Medication dosage calculation among nursing students: does digital technology make a difference? A literature review

Patient safety is a major component of nursing and following patients’ medication orders is one of the most important things nurses do [1‐3]. Several studies [4‐6] emphasized the difficulties nursing students face when calculating medication doses. More than half of nursing students around the world fail numeracy and medication dose calculation (MDC) assessments, indicating the global nature of this issue [6].

MDC mistakes are also one of the most common complications in modern medicine [7]. Many of the errors are avoidable and constitute a significant risk to patient safety [8]. These errors can have a variety of consequences for patients, ranging from loss of a medication’s beneficial effects to death. Aside from the ethical issues, MDC errors can result in higher healthcare expenses [9]. Medical errors are expected to cost at least 20 billion dollars in the United States [10] and in Sweden around SEK 8 billion per year [11].

According to Karlstedt, Wadensten, Fagerberg, and Pöder [12], nursing education does not provide sufficient competence in this field. Newly registered nurses are expected to be proficient in implementing safe pharmacotherapy immediately after graduation. According to nursing studies, nurses have inadequate MDC skills, and this is a global concern for the nursing profession [13, 14]. Evidence also suggests that nursing students lack a solid numeracy foundation [15‐17] and have had problems with mathematical competency for more than 40 years [15, 16]. Several studies conducted since the 1980s have discovered that a large proportion of nursing students are unable to pass arithmetic assessments [16, 18, 19]. As a result, it is critical to design learning strategies for recognizing the hurdles to success as well as to provide chances for nursing students to improve their numeracy skills.

MDC has always been taught with a chalkboard, pencil, and paper. More recently, a slew of studies investigating the use of novel approaches to teaching medicine administration in simulation have revealed that such methods are already ubiquitous. Several investigations [20, 21] have found that employing technology such as calculators, web-based education, simulations, forums, and Virtual Learning Environments improves nurses’ knowledge and abilities in nursing operations. Despite these advancements in technology, nursing students continue to have difficulty with numerical computation. To overcome these issues, various teaching methodologies have been explored, but no single system has been shown to be extremely effective.

The goal of this study is to find and critically assess the literature on how the usage of digital technologies influences the development of MDC skills in nursing students. Better MDC teaching recommendations for nursing students can be developed by gaining insights into the types of technologies that are being employed and their consequences.

The search process was initiated early by discussing different concepts within the research group, thereafter, delimiting relevant search terms. One of the authors (MAJ), a research librarian, undertook the literature search, but then the whole research team considered the evidence against exclusion and inclusion criteria regarding the overall topic in relation to nursing.

An early scoping search indicated that the phrase “information and communications technology” (ICT) was not sufficient to retain studies relevant to the present aim. ICT is a generic term for a multitude of applications concerning digital technology; it is used in a variety of fields and has different meanings. Therefore, other relevant terms representing different aspects of ICT were identified and included in the search strategy. The terms included in the search strategy are presented in Table 1. The terms were combined using Boolean operators and based on the principle (population OR population) AND (exposure OR exposure) AND (outcome OR outcome).

The following databases were utilized: Scopus (Elsevier), Academic Search Elite/Premiere (Ebsco), Cinahl (Ebsco), ERIC (Ebsco) and Web of Science (Thomson Reuter). MAJ performed the searches in April 2018 and again in August 2021. In the second search the database PubMed was included to broaden the search base. The complete search and inclusion process are described in a PRISMA flow chart (Fig. 1), and the search strategies are available as supplementary material. The search strategy was initially created in Scopus, using free text and a step-by-step method. The strategy was subsequently adapted to the other databases. The search was limited to peer-reviewed articles, both original and review articles. Only articles written in English were included. No citation searches were performed.

The formal searches retained 2843 articles (Scopus n = 831, Academic search elite/premiere n = 321, Cinahl n = 796, Web of Science n = 218, Eric n = 24, and PubMed = 653). Duplicates were removed using Endnote and the de-duplication method suggested by Bramer, Giustini, de Jonge, Holland and Bekhuis [22]. After de-duplication, 1376 articles remained. For the number of articles retrieved and screened for each search occasion (2018 and 2021) see the PRISMA flow chart in Fig. 1.

Prior to the screening process, a test screening was performed, in which all authors screened the title and abstract of 30 articles to ensure consistency in the process. The test screen showed discrepancies among the research group, which led to further clarification of the inclusion and exclusion criteria.

The screening process was performed in three steps. In the first step, the authors – using the web-based tool Rayyan (rayyan.​qcri.​org) – screened all titles and abstracts using the inclusion and exclusion criteria (Table 2). The screening was performed in teams of two, consisting of one researcher from educational science and one from medical and nursing science, thus ensuring that both perspectives were included. Each team member independently screened all titles and abstracts assigned to them. If the results showed any conflicts, i.e., uncertainties regarding inclusion, the team members discussed the papers in question before a final decision was reached. In the first step, 1199 articles were excluded, leaving 174 eligible for the second step.

In the second step, the full texts of the eligible articles were reviewed. The 174 articles were divided equally among the team members, ensuring that they did not read the articles they had screened in the first step. Each team member independently read the full articles. Any uncertainty concerning including or excluding an article was first discussed by the team members and later by all authors. The last discussion showed some minor differences, leading to a second assessment of a selected number of articles. Of the 174 articles, 146 were excluded for focusing on the wrong population, exposure or outcome, leaving 19 eligible for qualitative synthesis.

In the third step, each author separately read eight articles, grouping the main findings in a joint matrix. To ensure uniform presentation, one author (KSN) performed the compilation of all separate matrixes.

