Mediator Effects of Cognitive Load on Association between Self-Efficacy and Task Load in Intensive Care Unit Nurses

,


Introduction
Te demand for nursing services is constantly increasing, and nurses are the most direct personnel to provide nursing care to patients, especially in intensive care units (ICUs) where nursing activities are characterized by extremely demanding workloads [1,2].ICU nurses need to possess the ability to make clinical decisions quickly, handle complex changes in patient conditions, and use special equipment.Te complexity and urgency of the ICU environment often expose nurses to high-level mental task loads [1,3].Paula et al. [3] investigated 111 ICU nurses and 100% indicated a medium to high level of mental task load.Nursing administrators and researchers are paying more attention to the task load of nursing care.High-level task loads among ICU nurses can lead to physical and psychological problems, such as the increased risk of anxiety and depression, higher occupational fatigue and turnover rate, missed nursing care, and increased nursing errors [4,5].For example, Boehm and colleagues surveyed 268 ICU nurses and found that for every additional point increase in task load (rated on a scale of 0-10), adherence to bundle interventions among ICU nurses decreased by 53% [4].
Studies have shown that self-efcacy and cognitive load were related to an individual's mental task load [6][7][8].Selfefcacy is defned as the level of belief in one's ability to complete a heavy workload successfully [9].Cognitive load is defned as the total amount of cognitive resources that a person requires to process cognitive tasks [10] and is composed of intrinsic, extraneous, and germane cognitive load [11].Intrinsic cognitive load is defned as the demand for working memory in processing information elements and their interactions in related tasks [12].Extraneous cognitive load is generated by inappropriate activity presentation rather than the task itself [12].Germane cognitive load increases schema construction, therefore promoting the execution of the task [12,13].Broad evidence has shown that self-efcacy also afects an individual's cognitive load [14,15].Jiang et al. [14] reported that self-efcacy in English reading among students is negatively related to intrinsic cognitive load.However, the mediating efect of cognitive load between self-efcacy and task load is unclear.Terefore, the study aimed to investigate the mediator efects of cognitive load on the association between self-efcacy and task load of ICU nurses.

Study Design and Samples.
A cross-sectional questionnaire study design was conducted with convenience sampling at three tertiary hospitals between February and June 2023.
Registered nurses were eligible for the study if they consented to participate in this study.Nurses were excluded if they (a) had less than 1 year of intensive care experience, (b) currently did not work full-time in the ICU, and (c) had long leave with pay, such as sick or prenatal leave.A simple structural equation model (SEM) was used to analyze the relationship among self-efcacy, cognitive load, and task load, and a minimum of 200 patients was required to run the model as suggested by Shah and Goldstein [16].

Demographic Information.
Demographic information included age, gender, education level, marital status, number of children, living situation, years of ICU working experience, professional title, number of night shifts per month, and number of patients cared for during the whole shift.

National Aeronautics and Space Administration-Task
Load Index.Te NASA-TLX was used to assess task load and was developed by Hart [17] and was translated into Chinese by Liling Liang and colleagues [18].Te NASA-TLX has 6 items: mental demand, physical demand, temporal demand, performance, efort, and frustration.Each item is scored on a Likert scale from 0 (low load) to 20 (high load), and the total score ranges from 0 to 120, with higher scores indicating a higher level of task load.Te retest reliability of the scale was 0.806, and Cronbach's α coefcient was 0.707 [18].Content validity in the Chinese version of NASA-TLX was 0.900, and the split-half reliability was 0.808 [19].Te Chinese version of NASA-TLX has good consistency between each item and validity.
2.5.General Self-Efcacy Scale.Te GSES was developed by Schwarzer [9] and was translated into Chinese version by Caikang Wang et al. [20].Te GSES consists of 10 items, and each item is scored on a 4-point Likert scale ranging from 1 ("not at all true") to 4 ("exactly true").Example items are "If you try hard, you can always accomplish a task efciently" and "Even objected by others, you still can manage to get what you want."A single score for the GSES is computed using the total score of all items.A higher total score indicates a higher level of self-efcacy.Te Chinese version of the GSES has shown good reliability and validity for Chinese adults [20,21].Cronbach's α coefcient was 0.870, retest reliability was 0.830 (P < 0.001), and split-half reliability was 0.820 (n � 401, P < 0.001) [20].

