Art & science | patient safety
Medication errors: what hospital reports reveal about staff views Marja Härkänen and colleagues examine a study to help nurses reduce the risk of making mistakes when administering medicines to patients Correspondence
[email protected] Marja Härkänen is an early stage researcher at the Department of Nursing Science, University of Eastern Finland, Kuopio Hannele Turunen is a professor at the Department of Nursing Science, University of Eastern Finland, Kuopio, and a nurse director at Kuopio University Hospital Susanna Saano is a pharmacist at Kuopio University Hospital Katri Vehviläinen-Julkunen is a professor at the Department of Nursing Science, University of Eastern Finland, Kuopio, and a nurse director at Kuopio University Hospital Date of submission October 10 2012 Date of acceptance December 27 2012 Peer review This article has been subject to double-blind review and has been checked using antiplagiarism software Author guidelines www.nursingmanagement.co.uk
Abstract Nurses are generally responsible for administering medication to patients and are, therefore, able to monitor or report medication errors. However, nurses can sometimes be responsible for causing errors, so it is important that they understand the consequences of these mistakes and how to prevent them. This article reports the results of a study that analysed the views of nurses, pharmacists and physicians at a Finnish university hospital on the prevention of medication errors. The findings suggest errors can be prevented by improving work environments and allowing healthcare staff, particularly nurses, to concentrate on the task at hand when dealing with medicines. Keywords Medication errors, patient safety, reporting ADMINISTERING MEDICATION is one of the most common interventions in health care, but the complexity of the medication process can lead to unintended outcomes (Rozich et al 2003, McBride-Henry and Foureur 2006, Sheu et al 2008). In recent years the use of medicines has grown, and there has been an increase in new products available (Ndosi and Newell 2008). Rapid advances in technology and medicine have improved patient care, but the risk of errors has increased in this complex environment (Ndosi and Newell 2008, Sheu et al 2008). Promoting safety in the medication process is vital because adverse events have significant consequences, such as increasing hospital admissions, prolonging hospital stays, using additional resources, increasing the amount of time staff spend away from work and lowering patient satisfaction. And medication errors can
32 March 2013 | Volume 19 | Number 10
cause patients harm, the result of which can range from slight discomfort to death (Barker et al 2002, Ford et al 2006). Nurses care for patients in hospitals around the clock and are usually responsible for administering medication, so they have an important role in detecting, reporting and preventing errors (Miller et al 2006). However, nurses also cause errors (Barker et al 2002), so it is vital that they understand and can prevent them (Wakefield et al 2005). Information about medication errors and near misses can be used to identify risks and generate safer practice (Thomas and Panchagnula 2008). Error-reporting systems help gather data about these events (Savage et al 2005), the details of which are used in many international studies (Hicks et al 2004, Rudman et al 2005, Miller et al 2006, Pierson et al 2007, Thomas and Panchagnula 2008). However, because these studies have not analysed participants’ views on how to prevent errors, this study was designed to gather this information.
Study The data for this retrospective register study were collected from the Kuopio University Hospital web-based error-reporting database HaiPro. Most reporters on the database are nurses, although all professionals involved in patient care can use it to report detected errors and near misses. In 2010 there were 1,617 incident reports on the database, of which 671 were connected to medication and were therefore analysed for this study. Data analysis The error-reporting form on HaiPro contains a free-text section in which reporters can give their views about how this type of incident could have been prevented. This qualitative information was quantified and classified into NURSING MANAGEMENT
different categories according to content, then processed statistically using the SPSS for Windows 19.0 software program. Results were reported using frequencies and percentages, but their description incorporated direct quotes from staff about error prevention; some examples are given in Box 1.
