The associations of psychological burnout and time factors on medication errors in rotating shift nurses in Korea: A cross sectional descriptive study

Abstract

Aim

To describe the associations of psychological burnout and time factors on hospital nurses' medication errors.

Design

A cross‐sectional survey design was used.

Methods

A structured questionnaire pertaining to psychological burnout, time factors and medication error was administered to 200 bedside nurses working at two tertiary university hospitals in Korea. The associations between the psychological burnout, time factor and medication error were analysed with the zero‐inflated negative binomial regression for over‐dispersed and over‐abundant zeros count data.

Results

Higher psychological burnout, shorter meal time during duty and longer weekly overtime were associated with an increased likelihood of medication error of nurses working in tertiary university hospitals. For medication safety, nurse managers should provide appropriate administrative support to nurses to cope with psychological burnout of nurses. Work time management should also be considered as human factors to satisfy the needs of nurses, such as securing meal times and maintaining a low level of weekly overtime.

1. INTRODUCTION

The medication delivery process involves a series of complex and interconnected steps, ranging from drug prescription to preparation, dispensing and administration; and nurses are expected to adequately perform each of these tasks (Manias et al., 2019). According to National Coordinating Council for Medication Error Reduction and Prevention (2020), medication errors are defined as ‘any preventable event that may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the health care professional, patient, or consumer,’ which may also result in considerable economic burdens (Walsh et al., 2017).

Medication errors negatively impact health systems worldwide. For example, 10% of all hospital admissions in the United Kingdom result in adverse events caused by a medication error (Kongkaew et al., 2013) and 1.6 adverse events per 1000 patient days in United States (Choi et al., 2016). Across a 5‐year period in Australia, 5.73 medication errors per every 1000 bed‐days, or 0.56% per admission, was reported by researchers for a paediatric hospital (Manias et al., 2019). The economic impact for avoidable medication errors is reported to be as much as £ 98.5 million per year in the United Kingdom (Elliott et al., 2021) and $42 billion in the United States (Aitken & Gorokhovich, 2012). The additional cost of treatment attributed to medication errors is about $8600 per patient (Choi et al., 2016).

The causes of medication errors in hospitals can be grouped into person‐related factors, communication‐related factors within the treatment team or between treatment team and the patient, and environment‐related factors of medication processes (Manias et al., 2019). Among them, the most common person‐related factors are performance deficit and inadequate screening of patient, which account for nearly 30% of total medication errors (Manias et al., 2019). Fatigue, stress and psychological burnout are error provoking conditions that affect person‐related factors (Dall'Ora et al., 2016; Keers et al., 2013). As an essential part of the clinical environment, time is an additional important factor contributing to medication errors (Westley et al., 2020). Nurses work at the human limits and their capabilities as they engage in multiple tasks with frequent interruptions under high cognitive load and physical overload (DeLucia et al., 2009).

2. BACKGROUND

Psychological burnout is a prolonged psychological reaction comprising emotional exhaustion, a tendency to depersonalize client encounters and reduced personal accomplishment (Maslach, 1999). According to Maslach (1999), psychological burnout is a state caused by excessive workload and demand, low work control, inadequate reward, low social support, unfairness at the work, conflicting values. A high level of psychological burnout not only adversely affects the health of workers but also directly affects their job performance (Kakemam et al., 2021; Montgomery et al., 2021; Sullivan et al., 2022). For nurses, psychological burnout of nurses affects their vigilance and reduces their alertness (Dall'Ora et al., 2020), which results in poor job performance (Montgomery et al., 2021; Nantsupawat et al., 2016). In particular, nurses with high level of psychological burnout report feeling incapable of focusing on the individual needs of patients, and perceiving their work as inhuman and repetitive Dall'Ora et al. (2020). Thus, these nurses are more likely to make errors in the complex and multi‐step process of preparing and administering medication (Montgomery et al., 2021).

In the clinical context, time‐related factors are discussed as an important aspect of the human factors for nurses (Min et al., 2021). For example, objective work time factors, such as rest and meal time during duty, the number of consecutive shifts, overtime, and recovery time between the shift are related to nurses' occupational fatigue and care quality (Min et al., 2021). In addition to objective work time factors, subjectively perceived time pressure has an influence on nurses' work performance (Teng et al., 2010).

