HEALTH POLICY AND SYSTEMS

Barriers for Nurses to Safe Medication Managementin Nursing HomesTinne Dilles, BN, RN, MScN1, Monique M. Elseviers, MS, PhD2, Bart Van Rompaey, BN, MS, PhD3,Lucas M. Van Bortel, MD, PhD4, & Robert R. Vander Stichele, MD, PhD5

1 Junior researcher, assistant lector scientific research, University of Antwerp, department of nursing science, Antwerp, Belgium and Ghent University,Heymans Institute of Pharmacology, Ghent, Belgium2 Senior researcher, Lector Scientific Research, University of Antwerp, Department of Nursing Science, Antwerp, Belgium3 Rho Chi at large, Senior researcher, lector scientific research, University of Antwerp, Department of Nursing Science, Artesis University College ofAntwerp, Department of Health Sciences, Belgium4 Senior researcher, Head of the Unit Clinical Pharmacology, University of Ghent, Heymans Institute of Pharmacology, Ghent, Belgium5 Senior Researcher, Lector Clinical Pharmacology, University of Ghent, Heymans Institute of Pharmacology, Ghent, Belgium

Key wordsMedication management, safety management,

drug-related problem, role of nurses

CorrespondenceTinne Dilles, University of Antwerp, Department

of Nursing Science, Universiteitsplein 1 DR3.32,

B-2610 Wilrijk, Antwerp, Belgium or Ghent

University, Heymans Institute of Pharmacology,

De Pintenlaan 185, B-9000, Ghent, Beligum.

E-mail: Tinne.Dilles@ua.ac.be

Accepted December 19, 2010.

doi: 10.1111/j.1547-5069.2011.01386.x

Abstract

Purpose: This study aims to identify and compare the relevance of barri-ers that nurses in nursing homes experience in medication management inBelgium.Design: The mixed-method study started with an expert meeting in November2008 and was followed by a cross-sectional survey in February–March 2009,questioning 246 nurses and 270 nurse assistants in 20 nursing homes.Methods: Twelve nurses represented nursing homes in an expert meeting andlisted all barriers that might cause suboptimal medication management. Basedon the results, a survey was developed in which respondents could indicatewhether they were involved in a particular stage of the medication processand if yes, rate the relevance of the barriers in that stage on a continuousscale, varying from 1 = no barrier to 10 = strong barrier. Barriers scored 7 ormore were defined as strong.Findings: Nurses experienced a large number of barriers to safe medicationmanagement related to the nurse, organization, interdisciplinary cooperation,or to the patient and family. In preparing medication, medication administra-tion and monitoring, being interrupted, not knowing enough on interactions,and barriers in interdisciplinary cooperation caused the most hindrance. Ingeneral, barriers in medication monitoring scored the strongest.Conclusions: In order to improve safe medication management by tailoredinterventions in nursing homes, through the use of a standard questionnaire,nurses and nurse assistants can give an overview of barriers they experienceand rate their relevance. Nurses and nurse assistants had different opinionson the relevance of barriers, especially in the stage of medication monitoring.Job expectations in medication management were not always clear, creatingadditional barriers in medication safety.Clinical Relevance: This study provides an overview of potential barriers tosafe medication management in nursing homes, which can be addressed inpractice. The relevance scoring of the barriers enables prioritization.

Medication safety research is necessary to improve thequality of medication management and the preventionand detection of adverse drug events in the growing nurs-ing home population. In 1994, the World Health Organ-

isation defined an adverse drug event as any untowardmedical occurrence while a patient is taking medicationbut there does not necessarily have to be a causal rela-tionship with the treatment.

Journal of Nursing Scholarship, 2011; 43:2, 171–180. 171C© 2011 Sigma Theta Tau International

Barriers in Medication Safety Dilles et al.

Adverse drug events are common in nursing homes,and nursing home residents are vulnerable to such eventsdue to a high incidence of polypharmacy and changedpharmacokinetics and pharmacodynamics (Turnheim,2003). The latter issues refer to age-related changes in thefunctions and composition of the human body, which re-quire adjustments of medication selection and dosage forelderly individuals. Elseviers, Vander Stichele, and VanBortel (2010) recently described drug utilization in Bel-gian nursing homes, as well as the characteristics of theresidents and the institutions. They found that the aver-age number of chronic medications prescribed per nurs-ing home resident was 8, with some residents taking upto 20 different medications. Incidence rates of adversedrug events range from 1.19 to 7.26 per 100 resident-months (Handler, Wright, Ruby, & Hanlon, 2006). Someadverse drug events are preventable and are caused byerrors. Medication errors are those occurring in the med-ication use process that can result in preventable adversedrug reactions, adverse drug withdrawal events, or ther-apeutic failure (Handler et al., 2006).

