In the clinic, the existing literature is insufficient to counsel our infertile men on medication use. Most studies have flaws that limit their application to evidence-based practice. In this chapter, we discuss the limitations of the current literature and the challenges to designing more useful studies. Among the most important weaknesses of existing studies is lack of power; that is, too few men are included to draw conclusions about the existence and size of medication effects. Adequate power is particularly important when confirming an absence of medication effect. Bias is also a problem in most studies. Early studies were rarely randomized, placebo-controlled, or blinded; a common example is patients receiving different medication regimes based on the severity of their symptoms-making it impossible to attribute differences between treated and untreated men to the medications. Additional bias is introduced by failing to include other factors that influence the outcome in the experimental design. A uniform population amenable to randomization and placebo-control are experimental species, and useful information has been gained from these models. However, application to humans is limited by differences from other species in route of drug administration, absorption of the drug, concentration in the male genital tract tissues, and genital tract physiology. To a lesser degree, there is variation among individual men in their response to drugs. In addition, drugs in the same class may have different effects, limiting the applicability of data across drugs of a single class. Complicating matters further, a toxic medication may seem to improve fertility endpoints by improving a disease condition that diminishes fertility. Finally, drug interactions have not been studied, and actual fertility data (pregnancy/fecundity) in humans are rare. A healthy dose of skepticism is warranted when evaluating studies of medications and male reproductive health.