UvA-DARE (Digital Academic Repository)

Medication safety in pediatric care

Maaskant, Jolanda

Link to publication

Citation for published version (APA):

Maaskant, J. M. (2016). Medication safety in pediatric care

General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),
other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating
your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask
the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,
The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

Download date: 15 May 2017

 Chapter 4

High-alert medications for pediatric patients:

an international modified Delphi study

Jolanda Maaskant
Anne Eskes
Petra van Rijn-Bikker
Diederik Bosman
Wim van Aalderen
Hester Vermeulen

Expert Opinion on Drug Safety 2013;12:805-814.

Chapter 4

ABSTRACT
Introduction
The available knowledge about high-alert medications for children is limited. Because
children are particularly vulnerable to medication errors, a list of high-alert medication
specifically for children would help to develop effective strategies to prevent patient
harm. The objective of this study was to generate an internationally accepted list of
high-alert medications for a pediatric inpatient population from birth to 18 years old.

Methods
We conducted an international modified Delphi study and validated the results with
reports on medication incidents in children based on national data.

Results
The rating panel consisted of 34 experts from 13 countries. In total, 14 medications and
4 medication classes were included with the predefined level of consensus of 75%. The
high-alert medications were: amiodarone, digoxin, dopamine, epinephrine, fentanyl,
gentamycin, heparin, insulin, morphine, norepinephrine, phenytoin, potassium,
propofol and tacrolimus. The high-alert medication classes included in the final list
were: chemotherapeutic drugs, immunosuppressive medications, lipid/total parenteral
nutrition and opioids.

Conclusion
An international group of experts defined 14 medications and 4 medication classes
as high-alert for children. This list might be helpful as a starting point for individual
hospitals to develop their own high-alert list tailored to their unique situation.

58
 High-alert medications

INTRODUCTION
Patient harm as a result of medication errors (MEs) is one of the most common type of
adverse events in hospitalized patients [1]. Although MEs do not always cause patient
harm, studies have shown that 3-10% of such errors result in significant harm or even
contribute to death [1-4]. Children are believed to be at especially high risk of harm due
to MEs; according to estimates, MEs result in harm in children about three times more
often than in adults [5]. Therefore, the safe use of medication in children requires even
more precautions than in adult patients. It is important to note that not all adverse drug
events are the result of MEs, and are thus not preventable, but it is estimated that about
half of them are [2,6,7]. Nevertheless, until now interventions have led to only limited
improvements, and MEs continue to threaten patient safety [8].
Medications are very diverse and have a wide range of risk profiles. Those with a

Chapter 4
heightened risk of causing patient harm are known as high-alert medications; they
have serious consequences for patients when misused [9]. Attention for high-alert
medications is stressed by several leading organizations like the Institute of Healthcare
Improvement (IHI), Joint Commission International (JCI) and the Institute of Safe
Medication Practice (ISMP) [9-11]. A list of high-alert medications for the general patient
population has been developed by the ISMP, based on error reports and expert opinions
[9].
The available knowledge about high-alert medications for children is limited.
Analyses of pediatric medication errors voluntary reported in the USA and Canada
have resulted in two slightly different lists [12,13]. Although studies on MEs sometimes
report the medications involved, they seldom report the ones that must be considered
high-alert. Moreover, studies that reported on high-alert medications in pediatric and
neonatal intensive care units did not result in a uniform list [14,15]. Because children
are particularly vulnerable to MEs, in clinical practice a list of high-alert medications
specifically for children would help to develop and implement more effective strategies
to prevent patient harm.
Therefore, we aimed to generate an internationally accepted list of high-alert
medications for an inpatient pediatric population. The target group included all children
from birth until 18 years old without any underlying disease or syndrome.

59
Chapter 4

METHODS
Study design
Our study consisted of two parts. Firstly, we reviewed the literature to generate a
preliminary list of high-alert medications for children. Secondly, we used the modified
Delphi technique, which is considered to be an effective way to obtain consensus within
a group of experts [16]. High-alert medications are defined as medications that are
considered to cause patient harm when misused due either to a narrow therapeutic
window or to serious adverse events in the past. Harmful and unintended responses
to medications, at normal dosages and proper use, known as adverse drug reactions or
side effects, are excluded from this definition.
The Institutional Review Board of the Academic Medical Center in Amsterdam
reviewed the protocol and judged that it was not subject to medical ethical approval
according to the Dutch Medical Ethics Law. All responses were analyzed and reported
anonymously.

Review of literature reporting high-alert medication

We reviewed the literature to generate a list of possible high-alert medications. For this
purpose we systematically searched the Cochrane Library, MEDLINE and EMBASE to
identify studies published until 1 February 2012. We included all publications that aimed
to describe high-alert medications because they were identified or considered to be
harmful in hospitalized children from birth to 18 years old. No limitations were applied
to study design, publication date or language. We excluded publications investigating
safety profiles of predefined medications. We conducted a purely descriptive review
and therefore did not assess the publications on study quality. The results of this review
were used as the starting point for the Delphi rounds.

