Article

Africa Journal of Nursing and Midwifery https://doi.org/10.25159/2520-5293/6957

https://upjournals.co.za/index.php/AJNM/index ISSN 2520-5293 (Online) Volume 22 | Number 2 | 2020 | #6957 | 18 pages © Unisa Press 2020

Perceived Occurrence of Medication Administration Errors among Nursing Students at a Higher Education Institution in Western Cape, South Africa

Yousef A. Abusaksaka Hilda Vember

https://orcid.org/0000-0002-8242-8953 https://orcid.org/0000-0002-8006-6266

Cape Peninsula University of Technology, Cape Peninsula University of Technology,

South Africa South Africa

[email protected] [email protected]

Regis R. Marie Modeste

https://orcid.org/0000-0003-0329-3526

Cape Peninsula University of Technology,

South Africa

[email protected]

Abstract

The aim of the study was to determine awareness and perception of trends in the

occurrence of medication administration errors (MAEs) among nursing

students. A descriptive quantitative design was employed on two consecutive

days and a self-administrative questionnaire was used to collect data. The

validity and reliability of this instrument were tested and established by a pilot

study. Responses were collected from 291 nursing students at a higher education

institution (HEI) in Western Cape, South Africa. Non-probability proportional

quota sampling was used and the data was analysed with IBM SPSS® software.

The data was presented in graphs, percentages, means and standard deviation,

while inferential statistics was applied. The findings of the study revealed that

85.2% of the respondents were aware of MAE occurrence. There was no

significant difference between the respondents’ awareness of MAE occurrence

and their year of the study. The significant (p-value < 0.05) subscale for the

causes of MAE occurrence was the physician communication subscale (p-value

< 0.001). Moreover, the respondents perceived the top item to be using

abbreviations instead of writing out the prescription orders completely (p-value

< 0.001, mean = 4.85). The respondents disagreed that the pharmacy related

subscale and its items were causes of MAE occurrence. In conclusion, the

nursing students who participated in the study were aware of MAE occurrence

Abusaksaka, Vember and Marie Modeste

2

during their practice time. The causes of these errors as indicated by the

respondents are mentioned in the article. Therefore, the healthcare institutions

as well as HEIs must focus on treating these causes in order to reduce MAE

occurrence and enhance patient safety.

Keywords: medication errors; medication administration errors; nursing students

Introduction and Background

According to the World Health Organization (WHO 2011), the risk of being harmed

during the rendering of healthcare is much greater than the risk of being harmed by air

travel or nuclear plants, which are sources usually perceived as being considerably

dangerous (Iliffe n.d.). Errors that occur during the delivering of healthcare lead to about

7 000 deaths annually in the United States (US) (WHO 2011, 243), while in the United

Kingdom these errors are the main cause of 712 deaths every year and are indicated as

a contributory factor in 1 700 to 22 300 deaths per year (Elliott et al. 2018, 4). The

number of deaths is an average based on the incidence of medication errors (MEs)

worldwide, but there are no clear statistics from the developed countries (Haw, Stubbs

and Dickens 2014, 798; WHO 2011, 243). However, it is important to stress that there

are differences between the developed and developing countries (Iliffe n.d.).

The medication administration process consists of five steps, namely: (1)

prescribing/ordering the medication; (2) transcribing and verifying the order; (3)

preparing the medication and delivering it; (4) administering the medication by the

healthcare workers; and (5) monitoring and documenting the medication. Medication

errors may occur at any time during one or more of these steps (Gordon 2014, 18) and

MEs are classified according to prescription, transcription, dispensing, and

administration errors (Radley et al. 2013, 471). An error at the administration phase is

very critical because it is directly harmful to the patient and the possibility to correct it,

is limited. Understanding how and why medication administration errors (MAEs) occur,

represents the key to the intervention that minimises MAEs (Keers et al. 2013a, 1046).

Wakefield, Uden-Holman and Wakefield (2005, 475–489) classify the causes of MAEs

into four main groups, namely, MAEs caused by factors related to pharmacies,

manufacturers, physicians, and nurses. Wakefield, Uden-Holman and Wakefield (2005,

475–489) developed a questionnaire based on the four main groups to investigate

nurses’ perceptions regarding MAEs. This questionnaire served as a baseline for further

research regarding the causes of MAEs.

Nurses are responsible for administering medications to various in- and outpatients

daily. Hence, the chance of committing errors during this process is probable (Jones and

Treiber 2010, 240). So, the prevention of MAEs and ensuring patient safety are

Abusaksaka, Vember and Marie Modeste

3

important roles of the nurse, as they are part of the medication administration process

(Weant, Bailey and Baker 2014, 47).

