Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Fredrik Brounéus (00:16):
Welcome to
Drug Safety Matters, a podcast
by Uppsala Monitoring Centre,where we explore current issues
in pharmacovigilance and patientsafety.
Today, we'll be talking aboutmedication errors.
What are they, how big is theproblem, why do they occur and,
perhaps most importantly, howcan we prevent them from
occurring?
This and much more is what DrugSafety Matters host Federica
(00:40):
Santoro will be discussing withguests Ghita Benabdallah and
Loubna Alj from the NationalPharmacovigilance Centre of
Morocco, along with UppsalaMonitoring Centre's own, Alem
Zekarias.
My name is Fredrik Brouneus andyou'll be hearing more from me
as I'll be filling in forFederica during the upcoming
months.
But now, over to Federica andher guests for an in-depth
(01:10):
discussion on medication errors.
Federica Santoro (01:16):
Hi all, and
welcome to Drug Safety Matters.
I'm delighted to have you allon the show to talk about
medication errors today and,more broadly, about how to
prevent adverse drug reactions.
So I think we start with adefinition, just to get everyone
on the same page.
And, Loubna, I would start withyou.
What is a medication error?
Loubna Alj (01:36):
So there are several
definitions of medication
errors, but I will state theEuropean Medicine Agency's one.
Medication errors is defined asan unintended failure in the
drug treatment process thatleads to, or has the potential
to lead to, harm to the patient.
This definition has theadvantage to point out two
(01:58):
important characteristics ofmedication errors.
First of all, the fact thatmedication errors are unintended
.
This term is key to highlightthe non-intentional aspect of
medication errors, in contrastto other drug-related problems
that are intentional, such asmisuse, abuse or off-label use.
(02:21):
The second characteristic isthat not all medication errors
lead to harm.
Some of them are not associatedwith adverse reactions.
Others are potential, meaningthat there are circumstances
that could lead to theirhappening.
Obviously, all medicationerrors should be reported to
(02:44):
pharmacovigilance centers.
Federica Santoro (02:48):
So not all
medication errors automatically
lead to harm, but obviously wewill try to focus on those that
do harm patients, because theyhave an impact and we want to
prevent those especially.
It's not always easy topinpoint exactly why people make
mistakes, but has anyone lookedinto this when it comes to
(03:09):
medication errors?
So what are their most commoncauses?
Loubna Alj (03:15):
Studies show that
there are five most common
causes of medication errors.
First, look-alike andsound-alike medicines are a
leading cause.
They can occur during all thestages of the medication use:
prescribing, transcribing,dispensing and also
administering.
Medication e rrors could occurwhen medicines have similar
(03:39):
looking or similar soundingnames, look-alike boxes,
look-alike tablets or look-alikeblister packaging.
Second, medication errors occurwhen protocols and guidelines
for safe use of medications arelacking in healthcare settings,
especially those related tohigh-alert medications.
(04:01):
Third, when communication isunclear, medication errors could
happen, of course.
For example, illegiblehandwriting orders could lead to
dispensing errors and the useof error-prone abbreviations may
also result in those errors.
(04:21):
I can give an example of amisinterpreted trailing zero
after a decimal.
This could lead to a tenfolddrug dose administration error.
Fourth, interruptions ordistractions to nurses during
administration may also resultin giving a drug by an incorrect
(04:45):
route or giving a drug toanother patient.
Last but not least, lack ofallergy documentation in medical
records contribute toprescribing errors, as well as a
lack of pharmacist medicationorders review.
In oncology prescription reviewby pharmacist is important to
(05:09):
intercept medication errorsrelated to those calculation
errors for weight basedmedication.
In addition, receiving complexmedication regimens in cancer
patients may also lead tomedication errors.
Federica Santoro (05:26):
So there's a
few different causes, and
obviously that helps thenunderstand how one can intervene
to try and prevent those errorsfrom happening, and we'll go
into that later on.
But first I'd like to talkabout frequencies, and I
understand just as it'sdifficult to pinpoint causes.
It can be even harder then todetermine frequencies.
(05:48):
But, lem, you might have somenumbers for us.
How often do medication errorsoccur?