The rapidly changing patient conditions, unpredictable worsening of patients, multi‐disciplinary collaboration and multi‐step work procedures are all unique characteristics of nursing work conditions. As a result, nurses are constantly subjected to changes in workload and are expected to work with time constraints (Darawad et al., 2015). Even if two nurses work simultaneously, their workload may differ due to patient acuity, task complexity, and care requirements. Even within the same working hours, this variance in work complexity affects several time patterns, such as meal and rest breaks, in addition to the total working hours.

The quality of patient care is impacted by the availability of time for nurses to provide humanized care (Nejati et al., 2016); an essential requirement to reduce medication errors (Cho et al., 2016; Min et al., 2019). Yet, nurses are often unable to care for patients without substantial time pressures due to their high workload. For example, the mean break time (breaks and meals) reported by nurses was only 15.5 min (Min et al., 2021). As a result, the authors argue that high workloads result in time pressures that limited time for self‐care that contribute to severe occupational fatigue.

Time pressure refers to the lack of time, perceived by an individual, whilst performing work responsibilities or other tasks, this is a subjective experience of almost always rushing the given work, thereby causing job stress (Szollos, 2009). Particularly for nurses, time pressure results in increased stress (Darawad et al., 2015), which ultimately reduces their responsiveness to patient care (Park & Lee, 2019) and adversely limits their ability to anticipate patient needs (Irwin et al., 2013). Nurses who were more aware of the time pressure had lower intervention performance and perceived decision‐making time for patient safety as a waste (Yun & Son, 2019). As time pressure increased whilst on duty, nurses reported feeling a lack of time, which led to higher rate of care left undone, contributing to making errors (Cho et al., 2016; Wu et al., 2013). Consequently, time pressure may affect patient safety, negatively (Scott et al., 2006).

There is a known relationship between psychological burnout and the time available to nurses for patient care (Teng et al., 2010; Yun & Son, 2019). Teng et al. (2010) reported that time pressure was negatively associated with patient safety only in regards to nurses with more severe burnout. Another study reported that time pressure was significantly associated with patient safety due to the mediation effect of burnout (Yun & Son, 2019).

Evidence to date suggests a link between psychological burnout, the perception of time pressure, and medication errors in nurses (Cao & Naruse, 2019; Khamisa et al., 2016; Nantsupawat et al., 2016; Westley et al., 2020). Therefore, the purpose of this study was to understand the relationship between psychological burnout, time factors and medication errors in nurses working in hospitals.

2.1. Study design and sample

A cross‐sectional study design was used to survey nurses working 8‐h shifts in two tertiary university hospitals in metropolitan cities in South Korea. The inclusion criteria were (1) bedside nurses with assignment to patient care, (2) the adult care nurses, (3) nurses working in a general ward or intensive care unit (ICU), working in 8 h, 3 shift system. The exclusion criteria were (1) nurses with <3 months of nursing career (to rule out the adaptation periods of new nurses), (2) nurses working in management roles. A convenience sampling method was employed, and bedside nurses were recruited from two tertiary university hospitals with similar staffing context. The ratio of beds to nurses in the two hospital was same, ranging from 2.0 to 2.5. A minimum sample size of 207 was calculated with G*power program (Faul et al., 2007) assuming multiple regression analysis, α = 0.05, a medium effect size of 0.15 (Cohen, 2013), 0.90 power, 10 factors, and 20% drop out rate.

2.2. Data collection

Data were collected during December 2021 in ten hospital units, including six general wards and four intensive care units. Each unit was provided with questionnaires for the nurses to complete. The researcher explained the research purpose and ethical considerations in each unit during the day and evening shift changes. The questionnaire required 10–15 min to complete. Completed questionnaires were returned in sealed envelopes to maintain confidentiality.

2.3. Measures

The questionnaire used for this study was 41 items, including 14 demographic items, Maslach Burnout Inventory (22 items), time pressure perception measurement (5 items), objective time pressure (5 items), and medication error measurement (9 items). The questionnaire was pilot tested with 7 clinical nurses to ensure the readability and accuracy of the meaning of each item of measurements prior to distribution.

2.3.1. General and hospital‐related characteristics

As general and hospital related characteristics, age, sex, marital status, education level, hospital bed number, working department, clinical career in one's current department, number of assigned patient during day duty were collected.

