Introduction
Patient safety is an important issue in the healthcare field [1]. Many countries are strengthening patient safety through regulatory frameworks such as mandatory licensing systems for healthcare facilities to ensure compliance with established safety standards and promote quality healthcare delivery [2]. However, it has been estimated that 10% patients receiving medical services experience damage from patient safety incidents (PSIs) [3].
PSIs impact not only patients, but also healthcare providers (HPs), termed ‘second victims’ [4]. HPs involved in PSIs often experience emotional and mental distress [5, 6]. Documented adverse repercussions of PSIs include the need for assistance from colleagues, supervisors, institutions, and family members [4, 7], along with potential decisions to leave the organization [6, 7]. Second victim experiences (SVEs) can be subdivided into second victim distress (SVD), demand for support, and negative work-related outcomes [4‐7]. However, previous studies have presented a fragmented view of SVEs using only some attributes of second victims [7‐9]. Therefore, an integrated approach is required to understand causal relationships between SVD, demand for support, and negative work-related outcomes.
Anzeige
Studies have found that second victim experience is not limited to adverse events or sentinel events that cause serious harm to patients. It can also occur when a patient experiences a near miss that does not directly harm the patient [4, 10]. Therefore, active support and prevention activities at the institutional level have been emphasized to prevent HPs from second victim experience after a PSI [11].
Clinical nurses are more vulnerable to SVEs than other healthcare workers. Because they are directly involved in patient care, including medication administration, clinical nurses are also at the greatest risk of being blamed for a PSI [12]. In previous studies, 50–69% of clinical nurses reported symptoms of distress due to SVEs and more than 25% of them reported absenteeism and turnover intentions [9, 13]. However, only 5.6% of nurses reported to have received organizational support [13]. This indicates that nurse’s SVEs are often neglected. The concept of second victim was first introduced in 2000 [14]. Since then, studies have explored the phenomenon of second victim and its outcome factors, including current status of second victim [7], the process of overcoming second victim experience [4], psychological and physical symptoms of second victim [15], changes in nursing practice [16], and effects on the quality of life [17]. Nevertheless, only a few studies have explored antecedents of the second victim among nurses [8, 9]. The degree of SVE of an individual after PSI may vary depending on the culture of the organization or department [8, 9]. To mitigate the phenomenon of second victim among nurses, it is imperative to identify individual and organizational antecedents that might affect SVEs so that healthcare organizations could develop and implement various support strategies based on this information.
In this context, the significance of fostering a just culture (JC) within healthcare organizations has gained prominence. A JC is defined as a culture where trust is fostered by encouraging and rewarding the sharing of critical safety-related information, while maintaining a clear distinction between acceptable and unacceptable behaviors [18]. A JC can promote reporting of PSIs without assigning blame to individuals as it perceives such incidents as outcomes of shortcomings within organizational systems and structures rather than individual errors [18]. Furthermore, it embodies a culture of trust wherein nurses and managers engage in open communication regarding PSIs, seek and provide feedback, and engage in unfettered discussions concerning improvement strategies [19]. Nevertheless, by holding individuals responsible for incidents that involve intentional violations of explicit institutional rules or guidelines [19], it upholds a culture that adopts a balanced approach to cultivating trust among its members [20]. Hence, the concept of a JC distinguishes itself from patient safety culture which primarily centers on preempting patient safety events. This philosophy can be quantified through a specific subscale in established patient safety culture assessment tools - namely, a non-punitive response to errors [9, 10]. Studies have indicated that a non-punitive response to errors can help diminish distress experienced by nurses as second victims [8‐10]. Consequently, it is anticipated that a JC can help mitigate SVEs among nurses [21].
Identifying effects of a JC on SVEs is important to enlighten healthcare organizations about the efficacy of implementing such a culture to mitigate nurses’ experiences as second victims following PSIs. To date, studies have been conducted to gauge the level of JC, including examinations of its essence [22] and assessments of how nursing students and professors perceive it [23]. Additionally, studies have explored the impact of nursing work environment on JC [24] and effects of JC on activities related to patient safety management [25]. Nevertheless, to the best of our knowledge, the relationship between JC and SVEs has not been empirically demonstrated. Moreover, no studies have investigated repercussions of JC on such experiences.
