Analyzing feeding behavior, and in particular meal duration, can be used as a biological marker for temporomandibular joint (TMJ) inflammation/pain. The present study determined the specificity of meal duration as a measure of TMJ inflammation/pain in a rodent model. The model was also used to test the efficacy of dexamethasone (DEX) as a treatment for TMJ inflammation/pain that was induced by TMJ injection of complete Freund's adjuvant (CFA). In the first study, anesthetized male Sprague-Dawley rats housed in computerized feeding modules received bilateral intra-articular knee injections of CFA or saline. The next day, CFA-injected rats had significant knee swelling and impaired mobility. Food intake in the CFA-injected group was reduced over the next two days and this was due to reduced meal number with no change in meal size. Notably, meal duration was normal in both the CFA and saline knee-injected groups. In the second study, male rats were assigned to one of four groups: Group 1, no CFA and no DEX treatment; Group 2, no CFA and treatment with DEX (0.4 mg/kg i.m. once daily); Group 3, bilateral TMJ CFA injection and no DEX treatment; and Group 4, bilateral TMJ CFA injection and treatment with DEX. CFA significantly increased TMJ swelling and stress-induced chromodacryorrhea in Group 3, but treatment with DEX attenuated these effects in Group 4. Compared to the controls, meal duration was significantly lengthened 24 and 48 h post-CFA injection in Group 3, whereas DEX treatment attenuated TMJ swelling, chromodacryorrhea and normalized meal duration. The data demonstrate that meal pattern analysis, and in particular meal duration, can be used as a non-invasive specific measure of TMJ inflammation/pain and can be used as a marker of DEX treatment efficacy.