Review
An integrative review of drug errors in critical care
Journal of the Intensive Care Society 2016, Vol. 17(1) 63–72 ! The Intensive Care Society 2015 Reprints and permissions: sagepub.co.uk/ journalsPermissions.nav DOI: 10.1177/1751143715605119 jics.sagepub.com
Caroline C MacFie, Simon V Baudouin and Peter B Messer
Abstract Medication error is the commonest cause of medical error and the consequences can be grave. This integrative review was undertaken to critically appraise recent literature to further define prevalence, most frequently-implicated drugs and effects on patient morbidity and mortality in the critical care environment. Forty studies were compared revealing a markedly heterogeneous data set with significant variability in reported incidence. There is an important differentiation to be made between medication error (incidence 5.1–967 per 1000 patient days) and adverse drug event (incidence 1–96.5 per 1000 patient days) with significant ramifications for patient outcome and cost. The most commonly implicated drugs were cardiovascular, gastrointestinal, antimicrobial and hypoglycaemic agents. Beneficial interventions to reduce such errors include computerised prescribing, education and pharmacist input. The studies described provide insight into suboptimal management in the critical care environment and have implications for the development of specific improvement strategies and future training.
Keywords Medication errors, adverse drug event, medical errors, pharmaceutical preparations, critical care
Introduction Medical errors are increasingly high profile and the focus of both scientific investigation and potential litigation. Medical errors have been found to be the eighth commonest cause of death in the USA.1 Apart from substantial morbidity and mortality, there are also consequences on clinical governance, resource distribution and a substantial economic burden. Medication error (ME) is the commonest cause of medical error with previous estimates of incidence of 9.1/1000 patient days.2 The consequences of ME can range from minor to catastrophic, and cases of ME which have caused patient harm and even death are well reported. As such, the topic has attracted much interest from patient safety groups.3 Medication error is particularly pertinent in the critical care setting. Patients are vulnerable as a consequence of a reduced physiological reserve, reduced ability to metabolise drugs and alterations in pharmacodynamics. Polypharmacy is common along with prescriptions of drugs to which patients have not been previously exposed. The use of high risk substances and varied routes of administration is more prevalent and these factors occur in the setting of busy and highly pressurised environments.
Medication errors are often cited as being one of the most common sources of error within hospitals2,4,5 but the prevalence of such errors, the most frequently implicated drugs and the effects on patient morbidity and mortality have not been recently systematically reviewed. Previous studies have tended to focus on specific areas of error, e.g. parenteral drug use or prescriptions only,6–8 and investigated only those studies performed in the developed world.9 Initial review of the literature showed that published studies are very heterogeneous in trial methodology, study population, setting and methods of data presentation. The primary data did not lend themselves to a systematic review with meta-analysis. An integrative review was therefore undertaken to critically appraise the recent literature to determine the prevalence of ME in the critical care environment, Department of Anaesthesia & Critical Care, Royal Victoria Infirmary, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK Corresponding author: Caroline MacFie, Department of Anaesthesia & Critical Care, Royal Victoria Infirmary, Newcastle upon Tyne, UK. Email: [email protected]
64
Journal of the Intensive Care Society 17(1)
the drugs most commonly associated with ME and the consequences of ME.
Further, ‘snowballing’ from the reference lists of all relevant full text studies identified from the electronic search was undertaken.
Methods An integrative review is a reviewing process summarising available literature to provide a more comprehensive understanding of a clinical problem. It is more structured and rigorous than a narrative review in situations where the primary data do not support the use of a systematic review and meta-analysis. Integrative reviews present the current understanding of a subject and contribute to theory development. They have gained favour in academic nursing journals and represent a process well suited to many areas of current intensive care medical practice. The integrative review was structured to incorporate the five-step review framework described by Whittemore and Knafl10: 1. 2. 3. 4. 5.
Problem identification Literature search Data evaluation Data extraction, synthesis and analysis Presentation of results
Problem identification This has been described in the introduction. The aim of this integrative review is to answer the following questions: . What is the prevalence of ME in the setting of critical care? . What are the drugs most commonly associated with ME in the setting of critical care? . What are the consequences of ME in the setting of critical care?
Literature search Studies published from 1995 to June 2015 were identified using Medline, CINAHL and Embase databases. The following search strategy was used: MeSH terms: Medication Errors, Medical Errors, Pharmaceutical Preparations, Critical Care, Intensive Care, and Intensive Care Units were used in conjunction with title and abstract searches of ‘‘medication error’’, ‘‘medical error’’, mistake, ‘‘adverse event’’, drug, ‘‘level 2’’, ‘‘level two’’, ‘‘level 3’’, ‘‘level three’’, ‘‘critical care’’, ‘‘intensive care’’, ‘‘high dependency’’, hdu, icu, and ‘‘intensive treatment unit’’. The amalgamation of the three broad categories of error with drug with critical care gave the final results which were then limited to English Language, Humans and Adults. A hand search was also performed of personal files and reference lists of relevant review articles.
Data evaluation Relevant studies were identified by title, abstract and finally full text. Studies were included for review if the following predefined inclusion criteria were met: . Rates of ME reported . ME investigated and reported in a critical care setting . ME rates reported in isolation rather than in conjunction with other forms of medical error. Studies were excluded for review if the following predefined exclusion criteria were present: . Studies outside the setting of critical care or studies combining data from critical care and non-critical care settings . Studies of primarily paediatric populations . Review articles, correspondence, abstracts, dissertations, and editorials . Full text not available in English. Abstracts were reviewed separately by two of the authors, and any discrepancy between the decision on whether to include in the final analysis was resolved by discussion. Complete articles were then analysed for type of study, incidence of medication error and type of error investigated, associated adverse events and most commonly implicated drugs.
Definitions Medication error (ME): any mistake in the prescription, preparation or administration of a drug which does not necessarily cause harm.11 Adverse drug event (ADE): a medication error where harm occurs.11
Data extraction, synthesis and analysis For each study data were extracted to define study characteristics. A qualitative synthesis of the primary data was then undertaken to determine the prevalence and consequences of ME and the most commonly implicated medications.
Presentation of results Data are presented in narrative and tabular form.
Results Our primary literature search produced 572 abstracts which were reviewed to identify 85 studies for
MacFie et al. analysis. Forty-five studies were discarded as they did not meet the inclusion criteria for this study. Forty studies met our inclusion criteria and are summarised in Table 1. The findings of our literature review can be summarised within the following domains: study methodology; units used to report medication error incidence; overall incidence of medication error; severity of medication error; and drugs implicated in medication errors.
65 study as being required after 23.5% of ME.26 Serious harm (3.39 to 64.8% of ADEs)30,38 and fatal consequences (between 0.45 to 25% total ADEs)24,30 of ADEs were also described.
Drugs implicated in medication errors Drugs commonly implicated in this review are summarised in Table 2.
Effect of different interventions Study methodology Study methodology varied between studies. Data were gathered either prospectively or retrospectively or as part of an interventional study to identify medication error incidence prior to and following an intervention. Errors were identified using three main methodologies: chart review, voluntary reporting or direct observation.
Units used to report medication error incidence The units used to report medication errors also varied between studies. The most common units used to report medication error incidence were medication errors per 1000 patient days. Other units used included medication errors per 100 admissions, percentage of patients experiencing one or more medication errors, percentage of medication error as a proportion of overall drug administration or medication errors per 100 medication orders.
Overall incidence of error The incidence of ADE and ME varied widely between studies. The incidence of ADEs ranged widely from 1 to 96.5 per 1000 patient days37,38 or 1.3 to 21.1 when reported as per 100 admissions.25,29 When data were reported in different units, one study reported the incidence of ADEs as 34.1% of patients31 and another as 0.16 per 100 medication orders.24 The incidence of ME ranged from 5.1 to 967 per 1000 patient days,22,44 or when reported as per 100 admissions – 0.88–780.24,26 Percentage of medication error as a proportion of overall drug administration ranged from 1.96 to 69.7%.17,39 One study reported the incidence in units per 100 orders – 10 per 100 orders.24
Severity of medication error Severities of ADEs were described in terms of cost, monitoring implications and patient harm. Extra cost per patient varied between $34716 and $6647.14 Increased monitoring was reported in only one
Eleven of the studies compared pre- and post-intervention data. The types and effects of the various interventions are listed in Table 3.
