Here is a recap of the stories that appeared last week at Science-Based Medicine, a multi-author skeptical blog that separates the science from the woo-woo in medicine.

Do clinical trials work? It depends on what you mean by “work” (David Gorski) Evidence-based medicine tends to worship the randomized clinical trial as the only valid method of investigation. A recent article questioned whether clinical trials even work. They do, and they’re the best method we have; they will improve as science evolves. Subsets of patients may respond to a treatment that “doesn’t work” for a whole population, and we can learn to identify them.

The Business of Baby and the Monkey Business of Margulis (Harriet Hall) Journalist Jennifer Margulis has written a dreadful, dangerous book, “The Business of Baby.” She correctly identifies many areas where modern maternity care needs improvement, but she is anti-vaccine, pro-home births, and anti- pretty much everything mainstream doctors do, accusing them of acting only out of financial motives. She uses cherry-picked information, inflammatory language, distortions, and fallacious reasoning to support her “natural” agenda.

Researching SBM Online (Steven Novella) Dr. Novella offers a useful primer for doing your own research on science-based medicine topics. How to use Google, how to search PubMed, how to judge the quality of resources, and much more.

Doing Eric Merola a Favor (David Gorski) Merola is temporarily offering free access to his travesty of a “documentary” about cancer quack Burzynski. Viewers can judge for themselves.

I’ve been prescribed an antibiotic. Should I take a probiotic? (Scott Gavura) There is some (possibly biased) evidence that probiotics reduce the incidence of antibiotic-related diarrhea and can prevent C. diff infections. Probiotics offer promise, but we lack information about the most effective species, the optimal dose, the timing and duration of treatment.

Human Sex Determination: Psychic Sperm and the Gambler’s Fallacy… (Clay Jones) Cognitive biases, mental shortcuts, poor understanding of probability and statistics, and common illusions lead us to misinterpret phenomena like clustering. We fail to appreciate that randomness is lumpy. The gambler’s fallacy suggests that if a woman has 3 boys, her next child is more likely to be a girl. Not so: the likelihood is 50/50 for each birth.