- Email: [email protected]
The Relationship Between Nursing Experience and Education and the Occurrence of Reported Pediatric Medication Administration Errors
Journal of Pediatric Nursing (2016) xx, xxx–xxx
The Relationship Between Nursing Experience and Education and the Occurrence of Reported Pediatric Medication Administration Errors Kim Sears RN, PhD a,⁎, Linda O'Brien-Pallas PhD, RN b , Bonnie Stevens PhD, RN b,c , Gail Tomblin Murphy PhD, RN d a
School of Nursing, Queens University, Kingston, ON, Canada Lawrence S. Bloomberg Faculty of Nursing University of Toronto, Toronto, ON, Canada c Paediatric Nursing Research, The Hospital for Sick Children d WHO/PAHO Collaborating Centre & Professor School of Nursing, Dalhousie University, Halifax, NS, Canada b
Received 26 May 2015; revised 7 December 2015; accepted 11 January 2016
Key words: Pediatric; Medication error; Nursing; Reporting
Medication errors are one of the most common incidents in the hospitals. They can be harmful, and they are even more detrimental for pediatric patients. This study explored the relationship between nursing experience, education, the frequency and severity of reported pediatric medication administration errors (PMAEs). The data for this study were collected from a larger pan Canadian study. A survey tool was developed to collect self-reported data from nurses. In addition to descriptive statistics, a Poisson regression or a multiple linear regression was completed to address the research questions, and a Boneferrai correction was conducted to adjust for the small sample size. Results demonstrated that on units with more nurses with a higher level of current experience, more PMAEs were reported (p = .001), however; the PMAEs reported by these nurses were not as severe (p = .003). Implications to advance both safe medication delivery in the pediatric setting and safe culture of reporting for both actual and potential errors are identified. © 2016 Elsevier Inc. All rights reserved.
MEDICATION ERRORS ARE one of the most common types of hospital incidents, and they often have devastating consequences (Wilkins & Shields, 2008). Data from the 2005 National Survey of the Work Health of Nurses were weighted to represent all registered nurses in Canada delivering direct patient care. Nearly one-fifth of all registered nurses reported that their patients had experienced a medication error occasionally or frequently under their care (Wilkins & Shields, 2008). There are many factors that influence this number such
⁎ Corresponding author: Kim Sears RN, PhD. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.pedn.2016.01.003 0882-5963/© 2016 Elsevier Inc. All rights reserved.
as heavy workloads and overtime. It is important to recognize that nurses are the last person in the process of medication administration to resolve and intercept the errors; therefore, they must know the effect, rationale and compatibilities of the drug, and need to calculate the right dose for patients (Ofosu & Jarrett, 2015) along with having a work environment that supports this process. Nurses can administer up to 50 medications per shift (Grigg, Garrett, & Craig, 2011) and they spend up to 40 percent of their time administering medications (Hughes & Blegen, 2008). Medication errors have the potential to be harmful with pediatric patients over-represented as victims (Honey & Condren, 2010). Medication errors are prevalent in children due to individual calculations of medications developed for
2 adult consumption, and the propensity for reactions to small errors (Broussard, 2010; Chen, 2013; Kaufmann, Laschat, & Wappler, 2012). Additionally, medications on pediatric wards are often packaged and prepared for adults, and staff often lack nursing education and clinical experience in this setting (Broussard, 2010). In order to examine these factors, it is essential to first identify them.
Purpose of Study The data in this study were from the second phase of a larger study (Sears, O'Brien-Pallas, Stevens, & Murphy, 2013). The purpose of this phase of the study was to examine how nursing experience and education related to pediatric medication administration error occurrence. Specifically, the study explored the frequency of reported pediatric medication administration errors (PMAEs), the severity of errors, and the quality of nursing care at the time of a PMAE. A medication error was defined as “any preventable error that has occurred as a result of human mistakes or system flaw that occurred in the process of administering a medication resulting in harm or the potential for harm” (Institute of Medicine, 2000).
Research Questions The research questions underpinning this phase of the study were as follows: 1) What is the influence of unit nurse characteristics (experience and education) on the frequency of reported PMAEs?; 2) What is the influence of unit nurse characteristics (experience and education) on the severity of error?; and 3) What is the influence of unit nurse characteristics (experience and education) on the quality of nursing care at the time of a PMAE?
Background Level of Nursing Experience A descriptive, cross-sectional study conducted by Unver, Tastan, and Akbayrak (2012) revealed that novice nurses' lack of experience negatively affects the frequency of medication errors. Compared to more experienced nurses, novices were less able to recognize the causes of medication errors. Moreover, many nursing students who have had a reduced load in their clinical training often begin their work with heavy workloads (Saintsing, Gibson, & Pennington, 2011). This inexperience is seen to result in a decreased incidence of error reporting (Duffield et al., 2011; Saintsing et al., 2011). Around 75 percent of novice nurses commit medication errors and thus often require additional support in acute care environments (Saintsing et al., 2011). It is recommended that there should always be a consistent scope of expertise in their workplace environment (Saintsing et al., 2011). In an experimental study conducted by Simonsen, Daehlin, Johansson, and Farup (2014) comparing registered nurses with at least 1-year experience and bachelor students in their last term, registered nurses demonstrated better medication knowledge than the bachelor students. This indicates that increased experience is significantly correlated with decreased risk of medication error.
K. Sears et al.
Level of Nursing Education Research has shown that with higher proportions of baccalaureate nurses on a unit, comes lower levels of mortality and failure-to-rescue rates (Aiken, Clarke, Sloane, Lake, & Cheney, 2008). A baccalaureate education is associated with better patient and nurse outcomes (Blegen, Goode, Park, Vaughn, & Spetz, 2013). In 2013, Blegen and colleagues conducted a cross-sectional study using data from 21 University Health System Consortium hospitals to analyze the relationship between registered nurse (RN) education and patient outcomes. If nurses had a higher education, such as a baccalaureate degree or higher, the number of adverse events such as the occurrence of deep vein thrombosis or pulmonary embolism, congestive heart failure, decubitus ulcers, failure to rescue and the length of stay decreased significantly (Blegen et al., 2013). The Institute of Medicine (IOM) recommends increasing the number of RNs with baccalaureate degrees to 80 percent by 2020, following IOM's campaign in their report on the Future of Nursing (Institute of Medicine, 2011).
Frequency of Error Ghaleb, Barber, Franklin, and Wong (2010) conducted a prospective study of 444 pediatric patients, and 2955 medication orders to detect the incidence of medication errors on pediatric units. The researchers identified that 13.2 percent of the medication errors were prescribing errors and 19.1 percent were administration errors. Researchers have determined that medication errors are approximately three times more likely on pediatric units than on adult units (Woo, Kim, Chung, & Park, 2015). In a 5-week retrospective cohort study by Al-Jeraisy, Alanazi, and Abolfotouh (2011), analysis of 2,380 medication errors were conducted and 1,333 errors were identified, which equated to a 56 percent error rate. In another study that evaluated telephone enquiries to the National Poisons Information Service concerning in-hospital pediatric medication errors, medication error rates on pediatric units varied from between one per 5.8 admissions to one in 662 pediatric admissions (Tharian, Thompson, & Tuthill, 2010). Further, as previously reported from the first phase of this study (Sears et al., 2013), 245 (65.9 %) actual errors were reported and 127 (34.1%) potential errors were reported. Of the 372 errors reported, most are committed at the wrong time (45.2%), wrong dose (22%), other factors (18.5%), wrong medication (8.3%), wrong route (2.7%), wrong patient (1.9%), wrong time and dose (0.8%) and wrong patient and medication (0.5%).