Expert group
Prospective members of the international expert group were identified through
literature review, websites of medication safety organizations and recommendations
from known experts in the field. In addition, we applied the following inclusion criteria:
healthcare professionals were qualified as experts if (a) they had at least 5-years post-
qualification experience in pediatrics, (b) were educated to postgraduate level and (c)
were willing to participate. We included pediatricians, neonatologists, anesthetists,
pharmacists and pediatric nurses. To improve the generalizability and usefulness of our
final list, we aimed to create a sample of representative experts, homogeneous in the
field of medication safety issues, but from various geographical regions. We aimed to
include at least 30 experts to assure sufficient expertise and representativeness, even
if attrition occurred [17]. We sent a letter by email to the experts to explain the aim

60
 High-alert medications

of the study and the study design, and to request their participation. After informed
consent, the experts received three web-based questionnaires, one for each Delphi
round. We also collected data on the experts, such as their profession, education, years
of experience and employment or designation. To prevent attrition, we aimed at a quick
turnaround time, used personalized emails, reminded non-responding experts by email
and gave deadlines [18].

Data collection
Data were collected by questionnaire, one for each round. The questionnaires
were pretested in a pilot study. To this end, we sent the web-based questionnaires
to a pediatrician, a neonatologist, a pharmacist and a nurse. We asked them to
comment on the content, clarity and phrasing of the questions as well as the layout
of the questionnaire. The comments were used to improve the questionnaires. The

Chapter 4
participating experts received the link for the online questionnaire by email, using a
commercial available survey tool (SurveyMonkey, Inc. Europe, Sarl, Luxembourg). The
three questionnaires were distributed in May, June and July 2012. The questionnaires
included instructions for completion. The experts were asked to respond within three
weeks. Up to two reminders were sent per round to non-responders. The second and
third questionnaires were send within one week after the previous round was finished.
During all rounds the experts were asked to rate their opinions on a 5-point Likert
scale: “strongly disagree”, “disagree”, “neutral”, “agree” and “strongly agree”. Because no
standard threshold for consensus exists [19], we used the following definition: at least
75% of the experts rated a medication as high-alert with “agree” or “strongly agree”.
Consequently, if at least 75% of the experts rated “disagree” or “strongly disagree”, then
the corresponding medication was excluded from further discussion.

Delphi rounds
The first questionnaire consisted of three parts. The first part posed questions about
the baseline characteristics of the experts. In the second part we presented a list of
possible high-alert medications generated from the literature review. We presented
the list alphabetically and included a brief summary of the relevant literature. The first
question was to rate the medication as high-alert. In addition, we collected information
regarding the patient age groups, routes of administration and the organ systems in
which adverse events might occur in case of misuse. This additional information was
collected only during the first round. In the third part of the questionnaire, the experts
were asked to add any medications or medication classes that they considered to be
high-alert if they were not included on the initial list. The experts were given the option
of supporting their choices with scientific evidence, incident reports or experience.

61
Chapter 4

Medications on which consensus was reached in Delphi round 1, based on the 75%
level of agreement, were not discussed again in Delphi round 2.
Medications on which no consensus was reached, together with the additional
medications suggested in Delphi round 1, were included in the second questionnaire.
Once again, the list of medications was alphabetical. The medications rated in the first
round were listed with the group response results, described as percentages of the
experts’ scores on the 5-point Likert scale. We asked the experts to rate each medication
on which no consensus was reached, as well as the newly added medications.
Medications on which consensus was reached in Delphi round 2 were not discussed
again in Delphi round 3. Also, the medications that had been rated by the experts in the
first two rounds were not presented again in Delphi round 3.
We asked the experts to once again rate the medications that remained from Delphi
round 2. We presented the list in alphabetical order. The experts were asked to rate the
medication as high-alert according the 5-point Likert scale, as in previous rounds.

Validation
We validated our results with reports on medication incidents in children. For this
purpose, we used data from the Dutch incident registration (Foundation Portal for
Patient Safety/CMR). CMR is a nonprofit organization devoted entirely to the nationwide
registration and prevention of adverse events in healthcare, using a voluntary incident
reporting system. These incident reports are used to learn about errors, understand their
causes, and share this information with healthcare professionals across the country. ME
prevention and safe medication use is a main focus of the organization.

Analysis
Descriptive statistics were used to report our results. We expressed the level of agreement
in percentages of same answers to each question. Medications rated as “strongly agree”
or “agree” by at least 75% of the experts were included in the final list. We excluded
medications that were rated as “strongly disagree” or “disagree” by 75% of the experts. To
analyze the change in answers towards consensus, we described the dispersion of the
results in standard deviations (SD) and ranges. All analyses were performed using SPSS
software (PASW statistics version 18.0, IBM, Armonk, NY, USA).

62
 High-alert medications

RESULTS
Review of literature reporting high-alert medication
The literature search yielded 76 titles. After independently screening on titles and
abstracts, we excluded 63 publications because they did not report on high-alert
medications, did not involve a pediatric population and/or were not performed in a
hospital setting. Consequently, 13 publications were selected for further study [14,15,20-
30]. A manual search of the references in these articles yielded two additional possibilities
[12,31]. After studying the full text of the articles, we excluded 11 publications [20-31],
leaving 4 publications that met our inclusion criteria [12-15]. Patient and medication
characteristics were summarized, resulting in an initial list of 16 high-alert medications
in children. This initial list, with a short summary of the literature, was used in the Delphi
process.