Discussion The main findings of this integrative review into drug errors in a critical care setting were: forty studies reported the incidence of medication error or adverse drug event in critical care; these errors are common in critical care; there are wide variations in estimates of incidence reflecting diverse methodology of studies; the consequences are often severe; and the drugs implicated are both specific to critical care but also include those which are commonly implicated in noncritical care settings. Forty studies met the inclusion criteria. Not all studies specifically investigated the incidence of medication error as their primary outcome. Eleven of the studies were interventional studies investigating the impact of an intervention on the incidence of medication errors in critical care. Medication errors are relatively common (1 to 96.5 per 1000 patient days) in a critical care setting. Polypharmacy is common along with prescriptions of drugs to which patients have not been previously exposed. Medication reconciliation may not be prioritised and the use of high risk substances and varied routes of administration is more prevalent. All these factors occur in the setting of busy and highly pressurised environments. Perhaps the most interesting finding of this integrative review was the wide variation in nomenclature, quality and methods of data collection which clouds interpretation and comparison of studies, and does not allow for meta-analyses. ME is increasingly accepted to mean any mistake in the prescription, preparation or administration of a drug which does not necessarily cause harm in contrast to ADE that refers to a medication error where harm occurs.11 One of the sentinel studies in this field remains that of Bates et al.2 who emphasised the importance of consistent definition and explains the term ADE as, for example, a patient with first degree heart block inappropriately receiving a beta blocker resulting in complete heart block requiring pacing. This is in contrast to a potential ADE synonymous
Holland
Switzerland
2008
2009
2012
1996
2008
2004
2003
2002
2012
2012
Nuckols et al.14
Benkirane et al.15
Jiang et al.16
Nettleman and Bock17
Fahimi et al.18
Herout and Erstad19
Fraenkel et al.20
van den Bemt et al.21
Pagnamenta et al.22
Agalu et al.23
Ethiopia
Australia
USA
Iran
USA
China
Morocco
USA
USA
1995
USA
2014
Korea
Country
Carayon et al.13
2014
Year
Bates et al.2
Cho et al.
12
Reference
P, O, N&PhS #MDT*
P, VR, I (electronic prescribing, standardised labelling & pumps, pharmacist), N&PS #MDT
P, O, PhS #MDT
P, VR, I (computerised clinical information system), NS #MDT
P, CR – continuous infusions only, PhS #PH
P, O – IV only, PhS #PH
P, CR – missed doses only, NS #MDT
P, CR, I (pharmacist), P&PhS #MDT
P, O, VR, P&PhS #MDT
P, CR - IV only, N&PS #MDT
P, VR, CR, PS #National study group
P, VR, CR, NS #MDT
CR, O, NS #MDT
Designa #Study leads
Table 1. Summary of included studies.
Percentage
Other
56.35
21.9 academic hospital, 17.4 non-academic hospital
11.5
9.2
15.36
11.5
6.1
Pre 506 incidents Post 478
Pre 85 incidents Post 55
105.9
7.7
9.1
276
7.5
4.8
/100 admissions
51.8% administrations
44.6% administrations
9.4% of administrations
1.96% of administrations missed
53.6% prescriptions 19% administrations
Percentage
/1000 patient days
/1000 patient days /100 admissions
Incidence ME/potential ADE
Incidence ADE
Pre 1.68 errors/pt Post 0.46
Other
Antibiotics Analgesics Antiepileptics
GI
Antibiotics GI Steroids
Insulin Local anaesthetics
Opiates Sedatives Anticoagulants
Analgesics Antibiotics Sedatives
Drugs
Cost
(continued)
L-T 18.75% Death 2.5% Serious 51.8%
Cost
L-T 12% Fatal 1% Serious 30%
Severity (of ADEs unless otherwise stated)
2010
2005
2008
2006
2012
2011
2010
2012
2011
2010
Kane-Gill et al.26
Rothschild et al.27
Nuckols et al.28
Kane-Gill et al.29
Vazin and Delfani30
Reis and Cassiani31
Ford et al.32
Benoit et al.33
Seynaeve et al.34
Morimoto et al.35
2011
2013
24
Year
Aljadhey et al.25
Jennane et al.
Reference
Table 1. Continued.
Japan
Belgium
Switzerland
USA
Brazil
Iran
USA
USA
USA
USA
Saudi Arabia
Morocco
Country
P, CR, PS #MDT
P, CR, VT #MDT
P, CR, I (stream-lining medication process), PhS #MDT
P, O, I (education), PhS #PH
P, CR, N&PhS #MDT*
P, O, PhS#PH
R, VR, PhS #PH
R, CR, I (smart pumps) – IV only, N&PS #MDT
P, O, VR, CR, PS&VT #MDT
R, VR, PhS #PH
P, VR, CR, P&PhS #MDT
P, VR, CR, N&P&PhS #MDT
Designa #Study leads Percentage
Other
2.4
2.4
4.86
37.8
15.7
1.3
21.1
12.6
34.1% patients
0.16/100 orders
228
127.8
36.2
967
0.88
38.2
780
/100 admissions
/1000 patient days
/1000 patient days /100 admissions
Incidence ME/potential ADE
Incidence ADE
Pre 4.95% administrations Post 2.14%
CCU Pre 30.8% administrations Post 4% MICU: pre 20.8% Post 22.7%
7.6% administrations
Percentage 10/100 orders
Other
Antibiotics Sedatives NSAIDS
Hypoglycaemics
Opiates Sedatives Potassium
Antibiotics
Antibiotics Analgesics Anticoagulants
Opiates Insulin Sedatives
CVS Anticoagulants Antibiotics
Opiates Antibiotics Anticoagulants
Anticoagulants Antibiotics Antihypertensives
Antibiotics Anticoagulants Steroids
Drugs
(continued)
Serious 4%
Of ADRs: Serious 20%
Of medication errors: Fatal 0.45% Serious 3.39%
Increased monitoring
Serious 35% L-T 6%
Death 25%
Severity (of ADEs unless otherwise stated)
36
Jordan
Iran
2006
2013
2014
2012
2007
2000
1997
2006
2007
2004
Colpaert et al.7
Nguyen et al.8
Nazer et al.38
Vazin and Fereidooni39
Fahimi et al.40
Lustig41
Cullen et al.42
Valentin et al.43
Chacko et al.44
Ridley et al.5
UK
India
European
USA
Israel
Iran
Vietnam
Belgium
Holland
2010
Klopotowska et al.37
European
USA
Country
2009
2009
Year
Valentin et al.6
Deyoung et al.
Reference
Table 1. Continued
P, CR, not stated #MDT
P, VR, PS #Physician
P, VR, not stated #Multinational study group
P, VR, CR, N&P&PhS #MDT
P, CR, PhS #PH
P, O - IV infusions only, PhS #PH
P, O, PhS #PH
P, CR, PhS #MDT
P, O, I (pharmacist led training), N&PhS #PH
P, CR, I (electronic prescribing), P&PhS #MDT
P, I (pharmacist) prescription errors only, P&PhS #MDT
P, VR - IV only, PS #Multinational study group
P, O, I (barcode assisted administration), N&PhS #MDT*
Designa #Study leads Percentage
Other
1.6
96.5
4 pre, 1 post
0
10.6% of errors were ADEs
Of prescriptions: Pre 0.98% Post 0.16%
5.1
105
19
8.2
190.5 pre, 62.5 post
745
/100 admissions
/1000 patient days
/1000 patient days /100 admissions
Incidence ME/potential ADE
Incidence ADE
15% of prescriptions
7.8% of administrations
69.7% of administrations
Of administrations:Pre 64%, Post 48.9%
Of prescriptions: Pre 27% Post 3.4%
Pre 19.7% administrations Post 8.7%
Percentage
Other
Electrolytes Anticoagulants Paracetamol
CVS Hypoglycaemics
Benzodiazepines
CVS CNS
Hypoglycaemics Antibiotics Sedatives
Antibiotics CVS
Antibiotic Antithrombotics, Antiepileptics
Antibiotics Sedatives Vasopressors
GI CVS
Drugs
(continued)
Of medication errors 10.6% led to harm
Serious/ L-T 64.8% Fatal 1.1%
Maximum was temporary harm
Of total patients: 0.9% permanent harm/death
Severity (of ADEs unless otherwise stated)
Of medication errors: Serious 19%
Percentage
P, O, P&PS #National study group 2015 Shulman et al.46
UK
1999 Leape et al.45
USA
P, CR, I (pharmacist), N&P&PhS #MDT
Preventable: pre 10.4, post 3.5 All ADEs: 33 pre, 11.6 post
/100 admissions Designa #Study leads Country Year Reference
P ¼ prospective; CR ¼ chart review; VR ¼ voluntary reporting; O ¼ observational; I ¼ interventional; CCU ¼ Coronary Critical Care Unit; MICU ¼ Medical Intensive Care Unit; L-T ¼ long-term. Italics ¼ how the errors were scored. PS ¼ physician scored; PhS ¼ pharmacist scored; NS ¼ nurse scored; VT ¼ validated tool. #Study leads. #MDT ¼ multidisciplinary team led; #MDT* ¼ multidisciplinary team led not including physician involvement; #PH ¼ pharmacist led. Others as stated.