2.3.2. Psychological burnout

Maslach Burnout Inventory (MBI) (Maslach & Jackson, 1981) was used to measure the psychological burnout of hospital nurses. MBI is one of the most frequently used measure for nurses' psychological burnout, worldwide (Chen & Meier, 2021). MBI was validated through convergent validity and factor analysis at the development (Maslach & Jackson, 1981), and was a reliable and validated measure for South Korean nurses in a previous study (Kang & Kim, 2012). Kang and Kim (2012) assessed the convergent validity through the factor analysis and evaluated the discriminant validity, and suggested that the MBI could be used as a valid instrument to measure the psychological burnout level of Korean nurses. MBI consists of 22 items divided into three sub‐domains: emotional exhaustion (nine items), depersonalization (five items), and personal accomplishment (eight items). Each item is measured using 5‐point Likert scale (one: strongly disagree to five: strongly agree). As the personal accomplishment was reverse coded, higher scores indicate severe psychological burnout. The internal consistency reliabilities (Cronbach's α) of each sub‐domain were 0.87, 0.88 and 0.89, respectively, in the previous study (Kakemam et al., 2021). In current study, Cronbach's α was 0.89, 0.79 and 0.74 respectively.

2.3.3. Time factors

In this study, time factors were operationally defined as (1) subjectively perceived time pressure and (2) objectively measured work time characteristics.

1. Perceived time pressure

Time pressure was measured using time pressure perception measurement (Putrevu & Ratchford, 1997) revised by Teng et al. (2010). Teng et al. (2010) modified Putrevu and Ratchford's version to measure the subjective time pressure perception of hospital nurses. Teng et al. (2010) tested the reliability and validity of time pressure measurement through the confirmatory factor analysis, and the results were acceptable for hospital nurses (item factor loading coefficient of each item ranged from 0.77 ~ 0.99, and Cronbach’ α = 0.96). This version of time pressure was a reliable and validated measurement for Korean nurses in a prior study (Yun & Son, 2019). The authors went through translation‐reverse translation process (Yun & Son, 2019). Time pressure comprises five items with no sub domains. Each item was measured on a 5‐point Likert scale (one: strongly disagree to five: strongly agree). A higher score means higher perception of time pressure. The internal consistency reliabilities (Cronbach's α) of time pressure perception in a previous study was over 0.90 (Teng et al., 2010; Yun & Son, 2019). In the current study, Cronbach's α was 0.87.

• 2Objective work time characteristics

To measure the objective work time characteristics, information on rest time whilst on duty, meal time during duty and overtime during the last week was collected. The participants were asked to write the precise time (minutes) during the last shift (rest times, meal times) or during the previous week (over time).

2.3.4. Medication errors

We used the medication error measurement for Korean nurses developed by Park and Lee (2019), which comprised nine items. The validity of the current measurement was confirmed through the content validity index (>0.08) of the expert group (3 nursing professors and 3 clinical nurses with 10 years of clinical career and over), and construct validity with confirmatory factor analysis (factor loading 0.54 ~ 0.78, explained variance 0.458) was confirmed at the development (Park & Lee, 2019). For each item, subjects were asked to indicate the frequencies of medication error they experienced during the last 3 months. The nurses' responses were scored using the following scale: 1 point (one time), 2 point (twice), 3 point (three times), 4 point (four times) and 5 points (more than five times). If no medication error was experienced over 3 months, a score of 0 was assigned. The higher sum value signifies higher frequencies of medication errors. The internal consistency reliabilities (Cronbach's α) in a previous study was = 0.77 (Park & Lee, 2019). In the current study, Cronbach's α was 0.78.

2.4. Data analysis

Descriptive and inferential statistics were used to analyse the data. The general characteristics, psychological burnout and time factors were analysed using descriptive statistics: frequencies, percentages, mean (standard deviations), median, minimum, maximum, interquartile ranges (IQR), skewness and kurtosis.

The medication error variable was positively skewed due to many zeros, indicating no medication error occurred. As the distribution of the medication error was nonparametric, all comparisons were analysed using nonparametric methods. To identify the control variables, differences in the sum of medication errors, according to the general characteristics, were analysed using the Man–Whitney U test and the Kruskal–Wallis test. Spearman's rho(ρ) coefficients were used to analyse the correlation between medication errors, major research variables (psychological burnout, time factors) and control variables. The associations between the major research variables and medication errors were analysed with the zero‐inflated negative binomial regression for over‐dispersed and over‐abundant zero count data; Mediation error mean; 2.46, variance: 15.64, number of zeros: 93 (48.5%). (Green, 2021; Schober & Vetter, 2021). The associations of the predictors were given in terms of incidence rate ratio (IRR) with 95% confidence intervals (CI) (Schober & Vetter, 2021). Statistical significance for all comparison was defined as p < 0.05. Data were analysed using the STATA Statistical package (Version 17) (StataCorp.).

