Repeat abortion: new insights.

Tietze and Jain 4 years ago analyzed the phenomenon of repeat abortion by simple population models based on equal numbers of women at each year of age within the 15-44 year range each model having a constant abortion rate/1000 women per year starting with the year abortion is legalized and disregarding abortions prior to legalization. Using computer simulation methodology this paper attempts to expand the approach by Tietze and Jain by introducing realistic age pattern of abortion rates based on recent experience in the US and by disaggregating repeat abortions into 2nd 3rd and higher order terminations. Results are shown for 3 sets of model populations with total abortion rates (TAR) of 300 600 and 1200/1000 women corresponding to annual abortion rates of 10 20 and 40/1000 women 15-44 in a rectangular population i.e. one with equal numbers of women at each year of age. A TAR of 300 was slightly lower than the current TAR of the UK; a TAR of 600 approximates the TAR of the white population of the US; a TAR of 1200 is somewhat lower than the TAR of blacks in the US. The age pattern of abortion rates used for each of the models reflects the high abortion rates in the late teens and early 20s and their rapid decline among older women observed in the US in recent years. In this report hetergeneity with respect to the use of abortion was simulated by combining several segments with different TARs into a single model. The purpose of using arbitrarily selected segments to simulate heterogeneous cohorts or populations was to identify patterns and trends and to illustrate orders of magnitude. The technique cannot establish with precision what actually happened or predict what will happen in the future but confidence is enhanced if it can be shown that the estimates closely match observed values. It is concluded from the simulations that the increasing numbers and proportions of repeat abortions including higher order repeat abortions in the US can be adequately explained by the increasing numbers of women at risk as the heteroegeneity of the populations in regard to both the need for and the utilization of abortion services. Rather than a deterioration of contraceptive practice following abortion the same factors explain the higher levels of repeat abortion rates compared with 1st abortion rates.