as crucial.