This study assessed the use and clinical yield of diagnostic imaging (radiography, computed tomography, and medical resonance imaging) ordered to assist in the diagnosis of acute neck injuries presenting to emergency departments (EDs) in Kingston, Ontario, from 2002–2003 to 2009–2010.
Acute neck injury cases were identified using records from the Kingston sites of the Canadian National Ambulatory Care Reporting System. Use of radiography was analyzed over time and related to proportions of cases diagnosed with clinically significant cervical spine injuries.
A total of 4,712 neck injury cases were identified. Proportions of cases referred for diagnostic imaging to the neck varied significantly over time, from 30.4% in 2002–2003 to 37.6% in 2009–2010 (ptrend = 0.02). The percentage of total cases that were positive for clinically significant cervical spine injury (“clinical yield”) also varied from a low of 5.8% in 2005–2006 to 9.2% in 2008–2009 (ptrend = 0.04), although the clinical yield of neck-imaged cases did not increase across the study years (ptrend = 0.23). Increased clinical yield was not observed in association with higher neck imaging rates whether that yield was expressed as a percentage of total cases positive for clinically significant injury (p = 0.29) or as a percentage of neck-imaged cases that were positive (p = 0.77).
We observed increases in the use of diagnostic images over time, reflecting a need to reinforce an existing clinical decision rule for cervical spine radiography. Temporal increases in the clinical yield for total cases may suggest a changing case mix or more judicious use of advanced types of diagnostic imaging.