Chapter 4
Expert group
We invited 92 experts to participate, of whom 34 gave informed consent. Experts who
replied that they did not want to participate, or who did not respond to our invitation,
were removed from the mailing list. As summarized in Table 1, our expert group from
13 countries included pediatricians, a neonatologist, an intensivist, pharmacists and
pediatric nurses. Regarding postgraduate experience, 62% the experts reported more
than 15 years. Of the total group of 34 experts, 30 responded (88%) in the first round,
27 responded (79%) in the second round and 30 responded (88%) in the third round. In
total, 25 experts completed all three questionnaires (74%). One expert did not respond
at all (2%).

Delphi rounds
In the first round, we presented 16 medications to the experts. Consensus was achieved
on 8 medications, which were then included in the final list: digoxin, dopamine,
epinephrine, fentanyl, heparin, insulin, morphine and potassium. No medications were
excluded. The experts added 23 high-alert medications and 7 high-alert medication
classes.
In the second questionnaire we included the 8 medications remaining from the first
round for further consensus discussion, along with the 23 additional medications and 7
medication classes suggested in the first round. Consensus was achieved on 2 of these
medications (gentamycin and propofol) and 2 medication classes (chemotherapeutic
drugs and opioids), which were then included on the final list. Once again, no
medications were excluded.

63
Chapter 4

Table 1. Baseline characteristics of the experts

total round 1 round 2 round 3

N % N % N % N %
Number of experts 34 100 30 88 27 79 30 88
Country
Belgium 2 6 2 7 2 7 2 7
Canada 1 3 1 3 1 4 1 3
Estonia 1 3 1 3 - - 1 3
Germany 1 3 1 3 1 4 1 3
Ireland 1 3 1 3 1 4 1 3
Italy 1 3 1 3 - 1 3
Latvia 1 3 1 3 1 4 1 3
Luxembourg 1 3 1 3 1 4 1 3
the Netherlands 7 21 5 17 5 18 5 17
Norway 1 3 1 3 1 4 1 3
Slovenia 1 3 1 3 - - - -
United Kingdom 10 29 10 33 9 33 10 33
United States of America 6 18 4 14 5 18 5 17
Profession
Pediatrician 12 35 10 33 8 29 10 33
Neonatologist 8 24 8 27 6 22 8 27
Intensivist 1 3 1 3 1 3 1 3
Pharmacist 9 24 9 30 9 33 7 23
Nurse 4 15 2 7 3 11 4 14
Postgraduate experiencea
5 - 10 years 1 3 1 3 1 4 1 3
10 - 15 years 10 29 10 33 10 37 10 33
>15 years 21 62 19 64 16 59 19 64
Highest level of education
Bachelor’s degree 1 3 1 3 1 4 1 3
Master’s degree 6 18 3 10 4 15 2 7
Post-master’s degree (PhD) 12 35 11 37 9 33 12 40
Professor 11 32 11 37 9 33 11 37
Otherb 4 12 4 13 4 15 4 13
Organization
Academic Hospital 19 56 18 60 16 59 16 53
Teaching Hospital 5 15 4 13 4 15 5 17
University 4 12 3 10 2 7 3 10
Otherc 6 18 5 17 5 19 6 20
a
Information of two experts missing.
b
Doctor of Medicine (MD), Doctor of Pharmacy, post-graduate degree Clinical Pharmacology.
c
Non-teaching hospital, Nonprofit research foundation, Research consultancy, (National) regulatory agency.

64
 High-alert medications

In the third questionnaire we included the 22 medications and 5 medication classes

that remained from the first two rounds. In this final round, consensus was achieved
on 4 additional medications (amiodarone, norepinephrine, phenytoin, tacrolimus)
and 2 medication classes (immunosuppressive medications and lipid/total parenteral

nutrition).Figure 1. Study flow
The results are summarized in Figure 1.

Records identified through database Additional records identified through

literature review

search other sources

(N = 76 ) (N = 2)

Publications included in the review

(N = 4)

Chapter 4
Delphi round 1

Response rate 88%

consensus on 8 medications

31 medications and 7
medication classes included
in the second questionnaire
( )
Delphi round 2

Response rate 79%

consensus on 2 medications
and 2 medication classes

22 medications and 5
medication classes included
in the third questionnaire
Delphi round 3

( )

Response rate 88%

consensus on 4 medications
and 2 medication classes

Figure 1. Study flow

Final List
The experts reached consensus on 14 medications and 4 medication classes. The final
list of medications included amiodarone, digoxin, dopamine, epinephrine, fentanyl,
gentamycin, heparin, insulin, morphine, norepinephrine, phenytoin, potassium,
propofol and tacrolimus. The final list of medication classes included chemotherapeutic
drugs, immunosuppressive medications, lipid/total parenteral nutrition and opioids.