/1000 patient days /1000 patient days
Other
Incidence ME/potential ADE Incidence ADE
/100 admissions
Percentage
6.8% prescriptions
Other
Drugs
Severity (of ADEs unless otherwise stated)
Table 1. Continued
with ME which refers to, for example, a patient with a documented allergy receiving said drug but not reacting.2 Some studies investigate the so-called preventable ADEs as decided by an ICU pharmacist and physician,37 whilst others define minor, intercepted and serious medication prescription errors (MPE).7 In contrast, medication error was also described as any deviation in drug preparation and administration from the doctors’ prescription, hospital policies or manufacturer’s instructions such as diluted in saline as opposed to water.8 The variations in definition are numerous and are not all listed here. Review of the two extremes of reported ME/potential ADE per 1000 patient days included one study that reported only 5.1 MEs per 1000 patient days despite documenting anything from delayed dose of a drug to accidental intravenous administration of planned intrapleural streptokinase.44 Incident outcome was not separately differentiated between other non-drug-related incidents that occurred. At the other end of the spectrum with the most MEs of 967 per 1000 patient days was a study that included an extensive array of ‘discrepancies’ such as prescription omissions such as ‘‘route’’ not written next to a prescribed drug.20 This may partly explain why their results were disproportionately higher than the other studies analysed. Clearly, the frequency of prescribing errors would be expected to be higher than that of actual error or harm to the patient. The differing definitions and methodology are reflected in the wide range of estimates for the incidence of ME and ADE reported by this review. The variation in reported incidence of ADE from 1 to 96.5 per 1000 patient days may reflect that the former study37 used a prospective interventional study looking at impact of pharmacist intervention into medical prescribing, whereas the latter38 used chart review and incorporated adverse drug reactions (i.e. those that are unavoidable such as anaphylaxis) into their ADE figures. Furthermore, various studies examined a single aspect of drug administration6,14,18,19,28,40 such as parenteral medications only; solely missed doses,17 or a prescriptions-only study design.37 Eleven of the studies analysed incorporated preand post-intervention data (see Table 3). As far as conclusions can be drawn from the array of data, it seems that significant benefit arises from computerised prescribing, education and the welcome involvement of our pharmacist colleagues. The only intervention that appeared to cause increased patient risk was due to the so-called smart pumps, i.e. pumps that alert users when programmed to deliver duplicate infusions or doses out-with defined ranges. The proposed explanation was that the pumps led to greater identification of potential errors before harm resulted.28 Chart review was the most commonly used method of identifying error with 20 of the 40 studies analysed using this method, employing unblinded nurse
70
Journal of the Intensive Care Society 17(1)
investigators, pharmacists and physicians. Fourteen used voluntary reporting methods by nurses and pharmacists which could lead to under-estimation of the incidence of ME. Thirteen used a direct observational method by independent pharmacy students,
Table 2. Drugs implicated in medication errors (for example, 32% is the mean average for the five studies that documented percentage of their total error being related to cardiovascular drugs.). Drug category
Percentage of total study errors
Cardiovascular Gastrointestinal Antimicrobial Hypoglycaemic Analgesic Antithrombotic Sedative Antihypertensive Antiepileptic Steroids Electrolytes Local anaesthesia
32 26 22 22 18 18 16 16 11 10 8 6
Note: Studies documenting ME as well as subgroup ADEs are combined.
pharmacists, nurses or study physicians. Only one study used direct observation in conjunction with reporting in the per 1000 patient day format.27 It might follow that using this method might result in a higher reported incidence of ADEs and MEs and apart from one other study38 this was the case with 37.8 ADEs per 1000 patient days. Even using independent observational reporting the incidence of medication error still varied widely and was reported between 7.6 and 69.7% of administrations: 7.6% referred to potential opportunities for occurrence of error per patient while 69.7% referred to errors as a proportion of doses given.30,39 Severity of consequences resulting from an ADE was concerning, ranging from cost and increased monitoring implications, through serious harm (3.39–64.8% of ADEs) to fatal in between 0.45 and 25% total ADEs. Patients in critical care are particularly vulnerable as a consequence of a reduced physiological reserve, reduced ability to metabolise drugs and alterations in pharmacodynamics. New medications are introduced with potential for interaction and pre-existing medications may be omitted due to other clinical priorities. Drugs commonly implicated in this review were antibiotics, sedatives, analgesics, anticoagulants, hypoglycaemics, antiepileptics, gastrointestinal and cardiovascular medications. These medications are implicated in medication errors in wider patient populations and have attracted interest from patient safety
Table 3. Effect of interventions on ME/ADE. Type of intervention implemented
ME/ADE pre intervention
ME/ADE post intervention
Pharmacist – order reviews, drug monitoring
1.68 MEs per patient
0.46 MEs per patient
(3 studies)
190.5 MEs per 1000 patient days 4 ADEs per 1000 patient days 33 ADEs per 1000 patient days
62.5 MEs per 1000 patient days 1 ADE per 1000 patient days 11.6 ADEs per 1000 patient days
Computerised clinical information system
85 ADEs in 2 years
55 ADEs in 2 years
Electronic prescribing
ME in 27% prescriptions ADE in 0.98% prescriptions
ME in 3.4% prescriptions ADE in 0.16% prescriptions
Electronic prescribing, standardised labelling & pumps, pharmacist input
506 ADEs in study period
478 ADEs in study period
Smart pumps
4.78 ADEs per 1000 patient days
4.95 ADEs per 1000 patient days
Barcoding of drugs
ME in 19.7% administrations
ME in 8.7% administrations
Education programme
ME in 30.8% administrations (CCU) ME in 20.8% administrations (MICU)
ME in 4% administrations (CCU) ME in 22.7% administrations (MICU)
Pharmacist led training
ME in 64% administrations
ME in 48.9% administrations
Streamlining of medication process – standardising documentation
ME in 4.95% administrations
ME in 2.14% administrations
ME ¼ medical error; ADE ¼ adverse drug event; CCU ¼ coronary critical care unit; MICU ¼ medical intensive care unit.
MacFie et al. groups.3 The reasons for this are beyond the scope of this review, but may be due to the frequent use of these medications or may be inherent to them or their administration process. The question remains whether certain drugs are inherently more prone to error, or indeed harm, or whether it is increased usage of particular drugs or their prescription that is more problematic. This review covers recent literature on medication error and adverse drug events in a more diverse setting than previously published. Errors are common with potentially severe consequences, and research into this area requires standardised definitions and ideas regarding data collection to allow us to make more meaningful conclusions. A number of interventions have indicated improvement in service such as computer-aided drug ordering systems, training, standardised practices, and of vital importance the daily presence and input of critical care pharmacists as supported by the recent PROTECTED-UK study.46 Other practices that may warrant investigation into whether they can reduce error include colour coding of drug packaging and labels, bar-coding, standardised storage, double checking prior to administration (widely performed but with little evidence base), techniques to minimise interruptions (e.g. ‘‘Do Not Disturb, Drug Round’’ tabards for nurses – perhaps more relevant in the ward environment), and checklists. We recommend a further large-scale national prospective observational trial using set definitions as per the Institute of Medicine and clarified by Bates2,11 to help us progress our knowledge and hence our safety mechanisms in this fundamental area. By understanding the depth of the problem in the intensive care setting, and realising where the risks are greatest, we can endeavour to improve our practice and future training.
Conclusion Drug errors in critical care are common and have serious deleterious effects on morbidity and mortality. This review incorporated a specific integrative review framework that identified a particularly heterogeneous range of research methods and nomenclature in this area making data interpretation difficult. There is an important differentiation to be made between medication error (incidence 5.1– 967 per 1000 patient days) and adverse drug event (incidence 1–96.5 per 1000 patient days) with significant ramifications for patient outcome and cost. The most commonly implicated drug groups include cardiovascular, gastrointestinal, antimicrobial, hypoglycaemic and analgesic agents. The studies described in this review provide insight into suboptimal areas of patient care in the critical care environment and have implications for the development of specific improvement strategies and future training.
