Accepted: 8 March 2018

DOI: 10.1111/jocn.14353

ORIGINAL ARTICLE

Medication safety challenges in primary care: Nurses’

perspective

Hanan Khalil PhD, BPharm, MPharm, Senior lecturer/Pharmacy Academic1 |

1
Sarah Lee BSc, PgCert, MRes, Research Assistant

Faculty of Medicine, Nursing and Health

Sciences, Monash Rural Health, Monash
Aims and objectives: To identify the issues surrounding medication error reporting
University, Moe, Vic., Australia in community nursing and improvement strategies related to medication safety.

Correspondence
Background: Medication-related problems have been identified from various
Hanan Khalil, Faculty of Medicine, Nursing sources in the literature. Examples of these include incident reporting by healthcare
and Health Sciences, Monash Rural Health,
Monash University, Moe, Vic., Australia.
professionals, medico-legal and patient complaints and systematic identification of
Email: [email protected] organisational structure. Only a few studies report on the clinicians’ perceptions of
medication safety in community nursing and the challenges they face within their
workplace to implement medication safety initiatives.
Design: Qualitative design, using conversation-style interviews with experienced
registered nurses in primary care roles.
Methods: Using a general iterative approach of semantic analysis, our qualitative
research study was guided by an essentialist paradigm. Our method for understand-
ing included semi-structured in-depth interviews with 10 clinicians from a large
community care organisation in rural Victoria in Australia. We developed an inter-
view guide, which included open-ended questions on clinicians’ experiences, per-
ceived barriers and facilitators, and strategies to improve medication safety.
Results: Several barriers have been identified by healthcare practitioners that hinder
medication safety in primary care including culture differences between community
and hospital setting, politics within the healthcare system, lack of clarity around the
nurses’ roles and lack of error reporting. Other sources of errors cited by the partici-
pants were the lack of clarity or awareness of the processes and procedures of
medication incidents reporting for staff within the organisation experience. Lack of
education regarding medication safety, the dilemmas associated with reporting and
documentation are also significant barriers.

KEYWORDS
medication error, medications, medication safety, medications error reporting

1 | BACKGROUND identification of organisational structure (Department of Health

2000; Tsang, Bottle, Majeed, & Aylin, 2013; Wallace, Lowry, Smith,
Medication-related problems have been identified from various & Fahey, 2013).
sources in the literature including incident reporting by healthcare Adverse drug-related problems are associated with significant
professionals, medico-legal and patient complaints and systematic patient harm including morbidity, hospitalisation, increased costs of
health care and death in some instances (Andel, Davidow, Hollander,
Study type: A qualitative research study. & Moreno, 2012). The cost of medications errors was calculated to

2072 | © 2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jocn J Clin Nurs. 2018;27:2072–2082.
KHALIL AND LEE | 2073

be $US 19.5 billion annually in 2008 in the USA. This cost does not
include the expenses associated with use of the healthcare system
What does this paper contribute to the wider
including hospital admissions which was estimated to be $US 177.4
global clinical community?
billion each year in 2001 (Andel et al., 2012). Hence, medication
safety has been the focus of attention of many governments and • This study notes that the issues surrounding medication
several medication safety initiatives have been put in place to error reporting in community nursing are multifactorial,
address the problem of medications errors. Examples of these medi- complex and often institution-specific. Several challenges
cation safety initiatives include training packages, organisational were highlighted by community nurses who hinder medi-
structures and policy changes. Many of these initiatives had different cation safety in primary care including culture differences
success rates (Hartnell, MacKinnon, Sketris, & Fleming, 2012; between community and hospital setting, politics within
Nguyen, Connolly, & Wong, 2010; Roshanov et al., 2013; Stuijt the healthcare system, lack of clarity around the nurses’
et al., 2013). roles and responsibilities and error reporting.
As most prescribing activity takes place in the general practice • The development of clear guidelines detailing nurses’
(GP) environment, primary care has been the focus of recent efforts roles and responsibilities regarding medication administra-
to address the issue of medication safety. Errors in prescribing medi- tion in the community is crucial.
cations were among the most common type of errors in primary care
in the USA as well as other countries. Other common errors include
those associated with communication and diagnosis (Kostopoulou & study examining the types and causes of medications from nurses’
Delaney, 2007; : Kuo, Phillips, Graham, & Hickner, 2008). This is rel- perspectives revealed that 65% of the errors are the result of nurses’
evant to both general practitioners and nurse practitioners who are actions. The most common errors were due to lack of knowledge
eligible to prescribe medications in Australia. (Cheragi, Manoocheri, Mohammadnejad, & Ehsani, 2013).
The highest rates of medication errors tend to be found in This type of information is crucial in order to seek improvements
patients taking multiple medications and in the older population. in patient safety and recommend ways to improve the current clini-
Reasons for these errors have been studied in-depth over recent cal practice. Identifying barriers and facilitators to medication safety
years and have been identified that both individual factors (e.g., in community care will enable organisations to implement educa-
knowledge about medication and slips in attention) and organisa- tional programmes to improve practice.
tional factors (including poor communication, work structure, work-
€ gele,
load, training and supervision) are key influential (Lainer, Vo
€ nnichsen, 2015; Pearson & Aromataris, 2009).
Wensing, & So 2 | OBJECTIVES
Another contributing factor to the medication safety in primary
care is the difficulty in co-ordination and management of the clinical In this study, we explored nurses’ perceptions regarding barriers,
information related to each patient (Lainer et al., 2015; Pearson & facilitators and improvement strategies related to medication safety.
Aromataris, 2009). A study by Smith et al. of primary care general
practitioners reported that their missing information was evident in
nearly 13.5% of the visits undertaken by these practitioners. Exam- 3 | METHOD
ples of missing information included laboratory results, letters, infor-
mation related to patients’ medications and other documentation The study gained ethics approval from Monash University Ethics
related to radiology and physical examination (Smith et al., 2005). Committee (CF14/3987—2014002063), informed consent was
Community nursing has a broad range of responsibilities includ- obtained from all participants, and the procedures followed were in
ing treating and caring for sick people in the home after hospital dis- accordance with the National Health and Medical Research Council
charge, health promotion, illness prevention, policy development and of Australia National Statement on Ethical Conduct in Human
advocacy and community development. The challenges in rural and research. We used a qualitative design, using conversation-style
remote communities for many of these nurses include the diversity interviews with experienced registered nurses in primary care roles.
of clients, with a wide range of clinical presentations, long-distance The study attempted to explore both the attitudes and cultures
travel, lack of consistency in the organisational structure for allied of the workplace and the facilitators and barriers faced by healthcare
health services and access to quality information technology and practitioners in relation to medication safety. The underlying episte-
communication systems (Paliadelis, Parmenter, Parker, Giles, & Hig- mology stemmed from an essentialist paradigm. The essentialist
gins, 2012; Wong et al., 2017). paradigm assumes there is a unidirectional relationship between
While many studies reported on the types of adverse events and meaning, language and experience. It focusses on reporting experi-
the rate of errors (Pearson & Aromataris, 2009; Rosser et al., 2005), ences, meanings and assumed reality, which is evident in the data
only a few studies report on the clinicians’ perceptions of medication (Braun & Clarke, 2006).
safety in community nursing and the challenges they face within The study set out to generate phenomenological data, identify
their workplace to implement medication safety initiatives. A recent meanings of the data. The aim of the study was to ascertain the
2074 | KHALIL AND LEE