Medication management is complex; errors can oc-cur in all stages of the process (Figure 1) and differentprofessionals can be involved (physicians, pharmacists,nurses, and nurse assistants). In the system approach, er-rors are considered a consequence of system failures– thatis, weaknesses in organizational processes or work con-ditions and a lack of checks and balances (Kohn, Corri-gan, & Donaldson, 2000; Reason, 2000; Reason, Carthey,& de Leval, 2001). This means that different barriersthroughout the medication management process resultin errors that remain undetected prior to a patient takingmedication.

Barriers in the system that undermine safe medicationmanagement should be identified, and several reviewshave addressed factors contributing to medication errors.In a literature review, O’Shea (1999) initially describedpotential factors related to errors in medication admin-istration, upon which Armitage and Knapman (2003)expanded with reports published until 2003. In 2006,McBride-Henry and Foureur added more factors and di-vided them in system issues and personal issues. Systemissues consisted of a lack of adequate staffing, patient acu-ity levels, access to policy and medication information,physical environment, organizational culture, organiza-tional communication channels, organizational routines,pharmaceutical-related issues, and the incident-reportingculture. Personal issues were stated as the understand-ing of how errors occur, failure to adhere to policy andprocedure documents, number of hours on shift, distrac-tions, lack of knowledge about medications and dosagecalculations, workload, and the care delivery model

Figure 1. Themedicationprocess.Adaptedand translated fromDeClerq,

E., De Clerq, T., & Reynhout, D. (2008). De rol van de verpleegkundige in

farmacotherapie. Psychiatrie en verpleging, 84(1), 17–23.

172

Dilles et al. Barriers in Medication Safety

(Bride-Henry & Foureur, 2006). In 2009, Brady, Malone,and Fleming added some new factors, such as medicationreconciliation and medication distribution systems.

Reports comparing the importance of factors withinmedication administration are rare. Mayo and Duncanexplored nurses’ opinions in multiple settings and askedthem to rank 10 possible causes of medication errors.Physicians’ illegible or difficult-to-read handwriting wasperceived as the most frequent cause of medication er-rors, followed by distraction of nurses by other patients,coworkers, or events, and tiredness and exhaustion of thenurses (Mayo & Duncan, 2004). In the study of Tang,Sheu, Yu, Wei, and Chen (2005), hospital nurses wereasked to recall an error in which they had been person-ally involved. After a description of the error by the nurse,contributing factors were selected out of a list of 40 pos-sibilities. The need to solve other problems during med-ication administration was the most commonly selectedcause, and personal neglect contributed to medication er-rors twice as frequently as other categories.

Nurses play a key role in the medication process and inmedication safety. Therefore, in order to increase medica-tion safety, nurses’ opinions on barriers that impede safecare are very valuable. Nurses can locate important sys-tem weaknesses, allowing tailored interventions (Bride-Henry & Foureur, 2007). Furthermore, the contributionof nurses to medication safety projects– addressing prob-lems they experience– facilitates implementation later on(Weckman & Janzen, 2009).

Research on medication management in nursinghomes is uncommon. Moreover, most studies are lim-ited to the stage of medication administration and do notaddress other stages of the medication process, such asmonitoring the effects of medication. In order to improvethe safety of nurses’ medication management in nurs-ing homes, causes and risk factors for medication errorsand suboptimal medication management in this settingneed to be identified. Thus, the aim of our study was toidentify the barriers that nurses in nursing homes expe-rience in medication management and to compare theirimportance as perceived by nurses. Barriers were not re-stricted to those causing an error because the sum ofmany smaller barriers could also cause error or a barriermay cause intermittent errors. Furthermore, not only er-rors but also suboptimal care caused by barriers deservesattention.

Methods

The mixed-method study started with an expert meet-ing in November 2008. This was followed by a cross-sectional survey in February–March 2009.

