Canonical correlations between individual self-efficacy/organizational bottom-up approach and perceived barriers to reporting medication errors: a multicenter study

Patient-safety reporting systems have been used to detect the root causes of errors and to enhance patient safety and the economic efficiency of healthcare [1]. However, the number of errors reported by these means is less than the number of actual errors determined by experts [2]. Because under-reporting errors is a critical limitation to analyzing the process behind medication administration errors (MAEs) [3], understanding perceived barriers to error reporting is crucial in promoting reporting behavior. Several studies have shown individual and organizational factors for enhancing error- reporting [2, 3], and one study [4] proposed a conceptual model for determinants of MAE reporting with an individual perceived self-efficacy and organizational factors such as reporting process capability and support, organizational culture, management support, and regulatory authority.

Individual perceived self-efficacy is defined as an individual’s ability or competence to effectively cope with responding to MAEs [4]. Understanding MAEs reduces confusion about error definitions [5], raises levels of clinical reasoning about whether the error was harmful [6], and allows more frequent reporting of perceived and identified errors by healthcare providers [5]. Therefore, required competencies for medication safety were a knowledge and decision-making [7]. Collection of audit data and reporting of errors provide evidence-based knowledge of medication administration standards [8]. However, the information accumulated to measure knowledge is insufficient, owing to the lack of feedback on error-reporting systems [8]. Furthermore, while the majority of MAEs do not result in harm [9], some drugs, known as ‘high-alert’ drugs, are more likely to cause significant patient harm when used in error [10]. More attention should be paid to substances categorized as high-alert drugs by the US Institute for Safe Medication Practices [10] and a systematic review [11], including anticoagulants, cardiovascular drugs, insulins, and opiates, among others. Specific knowledge about high-alert medication, which is high -risk medication for nurses’ MAEs [12], can serve as the basic competence level when judging what is an MAE and deciding whether to report it.

Judging medication errors and the decision to report them may depend on nurses’ knowledge base and their decision-making ability; however, there is insufficient previous research on nurses’ decisions to report errors. In particular, barriers to reporting MAEs may depend on whether decision-making is analytic or intuitive [13]. According to analytical-intuitive decision-making theory [14], cognitive processes may move along the analytical-intuitive continuum over time. Generally, analytical decision-making is known as an appropriate decision-making style when information is organized in memory and the correct weight and importance is given to each piece of information [14, 15]. Intuitive decision-making is defined as an unconscious, not logically defensible, and non-stepwise process [14, 15]. However, decision-making cannot be the same across diverse nursing situations [15], and little is known about the relationship between nurses’ decision-making and error- reporting situations.

Regarding organizational factors, improving patient-safety culture and supportive leadership is a key organization-based strategy for enhancing error reporting [4, 16]. Safety culture has major effects on patient outcomes and a positive safety culture in healthcare is evidenced by the visibility of leaders and support for patient-safety initiatives [17]. However, because patient-safety culture starts at the highest levels of the healthcare organization [16], it is difficult to establish and maintain supportive management of medication safety [18]. Therefore, it can be helpful to utilize a bottom-up approach [19]. A bottom-up approach to improving patient-safety culture means that patient-safety awareness should be embedded at all staff levels and be seen in all actions [20], and collaboration among healthcare providers can establish effective communication [21] and create a positive patient-safety culture. Communication openness regarding mistakes and failures is a crucial factor in patient-safety; however, communication remains an adaptive challenge that has proven difficult to overcome in the sociotechnical viewpoint that defines healthcare [22]. Most of the current literature has focused on a top-down approach such as supportive leadership for enhancing error reporting [16‐18] and there has been limited research on bottom-up approaches such as collaboration.