Pryce-Miller and Emanuel (2010, 8) argue that nursing education institutions (NEIs)

should create a learning environment where experienced lecturers and professional

registered nurses conduct workshops for undergraduate nurses on the occurrence of

MAEs. Nursing students’ awareness of MAEs and patient safety is enhanced by early

exposure to the complex nature of the medication administration process. This should

be taught and practised in the clinical skills laboratory as continuous practice assists

nurses to improve patient outcomes (Ofosu and Jarrett 2015, 12). Practical training

should be fundamental for nursing students in order to facilitate a deeper understanding

of MAEs and their impact on patient safety (Hammoudi, Ismaile and Yahya 2017,

1038). Although lecturers spend time and effort on teaching students the protocols for

safe medication administration, students still commit MAEs (Valdez, De Guzman and

Escolar-Chua 2013, 222). In the literature, registered nurses are the focus of MAEs,

while nursing students’ MAEs remain unreported (Valdez, De Guzman and Escolar-

Chua 2013, 222). Therefore, the current study was conducted among the nursing

students at an HEI in Western Cape, South Africa to determine the main factors that

influence the occurrence of MAEs in order to enhance the patients’ safety and supress

the occurrence of these errors.

Problem Statement

Lack of the knowledge of MAEs has dire consequences for patients, nurses and

healthcare institutions. It severely compromises patient safety and may lead to disability

or even death. Therefore, it is imperative that nurses receive adequate training to

administer medication correctly to patients.

Aim and Objectives of the Study The aim of the study was to determine the awareness and perception of the occurrence

and reporting of MAEs among nursing students at an HEI in Western Cape, South

Africa.

The objectives of the study were to:

• determine the level of awareness of the occurrence and reporting of MAEs;

• determine the factors related to MAE occurrence as experienced by undergraduate nursing students at an HEI in Western Cape in the units at the

healthcare facilities where undergraduate students are placed.

Abusaksaka, Vember and Marie Modeste

4

Research Methodology

Research Design

A descriptive quantitative design was used in the study that was conducted among the

nursing students on 2 and 3 April 2017. Descriptive research is aimed at casting light

on current issues or problems through a process of data collection that enables

researchers to describe the situation more completely, in this case the occurrence and

reporting of MAEs as perceived by nursing students at an HEI in Western Cape.

Research Setting

The study was carried out at an HEI in Western Cape, South Africa. The nursing

department of the HEI has three sites where the basic undergraduate training programme

is offered. The campus included in the study for data collection has the largest number

of undergraduate students compared with the other two sites, which are situated in two

rural communities.

Population and Sampling

The target population for the study was approximately 563 second-, third- and fourth-

year undergraduate nursing students registered for the undergraduate nursing degree at

an HEI in Western Cape. All the students enrolled for the study programme were

approached to participate in the study.

Non-probability proportional quota sampling was used in the study. The purpose of

quota sampling is to draw a sample that has the same proportions or characteristics as

the whole population. With this sampling method, the researcher divided the population

group into three subgroups (strata) depending on their year of study (second-, third- and

fourth-year nursing students). Quota sampling with convenience technique was then

used to collect the data to achieve the desired sample size.

Sample Size

The researcher used the Sample Size Calculator (2016) to calculate the sample size, a

confidence level of 95%, and a confidence interval of 4. Moreover, the sample size was

calculated manually. Based on that, the sample size was estimated as 291, comprising

second-, third- and fourth-year undergraduate nursing students.

Data Collection

The present study used a self-administered questionnaire for the data collection.

Maintaining the quota distribution, the students were invited to participate in the study,

and 300 consent forms were distributed. The researcher and his supervisor met with all

Abusaksaka, Vember and Marie Modeste

5

the students, and the lecturer in each class introduced the researcher to the prospective

respondents. The data was collected from the population over two days on 2 and 3 April

2017. On the first day, data was collected from the second-year nursing students, then

on the following day, data was collected from the third- and fourth-year nursing

students. Three hundred consent forms were signed and 300 questionnaires distributed

to the respondents. Nine of the returned questionnaires were excluded because they were

incomplete. However, the sample size was achieved.

