The study objective was to compare spinal immobilization techniques to a vacuum mattress-splint (VMS) with respect to the incidence of symptoms generated by the immobilization process. We used a prospective, cross-over study in a university hospital setting. Participants consisted of 37 healthy volunteers without history of back pain or spinal disease. Interventions consisted of two phases. In Phase I, subjects were randomly assigned to be immobilized on either a wooden backboard or a mattress-splint for 30 min. The incidence and severity of any symptoms generated by the immobilization process were recorded. In Phase II, the two groups were again tested after a 2-week washout period, with the method of immobilization being reversed. Symptoms and severity were again recorded. Pain symptoms were confined to four anatomic sites: Occipital prominence, lumbosacral spine, scapulae, and cervical spine. After adjusting for the effect of order of exposure, subjects were 3.08 times more likely to have symptoms when immobilized on a backboard than when immobilized on the VMS. They were 7.88 times more likely to complain of occipital pain and 4.27 times more likely to complain of lumbosacral pain. Severity of occipital and lumbosacral pain was also significantly greater during backboard immobilization. We conclude that, when compared to a VMS, standard backboard immobilization appears to be associated with an increased incidence of symptoms in general and an increased incidence and severity of occipital and lumbosacral pain in particular.