Severity of Error The literature reports that 95 percent of medication errors are not reported. The five percent that is reported is related to life-threatening effects (Institute of Medicine, 2007). It is perceived, therefore, that medication errors, which are considered minor or less severe, would go unreported. Al-Jeraisy et al. (2011) studied 2,380 medication orders, in which 1,333 medication errors were recognized. 44.5 percent of errors were in infants younger than 1 year of age. Children under 5 years of age were at a higher risk of error. Furthermore,
Paediatric medication administration errors risk for infants under 1 year of age was more than double that of the population aged 65 to 70, which is the following at-risk group. It is evident that the pediatric population appears to be more vulnerable to adverse outcomes from medication errors than adults. The majority of the errors (78.8%) were classified as potentially harmful. When medication errors occur in the pediatric population, there is a higher rate of death associated with error compared to the adult population (Lacey, Smith, & Cox, 2008). In their cross-sectional study of 19,350 errors retrieved from the Med Marx database over a 5 year period, Hicks, Becker, and Cousins (2006) noted that 4.2 percent of (816 of 19,350) errors were identified as “harmful” (involving consequences for the patient ranging from temporary harm to death). The Med Marx database is the largest adverse drug event database in the United States, with over 1.3 million medication error records. Marcin et al. (2007) conducted an observational study that investigated medication errors within acutely ill and injured children and found a 39 percent medication error rate; 15.9 percent which were categorized as potentially harmful and 85.5 percent as innocuous. Woods, Thomas, Holl, Altman, and Brennan (2005) examined 3719 patient charts retrospectively from the states of Utah and Colorado, looking at adverse events for children. One percent of children in these two states (1200–2100 children) experienced a prolonged hospitalization or a disability related to an adverse event in hospital (Woods et al., 2005). This number was projected nationally in the United States and indicated that 70,000 hospitalized children experienced an adverse event such as a medication error each year and that 60 percent of these events are, in fact, preventable (Woods et al., 2005). As noted within the first phase of this study, there were both potential (127) and actual (245) PMAE occurrences reported. They were reported as a whole of 327 errors and were then separated into the following groups: 49.7% minimal, 30.1% significant, 0.05% serious, 0.14% potentially lethal, and 0.01% lethal (Sears et al., 2013).
Quality of Care Within the Pediatric Nursing Work Environment Nurses have a significant role in assessing errors because they are generally in a position to notice medication errors first hand, and can thereby take steps to reduce the risk of incorrect drug administration (Toruner & Uysal, 2012). Children especially, are vulnerable to adverse outcomes during hospitalization (The Joint Commission, 2008). Toruner and Uysal (2012) state that higher patient/nurse ratios and longer working hours negatively impact the clinical environment resulting in higher levels of distraction, reduced quality of care and medication errors. Inadequate nursing staff levels adversely affect patient outcomes, quality of care, patient safety, and the health of the nurses (LeMoal, 2014). In pediatrics, the number, quality and availability of nursing staff have been shown to relate to the number of pediatric medication errors (Karavasiliadou & Athanasakis, 2014). Increased workload on pediatric units may be responsible for
3 medication errors resulting in a busy environment, which can cause fatigue and distraction leading to an error in the medication administration process (Karavasiliadou & Athanasakis, 2014). In addition, a greater number of nursing hours per patient were shown to significantly reduce postoperative pediatric complications and improve the quality of care for hospitalized children (Mark, Harless, & Berman, 2007). Children suffer from patient safety problems with severe impacts (Woods et al., 2005). Moreover, patient safety events occur frequently in very young children. The most reported patient safety event is medication error, in which nurses are directly related (Hughes, 2008). Patient safety and quality of care are health system priorities; however to address the systemic issues within organizations, a systems approach is required (Hughes, 2008).
Methods Study Design This pan-Canadian study was descriptive and used a prospective design collecting spontaneous self-report survey data related to the occurrence of an actual and potential PMAE. This methodology permitted nurses to respond at the time of the PMAE in a confidential manner and to report potential errors. The independent variable is a nurse's level of experience and education. The dependent variable is the frequency of error, the severity of error, and the quality of nursing care. Responses from the survey tool were used in the data analysis. This study was approved by ethics through the university and at each of the three hospital sites.
Setting The data for this study were gathered from a sub-study of a larger pan-Canadian study entitled: “Understanding the costs and outcomes of nurse's turnover in Canadian hospitals” (O’Brien-Pallas, Tomblin Murphy, & Shamian, 2004). The setting in which data were collected involved three tertiary university-affiliated pediatric health care center in east, west and central Canada (Sears et al., 2013).
Sample The sample was chosen conveniently consisting of 18 previously randomly selected units from the pan-Canadian study. The sample consisted of 18 in–hospital units consisting of critical care (n = 8) and medical/surgical units (n = 10). The total number of children's beds in the study was 440. Work environment data were collected including hours of nursing overtime per unit from the Nursing Turnover Study units within the three sites. Units were randomly selected and there were approximately 30 randomly assigned patients on each unit. For units where there were only data from 30 patients or less, randomization was not conducted. There were 372 surveys collected in total.
Instrument The PMAE survey tool was developed to collect prospective self-reported data from registered nurses who voluntarily
4
K. Sears et al.
completed the survey whenever an actual (APMAEs) or potential (PPMAEs) error occurred (Sears, 2009). The definition of APMAE and PPMAE were included in the survey tool. The tool received face validity, content validity and construct validity. Face validity was attained through the validation of ten experts who critiqued it for accuracy, wording, and appropriateness. Content validity was established for this survey tool. Respondents were asked to rate the representativeness, clarity and comprehensiveness of each question (Rubio, Berg-Weger, Tebb, Lee & Raunch, 2003). Reviewers rated the content of each question on a scale of one to four (questions receiving a rating lower than three were reworked or removed). To calculate the overall Content Validity Index (CVI) the total number of items ranked at 3 or 4 by the reviewers was divided by the total number of items. The PMAE survey had a CVI of 0.95. Construct validity was established for this tool through the use of factor analysis. Factor analysis is a widely used approach to obtain an estimation of construct validity (Goodwin & Goodwin, 1999). The tool obtained data related to the type of error (APMAEs or PPMAEs), the medication administration right that was violated, the environmental factors that contributed to the error, the level of severity of the error and the quality of care that was delivered at the time of error. A Likert scale was used, therefore, participants selected from a range of ‘one’ (not at all) to ‘five’ (significantly). The tool was short and easy and took approximately 5 minutes to complete. It was also readily available to participants.
Data Analysis Data from the survey tool were linked to data collected from the same units during the sequential 3-month period of data collection in the second wave of the Nursing Turnover Study (O’Brien-Pallas et al., 2004). Data aggregated to the unit level were used to explore the influence of nurse characteristics such as their level of experience and education per unit on the frequency of reported PMAEs, the severity of the error, and the perceived quality of nursing care at the time of the error. In addition to descriptive statistics, a Poisson regression or a multiple linear regression was completed to address the research questions, and a Boneferrai correction was conducted to adjust for the small sample size. Data analysis was conducted in SPSS version 15 and the level of significance was set at alpha .05. Table 1
Results Nursing Experience There were two types of nursing experience explored and aggregated to the unit level. The first was overall nursing experience (referring to experience since entering into practice), and the second was current nursing experience (referring to their experience on their current unit) (Table 1). On average, nurses indicated having worked as an RN for 10.74 years, and having worked on their current unit for 6.93 years. Further, there were two types of nursing education explored and aggregated to the unit level. The first group were nurses that had identified their highest level of nursing education as a diploma, and the second group were nurses that identified that their highest level of nursing education was a bachelor of science in nursing (BScN). The percentage of nurses that identified that their highest level of nursing education was a diploma ranged from 17.40 percent to 60.90 percent with a mean of 33.39 percent (SD = 11.98). The percentage of nurses with a BScN degree ranged from 21.70 percent to 78.90 percent across the units, with a mean of 60.53 percent (SD = 16.40).
Frequency of Errors The average frequency of reported errors per unit over the 3-month period of data collection was 29.18 (SD = 9.86). The minimum response was one error per unit, and the maximum number of errors reported was 43. The results of the Poisson regression demonstrated that there was no statistically significant relationship between unit type and education and frequency of reported PMAEs. The results demonstrated that, on units where the nurses' current experience (length of time working on the study unit) was longer, there were more errors reported (p = .001) (Table 2).
Severity of Errors In this phase of the study, the severity of error was regressed onto the unit level of experience and education. The model demonstrated an R2 of .56 identifying that 56 percent of the variance in the dependant variable could be explained by the relationship to the independent variable (Table 3). The results of the multiple regression demonstrated that there was a statistically significant association between the current experience (length of time working on the study unit) and the severity of error (p = .003). The results demonstrate that
Descriptive Statistics Nursing Inputs Experience and Education (n = 18). Minimum statistic
Overall nursing Experience (in months) Current nursing experience (In months) Highest nursing education diploma Highest nursing Education degree
Maximum statistic
Mean
Std. deviation
Statistic
Std. error
Statistic
74.16
191.73
128.86
10.02
42.50
14.00
176.66
83.11
8.90
37.76
17.40
60.90
33.39
2.82
11.98
21.70
78.90
60.53
3.86
16.40
Paediatric medication administration errors Table 2
5
Poisson Regression Nurse Education and Experience With Frequency of PMAEs (n = 18).
Parameter
B
(Intercept) Unit med/surg Unit critical care Nursing education BScN Experience overall Experience current (Scale)
1.585 .285 0(a) .003 − .004 .005 1(b)
Std. error
95% Wald confidence interval
Hypothesis test
Lower
Upper
Wald chi square
Df
Sig.
