Although we could not find
estimates of fan use in the US population, we examined the
potential selection bias by evaluating characteristics associated with SIDS risks in study participants and nonparticipants in a previously published article.16 Using California birth certificate data, we compared the ORs for maternal age, maternal education, parity, birth weight, infant sex, and late initiation of prenatal care (> 5 months» gestation).
Importantly, our null effect
estimates from the random experiment differ substantially from those found from an analysis of CPS data, raising concerns about the
potential for
selection bias in non-experimental
estimates of returns.