71 Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
References 1. Kohn LT, Corrigan JM, Donaldson MS (eds); for the Committee on Quality of Health Care in America, Institute of Medicine. To err is human: Building a safer health system. Washington, DC: National Academy Press, 2000. 2. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA 1995; 274: 29–34. 3. http://www.england.nhs.uk/wp-content/uploads/2014/ 03/psa-sup-info-med-error.pdf 4. Cullen DJ, Bates DW, Leape LL, et al. Prevention of adverse drug events: a decade of progress in patient safety. J Clin Anesth 2000; 12: 600–614. 5. Ridley SA, Booth SA, Thompson CM, et al. Prescription errors in UK critical care units. Anaesthesia 2004; 59: 1193–1200. 6. Valentin A, Capuzzo M, Guidet B, et al. on behalf of the Research Group on Quality Improvement of the European Society of Intensive care Medicine (ESICM) and the Sentinel Events Evaluation (SEE) Study Investigators. Errors in administration of parenteral drugs in intensive care units: multinational prospective study. BMJ 2009; 338: b814. 7. Colpaert K, Claus B, Somers A, et al. Impact of computerized physician order entry on medication prescription errors in the intensive care unit: A controlled crosssectional trial. Crit Care 2006; 10: R21. 8. Nguyen HT, Pham HT, Vo DK, et al. The effect of a clinical pharmacist-led training programme on intravenous medication errors: A controlled before and after study. BMJ Qual Saf 2013; 0: 1–6. 9. Wilmer A, Louie K, Dodek P, et al. Incidence of medication errors and adverse drug events in the ICU: A systematic review. Qual Saf Health Care 2010; 19: e7. 10. Whittemore R, and Knafl K. The integrative review: Updated methodology. J Adv Nurs 2005; 52: 546–553. 11. Institute of Medicine. Preventing medication errors. Washington, DC: National Academies Press, 2007. 12. Cho I, Park H, Choi Y, et al. Understanding the nature of medication errors in an ICU with a computerized physician order entry system. PLoS ONE 2014; 9. 13. Carayon P, Wetterneck TB, Cartmill R, et al. Characterising the complexity of medication safety using a human factors approach: an observational study in two intensive care units. BMJ Qual Saf 2014; 23: 56–65. 14. Nuckols TK, Paddock SM, Bower AG, et al. Costs of intravenous adverse drug events in academic and nonacademic intensive care units. Med Care 2008; 46: 17–24. 15. Benkirane RR, Abouqal R, Haimeur CC, et al. Incidence of adverse drug events and medication errors in intensive care units: A prospective multicenter study. J Patient Saf 2009; 5: 16–22.
72 16. Jiang SP, Zheng X, Li X, et al. Effectiveness of pharmaceutical care in an intensive care unit from China. A pre- and post-intervention study. Saudi Med J 2012; 33: 756–762. 17. Nettleman MD, and Bock MJ. The epidemiology of missed medication doses in hospitalized patients. Clin Perform Qual Health Care 1996; 4: 148–153. 18. Fahimi F, Ariapanah P, Faizi M, et al. Errors in preparation and administration of intravenous medications in the intensive care unit of a teaching hospital: An observational study. Aust Crit Care 2008; 21: 110–116. 19. Herout PM, and Erstad BL. Medication errors involving continuously infused medications in a surgical intensive care unit. Crit Care Med 2004; 32: 428–432. 20. Fraenkel DJ, Cowie M, and Daley P. Quality benefits of an intensive care clinical information system. Crit Care Med 2003; 31: 120–125. 21. van den Bemt PML, Rijn R, van der Voort PHJ, et al. Frequency and determinants of drug administration errors in the intensive care unit. Crit Care Med 2002; 30: 846–850. 22. Pagnamenta A, Rabito G, Arosio A, et al. Adverse event reporting in adult intensive care units and the impact of a multifaceted intervention on drug-related adverse events. Ann Intens Care 2012; 2. 23. Agalu A, Ayele Y, Bedada W, et al. Medication administration errors in an intensive care unit in Ethiopia. Int Arch Med 2012; 5. 24. Jennane N, Madani N, Oulderrkhis R, et al. Incidence of medication errors in a Moroccan medical intensive care unit. Int Arch Med 2011; 4. 25. Aljadhey H, Mahmoud MA, Mayet A, et al. Incidence of adverse drug events in an academic hospital: a prospective cohort study. Int J Qual Health Care 2013; 25: 648–655. 26. Kane-Gill SL, Kowiatek JG, and Weber RJ. A comparison of voluntarily reported medication errors in intensive care and general care units. Qual Saf Health Care 2010; 19: 55–59. 27. Rothschild JM, Landrigan CP, Cronin JW, et al. The Critical Care Safety Study: The incidence and nature of adverse events and serious medical errors in intensive care. Crit Care Med 2005; 33: 1694–1700. 28. Nuckols TK, Bower AG, Paddock SM, et al. Programmable infusion pumps in ICUs: an analysis of corresponding adverse drug events. J Gen Intern Med 2008; 23: 41–45. 29. Kane-Gill S, Rea RS, Verrico MM, et al. Adverse-drugevent rates for high-cost and high-use drugs in the intensive care unit. Am J Health Syst Pharm 2006; 63: 1876–1881. 30. Vazin A, and Delfani S. Medication errors in an internal intensive care unit of a large teaching hospital: A direct observation study. Acta Medica Iranica 2012; 50: 425–432. 31. Reis AMM, and Cassiani SHDB. Adverse drug events in an intensive care unit of a university hospital. Eur J Clin Pharmacol 2011; 67: 625–632.
Journal of the Intensive Care Society 17(1) 32. Ford DG, Seybert AL, Smithburger PL, et al. Impact of simulation-based learning on medication error rates in critically ill patients. Intens Care Med 2010; 36: 1526–1531. 33. Benoit E, Eckert P, Theytaz C, et al. Streamlining the medication process improves safety in the intensive care unit. Acta Anaesthesiol Scand 2012; 56: 966–975. 34. Seynaeve S, Verbrugghe W, Claes B, et al. Adverse drug events in intensive care units: A cross-sectional study of prevalence and risk factors. Am J Crit Care 2011; 20: 131–140. 35. Morimoto T, Sakuma M, Matsui K, et al. Incidence of adverse drug events and medication errors in Japan: The JADE study. J Gen Intern Med 2011; 26: 148–153. 36. DeYoung JL, Vanderkooi ME, and Barletta JF. Effect of bar-code-assisted medication administration on medication error rates in an adult medical intensive care unit. Am J Health Syst Pharm 2009; 66: 1110–1115. 37. Klopotowska JE, Kuiper R, van Kan HJ, et al. Onward participation of a hospital pharmacist in a Dutch intensive care unit reduces prescribing errors and related patient harm: An intervention study. Crit Care 2010; 14: R174. 38. Nazer LH, Hawari F, and Al-Najjar T. Adverse drug events in critically Ill patients with cancer: Incidence, characteristics, and outcomes. J Pharm Pract 2014; 27: 208–213. 39. Vazin A, and Fereidooni M. Determining frequency of prescription, administration and transcription errors in internal intensive care unit of Shahid Faghihi hospital in Shiraz with direct observation approach. Iran J Pharmaceut Sci 2012; 8: 189–194. 40. Fahimi F, Sistanizad M, Abrishami R, et al. An observational study of errors related to the preparation and administration of medications given by infusion devices in a teaching hospital. Iran J Pharm Res 2007; 6: 295–299. 41. Lustig A. Medication error prevention by pharmacists – An Israeli solution. Pharm World Sci 2000; 22: 21–25. 42. Cullen DJ, Sweitzer BJ, Bates DW, et al. Preventable adverse drug events in hospitalized patients: A comparative study of intensive care and general care units. Crit Care Med 1997; 25: 1289–1297. 43. Valentin A, Capuzzo M, Guidet B, et al. Patient safety in intensive care: results from the multinational Sentinel Events Evaluation (SEE) study. Intens Care Med 2006; 32: 1591–1598. 44. Chacko J, Raju HR, Singh MK, et al. Critical incidents in a multidisciplinary intensive care unit. Anaesth Intens Care 2007; 35: 382–386. 45. Leape LL, Cullen DJ, Clapp MD, et al. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. JAMA 1999; 282: 267–270. 46. Shulman R, McKenzie CA, Landa J, et al. Pharmacist’s review and outcomes: Treatment-enhancing contributions tallied, evaluated, and documented (PROTECTED-UK). J Crit Care 2015; 30: 808–813.