2.5. Ethical consideration

This study was performed in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of “REDACTED”. Each participant provided an informed consent prior to completing the survey. Participants who had completed the questionnaire were compensated with US $5 for their time.

3. RESULTS

3.1. General characteristics of participants and differences of medication errors

200 questionnaires were collected and analysed of 207 sent (Response rate = 96.6%). Average age of the participants was 29.4 ± 6.1 years. The average clinical career duration was 45.7 ± 3.3 months. About 60.0% (n = 120) of participants were working in general ward. The average number of assigned patients during the duty was 10.0 ± 2.6 among general ward nurses, and 2.8 ± 0.5 among the ICU nurses. (Table 1).

TABLE 1.

Nurses' general characteristics and differences in medication errors (N = 200).

Characteristic	Category	n (%)	Medication errors
			Mean ± SD	U * or H **	p
Age (years)**; mean = 29.4 ± 6.1	≤24a	32 (16.0)	3.25 ± 4.41	21.13	<0.001
	25–27b	67 (33.5)	3.07 ± 4.24
	28–30c	45 (22.5)	3.15 ± 4.67
	31–34d	21 (10.5)	1.09 ± 1.72
	≥35e	35 (17.5)	0.49 ± 1.24
Sex*	Female	188 (94.0)	2.51 ± 4.02	1069.00	0.746
	Male	12(6.0)	1.67 ± 2.67
Marital status*	Unmarried	149 (74.5)	2.85 ± 4.22	3030.00	0.020
	Married	51 (25.5)	1.33 ± 2.77
Education level**	3‐year diploma	9 (4.5)	4.33 ± 4.66	4.23	0.117
	Bachelor's degree	158 (79.0)	2.47 ± 3.97
	≥Master's degree	33 (16.5)	1.87 ± 3.62
clinical career in current department (month)**; mean = 45.7 ± 3.3	≤1	34 (17.0)	3.47 ± 4.77	9.81	0.044
	2–3	69 (34.5)	2.63 ± 3.89
	4–6	59 (29.5)	2.55 ± 4.10
	7–9	24 (12.0)	1.50 ± 3.12
	≥10	14 (7.0)	0.35 ± 0.63
Working area*	General ward	120 (60.0)	3.23 ± 4.47	3311.00	<0.001
	Intensive care unit	80 (40.0)	1.31 ± 2.66
Position*	Staff nurse	135 (67.5)	2.99 ± 4.30	3219.00	0.001
	Charge nurse	65 (32.5)	1.36 ± 2.84
hospital size*	≤1000 beds	100 (50.0)	2.35 ± 3.81	4889.00	0.773
	≥1001 beds	100 (50.0)	2.57 ± 4.10
No. of assigned patient during duty (general ward)**; mean = 10.0 ± 2.6	4–6	12 (10.0)	0.92 ± 1.72	5.02	0.081
	7–9	37 (30.8)	2.95 ± 4.21
	≥10	71 (59.2)	3.76 ± 4.80
No. of assigned patient during duty (ICU)**; mean = 2.8 ± 0.5	1	4(5.3)	1.75 ± 2.87	1.11	0.572
	2	6(8.0)	0.50 ± 1.22
	3	65 (86.7)	1.31 ± 2.62

Abbreviations: H, Kruskal–Wallis's H; SD, standard deviation; U, Mann–Whitney's U.

^*Means the Mann‐Whitney U test was done to analyze the difference in medication errors according to a variable.

^**Means the Kruskal‐Wallis test was done to analyze the difference in medication errors according to the variable.

The medication errors differed by age, marital status, clinical career, working area and position. (Table 1).

3.2. Correlations between psychological burnout, time factors and medication errors

The participants' psychological burnout averaged 2.93 ± 0.50A out of 5. For the time factor, the time pressure averaged 4.08 ± 0.64 out of 5. The average rest time during duty was 9.72 ± 10.63 min, ranging from 0 to 30 min. The average mealtime was 16.36 ± 9.50 min, ranging from 0 to 30 min. The average overtime (during the last week) was 142.45 ± 142.04 min. The median value of medication errors was 1.00, ranging from 0 to 15. (IQR 3) (Table 2).

