There is increasing interest in the problem of designing clinical trials to assess the ability of treatments to modify the course of progressive neurological diseases rather than simply relieve symptoms. In the absence of reliable and valid biochemical or neuroimaging measures of disease progression, it is necessary to rely on clinical outcomes to demonstrate disease modification. A number of two-period designs have been introduced in the literature to evaluate the effects of treatment on progressive diseases, including the randomized start and randomized withdrawal designs. These designs, in theory, appear to be ideal for distinguishing between symptomatic and disease-modifying effects of an intervention, but there are many practical difficulties in their implementation. These designs will be reviewed and their associated logistical and analytic challenges will be discussed.