Injections in Patients with Osteoarthritis and other Musculoskeletal Disorders

Abstract

This study examined whether resident injection skills could be enhanced using synthetic injection models. A total of 30 physiatry residents underwent written and injection-model pretesting and posttesting. After randomization into injection-model and control groups, the experimental group trained by watching an instructional videotape and by using models that gave visual feedback on injection accuracy, whereas the control group studied technical aspects of injections. Immediately after patient injections, a blinded attending graded residents on the required level of verbal or manual assistance. The experimental group performed significantly better during patient injections as per first injection data (i.e., the scores obtained from performing a procedure for the first time on each particular body region; P = 0.013), total injection data (i.e., the mean scores obtained from performing all procedures on each particular body region; P = 0.017), and postrotation practical testing (P < 0.007) but not for didactic knowledge (postrotation written testing; P < 0.039). Data analysis by body region showed significant benefit only for occasionally performed patient injection procedures. The benefit was most evident in less experienced residents. Injection-model accuracy testing correlated with actual patient injections, both for first injections into each major body region (r = 0.52, P = 0.005) and for all injections (r = 0.55, P = 0.003). Consideration should be given for incorporating injection-model training into residency education, especially for residents with less injection experience and for occasionally performed procedures. The overall correlation between model practical testing and subsequent patient injection performance suggests that models can measure injection competence.