Anzeige
Therefore, this study aimed to perform an in-depth examination of the relationship between JC of healthcare organizations and SVEs of clinical nurses (i.e., SVD, demand for support, and negative work-related outcomes) through path analysis. Path analysis offers the advantage of delineating both direct and indirect effects among variables within a hypothetical model, thereby enhancing explanatory power by illustrating connections between exogenous and endogenous factors [26]. Based on previous studies, this study has the following hypotheses: (1) JC is directly related to SVD, demand for support; (2) SVD is directly related to negative work-related outcomes, demand for support; (3) demand for support is directly related to negative work-related outcomes; (4) JC is indirectly related to negative work-related outcomes via SVD; (5) JC is indirectly related to negative work-related outcomes via demand for support; and (6) JC is indirectly related to negative work-related outcomes via SVD and, in turn, demand for support (Fig. 1). Findings of this study may help inform interventions for mitigating SVEs of nurses and fostering the establishment of a JC within healthcare organizations. To further explore these relationships, this study will address the following research questions:
Fig. 1
Hypothetical model
×
![]()
1.
How does JC influence second victim distress among clinical nurses?
2.
How does JC affect the demand for support after patient safety incidents?
3.
How does JC indirectly impact negative work-related outcomes through second victim distress and demand for support?
Methods
Study design
This was a cross-sectional correlational study to determine effects of healthcare organization’s JC on SVEs perceived by clinical nurses involved in PSIs. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines to ensure rigorous reporting of observational studies. This study constructed a hypothetical model, tested the model’s fit and hypotheses, and analyzed direct and indirect pathways of factors affecting SVEs.
Setting and samples
Study participants were clinical nurses working in a ward or intensive care unit, a nursing unit that provides direct nursing care to inpatients in tertiary general hospitals in Korea. Nurses with at least one year of clinical experience who had experienced at least one PSI (near miss, adverse event, or sentinel event) within the last six months were eligible for inclusion. In this study, we selected nurses with more than one year of clinical experience. This is because it takes at least 8 to 12 months for nurses to adapt to their roles [27]. We also excluded managers, head nurses, educational nurses, and physician assistant nurses who were not directly involved in patient care. A purposive sampling method was employed to select participants who met these inclusion criteria, ensuring that the sample was representative of nurses with relevant experiences of PSIs. Additionally, a snowball sampling approach was used to complement the recruitment process by asking respondents to recommend other eligible participants.
It is recommended to have a minimum sample size of 100–150 for using the maximum likelihood method in path analysis and at least 10 times the number of participants per observation [28]. This study had a total of 17 variables which required a minimum sample size of 170. Factoring a dropout rate of 15%, 200 nurses were enrolled in this study. Of a total of 200 questionnaires collected through an online survey, 183 (91.5%) were used for analysis after excluding 17 questionnaires due to response errors.
Measurements
The permission to use tools in this study was obtained from original authors or translators.
JC
JC was measured using the JC Assessment Tool (JCAT) developed by Petschonek et al. [19] and translated by Han et al. [29] with proven validity and reliability. The tool consists of 27 questions encompassing six factors, including three questions on feedback and communication, five questions on openness of communication, five questions on balance, five questions on quality of event reporting process, four questions on continuous improvement, and five questions on trust. Each question was answered on a Likert scale ranging from 1 (“strongly disagree”) to 7 (“strongly agree”), with a higher score indicating a higher perception that the organization had a JC. The Cronbach’s α for internal consistency of JCAT in this study was 0.91.
Anzeige
SVEs
Korea-Second Victim Experience and Support Tool (K-SVEST) developed by Kim et al. [30] was used to quantify SVEs. This tool consists of three domains with a total of 28 questions to simultaneously measure SVD, demand for support, and negative work-related outcomes due to experience of PSIs.
SVD
SVD is one of the subscales of the K-SVEST. It consists of 12 items: four items on psychological distress, four items on physical distress, and four items on decreased professional self-efficacy. Each question was answered on a Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”), with a higher mean score indicating more severe distress from the second victim experience. The Cronbach’s α of this subscale in this study was 0.87.