Results Medication errors Most of the reports (82.6 per cent) came from registered nurses, while pharmacists reported 5.4 per cent and physicians only 2.5 per cent. Most incidents related to administration (39.9 per cent) or documentation (25.5 per cent) errors, and the most common types of incidents involved wrong doses (26.0 per cent) or omissions (24.0 per cent). Table 1, page 34, shows medication error types in different phases of the medication process. Approximately two thirds (69.2 per cent) of the incidents reached the patients, while one third were near misses. The majority of the incidents (65.7 per cent) caused no harm to patients, and only 0.3 per cent were estimated to have caused severe harm. Particular medicines were commonly linked to errors, most often psycho-pharmaceuticals or intravenous or intramuscular antibiotics. Box 1 Examples of staff views on preventing errors; direct quotes taken from the error reports ■ Attention and caution ‘More accuracy, diligence and focus on dispensing medicines. It is important to check the medicines before administering them.’ ■ Common management policies and guidelines ‘There is a guide to update the patient medication records, when a patient enters and exits the hospital. In this error situation, the instructions were not followed. The incident could have been prevented if the instructions had been followed.’ ■ Flow of information ‘The patient records information should be consulted. Entry interview has to be very careful especially with the patient’s allergies. The person who brings the patient must be familiar with the patient details; you cannot say: “I don’t know, I just came to work”.’ ■ Clear lines of responsibility ‘A patient demanding continuous medical treatment should be treated by an intravenous-licensed nurse. And one should have to focus only on those issues, not ten others at the same time.’ (Quotes translated from Finnish)
NURSING MANAGEMENT
Figure 1 Cited ways to prevent medication errors (n=671) No proposal Increasing attention and caution Increasing common guidelines Clarification of electronic patient records Avoiding the interference of medicines Checking medicines Flow of information More staff Clarity of documentation Opportunity to focus on the work Clear lines of responsibility Orientation of the staff Identifying the right patient Other 0
50
100
150
Staff views on preventing medication errors Staff gave different views on how medication errors have been prevented (Figure 1), but increased attention and caution (21.6 per cent) in different phases of the medication process was the most common view. Jointly agreed management policies and guidelines were mentioned in 12.4 per cent of cases, because they were not followed or there were not any. Staff thought that such policies and guidelines were particularly important because of the obscurity of roles, tasks or responsibilities in the workplace. Employees thought that the guidelines could help clarify roles, such as whether it is the responsibility of nurses or doctors to record details of new medication in electronic records. Proposals for the development of existing activities were also mentioned. Some of the comments described situations where existing guidelines were not followed. A new electronic medication record was introduced at Kuopio University Hospital at the end of 2009, but by 2010 users were still experiencing difficulties and lacked familiarity with the software program; for example, 7.7 per cent of reporters needed clarification about, or support with, using it. Avoiding confusion about the medicines was described by 7 per cent of reporters as important. Their comments refer to being unsure about medicines because of similar drug names or similar packaging, as well as confusion over dispensing or administering medicines to patients. March 2013 | Volume 19 | Number 10 33
Art & science | patient safety Table 1
Error types by percentage in different phases of the medication process (n=670) Administering
Documenting
Dispensing
Wrong patient
5.6
(15)
1.2
(2)
1.0
(1)
Wrong medicine
9.4
(25)
8.3
(14)
18.4
Wrong dose
28.5
(76)
15.4
(26)
Wrong route
1.1
(3)
0
Wrong timing
14.2
(38)
Omission
9.1
Discontinued medicine Wrong documenting
Prescribing 0
–
(18)
9.4
(8)
40.8
(40)
31.8
(27)
–
0
–
3.5
(3)
0
–
2.0
(2)
0
–
(51)
40.8
(69)
26.9
(29)
10.6
(9)
8.2
(22)
3.0
(5)
7.1
(7)
1.2
(1)
0.4
(1)
17.8
(30)
1.0
(1)
20.0
(17)
Wrong technique
12.0
(32)
5.3
(9)
0
–
10.6
(9)
Wrong duration of medication
0.4
(1)
6.5
(11)
0
–
10.6
(9)
Wrong form of medicine
0.4
(1)
0.6
(1)
0
–
2.4
(2)
Other
0.7
(2)
1.2
(2)
0
–
0
(169)
100
TOTAL
100
(267)
100
Checking medicines is fairly routine, but 6.3 per cent of reporters said that they believed that medication checks should be increased, particularly during busy periods when mistakes are easily overlooked, and even where there is a dose distribution system, or when pharmacists are working on the wards. Reporters said that double checking and checking medication with patients was important for error prevention, while 4.6 per cent suggested that increasing the verbal and written flow of information between colleagues, different professional groups, patients and different wards was important. Having enough nurses on a shift in relation to the number of patients was described as a factor in error prevention in 3.9 per cent of reports, and reporters suggested that lack of staff makes people rush and pay less attention to their tasks, which increases the risk of errors. Another 3.9 per cent 34 March 2013 | Volume 19 | Number 10
