Psychometric Properties of a Questionnaire that Evaluates the Perception of Nursing Personnel to Job Factors that May Contribute to Musculoskeletal Symptoms

Wednesday, 24 July 2013

Neusa M. C. Alexandre, RN, PhD
Department of Nursing, State University of Campinas - Faculty of Medical Sciences, Campinas, São Paulo, Brazil
M. Z. O. Coluci, PhD
Department of Nursing, Unicamp, Campinas, Brazil

Learning Objective 1: The learner will be able to know a new questionnaire: the “ErgoEnf - Questionnaire survey for ergonomic risk evaluation among nursing workers”

Learning Objective 2: The learner will be able to know the process of development and validation of the “ErgoEnf - Questionnaire

Purpose: . The objectives were to develop a questionnaire that evaluates the perception of nursing workers to job factors that may contribute to musculoskeletal symptoms, and to evaluate its psychometric properties.

Methods: Literature reviews, identification of instruments and interviews were performed. The development stages of domains and items had its content assessed by two expert committees. After that, a pre-test was performed. Data were obtained through interviews with 370 nursing workers. The “ErgoEnf - Questionnaire"was composed by 32 items distributed in four domains. The construct validity was analyzed by the factorial analysis, the construct validity (technique of the known groups: nursing workers and administrative workers) and by the convergent validity, comparing  the instrument with the Numeric Pain Scale, the SF-36, and the Work Role Functioning Questionnaire. Reliability was assessed through internal consistency and stability (test-retest). Quantitative analysis were performed: descriptive analysis, technique of confirmatory factor analysis, the Mann-Whitney test to compare known groups, the Spearman correlation test to correlate the new questionnaire scores with other instruments, Cronbach's alpha to study the internal consistency, and the ICC to investigate the stability.

Results: The confirmatory factor analysis indicated that the questionnaire indices are all tweaking satisfactory, confirming the pre-defined structure of the questionnaire. There was also a significant difference (p <0.01) between the responses of nursing staff and administrative area workers in all domains of ErgoEnf. The correlations between the new questionnaire and Numeric Pain Scale, SF-36 and WRFQ were all significant (p <0.01) and most of them of moderate magnitude. Cronbach's alpha ranged from 0.89 to 0.91 and ICC values ranged from 0.64 to 0.76.

Conclusion:

The process of development and validation of the “ErgoEnf - Questionnaire to evaluate ergonomic risk in nursing” was completed with results that indicate good psychometric properties for use in ergonomic studies.