Sarah writes that mothers randomized to the control group could have been
excluded from follow up, but to do so would involve a serious breach of
research standards. The only acceptable way to analyze results from a
randomized controlled trial is by 'intent to treat', which means that when
someone has been placed in one group, all data from their participation in
the trial go into the group they were randomized to.
Space does not permit a full explanation of why this is necessary, but it
is an essential point, and it was a scientific milestone when journals
finally got the point. Before that, articles were rejected because
participants in one group received treatment.that was not strictly
according to the research protocol. The research is infinitely more
valuable when deviation from protocol is reported, and the possible reasons
for discrepancy form a very important part of the discussion section of the
article.
The discussion section is where you see what kind of self-criticism and
reflection the authors have and much of the time it is the most valuable
part of the researxh.
To exclude everyone whose further course of treatment deviates from
protocol is close to scientiific fraud. Users of research need to
understand why this is so.
Rachel Myr, squinting at my touch screen in the sunny cockpit of a sailboat
near Kristiansand, Norway

             ***********************************************

Archives: http://community.lsoft.com/archives/LACTNET.html
To reach list owners: [log in to unmask]
Mail all list management commands to: [log in to unmask]
COMMANDS:
1. To temporarily stop your subscription write in the body of an email: set lactnet nomail
2. To start it again: set lactnet mail
3. To unsubscribe: unsubscribe lactnet
4. To get a comprehensive list of rules and directions: get lactnet welcome