An Instrument for Measuring Diferent Types of Cognitive
Load.Te MDT-CL was used to assess three types of cognitive load.Te MDT-CL was developed by Leppink and colleagues [12] and was translated into Chinese version by Zhang et al. [13].A ten-item questionnaire was presented for the measurement of intrinsic cognitive load (items 1, 2, and 3), extraneous cognitive load (items 4, 5, and 6), and germane cognitive load (items 7, 8, 9, and 10).Example items are "Te intensive care activity/activities presented in daily usual care was/were very complex" for intrinsic cognitive load, "Te instructions and/or explanations in daily usual care during implementing intensive care activity/activities were very unclear" for extraneous cognitive load, and "Daily usual care enhanced my understanding of the intensive care activity/activities covered" for germane cognitive load.Each item scores from 0 ("not at all") to 10 ("completely"), the higher the score, the higher the cognitive load.Te Chinese version MDT-CL has good reliability and validity, and Cronbach's α was 0.818; Cronbach's α in measurement of intrinsic, extraneous, and germane cognitive load was 0.879, 0.878, and 0.946, respectively [13].

Data Collection.
Before the study commencement, the investigators received research training about how to use standardized instruments to ensure the collected data were consistent and accurate.Data collectors recruited ICU nurses based on the inclusion and exclusion criteria, and the aim and procedure of the study were fully explained to all eligible nurses before the recruitment.Tis survey was conducted anonymously, utilizing online responses via mobile phones.To ensure the accuracy and efectiveness of the questionnaire, any questions related to the study were answered by the investigators during the completion of the questionnaire.All of the questions refer to the daily nursing activity that just fnished.For instance, "Te nursing activity presented in daily usual care that I perceived as very complex," "Te instructions and/or explanations in daily usual care were, in terms of clinical application, very 2 Journal of Nursing Management inefective," and "Daily usual care enhanced my knowledge and understanding of nursing."After completing the questionnaire, all data were entered into a database by two researchers who conducted extensive error and validity checks.Te quality of the questionnaires was checked, and any questionnaires with identical answers or missing answers to more than two-thirds of the questions were excluded.
2.8.Informed Consent and Ethical Considerations.Te institutional review board of the university approved this study (approval number: Z2019SY021).Written informed consent was obtained from all ICU nurses after the investigators introduced the study procedures in detail.Te eligible ICU nurses had the right to withdraw from the study at any time without any harmful consequences.
2.9.Data Analysis.Te SPSS version 25.0 (SPSS Inc., Chicago, IL) was used for data analyses, cases with missing data in self-efcacy, cognitive load, and task load, or with missing covariates (age, gender, educational level, and years of ICU working experience) >35% were excluded from the fnal analysis, and others were processed using multiple imputations if the missing is in random [22].Continuous variables were described as means and standard deviation (SD) for normal and medians and interquartile range for abnormally distributed data.Te relationship among self-efcacy, cognitive load (intrinsic, extraneous, and germane cognitive load), and task load (dichotomous variable) was verifed using the Pearson correlation test.Finally, the signifcant factors (P < 0.05) were selected to enter a multiple linear regression analysis.Te standardization coefcient (β) and standard error (SE) were used to express the strength of the association.Finally, all signifcant variables that were identifed by the multivariate linear regression were entered into the structural equation model (SEM).Harman's single-factor test was used to examine common method variance with self-report measures [23].Model 4 test in PROCESS in SPSS (an addon for SPSS) was used for mediation analysis of task load, self-efcacy, and cognitive load [24].Te bootstrap method with a 95% bias-corrected confdence interval (CI) was used to test the signifcance of the mediation efect.Te statistical signifcance was set at P < 0.05, with two-tailed testing.As shown in Table 2, statistically signifcant diferences were observed between task load and self-efcacy (r � −0.437, P < 0.001), task load and intrinsic cognitive load (r � 0.373, P < 0.001), and task load and extraneous cognitive load (r � 0.636, P < 0.001).However, no signifcant diferences were found between task load and germane cognitive load (r � −0.051, P > 0.05).