Alem Zekarias (05:55):
So medication
errors can happen throughout the
use of medicines, as just Lubnamentioned, which usually
includes the prescribing,dispensing administration, but
also the monitoring stages.
According to the report "GlobalBurden of Preventable
Medication-Related Harm inHealthcare that was published in
2023 by the World HealthOrganization, where they have
(06:15):
done a systematic review andmeta-analysis of 100 studies
globally, so in this studie wasboth studies from high income
settings but also low-incomecountries, and the pool
prevalence of medication areasthat actually have led to harm
was 5%, which is basically onein 20 patients, and one-fourth
(06:35):
of the harm was severe orpotentially life-threatening.
A nd the highest prevalencerates globally for preventable
medication-related harm are forpatients managed in geriatric
care units, as well as amongpatients in surgical care.
So, if we focus on theavoidable medication-related
harms on a global level, abouthalf of them occurred during the
(06:58):
ordering prescribing stage andabout 30-40% at the monitoring
reporting stage.
It is important, though, as youjust mentioned, that these are
just approximate numbers, sothere is no really uniform
collection of data or harmonizeddefinition, which, of course,
makes it harder to estimate howoften medication errors can
(07:19):
occur or occur.
So they happen everywhere andwe do have challenges with
underreporting due to differentreasons, including lack of
capacity to detect and reportthese types of problems.
Federica Santoro (07:31):
So, even
though the figures are an
approximation, they still givean idea of the degree of the
problem and obviously we'retalking about a pretty intense
burden on health care.
Let's go into that a littlemore, Alem.
What is exactly the impact ofmedication errors on healthcare?
So, how do they affect patients?
(07:53):
But how do they affectcommunities as well?
Alem Zekarias (07:57):
So the direct
impact is, of course, the harm,
pain and suffering for theindividual patient and its
family.
The patient may experienceemotional, but also
psychological discomfort and, asa consequence to this, of
course, lose trust andconfidence to the healthcare
system.
Except this, another importantaspect is the financial impact,
(08:19):
which it's also important toremember.
As previously mentioned,one-fourth of the harms due to
preventable medication errorswere severe or potentially
life-threatening, which requireresources from the healthcare
systems.
Federica Santoro (08:34):
Let's move on
to the nitty-gritty, so to say.
I'd like to talk about datacollection and analysis now,
which I'm sure will interest allthe pharmacovigilance
professionals out therelistening to this episode.
So, Ghita, how is data onmedication errors collected?
And, specifically, I'mwondering, does it end up in the
(08:57):
same pharmacovigilancedatabases that we use to collect
regular reports of adverse drugreactions?
Ghita Benabdallah (09:05):
Thank you,
Federica.
Data on medication errors couldbe collected from healthcare
professionals and patients oncethey are familiar with the
culture of reporting medicationerrors, and, in the second way,
preventable ADRs that are calledalso medication errors with
harm could be detected throughindividual case safety reports
(09:29):
that are received atpharmacovigilance centers by
using a tool which is enablingto identify preventable ADRs.
This is the preventabilitymethod, called P method, which
enables the identification ofany well-known risk factors that
will increase the likelihood ofADR occurrence, which are
(09:52):
leading to understand the rootcauses of this medication error
and will help to be able to putin place the right risk
minimization actions.
The P method is used only bypharmacovigilance staff within
pharmacovigilance centers and itcomes after the step of
(10:13):
causality assessments, whichenables to confirm the link
between the adverse event andthe medication that is taken.
Regarding these data that arecollected, yes, they are all
sent to the samepharmacovigilance database we
use to collect a regular ADRreport.
(10:34):
So medication error with harmand without harm have to be sent
to pharmacovigilance databases,so the national one that we
have in our countries and alsothe international one, VigiBase.
Federica Santoro (10:49):
And we'll dive
into that distinction between
the national and the globalperspective a little bit later.
I'm glad you brought up the Pmethod because we actually got a
question about that on socialmedia from one of our listeners.
Mohamed in Egypt asks (11:05):
is this
method effective in keeping
healthcare professionalsproactive and helping them
prevent any potential risks?
Ghita Benabdallah (11:14):
Yes, so the
question is very interesting.
