Virtual patient technology to teach pharmacists NOAC counselling – Questionnaire development, validation and pilot

Abstract

Abstract
Background Keele University has developed a virtual patient (VP) educational tool to teach pharmacists how to conduct non-vitamin K oral anticoagulant (NOAC) counselling. The VP’s use will be evaluated, but in order to do so, a validated data collection instrument is required. The aim of this study is to validate a data collection questionnaire for evaluating VP uses.

Summary of work The instrument is an online questionnaire in a pre- and post-intervention design which considers the VP’s usefulness, usability, and self-perceived ability changes. Ethical approval was obtained. Using a number of guidelines for instrument development the following steps were undertaken in the validation: literature review, questionnaire development, and pilot testing. The validation stage incorporated face and content validity, internal consistency testing using a Cronbach alpha test, and revisions.

Summary of results Minor grammatical and typographical amendments were identified and made throughout the instrument development process. The Likert scales were amended to be fully verbal and acquiescence bias was considered in questionnaire development, this helped assess internal validity. Cronbach Alpha results were significant (P=0.932 and P=0.872) and the questionnaire was usable from the perspectives of the participants. Content and face validity were incorporated into reviews of the study design and instruments, and both were deemed to be of a high level.

Discussion The Cronbach alpha test had significant results suggesting that the Likert scales of the questionnaire had a high level of internal consistency and in turn, reliability to be used in the wider study, meaning that the Likert items can be grouped together and used as a single overall score.1 Many previous VP evaluations used some form of a questionnaire with many incorporating pre- and post-tests and satisfaction surveys, often using Likert scales. Most of these studies evaluated uses of various VPs and especially considered usability and satisfaction. Although such studies provided a basis for the use of questionnaires they had limited instrument validation. Multiple studies appeared to deploy self-designed instruments and did not discuss any validation or piloting. Many studies did include pilots but with a focus on the technology rather than the instrument.

Conclusion The questionnaire had a high level of validity so it can be used for a larger scale study. The Cronbach alpha test was significant establishing the questionnaire’s internal consistency and reliability. The questionnaire is therefore valid for use, following minor amendments, to evaluate a VP.