The Development and Testing of the MISSCARE Survey OR.

Abstract

An acceptable, reliable, and valid survey instrument to measure missed nursing care in perioperative settings has not been developed. To develop and conduct psychometric testing of the MISSCARE Survey OR. Data were collected nationwide from 1,693 operating room (OR) nurses who completed the MISSCARE Survey OR. The survey contained two sections: Part A, "Elements of perioperative nursing care" (32 questions) and Part B, "Reasons for missing nursing care" (17 questions). The MISSCARE Survey OR demonstrated acceptability, as few respondents missed questions in Part A (0.1%-1.1%) and Part B (0.8%-1.3%). Exploratory factor analysis revealed five subscales in Part A (Legal, Preparation, Safety, Communication, and Closing) and four in Part B (Urgency, Staffing, Materials, and Teamwork). In Part A, the five-factor solution explained 44% of the variance. In Part B, the four-factor solution explained 53% of the variance. Alpha coefficients for subscales in Part A ranged from 0.71 to 0.84 and 0.74 to 0.90 for Part B. Validity was measured using content validity, criterion validity, and construct validity. A panel of OR nurse experts established content validity. Criterion validity compared hospitals with fewer than six ORs to hospitals with hospitals with more than six ORs where it was hypothesized aprior that nurses in hospitals with fewer ORs would have missed less care (X = -.123, standard error [SE] = .041, p = .003). Construct validity was tested through exploratory and confirmatory factor analyses (CFA). Correlation coefficients for Part A ranged from 0.34 to 0.73 and 0.60 to 0.73 for Part B. Overall model fit was acceptable: goodness-of-fit index (GFI) and CFA were greater than 0.90, standardized root mean square residual (SMRM) was less than 0.06, and root mean square error of approximation (RMSEA) less than 0.08. The MISSCARE Survey OR promises to be a reliable, valid indicator of the extent of and reasons for missed nursing care.