Sexual abstinence programs have the potential to reduce the incidence of un
planned pregnancies and sexually transmitted diseases (STDs) among adolesce
nts. Effectiveness measures are needed to help researchers assess the impac
t of sexual abstinence promotion programs on STD and pregnancy rates and to
enable comparisons of abstinence effectiveness with other contraception an
d STD prevention methods. Abstinence "failure rates" have been proposed as
one measure of program effectiveness. However, the concept of abstinence fa
ilure rates has not been adequately operationalized. The present study exam
ines a novel mathematical framework for estimating abstinence failure rates
, both theoretically and empirically. Examples are provided, and the advant
ages and disadvantages associated with the mathematical model-based approac
h are discussed.