Psychosocial, socio-demographic and symptom characteristics have been shown to be associated with treatment outcome in patients with craniomandibular dysfunction (CMD). This study was performed to assess to what extent symptoms and correlates of CMD change as a consequence of conservative treatment for CMD. Treatment outcome in a group of CMD patients treated with a stabilization splint (experimental group) was compared with that of a group of patients with CMD who were not treated for CMD (control group). Patients in the experimental group had fewer symptoms of CMD at the end of treatment. However, several symptoms and correlates of CMD also improved in the control group (severity of pain, joint noises, ear symptoms). It was therefore questioned whether all improvements in symptoms and correlates of CMD in the experimental group could be attributed to the treatment received. Results suggest that the main improvement that might be ascribed to therapy was a decrease in 'jaw symptoms'. There was a noticeable decrease in depression and an increased use of 'planned actions and rational thinking' as a coping style in the experimental group whereas these variables did not change in the control group. Implications and suggestions for further research are discussed.