As an nutritional epidemiologist, I would not want to see a repetition of research from so-called scientists that don't even know the basics of statistics.  Or perhaps really did know and employed the tactics that have been often used when there is a profit motive to be gained by trying to hide potential side effects.  If I were unethical and wanted to design a study and claim my product had no negative effect I would:

a) Make sure the sample size was really small. That way I would have a much better chance of claiming that there was no statistical significance.
b) Make sure to keep the duration of time between administration of the treatment and the measurement as short as possible so that there would be less time for large side effects to develop.
c) Make sure to pick outcome variables for side effects that are least likely to occur and most likely to be influenced by multiple factors other than the treatment.

Furthermore I am dumbfounded as to how anyone who has any smidgeon of a science background can fail to understand that if you have a study that has no statistical significance it proves NOTHING.  Zero, zip, nada.  They did NOT prove that early use for formula is safe with this study by any criteria known to epidemiology or statistics.

As for human ethics, it is not considered "ethical" to treat a normal condition ever.  If you treat a normal person who doesn't not have diabetes with insulin that would be unethical.  A weight loss of 5% is not at a level of medical necessity to warrant "treatment" with formula.  

As a reviewer of peer-reviewed journal articles, I would have put this study into the category of "not acceptable even with revisions".  I shudder at the thought of this particular group of researchers doing any further research on the topic for the following reasons:

a) they show a serious lack of understanding of existing evidence-based protocols for when supplementation is warranted
b) they show a serious disregard for basic statistics and epidemiology.

Finally, beyond epidemiology the neglected element now in those who dabble in statistics without fully understanding epidemiology is that you also have to establish the plausible relationships that you are investigating.  I can think of at least a dozen different interpretations of their findings, but the study was so abysmal that I didn't even bother to look further until seeing the shocking notion on Lactnet that these authors would be suitable for further investigation of a topic they so woefully misunderstood.

Sincerely, 

Susan E Burger, MHS, PhD, IBCLC

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

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