An integrative review of drug errors in critical care
Journal of the Intensive Care Society 2016, Vol. 17(1) 63–72 ! The Intensive Care Society 2015 Reprints and permissions: sagepub.co.uk/ journalsPermissions.nav DOI: 10.1177/1751143715605119 jics.sagepub.com
Caroline C MacFie, Simon V Baudouin and Peter B Messer
Abstract Medication error is the commonest cause of medical error and the consequences can be grave. This integrative review was undertaken to critically appraise recent literature to further define prevalence, most frequently-implicated drugs and effects on patient morbidity and mortality in the critical care environment. Forty studies were compared revealing a markedly heterogeneous data set with significant variability in reported incidence. There is an important differentiation to be made between medication error (incidence 5.1–967 per 1000 patient days) and adverse drug event (incidence 1–96.5 per 1000 patient days) with significant ramifications for patient outcome and cost. The most commonly implicated drugs were cardiovascular, gastrointestinal, antimicrobial and hypoglycaemic agents. Beneficial interventions to reduce such errors include computerised prescribing, education and pharmacist input. The studies described provide insight into suboptimal management in the critical care environment and have implications for the development of specific improvement strategies and future training.
Keywords Medication errors, adverse drug event, medical errors, pharmaceutical preparations, critical care
Introduction Medical errors are increasingly high profile and the focus of both scientific investigation and potential litigation. Medical errors have been found to be the eighth commonest cause of death in the USA.1 Apart from substantial morbidity and mortality, there are also consequences on clinical governance, resource distribution and a substantial economic burden. Medication error (ME) is the commonest cause of medical error with previous estimates of incidence of 9.1/1000 patient days.2 The consequences of ME can range from minor to catastrophic, and cases of ME which have caused patient harm and even death are well reported. As such, the topic has attracted much interest from patient safety groups.3 Medication error is particularly pertinent in the critical care setting. Patients are vulnerable as a consequence of a reduced physiological reserve, reduced ability to metabolise drugs and alterations in pharmacodynamics. Polypharmacy is common along with prescriptions of drugs to which patients have not been previously exposed. The use of high risk substances and varied routes of administration is more prevalent and these factors occur in the setting of busy and highly pressurised environments.
Medication errors are often cited as being one of the most common sources of error within hospitals2,4,5 but the prevalence of such errors, the most frequently implicated drugs and the effects on patient morbidity and mortality have not been recently systematically reviewed. Previous studies have tended to focus on specific areas of error, e.g. parenteral drug use or prescriptions only,6–8 and investigated only those studies performed in the developed world.9 Initial review of the literature showed that published studies are very heterogeneous in trial methodology, study population, setting and methods of data presentation. The primary data did not lend themselves to a systematic review with meta-analysis. An integrative review was therefore undertaken to critically appraise the recent literature to determine the prevalence of ME in the critical care environment, Department of Anaesthesia & Critical Care, Royal Victoria Infirmary, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK Corresponding author: Caroline MacFie, Department of Anaesthesia & Critical Care, Royal Victoria Infirmary, Newcastle upon Tyne, UK. Email: [email protected]
64
Journal of the Intensive Care Society 17(1)
the drugs most commonly associated with ME and the consequences of ME.
Further, ‘snowballing’ from the reference lists of all relevant full text studies identified from the electronic search was undertaken.
Methods An integrative review is a reviewing process summarising available literature to provide a more comprehensive understanding of a clinical problem. It is more structured and rigorous than a narrative review in situations where the primary data do not support the use of a systematic review and meta-analysis. Integrative reviews present the current understanding of a subject and contribute to theory development. They have gained favour in academic nursing journals and represent a process well suited to many areas of current intensive care medical practice. The integrative review was structured to incorporate the five-step review framework described by Whittemore and Knafl10: 1. 2. 3. 4. 5.
Problem identification Literature search Data evaluation Data extraction, synthesis and analysis Presentation of results
Problem identification This has been described in the introduction. The aim of this integrative review is to answer the following questions: . What is the prevalence of ME in the setting of critical care? . What are the drugs most commonly associated with ME in the setting of critical care? . What are the consequences of ME in the setting of critical care?
Literature search Studies published from 1995 to June 2015 were identified using Medline, CINAHL and Embase databases. The following search strategy was used: MeSH terms: Medication Errors, Medical Errors, Pharmaceutical Preparations, Critical Care, Intensive Care, and Intensive Care Units were used in conjunction with title and abstract searches of ‘‘medication error’’, ‘‘medical error’’, mistake, ‘‘adverse event’’, drug, ‘‘level 2’’, ‘‘level two’’, ‘‘level 3’’, ‘‘level three’’, ‘‘critical care’’, ‘‘intensive care’’, ‘‘high dependency’’, hdu, icu, and ‘‘intensive treatment unit’’. The amalgamation of the three broad categories of error with drug with critical care gave the final results which were then limited to English Language, Humans and Adults. A hand search was also performed of personal files and reference lists of relevant review articles.
Data evaluation Relevant studies were identified by title, abstract and finally full text. Studies were included for review if the following predefined inclusion criteria were met: . Rates of ME reported . ME investigated and reported in a critical care setting . ME rates reported in isolation rather than in conjunction with other forms of medical error. Studies were excluded for review if the following predefined exclusion criteria were present: . Studies outside the setting of critical care or studies combining data from critical care and non-critical care settings . Studies of primarily paediatric populations . Review articles, correspondence, abstracts, dissertations, and editorials . Full text not available in English. Abstracts were reviewed separately by two of the authors, and any discrepancy between the decision on whether to include in the final analysis was resolved by discussion. Complete articles were then analysed for type of study, incidence of medication error and type of error investigated, associated adverse events and most commonly implicated drugs.
Definitions Medication error (ME): any mistake in the prescription, preparation or administration of a drug which does not necessarily cause harm.11 Adverse drug event (ADE): a medication error where harm occurs.11
Data extraction, synthesis and analysis For each study data were extracted to define study characteristics. A qualitative synthesis of the primary data was then undertaken to determine the prevalence and consequences of ME and the most commonly implicated medications.
Presentation of results Data are presented in narrative and tabular form.
Results Our primary literature search produced 572 abstracts which were reviewed to identify 85 studies for
MacFie et al. analysis. Forty-five studies were discarded as they did not meet the inclusion criteria for this study. Forty studies met our inclusion criteria and are summarised in Table 1. The findings of our literature review can be summarised within the following domains: study methodology; units used to report medication error incidence; overall incidence of medication error; severity of medication error; and drugs implicated in medication errors.
65 study as being required after 23.5% of ME.26 Serious harm (3.39 to 64.8% of ADEs)30,38 and fatal consequences (between 0.45 to 25% total ADEs)24,30 of ADEs were also described.
Drugs implicated in medication errors Drugs commonly implicated in this review are summarised in Table 2.
Effect of different interventions Study methodology Study methodology varied between studies. Data were gathered either prospectively or retrospectively or as part of an interventional study to identify medication error incidence prior to and following an intervention. Errors were identified using three main methodologies: chart review, voluntary reporting or direct observation.
Units used to report medication error incidence The units used to report medication errors also varied between studies. The most common units used to report medication error incidence were medication errors per 1000 patient days. Other units used included medication errors per 100 admissions, percentage of patients experiencing one or more medication errors, percentage of medication error as a proportion of overall drug administration or medication errors per 100 medication orders.
Overall incidence of error The incidence of ADE and ME varied widely between studies. The incidence of ADEs ranged widely from 1 to 96.5 per 1000 patient days37,38 or 1.3 to 21.1 when reported as per 100 admissions.25,29 When data were reported in different units, one study reported the incidence of ADEs as 34.1% of patients31 and another as 0.16 per 100 medication orders.24 The incidence of ME ranged from 5.1 to 967 per 1000 patient days,22,44 or when reported as per 100 admissions – 0.88–780.24,26 Percentage of medication error as a proportion of overall drug administration ranged from 1.96 to 69.7%.17,39 One study reported the incidence in units per 100 orders – 10 per 100 orders.24
Severity of medication error Severities of ADEs were described in terms of cost, monitoring implications and patient harm. Extra cost per patient varied between $34716 and $6647.14 Increased monitoring was reported in only one
Eleven of the studies compared pre- and post-intervention data. The types and effects of the various interventions are listed in Table 3.