Factors Afecting Task
Load.Te self-efcacy, intrinsic cognitive load, and extraneous cognitive load were entered into regression analysis with enter forwards selection due to these factors being statistically related to task load.As shown in Table 3, after adjusting covariates (age, gender, educational level, and years of ICU working experience), the results indicated that only self-efcacy (β′ � −0.202, P < 0.001) and extraneous cognitive load (β′ � −0.520, P < 0.001) were statistically explained task load.

Structural Model.
Harman's single-factor test showed that the cumulative variance interpretation was 28.26%, which is lower than the critical standard of 40% [25], illustrating that common method variance was not signifcant in this study.

Discussion
In this cross-sectional design study, 253 ICU nurses were investigated to explore the mediator efects of cognitive load between self-efcacy and task load.We found that 42.69% of ICU nurses had high levels task load, and correlations analysis reported self-efcacy was negatively related to task load, but intrinsic and extraneous cognitive load were positively related to task load.However, after adjusting covariates in multiple linear regression analysis, only selfefcacy and extraneous cognitive load were statistically explained task load.A three-factor mediation model was constructed and tested, and the relationships between self-efcacy and task load were partially mediated by extraneous cognitive load among ICU nurses, accounting for 51.64% of the total efect.Te fndings imply that even though the perceived task load of ICU nurses was afected by both poor self-efcacy and high-level extraneous cognitive load from daily intensive care activities, the infuence of selfefcacy can be increased if extraneous cognitive load declines.Terefore, early assessment of self-efcacy and cognitive load for ICU nurses may beneft from targeted prevention strategies.
Te partially mediating role of extraneous cognitive load between overall self-efcacy and task load of ICU nurses was confrmed in our study.In other words, an individual's selfefcacy not only has a direct efect on task load but also has an indirect efect through extraneous cognitive load.ICU nurses with high-level self-efcacy are always able to cope with unexpected issues and come up with solving strategies [26,27].Tese abilities weaken the high level of the extraneous cognitive load caused by unclear instructions for nursing activities in daily nursing practice, thereby reducing the task load in terms of mental demand, physical demand, temporal demand, performance, efort, and frustration for ICU nurses.Highlighting the meaningfulness of high-level self-efcacy and low-level extraneous cognitive load might contribute to a decreased perceived task load.
Regarding the relationship between self-efcacy and perceived task load among ICU nurses, it has been found that high levels of self-efcacy were associated with lower levels of task load.Te fndings from this study were confrmed by other studies [7,8].Cayupe et al. [7] recruited 300 primary school teachers to examine the mediating role of job satisfaction between self-efcacy, life satisfaction, workload, and overall life satisfaction.Tey also found that self-efcacy was negatively related to workload (r � −0.43).In addition, Molero et al. [8] investigated 1307 nurses in a cross-sectional study and reported that workload had a signifcant negative relationship with perceived self-efcacy (r � −0.07; P < 0.01).
It is likely because ICU nurses with high-level of selfefcacy are more capable of coping with the workload brought on by the high-intensity working environment.Tey are more likely to adopt proactive strategies to alleviate challenges from intensive tasks, such as participating in education programs, fexibly adjusting resource allocation, generating efective work plans, and time management strategies, and having good team cooperation ability; these strategies could help ICU nurses alleviated the perceived task load [28][29][30].Terefore, the nursing administrator should provide job support to increase self-efcacy among ICU nurses, which would empower and motivate them to cope with an intensive task load.
Our study also examined the correlation between extraneous cognitive load and task load, indicating that the high level of extraneous cognitive load from unclear instructions in daily intensive care activities was positively related to higher perceived task load.Nasirizad and colleagues [31] conducted a cross-sectional study among 105 ICU nurses.Tey confrmed a signifcant relationship between cognitive and task load (P < 0.001).Based on the Cognitive Load Teory (CLT), the human cognitive resources are limited, and once the cognitive resources required for the cognitive tasks to be processed exceed the total cognitive resources of human beings, cognitive load increases and results in unsatisfactory task performance [10].Te extraneous cognitive load can be derived from suboptimal presentation methods and unclear instruction of activities [12].Terefore, nursing administrators should identify and modify the factors that increase the extraneous cognitive load in nursing practice, for example, by simplifying nursing processes and improving collaboration mechanisms.In addition, administrators can introduce appropriate information technology-assisted tools, such as intelligent devices, to improve the presentation mode of interventions, provide decision support, and alleviate the nurse's extraneous cognitive load [32,33].

