In countries such as Nepal, where abortion is severely restricted by law, most women seeking abortions and most people providing abortion services must do so clandestinely. Often, women in rural areas and poor women in urban areas who wish to terminate a pregnancy must obtain services from untrained professionals in environments failing to meet basic, minimal medical standards.1 Most of these abortions may thus be considered "unsafe."2

Due to the fear of prosecution or social stigma, information relating to abortion is often difficult to obtain from either providers or clients. In countries such as Nepal, patients who present at a hospital with an incomplete abortion or with abortion-related complications cannot be denied care, even though abortions are restricted. Thus, hospitals and clinical facilities serve as logical and convenient places to gather information regarding abortion, as patients in hospitals and clinics are often more willing to consent to be interviewed and to provide information. For this reason, the limited previous research on abortion in urban Nepal largely focused on patients in hospitals and clinics.

A 1984 study of five hospitals in and around the Kathmandu Valley identified 165 possible cases of induced abortion from among 1,576 abortion-related cases over a one-year period.3 The study also revealed that many abortion service providers in urban areas were untrained, and therefore lacked knowledge about dangerous, unhygienic and sometimes fatal practices.

A six-month prospective study carried out in four public hospitals and one private clinic in Kathmandu in 1992-1994 showed that of 1,241 abortion-related cases, 234 were identified as induced abortion.4 The researchers followed up 130 (56%) of the 234 women who had induced abortions twice in the subsequent 15 months to study the women's reproductive behavior.* Most women receiving abortion services from trained medical professionals for pregnancy termination were educated and sought first-trimester procedures. In contrast, women seeking services from untrained providers seemed prepared to take considerable risks to terminate unwanted pregnancies. Although postabortion contraceptive use was high, some women resorted to repeat abortions due to method failure or other factors.

Neither of these studies included women who sought early termination of pregnancy. Furthermore, these studies were mostly limited to public hospitals, where the cost of services is considerably lower than in private facilities.

According to a recent survey among health professionals in Asia, approximately one-third of women who undergo abortion develop medical complications requiring hospital admittance.5 Therefore, the hospital-based cases, even when respondents provided accurate and full information, represent only one extreme: women who experience complications of abortions. Among those who do not experience complications, the incidence of abortion and the reasons for seeking services remain largely unknown. To develop a comprehensive understanding of the demand for and supply of services, information from this population is also needed.

In this article, we present that side of the story: women in Nepal seeking pregnancy termination from qualified providers working in standard medical facilities. In other words, we report on the incidence of "safe" abortions in Nepal. The percentage of women seeking safe services is most likely considerably smaller than that of women seeking unsafe services. Also, women who receive safe abortions are generally more educated and are from relatively higher economic backgrounds.6Even so, it is still important to understand how and why these women obtain abortions. Only with both perspectives can efforts that encourage women to utilize safe practices be made more effective. We know of no previous studies that provide primary data on women seeking termination of pregnancy in Nepal.

Data and Methods

In 1997, data were collected by interview with and clinical examination of all women who attended a private clinic in Kathmandu to obtain induced abortion. In all, data were collected from 160 women. All of the abortions were by manual suction evacuation, were within 10 weeks of the last menstrual period and were performed by one senior female gynecologist. The average total cost for the services was Rs3,000 (approximately US$50) per client. After the procedure, patients were advised to return to the clinic after one week for a follow-up visit. During the subsequent visit, they also received contraceptive counseling.

