Patient safety culture in a COVID-19 ICU compared to a clinical-surgical ICU in the Brazilian Eastern Amazon: A cross-sectional study

1.Introduction

Patient safety (PS) is an important and complex public health issue [1]. Patient safety culture (PSC) is defined as a component of organizational culture that encompasses a set of individual or group values, attitudes, perceptions, competencies, and behaviors. Organizations with a strengthened safety culture are characterized by good communication among professionals, mutual trust, and a shared perception of the importance and effectiveness of preventive measures [2,3]. PSC is also an important strategy that favors the implementation of safe practices and the mitigation of adverse events (AE) [4]. As AE harms patients, it is therefore considered a failure in medical care processes, and its occurrence is related to the characteristics of specific populations and PSC [5].

The COVID-19 pandemic promoted changes in behavior and attitudes toward PS, due to new dynamics and higher stress levels among workers and managers. The pandemic created a context of excessive pressure on frontline health care workers and hospitals, increased workload, increased number of patients, lack of knowledge about the virus, and lack of infrastructure. This scenario impairs performance, affects the quality of health care, increases the incidence of adverse events, and promotes poor PS [6–9].

In this moment of crisis, organizational planning had to be created in a short time to improve the environment of professional practice and ensure the conditions for safe and quality care. Accordingly, research conducted to analyze PS in the pandemic context shows that the dimensions that most often show weaknesses are: staffing, hospital leadership support for safety, interdepartmental teamwork, handoffs and transitions, openness of communication, nonpunitive response to errors, and overall perception of PS [10–12].

Considering the changes that the pandemic has caused in the organization of services, the evaluation of PSC in hospitals, especially in the Intensive Care Units (ICU), is imperative, since at least 20% of patients with COVID-19 have to be admitted to the ICU due to acute respiratory distress syndrome (ARDS) [13] and the occurrence of SARs in the ICU is more likely due to the complexity of interventions, protocols, and the need to make decisions within a short time [14,15]. A study of 343 event reports from 71 hospitals in Pennsylvania showed that 17% of SARs occur in the ICU, resulting in the following: Exposure to COVID-19 positive or presumptive positive patients, missed/delayed care or treatment, missed/delayed tests or results, and wasted resources [16].

In developing countries, basic hospital safety indicators are usually incomplete or unavailable. The few data found indicate that the prevalence of adverse events in South American countries is much higher than in developed countries [17]. The safety culture within a country’s health organization is strongly influenced by factors such as cultural norms, expectations, race, religion, and other personal characteristics [18]. In Brazil, Ordinance No. 529 of 2013 was launched by the National PS Program (PNSP) and aims to provide subsidies for health institutions in the national territory to have a framework to implement and promote measures of security [19] and thus direct and unify/standardize actions.

Brazil is a country with a large territorial extension, and the Brazilian Eastern Amazon has a particular reality with a lack of research on safety culture. Therefore, this study aimed to assess the PSC from the perspective of the health team professionals in COVID-19 ICU compared to clinical-surgical ICU in hospital references in infectious diseases in the Brazilian North Amazon.

2.Methods

3.Results

The characteristics of the health team professionals are presented in Table 1. The individuals of COVID-19 ICU (47.86 ± 8.11) were older than clinical-surgical ICU (41.04 ± 10.48) (t = −2.568, p = 0.013). Health team professionals of COVID-19 ICU work less time in the hospital. The time of work of participants in their current specialty or profession did not present statistical differences (G1: 15.74 ± 7.32 years; G2: 15.55 ± 8.09; t = 0.083, p = 0.934).

Table 2 shows the frequency of PSC responses overall, as well as positive, neutral, and negative responses for both groups. In the domain “Supervisor/Manager Expectations and Actions Promoting PS” (p = 0.003), individuals from G1 were more negative and neutral, as well as for “Management Support for PS”, although G2 had a more positive “Overall Perceptions of PS”.

In the analysis of session items, the perspective on work area/unit and supervisor/manager session (Table 3) revealed that G2s most positively agreed with the statement that they are actively doing things to improve PS (p = 0.038). They also responded most positively that they get the most help from other professionals, most negatively that the supervisor/manager overlooks PS problems, and more neutral about when in “crisis mode” trying to do too much too quickly. G2 was more positive in thinking that there are problems in this unit.

The perspectives on communication (Table 4) showed that G2 was more neutral towards information about errors in their unit (p = 0.009). In addition, G2 more positively indicated that they discuss ways to prevent errors from recurring. G2 assumes that when a mistake is corrected before the patient, the event is never reported in the clinical surgical ICU.

Perspectives on the hospital (Table 5) showed that G1 was more negative in the view that hospital management provides a work climate that promotes PS (p = 0.047); more positive in agreeing that it is often unpleasant to work with staff from other hospital units (p = 0.008); and that problems often occur in the exchange of information across hospital units (p = 0.047). In addition, G2 was more positive about the actions of hospital management, indicating that PS is a top priority and that hospital units work well together to provide the best care to patients. G1 expressed a more neutral view, assuming that shift changes are problematic for patients at this hospital.