So at the individual level, ina case-by-case analysis, it is
not a proactive tool becausewhen there is an error it is
already there.
So this tool will enable toprevent in the future the
recurrence of medication errorsthat have been detected.
(11:36):
And at a public health level,it is part of an overall
medication error preventionstrategy.
For example, when we arelooking for preventable ADRs in
the global database, we areacting to be proactive.
Federica Santoro (11:52):
Thank you very
much.
I have another question for youfrom our social media followers
.
Sudarshan, who is based inIndia, asks how we can simplify
the identification and reportingof medication errors from a
patient's perspective.
So I guess he means basically,how can we make it easier for
(12:12):
patients to identify and reportthose errors?
Ghita Benabdallah (12:16):
Yes, so a
patient can report medication
errors if there is a reportingsystem in place.
So they have to be aware of thatand then they need to be
sensitized on that field.
So the reporting system inplace needs to be adapted to the
(12:37):
culture of the country and, forexample, in countries with an
oral culture, the reportingsystem could be the phone call
and it could be a solution andthis will be easier for patients
to report medication error bycalling, and this way of
reporting has also some benefitfor healthcare professionals
(12:59):
from pharmacovigilance centersbecause it allows them to gather
all information that is neededin one time.
So, regarding theidentification of medication
errors by patient, it dependsfirst on the engagement in their
own health care, because theyshouldn't solely rely on health
(13:21):
care professionals and theyshould know that they are able
to identify medication errorsand for that they should be
aware about the five R's.
When they are starting thetreatment they always have to
check that this is the rightpatient with the right drug at
(13:42):
the right dose and at the rightroute of administration and at
the right moment.
Federica Santoro (13:49):
Thank you, and
that leads nicely into my next
question.
I wanted to ask you, Ghita,about the barriers to reporting
medication errors in general.
You mentioned, obviously, thata system has to be in place,
that people have to be awarethat it exists and have to be
aware of how to report.
But let's consider it not justfrom a patient's perspective.
(14:12):
Underreporting of adverse drugreactions is a well-known and a
pretty big issue in thepharmacovigilance world, and
we've discussed it on thepodcast before.
Do medication errors gounreported for the same reasons
as regular adverse drugreactions, or are there
(14:35):
additional factors at play here?
Ghita Benabdallah (14:38):
Thank you,
Federica, for this question and,
as you just mentioned, yes, atthe basis all countries are
facing underreporting of adverseevent and, of course, also of
medication errors as well.
For the latter, for medicationerrors, this is mostly due to
the culture of patient safety,with the concept of non-blame
(15:02):
that is not implementedeverywhere.
So this leads to a fear toreport because there is a fear
of repercussion.
And, on the other hand, westill have patients who do not
know that they are allowed toreport because there is still
this lack of culture of patientsafety and of reporting.
Federica Santoro (15:28):
And going back
to that blame issue, is it a
problem that we call them errors?
Does that just feed into theblame and shame?
And is that then stoppingpeople from documenting these
issues even more?
Ghita Benabdallah (15:48):
Yes,
definitely, the term error would
be an element that could stopthe process of reporting,
because it's referring to theblame, to the shame, and there
is a repercussion on the careerof the healthcare professional.
So, it's very important toimplement a culture of non-blame
that focuses on the fact thaterrors are not due to the human
but rather to the system inwhich the human works.
(16:18):
And regarding that point, thereis a very famous psychologist,
James Reason, who stated alreadyin 1995, that we cannot change
the human condition, but we canchange the conditions under
which humans work.
And this is how we moved fromthis individual approach to the
(16:40):
systemic approach.
Federica Santoro (16:43):
Absolutely,
it's a very interesting and very
positive, constructive way toframe it.
Thanks for sharing that.
Now I'll let you rest yourvoice for a little bit, Ghita.
I'll move over to Alem.
Let's talk about coding.
So what should healthcare orpharmacovigilance professionals
keep in mind when they're codingmedication errors into their
(17:06):
database?
Alem Zekarias (17:09):
It's exactly the
same thing as you should
actually think about when youreport suspected adverse drug
events.
Try to report everything andtry to describe what happened.
So if you don't have theinformation needed, contact the
primary reporter, especiallywhen it comes to medication
errors, and ask the additionalquestions.