Demand for support
Demand for support is one of the subscales of the K-SVEST. It consists of 13 items, including four items on colleague support, four items on supervisor support, three items on institutional support, and two items on nonwork-related support. Each item was rated on a Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”), with a higher mean score indicating a higher support need as clinical nurses perceive insufficient support resources after experiencing PSIs. The Cronbach’s α of this subscale in this study was 0.79.
Negative work-related outcome
Negative work-related outcome is one of the subscales of K-SVEST. It consists of three questions: two questions on turnover intentions and one question on absenteeism. Each question was answered on a Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”), with a higher mean score indicating more negative work outcomes due to second victim experience. The Cronbach’s α of this subscale in this study was 0.86.
Demographic characteristics
Demographic characteristics of the study population included age, gender, marital status, religion, and education level. Job-related characteristics included total clinical career, weekly working hours, position, working type, employment type, work unit, annual income, and patient safety event experiences.
Data collection
This study included sensitive questions about nurses’ experiences of second victim from PSIs, which might make them reluctant to respond to direct data collection requests. To maintain anonymity of participants and enhance reliability and participation rate of this study, data were collected through an online survey. Data collection was conducted from December 28, 2022 to January 14, 2023. To recruit participants, the recruitment notice was publicized in nursing departments of tertiary general hospitals nationwide. For hospitals requiring specific procedures, such as obtaining institutional approval, necessary permissions were secured prior to data collection. An online survey was then distributed to eligible participants. To prevent under-representation, a ‘half-open interval’ or ‘multiplicity sampling’ strategy was employed. Specifically, the last question in the survey, “Would you be willing to recommend this study to colleagues or acquaintances who might be eligible to participate in this research?”, was asked. Respondents who answered “Yes, I am willing to recommend this survey” were provided with the survey link to share with eligible peers. Participants who accessed the survey through this recommendation were required to meet inclusion criteria of this study before proceeding. At the beginning of the survey, potential participants answered screening questions about their eligibility, including their employment at a tertiary general hospital, role as a staff or charge nurse in a ward or intensive care unit, PSI experience within the last six months, and at least one year of clinical experience. Participants who responded “No” to any of these questions were automatically directed to the end of the survey. Those who met all inclusion criteria were asked to provide informed consent before proceeding to the main survey.
Data analysis
Data analyses were performed using SPSS WIN 23.0 and AMOS 23.0 programs (IBM Corp. Armonk, NY, USA). The statistical significance level for this study was set at p < 0.05 with a 95% confidence interval (CI). Normality of the sample was tested using skewness and kurtosis. Descriptive statistics were generated for general characteristics of participants. Internal consistency of the survey instrument was determined by calculating Cronbach’s α. Pearson correlation coefficients, tolerance, and variance inflation factors (VIF) were calculated to rule out multicollinearity among independent variables. Autocorrelation of dependent variable was assessed using the Durbin-Watson value [31]. For goodness-of-fit and hypothesis testing of the path model, parameters of the model were estimated using the maximum likelihood method. The goodness of fit of the hypothesized model was checked with absolute goodness of fit indices such as normed χ2 (DF, CMIN) and root mean square error of approximation (RMSEA) and intermediate goodness of fit indices such as comparative fit index (CFI), Tucker-Lewis index (TLI), normed fit index (NFI), and adjusted goodness of fit index (AGFI) [26, 28]. To determine effects of JC on SVEs, path analysis was conducted to analyze significant path coefficients and direct/indirect/total effects between variables. Statistical significance test was set to 2,000 iterations using the bootstrapping approach. The explanatory power of exogenous variables on endogenous variables was calculated using squared multiple correlations (SMC). The significance of indirect effects was tested using the phantom variable modeling method [26, 28].
Anzeige
Ethical consideration
This study was approved by the Institutional Review Board (IRB) of Jeonbuk National University prior to data collection (IRB NO. JBNU 2022-11-001). All procedures were conducted in accordance with relevant guidelines and regulations, including the Declaration of Helsinki. The consent form in the online survey informed participants about the voluntary nature of participating in the study and their right to withdraw at any time. It also informed them that their responses would not be used for any purpose other than this study and that their responses would be confidential. After completing the survey, responses of each individual were stored and analyzed using a unique identification number to protect participant privacy.