We analyzed the women's social and demographic information and their reasons for seeking early pregnancy termination at the clinic. We compared the women's profiles with those of women from two other data sets: data from the Nepal Family Health Survey, carried out in 1996 as part of the Demographic and Health Surveys project;7 and data from 1998 on all gynecology patients admitted to the largest maternity hospital in Kathmandu.8

Finally, we examine here the extent to which the fertility behavior of women in Nepal differs from that of a "natural fertility" population (i.e., one in which the practice of contraception or abortion is negligible). In such societies, marital fertility varies by women's age according to a standard pattern. When marital fertility is controlled (through the use of contraceptives or induced abortion), the age-specific birthrate declines more steeply with age than is seen with the standard pattern—i.e., older women prevent births more rigorously than younger women because a higher percentage of the former do not want any more children.9 Using a standard age-specific schedule of natural fertility,10 we can calculate an index of the degree of control of marital fertility, based on age-specific marital fertility rates. In this analysis, we estimated such indexes for urban and rural women in Nepal and contrasted the resulting patterns with the standard pattern for a natural fertility population.


The frequency with which clients sought induced abortion was highest in mid-November to mid-December, the period immediately following Nepal's biggest national festival, which lasts nearly a month. The festival is a special time for families, as people who have been away often return home to be together. Mid-May to mid-June, the month immediately following the Nepalese New Year, was another period that had a high frequency of clients seeking induced abortions.

Thirty-six percent of the women surveyed were between 25 and 29 years of age (Table 1) and 43% had two children—generally one of each sex (not shown). A majority of the women were somewhat educated; two-fifths had more than a high school education, while only one-fifth were without any formal education. One-fifth of the women surveyed were employed in the service or business sector. The overwhelming majority (91%) lived within the Kathmandu Valley.

Just under half of the women (48%) reported using some method of contraception prior to becoming pregnant (Table 2). Condoms were the most frequently used method (15%), followed by hormonal injectables (13%), the pill (8%) and the IUD (6%). Three women (2%) reported using hormonal implants and one woman's husband had had a vasectomy (less than 1%).

Thirty-four percent of women surveyed cited the desire for no more children as their primary reason for seeking pregnancy termination (Table 2). Sixteen percent chose to undergo abortion because their most recent child was still too small for them to take care of another. Although birthspacing was the main concern for many of the women surveyed, 10% chose abortion because of poor economic condition, 11% because of contraceptive failure and 1% because they were not yet married.

For additional perspectives on the social and demographic characteristics of women who underwent induced abortion, we analyzed data from two other independent sources. The data for both rural and urban Nepal from the Nepal Family Health Survey are presented in Table 3. The survey included questions for ever-married women aged 15-49 to ascertain the prevalence of spontaneous as well as induced abortion.§ Few women reported ever having had an induced abortion; given the highly restrictive abortion laws, the data are most likely underreported. Nonetheless, these data refer to the most recent countrywide sample of women of reproductive age, and are presented to help us discern any basic characteristics and patterns.

The 1996 data show that a considerably higher percentage of younger women in urban areas than in rural areas reported ever having had an induced abortion. Since no distinction could be made as to where or from whom the services were obtained, the data most probably include both safe and unsafe abortions. Half of women living in urban areas who reported an abortion were between 15 and 29 years of age, in contrast to 34% of the women in rural areas. Parity was also considerably lower among women living in urban areas: Whereas 55% of women in urban areas who had had an abortion, had fewer than three living children, only 37% women in rural areas did so. Among urban women who ever had had an abortion, 42% had at least a high school education. In contrast, only 4% of women living in rural areas who had had an abortion had a high school education or higher. Although these data are limited, overall they indicate that urban women who report ever having had an abortion were relatively younger, of lower parity and more educated than rural women.

Table 3 also includes data from a 1998 survey of the abortion experiences of all gynecology patients admitted to the largest maternity hospital in Kathmandu. Of the 1,116 women admitted to the hospital because of postabortion complications who were interviewed for the study, 74 (7%) were clinically determined to have had an induced abortions (not shown).**

The women in this survey were of similar age and parity as the women of the 1996 all-Nepal survey; however, they differed somewhat with respect to educational background. A greater percentage of women from the 1998 survey than from the urban all-Nepal sample had little or no education (51% vs. 29%). One reason for this difference might be that the particular hospital is a highly subsidized referral hospital that tends to attract women of low socioeconomic background and from outside the Kathmandu Valley. In the 1998 urban hospital sample, nearly half (49%) of the patients were from outside the capital city; they were most probably referred because appropriate services were unavailable where they lived. Overall, these data clearly indicate a preference for fertility regulation at lower parity among urban women who have ever had an abortion.


