The study by Vincent Odekerken and colleagues is well designed and well executed, but the primary results cannot be interpreted to make recommendations for patient care-an outcome addressed only indirectly in the article. The results showed no differences in the effects of subthalamic nucleus or globus pallidus interna deep brain stimulation on the Academic Medical Center linear disability scale (ALDS) or in the incidence of adverse behavioural effects. However, the hypothesis testing was underpowered and, consequently, there is little assurance against the occurrence of a type II error (ie, not finding a difference when one truly exists). The larger than anticipated variance in the ALDS meant that the study had a less than 50% chance of avoiding a type II error. The sample size for the incidence of adverse effects would have required a difference in the incidence of such effects between the two groups of at least 20% to have a reasonable chance of avoiding atype II error. If the authors wanted to make a claim of equivalence between the two targets for deep brain stimulation, then other approaches would have been more suitable.
展开▼