(98)
100
– (85)
of reports noted that clarity of written information about prescribing, ordering and documenting should prevent errors. Freedom to concentrate on the task at hand during the medication process was described as important by 3.4 per cent of reporters, while the busyness of working environments, large numbers of disruptions and rushing were all described as risk factors. Clear lines of responsibility during the medication process was perceived as important by 2.4 per cent of respondents, who also noted that clear division of responsibilities could prevent duplication and reduce the risk that no one completes a task. Clear lines of responsibility were also associated with clarity about who should take charge of patient medication. Insufficient induction of new employees and lack of skills were described as risks in 2.4 per cent of reports. Better training to carry out tasks is needed NURSING MANAGEMENT
Preparing
Delivering
Storing
Ordering
Other
Total
0
–
0
–
0
–
0
–
0
–
2.7
(18)
16.7
(3)
25.0
(3)
0
–
0
–
0
–
10.6
(71)
16.7
(3)
0
–
0
–
12.5
(1)
25.5
(1)
26.0
(174)
0
–
0
–
0
–
12.5
(1)
0
–
1.0
(7)
0
–
8.3
(1)
0
–
0
–
0
–
6.1
(41)
5.6
(1)
8.3
(1)
0
–
0
–
25.0
(1)
24.0
(161)
0
–
0
–
0
–
0
–
0
–
5.2
(35)
11.1
(2)
41.7
(5)
0
–
37.5
(3)
0
–
8.8
(59)
11.1
(2)
8.3
(1)
100
(9)
0
–
50.0
(2)
9.6
(64)
–
0
–
0
–
0
–
0
–
3.1
(21)
0
–
0
–
0
–
0
–
0.6
(4)
(1)
0
–
37.5
(3)
0
–
2.2
(15)
(12)
100
(8)
100
0
0
–
38.9
(7)
100
(18)
8.3 100
(9)
by all professionals involved in medication, but some reporters said that often there was too little time to induct new staff. Finally, verification of the right patients during the medication process was described by 1.3 per cent of reporters as important for preventing errors, alongside ensuring that patients’ identities are highlighted before administration and during documentation.
Discussion Limitations and strengths There are limitations to this study. First, 22.4 per cent of reporters made no suggestions about how errors could have been prevented. Second, research suggests that only a minority of errors are ever noticed (Evans et al 2006, van den Bemt et al 2007) and only some of these are reported (Wakefield et al 2005, Evans et al 2006, McBrideHenry and Foureur 2006, Griffin and Resar 2009). NURSING MANAGEMENT
100
(8)
100
(670)
Therefore, it is assumed that the data represent only a few of the errors that occurred. Generalisation of the results is challenging because the data were reported and collected in one hospital and in one country, so issues related to culture could affect the results. However, the study’s strength, is that two experts, the principal investigator and the hospital pharmacist, worked together to analyse the material, all categorisation of the errors was agreed on, and a consensus reached. In addition, the study material was a one-year sample of the medication errors. Discussion of results The aim of the study was to analyse staff views, recorded on incident reports, on preventing medication errors. Most reporters were nurses and their views were considered valuable because of their ability to oversee the whole medication process. Physicians reported only one in 40 errors, which supports previous studies March 2013 | Volume 19 | Number 10 35
Art & science | patient safety This study suggests that medication errors can be reduced or prevented by enabling staff to focus their attention on the task that show their low participation in error reporting (Evans et al 2006, Miller et al 2006). Reported incidents caused few consequences for patients and only 0.3 per cent were thought to cause severe harm, which again supports previous error-reporting studies. For example, Miller et al (2006) analysed 1,010 reported errors, of which none were serious and most of which related to administration errors; similar results were found by Hicks et al (2004). In addition, wrong dose and omission, the most common types of error, were also found in other studies (Hicks et al 2004, Rudman et al 2005, Pierson et al 2007, Sheu et al 2008). Staff views on error prevention mostly concerned paying increased attention and exercising caution during different phases of the medication process by everyone involved. Significantly, reporters thought that they or others had acted negligently, rather than blaming the system or environment for errors, despite the fact that, in recent years, the focus of patient safety has moved away from blaming individuals. This study found that staff often thought they had to work more competently. There are many environmental and system issues that affect the medication process; for example, rushing, interruptions and disruptions make it difficult to concentrate and administer treatments to patients safely (Wakefield et