The aim of this study was to compare two different methods of evaluating the efficacy of known analgesics, currently available in the clinic, for the treatment of the chronic pain associated with rheumatoid arthritis. The paw pressure withdrawal threshold assay (Randall & Selitto, 1957) has long been used as a mechanical test of assessing hypersensitivity in models of acute inflammatory pain (Anseloni et al. 2003). The recent development of the weight bearing assay has seen the introduction of an alternative readout providing an objective assessment of incident hypersensitivity (Clayton et al. 1997). Using a rat model of chronic inflammatory joint pain (Donaldson et al. 1993), a monoarthritis was induced by injecting 150μl (1mg/ml) of Freund′s complete adjuvant (FCA) into the intra-articular space of the knee joint. Injections were done under brief anaesthesia (3% isofluorane with O2 at 1.5l.min-1). A series of six, blind and randomised experiments (n=40, for each experiment) were conducted assessing the analgesic potency of Rofecoxib (1, 3 and 10mg.kg-1), Etoricoxib (1, 3 and 10mg.kg-1) and Ibuprofen (3, 10 and 30mg.kg-1), using both the paw pressure withdrawal threshold and the weight bearing readouts. Drug or vehicle (DMSO 1%, distilled water 33% and PEG400 66%) were chronically dosed p.o. on days 13-17 post FCA injection and the above readouts were used to assess the ability of the test drug to reverse the induced hypersensitivity. The contralateral paw pressure withdrawal threshold readout can be taken to negate any compound-induced motor deficits being falsely concluded as analgesia. All data were transformed into an area under the curve (AUC) value, for days 13-17 for each dosing group. Using Statistica V.6 the AUC data were statistically analysed by an ANOVA followed by a Duncan′s post-hoc test. A difference in groups was only considered significant when P<0.05. The relationship between the AUC group means were then assessed using Pearson′s linear regression correlation analysis in order to produce a correlation coefficient (where 1 = positive correlation, 0 = no correlation and -1 = negative correlation) and an associated P value stating the statistical significance of the test. Animals were humanely killed at the end of the experiment. Results revealed strong correlations between the two readouts (Rofecoxib r = -0.9949, Etoricoxib r = -0.9834 and Ibuprofen r = -0.9514) with both assays producing similar 50% effective dose (ED50 values) for each drug. These results provide evidence that the weight bearing assay can be used alongside the paw pressure withdrawal threshold assay to increase confidence in the pre-clinical potency of novel analgesics and hopefully lead to improved clinical efficacy.