Medication administration errors from a nursing viewpoint: a formal consensus of definition and scenarios using a Delphi technique

Abstract

Aims and objectives To develop and achieve formal consensus on a definition of medication administration errors and scenarios that should or should not be considered as medication administration errors in hospitalised patient settings. Background Medication administration errors occur frequently in hospitalised patient settings. Currently, there is no formal consensus on a definition of medication administration errors or scenarios that should or should not be considered as medication administration errors. Design This was a descriptive study using Delphi technique. Methods A panel of experts (n = 50) recruited from major hospitals, nursing schools and universities in Palestine took part in the study. Three Delphi rounds were followed to achieve consensus on a proposed definition of medication administration errors and a series of 61 scenarios representing potential medication administration error situations formulated into a questionnaire. Results In the first Delphi round, key contact nurses’ views on medication administration errors were explored. In the second Delphi round, consensus was achieved to accept the proposed definition of medication administration errors and to include 36 (59%) scenarios and exclude 1 (1·6%) as medication administration errors. In the third Delphi round, consensus was achieved to consider further 14 (23%) and exclude 2 (3·3%) as medication administration errors while the remaining eight (13·1%) were considered equivocal. Of the 61 scenarios included in the Delphi process, experts decided to include 50 scenarios as medication administration errors, exclude three scenarios and include or exclude eight scenarios depending on the individual clinical situation. Conclusion Consensus on a definition and scenarios representing medication administration errors can be achieved using formal consensus techniques. Relevance to clinical practice Researchers should be aware that using different definitions of medication administration errors, inclusion or exclusion of medication administration error situations could significantly affect the rate of medication administration errors reported in their studies. Consensual definitions and medication administration error situations can be used in future epidemiology studies investigating medication administration errors in hospitalised patient settings which may permit and promote direct comparisons of different studies.

Which potential harms and benefits of using ginger in the management of nausea and vomiting of pregnancy should be addressed? a consensual study among pregnant women and gynecologists

“…Previous studies used panels ranging in size from 10 to 1000 [44]. The number of panelists in this study was larger or similar to those previously used in Delphi consensus studies on issues in healthcare [40, 41, 46]. The advantages of the Delphi technique includes maintaining anonymity of the panelists, possibility of including panelists from different geographic locations, reduces costs and efforts to bring the panelists together compared to focus groups, and ensures immunity against individual domination of the decision compared to nominal or focused groups [44, 47].…”

“…Guidelines on what clinicians should address during the clinical consultation when ginger is advised for NVP do not exist. We, therefore, believe that such lists developed through consensual methods might be beneficial in changing the behavior of physicians during the clinical consultation [14, 40, 43, 46, 48–50]. …”

“…In this study, both panels were composed of 50 members. The panel size used in this study was either comparable to or more than those used in previous studies [36, 40, 43, 46, 50]. The panel of physicians included participants of both genders, from different geographical locations, clinical practice settings, age groups, and experience periods.…”

“…The study was conducted between November 2016 and February 2017. As in previous Delphi consensus studies [14, 40, 41], we decided to achieve consensus in two consecutive iterative Delphi rounds.…”

“…Physicians and women were considered two different panels. Consensus was defined as in previous studies on issues in healthcare [40, 43]. When the median votes was between 7 and 9 and the IQR was between 0 and 2, consensus was said to have been achieved and the potential harm or benefit was included in the list of potential harms and benefits to be addressed during the clinical consultation.…”