ARTICLES Relationships Between Contraception and Abortion: A Review of the Evidence By Cicely Marston CONTEXT: The relationship between levels of contraceptive use and the incidence of induced abortion continues to and John Cleland provoke heated discussion, with some observers arguing that use of abortion decreases as contraceptive prevalence rises and others claiming that increased use of family planning methods causes abortion incidence to rise. At the time this article METHODS: was written, Cicely Abortion trends are examined in countries with reliable data on abortion and with contraceptive preva- Marston was a post- lence information from two points in time showing increases in contraceptive use. The role of changes in fertility in doctoral research mediating the relationship between abortion and contraception is also explored. fellow at the Centre for Population Studies, RESULTS: In seven countries—Kazakhstan, Kyrgyz Republic, Uzbekistan, Bulgaria, Turkey, Tunisia and Switzerland— London School of Hygiene & Tropical abortion incidence declined as prevalence of modern contraceptive use rose. In six others—Cuba, Denmark, Nether- Medicine; she is lands, the United States, Singapore and the Republic of Korea—levels of abortion and contraceptive use rose simultane- currently lecturer in ously. In all six of these countries, however, overall levels of fertility were falling during the period studied. After fertility social science and levels stabilized in several of the countries that had shown simultaneous rises in contraception and abortion, contracep- public health in the tive use continued to increase and abortion rates fell. The most clear-cut example of this trend is the Republic of Korea. Department of Social Science and Medicine, Imperial College, CONCLUSIONS: Rising contraceptive use results in reduced abortion incidence in settings where fertility itself is con- London. John Cleland stant. The parallel rise in abortion and contraception in some countries occurred because increased contraceptive use is professor of alone was unable to meet the growing need for fertility regulation in situations where fertility was falling rapidly. demography at the International Family Planning Perspectives, 2003, 29(1):6–13 Centre for Population Studies. Common sense and an elementary understanding of the ity starts to fall, an increasing proportion of couples want biological determinants of human reproduction indicate no more children (or want an appreciable delay before the that contraception and induced abortion represent alter- next child), and exposure to the risk of unintended preg- native means of achieving the same aggregate level of fer- nancy also increases as a result. In the early and middle phas- tility in a population. If fertility and its other determinants es of fertility transition, adoption and sustained use of ef- (sexual exposure, lactation and pathological infertility, for fective methods of contraception by couples who wish to example) remain constant, a rise in contraceptive use or in postpone or limit childbearing is still far from universal. effectiveness of use must lead to a decline in induced abor- Hence, the growing need for contraception may outstrip tion and vice versa. use itself;4 thus, the incidence of unintended and unwant- Why, then, does the relationship between levels of con- ed pregnancies rises, fueling increases in unwanted live births traceptive use and the incidence of induced abortion con- and induced abortion. In this scenario, contraceptive use tinue to provoke heated discussion?* And why do some and induced abortion may rise simultaneously. observers claim that increased contraceptive use leads to As fertility decreases toward replacement level (two births higher abortion rates?1 per woman), or even lower, the length of potential expo- The reason for the confusion stems from the observation sure to unwanted pregnancies increases further. For in- that, within particular populations, contraceptive prevalence stance, in a society in which the average woman is sexual- and the incidence of induced abortion can and, indeed, often ly active from ages 20 to 45 and wants two children, do rise in parallel, contrary to what one would expect. The approximately 20 of those 25 years will be spent trying to explanation for these counterintuitive trends is clear.2 In avoid pregnancy. Once use of highly effective contracep- societies that have not yet entered the fertility transition, tive methods rises to 80%, the potential demand for abor- both actual fertility and desired family sizes are high (or, to tion, and its incidence, will fall. Demand for abortion falls put it another way, childbearing is not yet considered to be to zero only in the “perfect contraceptive” population, in “within the calculus of conscious choice”3). In such soci- eties, couples are at little (or no) risk of unwanted preg- *See for example, U.S. Senate debate, 105th Congress, 1st Session, Feb. 25, 1997, 2:15 P.M., Vote No. 13, on Mexico City Policy. Those in favor of pass- nancies. The advent of modern contraception is associated ing the bill declared, “It is a very arguable assumption at best to say that the declining abortion rates [seen in selected developing countries] are a with a destabilization of high (or “fatalistic”) fertility pref- direct result of pregnancy prevention services.” A synopsis is on the Inter- erences. Thus, as contraceptive prevalence rises and fertil- net at <http://www.senate.gov/~rpc/rva/1051/105113.htm#HEADING/>. 