In most industrialized countries, disability and work absence due to occupational back pain have risen steadily in recent decades. Conventional views of the causes of this slow epidemic tend to fall into one of the following three areas: (i) the clinical pathology view, which attributes the level of pain and disability to either the severity of the initial injury or to psychosomatic conditions; (ii) the biomechanical exposures view, which attributes the problem to hazardous and preventable conditions of work; and (iii) the perverse incentives view, which suggests that reporting and disability are influenced by a combination of work dissatisfaction and accessible disability benefits. This paper reviews, from an epidemiologic perspective, the specific methodological hurdles faced during investigations of the etiology of occupational back pain. It is argued that methodological issues have contributed to the perpetuation of the three distinct but incomplete views of the problem. New research directions are suggested and a broader interdisciplinary perspective is proposed to help resolve the existing polemic.