A theory-based instrument to evaluate team communication in the operating room: balancing measurement authenticity and reliability.

BACKGROUND: Breakdown in communication among members of the healthcare team threatens the effective delivery of health services, and raises the risk of errors and adverse events. AIM: To describe the process of developing an authentic, theory-based evaluation instrument that measures communication among members of the operating room team by documenting communication failures.
METHODS: 25 procedures were viewed by 3 observers observing in pairs, and records of events on each communication failure observed were independently completed by each observer. Each record included the type and outcome of the failure (both selected from a checklist of options), as well as the time of occurrence and a description of the event. For each observer, records of events were compiled to create a profile for the procedure.
RESULTS: At the level of identifying events in the procedure, mean inter-rater agreement was low (mean agreement across pairs 47.3%). However, inter-rater reliability regarding the total number of communication failures per procedure was reasonable (mean ICC across pairs 0.72). When observers recorded the same event, a strong concordance about the type of communication failure represented by the event was found. DISCUSSION: Reasonable inter-rater reliability was shown by the instrument in assessing the relative rate of communication failures displayed per procedure. The difficulties in identifying and interpreting individual communication events reflect the delicate balance between increased subtlety and increased error. Complex team communication does not readily reduce to mere observation of events; some level of interpretation is required to meaningfully account for communicative exchanges. Although such observer interpretation improves the subtlety and validity of the instrument, it necessarily introduces error, reducing reliability. Although we continue to work towards increasing the instrument's sensitivity at the level of individual categories, this study suggests that the instrument could be used to measure the effect of team communication intervention on overall failure rates at the level of procedure.