.2893 .1537
2.152 .586
30.029 3.432
1.018 − .016
1 1
.000 .064
.0039 .0017 .0016
.010 .000 .008
.451 4.919 11.412
− .005 − .007 .002
1 1 1
.502 .027 .001
Dependent variable: frequency of medication error at unit level. Model: (Intercept), Newunitype, HNEBScN, Overall, Current, offset = 1.099. (a) Set to zero because this parameter is redundant. (b) Fixed at the displayed value.
units with a higher level of current nursing experience reported less severe errors.
Quality of Care The results of the multiple regression examining the impact of nurse characteristics (experience and education) per unit on the perceived quality of nursing care demonstrated that there is no statistically significant association between experience and education and perceived quality of nursing care at the time of the PMAE.
Discussion There was a statistically significant relationship between current experience and the frequency of reported PMAEs. The results demonstrate that nurses with more experience on a unit report more errors (p = .001). These findings are paralleled with the findings in a survey study by Kim, An, Kim, and Yoon (2007) where nurses with 5 to 10 years of experience reported a higher frequency of medication errors. According to a survey study conducted by Vojir, Blegen, and Vaughn (2003), they found that a nurse's personal experience was an important determinant of their ability to detect a medication administration error, thus leading to higher reporting rates. The results in this study demonstrated that errors were less severe on units in which there are nurses with more experience on the unit (p = .003). This finding was not anticipated as it
Table 3
Strengths and Limitations The strengths of this study include the use of up to date unit level data to determine the staff compliment in terms of experience and education on each unit. Moreover, the
Multiple Regression Nurse Education and Experience With Severity of Error (n = 18) Coefficients (a).
Model
1
was hypothesized that the reported errors would be more severe in nature. This result is also found in the multisite organizational longitudinal study by Chang and Mark (2009). They found that errors were less severe with increased nursing experience and this was statistically significant. In this study there was no link between nursing education and the frequency of reported PMAEs nor nursing education and the severity of errors. It is believed that the high response rate to the minimal and significant errors is related to the study design and the data collection; the confidential manner of error reporting may have increased response to reporting of less significant errors. However, this does not negate the need for nurses entering into pediatrics to be educated about safe medication delivery through theoretical and practical experience. Nurses with a higher level of education have been correlated with an increased frequency of medication error reporting (Stratton, Blegen, Pepper, & Vaughn, 2004), however there are other studies that have determined no correlation between the two (Mayo & Duncan, 2004). Most studies have shown however that higher education is correlated with safer drug dose calculations (Simonsen et al., 2014).
(Constant) Nursing education BScN Experience overall Experience current
(a) Dependent variable: severity of error. R2 = .56 (56%).
Unstandardized Coefficients
Standardized coefficients
B
Std. error
Beta
1.029 .019 .011 − .018
.902 .010 .004 .005
.328 .509 − .716
T
Sig.
1.141 1.813 2.544 − 3.613
.273 .091 .023 .003
6 PMAE survey collected prospective data at the time of the error occurrence. There are limitations to this study. The sample size was only 18 units. There was a need for data aggregation based on the nature of the research questions. There was an overreliance on error and potential error reporting as the denominator of the actual number of errors per unit which is virtually impossible to know with certainty. There was a reliance on self-reporting of the PMAEs. A small number of hospitals (3) were reported in this study. Lastly, there was a potential limitation as there was a lack of independence of unit level data because of data clustering. Even though the three selected Canadian hospitals were identified as comparable, another limitation to the survey was that the culture of the units was not examined. Further study is required in order to generalize these findings.
Error Reporting and Implications for Clinical Practice In our study, there is a statistically significant relationship between current experience on a unit and the frequency of reported PMAEs. It is evident that there are factors that impede nurses from reporting medication errors. In a study conducted by Prang and Jelsness-Jorgensen (2014), the researchers determined the barriers to incident reporting in which nurses are often challenged. Nurses reported that the environment and the role of the administrator (ex. nurse leader) was very important in the incident reporting procedure. Other factors include a deficiency in the amount of support and culture on the floors, uncertain outcomes and routines. Some nurses revealed that they had limited experience with incident reporting whereas others had been encouraged to report as much as possible. Furthermore, the anonymity of the nurses was not always respected, which deterred the nurses from reporting incidents. There are opportunities to introduce incident reporting within the undergraduate nursing education. A simulation scenario in the simulation lab in which an adverse event occurs may be a useful tool to educate students. It is thought that the opportunity to complete this form in a safe environment followed by a debriefing session on the benefit of completing the form and feedback to the group on how the scenario could be changed could provide valuable learning for students and assist with the advancement of a safety culture on the units. Error reporting can provide a means to improve the medication delivery system but lessons for quality improvement will be lost on nurses. Future generations of nurses must be taught the value of reporting within the education system and then further encouraged within the practice setting. Rather than having organizations attempt to implement strategies to increase the safe delivery of medication whenever an error occurs, an open system of feedback will permit organizations to be proactive. To improve medication safety, error-reporting strategies should include identifying errors, being accountable for one's mistakes, correcting unsafe conditions, and reporting systems' improvements (Hughes, 2008).
K. Sears et al. Historically, error reporting has been viewed as a means for assigning blame and handing out punishment (Cohen, 1999). Hartnell, MacKinnon, Sketris, and Fleming (2014) identified several factors that would increase the likelihood of incident reporting namely: reporting should be easy for staff, feedback should be timely, and up-to-date education about reporting should be provided. Further, the reports need to be confidential. To create this culture, confidential reporting is encouraged; and summaries of medication incident reports need to be shared with staff and the changes to practice need to be identified based on the issues within the system. Staff nurses and members of the healthcare team will not be as likely to complete incident reports unless they see the link between taking time from their busy workload to complete a report, dissemination of the findings and changes from these reports. Hill (2010) notes that staff experience and education both contribute to patient outcomes. It is acknowledged that this has become an ever-increasing challenge given the exodus of senior nurses through retirements and the entrance of numerous new graduates into the system. Further, nurses within the work environment need to feel that reporting a medication error regardless of whether it is an actual or potential error is valued and encouraged.
Conclusion Underreporting of medication administering errors is a threat to the quality of nursing care and patient outcomes. Results of our study demonstrate that there were more errors reported on units where the nurses' current experience was longer. It is evident that a change in culture is necessary in order to encourage nurses with less experience and education to report medication errors, especially in a pediatric environment where there is an increased risk. It is also important to properly educate and prepare nurses with less overall experience and current experience on a certain unit to be able to decrease the incidence of medication errors and increase incidence of error reporting. This study has implications for the future education and practical experience provided to nurses in the pediatric setting to advance their level of safe medication delivery. Further research on the implications of nursing education and experience on factors affecting patient care is required in order to identify potential system issues and determine ways to improve error reporting and decrease risks for medication errors.
References Aiken, L. H., Clarke, S. P., Sloane, D. M., Lake, E. T., & Cheney, T. (2008). Effects of hospital care environment on patient mortality and nurse outcomes. Journal of Nursing Administration, 38, 223–229. http://dx. doi.org/10.1097/01.NNA.0000312773.42352.d7. Al-Jeraisy, M. I., Alanazi, M. Q., & Abolfotouh, M. A. (2011). Medication prescribing errors in a pediatric inpatient tertiary care setting in Saudi Arabia. BMC Research Notes, 4, 1–6. http://dx.doi.org/10.1186/1756-0500-4-294. Blegen, M. A., Goode, C. J., Park, S. H., Vaughn, T., & Spetz, J. (2013). Baccaleureate education in nursing and patient outcomes. Journal of Nursing Administration, 43, 89–94. http://dx.doi.org/10.1097/NNA. 0b013e31827f2028.