Limitations and Future
Research.Tis study has several limitations.First, the ICU nurses were recruited from three tertiary hospitals in one region, which limits the generalisability of the study fndings to diferent populations and regions.In the future, expanding the sample collection is recommended to identify diferences between region levels.Second, the nature of the cross-sectional study limited the possibility of drawing causal relationships among the variables.To gain a better understanding of the relationship between self-efcacy, cognitive load, and task load in ICU nurses, prospective longitudinal studies were recommended to confrm causal relationships in the future.Finally, the investigated variables in this study were collected by self-report questionnaires, which may lead to recall bias.For future studies, objective    Journal of Nursing Management measurements of task load, such as pupil diameter, heart rate variability (HRV), and electroencephalography (EEG), are recommended [34,35].

Conclusions
In this cross-sectional study, we concluded that self-efcacy and lower extraneous cognitive load could help defend against the adverse impact of increased task load among ICU nurses.Self-efcacy infuences the task load, and extraneous cognitive load partially mediates the correlation.Terefore, nursing managers should strengthen the assessment and monitoring of self-efcacy and cognitive load in ICU nurses and tailor efective support strategies to reduce task load.To sum up, self-efcacy is a crucial resource in the ICU nursing work environment with positive impacts on well-being (e.g., reducing physical demand and frustration).

Implications for Nursing Management
Given that self-efcacy has benefcial consequences for nurses' cognitive load and task load, interventions directed at three factors can be carried out.Nursing administrators need to pay attention to the measurement of nurses' self-efcacy, cognitive load, and task load, focus on the presentation of nursing activities and other factors that may afect the source of task load for ICU nurses, and provide targeted intervention measures.
For example, nursing administrators should promote the development of strong cooperation among nursing teams (e.g., gratitude for the help and mutual appreciation) to work together in high-pressure and complex ICU work environments.Tese efective strategies for enhancing self-efcacy and improving the cognitive load of nurses can help them to reduce work burden and pressure, thereby improving their job satisfaction, physical and mental health, and work performance.

.
Te questionnaire consisted of four parts: (a) demographic information; (b) the Chinese National Aeronautics and Space Administration-Task Load Index (NASA-TLX); (c) the Chinese General Self-Efcacy Scale (GSES); and (d) an instrument for Measuring Diferent Types of Cognitive Load (MDT-CL).Cronbach's α coefcient for the combined questionnaires was 0.732, indicating that the Chinese version of the NASA-TLX, GSES, and MDT-CL had acceptable reliability for ICU nurses.

Figure 1 :
Figure 1: Mediation model of extraneous cognitive load on the relationship between self-efcacy and task load of ICU nurses.Note: All coefcients are signifcant (P < 0.001).

Table 1
). Te mean years of ICU working experience were 9.81 years (SD � 5.18), and most of the ICU nurses (86.6%) reported working night shifts.

Table 1 :
Te sociodemographic characteristics of ICU nurses (N � 253).Correlation among Major Variables.Te ICU nurses showed high levels of task load with an overall NASA-TLX score of 86.49 ± 15.05, and 42.69% (108/253) of ICU nurses had a high-level task load (overall NASA-TLX score ≥90).

Table 2 :
Descriptive statistics and correlations of the study variables (N � 253).

Table 3 :
Te results of multiple linear regression analysis on the infuencing factors of task load in ICU nurses (N � 253).

Table 4 :
Efect analysis of extraneous cognitive load as a mediator between self-efcacy and task load in ICU nurses (N � 253).Note.Adjusting for covariates, including age, gender, educational level, and years of ICU working experience.