65
Chapter 4

The consensus level ranged between 75% and 100%. Based on the predefined threshold
of 75%, we excluded 25 medications and 3 medication classes from the final list. The
results are summarized in Tables 2 and 3.

Table 2. High-alert medications, consensus ≥75%

Medications
Gentamycina 100% Dopaminea 83%
Digoxin 93% Tacrolimus 83%
Norepinephrine 93% Phenytoin a
83%
Potassiuma 89% Insulina 79%
Amiodarone 87% Morphine a
79%
Epinephrinea 86% Heparina 75%
Propofol 84% Fentanyl a
75%
Medication classes
Chemotherapeutic drugsa 84% Lipid/total parenteral nutritiona 79%
Immunosuppressive drugs a
86% Opioids a
76%
a
Considered high-alert in the Dutch medication incident reports system.

Table 3. High-alert medications, consensus <75%

Medications
Calcium 74% Domperidone 30%
Epoprostenol 73% Methylphenidate 30%
Theophylline 73% Salbutamol 30%
Vancomycin 70% Acetylsalicylic acid 20%
Phenobarbital 70% Dexamethasone 20%
Dinoprostone/alprostadil 63% Ibuprofen 20%
Midazolam 63% Iron dextran 20%
Paracetamol /acetaminophen 43% Linezolid 17%
Propanolol 41% Ceftriaxon 11%
Risperidone 40% Fytomenadion 10%
Escitalopram 37% Montelukast 7%
Sodium bicarbonate 37% Spirolactone 7%
Dextrose > 20% 33%
Medication classes
Anti-epileptic drugs 66% Anti-hypertensive drugs 28%
ADHD medications 35%

66
 Table 4. Systems in which dysfunction might occura

Blood Cardiovasc. Digestive Endocr. Immune Metab. Nervous Respirat. Reprod. Urinary
Medications
Amiodarone - 84% - 37% - - - - - -
Digoxin - 100% 25% - - - - - - -
Dopamine - 93% - - - - - - - -
Epinephrine - 92% - - - - 27% - - -
Fentanyl - 48% - - - - 63% 82% - -
Gentamycin - - - - - - 39% - - 87%
Heparin 96% - - - - - - - - -
Insulin - - - 93% - - 26% - - -
Morphine - 42% - - - - 69% 81% - -
Norepinephrine - 95% - - - - - - - -
Phenytoin - 27% - - - - 67% - - -
Potassium - 96% - - - - - - - -
Propofol - 77% - - - 29% 35% 35% - -
Tacrolimus - 33% - - 73% 27% - - - 40%
Medication classes
Chemotherapeutic drugs 80% 60% 40% 33% 73% 47% 60% 60% 60% 33%
Immunosuppressive drugs 33% - - - 75% 33% - - 33% -
Lipid/total par. nutrition 36% 29% 36% - 36% 93% - - - -
Opioids - 35% 29% - - - 59% 77% - -
a
Percentage of experts that consider the medication (group) high-alert, because misuse might result in dysfunction of the mentioned systems.
Only scores ≥ 25% are presented.

67
High-alert medications

Chapter 4
Chapter 4

Additional information
The experts considered the listed medications to high-alert for all routes of administration,
but considered intravenous administration as an increased risk for patients. Regarding
the age of the patients, most experts considered the included medications to be high-
alert for all ages. Neonates and infants were believed to be at higher risk than older
children. The experts also indicated the organ systems which are most affected in case
of misuse. These results are summarized in Table 4.
We analyzed the process of reaching consensus by comparing the dispersion of the
results between the first and second experts’ opinions. The mean SD changed from 0.98
to 0.75 and the mean range changed from 3.29 to 2.89. This indicates that during the
Delphi process the variation in the opinions decreased and the consensus increased.

Validation
We analyzed all medication incident reports for children from 43 hospitals in the
Netherlands, submitted to the CMR in the period April 2010 – September 2012. In total
1064 reports were analyzed regarding medication and patient harm related to the
incidents. Of the 18 medications and medication classes on the high-alert list, 4 (22%)
were confirmed with incidents that resulted in serious temporary harm or worse, while
9 (50%) were confirmed with incidents that were considered potentially harmful , i.e.
they could have resulted in serious temporary harm or worse.

DISCUSSION
In this Delphi study, experts from 13 countries generated a list of high-alert medications
for a pediatric inpatient population. Such medications are considered to cause patient
harm when misused due either to a narrow therapeutic window or to serious adverse
events in the past.
Our finding that epinephrine [4,9,14], norepinephrine [9] and dopamine [9,12,14,15]
should be considered as high-alert medications for children is consistent with previous
publications. We also found a high level of consensus on the medication class opioids
[9,32,33], and more specifically for the medications morphine [4,32] and fentanyl [12-
15]. In addition, the medications digoxin [4,9,14,15,34], heparin [4,9,12,14,15,35],
insulin [9,12-15] and potassium [9,12-14,36] have been reported as high-alert in
several publications. Chemotherapeutic drugs [9], total parenteral nutrition [4,9,37],
amiodarone [9] and propofol [9] were also identified previously as high-alert, and fatal
incidents were reported on tacrolimus and phenytoin [4]. However, we did not find any
publications that described gentamycin and the medication class immunosuppressive
drugs as high-alert. Despite the fact that salbutamol [12,13], ceftriaxone [12] and