Paediatric medication administration errors Broussard, L. (2010). Small size, big risk: Preventing neonatal and pediatric medication errors. Nursing for Women's Health, 14, 405–408. http://dx. doi.org/10.1111/j.1751-486X.2010.01580.x. Chang, Y. K., & Mark, B. A. (2009). Antecedents of severe and nonsevere medication errors. Journal of Nursing Scholarship, 41, 70–78. http:// dx.doi.org/10.1111/j.1547-5069.2009.01253.x. Chen, C. J. (2013). Medication errors in pediatrics. Pediatrics & Neonatology, 54, 3–4. http://dx.doi.org/10.1016/j.pedneo.2012.12.019. Cohen, M. R. (1999). Medication errors. Washington: American Pharmaceutical Association, 6.10–6.11 (7.1-7.11). Duffield, C., Diers, D., O’Brien-Pallas, L., Aisbett, C., Roche, M., ... Aisbett, K. (2011). Nursing staffing, nursing workload, the work environment and patient outcomes. Applied Nursing Research, 24, 244–255. http://dx.doi.org/10.1016/j.apnr.2009.12.004. Ghaleb, M. A., Barber, N., Franklin, B. D., & Wong, I. C. K. (2010). The incidence and nature of prescribing and medication administration errors in paediatric inpatients. Archives of Disease in Childhood, 95, 113–118. http://dx.doi.org/10.1136/adc.2009.158485. Goodwin, L. D., & Goodwin, W. L. (1999). Measurement myths and misconceptions. School Psychology Quarterly, 14, 408–427. Grigg, S. J., Garrett, S. K., & Craig, J. B. (2011). A process centered analysis of medication administration: Identifying current methods and potential for improvement. International Journal of Industrial Ergonomics, 41, 380–388. Hartnell, N., MacKinnon, N., Sketris, I., & Fleming, M. (2014). Identifying, understanding and overcoming barriers to medication error reporting in hospitals: A focus group study. BMJ Quality and Safety, 21, 361–368. http://dx.doi.org/10.1136/bmjqs-2011-000299. Hicks, R. W., Becker, S. C., & Cousins, D. D. (2006). Harmful medication errors in children: A 5-year analysis of data from the USP’s MEDMARX® program. Journal of Pediatric Nursing, 21, 290–298. http://dx.doi.org/10.1016/j.pedn.2006.02.002. Hill, K. S. (2010). Improving quality and patient safety by retaining nursing expertise. The Online Journal of Issues in Nursing, 15 (Retrieved May 6th, 2015. Retrieved from: http://www.nursingworld.org/MainMenuCategories/ ANAMarketplace/ANAPeriodicals/OJIN/TableofContents/Vol152010/ No3-Sept-2010/Articles-Previously-Topic/Improving-Quality-and-PatientSafety-.html). Honey, B. L., & Condren, M. (2010). Promoting safe medication use in pediatric patients. Research in Social and Administrative Pharmacy, 6, 269–271. http://dx.doi.org/10.1016/j.sapharm.2010.07.002. Hughes, R. G. (2008). Patient safety and quality: An evidence-based handbook for nurses. Rockville (MD): Agency for Healthcare Research and Quality (US). Hughes, R. G., & Blegen, M. A. (2008). Patient safety and quality: An evidencebased handbook for nurses. Chapter 37: Medication Administration Safety. Rockville (MD): Agency for Healthcare Research and Quality (US). Institute of Medicine (2000). Committee on quality of health care in America. In L. T. Kohn, J. M. Corrigan, & J. M. Donaldson (Eds.), To err is human building a safer health system. Washington, DC: National Academy Press. Institute of Medicine (2007). Preventing medication errors. Washington, DC: National Academies Press. Institute of Medicine (2011). The future of nursing: Leading change, advancing health. Washington, DC: National Academies Press. Karavasiliadou & Athanasakis (2014). An inside look into the factors contributing to medication errors in the clinical nursing practice. Health Science Journal, 8, 32–44. Kaufmann, J., Laschat, M., & Wappler, F. (2012). Medication errors in pediatric emergencies: A systematic analysis. Deutsches Ärzteblatt International, 109, 609–616. http://dx.doi.org/10.3238/arztebl.2012.0609. Kim, J., An, K., Kim, M. K., & Yoon, S. H. (2007). Nurses’ perception of error reporting and patient safety culture in Korea. Western Journal of Nursing Research, 29, 827–844. http://dx.doi.org/10.1177/0193945906297370. Lacey, S., Smith, J. B., & Cox, K. (2008). Chapter 15: Pediatric safety and quality. Rockville (MD): Agency for Healthcare Research and Quality.
7 LeMoal, L. (2014). Nurse-patient ratios in the Canadian context. Retrieved May 8th, 2015. Retrieved from: http://www.thinknursing.ca/sites/ thinknursing.ca/files/Nurse_Patient_Ratios.pdf Marcin, J. P., Dharmar, M., Cho, M., Seifert, L. L., Cook, J. L., ... Romano, P. S. (2007). Medication errors among acutely ill and injured children treated in rural emergency departments. Annals of Emergency Medicine, 50, 361–367. http://dx.doi.org/10.1016/j.annemergmed. 2007.01.020. Mark, B. A., Harless, D. W., & Berman, W. F. (2007). Nurse staffing and adverse events in hospitalized children. Policy, Politics & Nursing Practice, 8, 83–92. http://dx.doi.org/10.1177/1527154407303499 (Mayo and Duncan, 2004). Mayo, A. M., & Duncan, D. (2004). Nurse perceptions of medication errors: What we need to know for patient safety. Journal of Nursing Care Quality, 19, 209–217. O’Brien-Pallas, L., Tomblin Murphy, G., & Shamian, J. (2004). Understanding the costs and outcomes of nurses’ turnover in Canadian hospitals Proposal . Ofosu, R., & Jarrett, P. (2015). Reducing nurse medicine administration errors. Nursing Times, 111, 12–14. Prang, I. W., & Jelsness-Jorgensen, L. P. (2014). Should I report? A qualitative study of barriers to incident reporting among nurses working in nursing homes. Geriatric Nursing, 35, 441–447. http://dx.doi.org/10. 1016/j.gerinurse.2014.07.003. Rubio, D. M., Berg-Weger, M., Tebb, S. S., Lee, E. S., & Rauch, S. (2003). Objectifying content validity: Conducting a content validity study in social work research. Oxford Journals, 27, 94–104. Saintsing, D., Gibson, L. M., & Pennington, A. W. (2011). The novice nurse and clinical decision-making: How to avoid errors. Journal of Nursing Management, 19, 354–359. http://dx.doi.org/10.1111/j.1365-2834.2011. 01248. Sears, K. (2009). The relationship between the nursing work environment and the occurrence of reported paediatric medication administration errors. Toronto, ON: University of Toronto. Sears, K., O'Brien-Pallas, L., Stevens, B., & Murphy, G. T. (2013). The relationship between the nursing work environment and the occurrence of reported paediatric medication administration errors: A pan Canadian study. Journal of Pediatric Nursing, 28, 351–356. http://dx.doi.org/10. 1016/j.pedn.2012.12.003. Simonsen, B. O., Daehlin, G. K., Johansson, I., & Farup, P. G. (2014). Differences in medication knowledge and risk of errors between graduating nursing students and working registered nurses: Comparative study. BMC Health Services Research, 14, 1–11. http://dx.doi.org/10. 1186/s12913-014-0580-7. Stratton, K. M., Blegen, M. A., Pepper, G., & Vaughn, T. (2004). Reporting of medication errors by pediatric nurses. Journal of Pediatric Nursing, 19, 385–392. http://dx.doi.org/10.1016/j.pedn.2004.11.007. Tharian, K., Thompson, J. P., & Tuthill, D. P. (2010). Inpatient paediatric medication errors – What can be learned from enquiries made to the National Poisons Information Service (Cardiff Unit)? Welsh Paediatric Journal, 34, 19–21. The Joint Commission (2008). Preventing pediatric medication errors. (39) Retrieved from: http://www.jointcommission.org/assets/1/18/sea_39.pdf. Toruner, E. K., & Uysal, G. (2012). Causes, reporting, and prevention of medication errors from a pediatric nurse perspective. The Australian Journal of Advanced Nursing, 29, 28–35 (Retrieved May 12th, 2015. Retrieved from: http://search.informit.com.au/documentSummary;dn= 626038810538014;res=IELHEA). Unver, V., Tastan, S., & Akbayrak, N. (2012). Medication errors: Perspectives of newly graduated and experienced nurses. International Journal of Nursing Practice, 18, 317–324. http://dx.doi.org/10.1111/j. 1440-172X.2012.02052.x. Vojir, C. P., Blegen, M. A., & Vaughn, T. (2003). Nursing staff as estimators of unit medication error reporting. Communicating Nursing Research, 36. Wilkins, K., & Shields, M. (2008). Correlates of medication error in hospitals. Statistics Canada, 19.
8 Woo, Y., Kim, H. E., Chung, S., & Park, B. J. (2015). Pediatric Medication Error Reports in Korea Adverse Event Reporting System Database, 1989-2012: Comparing with adult reports. Journal of Korean Medical Science, 30, 371–377.
K. Sears et al. Woods, D., Thomas, E., Holl, J., Altman, S., & Brennan, T. (2005). Adverse events and preventable adverse events in children. Pediatrics, 115, 155–160. http://dx.doi.org/10.1542/peds.20040410.
