This chapter focuses on the various study designs used to estimate long-term treatment effects of disease-modifying therapies (DMTs) in multiple sclerosis (MS). It also discusses their strengths and weaknesses and the methodological challenges. Results suggest that sustained early treatment in MS patients can delay progression to significant disability and the studies support the concept that early treatment with DMTs has long-lasting effects. Long-term follow-up of patients have significant limitations, including loss of randomization and blinding, incomplete ascertainment, and the absence of an appropriate comparator. Long-term non-randomized observational trials (NROTs) include a wide range of study designs, such as prospective and retrospective cohort studies, case-control studies, and cross-sectional studies, with the common feature that any intervention studied is determined by clinical practice and not by the protocol. Several epidemiological and statistical methods are available to deal with confounding, in both design and analytical phases of NROTs.