Expert Meeting

An expert meeting was organized to obtain anoverview of barriers that nurses in nursing homescan experience in medication management. Twenty-five institutions in the province of Antwerp were se-lected from a list of Belgian nursing homes. Privateor public nursing homes of all sizes with both restbeds and high-intensity care beds were eligible. Theselection of nursing homes was accomplished throughconvenience sampling, taking into account the cardriving range because participation was voluntary. Inaddition, to avoid selection bias, the researcher whochose the nursing homes from the list did not know theinstitutions.

A letter was sent to the selected nursing homes ask-ing them to delegate one or two nurses or head nursesfor the expert meeting who could represent nursing staff.To further enable delegation, the topics to be discussedwere attached to the invitation. Attending nurses couldobserve practice, ask colleagues, or initiate team discus-sion in preparation for the meeting.

During the expert meeting, participants were dividedinto three groups to brainstorm barriers that might causesuboptimal medication management. Nurses from thesame institution were split up. The main question askedwas, “Which barriers that you and your colleagues ex-perience in nursing care increase the risk for incomplete,careless, or erroneous medication management?” In eachgroup, a moderator guided the discussion with referenceto the medication process scheme (see Figure 1), ensur-ing that each part of the process was addressed and thatall participants could share their experiences. All barri-ers mentioned were written down, and consensus wasnot required. Afterward, the results were presented forthe whole group, leaving the opportunity for discussionor remarks and decreasing the risk for wrong interpre-tations or incomplete results. These final presentationsand additions were audio-taped with the consent of theparticipants.

Analysis started with allocating the brainstorming re-port data to the concordant phase of the medica-tion process. Thereafter, data were thematically ana-lyzed by the first author. The results of the analysiswere compared with the audio-taped data to correctany wrong interpretations or incomplete results. A fi-nal check was performed by presenting the results ofthe analysis to the moderators of the groups, with thefinding that there appeared to be no misinterpretationsor uncertainties in the results requiring further inves-tigation. (The complete report of the qualitative partis in preparation for publication in the Dutch journalVerpleegkunde.)

173

Barriers in Medication Safety Dilles et al.

Cross-Sectional Survey

Based on the results of the expert meeting, a surveywas developed to invite a large number of nursing homenurses to appraise the relevance of the barriers. The goalwas to receive 200 surveys completed by nurses from 15to 20 nursing homes (3 to 4 per Flemish province). Insti-tutions were selected at random, stratified per province,from an official list of Belgian nursing homes. Institu-tions with fewer than 60 beds or without nursing beds(i.e., only rest beds) were excluded. In January 2009, sixnursing homes per province were selected and contacted.In the following month, three more nursing homes wereselected in provinces in which the goal had not beenreached. In addition to nurses, we also invited nurse assis-tants to participate because of their involvement in med-ication management in nursing homes. Since 2006, inBelgium, nurses can entrust nurse assistants with admin-istering oral medications if prepared and personalized bya nurse, a pharmacist, or a distribution system. Nurseshave to supervise the administration and be available forinformation and support. Nurse assistants have to reportto the nurses. It is outside the scope of practice in manycountries for nursing assistants to give medications so it isvery important that readers can understand your system.Subjects with less than 6 months’ experience or insuffi-cient understanding of Dutch were excluded.

To create a feasible survey, not all barriers reportedby the expert group were listed. Instead, we focused onthe three main stages of the medication process: prepar-ing medication, medication administration, and moni-toring medication effects. Three researchers selected 30barriers, taking into account improvement opportuni-ties. Participants had to indicate whether they were in-volved in a particular stage of the medication process,and if this was true, to rate the relevance of the barri-ers in that stage on a continuous scale, varying from 1(no barrier) to 10 (strong barrier; occurrence, hindrance,consequences).

Data were collected from February to March 2009.Paper questionnaires were distributed by a research as-sistant together with a cover letter explaining the pur-pose and the methods of the study and outlining howanonymity of the participants would be protected. Aftersigning informed consent and completion of the ques-tionnaire, documents were anonymously collected in aclosed box. Data were analyzed using the statistical pro-gram SPSS v. 16.0 (SPSS, Inc., Chicago, IL, USA). Theresearch population was described using the mean, stan-dard deviation, and t tests for continuous variables andproportions and chi-square tests for discrete variables.Relevance of the barriers, as indicated on a 10-point scale,was presented in box plots showing the median, range,

and interquartile range. We defined scores of 7 or moreas strong barriers. The proportions of nurses and nurseassistants scoring 7 or more were compared using chi-square tests for discrete variables. A p value < .05 wasconsidered significant.