In relation to PSG, the participants’ responses to the sentence “Please give your work area/unit in this hospital an overall grade on PS” showed no statistical difference (Excellent G1: 2 (7.4), G2: 3 (15.8); Very good G1: 9 (33.3), G2: 10 (52.6); Acceptable G1: 10 (37.0), G2: 5 (26.3); Poor G1: 5 (18.5), G2: 1 (5.3); Falling G1: 1 (3.7), G2: 0 (0.0); 𝜒²: 4.325; p = 0.364). However, in the comparison between the total score in the different sessions (Table 6), the score PSG was worse in G1 (p = 0.045).

Still, in the past 12 months, there was no statistical difference in the event reports for participants (No event reports G1: 14 (50), G2: 12 (57.1); 1 to 2 event reports G1: 10 (35.7), G2: 5 (23.8); 3 to 5 event reports G1: 2 (7.1), G2: 3 (14.3); 6 to 10 event reports G1: 2 (7.1), G2: 1 (4.8); 𝜒²: 1.382; p = 0.710).

Table 7 presents the correlation between the total score in the different sessions. The results indicate a moderate positive correlation, in which, for G2, the better the execution/performance in the supervisor manager with work area/unit session, and the better the execution/performance in hospital with frequency of events reported, the better the other session (p = 0.026 and p = 0.022, respectively). The highest PSG was related to worst communication (G1 and G2) and with the best performance against the frequency of events reported (G2) (p < 0.05).

4.Discussion

This study provides original information for understanding the level of PS associated with the COVID-19 pandemic in public hospitals of the Brazilian Eastern Amazon. The results suggest that the domains “Supervisor/Manager Expectations and Actions Promoting PS” and “Management Support to PS” may be a weak point for the clinical surgical ICU.

During the COVID-19 pandemic, despite an increase in patient volume and the complexity of their clinical conditions leading to a potential reduction in PS, there was an increase in patient safety incident reports (PSIRs) during the COVID-19 pandemic [21] due to greater management attention, as well as improved outcomes in some PSC areas, such as in the domains ‘feedback and communication about errors’, ‘communication openness’ and ‘frequency of event reported' [11].

A study evaluating PSC before the pandemic and after the second wave showed that there were overall more positive ratings than in the national study conducted before the pandemic, in which the dimension “expectation of actions by management/supervision of the service” was considered a strength, with 85.29% positive responses [10].

Studies using different methods suggest that in normal contexts, the ability of management and supervision to connect with professionals to identify vulnerabilities in ICU is vulnerable [22–24], which may be associated with difficulties in relationships between professionals and managers due to reactive actions in the face of punitive events to correct irregularities [25,26].

A systematic review with meta-analysis [27] that included 59 studies found that dimensions of management support for PS, supervisor/manager expectations and actions to promote PS, and overall perceptions of PS remained neutral in most general ICUs. In our study, the first area remained as a potential weakness compared with COVID-19 ICU.

In our study, it was also observed that a higher level of PSG was associated with poorer communication, although PSG was reflected in the improvement of error reporting for the clinical surgical ICU. It is well known that lack of communication leads to underreporting of events and thus underreporting of errors, which may hinder the establishment and implementation of error prevention strategies [11].

This could also be related to the positive assessment of the dimension supervisor/manager expectations and actions to promote the patient’s safety by COVID-19 ICU professionals, who were also more aware of PS problems in this unit, and therefore more proactive in doing something to improve PS. This suggests that the team feels supported in their work environment and is more willing to participate in teamwork for the benefit of PS [28], as positive responses may be due to professionals feeling an active part of the safety culture [29], due to mature management that creates an atmosphere of awareness and improvement, rather than just taking action after incidents occur [30].

Thus, the results of this study suggest that the implementation of safety culture strategies needs to be improved in many ways by facility leadership and the management of the institution, especially in non-specialized ICUs.

As a limitation of this study, we have the number of participants. However, our sample included the entire population of services interested in participating in this study, and there was no difference in the professional class between the groups. The study demonstrated external validity when considering the use of statistical power analyses. However, we address the need for further research on this topic.

It is important to remember that in Brazil, due to the high demand for hospitalizations and the characteristics of COVID-19, there was a need to implement specific ICUs for infected patients and to hire emergency specialists to act in the face of the pandemic, which can be illustrated in our sample in relation to the COVID-19 ICU professionals who had recently joined the hospital on a temporary contract, as opposed to the clinical-surgical ICU, which may have influenced the most positive evaluations in this group [31].

Therefore, studies at PS are useful to guide decision making and intervention management, where interventions should be directed toward goals to minimize risks and change the current situation of adverse events, thus modifying the practice of care. Especially at this time when the COVID-19 pandemic has received more attention than other diseases. Therefore, it is necessary that other services not related to COVID-19 continue to receive quality attention.

5.Conclusion

Our study revealed a moderate positive correlation between supervisor/manager and work area/unit session and between hospital and frequency of reported events. The highest PS grade was related to poorer communication and to better performance in the frequency of reported events only for the COVID-19 ICU. In addition, although there was no difference between groups in overall PSC, the domains “Supervisor/Manager Expectations and Actions Promoting PS” and “Management Support for PS” could be key points for PSC for the general ICU. This study highlights the need to balance attention between the different ICUs in times of crisis at PS, especially in terms of expectations and actions to promote PS by supervisors/managers to prevent the weakness of PS. Nevertheless, further studies should be conducted to support these findings.