Alzheimers disease imposes a significant public health burden that will only worsen as the population ages. Thus, there is considerable motivation to develop effective strategies to treat, or more ideally, prevent the disease. Epidemiologic evidence has suggested that non-steroidal anti-inflammatory drugs (or NSAIDs) may be neuro-protective. However, this evidence is controversial. Observational studies in humans have found that the use of NSAIDs is associated with a lower risk of developing Alzheimers disease. By contrast, randomized trials have reported that NSAIDs are not effective in treating patients with clinically established disease nor in preventing the onset of dementia among those who are cognitively normal or have mild cognitive impairment. In this article, we review the existing epidemiologic evidence on the relationship between NSAIDs and Alzheimers disease and discuss several hypotheses to explain the divergent findings.