It is an open access study though I would not recommend reading it unless you are up for a lot of math.
The authors of the study developed "mathematical models of the spread of self-medicative treatments for medical conditions to explore the factors that lead to treatments becoming widespread, and how a treatment's efficacy affects its rate of spread."
"This study offers a simple, novel and counter-intuitive hypothesis for the prevalence of ineffective medical treatments: unbiased copying of new treatments can frequently lead to the prevalence of ineffective practices because such treatments are demonstrated more persistently than efficacious alternatives, even when there is enhanced abandonment of ineffective cures."
I think I like how the conclusion was summarized in article on the New Scientist site -
Under a wide range of conditions, quack treatments garnered more converts than proven hypothetical medicines that offer quicker recovery, Tanaka found. "The very fact that they don't work mean that people that use them stay sick longer" and demonstrate a treatment to more people, he says.
Or to summarize, ineffective treatments tend to spread because of the fact that they leave the person suffering. You would think that a moderately intelligent person would stop preaching about the benefits of something that wasn't working for them but the conclusion was proven by mathematical models so it must be correct.
The thing that I find most humorous about this article is how it lumps complementary medicines, traditional remedies, and home cures in with witchcraft. Talk about a loaded association - that would be like listing a group of politicians and adding Hitler into their ranks.