Although several studies have demonstrated perceived barriers to error reporting, there is a gap in the empirical study of areas such as knowledge for errors, decision-making, and the organizational bottom-up approach. Furthermore, there may be combination effects of variables in relation to individual or organizational factors for perceived barriers to MAE reporting. Based on the fact that communication and pharmacological knowledge impact nurses’ decision-making concerning medication [23], research variables are expected to show something other than a bivariate relationship with perceived barriers to MAE reporting. Therefore, the current research questions were set as follows: (1) is there a relationship between an individual self-efficacy/organizational bottom-up approach set and perceived barriers to MAE reporting? And if so, (2) what factor has the strongest relationship for reducing perceived barriers to reporting MAE?

The study aimed to define the roles of individual/organizational factors in reducing perceived barriers to reporting MAEs using canonical correlations; therefore, the findings are discussed by focusing on the significant canonical variates among study variables.

We identified the perceived barriers to reporting MAEs according to the participants’ characteristics. Nurses with a master’s degree perceived lower barriers to reporting MAEs than the others, though it is hard to generalize the result because only three participants had a master’s degree. Education was a barrier for reporting MAEs in a previous research [34]. According to one study, more confident and autonomous and more experienced nurses might think that not all medication errors are severe enough to be reported [35]. However, because underreporting makes it impossible to accurately analyze the medication process behind errors or near misses, all errors, regardless of their severity, should be reported in order to ensure medication safety [2]. Emphasis on the benefits of reporting was provided to the nurses in continuing education or in their graduate coursework. To prove a stronger relationship, it is necessary to examine the relationship between educational status and barriers to MAE reporting after increasing the number of participants and to test the effectiveness of the educational program.

The highest perceived barrier to MAE reporting was fear, a result similar to a previous study [36]. Fear may come from Korean cultural attributes such as hierarchical culture and face-saving concern [36]. If nurses feel that they are working within a strongly hierarchical structure, they may be afraid of co-workers’ and unit managers’ potential reactions to MAE reporting, such as blame or punishment [6]. Face-saving is defined as an individual’s claimed sense of positive feelings in the context of social interactions [37]. Health professionals in conservative and collectivistic cultures are probably reluctant to report coworkers’ MAEs in order to maintain harmonious interrelationships [36]. This conservative culture and hierarchy may mean that trying to look for and report errors would be regarded as making their colleagues uncomfortable or even a betrayal. Therefore, chief executives in hospital organizations should try to build a non-punitive culture [38] and members of organizations need to think of reporting behavior as a way of improving patient-safety practice, not in terms of gaining or losing face.

There are two significant canonical variates in this data set, which means the variables in the individual self-efficacy/organizational bottom-up approach set are related to the variables in perceived barriers to MAE reporting. In the first significant pair of canonical variates, lower knowledge, intuitive decision-making, and good nurse-physician collaboration satisfaction were associated with a higher perception of disagreement over medication errors. Decision-making is based on how information is organized in memory and the weight given to the information, and insufficient knowledge might be related to more intuitive decision-making [15]. Intuitive decision-makers show rapid information processing, simultaneous cue use, and evaluation cue use, because this type of decision-making is based on immediate apprehension of relevant situations [39]. Since nurses use an intuitive approach in data collection and an analytic approach in data processing and the identification of problems [40], a mismatch between decision-making style and situation is present. Intuitive decision-makers cannot make decisions based on their insufficient error-related knowledge; therefore, nurses’ decision-making style may be transformed from intuitive to analytical. Supportive social and institutional contexts such as good collaboration satisfaction play an important role in decision-making [41]. In these conditions, when opportunities for discussion are provided, which may or may not be appropriate, intuitive decision-making would be fostered. To increase the precision of our understanding of their relationship, additional research needs to be conducted.