So the report try to code ascorrect, clear and complete as
(17:32):
possible, which is important forthe root cause analysis,
because understanding whatcaused the error and how the
error can be avoided isimportant to prevent future
unnecessary patient harm.
The more accurate data that hasbeen collected, the higher are
the chances for a successfulroot cause analysis and
(17:53):
appropriate risk minimizationactions relevant to your own
setting can be suggested andimplemented.
There are coding guidelines, asthe one being provided by the
Medical Dictionary forRegulatory Activities, also
referred as MedDRA, which can beused as a support for the
coding of adverse event terms.
Federica Santoro (18:13):
Okay, I'd like
to move on to data analysis now
, and we've touched on this hereand there.
We've mentioned the P method.
You mentioned root causeanalysis, so we've hinted at it
so far, but I'd really like toexplain for our listeners how
the analysis of medication errorreports happens.
(18:35):
So how are signals detected andassessed for these types of
reports?
Are there any special methods?
Do regular pharmacovigilancemethods apply?
Is it a combination?
Loubna, take it away.
Loubna Alj (18:51):
Thank you, Federica.
Yes, the regularpharmacovigilance pipeline does
apply, although there are someparticularities.
So detecting signals related tomedication errors may be
qualitative or quantitative.
However, when pharmacovigilancecenters have small medication
error databases, the qualitativeapproach is more applicable.
(19:12):
So the pharmacovigilance staffproceed to a manual screening of
medication errors reported tothem and look for those that
draw their attention.
I can here give you examples oftriggers of medication error
signals.
We can consider seriousmedication reports, potentially
(19:35):
serious medication error reportsand also clusters of medication
errors.
Please note that only oneserious medication error is
enough to trigger a signal.
For example, if you receive, asa PV center, a fatal outcome
with paracetamol in a child thatwas administered adult
(19:59):
formulation instead of pediatricone, how would you consider
this?
Is it a signal or not?
I'm sure you will consider itas a signal and you will
investigate the case tounderstand how and why the error
occurred and what should bedone to prevent it.
(20:20):
Now, when it comes to signalvalidation, medication errors
are, by definition, preventable.
That is why there is no need toconduct epidemiological studies
to validate medication errorssignals in contrast to adverse
drug reactions which occur undernormal conditio n of use.
Federica Santoro (20:55):
That's very
helpful, thank you.
Alem, could we talk a littlebit about the global perspective
?
So obviously here at UMC we'relooking into a global database,
VigiBase.
So what happens when we want toanalyze medication error
reports like that, that arecoming from all over the world?
Is something different?
Alem Zekarias (21:05):
So, as Loubna
just mentioned, there are two
approaches to do signaldetection: it's the quantitative
or the qualitative approach.
For databases with a largenumber of reports of suspected
adverse drug events, thequantitative approach, with
support of different statisticalmethods that we also apply here
at UMC, can be used.
And what's mainly differentbetween regular
(21:26):
pharmacovigilance and signaldetection, but also assessment
on suspected medicationaryreports, is actually what type
of data you collect and report.
In general pharmacovigilance,the reported data is primarily
for causality assessmentpurposes.
You want to understand if thereis a possible causal
relationship between thereported medicinal product and
(21:47):
the reported adverse event,while for medicationary related
problems, the main purpose isactually to understand the
reasons and factors behind thisinappropriate medication use.
It's first when you actuallyfully understand what happened
you have the possibility toprevent it from occurring again.
Federica Santoro (22:05):
And, on that
note, let's talk about
prevention.
Now, as we said at thebeginning, we are talking about
this ultimately because we wantto try and prevent those
mistakes from happening again.
And now, while I appreciatethat achieving zero mistakes and
zero harm is an impossible feat, we're only human after all,
(22:36):
what are the most effectivestrategies, though, to try and
prevent those errors?
Back to you, Ghita, for thisone.
Ghita Benabdallah (22:41):
So, there are
many strategies and it depends
on the country where it will beimplemented, but it always
starts by creating a culture ofpatient safety.
To be able to report in thisnon-blame atmosphere, having a
good communication with thehealthcare professional, between
(23:03):
them, and between healthcareprofessional and patient is also
very important important.