Results
General characteristics
In terms of PSI experience, 51.9% (n = 95) of participants had experienced only near misses without causing harm to patients, 6.6% (n = 12) had experienced adverse or sentinel events without near misses, and 41.5% (n = 76) had experienced both near misses and adverse or sentinel events (Table 1).
Table 1
General characteristics of the study population (N = 183)
Variables | n (%) | M ± SD (range) | ||||||
|---|---|---|---|---|---|---|---|---|
Sociodemographic characteristics | ||||||||
Gender | ||||||||
Male | 12 (6.6) | |||||||
Female | 171 (93.4) | |||||||
Age (years) | ||||||||
23 –<30 | 69 (37.7) | 31.9 ± 6.1 | ||||||
30 –<40 | 92 (50.3) | (23–53) | ||||||
≥ 40 | 22 (12.0) | |||||||
Marital status | ||||||||
Single | 119 (65.0) | |||||||
Married | 64 (35.0) | |||||||
Religion | ||||||||
No | 113 (61.7) | |||||||
Protestantism | 40 (21.9) | |||||||
Catholicism | 23 (12.6) | |||||||
Buddhism | 7 (3.8) | |||||||
Education level | ||||||||
College | 21 (11.5) | |||||||
University | 132 (72.1) | |||||||
Graduate school | 16 (8.7) | |||||||
≥ Masters | 14 (7.7) | |||||||
Job-related characteristics | ||||||||
Clinical career (years) | ||||||||
1 –<3 | 42 (23.0) | 7.06 ± 5.60 | ||||||
3 –<5 | 41 (22.4) | (1–30) | ||||||
5 –<10 | 57 (31.1) | |||||||
≥ 10 | 43 (23.5) | |||||||
Position | ||||||||
Staff | 147 (80.3) | |||||||
Charge | 36 (19.7) | |||||||
Working type | ||||||||
Non-shift | 25 (13.7) | |||||||
Shift | 158 (86.3) | |||||||
Employment type | ||||||||
Full-time worker | 177 (96.7) | |||||||
Contract worker | 6 (3.3) | |||||||
Nursing unit | ||||||||
General ward | 142 (77.6) | |||||||
ICU | 41 (22.4) | |||||||
Weekly working hours | ||||||||
< 30 | 5 (2.7) | |||||||
30 – < 40 | 80 (43.7) | |||||||
≥ 40 | 98 (53.6) | |||||||
Annual income (10,000 Won) | ||||||||
< 4,000 | 41 (22.4) | |||||||
4,000 – < 6,000 | 99 (54.1) | |||||||
≥ 6,000 | 43 (23.5) | |||||||
Patient safety incident experiences | ||||||||
Incident A | 95 (51.9) | |||||||
Incident B | 12 (6.6) | |||||||
Incident C | 76 (41.5) | |||||||
Descriptions and correlations of study variables
The overall mean score of participants’ perceived JC was 4.42 (SD = 0.76) out of 7. Overall mean scores of SVD, demand for support, and negative work-related outcomes were 3.21 (SD = 0.63), 2.69 (SD = 0.48), and 3.28 (SD = 0.97) out of 5, respectively (Table 2).
