Repeat Abortions Are Common in Vietnam, May Be Linked to Son Preference

First published online:

Although sex-selective abortion is illegal in Vietnam, the likelihood that an abortion is not a woman’s first may nonetheless be linked to the gender of her children, according to a recent study conducted in three Vietnamese regions.1 In this setting—a country with one of the world’s highest abortion rates and a high incidence of repeat abortion—the odds that an abortion was a repeat (rather than first) abortion were greater among women with at least two daughters than among those with only one (odds ratio, 2.9), and lower among those with no sons than among those who had one (0.4). Repeat abortion was also associated with women’s age, location and number of children.

The data come from a cross-sectional study of women obtaining abortion services at 62 health facilities selected to be representative of the three cities or provinces in which they were located: Hanoi, Khanh Hoa and Ho Chi Minh City. The study, conducted from August to December 2011, included all hospitals and health centers specializing in sexual and reproductive health services, as well as a random selection of 50% of general hospitals. All women obtaining an abortion at a facility in the province or municipality in which they lived were eligible to participate; nearly 100%, some 1,224 women, completed the survey. Women were asked about the abortion they had just obtained, past abortions, social and demographic characteristics, knowledge and attitudes about abortion and abortion services, reproductive and contraceptive history, and contraceptive and childbearing intentions. Multivariate analyses examined the gender of a woman’s children in two ways: by using variables for number of sons (0, 1 or 2–4), number of daughters (0, 1 or 2–4) and parity (0, 1, 2 or 3–8), or using a sex composition variable with nine categories (no children, no sons and one daughter, no sons and 2–4 daughters, etc.).

Respondents were, on average, 29 years old, and half lived in Ho Chi Minh City. The majority were married (80%), had at least a high school education (58%) and had one or more children (69%). Half of women reported that they had been using contraceptives, typically either the pill or condoms, at the time the aborted pregnancy was conceived. Eighty-seven percent of women planned to use a contraceptive method in the future, and 43% reported wanting no more children.

Thirty-two percent of participating women had had a previous abortion. Women having a repeat abortion were more likely than those having a first abortion to be married (90% vs. 75%), to be from Hanoi (51% vs. 35%), to have more than one child (68% vs. 33%), to want to stop childbearing (64% vs. 34%) and to have had a surgical procedure (86% vs. 73%). Women undergoing repeat abortion were also more likely to report an intention to use a user-independent contraceptive method (sterilization, IUD or injectable; 34% vs. 26%) and less likely to report an intention to use a user-dependent method (57% vs. 69%).

In a multivariate analysis that controlled for social and demographic characteristics, the odds of having a repeat abortion were lower among women living in Khanh Hoa (odds ratio, 0.5) or Ho Chi Minh City (0.6) than among those in Hanoi. The odds were higher among women aged 35 or older (3.0) than among those aged 20–24, and among women with two children (2.2) than among those who had one. The odds that an abortion was a repeat abortion were reduced if a woman had no sons (0.4) rather than one, and elevated if she had 2–4 daughters (2.9) rather than one. In the alternative approach that used nine categories for family sex composition but did not adjust for parity, the odds of a repeat abortion were lower among women with no children and those with no sons and one daughter (0.3 for each) than those with one son and one daughter.

The researchers conclude that preference for a son may drive some women’s abortion-related decisions and state that their findings are consistent with previously established "correlations in Vietnam among son preference, the high abortion rate, and the sex ratio imbalance at birth." The researchers also note the heavy dependence of the study population (and especially of those obtaining repeat abortions) on less effective contraceptive methods, including the pill and the condom, as well as a larger pattern of concurrently high levels of both contraceptive use and repeat abortion. They suggest that the Vietnamese national family planning program should improve the availability of long-acting, reversible contraceptives and "determine how to adequately support women presenting for terminations to adopt and continue using the contraceptive of their choice."—H. Ball


1. Ngo TD et al., Risk factors for repeat abortion and implications for addressing unintended pregnancy in Vietnam, International Journal of Gynecology and Obstetrics, 2014, 125(3):214–246.