A guide to interpreting epidemiologic studies on the etiology of back pain

Understanding disease etiology is key for effective preventive or therapeutic interventions. Knowledge about risk factors is useful to the clinician to answer patients' questions about the possible cause(s) of their presenting disorders. This article describes the three most common types of etiologic studies using examples from the published literature on the risk factors for back pain. Cohort studies typically follow a group of "healthy" people forward in time to assess disease outcome after risk factors have already been measured. Case-control studies use people selected on the basis of outcome status; risk factors are assessed after the fact. Although more prone to bias than cohort studies, case control studies are more common in back pain research because of the ease of examining several risk factors simultaneously and the expense of cohort studies. Prevalence studies use a random sample of people collected at a single point in time; consequently there is no predetermined number of "cases" and "controls," their numbers depending on the prevalence of disease and exposures in the samples. To assist in the interpretation of research findings this paper presents a general framework for assessing the strengths and weaknesses of an individual study. Case-control studies are discussed in more detail because they are so commonly encountered in the etiologic literature. Finally, because the evidence from a single study, no matter how well designed and executed, is never enough on its own to decide if a risk factor is causal or not, the paper concludes by briefly summarizing the criteria for inferring causation using the full body of available biologic and epidemiologic literature.