68
 High-alert medications

propanolol [15] have been reported as high-alert medications for children, we reached
only limited consensus on these medications in our study.
We compared the medications defined as high-alert in our study to the list of high-
alert medications of the ISMP. Out of the 14 medications on our list, 11 are also on the
ISMP list. Moreover, 3 of the high-alert medication classes on our list are similar to the
medication classes on the ISMP list [9].
The limited consensus on several medications might be explained by the international
character of the expert group, reflecting different medication practices in their countries
of origin. Therefore, it is important to realize that many factors contribute to the risk profile
of a medication, such as the toxic features, the availability of more than one preparation
or dose strength of the same medication, the availability of preparations for different
routes of administration of a same medication, dosages that require conversion of unit,
calculations or serial dilutions. The high-alert classification is also dependent on dosage,

Chapter 4
the familiarity of the healthcare professionals with the medication and the possibility
of careful monitoring. These situations might differ in various countries, resulting in
different risk profiles for the same medication. Consequently, despite the absence of full
consensus, it is still worthwhile to evaluate the risk profiles of the excluded medications
on a national and institutional level.
The list, resulting from this study, should be used only as a starting point as individual
hospitals develop their own high-alert list. Therefore, we stress a careful local review of
how the list relates to specific risks for a hospital. Error reports, literature on medication
safety and national incidents registrations (if available) should be used to review the
hospital list on a regular base [11].

Delphi process
We performed an international modified Delphi study to reach consensus on a subject
on which empirical evidence is difficult to collect. In this situation, the Delphi technique
is a validated and accepted methodology, because it gives equal weight to the opinion
of each expert and avoids domination by one expert in the consensus process [19]. In
addition, the Delphi method allows anonymous participation of experts across various
countries and with different backgrounds [19].
Although our study consisted of three rounds, medications and medication classes
were only presented to the experts twice. To explore the change in opinions towards
consensus, we followed the recommendations of Holey et al. and determined the
standard deviations and ranges that express the group agreement in each round [38].
The decrease in standard deviations and ranges between the first and second expert
ratings indicates a trend towards consensus. Nevertheless, another round might have
increased the consensus even more.

69
Chapter 4

Low response rates in research using questionnaires are a recognized problem,

and response rates lower than 70% should be investigated for potential non-response
bias [39]. Fortunately, we reached high response rates in every round and therefore
consider our results to be robust. Possible reasons for the high response rate may be
the explicit consent for participation, resulting in a panel of experts who recognized the
importance of the topic and considered themselves to be partners in the study. Also, the
quick turnaround time, the clear timeframe and the personalized reminders might have
kept the experts interested.
Because no standard threshold for consensus was available [19], we based the
75% threshold in this study on face validity of the results of the first round. Our final
results show that this threshold is disputable, with consensus levels just under or just
above the predefined threshold. Therefore, we presented all medications with this level
of consensus, allowing healthcare professionals to consider the list in their specific
situations.

Experts
The heterogeneity of the experts in terms of profession and country of origin made it
potentially more difficult to achieve consensus. However, the selection was based on
what we judged appropriate for an adequate, international representation, which we
believe strengthens our findings. Although we tried to have all continents represented,
the experts in the final group predominantly came from the USA or European countries.
Most participants originated from the United Kingdom (29%), the Netherlands (21%) or
the USA (18%). Experts from Asia, South America, Africa and Oceania were absent in our
study. This could limit the generalizability of our results.

Validation
No standard exists to establish the validity of the Delphi method, but the assumed
validity is part of the Delphi process itself, due to successive rounds (concurrent validity)
and by achieving consensus within a group of experts (content validity) [40,41]. In
addition, we increased the credibility of our results by comparing them with national
medication incidents in children that resulted in patient harm or were reported as
potential harmful.

Applications and suggestions for future research

Creating a list of high-alert medication alone does not prevent patient harm. However,
knowing which medications pose serious risks to children allows healthcare organizations
to develop and implement risk-reduction interventions. These interventions should
focus on additional safeguards throughout the entire medication process for high-alert

70
 High-alert medications

medications and pay attention to the specific safety threats for children.
Several studies have demonstrated a reduction in MEs after the introduction of a
computerized physician order entry system (CPOE) [42-44] and computerized clinical
decision support [42,45]. An additional strategy to reduce prescribing errors could be to
limit access to high-alert medication to special trained and experiences staff, a measure
that can be implemented by special authorizations in CPOE. Furthermore the prescribing
stage of high-alert medication should be supported by standardized prescriptions,
a weight-based dosage calculator and preventing overdosage by implementing
maximum dosing. In the administering stage risk reduction might be achieved by
barcode technology [46,47], that also should be limited to special authorized staff in
case of high-alert medication. Additional risk-reduction measures in the administering
stage are described as “fail safes”, for example smart pumps or different syringes for
oral and intravenous tubes [48]. The involvement of clinical pharmacists has proven to