attitudes and cultures of community nurses in a community health different to a hospital setting?” “What are the attitudes and views
setting in rural Victoria, Australia, and to gain an understanding of of staff regarding patient safety in the community?” “What factors
the barriers and facilitators to medication safety for community facilitate and inhibit reporting?” “What do you think are the major
nurses. Purposive sampling was used in accordance with the aim of factors that contribute to medication safety problems in the com-
the research. Morse (1991) states that “when obtaining purposeful munity?” These questions facilitated discussion of the participant’s
sample, the researcher selects a participant according to the needs perceptions regarding barriers, facilitators and improvement
of the study”. The researcher deliberately sought participants who strategies related to medication safety. The semi-structured sched-
had particular knowledge and experience but also employed maxi- ule allowed flexibility to meet demands of individual context
mum variation within the sample as described by Sandelowski, (Table 2).
1995;. Barbour (2001) argues that this type of recruitment enhances
sample coverage and provides the researcher a framework for
analysis. T A B L E 1 Participants’ characteristics
The researcher followed a topic guide that included the following Parameter Average/Range
themes: attitudes and culture of the workplace and the facilitators Age 40.8 (25–61)
and barriers to medication safety. Field notes were also taken by the
Gender F
researcher conducting the interviews. Interviews were transcribed
Profession Registered nurse educated at a
verbatim and rechecked for accuracy by the researcher for subse- baccalaureate level
quent analysis. Number of years practicing 16.3 (2.5–39)
Type of practice setting (Solo, Solo
Duo, group, unknown)
3.1 | Participants
Full-time equivalence 0.65 (0.4–0.9)
Ten female nurses (aged between 23–60 years) working in a commu- Number of patients managed a 9.8 (7–13)
nity health service took part in interviews conducted between Octo- day
ber 2016–December 2016. All participants having read and signed a
consent form were interviewed and audio-taped for 35–80 min.
Open-ended semi-structured questions that were informed by the
T A B L E 2 Interview questions
Reason’s Accident Causation model (Reason, 1995) were used flexi-
Attitudes and culture of the workplace
bly and adapted to meet demands of individual context. The Rea-
son’s Accident Causation model proposes that there are multiple 1. What are the attitudes and views of staff regarding patient safety
in the community? With regards to medication administration,
barriers within complex organisations such as hospitals to prevent
prescribing and monitoring?
adverse events. Examples of these barriers include policies and clini- 2. Is medication safety something you think about it while doing
cal guidelines. However, there are random weaknesses in these bar- your work?
riers that can lead to adverse events. 3. Is a culture within the community setting of ‘just do the job/get
The nurses interviewed were all registered nurses with a bac- through the list and visit a client and move on to another’?
4. Do you feel the community culture is different to a hospital set-
calaureate qualification. Their main roles within the organisation
ting?
range from treating and caring for sick people in the home after hos- 5. Is there a difference in medication safety guidelines and protocols
pital discharge, health promotion illness prevention and policy devel- in the community in comparison to the hospital? If so how does
opment regarding medication safety within the organisation. The this impact on you? The culture? Other staff? and attitudes
participating organisation is the largest community health service in towards patient safety in the community
6. Is it harder to implement medication safety in the community?
the region and is focussing on translating evidence into practice and
7. Can you describe how incidents are generally handled in your
hence was keen on taking part in the research study. It has almost workplace?
40 community nurses (Table 1). 8. What types of incidents are reported/not reported? What are the
staff’s/your feelings and attitudes towards reporting medication
errors?
3.2 | Procedure 9. What factors facilitate and inhibit reporting?
10. How do you think reporting of errors could be made easier for
Appropriate approvals from Monash University Human Research staff?
were obtained. Participants were given an information sheet at the
Facilitators and barriers to medications safety
time of interview, and informed consent was obtained at the inter-
1. What do you think are the major factors that contribute to medi-
view. Interviews followed a semi-structured schedule. Interview cation safety problems in the community?
questions were designed to prompt discussion of participant’s atti- 2. What do you think some of the challenges are to improving medi-
tudes and cultures of the workplace and what they believed some of cation safety practice in the community?
the facilitators and barriers to medication safety in their workplace 3. Do you have any suggestions of how medication safety in the
community could be improved?
are. Questions included: “do you feel the community culture is
KHALIL AND LEE | 2075