The Relationship Between Nursing Experience and Education and the Occurrence of Reported Pediatric Medication Administration Errors Kim Sears RN, PhD a,⁎, Linda O'Brien-Pallas PhD, RN b , Bonnie Stevens PhD, RN b,c , Gail Tomblin Murphy PhD, RN d a
School of Nursing, Queens University, Kingston, ON, Canada Lawrence S. Bloomberg Faculty of Nursing University of Toronto, Toronto, ON, Canada c Paediatric Nursing Research, The Hospital for Sick Children d WHO/PAHO Collaborating Centre & Professor School of Nursing, Dalhousie University, Halifax, NS, Canada b
Received 26 May 2015; revised 7 December 2015; accepted 11 January 2016
Key words: Pediatric; Medication error; Nursing; Reporting
Medication errors are one of the most common incidents in the hospitals. They can be harmful, and they are even more detrimental for pediatric patients. This study explored the relationship between nursing experience, education, the frequency and severity of reported pediatric medication administration errors (PMAEs). The data for this study were collected from a larger pan Canadian study. A survey tool was developed to collect self-reported data from nurses. In addition to descriptive statistics, a Poisson regression or a multiple linear regression was completed to address the research questions, and a Boneferrai correction was conducted to adjust for the small sample size. Results demonstrated that on units with more nurses with a higher level of current experience, more PMAEs were reported (p = .001), however; the PMAEs reported by these nurses were not as severe (p = .003). Implications to advance both safe medication delivery in the pediatric setting and safe culture of reporting for both actual and potential errors are identified. © 2016 Elsevier Inc. All rights reserved.
MEDICATION ERRORS ARE one of the most common types of hospital incidents, and they often have devastating consequences (Wilkins & Shields, 2008). Data from the 2005 National Survey of the Work Health of Nurses were weighted to represent all registered nurses in Canada delivering direct patient care. Nearly one-fifth of all registered nurses reported that their patients had experienced a medication error occasionally or frequently under their care (Wilkins & Shields, 2008). There are many factors that influence this number such
⁎ Corresponding author: Kim Sears RN, PhD. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.pedn.2016.01.003 0882-5963/© 2016 Elsevier Inc. All rights reserved.
as heavy workloads and overtime. It is important to recognize that nurses are the last person in the process of medication administration to resolve and intercept the errors; therefore, they must know the effect, rationale and compatibilities of the drug, and need to calculate the right dose for patients (Ofosu & Jarrett, 2015) along with having a work environment that supports this process. Nurses can administer up to 50 medications per shift (Grigg, Garrett, & Craig, 2011) and they spend up to 40 percent of their time administering medications (Hughes & Blegen, 2008). Medication errors have the potential to be harmful with pediatric patients over-represented as victims (Honey & Condren, 2010). Medication errors are prevalent in children due to individual calculations of medications developed for
2 adult consumption, and the propensity for reactions to small errors (Broussard, 2010; Chen, 2013; Kaufmann, Laschat, & Wappler, 2012). Additionally, medications on pediatric wards are often packaged and prepared for adults, and staff often lack nursing education and clinical experience in this setting (Broussard, 2010). In order to examine these factors, it is essential to first identify them.
Purpose of Study The data in this study were from the second phase of a larger study (Sears, O'Brien-Pallas, Stevens, & Murphy, 2013). The purpose of this phase of the study was to examine how nursing experience and education related to pediatric medication administration error occurrence. Specifically, the study explored the frequency of reported pediatric medication administration errors (PMAEs), the severity of errors, and the quality of nursing care at the time of a PMAE. A medication error was defined as “any preventable error that has occurred as a result of human mistakes or system flaw that occurred in the process of administering a medication resulting in harm or the potential for harm” (Institute of Medicine, 2000).
Research Questions The research questions underpinning this phase of the study were as follows: 1) What is the influence of unit nurse characteristics (experience and education) on the frequency of reported PMAEs?; 2) What is the influence of unit nurse characteristics (experience and education) on the severity of error?; and 3) What is the influence of unit nurse characteristics (experience and education) on the quality of nursing care at the time of a PMAE?
Background Level of Nursing Experience A descriptive, cross-sectional study conducted by Unver, Tastan, and Akbayrak (2012) revealed that novice nurses' lack of experience negatively affects the frequency of medication errors. Compared to more experienced nurses, novices were less able to recognize the causes of medication errors. Moreover, many nursing students who have had a reduced load in their clinical training often begin their work with heavy workloads (Saintsing, Gibson, & Pennington, 2011). This inexperience is seen to result in a decreased incidence of error reporting (Duffield et al., 2011; Saintsing et al., 2011). Around 75 percent of novice nurses commit medication errors and thus often require additional support in acute care environments (Saintsing et al., 2011). It is recommended that there should always be a consistent scope of expertise in their workplace environment (Saintsing et al., 2011). In an experimental study conducted by Simonsen, Daehlin, Johansson, and Farup (2014) comparing registered nurses with at least 1-year experience and bachelor students in their last term, registered nurses demonstrated better medication knowledge than the bachelor students. This indicates that increased experience is significantly correlated with decreased risk of medication error.
K. Sears et al.
Level of Nursing Education Research has shown that with higher proportions of baccalaureate nurses on a unit, comes lower levels of mortality and failure-to-rescue rates (Aiken, Clarke, Sloane, Lake, & Cheney, 2008). A baccalaureate education is associated with better patient and nurse outcomes (Blegen, Goode, Park, Vaughn, & Spetz, 2013). In 2013, Blegen and colleagues conducted a cross-sectional study using data from 21 University Health System Consortium hospitals to analyze the relationship between registered nurse (RN) education and patient outcomes. If nurses had a higher education, such as a baccalaureate degree or higher, the number of adverse events such as the occurrence of deep vein thrombosis or pulmonary embolism, congestive heart failure, decubitus ulcers, failure to rescue and the length of stay decreased significantly (Blegen et al., 2013). The Institute of Medicine (IOM) recommends increasing the number of RNs with baccalaureate degrees to 80 percent by 2020, following IOM's campaign in their report on the Future of Nursing (Institute of Medicine, 2011).
Frequency of Error Ghaleb, Barber, Franklin, and Wong (2010) conducted a prospective study of 444 pediatric patients, and 2955 medication orders to detect the incidence of medication errors on pediatric units. The researchers identified that 13.2 percent of the medication errors were prescribing errors and 19.1 percent were administration errors. Researchers have determined that medication errors are approximately three times more likely on pediatric units than on adult units (Woo, Kim, Chung, & Park, 2015). In a 5-week retrospective cohort study by Al-Jeraisy, Alanazi, and Abolfotouh (2011), analysis of 2,380 medication errors were conducted and 1,333 errors were identified, which equated to a 56 percent error rate. In another study that evaluated telephone enquiries to the National Poisons Information Service concerning in-hospital pediatric medication errors, medication error rates on pediatric units varied from between one per 5.8 admissions to one in 662 pediatric admissions (Tharian, Thompson, & Tuthill, 2010). Further, as previously reported from the first phase of this study (Sears et al., 2013), 245 (65.9 %) actual errors were reported and 127 (34.1%) potential errors were reported. Of the 372 errors reported, most are committed at the wrong time (45.2%), wrong dose (22%), other factors (18.5%), wrong medication (8.3%), wrong route (2.7%), wrong patient (1.9%), wrong time and dose (0.8%) and wrong patient and medication (0.5%).