Anzeige
Table 2
List of study variables and correlations of related variables (N = 183)
Variables | Just culture | Second victim distress | Demand for support | Negative work-related outcome | VIF | Tolerance | ||||
|---|---|---|---|---|---|---|---|---|---|---|
r(p) | r(p) | r(p) | r(p) | |||||||
Just culture | 1.94 | 0.52 | ||||||||
Second victim distress | − 0.29 (< 0.001) | 1.13 | 0.88 | |||||||
Demand for support | − 0.69 (< 0.001) | 0.33 (< 0.001) | 1.99 | 0.50 | ||||||
Negative work-related outcome | − 0.37 (< 0.001) | 0.72 (< 0.001) | 0.40 (< 0.001) | |||||||
Mean ± SD | 4.42 ± 0.76 | 3.21 ± 0.63 | 2.69 ± 0.48 | 3.28 ± 0.97 | ||||||
Possible range | 1–7 | 1–5 | 1–5 | 1–5 | ||||||
Min | 3.04 | 1.58 | 1.15 | 1.00 | ||||||
Max | 7.00 | 4.92 | 4.08 | 5.00 | ||||||
Standardized skewness | 0.54 | -0.05 | -0.30 | -0.29 | ||||||
Standardized kurtosis | 0.66 | 0.01 | 0.79 | -0.55 |
Participants’ perceived JC showed significant negative correlations with SVD (r = -0.29, p < 0.001), demand for support (r = -0.69, p < 0.001), and negative work-related outcomes (r = -0.37, p < 0.001). SVD showed significant positive correlations with demand for support (r = 0.33, p < 0.001) and negative work-related outcomes (r = 0.72, p < 0.001). Demand for support showed a significant positive correlation with negative work-related outcomes (r = 0.40, p < 0.001) (Table 2).
Hypothesized model fit
Normality and multicollinearity
Absolute values for skewness (-0.30–0.54) and kurtosis (-0.55–0.79) of study variables met their respective normality criteria. The VIF ranged from 1.13 to 1.99, all less than 10. Tolerance ranged from 0.50 to 0.88, all greater than 0.1. Inter-variable correlation coefficient ranged from − 0.69 to 0.33, all less than 0.8, indicating no multicollinearity among measured independent variables. In addition, the Durbin-Watson index was 1.98 (dU = 1.79 < d < 4-dU = 2.21), confirming that dependent variables were independent without autocorrelation [31]. Thus, the model fulfilled the assumption of path analysis (Table 2).
Goodness of fit test
The maximum likelihood method was used to test the fit of the hypothesized model. Results of goodness-of-fit indices of the model were: normed χ2 (1, 2.53) = 2.53 (p = 0.112), RMSEA = 0.09, CFI = 0.99, TLI = 0.97, NFI = 0.99, and AGFI = 0.93. These results indicated that the model was appropriate for this study [26, 28].
Path analysis: testing hypotheses and variable effects
In this study, path analysis was conducted to test the hypothesized model for effects of a healthcare organization’s JC perceived by clinical nurses who experienced PSIs on their SVEs. Results are as follows (Table 3; Fig. 2).
Table 3
Direct, indirect, and total effects of the hypothetical model (N = 183)
Endogenous variables | Exogenous variables | Hypothetical model | Direct effect | Indirect effect | Total effect | SMC | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
B | SE | CR | p-value | β (p) | 95% CI | β (p) | 95% CI | β (p) | 95% CI | |||||||
Second victim distress | ← | Just culture | -0.24 | 0.06 | -4.01 | < 0.001 | − 0.29 (0.001) | − 0.44–-0.14 | - | - | - | − 0.29 (0.001) | − 0.44–-0.14 | 0.08 | ||
Demand for support | ← | Just culture | -0.41 | 0.04 | -11.89 | < 0.001 | − 0.65 (0.001) | − 0.75–-0.54 | − 0.04 (0.016) | − 0.09–-0.01 | − 0.69 (0.001) | − 0.77–-0.59 | 0.50 | |||
← | Second victim distress | 0.11 | 0.04 | 2.56 | 0.011 | 0.14 (0.025) | 0.01–0.27 | - | - | 0.14 (0.025) | 0.01–0.27 | |||||
Negative work-related outcome | ← | Just culture | - | - | - | - | - | - | − 0.32 (0.001) | − 0.44–-0.17 | − 0.32 (0.001) | − 0.44–-0.17 | 0.54 | |||
← | Second victim distress | 1.02. | 0.08 | 12.44 | < 0.001 | 0.66 (0.001) | 0.56–0.75 | 0.03 (0.018) | 0.00–0.07 | 0.68 (0.001) | 0.59–0.76 | |||||
