An evidence-based medicine approach was applied to evaluate analytic studies of exogenous risk factors for amyotrophic lateral sclerosis (ALS) published since 1991. Classification systems for evaluating the literature and for drawing conclusions based on the class of available evidence were developed, modeled on those used by national societies. Considerations regarding the impact on the general public of confirming a role for putative risk factors were made explicit. There was evidence in support of smoking being a probable ('more likely than not') risk factor for ALS. Smoking has broad public health impact, no redeeming features, and is a modifiable risk factor. Evidence supported the conclusion that the following were probably not risk factors for ALS: trauma, physical activity, residence in rural areas and alcohol consumption. Evidence-based medicine methodology does not permit calculation of the magnitude of type I or type II errors in drawing these conclusions. New evidence may change these conclusions. Recommendations for future research include: draw on clinical trial methodology in designing future, confirmatory, case-control studies; consider utilizing cohort studies, recognizing the longer timelines for these to come to fruition; accord priority to investigating putative risk factors with greatest public health impact. Advances in study methodology may lead to development of finite research cycles for individual putative risk factors for sporadic ALS.