TABLE 2.

The descriptive statistics for psychological burnout, time factors and medication errors (N = 200).

Variable	Range	Mean ± SD	Median	Min	Max	IQR	Skewness	Kurtosis
Psychological burnout (overall)	1–5	2.93 ± 0.50	3.00	1.59	4.18	0.68	−0.26	−0.17
Emotional exhaustion	1–5	3.46 ± 0.68	3.56	1.56	5.00	0.89	−0.35	−0.07
Depersonalization	1–5	2.42 ± 0.73	2.40	1.00	5.00	0.80	0.16	0.10
Personal accomplishment	1–5	2.66 ± 0.47	2.63	1.38	4.25	0.75	0.26	0.36
Time pressure	1–5	4.08 ± 0.64	4.00	2.60	5.00	1.00	−0.05	−0.95
Rest time during duty (min)	0–30	9.72 ± 10.63	10.00	0.00	30.00	20.00	0.68	−0.86
	Category (min)	n (%)
	0	88 (44.0)
	≤10	36 (18.0)
	11–20	32 (16.0)
	21–30	44 (22.0)
Mealtime during duty (min)	0–30	16.36 ± 9.50	15.00	0.00	30.00	10.00	−0.06	−0.80
	Category (min)	n (%)
	0	25 (12.5)
	≤10	40 (20.0)
	11–20	86 (43.0)
	21–30	49 (24.5)
Overtime (during the last week) (min)	0–840	142.45 ± 142.04	120.00	0.00	840.00	120.00	1.80	3.96
	Category (h)	n (%)
	0	23 (11.5)
	≤ 2	110(55.0)
	2.1–5	45 (22.5)
	≥5.1	22 (11.0)
Medication errors	0–45	2.46 ± 3.95	1.00	0.00	15.00	3.00	1.83	2.34

Abbreviations: IQR, Interquartile range; Max, Maximum value; Min, Minimum value; SD, standard deviation.

Medication errors of participants were significantly correlated with age (negative) (r = −0.29, p < 0.001), clinical career in current department (negative) (r = −0.18, p = 0.013), time pressure (positive) (r = 0.42, p < 0.001), rest time during duty (negative) (r = −0.29, p < 0.001) and mealtime during duty (negative) (r = −0.39, p < 0.001), and overtime during the last week (positive) (r = 0.53, p < 0.001). Medication errors was significantly correlated with psychological burnout, emotional exhaustion (positive) (r = 0.49, p < 0.001), depersonalization (positive) (r = 0.31, p < 0.001), personal accomplishment (positive) (r = 0.33, p < 0.001). (Table 3).

TABLE 3.

Correlations between general characteristics, time factors, psychological burnout and medication errors (N = 200).

Variable	Medication error
	r	ρ
General characteristic
Age	−0.29	<0.001
Clinical career in current department	−0.18	0.013
Psychological burnout (overall)	0.41	<0.001
Emotional exhaustion	0.49	<0.001
Depersonalization	0.31	<0.001
Personal accomplishment	0.33	<0.001
Time factors
Time pressure	0.42	<0.001
Rest time during duty	−0.29	<0.001
Mealtime during duty	−0.39	<0.001
Overtime (during the last week)	0.53	<0.001

3.3. The associated factors on medication errors

The covariates were the variables that showed significant differences in the bivariate analysis. As the control variable, age, working department were chosen. The position and clinical career were excluded due to high correlation with age (age‐clinical career: Pearson's r = 0.631, p < 0.001, age‐position: Spearman's ρ = 0.670, p < 0.001).

The zero‐inflated negative binomial regression analysis was performed to examine which factors predicted the frequency of medication errors, particularly as it presents overdispersions and many zero values. The logit model was applied. Table 4 shows the zero‐inflated negative binomial regression model results. A likelihood‐ratio test was done to identify the model fit and it appeared appropriate (Chi‐square value: 26.07, p < 0.001). The final zero‐inflated negative binomial regression model's AIC value was 500.52 compared with the zero‐inflated Poisson regression model’ AIC: 524.58 revealing the appropriateness of the model.

TABLE 4.

The relationship between medication errors and psychological burnout, time factor (N = 200).

