Stop me if I am utterly misreading this, but surely several of the outcomes of the study *were* statistically significant? According to Table 2, the P-values for breastfeeding rates at 3 months and for EBF rates at both 2 months and 3 months were all less than 0.05, which would indeed make those outcomes statistically significant at the conventionally-chosen 95% level.

There is in fact *massive* precedent for research into treatment of people at high risk of developing a problem to see whether pre-emptive treatment reduces the risk of the problem developing. Yes, to take your example, there has indeed been research into the efficacy of using diabetic treatment (metformin) in people at high risk of developing diabetes.

Curious as to the ways in which you feel the authors demonstrated a lack of understanding of statistics? (The obvious one - the small sample size - doesn't seem to have been their decision, since they assessed a much larger sample for inclusion but the majority of the mothers approached refused enrollment.)


Best wishes,

Dr Sarah Vaughan
MBChB MRCGP

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

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