observational studies: propensity score analysis of non-randomized data

The randomized controlled trial (RCT) is considered to be the "gold standard" for providing evidence on drug efficacy. However, particularly for answering long-term questions in chronic diseases such as multiple sclerosis (MS), RCTs are often not feasible because of their size, duration, ethical constraints and costs. Data derived from observational studies complement information provided by RCTs. A major issue is that observational studies are more exposed and prone to biases, which can partly be addressed through rigorous study design or statistical analysis. Propensity score (PS) techniques are the most frequently used. PS is the probability that an individual would receive a certain treatment based on his/her pretreatment characteristics. This score is being widely used in many therapeutic areas and also in MS to adjust for the uncontrolled assignment of treatment in observational studies. However, since PS cannot adjust for unmeasured or unknown confounders, the conclusions from an observational study may not be considered as strong as those from RCTs.