Chapter 4
increase medication safety [49,50]. They can play an essential role in the medication
process of high-alert medication by performing the preparation of high-alert medication
and providing staff with timely consultation. Clinical pharmacist also should initiate
standardization and monitoring high-alert medications. Double checking is assumed to
reduce MEs and is nowadays standard nursing policy [51]. Indeed, nursing staff believe
that double checking, if done properly, increases medication safety. Unfortunately,
practical problems hamper the double checking process, for example interruptions,
staff shortages or an emergency situation, and some nurses prefer double checking
limited to high-risk patients and high-alert medications [51,52]. Prevention strategies
such as the use of a unit dose dispensing system, dedicated nurses and the involvement
of parents might contribute to medication safety, but robust evidence is limited [53-55].
Obviously, a fair number of risk-reduction strategies are described, although some
are considered more effective than others [48]. For example, maximizing access to
information and limited access to high-alert medication seem to be better safeguards
than education and the use of reminders [48]. A combination of different strategies is
recommended to reach an optimal effect, include all stages of the medication process
and reach all healthcare professionals involved [10,48]. In addition, interventions like
executive walk rounds and communication tools, aiming to create a positive patient
safety culture should be considered, as this has been reported to be associated with
enhanced patient safety [56,57]. Successful implementation of ISMP recommended
high-alert interventions into clinical practice have been reported [26,52]. The next step
will be to underpin these interventions and their implementation with evidence, based
on sound methodology [58].
Our list of high-alert medications should be viewed as a dynamic list that needs
regular updates. New medications, new formularies and growing knowledge on

71
Chapter 4

medication safety for children might have implications for the list presented in our
study. Sharing knowledge of harmful incidents, for example via national incidents
registrations, should be used as a valuable supplement to scientific research.

CONCLUSION
In a three-round Delphi study, consensus was reached on 14 medications and 4
medication classes that are considered high-alert for children. Medications and
medication classes on which consensus was less than our threshold of 75% still might
be handled as high-alert, depending on local situations. Our results might be helpful as
a starting point as individual hospitals develop their own high-alert list.

Acknowledgements
We would like to thank the experts for taking part in the Delphi procedures (Appendix).
We also thank Elvira Zwart for helping us to identify experts and Arianne van Rhijn and
David Opstelten for their support with the validation.

Funding
Neither this research project nor the authors received any grant from any agency in the
public, commercial or not-for-profit sector.

72
 High-alert medications

REFERENCES
1. de Vries EN, Ramrattan MA, Smorenburg SM, et al. The incidence and nature of in-hospital adverse events:
a systematic review. Qual Saf Health Care 2008;17:216-223.
2. Zegers M, de Bruijne MC, Wagner C, et al. Adverse events and potentially preventable deaths in Dutch
hospitals: results of a retrospective patient record review study. Qual Saf Health Care 2009;18:297-302.
3. Kale A, Keohane CA, Maviglia S, et al. Adverse drug events caused by serious medication administration
errors. BMJ Qual Saf 2012;21:933-938.
4. Cousins D, Clarkson A, Conroy S, et al. Medication errors in children - an eight year review using press
reports. Paed Perinat Drug Ther 2002;5:52-58.
5. Kaushal R, Bates DW, Landrigan C, et al. Medication errors and adverse drug events in pediatric inpatients.
JAMA 2001;285:2114-2120.
6. von Laue NC, Schwappach DL, Koeck CM. The epidemiology of preventable adverse drug events: a review
of the literature. Wien Klin Wochenschr 2003;115:407-415.
7. Kunac DL, Kennedy J, Austin N, et al. Incidence, preventability, and impact of Adverse Drug Events (ADEs)
and potential ADEs in hospitalized children in New Zealand: a prospective observational cohort study.