Interviews were audio-recorded, identifying details were

4.3 | Culture
removed, and interviews were transcribed by a transcription com-
pany and then rechecked for accuracy by the researchers. The theme of culture is presented in a number of different ways in
this study: first, in terms of language, specifically that of medication
administration or as some termed it, “supervision of medication.”
3.3 | Data analysis
Community nurses felt they lacked clarity and certainty over their
Thematic analysis is an analytical methodology that allows for a rich role because of the variance of language, with some using the terms
and detailed account and interpretation of complex data. It permits medication supervision which meant they were simply there to
the researcher to search for themes and patterns both within a data observe a patient taking their medication while others suggested
item and across a data set, focussing on the content of the data and that the term medication administration could be interpreted as
reflecting on the reality and therefore consistent with the essentialist administering medication to patients. Clinicians interpreted the lan-
paradigm that was used to analyse the resulting interview data guage differently, which had a knock-on effect in terms of the role
(Braun & Clarke, 2006). they felt they had in community work specifically medication admin-
Thematic analysis began once data collection was complete. An istration:
iterative process of semantic analysis was applied to the data corpus
using an inductive approach until theoretical saturation was I was just discussing that supervision and administration
achieved. Patton (1990) describes this inductive method as a way to on our drug charts can mean different things to other
identify themes that are strongly linked to the data. staff. So I might supervise but another client might
The researchers immersed themselves in the data and read and administer, there’s no clear description of what each
re-read in an iterative process to familiarise themselves and make means. And does administration mean physically putting
sense of the data and noted any initial analytic observations. This the tablets in that client’s mouth or does it mean just
approach allowed themes and patterns to be generated in the data. popping it out of a Webster pack?.
For internal coherence and comprehensiveness, two researchers
actively coded the data and identified initial codes and themes in One of the most frequently discussed topics within the interview
the data. To identify new codes, the researchers read through each was the difference between the cultures in a hospital setting com-
data item line by line, working systematically through the data set pared to that of a community setting. Some of the participants had
and coded important features and patterns in the data. Codes were worked previously in the hospital setting or were still working in a
organised into coherent categories; this was a recursive process that hospital setting as well as a community setting and were able to dif-
was continued until all codes and themes identified were categorised ferentiate between the two contexts. They highlighted the vastly dif-
and refined. Interpretation of themes and codes was conducted by a ferent roles they have in the community particularly around
process of repeat checking against the data set to ensure an accu- medication administration. Nurses in the community are only allowed
rate representation of the data corpus. Connections and meanings
were identified both between and within themes and subthemes. T A B L E 3 Summary of results
The researchers systematically conceptualised the overarching
Theme Subthemes
themes, and any discrepancies were discussed.
Culture 1. Language
2. Hospital versus community setting
4 | RESULTS 3. Politics in the workplace
Causes of medication error 1. Prepackaged medications
4.1 | Participants’ characteristics 2. Lack of sense of responsibility

Ten clinicians from one organisation agreed to be interviewed for 3. Documentation

this study. The characteristics of the participants are shown in 4. Logistical challenges
Table 1. They were all females and aged from 25–61 years. All par- 5. Patients vulnerability
ticipants worked part-time, with many of them working between 6. Lack of guidelines
one day–4 days a week, which is typical of primary care nursing in 7. Workload
Australia. 8. Lack of education and training
There were a total of four main themes reported by participants: Causes of nonreport 1. Complicated reporting process
culture, causes of medication errors, causes of nonreport and strate-
2. Delayed feedback
gies for improvement.
3. Defensiveness
4. Fear
4.2 | Findings Strategies for improvement 1. Development of guidelines
2. Improved documentation procedures
A summary of the themes and subthemes is shown in Table 3.
2076 | KHALIL AND LEE