Discussion The main findings of this integrative review into drug errors in a critical care setting were: forty studies reported the incidence of medication error or adverse drug event in critical care; these errors are common in critical care; there are wide variations in estimates of incidence reflecting diverse methodology of studies; the consequences are often severe; and the drugs implicated are both specific to critical care but also include those which are commonly implicated in noncritical care settings. Forty studies met the inclusion criteria. Not all studies specifically investigated the incidence of medication error as their primary outcome. Eleven of the studies were interventional studies investigating the impact of an intervention on the incidence of medication errors in critical care. Medication errors are relatively common (1 to 96.5 per 1000 patient days) in a critical care setting. Polypharmacy is common along with prescriptions of drugs to which patients have not been previously exposed. Medication reconciliation may not be prioritised and the use of high risk substances and varied routes of administration is more prevalent. All these factors occur in the setting of busy and highly pressurised environments. Perhaps the most interesting finding of this integrative review was the wide variation in nomenclature, quality and methods of data collection which clouds interpretation and comparison of studies, and does not allow for meta-analyses. ME is increasingly accepted to mean any mistake in the prescription, preparation or administration of a drug which does not necessarily cause harm in contrast to ADE that refers to a medication error where harm occurs.11 One of the sentinel studies in this field remains that of Bates et al.2 who emphasised the importance of consistent definition and explains the term ADE as, for example, a patient with first degree heart block inappropriately receiving a beta blocker resulting in complete heart block requiring pacing. This is in contrast to a potential ADE synonymous
Holland
Switzerland
2008
2009
2012
1996
2008
2004
2003
2002
2012
2012
Nuckols et al.14
Benkirane et al.15
Jiang et al.16
Nettleman and Bock17
Fahimi et al.18
Herout and Erstad19
Fraenkel et al.20
van den Bemt et al.21
Pagnamenta et al.22
Agalu et al.23
Ethiopia
Australia
USA
Iran
USA
China
Morocco
USA
USA
1995
USA
2014
Korea
Country
Carayon et al.13
2014
Year
Bates et al.2
Cho et al.
12
Reference
P, O, N&PhS #MDT*
P, VR, I (electronic prescribing, standardised labelling & pumps, pharmacist), N&PS #MDT
P, O, PhS #MDT
P, VR, I (computerised clinical information system), NS #MDT
P, CR – continuous infusions only, PhS #PH
P, O – IV only, PhS #PH
P, CR – missed doses only, NS #MDT
P, CR, I (pharmacist), P&PhS #MDT
P, O, VR, P&PhS #MDT
P, CR - IV only, N&PS #MDT
P, VR, CR, PS #National study group
P, VR, CR, NS #MDT
CR, O, NS #MDT
Designa #Study leads
Table 1. Summary of included studies.
Percentage
Other
56.35
21.9 academic hospital, 17.4 non-academic hospital
11.5
9.2
15.36
11.5
6.1
Pre 506 incidents Post 478
Pre 85 incidents Post 55
105.9
7.7
9.1
276
7.5
4.8
/100 admissions
51.8% administrations
44.6% administrations
9.4% of administrations
1.96% of administrations missed
53.6% prescriptions 19% administrations
Percentage
/1000 patient days
/1000 patient days /100 admissions
Incidence ME/potential ADE
Incidence ADE
Pre 1.68 errors/pt Post 0.46
Other
Antibiotics Analgesics Antiepileptics
GI
Antibiotics GI Steroids
Insulin Local anaesthetics
Opiates Sedatives Anticoagulants
Analgesics Antibiotics Sedatives
Drugs
Cost
(continued)
L-T 18.75% Death 2.5% Serious 51.8%
Cost
L-T 12% Fatal 1% Serious 30%
Severity (of ADEs unless otherwise stated)
2010
2005
2008
2006
2012
2011
2010
2012
2011
2010
Kane-Gill et al.26
Rothschild et al.27
Nuckols et al.28
Kane-Gill et al.29
Vazin and Delfani30
Reis and Cassiani31
Ford et al.32
Benoit et al.33
Seynaeve et al.34
Morimoto et al.35
2011
2013
24
Year
Aljadhey et al.25
Jennane et al.
Reference
Table 1. Continued.
Japan
Belgium
Switzerland
USA
Brazil
Iran
USA
USA
USA
USA
Saudi Arabia
Morocco
Country
P, CR, PS #MDT
P, CR, VT #MDT
P, CR, I (stream-lining medication process), PhS #MDT
P, O, I (education), PhS #PH
P, CR, N&PhS #MDT*
P, O, PhS#PH
R, VR, PhS #PH
R, CR, I (smart pumps) – IV only, N&PS #MDT
P, O, VR, CR, PS&VT #MDT
R, VR, PhS #PH
P, VR, CR, P&PhS #MDT
P, VR, CR, N&P&PhS #MDT
Designa #Study leads Percentage
Other
2.4
2.4
4.86
37.8
15.7
1.3
21.1
12.6
34.1% patients
0.16/100 orders
228
127.8
36.2
967
0.88
38.2
780
/100 admissions
/1000 patient days
/1000 patient days /100 admissions
Incidence ME/potential ADE
Incidence ADE
Pre 4.95% administrations Post 2.14%
CCU Pre 30.8% administrations Post 4% MICU: pre 20.8% Post 22.7%
7.6% administrations
Percentage 10/100 orders
Other
Antibiotics Sedatives NSAIDS
Hypoglycaemics
Opiates Sedatives Potassium
Antibiotics
Antibiotics Analgesics Anticoagulants
Opiates Insulin Sedatives
CVS Anticoagulants Antibiotics
Opiates Antibiotics Anticoagulants
Anticoagulants Antibiotics Antihypertensives
Antibiotics Anticoagulants Steroids
Drugs
(continued)
Serious 4%
Of ADRs: Serious 20%
Of medication errors: Fatal 0.45% Serious 3.39%
Increased monitoring
Serious 35% L-T 6%
Death 25%
Severity (of ADEs unless otherwise stated)
36
Jordan
Iran
2006
2013
2014
2012
2007
2000
1997
2006
2007
2004
Colpaert et al.7
Nguyen et al.8
Nazer et al.38
Vazin and Fereidooni39
Fahimi et al.40
Lustig41
Cullen et al.42
Valentin et al.43
Chacko et al.44
Ridley et al.5
UK
India
European
USA
Israel
Iran
Vietnam
Belgium
Holland
2010
Klopotowska et al.37
European
USA
Country
2009
2009
Year
Valentin et al.6
Deyoung et al.
Reference
Table 1. Continued
P, CR, not stated #MDT
P, VR, PS #Physician
P, VR, not stated #Multinational study group
P, VR, CR, N&P&PhS #MDT
P, CR, PhS #PH
P, O - IV infusions only, PhS #PH
P, O, PhS #PH
P, CR, PhS #MDT
P, O, I (pharmacist led training), N&PhS #PH
P, CR, I (electronic prescribing), P&PhS #MDT
P, I (pharmacist) prescription errors only, P&PhS #MDT
P, VR - IV only, PS #Multinational study group
P, O, I (barcode assisted administration), N&PhS #MDT*
Designa #Study leads Percentage
Other
1.6
96.5
4 pre, 1 post
0
10.6% of errors were ADEs
Of prescriptions: Pre 0.98% Post 0.16%
5.1
105
19
8.2
190.5 pre, 62.5 post
745
/100 admissions
/1000 patient days
/1000 patient days /100 admissions
Incidence ME/potential ADE
Incidence ADE
15% of prescriptions
7.8% of administrations
69.7% of administrations
Of administrations:Pre 64%, Post 48.9%
Of prescriptions: Pre 27% Post 3.4%
Pre 19.7% administrations Post 8.7%
Percentage
Other
Electrolytes Anticoagulants Paracetamol
CVS Hypoglycaemics
Benzodiazepines
CVS CNS
Hypoglycaemics Antibiotics Sedatives
Antibiotics CVS
Antibiotic Antithrombotics, Antiepileptics
Antibiotics Sedatives Vasopressors
GI CVS
Drugs
(continued)
Of medication errors 10.6% led to harm
Serious/ L-T 64.8% Fatal 1.1%
Maximum was temporary harm
Of total patients: 0.9% permanent harm/death
Severity (of ADEs unless otherwise stated)
Of medication errors: Serious 19%
Percentage
P, O, P&PS #National study group 2015 Shulman et al.46
UK
1999 Leape et al.45
USA
P, CR, I (pharmacist), N&P&PhS #MDT
Preventable: pre 10.4, post 3.5 All ADEs: 33 pre, 11.6 post
/100 admissions Designa #Study leads Country Year Reference
P ¼ prospective; CR ¼ chart review; VR ¼ voluntary reporting; O ¼ observational; I ¼ interventional; CCU ¼ Coronary Critical Care Unit; MICU ¼ Medical Intensive Care Unit; L-T ¼ long-term. Italics ¼ how the errors were scored. PS ¼ physician scored; PhS ¼ pharmacist scored; NS ¼ nurse scored; VT ¼ validated tool. #Study leads. #MDT ¼ multidisciplinary team led; #MDT* ¼ multidisciplinary team led not including physician involvement; #PH ¼ pharmacist led. Others as stated.
