There is uncertainty over whether the patient group in which acute pain studies are conducted (pain model) has any influence on the estimate of analgesic efficacy. Data from four recently updated systematic reviews of aspirin 600/650 mg, paracetamol 600/650 mg, paracetamol 1000 mg and ibuprofen 400 mg were used to investigate the influence of pain model. Area under the pain relief versus time curve equivalent to at least 50% maximum pain relief over 6 h was used as the outcome measure. Event rates with treatment and placebo, and relative benefit (RB) and number needed to treat (NNT) were used as outputs from the meta-analyses. The event rate with placebo was systematically statistically lower for dental than postsurgical pain for all four treatments. Event rates with analgesics, RB and NNT were infrequently different between the pain models. Systematic difference in the estimate of analgesic efficacy between dental and postsurgical pain models remains unproven, and, on balance, no major difference is likely.