In the second pair, analytic decision-making and higher nurse-physician collaboration were associated with lower perceived barriers to error reporting. In clinical situations, smooth communication and cooperation not only enhance perceptions of being a team, lead to positive patient outcomes, and limit MAEs, but also promote informal error reporting among colleagues [6], and foster collaboration in decision-making [42]. In a previous study, when nurses were provided with potential error-driven situations, they would informally report errors as a means of clarifying and validating their opinion with their colleagues [6]. In addition, because formal error reporting is also regarded as a form of bottom-up communication [43], we suggest that staff-led communication and better teamwork must be established for better informal and formal error reporting. Analytical decision-making also needs to be carried out by healthcare professionals to reduce perception of barriers to reporting MAEs [15]. One strategy for supporting analytical decision-making may be the development of a clinical decision-making supporting system (CDSS) for error reporting. A CDSS is a computerized system that uses case-based reasoning rooted in treatment guidelines to assist clinicians in clinical decision-making [44]. It can improve medication safety by improving the process of care, such as by reducing medication errors and improving patient outcomes [45]. It is essential to use consistent terms [46, 47], such as those provided by a standardized taxonomy, when adopting a CDSS in error-reporting. It appears likely that a CDSS based on a medication-error taxonomy would facilitate analytic decision-making for error -reporting.

Better knowledge about medication administration and good nurse-physician collaboration satisfaction were the stronger factors in each significant pair of canonical variates for reducing perceived barriers to MAE reporting. According to a literature review, nurses’ use of clinical reasoning to maintain safe medication administration was inadequate and clinical reasoning required in-depth knowledge [48]. Based on this study, knowledge about medication administration will not only be used in clinical reasoning for safe medication, but also as a precursor of error reporting. However, it is noteworthy that perception of disagreement over medication errors was the lowest barrier, even though the level of knowledge in this study was not high. Cognitive dissonance between perceived and real knowledge level is a possible reason. According to metacognition theory, if learners can accurately perceive the status of their knowledge, they will tend to think of themselves as good learners and complete work more efficiently [49]. Therefore, increased metacognition for knowledge of medication administration is needed.

A top-down approach with leadership from senior personnel takes time to create better patient-safety [50] and needs to be converted into a bottom-up approach to foster teambuilding and an open, fair environment. Better collaboration has been linked specifically to safety-relevant performance [51] such as fewer hospital-acquired infections [52]. According to a recent study, nurses perceive physicians not as team members, but as independent healthcare workers [21]. In addition, nurses who perceive more respect perceive higher levels of nurse-physician collaboration [21]. When nurses work in an environment that supports and recognizes their professional role, they may feel respected, and therefore more confident participating in decision-making during the collaboration process [53] It is helpful if hospital administrators create management regulations for collaborative nurse-physician decision-making and to maximize nurses’ involvement in the decision-making process [21]. However, self-help efforts for respectful communication between healthcare providers, which may be the foundation for collaboration, are even more important. Therefore, to improve nurse-physician communication, communication tools such as SBAR and cross-training/interprofessional educational programs would be helpful [53].

Although this study provides meaningful results and findings, there are several limitations. A convenient sampling may suggest possible sampling bias; therefore, generalizability of the findings of this study to other settings is limited. Another important limitation is the relatively low reliability of medication administration knowledge, which means a low proportion of true variability to the total obtained variability. Therefore, instrument refinement and additional studies are necessary.

Despite these limitations, this study showed the possible influence of an individual self-efficacy/organizational bottom-up approach on reducing perceived barriers to MAE reporting. The findings suggest three recommendations for the clinical and policy phases. In the clinical phase, medication administration education, including high-alert drug administration training for nurses, may reduce disagreement about whether errors have occurred. In addition, a bottom-up approach strategy such as improving collaboration with interactive communication may facilitate MAE reporting. In the policy phase, administrators need to make an effort to promote analytical decision-making for reporting MAEs. Hospital leaders who construct and implement evidence-based medication error reporting systems that support decision-making can reduce the perceived barriers to reporting MAEs. The present study sheds light on the broader view of reducing perceived barriers to reporting MAEs and suggests further research that examines the effectiveness of the experimental study to extend the strategy of enhancing MAE reporting.