Pictograms for safer medication handling by health care workers: a validation study in nursing students in Poland

For drug therapy to be optimal, the right patient should receive the right dose of the right drug, in the right way, at the right time [1]. Unfortunately, these conditions are often not met in real-life situations, leading to a high number of medication errors [2, 3]. These errors can be clinically inconsequential or be seriously harmful to patients resulting in hospitalization, life-threatening situations, disability, or even death [1‐4]. Nevertheless, they also represent an economic burden [5, 6], a source of trauma for the staff and weaken the patient-healthcare provider relationship [4]. Despite this, they are still widespread [7], especially during prescription and administration [2, 3]. The medication errors occuremainly due to deficiencies in training, overwork of the healthcare workers, lack of communication, policies inadequate to the actual practice, lack of consistency in procedures or the absence of pharmaceutical follow-up [4].

In recent years, medication error management has undergone a positive evolution from a blame-based approach to a systems approach that seeks to identify and address the underlying causes of medication error [8]. Thus, the focus is now on the provision of protocols, tools and resources designed to help decrease medication administration errors such as bar-coding systems, weight-based dosing, double- or triple-checking systems, increased pharmacist involvement, avoiding abbreviations, etc.[3‐5] One of the key points identified to reduce the risk of medication errors is to provide access to critical characteristics of drug during administration, however, the time constraints faced by healthcare professionals require these reminder systems to be concise [4].

Thus, pictograms could be an interesting alternative because of their visual impact and their ability to convey information in a concise way regardless of language skills [9]. According to various cognitive theories [10‐12], pictograms can increase recall of instructions, especially when combined with training. The cautionary pictograms of the Globally Harmonized System for the classification and labelling of chemicals (GHS) have already been used by the Workplace Hazardous Materials Information System (WHMIS) to increase workplace safety during the handling of chemicals [13]. A similar tool could improve the handling of medication by healthcare professionals, especially during the administration phase [14, 15]. However, the interpretation of a pictogram is person-dependent and therefore likely to be different according to culture, profession, age, etc. Thus, after the design and implementation of a pictogram, it is essential to complete a validation step of the pictogram understanding by the target population [16].

In a previous study conducted in Canada in collaboration with the Institute for Safe Medication Practices (ISMP), researchers have identified key medication safety categories that could benefit from the implementation of safety pictograms, including look-alike, sound-alike drugs, concentrated electrolyte solutions, and medications that have a high incidence of dosage/calculation error [17]. Pictograms for each of the nine key medication safety issues have been developed by a panel of international healthcare workers. In their study, Vaillancourt et al.[18] took a first step towards validation of these nine pictograms in a sample of pharmacy students in Canada. These students were able to guess the meaning of four of the nine pictograms and to recall the meanings of seven of the pictograms. These results suggest that at least two of the medication safety pictograms are not easily understood. However, pharmacists and therefore pharmacy students may be less aware of medication administration safety because they do not administer medications.

To further test the medication safety pictograms, nursing students in their final stages of study appear to be a good population. Indeed, as future nurses, they will be the health care professionals most often responsible for administering medications in hospitals, they have already completed hospital internships and received the majority of their education. Furthermore, medication safety pictograms must be understood by the most junior and least experienced health care professionals who administer medications because they are at risk of making errors [4].

This research project aimed to determine the comprehensibility, representativeness, and rate of recall for nine medication safety pictograms in a sample of nursing students in Poland in order to validate these images.

This study was conducted by Cardinal Stefan Wyszyński University, Warsaw, Poland among nursing students at the Hipolit Cegielski State University of Applied Sciences, Gniezno, Poland.

Although pictograms can be an effective tool in preventing medication errors during administration, they can have the opposite effect if the message they convey is not clearly perceived. It was therefore important to test the pictograms on students and healthcare professionals responsible for medication administration prior to their implementation. In phase 1, when participants were first presented with the pictograms and without additional information about their meaning, six of the nine pictograms were correctly understood by at least 66.7% of participants. In phase 2, four weeks after the intended meaning of the pictograms was revealed, participants were significantly more able to correctly interpret the pictograms. Indeed, all nine pictograms were correctly understood by almost all participants, which is well above the minimum 66.7% required by the ISO standard. Thus, the three pictograms that did not meet the ISO standard during phase 1 were finally considered as valid during the recall phase. This demonstrates that with training on the meaning of the pictograms, even if brief, these pictograms can be understood and easily memorized by health care professionals, thus be able to convey the intended messages. Representativeness scores also increased significantly in phase 2, reaching a total of 99.5% scores ≥ 5. This indicates that the training not only leads to high recall of the meaning of the pictogram, but also strengthens participants' perception of the relationship between the pictogram and the intended meaning, to the extent that it becomes obvious for them.