Severity of Error The literature reports that 95 percent of medication errors are not reported. The five percent that is reported is related to life-threatening effects (Institute of Medicine, 2007). It is perceived, therefore, that medication errors, which are considered minor or less severe, would go unreported. Al-Jeraisy et al. (2011) studied 2,380 medication orders, in which 1,333 medication errors were recognized. 44.5 percent of errors were in infants younger than 1 year of age. Children under 5 years of age were at a higher risk of error. Furthermore,
Paediatric medication administration errors risk for infants under 1 year of age was more than double that of the population aged 65 to 70, which is the following at-risk group. It is evident that the pediatric population appears to be more vulnerable to adverse outcomes from medication errors than adults. The majority of the errors (78.8%) were classified as potentially harmful. When medication errors occur in the pediatric population, there is a higher rate of death associated with error compared to the adult population (Lacey, Smith, & Cox, 2008). In their cross-sectional study of 19,350 errors retrieved from the Med Marx database over a 5 year period, Hicks, Becker, and Cousins (2006) noted that 4.2 percent of (816 of 19,350) errors were identified as “harmful” (involving consequences for the patient ranging from temporary harm to death). The Med Marx database is the largest adverse drug event database in the United States, with over 1.3 million medication error records. Marcin et al. (2007) conducted an observational study that investigated medication errors within acutely ill and injured children and found a 39 percent medication error rate; 15.9 percent which were categorized as potentially harmful and 85.5 percent as innocuous. Woods, Thomas, Holl, Altman, and Brennan (2005) examined 3719 patient charts retrospectively from the states of Utah and Colorado, looking at adverse events for children. One percent of children in these two states (1200–2100 children) experienced a prolonged hospitalization or a disability related to an adverse event in hospital (Woods et al., 2005). This number was projected nationally in the United States and indicated that 70,000 hospitalized children experienced an adverse event such as a medication error each year and that 60 percent of these events are, in fact, preventable (Woods et al., 2005). As noted within the first phase of this study, there were both potential (127) and actual (245) PMAE occurrences reported. They were reported as a whole of 327 errors and were then separated into the following groups: 49.7% minimal, 30.1% significant, 0.05% serious, 0.14% potentially lethal, and 0.01% lethal (Sears et al., 2013).
Quality of Care Within the Pediatric Nursing Work Environment Nurses have a significant role in assessing errors because they are generally in a position to notice medication errors first hand, and can thereby take steps to reduce the risk of incorrect drug administration (Toruner & Uysal, 2012). Children especially, are vulnerable to adverse outcomes during hospitalization (The Joint Commission, 2008). Toruner and Uysal (2012) state that higher patient/nurse ratios and longer working hours negatively impact the clinical environment resulting in higher levels of distraction, reduced quality of care and medication errors. Inadequate nursing staff levels adversely affect patient outcomes, quality of care, patient safety, and the health of the nurses (LeMoal, 2014). In pediatrics, the number, quality and availability of nursing staff have been shown to relate to the number of pediatric medication errors (Karavasiliadou & Athanasakis, 2014). Increased workload on pediatric units may be responsible for
3 medication errors resulting in a busy environment, which can cause fatigue and distraction leading to an error in the medication administration process (Karavasiliadou & Athanasakis, 2014). In addition, a greater number of nursing hours per patient were shown to significantly reduce postoperative pediatric complications and improve the quality of care for hospitalized children (Mark, Harless, & Berman, 2007). Children suffer from patient safety problems with severe impacts (Woods et al., 2005). Moreover, patient safety events occur frequently in very young children. The most reported patient safety event is medication error, in which nurses are directly related (Hughes, 2008). Patient safety and quality of care are health system priorities; however to address the systemic issues within organizations, a systems approach is required (Hughes, 2008).
Methods Study Design This pan-Canadian study was descriptive and used a prospective design collecting spontaneous self-report survey data related to the occurrence of an actual and potential PMAE. This methodology permitted nurses to respond at the time of the PMAE in a confidential manner and to report potential errors. The independent variable is a nurse's level of experience and education. The dependent variable is the frequency of error, the severity of error, and the quality of nursing care. Responses from the survey tool were used in the data analysis. This study was approved by ethics through the university and at each of the three hospital sites.
Setting The data for this study were gathered from a sub-study of a larger pan-Canadian study entitled: “Understanding the costs and outcomes of nurse's turnover in Canadian hospitals” (O’Brien-Pallas, Tomblin Murphy, & Shamian, 2004). The setting in which data were collected involved three tertiary university-affiliated pediatric health care center in east, west and central Canada (Sears et al., 2013).
Sample The sample was chosen conveniently consisting of 18 previously randomly selected units from the pan-Canadian study. The sample consisted of 18 in–hospital units consisting of critical care (n = 8) and medical/surgical units (n = 10). The total number of children's beds in the study was 440. Work environment data were collected including hours of nursing overtime per unit from the Nursing Turnover Study units within the three sites. Units were randomly selected and there were approximately 30 randomly assigned patients on each unit. For units where there were only data from 30 patients or less, randomization was not conducted. There were 372 surveys collected in total.
Instrument The PMAE survey tool was developed to collect prospective self-reported data from registered nurses who voluntarily
4
K. Sears et al.
completed the survey whenever an actual (APMAEs) or potential (PPMAEs) error occurred (Sears, 2009). The definition of APMAE and PPMAE were included in the survey tool. The tool received face validity, content validity and construct validity. Face validity was attained through the validation of ten experts who critiqued it for accuracy, wording, and appropriateness. Content validity was established for this survey tool. Respondents were asked to rate the representativeness, clarity and comprehensiveness of each question (Rubio, Berg-Weger, Tebb, Lee & Raunch, 2003). Reviewers rated the content of each question on a scale of one to four (questions receiving a rating lower than three were reworked or removed). To calculate the overall Content Validity Index (CVI) the total number of items ranked at 3 or 4 by the reviewers was divided by the total number of items. The PMAE survey had a CVI of 0.95. Construct validity was established for this tool through the use of factor analysis. Factor analysis is a widely used approach to obtain an estimation of construct validity (Goodwin & Goodwin, 1999). The tool obtained data related to the type of error (APMAEs or PPMAEs), the medication administration right that was violated, the environmental factors that contributed to the error, the level of severity of the error and the quality of care that was delivered at the time of error. A Likert scale was used, therefore, participants selected from a range of ‘one’ (not at all) to ‘five’ (significantly). The tool was short and easy and took approximately 5 minutes to complete. It was also readily available to participants.
Data Analysis Data from the survey tool were linked to data collected from the same units during the sequential 3-month period of data collection in the second wave of the Nursing Turnover Study (O’Brien-Pallas et al., 2004). Data aggregated to the unit level were used to explore the influence of nurse characteristics such as their level of experience and education per unit on the frequency of reported PMAEs, the severity of the error, and the perceived quality of nursing care at the time of the error. In addition to descriptive statistics, a Poisson regression or a multiple linear regression was completed to address the research questions, and a Boneferrai correction was conducted to adjust for the small sample size. Data analysis was conducted in SPSS version 15 and the level of significance was set at alpha .05. Table 1
Results Nursing Experience There were two types of nursing experience explored and aggregated to the unit level. The first was overall nursing experience (referring to experience since entering into practice), and the second was current nursing experience (referring to their experience on their current unit) (Table 1). On average, nurses indicated having worked as an RN for 10.74 years, and having worked on their current unit for 6.93 years. Further, there were two types of nursing education explored and aggregated to the unit level. The first group were nurses that had identified their highest level of nursing education as a diploma, and the second group were nurses that identified that their highest level of nursing education was a bachelor of science in nursing (BScN). The percentage of nurses that identified that their highest level of nursing education was a diploma ranged from 17.40 percent to 60.90 percent with a mean of 33.39 percent (SD = 11.98). The percentage of nurses with a BScN degree ranged from 21.70 percent to 78.90 percent across the units, with a mean of 60.53 percent (SD = 16.40).
Frequency of Errors The average frequency of reported errors per unit over the 3-month period of data collection was 29.18 (SD = 9.86). The minimum response was one error per unit, and the maximum number of errors reported was 43. The results of the Poisson regression demonstrated that there was no statistically significant relationship between unit type and education and frequency of reported PMAEs. The results demonstrated that, on units where the nurses' current experience (length of time working on the study unit) was longer, there were more errors reported (p = .001) (Table 2).
Severity of Errors In this phase of the study, the severity of error was regressed onto the unit level of experience and education. The model demonstrated an R2 of .56 identifying that 56 percent of the variance in the dependant variable could be explained by the relationship to the independent variable (Table 3). The results of the multiple regression demonstrated that there was a statistically significant association between the current experience (length of time working on the study unit) and the severity of error (p = .003). The results demonstrate that
Descriptive Statistics Nursing Inputs Experience and Education (n = 18). Minimum statistic
Overall nursing Experience (in months) Current nursing experience (In months) Highest nursing education diploma Highest nursing Education degree
Maximum statistic
Mean
Std. deviation
Statistic
Std. error
Statistic
74.16
191.73
128.86
10.02
42.50
14.00
176.66
83.11
8.90
37.76
17.40
60.90
33.39
2.82
11.98
21.70
78.90
60.53
3.86
16.40
Paediatric medication administration errors Table 2
5
Poisson Regression Nurse Education and Experience With Frequency of PMAEs (n = 18).
Parameter
B
(Intercept) Unit med/surg Unit critical care Nursing education BScN Experience overall Experience current (Scale)
1.585 .285 0(a) .003 − .004 .005 1(b)
Std. error
95% Wald confidence interval
Hypothesis test
Lower
Upper
Wald chi square
Df
Sig.