← | Demand for support | 0.36 | 0.11 | 3.35 | < 0.001 | 0.18 (0.010) | 0.05–0.31 | - | - | 0.18 (0.010) | 0.05–0.31 | |||||
Significance Testing Indirect Effects | ||||||||||||||||
Path | Phantom variables | Indirect effect | ||||||||||||||
B (p) | 95% CI | |||||||||||||||
Just culture → Second victim distress → Negative work-related outcome | P1 → P12 | -0.24 (0.001) | − 0.37–-0.12 | |||||||||||||
Just culture → Demand for support → Negative work-related outcome | P2 → P22 | -0.15 (0.010) | − 0.26–-0.04 | |||||||||||||
Just culture → Second victim distress → Demand for support → Negative work-related outcome | P1 → P13 → P14 | -0.01 (0.013) | − 0.03–-0.01 | |||||||||||||
Fig. 2
Path diagram of the hypothetical model
×
![]()
All five direct effect paths in the hypothesized model were statistically significant. JC had statistically significant effects on SVD (β = -0.29, p = 0.001, 95% CI [-0.44, -0.14]) and demand for support (β = -0.65, p = 0.001, 95% CI [-0.75, -0.54]). SVD had statistically significant effects on demand for support (β = 0.14, p = 0.025, 95% CI [0.01, 0.27]) and negative work-related outcomes (β = 0.66, p = 0.001, 95% CI [0.56, 0.75]). Demand for support had a statistically significant effect on negative work-related outcomes (β = 0.18, p = 0.010, 95% CI [0.05, 0.31]) (Table 3; Fig. 2).
In addition, all three indirect effect paths in the model were statistically significant. The indirect effect of JC on demand for support through SVD was statistically significant (β = -0.04, p = 0.016, 95% CI [-0.09, -0.01]). The indirect effect of JC on negative work-related outcomes through SVD and demand for support was also statistically significant (β = -0.32, p = 0.001, 95% CI [-0.44, -0.17]). In addition, the indirect effect of SVD on negative work-related outcomes through demand for support was statistically significant (β = 0.03, p = 0.018, 95% CI [0.00, 0.07]).
Using phantom variable modeling to test the statistical significance of the indirect effect of JC on negative work-related outcomes, the path of “JC → SVD → negative work-related outcomes” was represented by phantom variable “P1” → “P12”. The indirect effect of JC on negative work-related outcomes was statistically significant (B = -0.24, p = 0.001, 95% CI [-0.37, -0.12]). This indicates that fair culture can affect negative work-related outcomes through SVD, with SVD having a mediating effect. In addition, the path of “JC → demand for support → negative work-related outcomes” was represented by the phantom variable “P2” → “P22”. The indirect effect of JC on negative work-related outcomes was also statistically significant (B = -0.15, p = 0.010, 95% CI [-0.26, -0.04]). This indicates that fair culture can affect negative work-related outcomes through demand for support, with demand for support having a mediating effect. The path of “JC → SVD → demand for support → negative work-related outcomes” was represented by phantom variables “P1” → “P13” → “P14”. The indirect effect of JC on negative work-related outcomes was statistically significant (B = -0.01, p = 0.013, 95% CI [-0.03, -0.01]). This indicates that JC can affect negative work-related outcomes through SVD and demand for support, indicating a dual mediation effect of SVD and demand for support.
Discussion
This study attempted to establish a hypothetical model, test the model’s fit, and identify direct and indirect effects of factors affecting second victim’s perceived JC of healthcare organizations and experiences of clinical nurses as second victims of PSIs. Five direct effect paths and three indirect effect paths were supported among eight paths proposed in the hypothesized model. In other words, JC was found to affect SVD and demand for support. In addition, pathways through which these mediators could directly or indirectly affect negative work-related outcomes were identified.
