Episode notes
The weekly consolidation for the week of September twenty-first, covering four episodes — Monday's sweep, Parkinson disease, pediatric headache, and the GLP-1 outcomes dive. It leads with six things worth keeping: the REM sleep behavior disorder question as the highest-yield prodromal screen in neurology, the reversal on when to start levodopa, the CHAMP placebo response and what it means for any pediatric preventive you start, the tirzepatide cardiovascular approval that rests on noninferiority rather than superiority, the finding that about half of patients on long-term opioids succeed with a well-supported taper and nothing added, and the rule that pediatric migraine criteria are not the adult criteria. It closes on three appraisal threads that ran through every episode this week: composite endpoints where one component carries the result, exposure definitions that manufacture benefit through immortal time, and implausible secondary findings functioning as failed negative controls.