Before data collection commenced, the purpose and aims of the study were explained

to the respondents. A written information sheet was also given to all respondents and

attached to the consent form. The respondents were made aware that they had the right

to withdraw from the study, even after they had provided informed consent, without

being victimised or with any negative consequences. The respondents were given 15

minutes (as determined by the pilot study) to complete the questionnaire after reading it

and having it explained by the researcher and his supervisor. All of the respondents

returned their questionnaire before the end of 15 minutes.

The questionnaires were collected after completion and coded anonymously by the

researcher as second year (2.000), third year (3.000) and fourth year (4.000). After

extraction of the data, the completed questionnaires were stored safely in a locked safe

in the nursing department at the HEI where the researcher is a registered student.

Research Instrument

The data collection instrument in the present study consisted of a self-administered

questionnaire. The questionnaire was used to determine the perceived occurrence of

MAEs among the nursing students.

The questionnaire for the study was divided into three sections in alphabetical order.

Parts A and B were developed by the researcher. Part C had been used in previous

studies and developed by Dr Bonnie Wakefield of the Sinclair School of Nursing at the

University of Missouri, US. Permission to use the questionnaire was obtained from the

developer.

Part A collected the respondents’ demographic characteristics (age, gender, marital

status and year of study) and comprised items 1, 2, 3 and 4. Part B outlined the

respondents’ awareness of the occurrence and reporting of MAEs and comprised Item

5. Part C outlined the causes of MAEs and comprised 20 items on MAE-related causes

with a 6-point Likert scale (1 = strongly disagree to 6 = strongly agree). This part of the

questionnaire was further classified into four subscales: physician communication

causes (6 items: questions 10, 11, 12, 13, 14 and 18); medication package causes (3

Abusaksaka, Vember and Marie Modeste

6

items: questions 7, 8 and 9); pharmacy related causes (3 items: questions 15, 16 and 17);

and nurse related causes (8 items: questions 19, 20, 21, 22, 23, 24, 25 and 26).

Pilot Study

A pilot study was employed in the study and conducted at one campus of the HEI under

study. The pilot study was used to test the various components of the questionnaire in

terms of the validity and reliability of the instruments to the nursing students. The pre-

testing questionnaire was conducted on one day during the data collection period. The

respondents in the pilot study included 15 second-, third- and fourth-year nursing

students who met the inclusion criteria for the study. The respondents of the pilot study

were not included in the sample size of the study.

Consent forms with the information sheet about the study were distributed to the

respondents prior to the questionnaire sheets. Agreement to participate in the pilot study

was obtained using the consent form. The researcher explained to the respondents the

purpose of the study and the contents of the informed consent. The researcher informed

the respondents that they were free to participate in the study or to withdraw at any time

even after completion of the consent form. A total of 15 questionnaires were distributed.

All the distributed questionnaires were returned. The respondents indicated that the

questionnaire was clear, easy to read and easily understood. The questionnaires took

approximately 10–15 minutes to complete. The results of the pilot study indicated that

no corrections or adjustments to the existing instrument were necessary. The

Cronbach’s alpha for the whole instrument was 0.83; 0.84 for the medication package

subscale; 0.85 for the physician communication subscale; 0.89 for the pharmacy related

subscale; and 0.81 for the nurse related subscale.

Data Analysis

The collected data was coded and tabulated. The Statistical Package for Social Sciences

(SPSS Version 24) was used to analyse the data. Microsoft Excel was utilised for

graphical presentation such as line and bar charts. The analysis was performed under

the supervision of, and with consultation and support from, the statistician at the HEI

where the research was done. In the study, descriptive and inferential statistics were

conducted to analyse the collected data.

Ethical Considerations

Permission for the study was obtained and renewed annually from the Research Ethics

Committee of the Faculty of Health and Wellness Sciences at the Cape Peninsula

University of Technology (CPUT) (reference number: CPUT/HW–REC 2016/H23). An

informed consent was obtained from each respondent. Anonymity and confidentiality

were assured. The respondents’ names were not recorded, and the data was numerically

Abusaksaka, Vember and Marie Modeste

7

coded. Only the researcher, his supervisor, co-supervisor, and the statistician had access

to the data.

Results

Demographic Characteristics

In the current study, 290 respondents indicated their gender. Most of the respondents

(81.3%, n = 236) were female. The majority of the respondents (97.9%, n = 285)

responded to the age item and the mean of their ages was 23.49, with a minimum age

of 19 years and a maximum age of 43 years. Most of the respondents (81.7%, n = 233)

were 25 years or younger. The majority of the respondents (88.7%, n = 258) reported

their status as single. The distribution of the respondents was similar over the three years

of study, with 110 (37.8%) in second year, 93 (32.0%) in third year and 88 (30.2%) in

fourth year.