6 International Family Planning Perspectives which women are protected by absolutely effective con- FIGURE 1. Rate of induced abortion and cumulative per- traceptive use at all times, except for the relatively short pe- centage of women using an IUD, by years since first birth, riods when they want to conceive, are pregnant or are pro- Shanghai, China 5 tected by lactational amenorrhea. Because such a state of Rate % perfect protection is never actually achieved, a residual de- 20 90 mand for abortion always exists, although its magnitude 80 varies considerably among low-fertility societies, accord- 70 ing to levels of contraceptive use and choice of methods. 15 60 The purpose of this article is to summarize what is known IUD use (cumulative %) about the relationship between abortion and contraception. 50 10 Rate of induced abortion We start with a description of a recently proposed model of (per 1,000 mos.) 40 the relationship,6 and provide empirical illustrations to as- 30 sess the validity of this model. We then review trends over 5 20 time in the incidence of abortion and contraceptive use for 10 specific countries based on published articles. Finally, we 0 0 present a comprehensive examination of such trends in all 12345≥6 countries thought to possess reliable trend data on abortion Years since first birth and contraceptive use and in which major changes in con- traceptive prevalence or effectiveness have been recorded. Source: reference 9. MODEL AND EMPIRICAL ILLUSTRATIONS The expected relationship between abortion and con- Bongaarts and Westoff have identified and described math- traception can be seen in the first of our three illustrations, ematically parameters that account for the relationship be- which comes from a prospective study of married couples tween contraception and abortion levels.7 They show that in Shanghai, China.9 The women have all had one child and, the abortion rate (the number of abortions per 1,000 because of the one-child policy, second births are extremely women of reproductive age) in a population is related to rare. Following the birth of their first child, many women the number of years in which women are both fecund and initially use relatively ineffective methods: withdrawal, pe- exposed to the risk of childbearing by being sexually ac- riodic abstinence and condoms. With each successive year tive, the reproductive time taken for each live birth and the following childbirth, use of these methods is progressive- reproductive time for each abortion. The latter two time pe- ly replaced by use of the IUD—a highly effective method. riods constitute waiting time to conception, pregnancy and The proportion of women using IUDs rises from 20% in the postpregnancy period of insusceptibility. Abortion rates the first postnatal year to more than 75% in the fifth post- are also related to the prevalence and effectiveness of con- natal year (Figure 1). In the first year following childbirth, traceptive use, and the probability of aborting unintended the induced abortion rate is 16 abortions per 1,000 months pregnancies. The total fertility rate (TFR) equals the num- of exposure. By the fifth year, the rate is close to zero, thus ber of births a woman would expect to have over her life- providing a vivid demonstration of the trade-off between time under prevailing fertility rates, and it comprises two contraceptive effectiveness and induced abortion. components: intended births and unintended births. The The second illustration involves a compilation of data total abortion rate is the number of abortions a woman on abortion and contraceptive use for a set of countries hav- would expect to have under prevailing abortion rates, and ing approximately the same level of fertility and reliable in- can be linked mathematically to the TFR using the para- formation on contraception and abortion. The countries meters just described. selected had a TFR between 1.7 and 2.2 births per woman, The mathematical links between the TFR and abortion because this choice maximized the number of countries rates and the derivation of these links have been described and years that could be included. Relevant information was in detail.8 By varying the different parameters one by one, obtained for a total of 36 time points from 11 countries. the authors examine the relative effect of each factor on the We included all available data points from all periods in abortion rate. The practical implications of the model in- which the TFR was within the specified range. clude the following: Early onset of sexual activity, leading Figure 2 (page 8) shows a plot of abortion rates and preva- to a longer sexual and reproductive span, will tend to be lence of modern contraceptive use.