to supervise patients taking their medications as opposed to admin- the nurses not being able to correctly identify what they are giving
istering the medication directly, because in the community the client the clients. This is significant given that these devices are specifically
is in their home and therefore had control over their medication designed to reduce error and risk:
nurses felt they played a different role in the patients’ care. More-
over, community nurses have to be familiar with the different medi- I prefer the standardised ones we use in the hospital for
cation chart types from a range of hospital or general practices medication, I much prefer them. I much prefer medica-
when working in the community because there was no continuity of tions out of pharmacy packaging than a Webster pack,
care between hospitals and community settings: so that I know that that’s the medication I’m actually
giving and not just oh well, it says that’s what it is.
I feel sorry for the new people that are coming in . . .
community system. I’ve been working both jobs now for Some participants perceived other major causes of medication
like nearly 10 years, so I’m sort of aware of the differ- errors are staff carelessness or complacency in community nursing.
ences. But yeah, it’s very, very different, and I think it’s Most staff have a large number of clients to care for in any one shift
a big culture shock in regards to learning that like things and lack time to get through the lists. Many participants highlighted
actually aren’t what they seem. that they have to go through a list of clients by the end of their
shift, and therefore, they may choose to shorten their routine check-
There was a sense that some participants felt that risk of harm ing in order to visit all the clients they are assigned to for the day:
to a patient was much lower in the community than in a hospital
and so because of the perceived risk of error they became less vigi- Absolutely, absolutely, you can become complacent, you
lant and careless. Participants believed if a medication error was to absolutely can, particularly when you’re really busy. And
occur, it would not be too serious or cause too much harm to the you do, and you know, I’d be lying if I said I didn’t, you
patient: just – you run in and they’ve got five tablets ordered,
and there’s five tablets in the Webster, well here’s your
If someone makes a mistake we always tell the General tablets, see you later, have a nice day. Yeah, I mean
Practitioner (GP) what’s happened and we always tell that’s not a good position, but it happens.
the client to go to the GP for a review. It’s not like any-
thing bad is going to happen. It’s not like we’re in a hos- One participant felt that duty of care varies among staff and that
pital. I don’t think we can make a massive mistake like people lose their sense of duty of care when they come to do com-
we could in the hospital. So our risks are a little bit munity work:
lower in the community setting.
Yeah. Just, it comes down to the individual I think and
Many participants discussed that they believed there were cer- what their duty of, how they see their duty of care to
tain organisational politics within the healthcare system regarding their clients. I think a lot of people lose their duty of
staff engaging with each other, moral dilemmas about their responsi- care when they come out to communitySubsequently
bility towards medication incidents and the consequences when inci- the significant workloads was perceived by staff to be
dents were reported. Staff intimidation was also an issue when the cause of losing their duty of care.they’re there to do
incidents occurred and played a factor of whether staff choose to their work and earn their money and that’s it and there’s
report or not report a medication incident: no, what’s the word, duty of care to their patient’s.
There’s, in the hospital obviously you get a lot more duty
Oh there’s a lot of politics with ambulatory care but I of care to your patients whereas out in the community
stay out of it if it’s got nothing to do with me. So I put there’s a lack of duty of care I reckon.
my foot down and say it’s got nothing to do with me.
More than happy for people to debrief but you can see Participants strongly believed that a significant cause of medica-
the impact it does have on some of them because some tion errors in the community was documentation, either poorly kept
of them don’t get very happy and they can’t, they show documentation or trying to manage the different forms of it. There
it in the work place. So it doesn’t really help the staff was no consistency in approach:
and there’s just a few people that aren’t very kind.
In the community, medication charts come in all differ-
ent forms, they could just be a letter, it could be a chart,
it could be our chart that we faxed to them that they’ve
4.3.1 | Causes of medication errors
written on, they’re all different.There’s nothing universal
Staff mentioned that dose administration aids are a huge source of about our medication process in the community as such,
errors when medications are administered from them. This is due to which I think could be improved. It’s, it’s working but I
KHALIL AND LEE | 2077

think it could be improved a lot, compared to the hospi- would ask for something to be done, nurses felt conflicted by want-
tal. ing to follow organisations policy and procedure but found there
was ambiguity resulting from scenarios that occur in community
The importance of good documentation was acknowledged by nursing. Most of the participants mentioned that if clearer guidelines
participants but they felt there was no system in place. Some files existed in the workplace, they would serve as a validation of their
were kept as both paper and electronic documents, and this lack of work and a backup for any misinterpretations:
consistency and complicated process was a cause of medication
errors: So rather than having us battle almost it’s more like
we’re all on the same page and we all know what’s
Well there probably is a billion things we could probably expected and this is what the nurses at *organisation*
do but it all comes down to how we document and keep will do. They will work within these guidelines. And oh,
people’s files as well. It’s just a bit all over the place. okay well if I want to work with the nurses I have to do
We’ve got paper files and then we’ve got computer that too. And they might have their own guidelines
files.We don’t have standardised charts, so you can have about things and then we can fit them in together. But
orders that all over the shop and you’ve got no idea it would just really help. I think it would give us more
what they’re even on. You’ve got to wade through validity and sort of more credence I suppose to the doc-
paperwork to try and find the orders, and then you find tors to say well this is the way we feel best practice is.
it, and that was last month’s not this month’s, and so, a
lot of it’s our bookkeeping, if you like, and we don’t have Workload was another cause of medication errors. Nurses
standardised drug charts, so I think that’s an issue. And I described feeling like they are under a lot more pressure in a com-
think – and because we work with a computer and a munity setting, part of this pressure stemmed from chasing orders
paper, you just – I don’t know, it’s just doubling up. and challenges of communicating with other health professionals, in
particular with GP’s which was described as being stressful and diffi-
One of the challenges participants faced was the process of cult at times and something which was out of their control:
obtaining a prescription for patients and before administering medi-
cations. Participants believed they were stuck between trying to fol- You can have really crappy days, like really crappy days.
low correct policies and procedures and assisting patients to access Most days are fine, but you can have a really long day
their medication. The logistical challenge nurses faced in a commu- of just chasing orders, scripts, drugs, whatever it is, it
nity setting inevitably resulted in them waiting a number of days for can be a lengthy process and can make you in a bad
a doctor to sign off on a prescription. Participants said often doctors mood obviously, make you stressed, feel pressured, and
would just write a note on a piece of paper for the patient, but as or it’s things that are out of our control.I’d say GPs are
this was not the correct documentation, nurses could not administer a bit of a hindrance too to be honest, they like just fax
it and the process of trying to contact a GP was a challenging one, that off and you haven’t signed it, you haven’t dated it,
this caused a lot of internal conflict: hasn’t got your letterhead on it, come on I need more
than that.you’ve got to run around and, and do all those
I could just say, ‘Look, I’ve only got an order for 14’. A things. So trying to prevent an error from happening but
couple of days of higher blood sugars is not going to then in that process of you rushing around to get to
change anything. But in reality, I just think they’ve gone everyone else, another error could happen. So it’s defi-
to the doctor, they’ve gotten that piece of information nitely a lot more pressured.
to try and help them, and here I am holding them back
for another 3 days until I can get the doctor’s order in. Participants felt that there were not the same training opportuni-
It’s a really hard position to be in. ties in the rural community setting as there is in a hospital setting. In
a hospital there is an onsite educator and training courses delivered
Participants also believed that one of the most significant causes frequently; however, there are not the same opportunities in com-
of medication errors in the community is that there is lack of clear munity work:
community nursing specific guidelines. There was a lack of clarity,
and current guidelines were not fit for purpose. The lack of guideli- In a hospital there’s trainings every day, so for handover
nes caused many challenges for nurses, it meant that other health- there’s usually a bit of a, education session, then there’s
care workers were not clear of the nurses’ roles, the lack of also education sessions held at the hospital perhaps
understanding of nurses’ roles by other health professionals including after work or here they do the important ones, as I said
general practitioners and pharmacists resulted in conflict of expecta- the annual ones. If I wanted to go to a specific medica-
tions and responsibility in patient care. Nurses felt they had nothing tion education session, I might have to go to Melbourne
to refer to in order to support and validate their actions When a GP and spend $300 on a course for a day and then perhaps
2078 | KHALIL AND LEE

