Thursday, June 17, 2010

Clinical Research Oriented Workshop (CROW) Meeting: June 16, 2010

Present: Abby Crocker, Kairn Kelley, Ben Littenberg, Carole McBride, Charlie McLean

1. Round table discussion
2. Discussion of Kairn's F31 proposal development for August submission
Lessons of the day:
  • If you select the worst cases to study, you will see improvement no matter what the intervention. (Regression to the mean) RCTs avoid over-interpreting this improvement by comparing the amount of change between groups.
  • Don't conflate marketing with evidence.
  • Design research that answers the question you intend to ask.
Assignment: Explain regression to the mean to an innocent bystander (any volunteers? I asked my husband if he'd like me to explain regression to the mean to him, and he politely declined.)

Action for Kairn: evaluate the existing evidence for Fast ForWord intervention before deciding which question to ask.

Next Week: Run Charts (Control Charts) discussion led by Abby. Ben requests that all attending bring a sequence of measures over time for demonstration. Could be your weight, the temperature at noon... anything where the collection method was the same each time. The data could be in Excel or on a piece of paper!

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.