.2893 .1537
2.152 .586
30.029 3.432
1.018 − .016
1 1
.000 .064
.0039 .0017 .0016
.010 .000 .008
.451 4.919 11.412
− .005 − .007 .002
1 1 1
.502 .027 .001
Dependent variable: frequency of medication error at unit level. Model: (Intercept), Newunitype, HNEBScN, Overall, Current, offset = 1.099. (a) Set to zero because this parameter is redundant. (b) Fixed at the displayed value.
units with a higher level of current nursing experience reported less severe errors.
Quality of Care The results of the multiple regression examining the impact of nurse characteristics (experience and education) per unit on the perceived quality of nursing care demonstrated that there is no statistically significant association between experience and education and perceived quality of nursing care at the time of the PMAE.
Discussion There was a statistically significant relationship between current experience and the frequency of reported PMAEs. The results demonstrate that nurses with more experience on a unit report more errors (p = .001). These findings are paralleled with the findings in a survey study by Kim, An, Kim, and Yoon (2007) where nurses with 5 to 10 years of experience reported a higher frequency of medication errors. According to a survey study conducted by Vojir, Blegen, and Vaughn (2003), they found that a nurse's personal experience was an important determinant of their ability to detect a medication administration error, thus leading to higher reporting rates. The results in this study demonstrated that errors were less severe on units in which there are nurses with more experience on the unit (p = .003). This finding was not anticipated as it
Table 3
Strengths and Limitations The strengths of this study include the use of up to date unit level data to determine the staff compliment in terms of experience and education on each unit. Moreover, the
Multiple Regression Nurse Education and Experience With Severity of Error (n = 18) Coefficients (a).
Model
1
was hypothesized that the reported errors would be more severe in nature. This result is also found in the multisite organizational longitudinal study by Chang and Mark (2009). They found that errors were less severe with increased nursing experience and this was statistically significant. In this study there was no link between nursing education and the frequency of reported PMAEs nor nursing education and the severity of errors. It is believed that the high response rate to the minimal and significant errors is related to the study design and the data collection; the confidential manner of error reporting may have increased response to reporting of less significant errors. However, this does not negate the need for nurses entering into pediatrics to be educated about safe medication delivery through theoretical and practical experience. Nurses with a higher level of education have been correlated with an increased frequency of medication error reporting (Stratton, Blegen, Pepper, & Vaughn, 2004), however there are other studies that have determined no correlation between the two (Mayo & Duncan, 2004). Most studies have shown however that higher education is correlated with safer drug dose calculations (Simonsen et al., 2014).
(Constant) Nursing education BScN Experience overall Experience current
(a) Dependent variable: severity of error. R2 = .56 (56%).
Unstandardized Coefficients
Standardized coefficients
B
Std. error
Beta
1.029 .019 .011 − .018
.902 .010 .004 .005
.328 .509 − .716
T
Sig.
1.141 1.813 2.544 − 3.613
.273 .091 .023 .003
6 PMAE survey collected prospective data at the time of the error occurrence. There are limitations to this study. The sample size was only 18 units. There was a need for data aggregation based on the nature of the research questions. There was an overreliance on error and potential error reporting as the denominator of the actual number of errors per unit which is virtually impossible to know with certainty. There was a reliance on self-reporting of the PMAEs. A small number of hospitals (3) were reported in this study. Lastly, there was a potential limitation as there was a lack of independence of unit level data because of data clustering. Even though the three selected Canadian hospitals were identified as comparable, another limitation to the survey was that the culture of the units was not examined. Further study is required in order to generalize these findings.
Error Reporting and Implications for Clinical Practice In our study, there is a statistically significant relationship between current experience on a unit and the frequency of reported PMAEs. It is evident that there are factors that impede nurses from reporting medication errors. In a study conducted by Prang and Jelsness-Jorgensen (2014), the researchers determined the barriers to incident reporting in which nurses are often challenged. Nurses reported that the environment and the role of the administrator (ex. nurse leader) was very important in the incident reporting procedure. Other factors include a deficiency in the amount of support and culture on the floors, uncertain outcomes and routines. Some nurses revealed that they had limited experience with incident reporting whereas others had been encouraged to report as much as possible. Furthermore, the anonymity of the nurses was not always respected, which deterred the nurses from reporting incidents. There are opportunities to introduce incident reporting within the undergraduate nursing education. A simulation scenario in the simulation lab in which an adverse event occurs may be a useful tool to educate students. It is thought that the opportunity to complete this form in a safe environment followed by a debriefing session on the benefit of completing the form and feedback to the group on how the scenario could be changed could provide valuable learning for students and assist with the advancement of a safety culture on the units. Error reporting can provide a means to improve the medication delivery system but lessons for quality improvement will be lost on nurses. Future generations of nurses must be taught the value of reporting within the education system and then further encouraged within the practice setting. Rather than having organizations attempt to implement strategies to increase the safe delivery of medication whenever an error occurs, an open system of feedback will permit organizations to be proactive. To improve medication safety, error-reporting strategies should include identifying errors, being accountable for one's mistakes, correcting unsafe conditions, and reporting systems' improvements (Hughes, 2008).
K. Sears et al. Historically, error reporting has been viewed as a means for assigning blame and handing out punishment (Cohen, 1999). Hartnell, MacKinnon, Sketris, and Fleming (2014) identified several factors that would increase the likelihood of incident reporting namely: reporting should be easy for staff, feedback should be timely, and up-to-date education about reporting should be provided. Further, the reports need to be confidential. To create this culture, confidential reporting is encouraged; and summaries of medication incident reports need to be shared with staff and the changes to practice need to be identified based on the issues within the system. Staff nurses and members of the healthcare team will not be as likely to complete incident reports unless they see the link between taking time from their busy workload to complete a report, dissemination of the findings and changes from these reports. Hill (2010) notes that staff experience and education both contribute to patient outcomes. It is acknowledged that this has become an ever-increasing challenge given the exodus of senior nurses through retirements and the entrance of numerous new graduates into the system. Further, nurses within the work environment need to feel that reporting a medication error regardless of whether it is an actual or potential error is valued and encouraged.
Conclusion Underreporting of medication administering errors is a threat to the quality of nursing care and patient outcomes. Results of our study demonstrate that there were more errors reported on units where the nurses' current experience was longer. It is evident that a change in culture is necessary in order to encourage nurses with less experience and education to report medication errors, especially in a pediatric environment where there is an increased risk. It is also important to properly educate and prepare nurses with less overall experience and current experience on a certain unit to be able to decrease the incidence of medication errors and increase incidence of error reporting. This study has implications for the future education and practical experience provided to nurses in the pediatric setting to advance their level of safe medication delivery. Further research on the implications of nursing education and experience on factors affecting patient care is required in order to identify potential system issues and determine ways to improve error reporting and decrease risks for medication errors.
References Aiken, L. H., Clarke, S. P., Sloane, D. M., Lake, E. T., & Cheney, T. (2008). Effects of hospital care environment on patient mortality and nurse outcomes. Journal of Nursing Administration, 38, 223–229. http://dx. doi.org/10.1097/01.NNA.0000312773.42352.d7. Al-Jeraisy, M. I., Alanazi, M. Q., & Abolfotouh, M. A. (2011). Medication prescribing errors in a pediatric inpatient tertiary care setting in Saudi Arabia. BMC Research Notes, 4, 1–6. http://dx.doi.org/10.1186/1756-0500-4-294. Blegen, M. A., Goode, C. J., Park, S. H., Vaughn, T., & Spetz, J. (2013). Baccaleureate education in nursing and patient outcomes. Journal of Nursing Administration, 43, 89–94. http://dx.doi.org/10.1097/NNA. 0b013e31827f2028.