Each of the three data sources examined here has relative strengths and weaknesses. Although the clinical data lack the necessary denominators to allow us to estimate the percentage of women seeking services for early termination of pregnancy, those data provide information about women seeking services. The data on women attending clinics and hospitals because of postabortion complications represent only those who experience morbidity, yet they provide insight into what might be happening in the communities. The survey among all women of reproductive age is helpful in that it provides the overall prevalence or incidence of abortion (lifetime or for a specific time period), but surveys of spontaneous and induced abortions are known to suffer from a large degree of underreporting.11 The reported pattern of incidence may be relatively more correct than the level of incidence.

In view of their relative strengths and weaknesses, the three data sources utilized here in understanding the profile of women seeking abortion in urban Nepal should be considered as complementary. They reveal a generally consistent pattern: Women in Nepal seeking abortion services are young, and the majority already have at least one child.

The demand for services must be viewed in the context of changes in reproductive attitudes and preferences, especially in the urban population of Nepal. As of 1996, ever-married women aged 15-49 living in urban areas desired 2.4 children, on average. The total fertility rate among all women (based on data from women aged 15-49 for the three years preceding the survey) was 2.9 lifetime births per woman (although it was 4.8 per woman in rural Nepal). However, the "wanted" fertility rate, which indicates the total fertility rate if all unwanted births were avoided, was only 1.9 lifetime births per woman in urban Nepal.12 Clearly, there is an important difference between the actual and wanted fertility rate.

A significant percentage of women in urban Nepal are practicing contraception. As of 1996, just over 45% of currently married women of reproductive age in urban areas were using a modern method.13 Consequently, the fertility patterns of urban women are very different from those of the "natural fertility" population.†† Figure 1 shows just how markedly fertility control has changed among married couples in urban Nepal. The overall degree of fertility control within marriage is 1.5 for urban Nepal (but is 0.6 for rural Nepal), based on the application of the Coale-Trussell model. The pattern of urban Nepal indicates just the opposite of the natural fertility regime, suggesting the emergence of a significant degree of fertility control within marriage in the urban population.

Obviously, a culture of contraception and small family size has become more common among women in urban Nepal. The data suggest that levels of effective contraceptive use could be even higher if certain improvements were made: If people using contraceptives, particularly those who do not want any more children, would either use more effective methods or use methods, such as condoms, more effectively; if there were increased promotion of emergency contraception to decrease the percentage of women who cite "unplanned" pregnancy as their primary reason for seeking abortion; and if there were increased availability of contraceptive counseling for those who do not want additional children.


A desired family size of two children is now the norm in urban Nepal. This is a very different situation than 10-15 years ago. Clearly, fertility transition is underway in urban Nepal, with fertility being controlled within marriage by contraceptive use and to some extent by induced abortion. In urban Nepal, a significant percentage of women cite the desire for no more children as their primary reason for seeking pregnancy termination, particularly in the capital city of Kathmandu. The overwhelming majority of women seeking termination have at least one living child.

To improve abortion prevention, the data clearly suggest the need for increased use of contraceptive methods, including emergency contraception. On the curative side, the availability of menstrual regulation by manual vacuum aspiration should be considered the first and foremost policy option for addressing the problem of unwanted pregnancy in Nepal. Menstrual regulation has been acceptable to women around the world, including in countries like Bangladesh, where abortion is not permitted.14 Easier and regular availability of menstrual regulation services could go a long way toward reducing health problems and misery for many women in Nepal.‡‡ Without access to such services, only a small percentage of Nepalese women, mostly those of better socioeconomic conditions, will have access to safe abortion services; other women will likely have no choice but to submit to the inherently high-risk and high-priced clandestine abortions.