Variable	IRR	SE	z	p	95% CI
Constant	5.14	4.79	1.76	0.079	0.83, 31.94
General characteristics
Age	0.98	0.02	−0.95	0.340	0.95, 1.37
Working department (ref. general ward)	0.98	0.17	−0.12	0.907	0.70, 1.30
Burnout	1.49	0.24	2.45	0.014	1.08, 2.05
Time factor
Time pressure	0.82	0.14	−1.19	0.233	0.60, 1.13
Rest time during duty	0.98	0.01	−1.42	0.155	0.96, 1.01
Mealtime during duty	0.97	0.01	−3.55	<0.001	0.95, 0.98
Overtime (weeks)	1.00	0.00	3.64	<0.001	1.00, 1.00

Abbreviations: CI, Confidence interval; IRR, Incidence rate ratio; SE, Standard error.

Higher psychological burnout (IRR = 1.49, p = 0.014), shorter mealtime during duty (IRR = 0.97, p < 0.001), longer weekly overtime (IRR = 1.00, p < 0.001) were associated with increased likelihood of medication error. (Table 4).

4. DISCUSSION

This study explored the association between psychological burnout, time factors and the medication errors among nurses working in Korean tertiary university hospitals. Our major findings are as follows; higher psychological burnout, shorter meal times whilst on duty and longer weekly overtime were associated with an increased likelihood of medication error of nurses among nurses operating under the controlling nurses' characteristics.

The level of psychological burnout was relatively high level, which is consistent with the results of a previous cross‐national investigation study (Poghosyan et al., 2010). Higher level of psychological burnout are also correlated with poor quality of patient care (Kakemam et al., 2021; Poghosyan et al., 2010) and increased medication error (Kakemam et al., 2021; Nantsupawat et al., 2016). Nurses with higher level of psychological burnout tend to have less skills and less desire to provide high‐quality nursing care (Maslach, 1999). As result, these nurses tend to be less engaged in the job (Kakemam et al., 2021). Nurses experiencing psychological burnout are likely to have negative perceptions about their jobs, and these negative thoughts could interfere with the perception of reality, and could induce an erratic or unpredictable state, and subsequently cause a flawed performance at the given tasks (Bagnasco et al., 2020). As nurses become psychologically burnt‐out, they may treat medication as a simple, repetitive and mechanical process without deliberate thoughts of working with patients, critical thought, or any professional regard (Gandi et al., 2011). In turn, this can increase the risk of medication errors (Gandi et al., 2011).

Managerial support and favourable working environment may positively influence the negative consequence of patient safety, caused by nurses' psychological burnout (Halbesleben et al., 2013; Montgomery et al., 2021). According to a previous study, low managerial support for nurses had worsened their level of psychological burnout level (28.6% shared variance); whereas greater managerial support decreased the likelihood of medication errors (Khatatbeh et al., 2021). Moreover, the joy in work strategy is a feasible alternative to improving the level of psychological burnout among nurses (Perlo et al., 2017); this could allow the nurses to feel physically and psychologically safe, and to develop resilience by linking the meaning and purpose to their job (Carter & Hawkins, 2019; Perlo et al., 2017), which ultimately could help improve their level of psychological burnout (Fitzpatrick et al., 2019). Therefore, nursing managers should seek managerial support strategies which utilize the joy in work strategy, for nurses so that they could discover purpose and meaning in their daily patient care and medication processes, and thus gain a sense of accomplishment from their jobs.

As another significant result, time factors were related to medication errors. Particularly, the shorter meal time and longer weekly overtimes were associated with increased likelihood of medication errors. This finding was consistent with the results of previous studies (Bae, 2013; Liu et al., 2012; MacPhee et al., 2017). MacPhee et al. (2017) reported that shorter meal break and longer overtime increased the frequency of medication errors. In U.S. nurses working more than 40 h per week have reported increased medication error (odds ratio; 3.71, p < 0.05) (Bae, 2013). Similar results were reported by Liu et al. (2012), who studied the associations of nurse time factor on nurse‐sensitive outcomes, including near misses in medication.

Meanwhile, break time whilst on duty was not a factor associated with the medication errors among nurses in the current study, unlike a previous study, which reported fewer breaks including meal time whilst on duty, led to more medication errors (Min et al., 2019). This may indicate that meal time is a more of a sensitive indicator of medication error than rest time whilst on duty.

Time pressure was not an associated factor of medication errors in this study, either. In this study, the mean value of time pressure of nurses was very high. In detailed data analysis, we found that the percentage of nurses who rated higher score than 4 (out of 5) was 57.5%, leaving a left‐long tailed distribution, which means most of the participant were feeling severe time pressure. Considering the participants mean meal times, rest times and overtime, the participants' high level of time pressure was understandable. The small variance in time pressure felt by the participants may decrease the influential power of time pressure against the mediation errors in the regression model. Notably, there may be a measurement error in time pressure, which could result in time pressure being insignificant.