Chapter 4
Paediatr Drugs 2009;11:153-160.
8. Landrigan CP, Parry GJ, Bones CB, et al. Temporal trends in rates of patient harm resulting from medical
care. N Engl J Med 2010;363:2124-2134.
9. Institute for Safe Medication Practices. High Alert Medications. Available: www.ismp.org/tools/
institutionalhighAlert.asp, accessed June 2013.
10. Federico F. Preventing harm from high-alert medications. Jt Comm J Qual Patient Saf 2007;33:537-542.
11. Joint Commission International Accreditation Standards for Hospitals (including standards for academic
medical center hospitals), 4the edition, Joint Commission Resources, Oakbrook Terrace, Illinois 60181,
United States of America, 2012.
12. Santell JP, Hicks R. Medication errors involving pediatric patients. Jt Comm J Qual Patient Saf 2005;31:348-
353.
13. Colquhoun M, Orrbine E, Sheppard I, et al. National collaborative: Top five drugs reported as causing harm
through medication error in pediatrics. Dynamics 2009;20:20-22.
14. Franke HA, Woods DM, Holl JL. High-alert medications in the pediatric intensive care unit. Pediatr Crit Care
Med 2009;10:85-90.
15. Stavroudis TA, Shore AD, Morlock L, et al. NICU medication errors: identifying a risk profile for medication
errors in the neonatal intensive care unit. J Perinatol 2010;30:459-468.
16. Jones J, Hunter D. Consensus methods for medical and health services research. BMJ 1995;311:376-380.
17. Akins RB, Tolson H, Cole BR. Stability of response characteristics of a Delphi panel: application of bootstrap
data expansion. BMC Med Res Methodol 2005;5:37-49.
18. Edwards PJ, Roberts I, Clarke MJ, et al. Methods to increase response to postal and electronic questionnaires.
Cochrane Database Syst Rev 2009, Issue 3. Art. No.: MR000008. DOI: 10.1002/14651858.MR000008.pub4.
19. Keeney S, Hasson F, McKenna H. Consulting the oracle: ten lessons from using the Delphi technique in
nursing research. J Adv Nurs 2006;53:205-212.
20. Alexander DC, Bundy DG, Shore AD, et al. Cardiovascular medication errors in children. Pediatrics
2009;124:324-332.
21. Blix HS, Viktil KK, Reikvam A, et al. The majority of hospitalised patients have drug-related problems:
results from a prospective study in general hospitals. Eur J Clin Pharmacol 2004;60:651-658.
22. Brown M. Medication safety issues in the emergency department. Crit Care Nurs Clin North Am 2005;17:65-69.
23. Ferranti J, Horvath MM, Cozart H, et al. Reevaluating the safety profile of pediatrics: a comparison of
computerized adverse drug event surveillance and voluntary reporting in the pediatric environment.
Pediatrics 2008;121:e1201-e1207.

73
Chapter 4

24. Rinke ML, Shore AD, Morlock L, et al. Characteristics of pediatric chemotherapy medication errors in a
national error reporting database. Cancer 2007;110:186-195.
25. Trinkle R, Wu JK. Errors involving pediatric patients receiving chemotherapy: a literature review. Med
Pediatr Oncol 1996;26:344-351.
26. Tham E, Calmes HM, Poppy A, et al. Sustaining and spreading the reduction of adverse drug events in a
multicenter collaborative. Pediatrics 2011;128:e438-e445.
27. Juntti-Patinen L, Neuvonen PJ. Drug-related deaths in a university central hospital. Eur J Clin Pharmacol
2002;58:479-482.
28. Silva MD, Rosa MB, Franklin BD, et al. Concomitant prescribing and dispensing errors at a Brazilian hospital:
a descriptive study. Clinics (Sao Paulo) 2011;66:1691-1697.
29. Billman G. Evidence-based improvements prevent high-risk errors. Harm isn’t random. J Nurs Adm
2004;34:7-8 (suppl).
30. Johnston PE, France DJ, Byrne DW, et al. Assessment of adverse drug events among patients in a tertiary
care medical center. Am J Health Syst Pharm 2006;63:2218-2227.
31. Bar-Oz B, Levichek Z, Koren G. Medications that can be fatal for a toddler with one tablet or teaspoonful:
a 2004 update. Paediatr Drugs 2004;6:123-126.
32. Doherty C, McDonnell C. Tenfold medication errors: 5 years’ experience at a university-affiliated pediatric
hospital. Pediatrics 2012;129:916-924.
33. Morton NS, Errera A. APA national audit of pediatric opioid infusions. Paediatr Anaesth 2010;20:119-125.
34. Afroukh N, Vasselon-Raina M, Hida H. Medication errors: analysis of a digoxin overdose case in neonatology
unit. Archives de Pediatrie 2011;18:1076-1080.
35. Binder L, Wilson DF. Heparin overdose scare in 14 babies at Texas hospital. Healthcare Benchmarks Qual
Improv 2008;15:89-90.
36. Grissinger M. Potassium chloride injection still poses threats to patients. Pharmacy and Therapeutics
2011;36:241, 302.
37. Hicks RW, Becker SC, Chuo J. A summary of NICU fat emulsion medication errors and nursing services:
data from MEDMARX. Adv Neonatal Care 2007;7:299-308.
38. Holey EA, Feeley JL, Dixon J, et al. An exploration of the use of simple statistics to measure consensus and
stability in Delphi studies. BMC Med Res Methodol 2007;7:52-62.
39. Bose J. Non response bias analyses at the National Center for Education Statistics. National Center for
Education Statistics, Washington DC, Unites States of America, 2001.
40. Hasson F, Keeney S, McKenna H. Research guidelines for the Delphi survey technique. J Adv Nurs
2000;32:1008-1015.
41. Streiner DL, Norman GR. Health measurement scales: a practical guide to their development and use.
Oxford: Oxford University Press, United Kingdom, 2002.
42. Kaushal R, Shojania KG, Bates DW, et al. Effects of computerized physician order entry and clinical decision
support systems on medication safety: a systematic review. Arch Intern Med 2003;163:1409-1416.
43. Eslami S, de Keizer NF, Abu-Hanna A. The impact of computerized physician medication order entry in
hospitalized patients: a systematic review. Int J Med Inform 2008;77:365-376.
44. Walsh KE, Landrigan CP, Adams WG, et al. Effect of computer order entry on prevention of serious
medication errors in hospitalized children. Pediatrics 2008;121:e421-e427.
45. Durieux P, Trinquart L, Colombet I, et al. Computerized advice on drug dosage to improve prescribing
practice. Cochrane Database Syst Rev 2008, Issue 3. Art. No.: CD002894. DOI: 10.1002/14651858.CD002894.
pub2
46. Poon EG, Keohane CA, Yoon CS, et al. Effect of bar-code technology on the safety of medication
administration. N Engl J Med 2010;362:1698-1707.

