There is a strong policy agenda and emerging research base for medication safety and the reduction of medication errors across the developed world. Bates (2000) has argued, using a range of epidemiological studies that medication errors are common and that adverse drug events are often preventable. A key stage in the delivery of medicines to patients is administration – it is the final opportunity for clarifying whether the entire process has been carried out effectively. It is usually carried out by nurses. Additionally, nurses are also prescribing medicines – an initiative which is likely to expand in the UK (Medicines & Health Care Products Regulatory Agency 2005). Many factors contribute to the ineffective administration of medicines, which include mathematical ability (Ridge et al. 1995) and pharmacological knowledge (Leape et al. 1995). By designing and evaluating a test which assesses nurse's pharmacological knowledge and calculation skills in a multi-site study, Grandell-Niemi (2005) can make a pragmatic contribution to nurse education and ultimately patient safety.
The concept of ‘pharmacological skills’ is operationalized by Grandell-Niemi et al. through reference to three components: giving patients treatment-related advice, giving medicines correctly and evaluating any adverse effects. The subsequent literature review includes a range of salient research papers but by the author's admission, only from nursing journals. However, as the delivery of medicines to patients is an inherently multi-disciplinary activity, research papers specific to this topic are also evident in non-nursing journals. For example, Rowe et al. (1998) in the Archives of Disease in Childhood examined 64 paediatric residents’ calculation skills through testing and then compared their results with length of training. Whilst aware that the key words used by the authors might not identify such papers, I would at least argue for the inclusion of Taxis and Barber (2003a,b) who produced an informative study of nurses’ i.v. medication errors. Although it is contended elsewhere that this study was not a true ethnography as originally described (Armitage & Hodgson 2004), Taxis and Barber systematically collected and analysed a range of observations identifying preparation of wrong drugs, wrong volumes and wrong doses – particular to multiple step preparations; and wrong administration rates – particular to bolus doses (Taxis & Barber 2003a,b). Studies such as these make a useful contribution to the conceptualization of pharmacological skills and their results may suggest that some nurses have limited pharmacological knowledge and skills, especially for prescribing purposes. Further study is necessary so as to compare the pharmacological skills of qualified staff, and constructive objective and reliable self-assessments with additional independent assessment.
Grandell-Niemi et al. devised a three part questionnaire which included the Medication Calculation Skills Test, a term which does not obviously suggest that pharmacological as well as mathematical skills are tested. Two other sections addressed: the participant's personal and professional characteristics, and self-assessment of basic and higher level pharmacological skills using a Likert-type scale. A sample of 364 qualified and 282 graduating student nurses provided a response rate of 68% and 70%, respectively. The author's attention to confidentiality through anonymity and voluntariness may have contributed to this rate and, although encouraging, begs the following questions – did they those who failed to respond have very limited pharmacological skills or as is sometimes the case in medical incident reporting, did they hold a fear of revealing their flaws (Waring 2005)?
Content validity of the instrument was assessed using both literature and practitioner feedback. Cronbach's coefficient alpha was used to evaluate the reliability of the instrument at pilot stage, giving a value range of 0·72–0·92 – the authors rightly claiming this to be adequate based on the number of items. A value at the lower end of the range above was achieved in the study proper – which is equally adequate.
However, there was a threat to construct validity in the data collection process, clearly acknowledged by the authors and largely because the qualified nurses did not complete the test/questionnaire under the same conditions. Whilst the reader is informed that a lack of consistent patterns in the completed returns failed to indicate participant cooperation, because the qualified staff did not take the questionnaire/test under classroom conditions (like their student counterparts), they would still have had more time to consider their answers and presumably the opportunity to refer to textbooks. It would have been helpful to know whether any time constraints were set on both groups for returning the completed test/questionnaire, and whether both groups were asked to complete the test/questionnaire as an unseen examination.
Descriptive and non-parametric statistical measures were used to analyse and compare the data sets with according rationales. The results showed that both groups of nurses were regularly involved in medication administration and related calculations. However, the students were unhappy with the volume of preparation they received, despite showing less interest in the topic than qualified staff. On the self-rating scales, pharmacology was not viewed as an easy topic – pharmacodynamics and pharmacokinetics were suggested to be least well known. Personal experience of feedback from nursing students studying pharmacology here in the UK concurs with the cited work of Clancy et al. (2000) and Manias and Bullock (2002), providing a similar picture which is perhaps partly attributable to the low number of nursing students who have familiarity the natural sciences (e.g. chemistry) at advanced level (Leathard 2001). In line with this perspective, those with a strong history of mathematical ability achieved a higher score than their less mathematically able peers. The qualified nurse's self-rated sum variable, mean score and actual test mean score were higher than the student nurses, but because of the previously acknowledged limitation concerning validity, this finding should be treated cautiously.
Notably, those in both groups, sufficiently motivated to maintain their pharmacological skills, found pharmacological interesting. It was also found that the student's self-rated skills had a closer correlation with their actual skills than their qualified counterparts, which may be related to their constant exposure to teaching, learning and, moreover, assessment.
The findings and subsequent discussion generate a number of further questions related to additional research and nurse education:
- •What pharmacology theory and related practical skills are taught in the Finnish pre- and postregistration nursing curricula, compared with the curricula of other countries, and what is the comparative effectiveness?
- •Why do qualified nurses, who are apparently more confident about their skills, find pharmacology more interesting?
- •Does clinical experience per se really make a difference to pharmacological competence?
The first question might be answered though international research collaborations – the foundation of which could be laid in conjoined publication and commentary processes, as here. Evaluating the effectiveness of any educational input may provide another opportunity to use the MCS. The last two questions – linked to the reality of nursing practice – stimulate the debate about non-didactic teaching methods in pre-registration education such as those advocated by Weeks et al. (2000) where drug dosage errors are addressed by taking a constructivist approach (von Glaserfield 1989). Weeks et al. concluded that nursing students require conceptual knowledge, i.e. knowing how to construct a problem prior to the resolution of any numerical calculation achieved by giving students words as well as numbers to describe each step of a calculation process. This approach could be further enhanced by the use of problem-based learning (PBL) where students are given medication-related scenarios such as simulated incorrect prescription orders, the medications themselves, and the option of discussing the problem with practitioners in a clinical setting. Students are then required to establish learning outcomes, diagnose the inherent problems and demonstrate a rational solution to their peers and supervisors (Boud & Feletti 1997). Mindful of Grandell-Niemi's findings, PBL because of its self-directed nature, would have to follow key note lectures delivering relevant foundation theory.
If skill deficits are being constantly demonstrated in this crucially important element of practice, policy makers might reflect on a periodic, mandatory examination of pharmacological skills (comparable with examination of resuscitation skills in some organizations) to satisfy a nurse's fitness to practice. Research such as that by Grandell-Niemi et al. can provide the momentum for such initiatives, particularly when applied to pragmatic yet creative learning methods, which could be further developed through international collaborations.