stay the night or find my way back home.Participants fear associated with the repercussions of reporting medication errors
also said they find it challenging to be approved for at an organisation level. Other reasons were fear of discussion of
study leave to attend training courses. incidents in the staff annual performance as it was considered a
cause for staff conflict and stress in the organisation:

Fear of the backlash what’s going to happen, that would

4.3.2 | Causes of nonreport
probably be the big one; I’ve made an error, what’s going
The reporting process was discussed by most participants as a bar- to happen and quite clearly the first time you know that
rier to medication safety rather than a facilitator. Nearly all partici- you’ve made a mistake and they could take that further
pants mentioned that the programme used to report errors is not if they wanted to I suppose if it was something serious.It
user-friendly and does not allow for near-miss’ entries: would be the fear of having to go and speak to the man-
ager about the fact that you’ve made an error. Nobody
I think near misses probably don’t get as reported as likes to think they’re making errors because we do put
much as they should, as I said time, if you’ve been run- patients first, so if you made an error you’d feel a) you’d
ning around all day, over a near miss situation, by the feel guilty, you’d want to make sure that person was
time you get back to the office, you’re late already, you’re okay and then there’s the fear of yeah, having to face
working overtime, you might not be approved for over- the manager.Look it’s certainly not something I think
time, so you’re just working for free. You’ve got to do all about 24/7 but it’s always on the back of my mind and
your notes and everything, and then it’s like, oh I’ve got if it, an error did occur who, who would be in trou-
to do a RiskMan, some people probably just don’t do it, ble.Participants felt that the process of reporting was a
you probably would just skip over it, it didn’t happen, it scary one, when instead it should be treated as a learn-
didn’t get to the patient, there wasn’t really an error, it ing opportunity to try and help staff develop. One par-
just was a pain in the arse.Not user friendly – I don’t like ticipant commented that the process of reporting was
Riskman. I find it very hard to choose an appropriate viewed as’victimising’. She went on to discuss that the
option a lot of the time and you find yourself repeating process was not discussed openly and this causes the
the same information like through it. I just find it – there fear.It is so closed and doesn’t get brought up to every-
should be a simpler process – more time friendly because one and names and things don’t need to be relevant, it
you’ve got to jump on that thing and do a thing and at just can be an education. I think if that was more preva-
the end of the day you’re just like oh god because you lent, perhaps that’s stigma might go away a little bit,
know you are going to be there for half an hour doing it. but definitely you’ve got to go sit with a manager and
half the time you don’t even know what it’s about
When the reporting process was discussed with participants that because you’ve just made a mistake and you might not
the process to of reporting stops once an error had been uploaded have picked up on it.
on the system, they do not receive feedback from senior staff and
many felt they had no indication of how many errors had occurred
in their department:
4.4 | Strategies for improvement
But no, right now I wouldn’t have a clue how many Many participants suggested a few strategies to improve medication
errors there’ve been in the last six months or not. I safety in community nursing. These included the development of
wouldn’t have a clue. clear guidelines detailing nurses’ roles and responsibilities regarding
medication administration in the community. The guidelines would
Reporting was mentioned by some participants as a mechanism act as support for their roles and a legal document. Participants
of protecting themselves from litigation and an obligation by the emphasised that they need to be involved in the process of establish-
workplace to improve medication safety practices although the cul- ing guidelines so they can contribute to the practicalities of them:
ture surrounding reporting was an unsupported one:
Yes having a massive discussion whether it’s a guideline,
A lot of people can get a bit narky if someone picks up pointers put out to them and then we discuss it and do
someone else’s medication error but at the end of the the pros and cons and will this work, won’t it work. I
day it’s got to be reported for the safety of the client. think we’re willing to try, we’re always trying different
So the cultures probably not the greatest in that aspect. things, but it gets frustrating when we get told to do it
one way and you just go, it’s not even practical, it’s
Participants suggested that there was a huge amount of fear really not. So I think we try our best but there’s defi-
associated with reporting. Reasons for this included the stigma and nitely room for improvement.
KHALIL AND LEE | 2079