74
 High-alert medications

47. Morriss FH, Abramowitz PW, Nelson SP, et al. Effectiveness of a barcode medication administration system
in reducing preventable adverse drug events in a neonatal intensive care unit: a prospective cohort study.
J Pediatr 2009;154:363-368.
48. ISMP. Your high-alert medication list, relatively useless without associated risk reduction strategies.
Available: www.ismp.org/newletters/acutecare, accessed June 2013.
49. Kaushal R, Bates DW, Abramson EL, et al. Unit-based clinical pharmacists’ prevention of serious medication
errors in pediatric inpatients. Am J Health Syst Pharm 2008;65:1254-1260.
50. Zhang C, Zhang L, Huang L, et al. Clinical pharmacists on medical care of pediatric inpatients: a single-
center randomized controlled trial. PLoS ONE [Electronic Resource] 2001;7:e30856.
51. Alsulami Z, Conroy S, Choonara I, et al. Double checking the administration of medicines: what is the
evidence? A systematic review. Arch Dis Child 2012;97:833-837.
52. Graham S, Clopp MP, Kostek N, et al. Implementation of a high-alert medication program. Perm J
2008;12:15-22.
53. Miller MR, Robinson KA, Lubomski LH, et al. Medication errors in paediatric care: a systematic review of
epidemiology and an evaluation of evidence supporting reduction strategy recommendations. Qual Saf
Health Care 2007;16:116-126.

Chapter 4
54. Chedoe I, Molendijk HA, Dittrich ST, et al. Incidence and nature of medication errors in neonatal intensive
care with strategies to improve safety: a review of the current literature. Drug Saf 2007;30:503-513.
55. Conroy S, Sweis D, Planner C, et al. Interventions to reduce dosing errors in children: a systematic review
of the literature. Drug Saf 2007;30:1111-1125.
56. Morello RT, Lowthian JA, Barker AL, et al. Strategies for improving patient safety culture in hospitals: a
systematic review. BMJ Qual Saf 2013;22:11-18.
57. Weaver SJ, Lubomski LH, Wilson RF, et al. Promoting a culture of safety as a patient safety strategy: a
systematic review. Ann of Intern Med 2013;158:369-374 (suppl).
58. Fan E, Laupacis A, Pronovost PJ, et al. How to use an article about quality improvement. JAMA
2010;304:2279-2287.

75
Chapter 4

APPENDIX. EXPERTS PARTICIPATING IN THE DELPHI ROUNDS

K Allegaert (University Hospitals Leuven), JN van den Anker (George Washington
University School of Medicine and Health Sciences, Johns Hopkins University School of
Medicine, Erasmus Medical Center-Sophia Children’s Hospital), D Apele-Feimane (State
Agency of Medicines of Latvia, European Medicine Agency), CE de Beaufort (Clinique
Pediatrique Luxembourg), l ten Berg-Lammers (Academic Medical Center Amsterdam),
SM Chapman (Great Ormond Street Hospital for Children NHS Foundation Trust), AC
Ceci (University of Bari), K Connolly (European Medicine Agency), MJ Doman (Plymouth
University), FK Engels (Erasmus Medical Center-Sophia Children’s Hospital), F Federico
(Institute for Healthcare Improvement), HA Franke (University of Arizona, University
of Arizona Healthcare Network, Tucson Medical Center), M Ghaleb (University of
Hertfordshire), JB van Goudoever (Emma Children’s Hospital - Academic Medical Center
Amsterdam, VU Medical Center), KM Gura (Boston Children’s Hospital, Massachusetts
College of Pharmacy & Health Sciences), M de Hoog (Erasmus Medical Center-Sophia
Children’s Hospital), SM Jarvis (Bristol Royal Hospital for Children), J Jazbec (Medical
Centre Ljubljana), MN Lub- de Hooge (University Medical Center Groningen), I Lutsar
(University of Tartu), E la Mache (European Medicine Agency), J Mante (Great Ormond
Street Hospital for Children), MR Miller (Johns Hopkins University), A Neubert (University
Hospital Erlangen), K Norga (Antwerp University Hospital), AJ Nunn (Liverpool John
Moores University) , M Offringa (Child Health Evaluative Sciences, Research Institute, The
Hospital for Sick Children), H Rabe (Brighton & Sussex Medical School, Brighton & Sussex
University Hospital), BA Semmekrot (Canisius-Wilhelmina Hospital Nijmegen), A Soe
(Medway Maritime Hospital Kent) , S Thayyil (University College Hospital London), TW
de Vries (Medical Centre Leeuwarden), SW Wang (Statens Legemiddelverk, Norwegian
Medicines Agency).

76