al 2005, Sheu et al 2008, Brady et al 2009, Palese et al 2009, Hewitt 2010). Many reporters highlighted insufficient staff resources as a risk and said that more staff and more peaceful environments were required to reduce errors. Interruptions during nurses’ drug rounds are notably common. For example, Biron et al (2009) observed 6.3 work interruptions per hour during medication administration, while Palese et al (2009) noted one interruption every third drug given. Research suggests that an important way to avoid this problem is to create a calm working atmosphere. Anthony et al (2010), for example, found that after implementation of a ‘no interruption zone’ system during medication preparation, interruptions decreased by 40.9 per cent. Verifying and checking drugs are vital for error prevention, particularly at the administration stage, 36 March 2013 | Volume 19 | Number 10
when there are many opportunities for mistakes while staff ensure the ‘right dose of the right drug is given to the right patient by the right route at the right time’ (Wulff et al 2011); distraction might partly explain why these ‘five rights’ are not observed. Furthermore, heavy workloads, long shifts and fatigue can affect nurses’ focus and might explain why the ‘five rights’ are no longer a part of the safe medication administration routine (Hewitt 2010). Reporters offered opinions on how medication processes might be changed. For example, some said drugs with similar sounding names or similar packaging caused problems and these are issues that have been reported previously (Wakefield et al 2005, Brady et al 2009, Hewitt 2010). The pharmaceutical industry should investigate the potential for eliminating confusion over drug names and packaging. Hospitals should ensure the appropriate and safe storage of medicines, and make sure staff are aware of the potential risk to patients if there is confusion over drugs or packaging. Lack of thorough patient identification is another risk factor during the medication process; even if staff are familiar with patients, they must check identities to ensure the right patients receive the right care (World Health Organization (WHO) 2007). The Institute for Safe Medication Practices (2011) recommends that at least two staff check patients’ identities during medication administration. Verbal and written information are risk factors, and various studies have highlighted problems with transfer of information, and with obscurity and inconsistency of medication information. Tjia et al (2009), for example, found medication discrepancies when patients were moved between institutions or care settings in almost three out of four admissions, which accounted for one in five medications prescribed on admission. Meanwhile, Cornish et al (2005) found that 53.6 per cent of patients had at least one medication discrepancy at the time of hospital admission, and Vira et al (2006) found that 60 per cent of patients before and after admission had at least one unintended variance between prescription and pre-admission medication use, as well as with inpatient medication at discharge. Medication variances during hospital admission and discharge are common and clinically important. Effective communication is crucial to patient safety (WHO 2008), and these findings emphasise the importance of improving communication and co-operation. NURSING MANAGEMENT
Conclusion Improving safety in the medication process is important for reducing the risk of harm to patients and for increasing the quality of health care. Voluntary web-based error-reporting systems are important tools for gathering information on medication errors because they enable healthcare staff to offer ideas about how to prevent mistakes. This study suggests that medication errors can be reduced or prevented by enabling staff to focus attention on the tasks at hand and by improving
work environments. Verifying and checking actions, fluent communication and staff co-operation are also vital. Hospital systems should encourage safe medication practice by providing the necessary resources, relating to workforce, education, induction and the provision of interruption-free zones. In many countries, legislation and guidelines are in place to promote patient safety, and these should be implemented, evaluated and followed in practice.
Online archive For related information, visit our online archive of more than 7,000 articles and search using the keywords Conflict of interest None declared
References Anthony K, Wiencek C, Bauer C et al (2010) No interruptions please: impact of a no interruption zone on medication safety in intensive care units. Critical Care Nurse. 30, 3, 21-29. Barker K, Flynn E, Pepper G et al (2002) Medication errors observed in 36 health care facilities. Archives of Internal Medicine. 162, 16, 1897-1903. Biron A, Lavoie-Tremblay M, Loiselle C (2009) Characteristics of work interruptions during medication administration. Journal of Nursing Scholarship. 41, 4, 330-336.
Griffin F, Resar R (2009) IHI Global Trigger Tool for Measuring Adverse events. Second edition. tinyurl.com/toolwhite013 (Last accessed: February 4 2013.)