Results

Definition of Barriers to Safe MedicationManagement by Nurses– Expert Meeting

In the expert meeting, six nursing homes participated:two private and four public. The number of beds rangedfrom 90 to 214. The six nursing teams were representedby a total of 12 nurses (4 of whom were male), ofwhom 9 held a position as head nurse or similar level.The expert meeting resulted in a large list of barriersspread over the whole medication process, from receiv-ing a prescription to the monitoring and reporting of ef-fects and side effects. The barriers could be divided intofour relational categories: nurses, interdisciplinary coop-eration, organizational culture or structure, and patientor family. The results are summarized in Figure 2. Afull report of the qualitative results will be publishedelsewhere.

Description of the Survey Respondents

A total of 246 nurses (response rate 67%) and 270nurse assistants (response rate 46%), employed in 20nursing homes (13 private and 7 public), with a meanof 104 beds (range 65–191), scored the relevance of thebarriers. All nursing homes cared for residents both withand without dementia, except for one, which was re-stricted to older people with dementia. Of the respon-dents, 48% were nurses (0.6% master’s degree, 14%bachelor’s degree, 33% diploma level) and 7% held aposition as a head nurse (Table 1). Nurse assistantsrepresented 52%.

Involvement in Medication Management

Of the nurses, 92% participated in preparing medica-tion for administration, with 62% involved at least oncea week. Of the nurse assistants, 25% participated, with13% involved at least once a week (p < .001). With re-gard to medication administration, 98% of the nurseswere involved, with 87% participating at least once aweek. The proportion of nurse assistants involved inmedication administration was 87%, with 70% partici-pating at least once a week (p < .001).

Eighty percent of the nurses believed that follow-ing the therapeutic effects of medication was part of

174

Dilles et al. Barriers in Medication Safety

Figure 2. Definition of barriers to safe medication care.

Table 1. Characteristics of the Research Population

Total Nurses Nurse assistants

Characteristic N = 516 n = 246 n = 270 p value of difference

Mean years of age (SD) 38.4 (10.2) 40.2 (9.0) 36.8 (11.0) <.001

Men (%) 7.4 3.7 11.4 .001

Full time regimen (%) 37.2 47.2 28.1 <.001

Shifts .052

Only days (%) 72.1 67.1 76.7

Only nights (%) 11.2 13.4 9.3

Mean years of experience (SD)

In health care 15.8 (10.0) 17.8 (9.2) 13.8 (10.4) <.001

In nursing home 10.5 (8.8) 10.9 (8.3) 10.1 (9.2) .300

p-values were obtained by t tests for continuous variables and by chi-square tests for discontinuous variables.

their job, and 95% felt as if it should be a part oftheir job. The percentages for the nurse assistants weremuch lower: 37% (p < .001) and 55% (p < .001),respectively.

With respect to side effects, 76% of nurses believed thatmonitoring side effects was part of their job and 92% feltit ought to be, whereas only 45% of the nurse assistants

believed that monitoring side effects was part of their job(p < .001) and 62% felt it should be part of their job de-scription (p < .001).

An overview of the involvement in medication man-agement is presented in Table 2. Only those involvedin a particular stage rated the relevance of the associatebarriers.

175

Barriers in Medication Safety Dilles et al.

Table 2. Involvement in the Medication Process

Nurses Nurse p value of

(%) assistants (%) difference

In preparation 227 (92) 67 (25) <.001

In administration 241 (98) 235 (87) <.001

In monitoring effectsa 236 (96) 172 (64) <.001

aAll respondents who believed it was a part of their job to monitor effects

ofmedication, orwho felt like it shouldbe,were considered tobe involved.

General Relevance Scoring of Barriersin the Medication Process

Box plots show the general relevance scoring of barri-ers in the medication process (Figure 3). There is a varia-tion in the relevance of barriers as experienced by nursesand nurse assistants with barriers evaluating the effects ofmedication scoring stronger compared with those in med-ication administration, which scored stronger than thosein medication preparation. Remarkably, nurses’ scoresranged from 1 to 10 for all barriers.

Relevance Scoring of Barriersin Different Stages

The proportion of respondents scoring barriers to safemedication management was highly important (7 to 10)and allowed comparison between nurses’ and nurse assis-tants’ opinions (Table 3). When investigating the scoresof the head nurses separately, there was only one signif-icant difference: head nurses scored hindrance of unclearmedication charts slightly lower. Yet scores on this fac-tor were very low in all groups, indicating it was not arelevant barrier.