/1000 patient days /1000 patient days
Other
Incidence ME/potential ADE Incidence ADE
/100 admissions
Percentage
6.8% prescriptions
Other
Drugs
Severity (of ADEs unless otherwise stated)
Table 1. Continued
with ME which refers to, for example, a patient with a documented allergy receiving said drug but not reacting.2 Some studies investigate the so-called preventable ADEs as decided by an ICU pharmacist and physician,37 whilst others define minor, intercepted and serious medication prescription errors (MPE).7 In contrast, medication error was also described as any deviation in drug preparation and administration from the doctors’ prescription, hospital policies or manufacturer’s instructions such as diluted in saline as opposed to water.8 The variations in definition are numerous and are not all listed here. Review of the two extremes of reported ME/potential ADE per 1000 patient days included one study that reported only 5.1 MEs per 1000 patient days despite documenting anything from delayed dose of a drug to accidental intravenous administration of planned intrapleural streptokinase.44 Incident outcome was not separately differentiated between other non-drug-related incidents that occurred. At the other end of the spectrum with the most MEs of 967 per 1000 patient days was a study that included an extensive array of ‘discrepancies’ such as prescription omissions such as ‘‘route’’ not written next to a prescribed drug.20 This may partly explain why their results were disproportionately higher than the other studies analysed. Clearly, the frequency of prescribing errors would be expected to be higher than that of actual error or harm to the patient. The differing definitions and methodology are reflected in the wide range of estimates for the incidence of ME and ADE reported by this review. The variation in reported incidence of ADE from 1 to 96.5 per 1000 patient days may reflect that the former study37 used a prospective interventional study looking at impact of pharmacist intervention into medical prescribing, whereas the latter38 used chart review and incorporated adverse drug reactions (i.e. those that are unavoidable such as anaphylaxis) into their ADE figures. Furthermore, various studies examined a single aspect of drug administration6,14,18,19,28,40 such as parenteral medications only; solely missed doses,17 or a prescriptions-only study design.37 Eleven of the studies analysed incorporated preand post-intervention data (see Table 3). As far as conclusions can be drawn from the array of data, it seems that significant benefit arises from computerised prescribing, education and the welcome involvement of our pharmacist colleagues. The only intervention that appeared to cause increased patient risk was due to the so-called smart pumps, i.e. pumps that alert users when programmed to deliver duplicate infusions or doses out-with defined ranges. The proposed explanation was that the pumps led to greater identification of potential errors before harm resulted.28 Chart review was the most commonly used method of identifying error with 20 of the 40 studies analysed using this method, employing unblinded nurse
70
Journal of the Intensive Care Society 17(1)
investigators, pharmacists and physicians. Fourteen used voluntary reporting methods by nurses and pharmacists which could lead to under-estimation of the incidence of ME. Thirteen used a direct observational method by independent pharmacy students,
Table 2. Drugs implicated in medication errors (for example, 32% is the mean average for the five studies that documented percentage of their total error being related to cardiovascular drugs.). Drug category
Percentage of total study errors
Cardiovascular Gastrointestinal Antimicrobial Hypoglycaemic Analgesic Antithrombotic Sedative Antihypertensive Antiepileptic Steroids Electrolytes Local anaesthesia
32 26 22 22 18 18 16 16 11 10 8 6
Note: Studies documenting ME as well as subgroup ADEs are combined.
pharmacists, nurses or study physicians. Only one study used direct observation in conjunction with reporting in the per 1000 patient day format.27 It might follow that using this method might result in a higher reported incidence of ADEs and MEs and apart from one other study38 this was the case with 37.8 ADEs per 1000 patient days. Even using independent observational reporting the incidence of medication error still varied widely and was reported between 7.6 and 69.7% of administrations: 7.6% referred to potential opportunities for occurrence of error per patient while 69.7% referred to errors as a proportion of doses given.30,39 Severity of consequences resulting from an ADE was concerning, ranging from cost and increased monitoring implications, through serious harm (3.39–64.8% of ADEs) to fatal in between 0.45 and 25% total ADEs. Patients in critical care are particularly vulnerable as a consequence of a reduced physiological reserve, reduced ability to metabolise drugs and alterations in pharmacodynamics. New medications are introduced with potential for interaction and pre-existing medications may be omitted due to other clinical priorities. Drugs commonly implicated in this review were antibiotics, sedatives, analgesics, anticoagulants, hypoglycaemics, antiepileptics, gastrointestinal and cardiovascular medications. These medications are implicated in medication errors in wider patient populations and have attracted interest from patient safety
Table 3. Effect of interventions on ME/ADE. Type of intervention implemented
ME/ADE pre intervention
ME/ADE post intervention
Pharmacist – order reviews, drug monitoring
1.68 MEs per patient
0.46 MEs per patient
(3 studies)
190.5 MEs per 1000 patient days 4 ADEs per 1000 patient days 33 ADEs per 1000 patient days
62.5 MEs per 1000 patient days 1 ADE per 1000 patient days 11.6 ADEs per 1000 patient days
Computerised clinical information system
85 ADEs in 2 years
55 ADEs in 2 years
Electronic prescribing
ME in 27% prescriptions ADE in 0.98% prescriptions
ME in 3.4% prescriptions ADE in 0.16% prescriptions
Electronic prescribing, standardised labelling & pumps, pharmacist input
506 ADEs in study period
478 ADEs in study period
Smart pumps
4.78 ADEs per 1000 patient days
4.95 ADEs per 1000 patient days
Barcoding of drugs
ME in 19.7% administrations
ME in 8.7% administrations
Education programme
ME in 30.8% administrations (CCU) ME in 20.8% administrations (MICU)
ME in 4% administrations (CCU) ME in 22.7% administrations (MICU)
Pharmacist led training
ME in 64% administrations
ME in 48.9% administrations
Streamlining of medication process – standardising documentation
ME in 4.95% administrations
ME in 2.14% administrations
ME ¼ medical error; ADE ¼ adverse drug event; CCU ¼ coronary critical care unit; MICU ¼ medical intensive care unit.
MacFie et al. groups.3 The reasons for this are beyond the scope of this review, but may be due to the frequent use of these medications or may be inherent to them or their administration process. The question remains whether certain drugs are inherently more prone to error, or indeed harm, or whether it is increased usage of particular drugs or their prescription that is more problematic. This review covers recent literature on medication error and adverse drug events in a more diverse setting than previously published. Errors are common with potentially severe consequences, and research into this area requires standardised definitions and ideas regarding data collection to allow us to make more meaningful conclusions. A number of interventions have indicated improvement in service such as computer-aided drug ordering systems, training, standardised practices, and of vital importance the daily presence and input of critical care pharmacists as supported by the recent PROTECTED-UK study.46 Other practices that may warrant investigation into whether they can reduce error include colour coding of drug packaging and labels, bar-coding, standardised storage, double checking prior to administration (widely performed but with little evidence base), techniques to minimise interruptions (e.g. ‘‘Do Not Disturb, Drug Round’’ tabards for nurses – perhaps more relevant in the ward environment), and checklists. We recommend a further large-scale national prospective observational trial using set definitions as per the Institute of Medicine and clarified by Bates2,11 to help us progress our knowledge and hence our safety mechanisms in this fundamental area. By understanding the depth of the problem in the intensive care setting, and realising where the risks are greatest, we can endeavour to improve our practice and future training.
Conclusion Drug errors in critical care are common and have serious deleterious effects on morbidity and mortality. This review incorporated a specific integrative review framework that identified a particularly heterogeneous range of research methods and nomenclature in this area making data interpretation difficult. There is an important differentiation to be made between medication error (incidence 5.1– 967 per 1000 patient days) and adverse drug event (incidence 1–96.5 per 1000 patient days) with significant ramifications for patient outcome and cost. The most commonly implicated drug groups include cardiovascular, gastrointestinal, antimicrobial, hypoglycaemic and analgesic agents. The studies described in this review provide insight into suboptimal areas of patient care in the critical care environment and have implications for the development of specific improvement strategies and future training.