Eighteen research articles met the inclusion criteria in that they addressed the issue of developing nursing students’ MDC skills using digital technologies. An outline of the designs, approaches, aims, and samples of the included articles are presented in Table 3. The findings entail a narrative description of the results of the included papers and are presented as follows: Descriptions of the various digital tools used in MDC and their effects. The digital tools are: calculators (n = 3), online tutorial (n = 1), personal digital assistants (n = 4), web-based courses (n = 8) and simulations (n = 2). Most studies had a descriptive or experimental design.

The results showed that web-based courses (n = 8) were the digital technologies most frequently used to develop nursing students’ MDC skills. However, these web-based courses had limited impacts. A number of the studies reported no differences between the intervention group and the control group. For instance, Van et al. [37] and Valizadeh et al. [36] reported no significant difference between the two groups in terms of medicinal calculation ability after training. Similar findings were reported by Mackie and Bruce [4], who presented posttest data indicating that all three types of errors (conceptual, procedural, and unit errors) were still being made by students. Students in Maag’s [30] intervention group demonstrated equal posttest and retest knowledge of math, and their math self-efficacy scores were the same, hence the intervention did not impede students’ learning.

On the other hand, a number of studies highlighted positive effects on students’ medication calculations. Cunningham and Roche [23] showed how all 52 students achieved a score of 90% by the second retake test. The group had participated in the standard program the preceding semester, at which time it took five retakes for the entire group of students to pass with a score of 90%. The intervention results indicated that there was a statistically significant between-group difference in performance outcomes. Their intervention consisted of practice MDC tests, where students were using the Web-CT platform and one-on-one tutorial sessions. The program provided students with feedback on answers to enhance learning, and students could access it as often as they wished. Because it was web-based it was easy to access. Perhaps the positive findings can be explained by the feedback given to students, e.g., increasing students’ understanding of the problem, or by the fact that the students using the Web-CT platform had ‘participated in the standard program the preceding semester.’ Students learning can be shallow, when they only manage the demands and goals of the course, or they can be immersive, encouraging them to apply a deep-learning approach [39]. Within an immersive learning approach, students can gain an understanding and ability to relate their ideas to previous knowledge and recognize patterns. In summary, only a few studies using a web-based intervention to develop students’ learning of MDC skills showed significant differences between the intervention group and control group. Based on these findings, it is reasonable to wonder why the studies used different kinds of educational interventions, but only one [23] could show a positive effect. The findings of Ramjan et al.’s [33] study, similarly, highlighted the importance of providing visually enhanced ‘hands-on,’ ‘authentic’ and clinically contextualized interventions for numeracy learning.

Calculators have been used for decades to improve students’ MDC, and Hembree and Dessart’s [40] meta-analysis showed how use of calculators improves students’ basic skills regarding both problem-solving and accuracy. However, in nursing education, MDC has traditionally been taught using paper and pencil, despite developments in other curricula where calculators are taken for granted [41]. In our literature review, three studies [28, 34, 35] examined the effect of calculators on students’ mathematical skills. Only one of them [34] could show that calculator use was associated with fewer arithmetic errors, but with more conceptual errors. The others [28, 35] showed no difference between students using and those not using calculators. However, Grugnetti et al. [28] demonstrated how nursing students overall had significant gaps in mathematical competence. This is an interesting finding. Do nursing students as a group has gaps in their mathematical competence? Several studies have highlighted how educators are continually confronted with nursing students who lack basic mathematical skills [42‐44]. The absence of effects of implementing different mathematical aids, such as calculators, could be explained by the fact that though calculators are used to facilitating students’ calculations, they require the user to have sufficient basic mathematical knowledge [41].

Four studies [26, 27, 31, 32] assessed whether PDA technology (such as hand-held computers, iPods, etc.) could reduce nursing students’ medication errors. Greenfield [27] showed that use of PDAs has the potential to reduce nursing students’ calculation errors. Similar findings were presented by Pereira et al. [32], who showed that introducing nursing students to the application CalcMed and using CalcMed led to significantly fewer errors among students. Previous studies in other teaching contexts [45] have shown that ICT can enhance students’ mathematical abilities [46]. Some of the studies evaluating the effect of different PDAs have addressed cell phone use in higher education [46], but many studies focused on lower levels of education, such as kindergarten [47] and compulsory school [48]. This focus may be based on technological developments and the recent trends in many Western countries where every pupil at school receives a personal portable computer or tablet PC [49] – a phenomenon that, in turn, has attracted the attention of researchers.

The method of simulation has been used more frequently in recent years [24, 25], and it seems that simulation is the method that has the best effect on students’ learning. This is an area that needs to be explored further using larger intervention groups and clearer randomizations. In addition to this, we need to examine students’ prior knowledge of mathematics. If we combine a mapping assessment of each student’s mathematical knowledge, based on which we can improve this knowledge, with the best learning tools, then perhaps we can near our ultimate goal: safe medication and, thereby, safe patient care.

Methodologically, twelve of the 18 analyzed studies are randomized experimental designs with interventions and control groups, and in one the students self-selected their group depending on whether or not they had a PDA [27]. Four studies [20, 23, 33, 34] did not use control groups but whole-group approaches and one was a pilot study aimed at designing an intervention study [4], while Wright [38] used two groups in a quasi-experimental approach, never mentioning how the students were grouped. Thus, the studies seem to align with traditional research designs focused on the medical and healthcare sector as well as to have a sound methodological foundation, although some methodological questions may be raised concerning the study by Cunningham and Roche [23].

Finally, the reviewed articles draw on digital technologies such as web-based platforms, e-learning modules, calculators, simulation, PDAs, such as hand-held computers, among other devices. However, in the present review, we found no study focusing on, for instance, the use of virtual reality and head-mounted displays to create an immersive experience of MDC situations. Here, new technologies may bring new opportunities to enhance nursing students’ MDC skills globally.