Medication Preparation

The most important barrier was being interruptedwhen medications were being prepared. Over 40% of re-spondents scored relevance of the barrier as 7 or higher(median = 6). The statements “I do not have enoughknowledge on which medications can be crushed” and“Work pressure is too high to prepare medications withcare” were rated 7 or higher by about 24%. On the otherhand, many respondents did not judge the barriers inthese statements as important, resulting in a median un-der 5. The opinion of nurses and nurse assistants corre-sponded, except for knowledge on crushing medicationsand medication calculation, which were scored as a prob-lem by relatively more nurse assistants.

Medication Administration

More than 30% scored 7 or more for lack of timeto double-check medication before administration and

Figure 3. Description of the relevance of barriers to safemedication care

in nursing homes. Relevance is rated by nurses and nurse assistants on a

10-point scale with 1 = no barrier and 10 = very strong barrier. Red

Lines represent themiddle of the scale (5) and the cut off (7) used to define

strong barriers.

not knowing enough on interactions between medica-tion and food or between different medications when ad-ministered together. These factors impeded medication

176

Dilles et al. Barriers in Medication Safety

Table 3. Relevance of Barriers to Safe Medication Care in Nursing Homes and the Differences Between Nurses’ and Nurse Assistants’ Opinions

Medication preparation

Total Nurses Nurse assistants

N = 294 n = 227 n = 67

% ≥ 7 % ≥ 7 % ≥ 7 p value of differencea

Being interrupted 42.8 44.2 37.3 .340

High work pressure 24.6 24.2 25.9 .795

Insufficient knowledge on crushing 24.4 21.5 35.0 .031

Way in which drugs are prepared 14.9 14.8 15.5 .891

Unclear medication charts 8.8 8.9 8.5 .913

Difficulties to calculate doses 7.7 5.6 16.1 .009

Guidelines on the topic are unclear 3.9 4.1 3.4 .833

Medication administration

Total Nurses Nurse assistants

N = 476 n = 241 n = 235

% ≥ 7 % ≥ 7 % ≥ 7 p value of difference

Lack of time for double-checking medication 37.0 43.9 29.0 .001

Not knowing enough on interactions with food 36.0 34.9 37.3 .586

Not knowing enough about what medication can be given together 31.3 28.5 34.4 .171

Not enough guidelines of physicians on the topic 26.3 23.6 29.4 .175

Not enough communication in the team on the topic 20.2 23.1 17.0 .106

Not knowing the correct administration time 19.5 19.4 19.6 .952

(f.e. before, during or after a meal)

High work pressure 18.0 18.5 17.4 .769

Lack of time to check the actual intake 18.1 17.6 18.7 .774

Shortage of sources of information on the topic 16.9 16.3 17.5 .734

Not knowing enough on the usage of inhalation devices 14.0 8.8 19.7 .001

Difficulties to evaluate selfcare abilities of residents 9.7 8.8 10.7 .492

Medication monitoring

Total Nurses Nurse assistants

N = 408 n = 236 n = 172

% ≥ 7 % ≥ 7 % ≥ 7 p value of difference

Not enough information of the physicians 40.5 42.4 37.9 .363

Not enough interdisciplinary communication on evaluating side-effects 28.4 37.4 15.3 ≤.001

Not enough attention to reporting observations 28.2 34.2 20.0 .002

Not knowing enough on side-effects 35.5 33.5 38.4 .307

Not enough interdisciplinary communication on evaluating therapeutic effects 26.6 32.2 18.4 .002

Lack of time to perform the task with care 29.7 30.5 28.5 .662

Difficulties in communicating with physicians 23.5 22.7 24.6 .675

Limited accessibility of physicians 17.9 19.6 15.6 .302

Not feeling responsible for the evaluation of the effects 16.0 18.4 12.9 .136

Not knowing enough on therapeutic effects 24.9 16.9 36.1 ≤.001

Limited accessibility of pharmacists 10.0 8.5 12.0 .244

ap-value of difference between nurses and nurse assistants.

administration the most. Furthermore, about 26%thought that more physician instructions on medicationadministration would be an improvement, and about20% experienced a barrier in a lack of communication onthis topic within the team of nurses and nurse assistants.