71 Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
References 1. Kohn LT, Corrigan JM, Donaldson MS (eds); for the Committee on Quality of Health Care in America, Institute of Medicine. To err is human: Building a safer health system. Washington, DC: National Academy Press, 2000. 2. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA 1995; 274: 29–34. 3. http://www.england.nhs.uk/wp-content/uploads/2014/ 03/psa-sup-info-med-error.pdf 4. Cullen DJ, Bates DW, Leape LL, et al. Prevention of adverse drug events: a decade of progress in patient safety. J Clin Anesth 2000; 12: 600–614. 5. Ridley SA, Booth SA, Thompson CM, et al. Prescription errors in UK critical care units. Anaesthesia 2004; 59: 1193–1200. 6. Valentin A, Capuzzo M, Guidet B, et al. on behalf of the Research Group on Quality Improvement of the European Society of Intensive care Medicine (ESICM) and the Sentinel Events Evaluation (SEE) Study Investigators. Errors in administration of parenteral drugs in intensive care units: multinational prospective study. BMJ 2009; 338: b814. 7. Colpaert K, Claus B, Somers A, et al. Impact of computerized physician order entry on medication prescription errors in the intensive care unit: A controlled crosssectional trial. Crit Care 2006; 10: R21. 8. Nguyen HT, Pham HT, Vo DK, et al. The effect of a clinical pharmacist-led training programme on intravenous medication errors: A controlled before and after study. BMJ Qual Saf 2013; 0: 1–6. 9. Wilmer A, Louie K, Dodek P, et al. Incidence of medication errors and adverse drug events in the ICU: A systematic review. Qual Saf Health Care 2010; 19: e7. 10. Whittemore R, and Knafl K. The integrative review: Updated methodology. J Adv Nurs 2005; 52: 546–553. 11. Institute of Medicine. Preventing medication errors. Washington, DC: National Academies Press, 2007. 12. Cho I, Park H, Choi Y, et al. Understanding the nature of medication errors in an ICU with a computerized physician order entry system. PLoS ONE 2014; 9. 13. Carayon P, Wetterneck TB, Cartmill R, et al. Characterising the complexity of medication safety using a human factors approach: an observational study in two intensive care units. BMJ Qual Saf 2014; 23: 56–65. 14. Nuckols TK, Paddock SM, Bower AG, et al. Costs of intravenous adverse drug events in academic and nonacademic intensive care units. Med Care 2008; 46: 17–24. 15. Benkirane RR, Abouqal R, Haimeur CC, et al. Incidence of adverse drug events and medication errors in intensive care units: A prospective multicenter study. J Patient Saf 2009; 5: 16–22.
72 16. Jiang SP, Zheng X, Li X, et al. Effectiveness of pharmaceutical care in an intensive care unit from China. A pre- and post-intervention study. Saudi Med J 2012; 33: 756–762. 17. Nettleman MD, and Bock MJ. The epidemiology of missed medication doses in hospitalized patients. Clin Perform Qual Health Care 1996; 4: 148–153. 18. Fahimi F, Ariapanah P, Faizi M, et al. Errors in preparation and administration of intravenous medications in the intensive care unit of a teaching hospital: An observational study. Aust Crit Care 2008; 21: 110–116. 19. Herout PM, and Erstad BL. Medication errors involving continuously infused medications in a surgical intensive care unit. Crit Care Med 2004; 32: 428–432. 20. Fraenkel DJ, Cowie M, and Daley P. Quality benefits of an intensive care clinical information system. Crit Care Med 2003; 31: 120–125. 21. van den Bemt PML, Rijn R, van der Voort PHJ, et al. Frequency and determinants of drug administration errors in the intensive care unit. Crit Care Med 2002; 30: 846–850. 22. Pagnamenta A, Rabito G, Arosio A, et al. Adverse event reporting in adult intensive care units and the impact of a multifaceted intervention on drug-related adverse events. Ann Intens Care 2012; 2. 23. Agalu A, Ayele Y, Bedada W, et al. Medication administration errors in an intensive care unit in Ethiopia. Int Arch Med 2012; 5. 24. Jennane N, Madani N, Oulderrkhis R, et al. Incidence of medication errors in a Moroccan medical intensive care unit. Int Arch Med 2011; 4. 25. Aljadhey H, Mahmoud MA, Mayet A, et al. Incidence of adverse drug events in an academic hospital: a prospective cohort study. Int J Qual Health Care 2013; 25: 648–655. 26. Kane-Gill SL, Kowiatek JG, and Weber RJ. A comparison of voluntarily reported medication errors in intensive care and general care units. Qual Saf Health Care 2010; 19: 55–59. 27. Rothschild JM, Landrigan CP, Cronin JW, et al. The Critical Care Safety Study: The incidence and nature of adverse events and serious medical errors in intensive care. Crit Care Med 2005; 33: 1694–1700. 28. Nuckols TK, Bower AG, Paddock SM, et al. Programmable infusion pumps in ICUs: an analysis of corresponding adverse drug events. J Gen Intern Med 2008; 23: 41–45. 29. Kane-Gill S, Rea RS, Verrico MM, et al. Adverse-drugevent rates for high-cost and high-use drugs in the intensive care unit. Am J Health Syst Pharm 2006; 63: 1876–1881. 30. Vazin A, and Delfani S. Medication errors in an internal intensive care unit of a large teaching hospital: A direct observation study. Acta Medica Iranica 2012; 50: 425–432. 31. Reis AMM, and Cassiani SHDB. Adverse drug events in an intensive care unit of a university hospital. Eur J Clin Pharmacol 2011; 67: 625–632.
Journal of the Intensive Care Society 17(1) 32. Ford DG, Seybert AL, Smithburger PL, et al. Impact of simulation-based learning on medication error rates in critically ill patients. Intens Care Med 2010; 36: 1526–1531. 33. Benoit E, Eckert P, Theytaz C, et al. Streamlining the medication process improves safety in the intensive care unit. Acta Anaesthesiol Scand 2012; 56: 966–975. 34. Seynaeve S, Verbrugghe W, Claes B, et al. Adverse drug events in intensive care units: A cross-sectional study of prevalence and risk factors. Am J Crit Care 2011; 20: 131–140. 35. Morimoto T, Sakuma M, Matsui K, et al. Incidence of adverse drug events and medication errors in Japan: The JADE study. J Gen Intern Med 2011; 26: 148–153. 36. DeYoung JL, Vanderkooi ME, and Barletta JF. Effect of bar-code-assisted medication administration on medication error rates in an adult medical intensive care unit. Am J Health Syst Pharm 2009; 66: 1110–1115. 37. Klopotowska JE, Kuiper R, van Kan HJ, et al. Onward participation of a hospital pharmacist in a Dutch intensive care unit reduces prescribing errors and related patient harm: An intervention study. Crit Care 2010; 14: R174. 38. Nazer LH, Hawari F, and Al-Najjar T. Adverse drug events in critically Ill patients with cancer: Incidence, characteristics, and outcomes. J Pharm Pract 2014; 27: 208–213. 39. Vazin A, and Fereidooni M. Determining frequency of prescription, administration and transcription errors in internal intensive care unit of Shahid Faghihi hospital in Shiraz with direct observation approach. Iran J Pharmaceut Sci 2012; 8: 189–194. 40. Fahimi F, Sistanizad M, Abrishami R, et al. An observational study of errors related to the preparation and administration of medications given by infusion devices in a teaching hospital. Iran J Pharm Res 2007; 6: 295–299. 41. Lustig A. Medication error prevention by pharmacists – An Israeli solution. Pharm World Sci 2000; 22: 21–25. 42. Cullen DJ, Sweitzer BJ, Bates DW, et al. Preventable adverse drug events in hospitalized patients: A comparative study of intensive care and general care units. Crit Care Med 1997; 25: 1289–1297. 43. Valentin A, Capuzzo M, Guidet B, et al. Patient safety in intensive care: results from the multinational Sentinel Events Evaluation (SEE) study. Intens Care Med 2006; 32: 1591–1598. 44. Chacko J, Raju HR, Singh MK, et al. Critical incidents in a multidisciplinary intensive care unit. Anaesth Intens Care 2007; 35: 382–386. 45. Leape LL, Cullen DJ, Clapp MD, et al. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. JAMA 1999; 282: 267–270. 46. Shulman R, McKenzie CA, Landa J, et al. Pharmacist’s review and outcomes: Treatment-enhancing contributions tallied, evaluated, and documented (PROTECTED-UK). J Crit Care 2015; 30: 808–813.
