A new randomized control trial of Text4Baby in military families returned mixed results -- women who engaged in the intervention were more aware and knowledgable about healthy pregnancy strategies, but, at least during the four-week study period, this was not reflected by any significant change in behavior. Study author Douglas Evans thinks a follow-up study to be published later this year will likely show that the behavior change came in later months.

Text4Baby is a well-known text messaging service for new and expected mothers, which boasts that it's reached more than 700,000 new mothers since its 2010 launch.  It has also been validated in a number of studies from San Diego to Russia. However, according Evans, of the Milken Institute School of Public Health at The George Washington University, few of the studies so far have been randomized control trials that compare patients who use Text4Baby to those who don't.

"The other studies that have been done have all been basically observational types of studies, much less rigorous designs, he said. "The work that I've published is, to my knowledge, the only randomized control trials. ... The other studies are really just asking people who participated in Text4Baby what they thought of it, what their reactions to it were. ... Here, the result of this study is we can say that Text4Baby actually has an effect because we have a comparison to the absence of Text4Baby."

The latest study was conducted with 943 women at Madigan Army Medical Center in Tacoma, Washington from December 2011 to September 2013. They were interviewed before signing up for Text4Baby and again after four weeks. 

First of all, researchers found adherence to the study was very high -- only 1.9 percent of women dropped out of the service at any point during the four weeks. In terms of attitudes and knowledge, Text4Baby participants were more likely than non-participants to have a positive change in attitude on both the importance of visiting a primary care provider during pregnancy and the importance of not drinking during pregnancy. When results were adjusted for marital status and race, there was also an improvement in apprehension of the importance of prenatal vitamins.

The survey included self-reported behavior questions as well as knowledge and attitude ones, but, four weeks in, there were no significant behavioral results. However, Evans attributes this to the relative brevity of the four-week interval.

"There's more to come from this project. We have data on our sample, the portion of the sample we were able to follow, all the way through their post-partum survey. This paper looks at two time points of data; the final paper, that we hope to have out in several months, looks at four points of data. There we'll be able to answer the question of whether or not Text4Baby produced behavior change, and that's really the bigger question that we're hoping to answer one way or another."