The lack of time to check medications a second timewas evaluated as more important by nurses comparedwith nurse assistants. Significantly more nurse assistantsexperienced a lack of knowledge on the usage of inhala-tion devices as a relevant barrier.

177

Barriers in Medication Safety Dilles et al.

Monitoring of Medication Effects

In contrast to the first two categories, with regardto the evaluation of effects, opinions of nurses andnurse assistants on the importance of barriers wereclearly different. Nurses rated barriers on interdisciplinarycommunication, support by physicians’ information, andattention to reporting and observing effects as the mostimportant. Nurse assistants rated the barriers on inter-disciplinary communication and attention to observingand reporting lower, but rated lack of knowledge on boththerapeutic effects and side effects more highly. However,34% of nurses also rated their knowledge of side effectsas a barrier as 7 or more.

Discussion

Throughout the whole medication process, nurses ex-perienced a large number of barriers to safe medicationmanagement related to the nurse, the organization, in-terdisciplinary cooperation, or the patient and family. Ofthe 30 barriers rated for relevance, 15 were scored ashighly relevant (7 to 10) by over 20% of nurses, in-dicating that these barriers were experienced as majorhindrances. Thus, should we still be surprised about therisk of errors? Applying the system approach (Reason,2000; Reason et al., 2001), we need to break down thesebarriers to increase medication safety. Nurses and nurseassistants can help identify barriers and prioritize ac-tions, taking into account their legal practice, educa-tional preparation, and actual involvement in medicationmanagement.

Nurses’ Involvement in MedicationManagement

Nurses are not always sure how far their responsi-bilities reach. In our study, some doubts were voicedby nurses on the responsibility for monitoring effects ofmedication. While most felt like as if it should be part oftheir job, many thought it should not be at this moment.The situation was even worse with regard to monitoringside effects: A quarter of the nurses believed it was notpart of their job. Those who believed it was not part oftheir job will probably not make efforts to monitor medi-cation side effects.

Nurse Assistants’ Involvement in MedicationManagement

Because of differences in educational background, legalobligations, and practical job expectations, involvementin stages of the medication process differed significantly

between nurses and nurse assistants. However, differen-tiation in tasks was not always strong enough as 25% ofthe nurse assistants participated in preparing medicationfor administration, a task they are not legally allowed todo. Legally, in Belgium, nurse assistants can help nursesregarding patient care in activities coordinated by a nursein a structured team. However, nurse assistants can onlycarry out nurse activities with which they are entrusted,and this is a limited list. Listed activities in medicationmanagement include helping the patient take oral medi-cation after it has been prepared and personalized by anautomatic dispensing system, a nurse, or a pharmacist,and informing and advising the patient and family aboutthis activity. Nurse assistants are also obligated to reportto the nurse the same day. On the other hand, nursesare responsible for the correctness of activities and theway they are executed, and have to be available for in-formation and support. Because of a shortage of nurses innursing homes, nurse assistants perform more tasks thanthey are legally allowed to perform. Consequently, nurs-ing home administrators and nurses have to be awareof their responsibility and the potentially higher risk formedication errors when deploying professionals to per-form a task for which they are not educated.

Barriers in Medication Management

In general, barriers in monitoring the side effects ofmedication were experienced as more important thanthose in medication administration, which in turn wereexperienced as more important than in preparing medica-tion. In medication preparation, medication administra-tion and medication monitoring, being interrupted, notknowing enough on interactions, and barriers to inter-disciplinary cooperation (informing, reporting, and thefrequency of communication) were found to impede themost.

When comparing the barriers experienced by nursesand nurse assistants in the three stages of the medica-tion process that were investigated quantitatively, nurseassistants reported a higher importance for lack of knowl-edge. Nurses also indicated a higher importance of barri-ers in interdisciplinary cooperation and, although not sig-nificant, a feeling of lack of responsibility in monitoringmedication effects. Differences in educational preparationand in job responsibilities can cause these differences inopinion. In practice, nurses have more responsibility forinterdisciplinary cooperation, which likely increases theimportance of barriers on that topic for them. Further-more, professionals should be more aware of their ownand others’ professional responsibilities. Although uncer-tainty on job content was not a rated barrier, it can def-initely be considered a barrier; for instance, when one

178

Dilles et al. Barriers in Medication Safety

does not know the details of one’s own job responsibili-ties, no clear agreements on interdisciplinary cooperationcan be made.