Paediatric medication administration errors Broussard, L. (2010). Small size, big risk: Preventing neonatal and pediatric medication errors. Nursing for Women's Health, 14, 405–408. http://dx. doi.org/10.1111/j.1751-486X.2010.01580.x. Chang, Y. K., & Mark, B. A. (2009). Antecedents of severe and nonsevere medication errors. Journal of Nursing Scholarship, 41, 70–78. http:// dx.doi.org/10.1111/j.1547-5069.2009.01253.x. Chen, C. J. (2013). Medication errors in pediatrics. Pediatrics & Neonatology, 54, 3–4. http://dx.doi.org/10.1016/j.pedneo.2012.12.019. Cohen, M. R. (1999). Medication errors. Washington: American Pharmaceutical Association, 6.10–6.11 (7.1-7.11). Duffield, C., Diers, D., O’Brien-Pallas, L., Aisbett, C., Roche, M., ... Aisbett, K. (2011). Nursing staffing, nursing workload, the work environment and patient outcomes. Applied Nursing Research, 24, 244–255. http://dx.doi.org/10.1016/j.apnr.2009.12.004. Ghaleb, M. A., Barber, N., Franklin, B. D., & Wong, I. C. K. (2010). The incidence and nature of prescribing and medication administration errors in paediatric inpatients. Archives of Disease in Childhood, 95, 113–118. http://dx.doi.org/10.1136/adc.2009.158485. Goodwin, L. D., & Goodwin, W. L. (1999). Measurement myths and misconceptions. School Psychology Quarterly, 14, 408–427. Grigg, S. J., Garrett, S. K., & Craig, J. B. (2011). A process centered analysis of medication administration: Identifying current methods and potential for improvement. International Journal of Industrial Ergonomics, 41, 380–388. Hartnell, N., MacKinnon, N., Sketris, I., & Fleming, M. (2014). Identifying, understanding and overcoming barriers to medication error reporting in hospitals: A focus group study. BMJ Quality and Safety, 21, 361–368. http://dx.doi.org/10.1136/bmjqs-2011-000299. Hicks, R. W., Becker, S. C., & Cousins, D. D. (2006). Harmful medication errors in children: A 5-year analysis of data from the USP’s MEDMARX® program. Journal of Pediatric Nursing, 21, 290–298. http://dx.doi.org/10.1016/j.pedn.2006.02.002. Hill, K. S. (2010). Improving quality and patient safety by retaining nursing expertise. The Online Journal of Issues in Nursing, 15 (Retrieved May 6th, 2015. Retrieved from: http://www.nursingworld.org/MainMenuCategories/ ANAMarketplace/ANAPeriodicals/OJIN/TableofContents/Vol152010/ No3-Sept-2010/Articles-Previously-Topic/Improving-Quality-and-PatientSafety-.html). Honey, B. L., & Condren, M. (2010). Promoting safe medication use in pediatric patients. Research in Social and Administrative Pharmacy, 6, 269–271. http://dx.doi.org/10.1016/j.sapharm.2010.07.002. Hughes, R. G. (2008). Patient safety and quality: An evidence-based handbook for nurses. Rockville (MD): Agency for Healthcare Research and Quality (US). Hughes, R. G., & Blegen, M. A. (2008). Patient safety and quality: An evidencebased handbook for nurses. Chapter 37: Medication Administration Safety. Rockville (MD): Agency for Healthcare Research and Quality (US). Institute of Medicine (2000). Committee on quality of health care in America. In L. T. Kohn, J. M. Corrigan, & J. M. Donaldson (Eds.), To err is human building a safer health system. Washington, DC: National Academy Press. Institute of Medicine (2007). Preventing medication errors. Washington, DC: National Academies Press. Institute of Medicine (2011). The future of nursing: Leading change, advancing health. Washington, DC: National Academies Press. Karavasiliadou & Athanasakis (2014). An inside look into the factors contributing to medication errors in the clinical nursing practice. Health Science Journal, 8, 32–44. Kaufmann, J., Laschat, M., & Wappler, F. (2012). Medication errors in pediatric emergencies: A systematic analysis. Deutsches Ärzteblatt International, 109, 609–616. http://dx.doi.org/10.3238/arztebl.2012.0609. Kim, J., An, K., Kim, M. K., & Yoon, S. H. (2007). Nurses’ perception of error reporting and patient safety culture in Korea. Western Journal of Nursing Research, 29, 827–844. http://dx.doi.org/10.1177/0193945906297370. Lacey, S., Smith, J. B., & Cox, K. (2008). Chapter 15: Pediatric safety and quality. Rockville (MD): Agency for Healthcare Research and Quality.
7 LeMoal, L. (2014). Nurse-patient ratios in the Canadian context. Retrieved May 8th, 2015. Retrieved from: http://www.thinknursing.ca/sites/ thinknursing.ca/files/Nurse_Patient_Ratios.pdf Marcin, J. P., Dharmar, M., Cho, M., Seifert, L. L., Cook, J. L., ... Romano, P. S. (2007). Medication errors among acutely ill and injured children treated in rural emergency departments. Annals of Emergency Medicine, 50, 361–367. http://dx.doi.org/10.1016/j.annemergmed. 2007.01.020. Mark, B. A., Harless, D. W., & Berman, W. F. (2007). Nurse staffing and adverse events in hospitalized children. Policy, Politics & Nursing Practice, 8, 83–92. http://dx.doi.org/10.1177/1527154407303499 (Mayo and Duncan, 2004). Mayo, A. M., & Duncan, D. (2004). Nurse perceptions of medication errors: What we need to know for patient safety. Journal of Nursing Care Quality, 19, 209–217. O’Brien-Pallas, L., Tomblin Murphy, G., & Shamian, J. (2004). Understanding the costs and outcomes of nurses’ turnover in Canadian hospitals Proposal . Ofosu, R., & Jarrett, P. (2015). Reducing nurse medicine administration errors. Nursing Times, 111, 12–14. Prang, I. W., & Jelsness-Jorgensen, L. P. (2014). Should I report? A qualitative study of barriers to incident reporting among nurses working in nursing homes. Geriatric Nursing, 35, 441–447. http://dx.doi.org/10. 1016/j.gerinurse.2014.07.003. Rubio, D. M., Berg-Weger, M., Tebb, S. S., Lee, E. S., & Rauch, S. (2003). Objectifying content validity: Conducting a content validity study in social work research. Oxford Journals, 27, 94–104. Saintsing, D., Gibson, L. M., & Pennington, A. W. (2011). The novice nurse and clinical decision-making: How to avoid errors. Journal of Nursing Management, 19, 354–359. http://dx.doi.org/10.1111/j.1365-2834.2011. 01248. Sears, K. (2009). The relationship between the nursing work environment and the occurrence of reported paediatric medication administration errors. Toronto, ON: University of Toronto. Sears, K., O'Brien-Pallas, L., Stevens, B., & Murphy, G. T. (2013). The relationship between the nursing work environment and the occurrence of reported paediatric medication administration errors: A pan Canadian study. Journal of Pediatric Nursing, 28, 351–356. http://dx.doi.org/10. 1016/j.pedn.2012.12.003. Simonsen, B. O., Daehlin, G. K., Johansson, I., & Farup, P. G. (2014). Differences in medication knowledge and risk of errors between graduating nursing students and working registered nurses: Comparative study. BMC Health Services Research, 14, 1–11. http://dx.doi.org/10. 1186/s12913-014-0580-7. Stratton, K. M., Blegen, M. A., Pepper, G., & Vaughn, T. (2004). Reporting of medication errors by pediatric nurses. Journal of Pediatric Nursing, 19, 385–392. http://dx.doi.org/10.1016/j.pedn.2004.11.007. Tharian, K., Thompson, J. P., & Tuthill, D. P. (2010). Inpatient paediatric medication errors – What can be learned from enquiries made to the National Poisons Information Service (Cardiff Unit)? Welsh Paediatric Journal, 34, 19–21. The Joint Commission (2008). Preventing pediatric medication errors. (39) Retrieved from: http://www.jointcommission.org/assets/1/18/sea_39.pdf. Toruner, E. K., & Uysal, G. (2012). Causes, reporting, and prevention of medication errors from a pediatric nurse perspective. The Australian Journal of Advanced Nursing, 29, 28–35 (Retrieved May 12th, 2015. Retrieved from: http://search.informit.com.au/documentSummary;dn= 626038810538014;res=IELHEA). Unver, V., Tastan, S., & Akbayrak, N. (2012). Medication errors: Perspectives of newly graduated and experienced nurses. International Journal of Nursing Practice, 18, 317–324. http://dx.doi.org/10.1111/j. 1440-172X.2012.02052.x. Vojir, C. P., Blegen, M. A., & Vaughn, T. (2003). Nursing staff as estimators of unit medication error reporting. Communicating Nursing Research, 36. Wilkins, K., & Shields, M. (2008). Correlates of medication error in hospitals. Statistics Canada, 19.
8 Woo, Y., Kim, H. E., Chung, S., & Park, B. J. (2015). Pediatric Medication Error Reports in Korea Adverse Event Reporting System Database, 1989-2012: Comparing with adult reports. Journal of Korean Medical Science, 30, 371–377.
K. Sears et al. Woods, D., Thomas, E., Holl, J., Altman, S., & Brennan, T. (2005). Adverse events and preventable adverse events in children. Pediatrics, 115, 155–160. http://dx.doi.org/10.1542/peds.20040410.