Having a very well implementedvigilance system that will react
and act quickly is also the keyelement for putting in place
this strategy.
Regarding the training, it'svery important to strengthen
(23:27):
training; general training, butalso specific training, as, for
example, training in prescribing.
And within the health system,it's important to be proactive
and implement therapeuticprotocols for clinical
management and also focus onhigh alert medications.
(23:48):
And finally, it's veryimportant to be part of
medication error networks toshare prevention strategies.
Federica Santoro (23:59):
How about
patient engagement?
We mentioned this earlier, buthow can it specifically help to
prevent medication errors?
Do you have any real lifeexamples of how that can work?
Ghita Benabdallah (24:11):
So, there are
some examples, but patient
engagement is very importantit'snot possible to rely important,
because, as I said before, it'son healthcare professionals.
And when the patients areengaged in their own healthcare,
they could, for example, by thedouble-check of the five R's,
(24:32):
prevent the occurrence ofmedication errors.
And, as a real-life example, Icould say that the prevention of
an immunoallergic reaction bythe double-check of the patient
just before the administrationof ibuprofen at the hospital.
Here, the patient was allergicto ibuprofen and it wasn't
(24:53):
mentioned before.
Federica Santoro (24:55):
We have a
couple of more questions from
our social media followers.
This clearly was a topic thatresonated very much with our
audience; when we advertised iton social, we got quite a few
responses.
So, Ghita, here's anotherquestion from Sudarshan in India
.
He asks about technologicalinnovations and can you imagine
(25:18):
how they could help preventmedication errors.
He cites an interesting exampleof a medication packaging that,
for example, could change colorif stored incorrectly.
Ghita Benabdallah (25:31):
Yes, yes.
So, there are many actions tobe implemented to prevent
medication errors, as having theTall Man lettering on the boxes
of medicines to avoidmedication errors that are due
to look- alike or sound-a likedrug names.
We can use the pictograms, wecould have adapted storage, we
(25:54):
could use different storage, andthis is an action that is easy
to implement and is efficient toprevent medication errors.
The barcodes, when available,will help to reduce medication
error, as well as someartificial intelligence tools as
e-prescribing, that can be usedto decrease medication errors
(26:19):
and prevent human errors.
Federica Santoro (26:23):
Great that you
mentioned AI, because another
listener, Paula in the UK, had aquestion on that.
So can artificial intelligencehelp both patients and
healthcare professionals reducethese errors?
Ghita Benabdallah (26:37):
The use of AI
for medication management and
in reducing medication errorswould help identify potential
medication errors that couldhappen during the whole delivery
process.
Regarding the AI, it isimportant that the tools to be
used have to be validated andthat they will be used in a
(27:02):
rational way, without outpassingthe role of healthcare
professionals.
The AI used to preventmedication errors needs to stay
a tool to help and could notreplace the role of healthcare
professionals.
Alem Zekarias (27:15):
There are some
existing support already, AI to
support, like barcode scanning,for example, but also this
robotic dispensing errors, whichbasically using AI that
accurately dispense medications,which studies have seen reduces
the risk for human error indosage and labeling.
Then we have something calledclinical decision support system
, of course, that providereal-term alerts, reminders to
(27:36):
the healthcare professionalsabout dosage errors, incorrect
prescriptions, also calledautomated prescription
verification.
That can basically an AIalgorithm that can cross -check
prescriptions against patientrecords, medical history and
known drug interactions, to flagpotential errors before
medications are being dispensed.
But also there are ways youknow, if you look at the
(28:00):
analysis part, how AI cansupport.
One way could be using machinelearning for pattern recognition
, which can be applied toanalyze reported safety data to
identify patterns of medicationerrors and suggest interventions
or processes to the healthcareproviders.
Federica Santoro (28:15):
Thank you both
for that.
I hope that answers ourlisteners questions and gives
them an idea of where technologyis headed in this field.
Now, we're nearly at the end ofthe interview, but before I let
you go, I'd like to hear yourreflections from a course on
medication errors that yourecently led.