The relevance of barriers on interdisciplinary coop-eration and work pressure was scored differently indifferent stages of the medication process. Barriers ininterdisciplinary cooperation can be influenced by the be-lief that nurses should know about medication admin-istration, but they expect more cooperation with physi-cians and pharmacists in monitoring effects. Remarkably,time barriers have been judged differently in differentstages. For example, in medication administration, a lackof time to double-check medication scored the highest,while a lack of time to check the actual intake of a medi-cation scored much lower. We suggest three possible ex-planations. Firstly, nurses might have shown in the re-sults that checking the intake is a priority for them overchecking the correctness of the medication once more,giving them enough time to check the intake, but leavingno time for double-checking the medication. Secondly,it might be a consequence of nurses’ resistance towardrigorous rules; double-checking medication might be ex-perienced as an obligation, while checking the intake isa part of direct patient care. Finally, the amount of timeneeded to perform a particular task might be important,although there appears to be no evidence for this fromthe results.

Comparison With Earlier Studies

Both Crespin et al. (2010) and Vogelsmeier, Scott-Cawiezell, and Zellmer (2007) studied barriers in nurs-ing homes. Crespin et al. searched for contributingfactors to medication errors, while Vogelsmeier et al.explored staff perceptions and concerns about the med-ication use process in interviews and focus groups. Or-dered by frequency, contributing factors were (a) hu-man errors; (b) transcription errors, (c) distractions, (d)following faulty policies and procedures, (d) poor com-munication, incorrect pharmacy dispensing; medicationunavailable; medication name (e) confusion; inadequateinformation, and (f) wrong medication delivered. In anoverview of the multiple barriers to safe medication prac-tices as perceived by nursing home staff, communicationdeficiencies were the most common theme throughoutthe whole medication use process (Vogelsmeier et al.,2007).

The focus on all barriers is wider than the focus oncontributing factors to medication errors. Although bar-riers may contribute to errors, they can also just hinderan efficient medication process or adverse drug event de-tection. In all studies, the same barriers seem to occur(including ours), but sometimes new factors are added,

contributing to the complex basis for medication errors.While our study does not add any new main factors, itincreases the sensitivity of the main factors, stemmingfrom the nurses’ own experiences. In this way, a lack ofknowledge was restricted to specific topics; for example,interactions between medication and food and communi-cation problems were split into terms of accessibility, fre-quency of interdisciplinary communications, and difficul-ties in communicating (e.g., mutual respect or language).This sensitivity can facilitate very specific, adjusted in-terventions to improve medication safety. Moreover, ourstudy combined research based on qualitative nurse ex-periences with a quantitative relevance rating on a largerscale, enabling prioritization. Communication is crucialto safe medication management. The importance of com-munication between physicians and nurses in our studycorresponds to the results of Vogelsmeier et al. (2007) innursing homes.

Strengths and Weaknesses

The results of our study correspond partially with ear-lier studies. However, our study had a specific focuson nurses’ experiences on barriers in nursing homes.Furthermore, we looked for all barriers and not retro-spectively for factors that contributed to a recorded med-ication error, and in contrast to most other studies, ourinvestigation was not limited to the stage of medicationadministration. Finally, the questioning of nurses andnurse assistants allowed comparison between groups.

A limitation of our study was the impracticality toinclude all sensitive factors extracted from the expertmeeting in a questionnaire for relevance rating; the ques-tionnaire would have been too long for respondents tocomplete accurately. Another limitation was the variationin respondents’ relevance ratings. Although the propor-tion of respondents rating 7 or higher on a 10-point scaleclearly differed, medians seldom passed 5 and opinionsalways ranged from 1 to 10, demonstrating contrasts inthe strength of barriers. This variation could not be easilyexplained. Work regimen, educational preparation, pri-vate or public institution, and function all seemed to havean influence, yet creating sum scores for separate bar-rier scores to compare groups is not statistically correctbecause separate barriers are not part of a concept to bemeasured.

