Safety culture of multidisciplinary teams from neonatal intensive care units of public hospitals*

Objective analyze the safety culture of multidisciplinary teams from three neonatal intensive care units of public hospitals in Minas Gerais, Brazil. Method a cross-sectional survey conducted with 514 health professionals, using the Hospital Survey on Patient Safety Culture; data were subjected to a descriptive statistical analysis in software R-3.3.2. Results the findings showed that none of the dimensions had a positive response score above 75% to be considered as a strength area. The dimension ‘Nonpunitive response to error’ was classified as a critical area of the patient safety culture, present in 55.45% of the responses. However, areas with potential for improvements were identified, such as ‘Teamwork within units’ (59.44%) and ‘Supervisor/manager’s expectations and actions to promote patient safety’ (49.90%). Conclusion none of the dimensions was considered as a strength area, which indicates safety culture has not been fully implemented in the evaluated units. A critical look at the weaknesses of the patient safety process is recommended in order to seek strategies for the adoption of a positive safety culture to benefit patients, family members and health professionals.


Introduction
Patient safety is one of the critical pillars of health care quality and discussions about it have been strengthened after the publication of the American report To err is human: building a safer health system, which highlights the great number of errors and damages involved in health care (1) .
After that, studies on safety culture assessment and impact on health management have been considered crucial for the development of safe care, with emphasis on learning, continuous improvement and nonpunitive response to error (2) . Safety culture is characterized as the product of individual and collective values, attitudes, skills and behavior patterns, which determine the commitment, style and proficiency of a healthy and safe organization (3) .
Safety culture in health care settings is usually assessed through quantitative questionnaires based on individual items and a combination of dimensions (2)(3)(4) .
One study reports that institutions with a positive safety culture offer safe and better quality of care to their patients. In addition, better rates in safety culture assessments may help reduce occurrences of infection and adverse events (4) .
Patient safety can be influenced by the work culture of the multidisciplinary team involved. A study reports that many elements of work culture directly affect health care, especially due to the way health professionals see patient safety and perform their work (5) .
In settings such as neonatal intensive care units (NICUs), where patients are more vulnerable and the daily routine of the multidisciplinary team involves many error-prone processes (6) , analyzing the safety culture becomes critical to identify areas with potential for improvements.
Then, studies that measure the safety culture in   Values of >0.5 were considered of good reliability.
The responses provided in each dimension were classified in areas of strength or critical areas (7)(8) . Areas of strength were those presenting 75% of 'strongly agree/agree' or 'almost always/always' responses to positively worded questions, and 'strongly disagree/ disagree' or 'never/almost never' for negatively worded questions. Critical areas were those presenting 50% or more participants answering negatively with 'strongly disagree/disagree' or 'never/almost never' for positively worded questions, and 'strongly agree/ agree' or 'almost always/always' for negatively worded questions (7)(8) .   According to the guidelines from the Agency for Health Research and Quality, data obtained in this study did not show any dimension with a positive response score above 75% to be considered a strength area. That is, of the 12 dimensions evaluated, 11 were characterized as weakness or opportunities for improvement, and none as a strength area. However, some dimensions presenting a higher percentage of positive responses and the items of these dimensions received a better evaluation (8) . and 39.8% considered safety as 'very good' (Figure 2).   (Table 2).

Discussion
The results show that safety culture is not fully established in the NICUs, which is similar to other studies (6,9) . However, some dimensions presented the highest percentage of positive responses, but also below 75%, among them, the dimensions of 'Teamwork within units' and 'Supervisor/manager's expectations and actions to promote patient safety.' International studies have reported similar findings (10)(11) , as well as Brazilian studies (12) . Despite regional cultural specificities, the to error' received 56% negative response and was the dimension with the worst evaluation (9) . It is evident a culpability culture blames an individual for an error, discouraging him/her to report the error and, consequently, prevents organizational learning from such occurrence (13) .
Regarding the dimension of 'General perception of patient safety,' the item with the worst evaluation was 'Patient safety is never compromised due to the greater amount of work to be performed,' probably due to the professional's perception of the daily workload in the unit and the insufficient staff to meet the demand of care provision. A study with a multidisciplinary team from eight public hospitals in the region of Murcia, Spain (10) , showed similar results to this study, with a high percentage of negative responses in this dimension.
Despite not showing strength areas for patient safety, but critical areas only, most professionals classified patient safety as 'acceptable' and 'very good.'

A study conducted in a public general hospital in Minas
Gerais, Brazil, also found similar assessment of patient safety, ranging from 'acceptable' (43%) to 'very good' (40%) (14) .
Regarding events reported, most professionals responded they had not filled out any event form. This situation is even more alarming when a consensus is observed among experts in the subject stating the reported numbers of adverse events are a very modest estimate versus the actual number (1) . In addition, the number of adverse events reported should not be the responsibility of a single professional category, as found in this study. The responsibility for safety should be equally shared by all teams.  (11) . In contrast, a study conducted in a chain of public hospitals in the region of Murcia, Spain, showed that nurses were more positive about safety assessment than physicians (15) . to 20 reported events (80% and 20%, respectively) (5) .
Sometimes, a nurse has the responsibility to report events, as he/she is considered the most capable leader to manage adverse reports of events and encourage the team (16) . A study reports the need to encourage the communication of events by the multidisciplinary team in order to collectively develop strategies for error prevention and promotion of a consolidated safety culture (16) . The authors emphasize that, despite the fact that error reporting is a responsibility of the whole team, the hospital management needs to assume a leadership position, encouraging and implementing a safety culture that addresses errors in a systemic and non-punitive manner (6) .
The lack of reports from technicians/assistants and physicians was probably due to 'corporatism,' fear or lack of knowledge of error reporting systems, and due to the perception that incident reporting may not result in improvements (17) .  (18)(19) .
This way, studies suggest that senior management commitment to support the development of a patient safety culture, the use of information technology and simulators to reduce errors, incentive to error reporting practices, and educational practices are essential for enhancing a safety culture (19)(20) .
Regarding the insertion of 'patient safety' in the organizational environment and, consequently, in the organizational culture, it should be noted that it is influenced by the labor and power relations existing among the various professional profiles that constitute a hospital environment (18) . Then, in order to establish a patient safety culture across several professional categories, the managers in charge should lead this multidisciplinary team and promote a work environment based on dialogue and learning. Another aspect to be considered is that event reporting should be incorporated into the daily routine of these professionals, establishing a culture of learning. A national study emphasizes the need for institutional investments in the promotion and development of safe health systems (12) .

Conclusion
The findings of this study did not present any of the