A JC was identified as a variable that could directly influence SVD and demand for support, with perception of a JC being associated with lower levels of SVD and demand for support among nurses who experienced a PSI, consistent with previous findings showing that higher perceived levels of “non-punitive response to errors” having similar attributes to a JC was related to lower levels of SVD and higher levels of second victim support provided to nurses [8‐10]. A JC not only focuses on systemic improvements in the work environment rather than punishing individuals for PSIs, but also holds individuals accountable for institutional compliance [19]. Because of this balancing act, an organization’s JC is necessary to provide a support system to reduce SVD experienced after a PSI [32]. Reducing punitive responses to PSIs and providing support through interactions with colleagues, supervisors, and the organization can help mitigate SVD [10]. A study comparing the perception of JC among international nurses has found that nurses who have received support after SVEs have a higher perception of JC than nurses who have no experience as second victims [33]. Our findings suggest that implementing proactive measures to establish a JC in medical institutions is a key imperative to reduce SVD and demand for support among nurses. Therefore, healthcare leaders must recognize the importance of a JC and implement policies to promote its awareness and integration within their organizations. Providing administrative support to embed a JC in institutional culture can help reduce SVD and demand for support among nurses.
SVD was found to have the most substantial direct influence on negative work-related outcomes. It also had a direct impact on demand for support. Our findings were consistent with those of previous studies [7, 9, 34] showing that the more severe the SVD of nurses who experienced a PSI, the higher the negative work-related outcomes and demand for support. Negative work-related outcomes such as nurse turnover or absenteeism not only entail significant costs for healthcare organizations, but also have adverse implications for patient safety [10]. Therefore, it is imperative to prevent secondary victimization of nurses involved in a PSI [35]. While it is essential for healthcare institutions to offer various educational programs aimed at preventing patient safety incidents, it is also essential to provide training that equips nurses with strategies to overcome SVEs [36]. For nurses experiencing SVD, empathetic support from colleagues and active support from institutions through work assignments or work adjustments, counseling support, and improvement of the working environment will be helpful [9]. A previous study [13] has found that support needed varies depending on the degree of SVD. Therefore, it is important to provide individualized support to nurses involved in a PSI based on the assessment of SVD.
In addition, there was a significant indirect effect of JC on negative work-related outcomes mediated via SVD. This suggests that a more robust implementation of a JC within an organization may help reduce SVD, consequently leading to diminished negative work-related outcomes. This study provides empirical evidence that SVD, a factor having the most significant influence on negative work-related outcomes, can be mitigated by establishing a JC in healthcare organizations. This necessitates a systematic approach grounded on a JC rather than focusing on assigning individual responsibility when such incidents occur.
Demand for support was identified as a variable that could directly affect negative work-related outcomes. The higher the demand for support among nurses who experienced PSIs, the higher the negative work-related outcomes. Our results were similar to those of previous studies showing that second victim support was a significant predictor of reducing negative work-related outcomes [7, 9, 34]. High demand for support among nurses may indicate a lack of adequate support from colleagues, supervisors, organizations, family, or friends following a PSI, which can lead to negative work-related outcomes such as turnover or absenteeism. A previous study on international nurses [9] has found a direct effect of organizational support on nurses’ absenteeism, but not on their turnover intentions. Future research should aim to identify types of support systems that are effective in reducing negative work-related outcomes, including absenteeism and turnover intentions, in nurses.
In addition, the indirect effect of JC on negative work-related outcomes mediated via demand for support was significant. This implies that a more robust implementation of a JC within the organization can lead to a decreased demand for support among nurses by enhancing ‘feedback and communication’, ‘openness of communication’, ‘continuous improvement’, and so on, ultimately resulting in fewer negative work-related outcomes. It is evident that a healthcare organization with a JC already has an organizational framework to address PSIs in a systemic manner rather than on an individual basis. This entails open communication regarding incidents, provision of feedback, and execution of ongoing improvement initiatives. Empirical evidence from this study suggests that a more robust implementation of a JC within the organization can alleviate nurses’ demand for support and reduce negative work-related outcomes.
The indirect effect of JC on negative work-related outcomes was significant through dual mediation of SVD and demand for support. This suggests that the perception of a JC in an organization can reduce SVD and demand for support, which in turn can reduce negative work-related outcomes. While previous studies have reported a mediating effect of second victim support on the relationship between SVD and negative work-related outcomes [7], the present study analyzed the causal relationship between JC, SVD, demand for support, and negative work-related outcomes, thus providing a scientific basis for the positive influence of JC in healthcare organizations on SVEs of nurses involved in PSIs.