Conclusions and Implications

In order to improve safe medication managementin nursing homes by tailored interventions, this studyshowed that using a standard questionnaire, nurses andnurse assistants can give an overview of barriers they

179

Barriers in Medication Safety Dilles et al.

experience and rate their relevance. Throughout thewhole medication process, nurses experienced a largenumber of barriers to safe medication management re-lated to the nurse, organization, interdisciplinary cooper-ation, or patient and family, all of which threaten patientsafety. In preparing medication, medication administra-tion and medication monitoring, being interrupted, notknowing enough on interactions, and barriers to interdis-ciplinary cooperation (informing, reporting, and the fre-quency of communication) created the most hindrance.Nurses and nurse assistants had different opinions on therelevance of barriers, especially in the stage of medica-tion monitoring. Job expectations in medication manage-ment were not always clear, creating additional barriersto medication safety.

Management teams can use a similar question-naire approach to identify barriers in their institutions.Furthermore, for the strongest barriers in our study, im-provement strategies can be investigated or developed. Inaddition, special attention is required to clarify the roleswith regard to interdisciplinary cooperation in nursinghomes. Further research would also help nursing homeadministrators to define causes of barriers and to selectand implement improvement strategies.

Acknowledgments

Special thanks to Kateleine Panneels, who participatedin the research project as a study assignment.

Clinical Resources� Institute for Safe Medication Practices: http://

www.ismp.org/� Belgian Centre for Pharmacotherapeutic Informa-

tion: http://www.bcfi.be/

References

Armitage, G., & Knapman, H. (2003). Adverse events in drug

administration: A literature review. Journal of Nursing

Management, 11, 130–140.

Brady, A. M., Malone, A. M., & Fleming, S. (2009). A

literature review of the individual and systems factors that

contribute to medication errors in nursing practice. Journal

of Nursing Management, 17, 679–697.

Crespin, D. J., Modi, A. V., Wei, D., Williams, C. E., Greene,

S. B., Pierson, S., Hansen, R. A. (2010). Repeat medication

errors in nursing homes: Contributing factors and their

association with patient harm. American Journal of Geriatric

Pharmacotherapy, 8, 258–270.

Edwards, I. R., & Biriell, C. (1994). Harmonisation in

pharmacovigilance. Drug Safety, 10(2), 93–102.

Elseviers, M. M., Vander Stichele, R. R., & Van Bortel, L. M.

(2010). Drug utilization in Belgian nursing homes: Impact

of residents’ and institutional characteristics.

Pharmacoepidemiology and Drug Safety, 19, 1041–1048.

Handler, S. M., Wright, R. M., Ruby, C. M., & Hanlon, J. T.

(2006). Epidemiology of medication-related adverse events

in nursing homes. American Journal of Geriatric

Pharmacotherapy, 4, 264–272.

Kohn, L. T., Corrigan, J. M., & Donaldson, M. S. (2000). To err

is human. Building a safer health system. Washington, DC:

Institute of Medicine.

Mayo, A. M., & Duncan, D. (2004). Nurse perceptions of

medication errors. What we need to know for patient

safety. Journal of Nursing Care Quality, 19, 209–217.

McBride-Henry, K., & Foureur, M. (2006). Medication

administration errors: Understanding the issues. Australian

Journal of Advanced Nursing, 23, 33–41.

McBride-Henry, K., & Foureur, M. (2007). A secondary care

nursing perspective on medication administration safety.

Journal of Advanced Nursing, 60, 58–66.

O’Shea, E. (1999). Factors contributing to medication errors:

A literature review. Journal of Clinical Nursing, 8, 496–504.

Reason, J. (2000). Human error: Models and management.

British Medical Journal, 320, 768–770.

Reason, J. T., Carthey, J., & de Leval, M. R. (2001).

Diagnosing “vulnerable system syndrome”: An essential

prerequisite to effective risk management. Quality in Health

Care, 10(Suppl. 2), ii21–ii25.

Tang, F., Sheu, S., Yu, S., Wei, I., & Chen, C. (2005). Nurses

relate the contributing factors involved in medication

errors. Journal of Clinical Nursing, 16(3), 447–457.

Turnheim, K. (2003). When drug therapy gets old:

Pharmacokinetics and pharmacodynamics in the elderly.

Experimental Gerontology, 38, 843–853.

Vogelsmeier, A., Scott-Cawiezell, J., & Zellmer, D. (2007).

Barriers to safe medication administration in the nursing

home– Exploring staff perceptions and concerns about the

medication use process. Journal of Gerontological Nursing, 33,

5–12.

Weckman, H., & Janzen, S. (2009). The critical nature of early

nursing involvement for introducing new technologies.

Online Journal of Issues in Nursing, 14(2). DOI:

10.3912/OJIN.Vol14No02Man02.

180