So this was a course that youdeveloped with some other
(28:37):
colleagues of mine here at UMC,and it was designed to help
pharmacovigilance center staff,who maybe lack procedures for
collecting and analyzingmedication errors, to get them
started on that journey.
So I was just wondering whatdid you, as instructors, learn
(29:00):
from them, the participants,regarding the needs and the
knowledge gaps in thisparticular field of
pharmacovigilance?
And I'd say we start with you,Loubna.
Loubna Alj (29:12):
I think that
additional efforts are needed to
remind the pharmacovigilancecommunity that medication errors
are part of thepharmacovigilance scope.
The medication errors courseorganized by the UMC was a great
opportunity for national PVcenters to remind them, and to
(29:35):
stress the fact that they shouldimplement or enhance their
medication errors activity tobetter contribute to medication
safety.
Moreover, there is still someconfusion to distinguish
medication errors from otherdrug-related problems.
Always remember that medicationerrors are not deliberately
made by healthcare professionalsor patients.
(29:55):
As stated in the definition,they are unintended.
I would like also to add thatlooking for causes of medication
errors should be systematicallyperformed and keep in mind that
it is important to learn fromerrors, regardless of their
seriousness.
Indeed, every medication erroris an opportunity to improve
(30:18):
medication safety and patientsafety worldwide.
Federica Santoro (30:22):
Thank you.
Ghita, is there any otherreflection you'd like to add?
Ghita Benabdallah (30:26):
Thank you
very much, and thank you Loubna.
I believe that Loubna saideverything, so I will just add
that we have seen thatparticipants know that the
burden of the problem is there.
They are aware about that and Iwould like just to say that all
(30:46):
the tools are available and theparticipants need more training
and they need training,specific training, workshops,
and working groups to be able toimplement the tools that were
presented.
Federica Santoro (31:06):
Thank you for
that.
And finally, Alem, any lastthoughts?
Alem Zekarias (31:11):
I can just agree
with both my colleagues, Loubna
and Ghita, because medicationerrors are part of
pharmacovigilance, but despitethat, we still lack quite a lot
of suspected or virtualmedication- related reports from
several parts of the world inVigiBase.
So working with the medicationarea requires a different type
of mindset, different processesand guidelines, and it's not
(31:31):
easy as you may think, which wasevident during the course.
There was definitely acuriosity and an interest to
start tackling these types ofproblems as part of the
pharmacovigilance mission, whichis great, especially if you
manage to do this in closecollaboration with relevant
stakeholders, including thepatients.
But it may be hard to know howand where to start.
(31:52):
So I hope this course hasenabled and motivated healthcare
professionals working indifferent settings around the
world to start to identify,report, analyze and communicate
medication errors as part oftheir responsibility to ensure
safe use of medicinal productsfor their patients.
Federica Santoro (32:10):
Thank, you
very much.
I'm sure the course has helpedand I hope that this podcast,
where we tried to give anoverview of the field of
medication errors, will do therest and set people on a course
to start collecting andanalyzing medication errors more
and more.
Thank you very much, all threeof you, for sharing your time
(32:33):
with me.
It was such a pleasure to talkto you.
Alem Zekarias (32:36):
Thank you so much
.
Loubna Alj (32:37):
Thank you, Federica
.
Ghita Benabdallah (32:39):
Thank you,
Federica, thank you so much.
Fredrik Brounéus (32:51):
And if you'd
like to learn more about
medication errors andpreventable adverse drug
reactions, we've put togethersome useful links for you in the
episode show notes.
Apart from these conversationswith experts, we offer a series
of audio stories from UppsalaMonitoring Center's
pharmacovigilance news sitecalled Uppsala Reports Long
Reads, so why not check that outtoo?
That's all for now, but we'llbe back soon with more on
(33:12):
medicines safety.
In the meantime, we'd love tohear from you.
Reach out on Facebook, LinkedInand X, send comments or
suggestions for the show orquestions for our guests next
time we open up for that.
And visit our website to learnmore about what we do to promote
safer use of medicines andvaccines for everyone everywhere
.
If you like the podcast, pleasesubscribe to make sure you
(33:36):
won't miss an episode and spreadthe word so other listeners can
find us too.
For Drug Safety Matters.
I'm Fredrik Brouneus.
Thanks for listening.