Selective Fecundability and Contraceptive Effectiveness

Abstract

A study based on an extended model using the Type I-geometric distribution fitted by Jain to a sample of 2190 Taichung women was conducted to confirm Tietzes findings that in the absence of contraception the fecundability composition of a group of women changes rapidly by a selective process whereby more fecund women become pregnant leaving behind an increasingly subfecund group. Tabulated results show that the process becomes progressively slower with increasingly effective contraception. Mean values of fecundability are given according to effectiveness of contraception and months of fecundable exposure without pregnancy. The proportion of women accidentally conceiving within 2 years of contraceptive use decreases with increasing contraceptive effectiveness and with the length of preceding period of nonprotection during which the proportion of subfecund women is increased. After 4 years of nonprotection the rate of accidental pregnancy is reduced by a factor of 3. Potential sources of bias facing any comparison of contraceptive methods such as unequal length of observation of women using various methods and amount of overlap or interval between time of contraceptive initiation and amenorrhea are discussed with respect to family planning programs. IUD effectiveness for different age groups of women (under 24-39) in Taichung is discussed. An extensive appendix describes mathematical details relating to the model.