Sources of non-compliance with clinical practice guidelines in trauma triage: a decision science study
1 Department of Critical Care Medicine, The CRISMA Center (Clinical Research, Investigation, and Systems Modeling of Acute Illness), University of Pittsburgh, Scaife Hall, 3550 Terrace Street, Pittsburgh, PA, USA
2 Department of Surgery, University of Pittsburgh, F1266, 200 Lothrop Street, Pittsburgh, PA, USA
3 RAND Corporation, Suite 600, 4570 Fifth Avenue, Pittsburgh, PA, USA
4 Department of Social and Decision Sciences, Carnegie Mellon University, Pittsburgh, PA, USA
5 Department of Medicine, University of Pittsburgh, Suite 200, 200 Meyran Avenue, Pittsburgh, PA, USA
Implementation Science 2012, 7:103 doi:10.1186/1748-5908-7-103Published: 25 October 2012
United States trauma system guidelines specify when to triage patients to specialty centers. Nonetheless, many eligible patients are not transferred as per guidelines. One possible reason is emergency physician decision-making. The objective of the study was to characterize sensory and decisional determinants of emergency physician trauma triage decision-making.
We conducted a decision science study using a signal detection theory-informed approach to analyze physician responses to a web-based survey of 30 clinical vignettes of trauma cases. We recruited a national convenience sample of emergency medicine physicians who worked at hospitals without level I/II trauma center certification. Using trauma triage guidelines as our reference standard, we estimated physicians’ perceptual sensitivity (ability to discriminate between patients who did and did not meet guidelines for transfer) and decisional threshold (tolerance for false positive or false negative decisions).
We recruited 280 physicians: 210 logged in to the website (response rate 74%) and 168 (80%) completed the survey. The regression coefficient on American College of Surgeons – Committee on Trauma (ACS-COT) guidelines for transfer (perceptual sensitivity) was 0.77 (p<0.01, 95% CI 0.68 – 0.87) indicating that the probability of transfer weakly increased as the ACS-COT guidelines would recommend transfer. The intercept (decision threshold) was 1.45 (p<0.01, 95% CI 1.27 – 1.63), indicating that participants had a conservative threshold for transfer, erring on the side of not transferring patients. There was significant between-physician variability in perceptual sensitivity and decisional thresholds. No physician demographic characteristics correlated with perceptual sensitivity, but men and physicians working at non-trauma centers without a trauma-center affiliation had higher decisional thresholds.
On a case vignette-based questionnaire, both sensory and decisional elements in emergency physicians’ cognitive processes contributed to the under-triage of trauma patients.