Another strategy highlighted by the participants was a process to for the benefit of medication administration aids to improve patients’
ensure that appropriate documentation of medications and medical adherence to their medications.” The results were confounded by
history was in a patients’ file. the quality of the evidence and the small number of studies included
in the review.
Participants reported both individual and systemwide factors as
5 | DISCUSSION causes for medication errors. Systemwide factors included the use of
prepackaged medication, documentation procedures, logistical chal-
Medication use is a complex process that includes medication pre- lenges involved in community care, lack of clear community-specific
scribing, order processing, dispensing, administration and medication guidelines and inaccessible training opportunities for staff. On an
monitoring. Any failure in this process will present a compromise in individual level, participants reported a lack of sense of responsibility
the medication safety process. was a cause for medication errors and also reported high workloads.
The study identified that nurses working in a community setting A study by Hartnell et al. (2012) found that both individual and sys-
are faced with a unique set of challenges as summarised in Table 4. temwide factors are major contributors to medications errors. Exam-
First, there are key culture differences both between and within a ples of these include overworked nurses and lack of appropriate
community setting and the clinicians who work within it. The culture documentation. This is also consistent with the current study, partici-
differences found in this study included the culture of working in a pants discussed that a lack of time with each patient and high work-
community setting compared to a hospital setting, one challenge loads contributed to medication errors. Another consequence of high
found was the lack of access, challenging nature of communicating workloads was the lack of motivation to complete an incident report
with other health professionals such as GP’s and the challenges at the end of working day.
involved in trying to contact and communicate with other health Participants also expressed the biggest challenges in community
professional, within a community setting they are not as available as work and contributor of medication errors were both the logistical
onsite doctors at a hospital. Loeb, Bayliss, Candrian, and Binswanger challenges (navigating between geographical areas to access pharma-
(2016) also found that local system factors such as insufficient clini- cists and GP’s) and the lack of understanding from other healthcare
cal support, challenges communicating with other healthcare profes- professionals, in particular general practitioners of the nurses’ role
sionals and productivity demands are all major contributors to and responsibilities. The lack of clarity around who has legal respon-
medication errors in the community. sibility if a medication error was to occur caused distrust. The impor-
Participants believed that the tensions within primary care con- tance of trust between healthcare providers was noted by Bradley
tributed to an unsupportive culture. A review by McInnes, Peters, et al. (2008) who described trust as a fundamental component to
Bonney, and Halcomb (2015) found that confusion around nurse’s achieve collaborative working among healthcare providers. Partici-
scope of practice, hierarchical structures, territorialism, medico-legal pants in this study suggested that there would often be errors in the
obligations and poor communications are all barriers to working in prescribing of medication (i.e., incorrect dose or missing information),
primary care and can create a sense of hierarchical constraints at the they would struggle to contact GP’s to discuss changes in prescrip-
workplace. tions, and there was no clarity of who held legal responsibility if an
It was noted that the use of medication administration aids was error was to occur. This highlighted the challenges of community
not favoured by community nurses as they led to errors in some work and the difficulty with contacting other health professionals
instances and lack of knowledge regarding identification of what such as GPs and pharmacist, as they are not on hand. Nurses
medications are packed inside them. A review by Paterson, Kinnear, believed that they would be made culpable for any problems that
Bond, & McKinstry (2017), suggested that “there is limited evidence may arise as a result of a dispensing errors, this caused a lot of ten-
sion within the nurses and GP relationship and a lack of trust. Taran
T A B L E 4 Summary of findings (2011) highlighted that effective communication is essential to
patient care and that clear organisational policies are important in
Summary of results
clarifying the line of communications between the various health
• Many barriers and challenges hinder medication safety in commu-
professionals.
nity nursing.
Participants also discussed that they believed major sources of
• Culture differences and politics in the workplace create challenges,
which may contribute to the occurrence of medication errors in errors stem from the lack of lack of clarity or staff awareness of the
community nursing organisations. processes and procedures for staff within the organisation experi-
• Eight different causes of medication errors were identified in this ence in regard to medication errors but that they were unable to
study
access onsite. Furthermore, education to upskill and often study
• Barriers to nonreporting of medication errors include the compli-
leave were not approved due to staff shortages. Lack of education
cated reporting process, delayed feedback, defensives and fear.
• Nurses identified strategies that could reduce medication errors and regarding medication safety, reporting and the dilemmas associated
increase reporting of errors, these include development of commu- with reporting and documentation are also significant barriers. Loeb
nity-specific guidelines and improving documentation procedures et al. (2016) also found that local system factors such as insufficient
within the organisation. clinical support, challenges communicating with other healthcare
2080 | KHALIL AND LEE