Palese A, Sartor A, Costaperaria G et al (2009) Interruptions during nurses‘ drug rounds in surgical wards: observational study. Journal of Nursing Management. 17, 2, 185-192.
Tjia J, Bonner A, Briesacher B et al (2009) Medication discrepancies upon hospital to skilled nursing facility transitions. Journal of General Internal Medicine. 24, 5, 630-635.
Hewitt P (2010) Nurses‘ perceptions of the causes of medications errors: an integrative literature review. Medsurg Nursing. 19, 3, 159-167.
Pierson S, Hansen R, Greene S et al (2007) Preventing medication errors in long-term care: results and evaluation of a large scale web-based error reporting system. Quality and Safety in Health Care. 16, 4, 297–302.
van den Bemt P, Robertz R, De Jong A et al (2007) Drug administration errors in an institution for individuals with intellectual disability: an observational study. Journal of Intellectual Disability Research. 51, 7, 528-536.
Rozich J, Haraden C, Resar R (2003) Adverse drug event trigger tool: a practical methodology for measuring medication related harm. Quality and Safety in Health Care. 12, 3, 194-200.
Vira T, Colquhoun M, Etchells E (2006) Reconcilable differences: correcting medication errors at hospital admission and discharge. Quality and Safety in Health Care. 15, 2, 122-126.
Hicks RW, Cousins DD, Williams RL (2004) Selected medication-error data from USP’s Medmarx program for 2002. American Journal of Health-System Pharmacy. 61, 10, 993-1000.
Brady A-M, Malone A-M, Fleming S (2009) A literature review of the individual and system factors that contribute to medication errors in nursing practise. Journal of Nursing Management. 17, 6, 679-697.
Institute for Safe Medication Practices (2011) Oops, Sorry, Wrong Patient! A Patient Verification Process is Needed Everywhere, Not Just at the Bedside. Medication Safety Alert. tinyurl.com/acutecare013 (Last accessed: February 4 2012.)
Cornish P, Knowles S, Marchesano R et al (2005) Unintended medication discrepancies at the time of hospital admission. Archives of Internal Medicine. 165, 4, 424-429.
McBride-Henry K, Foureur M (2006) Medication administration errors: understanding the issues. Australian Journal of Advanced Nursing. 23, 3, 33-41.
Evans S, Berry J, Smith B et al (2006) Attitudes and barriers to incident reporting: a collaborative hospital study. Quality and Safety in Health Care. 15, 1, 39-43.
Miller M, Clark J, Lehmann C (2006) Computer based medication error reporting: insight and implications. Quality and Safety in Health Care. 15, 3, 208-213.
Ford C, Killebrew J, Fugitt P et al (2006) Study of medication errors on a community hospital oncology ward. Journal of Oncology Practise. 2, 4, 149-154.
Ndosi M, Newell R (2008) Nurses’ knowledge of pharmacology behind drugs they commonly administer. Journal of Clinical Nursing. 18, 4, 570-580.
NURSING MANAGEMENT
Rudman W, Bailey J, Hope C et al (2005) The impact of a web-based reporting system on the collection of medication error occurrence data. Advances in Patient Safety. 3, 193-205. Savage S, Schneider P, Pedersen C (2005) Utility of an online medication-error-reporting system. American Journal of Health-System Pharmacy. 62, 21, 2265-2270. Sheu S-J, Wie I-L, Chen C-H et al (2008) Using snowball sampling method with nurses to understand medication administration errors. Journal of Clinical Nursing. 18, 4, 559-569. Thomas A, Panchagnula U (2008) Medicationrelated patient safety incidents in critical care: a review of reports to the UK National Patient Safety Agency. Anaesthesia. 63, 7, 726-733.
Wakefield B, Uden-Holman T, Wakefield D (2005) Development and validation of the medication administration error reporting survey. Advances in Patient Safety. 4, 475-488. Wulff K, Cummings G, Marck P et al (2011) Medication administration technologies and patient safety: a mixed-method systematic review. Journal of Advanced Nursing. 67, 10, 2080-2095. World Health Organization (2007) Patient Identification. Patient Safety Solutions. WHO, Geneva. World Health Organization (2008) Patient Safety Workshop. Learning from Error. WHO, Geneva.
March 2013 | Volume 19 | Number 10 37