Respondents’ Awareness of MAE Occurrence across Each Year of Study

The respondents’ year of study was the only demographic variable that showed a

difference with the respondents’ awareness of MAE occurrence. As the study related to

the nursing students’ awareness of MAE occurrence, the researcher checked if there was

any relation between the students’ awareness and their year of study.

Table 1: Respondents’ awareness of MAE occurrence across each year of study

Are you aware of any medication administration errors in the health services where you are placed?

Year of study No Yes Total

Second year 16 (14.7%) 93 (85.3%) 109

Third year 18 (19.4%) 75 (80.6%) 93

Fourth year 9 (10.2%) 79 (89.8%) 88

Total 43 247 290

As presented in Table 1, the respondents’ awareness of MAE occurrence in each year

of their study is highlighted. The majority (n = 290) responded to the question, except

for one who did not answer this question.

A similar proportion of the respondents were aware of MAE occurrence during their

clinical placements in the three years of study. The highest percentage of awareness of

MAE occurrence among the respondents was noted among the fourth-year students

(89.8%, n = 79).

Abusaksaka, Vember and Marie Modeste

8

Relationship between the Respondents’ Year of Study and Their Awareness of

MAE Occurrence

Table 2 shows the result of the chi-square test to determine if there was a significant

difference between the respondents’ year of study and their awareness of MAE

occurrence during their practice time.

Table 2: Results of chi-square tests for the respondents’ awareness of MAE

occurrence across years of study

Chi-square tests

Value Df Asymptotic significance (2–sided)

Pearson chi-square 2.986* 2 0.225

Likelihood ratio 3.027 2 0.220

Linear-by-linear

association

0.610 1 0.435

No. of valid cases 290

Note:

*0 cells (0.0%) have an expected count less than 5. The minimum expected count is 13.05.

There was no significant difference between the respondents’ year of study and their

awareness of MAE occurrence as indicated by the p-value in Table 2. In other words,

the respondents’ awareness of MAE occurrence was independent of their year of study.

Causes of MAE Occurrence

Subscales of Respondents’ Perceptions of the Causes of MAE Occurrence

Figure 1 shows the subscales of the respondents’ perceptions of the causes of MAE

occurrence.

Abusaksaka, Vember and Marie Modeste

9

Figure 1: Subscales of the respondents’ perceptions of the causes of MAE occurrence

The physician communication subscale was ranked at the top subscale and was agreed

on by over two-thirds of the respondents. Moreover, the respondents indicated that the

medication package and nurse related subscales were at the same level of agreement.

Significant Subscales of the Causes of MAE Occurrence

Table 3 presents the results of one sample t–test for the four main subscales of the causes

of MAE occurrence. All the subscales were significantly different from the neutral level

of the 6-point Likert scale (3.5). The observed point was the pharmacy related subscale

which was significantly lower than the neutral level. This is an indication that this

subscale was not a significant cause of MAE occurrence among the respondents. The

other three subscales were significantly higher than the neutral level of the 6-point

Likert scale and thus reflected a significant cause of MAE occurrence.

Table 3: Significant subscales of the causes of MAE occurrence

One-sample t-test

Subscale

Test value = 3.5

t Df Sig. (2– tailed)

Mean difference

95% confidence interval of the difference

Lower Upper

0%

20%

40%

60%

80%

100%

Medication package Physician communication

Pharmacy related Nurse related

Agree Disagree

Abusaksaka, Vember and Marie Modeste

10

Physician

communication

12.459 290 < 0.001 0.72125 0.6073 0.8352

Medication package 3.193 290 0.002 0.24914 0.0956 0.4027

Pharmacy related –28.445 290 < 0.001 –1.73654 –1.8567 –1.6164

Nurse related 6.537 290 < 0.001 0.35880 0.2508 0.4668

Significant Difference Items of the Causes of MAE Occurrence

Table 4 presents the results of the one-sample t-test for the items.

Table 4: The significant causes of MAE occurrence

One-sample t-test

Item

Test value = 3.5

Mean Std. deviation

Std. error mean t

95% confidence interval of the difference

Df Sig. (2– tailed)

The names of many

medications are similar.

3.50 1.611 0.095 –0.036 289 0.971

Different medications look

alike.

3.99 1.667 0.098 4.976 290 < 0.001

The packaging of many

medications is similar.

