EVALUATING ACCEPTANCE STRATEGIES FOR TIMING OF POSTPARTUM CONTRACEPTION

What is the best time after childbirth to accept contraception? The problem is to reduce overlap with postpartum anovulation, during which contraception is redundant, but simultaneously to reduce the proportion of women conceiving before contraception is inaugurated. Efficiencies of three classes of acceptance strategies at resolving this dilemma are investigated. Particular attention is paid to the postpartum acceptance rule because of its special administrative advantages. Its efficiency relative to those of other rules is found to be enhanced by conditions of short postpartum anovulation, a high proportion of ovulatory rather than anovulatory first menstrual cycles, a high level of natural fecundability, and, especially, a high level of contraceptive continuation.; It is difficult for family planning programs to determine how soon after childbirth couples should be encouraged to start contraception. There may be different optimal acceptance strategies for various circumstances, and determining which strategy is best for a given target population could have considerable significance. The problems to be addressed are: 1) how can these acceptance strategies be evaluated to determine which is best; and 2) are there different optimal strategies for different circumstances. Conditions favoring early (low-T) strategies are short anovulation, high proportion of ovulatory 1st cycles, high natural fecundability, and good continuation. Favoring high-T rules, and particularly the postamenorrheic policy, are lengthy anovulation, low natural fecundability, low frequency of ovulatory 1st cycles, and poor continuation. The present analysis is based on a recent extension of the algebra designed to encompass the set of mixed-T strategies (Potter et al., 1978) allowing evaluation of optimum strategies under a variety of circumstances. Each cohort of women is followed from childbirth through anovulatory and fecundable periods to next conception by means of a discrete-time, expected value probability model. The results show clearly that there are variations in optimal acceptance strategies, although the differences are often small. From a policy perspective, it might be most useful to focus attention on the postpartum strategy which ordinarily has substantial administrative advantages.