Validity and Reliability of a Newly Developed Instrument to Measure Nursing Workarounds During Medication Administration.

Abstract

Background and Purpose: According to patient safety literature, workarounds are used when processes are not clear and may not match the intended workflows. There is no available quantitative instrument to measure the type and frequency of workarounds when nurses administer medications. The purpose of this study was to assess the psychometric properties of a newly developed instrument that measured the type and frequency of workarounds when nurses administer medications to patients. Methods: Items for the newly developed instrument were derived based on the concept analysis of workarounds, and the psychometric evaluation included content validity, face validity, item analysis, dimensionality, reliability, and construct validity testing. The instrument was administered to registered nurses in an acute care hospital in Northern Virginia. Results: Psychometric evaluation of the newly developed instrument demonstrated adequate content and face validity. Based on exploratory factor analysis using principal axis factoring of the 18 items, 12 items were retained comprising three subscales: (a) defining characteristics, (b) type of workarounds, and (c) frequency of workarounds. Cronbach's alpha ranged from .83 to .92 for the three subscales. As hypothesized, convergent validity was supported by Spearman Rho correlations ranging from .27 to .47 among the Halbesleben, Rathert, and Bennett total and two subscales. Divergent validity was supported with Spearman Rho correlations ranging from -.09 to .15 with two other Halbesleben, Rathert, and Bennett subscales. Conclusions: The findings provided beginning evidence for the reliability and validity of the newly developed Savage Barcode-Assisted Medication Administration Workarounds Tool comprised of three subscales-defining characteristics, types of workarounds, and frequencies of workarounds.