professionals and productivity demands are all major contributors to lack of open culture in regard to incident reporting have been dis-
medication errors in the community. Participants discussed that they cussed in depth in previous literature (e.g., Phipps et al., 2009). It can
were expected to travel into metropolitan areas for training courses be argued that comments made by participants in this study are con-
and believed that community work did not offer the same clinical sistent with findings of previous studies and support the concept that
support that a hospital setting did. fear/blame culture in the workplace creates a barrier to the reporting
This study identified a range of challenges nurses face when of medication errors and learning from adverse events in health care.
working in the community and the barriers and causes to nonreport- Staff do not want to cause conflict and are fearful of the backlash and
ing medication errors. Participants reported that the process is time- consequences and thus engage in nonreporting. Moreover, the study
consuming and not user-friendly; thus, clinicians would often not suggests the usability of reporting mechanisms is questionable and
report to avoid having to use the system. Hartnell et al. (2012) sta- needs to be served as an educational tool for the nurses involved.
ted similar findings that reporter burden was a barrier, the process Our study notes that the issues surrounding medication error
itself of reporting, the extra time and work it required and the unfa- reporting are multifactorial, complex and often institution-specific.
vourable characteristics of the incident report form all created a bar- Participants did however note changes to improve the current sys-
rier to reporting. Participants additionally expressed that poor tem of medication safety and the reporting of medication errors in
documentation led to medication errors. Examples of this included, the primary care, specifically, clear guidelines detailing nurses’ roles
incorrect orders left in charts, poor handwriting and both electronic and responsibilities regarding medication administration in the com-
and paper documentation that was not consistent. This was again munity. The guidelines will act as support for their roles and a legal
reflected in Hartnell’s study, that factors such as procedures and document, should any unexpected error occur in the community
management were contributors to medication errors. The study resulting in patients’ harm.
found that reporting should be made as easy as possible (forms Finally, the findings in this study reflect the factors discussed in the
should be accessible and straightforward). review of the literature addressing patient safety in primary care pre-
Participants also felt they would be more motivated to report pared for the Australian Commission on Safety and Quality in Health
errors if they received timely and honest feedback. Loeb et al. Care. It was found that a number of factors are necessary to ensure
(2016) study reinforced the importance of people receiving up-to- patient safety. The authors cited that adequate systems for reporting
date education about all aspects of the medication error reporting errors needed to be organised on a national scale and any implementa-
process at their institution. tion methods to improve patient safety needed to come from the
Defensiveness and fear were key contributors to nonreporting for ground up to avoid any risks identified through reporting. Other recom-
primary care nurses in this study. A study by Phipps, Noyce, Parker, mendations including education of health professionals, system aids to
and Ashcroft (2009) addressed the challenges facing pharmacists in increase efficiency such as computers, improving communication, com-
the community regarding medication safety. The authors cited that munity-specific guidelines in medication administration and more care
one barrier to reporting is the consequences of reporting, such as co-ordination across the various care providers. The authors have also
medico-legal consequences (Phipps et al., 2009). This fear was highlighted that blame and litigation are detrimental to advancing
reflected in the current study, the worry that accompanies reporting patient safety (Khalil & Lee, 2018; Pearson & Aromataris, 2009).
and the fear of the consequences, thus resulting in nurses not report-
ing some errors through fear of losing their registration and backlash
from colleagues and managers. There are similarities between both 6 | STUDY LIMITATION
professions as they both work autonomously with limited supervision
and are required to establish their own responsibility for risk. Findings may be limited as the sample was confined to a small group
The nurses however were aware of possible consequences of of community nurses working in rural Victoria, Australia. The
not reporting. A study by Birdwood and Kainer (2015), suggested researchers also acknowledge that the sampling method and the fact
that nurses are required to manage risk by following professional that the study consisted of all female nurses and therefore male
protocols prioritising patient care and professional accountability, nurses had no representation in the sample. A further limitation to
which does not necessarily include unpredictable and dangerous cir- the study is the nurses interviewed were recruited by managers,
cumstances in their everyday practice. This is because professional which may have led to coercion.
protocols cannot anticipate every eventuality in clinical practice cre-
ating the fear of professional discipline for nurses.
Phipps et al. (2009) describe a key barrier for pharmacists in the 7 | CONCLUSION
reporting of errors was moral conflict. Within the current study, this
was reported frequently as defensiveness and fear, nurses felt they Our study identified many challenges for medication safety in pri-
faced a moral dilemma between doing the right thing by the patient, mary care. These include culture differences between community
staying within legal boundaries as well as trying to avoid staff conflict. and hospital setting, politics within the healthcare system, lack of
Participants did not feel that there was an open culture in discussing clarity around the nurses’ roles and responsibilities and error report-
incidents and rather there was a fear culture. The consequences of a ing. Any measures to improve medication safety in primary care
KHALIL AND LEE | 2081

