Development of a medical error scale for nurses in Turkey.

Background: Medical errors can have an adverse effect on patients, health care providers and health care organizations. Determining the likelihood of such errors is important to implement appropriate and effective solutions to minimize errors.

Aims: The aim of this study was to develop a valid and reliable scale to evaluate the likelihood of medical errors by Turkish nurses.

Methods: The draft scale (with 94 items) was developed based on primary references and the opinions of nursing experts. Content validity was assessed using 15 nursing experts. Construct validity of the scale was assessed with exploratory and confirmatory factor analyses using 298 nurses at a university hospital in Trabzon, Turkey. To assess test-retest reliability of the scale, another group of 50 nurses were included.

Results: The content validity index of the scale was 0.82, Cronbach alpha was 0.89, and item-total correlation values ranged from 0.31 to 0.54. Kaiser-Meyer-Olkin was 0.81, Bartlett test was 5909.75, P < 0.0001, and the anti-image correlations ranged between 0.63 and 0.90. In the four rotations done with varimax rotation, 42 items were excluded because their factor loadings were less than 0.45. The final scale had 43 items and six subscales: falls, blood and blood products transfusion, medication practices, care practices, communication, and other controlled practices. The six-subscale structure was confirmed by confirmatory factor analysis, and the fit between the scale and its subscales was good.

Conclusion: The scale is a valid and reliable tool to collect consistent data on medical errors in the patient-related practices of nurses.