The pictograms had previously been submitted to validation test in a sample of Canadian pharmacy students [18]. It is expected that the nursing students have not been exposed to these pictograms as these have not been used in clinical practice. Overall, Polish nursing students were more likely to understand the pictograms than Canadian pharmacy students, suggesting that nursing students' exposure to medication administration enhances their appreciation of pictograms. Indeed, pharmacy students correctly guessed four pictograms in phase 1 and seven out of nine in phase 2. Two ("Concentrated electrolyte formulation" and "Neuromuscular blocking agent") of the three pictograms that did not meet ISO standards in phase 1 among nursing students also did not meet standards among pharmacy students, confirming that the messages conveyed by these two pictograms are particularly complex to understand. However, the "Concentrated electrolyte formulation" pictogram was more likely to be guessed by nurses, suggesting that the hypothesis by Vaillancout et al. that first- and second-year pharmacy students did not comprehend it due to their educational stage was likely. The "Neuromuscular blocking agent" pictogram was similarly guessed by Canadian pharmacy students. The pictogram "Drug requiring airway management" was correctly guessed by 69.3% of pharmacy students compared with 57.6% of nursing students, but this difference could be explained by the different sample size. In phase 2 among pharmacy students, two pictograms, "Concentrated electrolyte formulation" and "Medication with a significant risk of harm if administered improperly", were still correctly interpreted by less than 66.7% of pharmacy students, whereas all pictograms were largely consistent with ISO standards among nursing students. This better understanding of the pictograms by the nursing students may be explained by the fact that they are directly involved in the process of administration, more aware of the issues and therefore more likely to remember them, but also by the fact that they had more years of education than the first- and second-year pharmacy students. However, these comparisons must be taken with the necessary hindsight, as the transparency response raters were different between the two studies and could therefore gave different corrections.

The "Drug that requires airway management" pictogram did not meet the guessability requirements in phase 1. The image was easily understood, but the concept comprehension appeared problematic. This was confirmed during the translucency test with only 50.0% of representativeness scores ≥ 5 and recommendations given by participants that the intended meaning should be adjusted because it did not match the image. This confusion could be addressed by adding text with the pictogram to avoid confusion, along the same model as the WHMIS pictograms [13]. Indeed, WHMIS pictograms include a "signal word" to alert the reader to a potential hazard and a "hazard statement" describing the nature of the hazard. After the recall phase, guessability met ISO requirements with 100% correct guesses and representativeness score ≥ 5, suggesting that training was sufficient to improve the clarity of the concept for participants.

The fact that the pictogram depicting "Neuromuscular blocking agent" did not meet the guessability requirements in phase 1 was due to poor understanding of the image. This was confirmed in the translucency phase with only 65.2% of representativeness scores ≥ 5 and nine out of ten recommendations given by participants reporting that the pictogram was unclear. In contrast, in the recall phase, confusion was no longer made with 100% correct guesses and 98.5% representativeness scores ≥ 5, thus training (during phase 1 of the study) ultimately resulted in correct recall of the intended meaning without need for redesign.

The "Concentrated Electrolyte Formulation" pictogram did not meet the transparency requirement in phase 1. Participants could identify electrolytes, however they did not associate it the concept of high concentration. Nevertheless, the phase 1 translucency results were consistent with ISO standards, suggesting that the addition of a "hazard statement" describing the nature of the hazard should be sufficient to make the pictogram understandable. Alternatively, the more transparent pictogram needs to be proposed.

Although these pictograms were designed in North America, the Polish nursing students were able to understand all of them well enough to meet the ISO requirements in phase 2. Thus, despite differences in culture, representations and language, the pictograms are understandable. This result confirms the relevance of the design of the pictograms to the actual practices and challenges encountered by nurses. It could be interesting to test these pictograms in other countries and with other health professionals administering medication.

The nine medication safety pictograms can be considered as valid. However, further research is needed to check transparency and the understandability of pictograms in the group of hospital medical workers. It is important to assess the legibility of pictograms in everyday clinical practice, including their size and placement, taking into account the large amount of work, pressure and the rush. Research combining iteration testing of pictogram design as well as pictograms in combination with the hazard statements will be of particular importance.

Since some of these pictograms are complicated to understand, adding a hazard statement with the pictograms would greatly improve their comprehensibility. Nevertheless, it is necessary to associate pictograms with training prior to their implementation in order to facilitate the memorization of their meaning.