need to focus on multifaceted interventions addressing more than Bradley, F., Elvey, R., Ashcroft, D. M., Hassell, K., Kendall, J., Sibbald, B.,
one barrier as most of the barriers identified are interrelated. & Noyce, P. (2008). The challenge of integrating community pharma-
cists into the primary health care team: A case study of local pharma-
ceutical services (LPS) pilots and interprofessional collaboration.
Journal of Interprofessional Care, 22(4), 387–398.
8 | RELEVANCE TO CLINICAL PRACTICE Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology.
Qualitative Research in Psychology, 3(2), 77–101.
Cheragi, M. A., Manoocheri, H., Mohammadnejad, E., & Ehsani, S. R.
The results from this study highlighted several issues in community
(2013). Types and causes of medication errors from nurse’s view-
nursing affecting medication safety. These were the culture differ- point. Iranian Journal of Nursing and Midwifery Research, 18(3), 228.
ence between hospital and community especially when it comes to Department of Health (2000). An organisation with a memory. London:
language use and staff expectation around medication administration HMSO.
Hartnell, N., MacKinnon, N., Sketris, I., & Fleming, M. (2012). Identifying,
and supervision, the causes of medication errors especially due to
understanding and overcoming barriers to medication error reporting in
the complacence of staff, tremendous fear of backlash due to report- hospitals: A focus group study. BMJ Quality & Safety, 21(5), 361–368.
ing and finally the strategies for improvement, which address clarity Khalil, H., & Lee, S. (2018). The implementation of a successful medica-
around roles and responsibilities using guidelines. tion safety program in a primary care. Journal of Evaluation in Clinical
Practice, 24(2), 403–407. https://doi.org/10.1111/jep.12870
Strategies to overcome the reporting dilemma would be to pro-
Kostopoulou, O., & Delaney, B. (2007). Confidential reporting of patient
vide clear guidance about reporting procedure and to ensure a
safety events in primary care: Results from a multilevel classification
friendly approach to reporting errors that encompass education and of cognitive and system factors. Quality and Safety in Health Care, 16
improving knowledge. (2), 95–100.
Furthermore, having appropriate medication safety initiatives Kuo, G. M., Phillips, R. L., Graham, D., & Hickner, J. M. (2008). Medica-
tion errors reported by US family physicians and their office staff.
within healthcare organisations are essential in clarifying the issues
Quality and Safety in Health Care, 17(4), 286–290.
raised by the participants, and also, they would serve as support Lainer, M., Vo € gele, A., Wensing, M., & So € nnichsen, A. (2015). Improving
documents in cases of litigation. Examples of these initiatives would medication safety in primary care. A review and consensus procedure
include medication safety guidelines, medication safety committees by the LINNEAUS collaboration on patient safety in primary care.
European Journal of General Practice, 21(sup1), 14–18.
and clear policies from management.
Loeb, D. F., Bayliss, E. A., Candrian, C., & Binswanger, I. A. (2016). Pri-
mary care providers’ experiences caring for complex patients in pri-
mary care: A qualitative study. BMC Family Practice, 17(1), 34.
ACKNOWLEDGEMENT
McInnes, S., Peters, K., Bonney, A., & Halcomb, E. (2015). An integrative
review of facilitators and barriers influencing collaboration and team-
The authors would like to thank the participants for their time in
work between general practitioners and nurses working in general
taking part of this study. practice. Journal of advanced nursing, 71(9), 1973–1985.
Morse, J. M. (1991). Approaches to qualitative-quantitative methodologi-
cal triangulation. Nursing research, 40(2), 120–123.
CONFLICT OF INTEREST Nguyen, E. E., Connolly, P. M., & Wong, V. (2010). Medication safety ini-
tiative in reducing medication errors. Journal of nursing care quality,
None to declare. 25(3), 224–230.
Paliadelis, P. S., Parmenter, G., Parker, V., Giles, M., & Higgins, I. (2012).
The challenges confronting clinicians in rural acute care settings: A
CONTRIBUTIONS participatory research project. Rural and Remote Health, 12(2), 1–12.
Paterson, M., Kinnear, M., Bond, C., & McKinstry, B. (2017). A systematic
Study design: HK, data collection and analysis: HK, SL; and manu-
review of electronic multi-compartment medication devices with
script preparation: HK, SL. reminder systems for improving adherence to self-administered medi-
cations. International Journal of Pharmacy Practice, 25(3), 185–194.
Patton, M. Q. (1990). Qualitative evaluation and research methods. New-
ORCID bury Park, CA: Sage.
Pearson, A., & Aromataris, E. (2009). Patient safety in primary healthcare:
Hanan Khalil http://orcid.org/0000-0002-3302-2009 A review of the literature. Adelaide, SA: Australian Commission on
Safety and Quality in Health Care.
Phipps, D. L., Noyce, P. R., Parker, D., & Ashcroft, D. M. (2009). Medica-
tion safety in community pharmacy: A qualitative study of the
REFERENCES
sociotechnical context. BMC Health Services Research, 9(1), 158.
Andel, C., Davidow, S. L., Hollander, M., & Moreno, D. A. (2012). The Reason, J. (1995). Understanding adverse events: Human factors. Quality
economics of health care quality and medical errors. Journal of Health and Safety in Health Care, 4(2), 80–89.
Care Finance, 39(1), 39. Roshanov, P. S., Fernandes, N., Wilczynski, J. M., Hemens, B. J., You, J.
Barbour, R. S. (2001). Checklists for improving rigour in qualitative J., Handler, S. M., . . . Garg, A. X. (2013). Features of effective com-
research: A case of the tail wagging the dog? British Medical Journal, puterised clinical decision support systems: Meta-regression of 162
322(7294), 1115. randomised trials. BMJ, 346, f657.
Beardwood, B. A., & Kainer, J. M. (2015). Exploring risk in professional Rosser, W., Dovey, S., Bordman, R., White, D., Crighton, E., & Drummond,
nursing practice: An analysis of work refusal and professional risk. N. (2005). Medical errors in primary care: results of an international
Nursing Inquiry, 22(1), 50–63. study of family practice. Canadian Family Physician, 51(3), 386–387.
2082 | KHALIL AND LEE

Sandelowski, M. (1995). Sample size in qualitative research. Research in Wallace, E., Lowry, J., Smith, S. M., & Fahey, T. (2013). The epidemiology
nursing & health, 18(2), 179–183. of malpractice claims in primary care: A systematic review. British
Smith, P. C., Araya-Guerra, R., Bublitz, C., Parnes, B., Dickinson, L. M., Medical Journal Open, 3(7), e002929.
Van Vorst, R., & Pace, W. D. (2005). Missing clinical information dur- Wong, A. K. S., Ong, S. F., Matchar, D. B., Lie, D., Ng, R., Yoon, K. E., &
ing primary care visits. JAMA, 293(5), 565–571. Wong, C. H. (2017). Complexities and challenges of Singapore nurses
Stuijt, C., Klopotowska, J. E., Driel, C. K., Le, N., Binnekade, J., Kleij, B., & providing postacute home care in multicultural communities: A
Bemt, P. (2013). Improving medication administration in nursing grounded theory study. Journal of Transcultural Nursing. https://doi.
home residents with swallowing difficulties: Sustainability of the org/10.1177/1043659617736884.
effect of a multifaceted medication safety programme. Pharmacoepi-
demiology and drug safety, 22(4), 423–429.
Taran, S. (2011). An examination of the factors contributing to poor com-
munication outside the physician-patient sphere. McGill Journal of How to cite this article: Khalil H, Lee S. Medication safety
Medicine, 13(1), 86. challenges in primary care: Nurses’ perspective. J Clin Nurs.
Tsang, C., Bottle, A., Majeed, A., & Aylin, P. (2013). Adverse events
2018;27:2072–2082. https://doi.org/10.1111/jocn.14353
recorded in English primary care: Observational study using the Gen-
eral Practice Research Database. British Journal of General Practice, 63
(613), e534–e542.