This paper examines the development of and some logistical and methodological issues surrounding the use of animal models of chronic pain. The first section addresses the emergent move towards mechanism-based and disease-related animal models of chronic pain that has accelerated since the late 1980s following publication of Bennett and Xie's (Pain 33:87-107, 1998) paper on chronic constriction injury of the sciatic nerve and Stein et al.'s (Pharmacol Biochem Behav 31:445-451, 1988) paper on unilateral hind paw inflammation with complete Freund's adjuvant. The discussion covers vast areas of chronic pain models developed over the past 50 years, starting with the numerous neuropathic, inflammatory and central pain models, as well as the growing number of models developed to study various forms of chronic pain from chronic back pain to visceral pain. It also examines the advantages and disadvantages of tonic pain models, mechanism-based and disease-related models of chronic pain, including issues related to the novel discovery of injury- or disease-related pathophysiological processes, the expansion of testing repertoires, and the successes and failures in the translation of analgesic development from animal preclinical models to human chronic pain conditions. The second section addresses experimental design considerations in the implementation of one of the 3Rs for the use of animal models of chronic pain; that is methods employed to reduce the number of animals used. The discussion covers various issues including the advantages and disadvantages of repeated dose designs and within-group drug testing, including incremental dosing schedules, and crossover designs. It also examines concerns surrounding the stability of symptoms and measures, including varying durations of multiple symptoms and the potential development of nociceptive sensitization, as well as possible use-dependent alterations in drug sensitivity and time-dependent changes in pain processes in specific animal models.