Lack of time during work is a critical predictor of care left undone including medication work (Ausserhofer et al., 2014; Cho et al., 2016). When nurses experience high workloads, it limits their capacity to provide safe care to the patients (Min et al., 2019). In multiple previous studies on the relations of working hours and patient safety, the importance of maintaining an appropriate level of staffing is emphasized (Cho et al., 2016). Accordingly, health policies in Korea regulate hospitals to hire a certain number of nurses (Health insurance review and Assessment Service, 2021). Most tertiary general hospitals, including the affiliated hospitals of the participants in this study, maintained a ratio of 2.0–2.5 beds to nurses. However, it is a concern that basic necessary time such as meal time is not adequately guaranteed for the nurses. In Korea, 30‐min and 1‐h breaks must be provided per 4‐ and 8‐h shifts, respectively (Labor Standards Act in Korea, 2021). However, as seen in previous studies (Cho et al., 2016; Min et al., 2019) and our study, it is challenging to guarantee such minimum break time.

Meanwhile, actual time consuming for specific tasks or break, meal, overloaded work is important in the terms of nursing context (Min et al., 2019). In a five‐year long study on medication errors in a large regional hospital of Australia, medication errors were associated by temporal factors, and were frequently observed at certain times of the year and certain times of the day (Isaacs et al., 2021). As work in hospitals is often concentrated at specific times due to patient's needs, and workflow is not evenly distributed during working hours; therefore, job demands and time resources that fluctuate within working hours must be considered, along with the patient–nurse ratio (Isaacs et al., 2021; Valentin et al., 2006). Therefore, the recent staffing‐related problems are not only hard structural issues of the number of nurse staffing but practical and manageable issues of effectively assigning the nurses and their job. There is a need to focus not only on the nursing staffing but also on the composition of time used by the nurses during duty. Thus, it is important to assess the amount of time spent on different tasks to prepare more specific and realistic countermeasures to ensure minimum rest and meal time and reduce overtime hours.

Several limitations of the current study must be considered. First, this study had limited external validity because nurses working at two tertiary university hospitals with similar characteristics were conveniently sampled to control for possible other exogenous variables, such as the number of nurses to patients, severity of patients and regulations of hospitals, which may affect medication errors. Second, unlike objective time factors, subjective time pressure did not show significant association with the dependent variables. Further research should be considered, using subjective time pressure measures with higher specificity, sensitivity, and measurement error.

5. CONCLUSIONS

In conclusion, time factors, and psychological burnout, were associated with the medication errors for nurses working at tertiary university hospitals.

The nurse manager should help the nurses could recognize the meaning of their work in the routine job, and offer proper managerial support to them to deal with their psychological burnout in the clinical field. On the other hand, the nurse manager should monitor the actual time frame of the nurse. Securing minimum meal time is a basic need for every nurse, and directly relates to the nurses' work performance such as medication error. Therefore, the nurse manager should not consider only the number of nursing staff but also their actual time frame during the duty. The creative and efficient way of the nursing delivery system, job assignment and job analysis strategies should be considered in the way of the clinical context.

AUTHOR CONTRIBUTIONS

Study design: Kyoungja Kim, Cheongin Im, Suyoung Song. Data Collection: Kyoungja Kim, Suyoung Song. Data Analysis: Kyoungja Kim. Study Supervision: Kyoungja Kim. Manuscript writing: Kyoungja Kim, Cheongin Im, Suyoung Song. Critical revisions for important intellectual content: Kyoungja Kim.

FUNDING INFORMATION

This work was supported by INHA UNIVERSITY Research Grant (2020).

CONFLICT OF INTEREST STATEMENT

The author declares no actual or potential conflicts of interests.

ETHICAL APPROVAL

The current study was approved by the Institutional Review Board of “Inha University Hospital (Clearance number 2021‐09‐017). This study was performed in accordance with the Declaration of Helsinki.

Im, C. , Song, S. , & Kim, K. (2023). The associations of psychological burnout and time factors on medication errors in rotating shift nurses in Korea: A cross sectional descriptive study. Nursing Open, 10, 5550–5559. 10.1002/nop2.1794

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.