Collectively, findings of this study suggest that JC can reduce SVD and demand for support to improve negative outcomes of SVEs while maintaining a long-term nurse’s tenure and further promote a culture of trust and responsibility that can improve patients’ treatment outcomes. Currently, many countries are promoting the establishment of a JC in pursuit of high credibility [37]. However, actual effects of JC on patient safety, quality in healthcare, and healthcare personnel involved in PSIs remains unclear. In this context, this study holds significance in that it empirically illustrates positive impacts of a JC. Moreover, this is the first study to demonstrate direct and indirect effects of JC in diminishing SVEs of clinical nurses following PSIs. Our findings underline the need for concerted efforts to establish a JC in healthcare organizations through active participation of employees. In addition, clinical nurses should work together to establish a balanced and trustworthy organizational culture that pursues a JC by fulfilling their individual responsibilities. Additionally, based on findings of this study, further research is needed to develop a more comprehensive theoretical framework on the relationship between JC and SVEs.
This study adds to existing evidence that SVEs of nurses who experience PSIs can negatively impact healthcare organizations and patient safety. It is hoped that medical institutions will adopt established practices such as an organizational-level patient safety event response process [38] or establish a secondary victim support team [39], both of which have been shown to be effective internationally. Existing peer support programs shown to be effective in addressing SVEs [36] should be actively implemented to provide effective support for nurses experiencing SVEs. Moreover, active engagement and support from both the government and medical institutions are required to facilitate the development and implementation of such programs.
This study has some limitations. First, the study design did not allow for an isolated assessment of individual impacts of each of the six sub-attributes of JC on SVEs, specifically SVD, demand for support, and negative work-related outcomes. Second, the use of online surveys with predominantly closed-ended questions often limits the ability to capture nuanced insights into a topic. Third, although this study’s methodology, which used quantitative analysis, provides a lot of easily analyzable data, it lacked insight into why the phenomenon under study occurred and its context. Fourth, this was a cross-sectional study that did not capture longitudinal variations in study variables over time. Despite these limitations, our study is significant as it represents the initial empirical exploration of linkages between JC and SVEs within the context of clinical nursing practice.
Conclusions
This study demonstrated the impact of a healthcare organization’s JC on SVEs perceived by clinical nurses involved in PSIs, with JC being the main factor having direct effects on SVD and demand for support. JC had an indirect effect on negative work-related outcomes via SVD and demand for support. PSIs are inevitable in healthcare settings. To prevent SVD and negative work-related outcomes, medical organizations should establish a JC through a systematic approach to improvement rather than blaming individuals. In addition, strategies should be implemented to monitor nurses who have experienced PSIs as second victims and develop a support system to assist nurses who experience SVD.
In light of our study findings, the following recommendations are proposed. First, it is recommended to assess the SVD of nurses who have experienced PSIs and to develop and support customized interventions to reduce negative work-related outcomes in clinical settings. Second, a team dedicated to second victim support needs to be established in clinical settings to reduce SVD and negative work-related outcomes of nurses and other healthcare workers. Third, future research should adopt mixed method approaches to gain a deeper understanding of individual perceptions and contextual factors. Fourth, a longitudinal study needs to be conducted to assess enduring negative work-related outcomes experienced by nurses with SVEs.
Acknowledgements
The authors thank all nurses who participated in the survey for this study.
Declarations
Ethics approval and consent to participate
This study was approved by the Institutional Review Board (IRB) of Jeonbuk National University prior to data collection (IRB NO. JBNU 2022-11-001). The consent form in the online survey informed participants about the voluntary nature of participation in this study and their right to withdraw at any time. It also informed them that their responses would not be used for any purpose other than this study and that their responses would be confidential. After completing the survey, responses of each individual were stored and analyzed using a unique identification number to protect participant privacy.
Consent for publication
Not applicable.
Competing interests
Sunmi Kim, Seohee Jeong, and Seokhee Jeong have been the members of Global Korea Nursing Foundation-Korea (GKNF) but has no role in review process of funding. Except for that, no potential conflict of interest relevant to this article was reported.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