3.76 1.657 0.097 2.644 288 0.009

Physicians’ medication orders

are not legible.

4.65 1.650 0.097 11.819 288 < 0.001

Physicians’ medication orders

are not clear.

4.67 1.514 0.089 13.151 289 < 0.001

Physicians change orders

frequently.

3.88 1.448 0.085 4.501 289 < 0.001

Abbreviations are used instead

of writing the orders out

completely.

4.85 1.554 0.091 14.840 290 < 0.001

Verbal orders are used instead

of written orders.

2.97 1.733 0.102 –5.253 289 < 0.001

Pharmacy delivers incorrect

doses to this unit.

1.90 1.226 0.072 –22.232 289 < 0.001

Pharmacy does not prepare the

medication correctly.

1.72 1.116 0.065 –27.247 290 < 0.001

Pharmacy does not label the

medication correctly.

1.68 1.112 0.065 –27.929 289 < 0.001

Abusaksaka, Vember and Marie Modeste

11

The means of 12 of the 20 mentioned causes of MAE occurrence were significantly

higher than the neutral mean of the 6-point Likert scale (3.5). The means of these causes

ranged between 3.76 (+1.657) and 4.85 (+1.554). The top three significant causes with

the highest means were the use of abbreviations (4.85±1.554), unclear physician orders

(4.67±1.514), and illegible physician orders (4.65±1.619). Similar packaging of many

medications ranked at the bottom of the significant causes with a mean of (3.76±1.657).

Discussion

The current study aimed to determine nursing students’ awareness of MAE occurrence

and what they perceived to be the causes of MAE occurrence during their practice time.

The study found most of the respondents were aware of MAE occurrence during their

practice. This finding has been reported worldwide. In the emergency department of the

Imam Khomeini Hospital in Iran, less than half of the nurses had committed MEs

Poor communication between

nurses and physicians or

between the nursing student and

the supervisor.

4.32 1.464 0.086 9.589 290 < 0.001

Many patients are on the same

or similar medications.

4.51 1.446 0.085 11.936 290 < 0.001

On this unit, there is no easy

way to look up information on

medications.

3.40 1.758 0.103 –0.984 290 0.326

Nurses or the nursing students

get pulled between teams and

from other units.

3.84 1.688 0.099 3.432 288 0.001

Nurses or the nursing students

do not adhere to the approved

medication administration

procedure.

3.07 1.686 0.099 –4.319 289 < 0.001

Nurses or the nursing students

are interrupted while

administering medications to

perform other duties.

4.63 1.619 0.095 11.896 290 < 0.001

Unit staffing levels are

inadequate.

4.24 1.600 0.094 7.929 289 < 0.001

All medications for one team of

patients cannot be passed

within an accepted time frame.

3.99 1.651 0.097 5.014 289 < 0.001

Nurse or the nursing student is

unaware of a known allergy.

3.19 1.789 0.105 –2.999 290 0.003

Abusaksaka, Vember and Marie Modeste

12

(Ehsani et al. 2013, 1). Another study conducted by Feleke, Mulatu and Yesmaw (2015)

indicated that MAEs occurred in about half of administered medication among nurses

in Ethiopia. According to Ojerinde and Adejumo (2014, 22), more than half of the

nurses in Nigeria were involved in ME occurrence. These errors occurred because of a

shortage of nurses, insufficient pharmacological information, nurses’ interruption

during medication administration, physician communication causes, nurse related

causes, pharmacy related causes, and medication package causes (Aboshaiqah 2014, 63;

Al-Youssif, Mohamed and Mohamed 2013, 61; Blignaut 2015, 162; Ehsani et al. 2013,

1; Hanna 2014, 41; Wakefield, Uden-Holman and Wakefield 2005, 484).

Four subscales for MAE occurrence were mentioned in the current study. Physician

communication represented the main subscale of the causes of MAE occurrence.

Moreover, the result of the one-sample t-test reflected the fact that this subscale was a

significant cause of MAEs as perceived by the respondents. This finding is supported

Aboshaiqah (2014, 63), who reported the physician communication subscale as the main

cause of MAE occurrence. Moreover, the physician communication subscale was

reported by other studies as the second cause (Hanna 2014, 41), or the fourth cause (Al-

Youssif, Mohamed and Mohamed 2013, 61) of MAE occurrence. Furthermore, other

studies noted some cases related to physician communication as a reason for MAE

occurrence. Blignaut (2015, 162–182) noted communication lapses between nurses and

physicians as a cause of MAE occurrence among the respondents