Abortion in Vietnam: Measurements, Puzzles, and Concerns Daniel Goodkind Studies in Family Planning, Vol. 25, No. 6. (Nov. - Dec., 1994), pp. 342-352. Stable URL: http://links.jstor.org/sici?sici=0039-3665%28199411%2F12%2925%3A6%3C342%3AAIVMPA%3E2.0.CO%3B2-1 Studies in Family Planning is currently published by Population Council.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/popcouncil.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Tue Feb 12 02:09:53 2008

Abortion in Vietnam: Measurements, Puzzles, and Concerns Daniel Goodkind This report summarizes current knowledge about abortion in Vietnam, drawing upon government statistics, survey data, and fieldwork undertaken by the author in Vietnam throughout 1993 and part of 1994. The official total abortion rate in Vietnam in 1992 was about 2.5 per woman, the highest in Asia and worrisome for a c o u n t y with a still-high total fertility rate of 3.7 children per woman. Vietnamese provinces exhibited substantial variation in both the rate of abortion and the type of procedures performed. Among the hypotheses explored to explain Vietnam's high rate of abortion are the borrowing of family planning strategies from other poor socialist states where abortion is common; current antinatal population policies that interact with a lack of contraceptive alternatives; and a rise in pregnancies among young and unmarried women in the wake of recent free-market reforms. Because family-size preferences are still declining, abortion rates may continue to increase unless the incidence of unwanted pregnancy can be reduced, a goal that Vietnamese population specialists are seeking to achieve. (STUDIES IN FAMILY PLANNING 1994; 25,6: 342-352)

In Vietnam, as in many other countries in the world, the transition to lower rates of fertility may be attributed, in part, to rising rates of abortion. Unfortunately, despite much discussion and interest regarding abortion in Vietnam, systematic study of abortion levels and trends at the national and provincial levels has yet to be undertaken. Nor has there been a rigorous examination of the critical issue that logically should precede such an analysis -the quality of the abortion statistics. A focused study of such issues is badly needed, because confusion and uncertainty exist about the levels and trends of abortion in Vietnam, even among those currently engaged in such research. For instance, the number of annual abortions reported to have occurred in Vietnam around 1991 has ranged wildly in recent literaturefrom 1.1, to 1.6, to 2.6 million.' This report summarizes current knowledge regarding abortion in Vietnam, drawing upon government reg-

Daniel Goodkind, Ph.D. is Postdoctoral Fellow, Demography Department, Research School of Social Sciences, The Australian National University, Canberra ACT 200, Australia. Beginning in January, the author may be addressed at Brown University, Department of Sociology, P.O. Box 1916, Providence, R.I. 029120.

342 Studies in Family Planning

istration statistics, a variety of survey data, and the author's observations and fieldwork in Vietnam over the course of 1993 and part of 1994. The fieldwork included visits to rural villages and provincial health centers in several provinces in the northern Red River delta as well as in the South central area around Hue and QuangNam DaNang. Interviews and conversations with population specialists as well as abortion-service providers helped to clarify this analysis, although for a variety of reasons some of these sources are left unreferenced.

Historical Background An official population policy was outlined in the North of Vietnam in 1963, well before national reunification under socialism occurred in 1975 (Jones, 1982). Thus, Vietnam's first population policy preceded China's by several years and was relatively concurrent with those of more urbanized areas in East Asia such as Taiwan, Hong Kong, and Singapore. The incentive for the early delineation of such a policy in the largely rural and poor North Vietnam was the rapid population growth in the Red River delta, a long-settled area historically plagued by shortages of food (Le and Rambo, 1993).The stated goal of the policy was that "in the coming years, in all of North Vietnam the population growth rate must be

brought down from 3.5 percent to 2.5 percent, then 2 percent, and each family will have only 2-3 children with 5-6 years birth spacing" ("Guidance on Birth Control Activities," Directive No. 99-TTg of the Prime Minister's Office, 10 October 1963; quoted in Vu, 1992: 41). Formal attempts were not made to reach such goals, but the IUD and condom were increasingly promoted, and abortion services were available at some health facilities (Vu, 1992). However, the war of reunification between 1965 and 1975 reduced the human and material resources necessary for the formal pursuit of family planning programs. In the pre-unification South, some combination of family planning services was available in urban areas, although pronatalist policies instituted by the French in the early twentieth century and again by the Catholicbased Southern regime after Vietnam's partition in 1954 did not encourage family planning. Rural areas in the South had far less access to such services, which were particularly difficult to obtain because of wartime disruptions. For this and other reasons, fertility levels in the North of Vietnam were lower than they were in the South during the latter 1960s and the 1970s (Vu and Hanenberg, 1989; Allman et al., 1991; Goodkind, 1994a). The United States Agency for International Development (USAID) had increased its sponsorship of population and family planning programs in the South beginning in 1971, although such programs were left incomplete at the time of Vietnam's reunification (USAID, 1975). After reunification in 1975, Vietnam set about standardizing the availability of contraceptives. The Family Law of 1977 attempted to promote the protection of women's rights, which created a favorable climate for contraceptive use (Cimit, 19891, even if other factors led to a downturn in women's overall well-being after reunification (Goodkind, 199413). The Ministry of Health and the Institute for the Protection of Mothers and Newborns played the major roles in the management and implementation of such services. The 1979 census revealed a rapidly growing population of 54 million (about 72 million by 1994), which heightened concerns about family planning. In 1981,the Ministry of Health increased the availability of contraceptives and abortion services at some state-run medical facilities. Government favor toward family planning was indicated by the sharp increase in the target for use of contraceptives and abortion in 1981 compared to the previous year (Jones,1982). In 1984, the formal attempt to coordinate the family planning program with national goals to limit population growth resulted in the formation of the National Committee for Population and Family Planning (NCPFP). In late 1988, Vietnam officially introduced a national "oneor two-child" policy UPRS, 1989),which formalized a sys-

tem of childbearing guidelines, free provision of contraceptives and abortion, and economic incentives and disincentives that have been implemented in certain areas of Vietnam (Goodkind, 19944. The enactment of these population policies in the 1980s occurred at about the same time as Vietnam's official shift away from a centrally planned socialist system toward a free-market economy. These policies, known as Doi Moi (renovation), were formally instituted beginning in 1986, with the subsequent decollectivization of agriculture spelled out more fully in 1988 by Decree 10 (Forbes et al., 1991; Turley and Seldon, 1993h2Since 1989, the private sector has grown, and medicine, contraceptives, and abortion services have become increasingly available outside of public health facilities (Allen, 1993). Abortions can now be obtained through private clinics, for a fee of up to 60,000 dong (about US$6.00) or more. Government subsidies to private abortion providers began in 1991. This policy was evidently intended less as an incentive to encourage such procedures than as an attempt to incorporate abortion within the subsidized medical delivery system, with the standard amount of subsidization set at 2,000 dong (about 20 cents). No one knows what proportion of all abortions are currently obtained through private providers, although one knowledgeable observer suggests that 15 percent may be a reasonable guess. Clearly, the majority of abortions are still obtained through public facilities, which are available in cities, prosperous villages, and even remote rural areas through the employment of mobile teams. Rural women still prefer, however, to obtain services in district or provincial hospitals because of their better hygiene, safety, and training of staff; women are commonly referred to larger hospitals for more complicated, later-trimesterprocedures. The geography of abortion use is indicated by data from the 1988 Vietnam Demographic and Health Survey, which revealed that 4 percent of all abortions took place at central hospitals, 18 percent at provincial hospitals, 74 percent at districtlevel polyclinics, and only 4 percent at commune health facilities (NCPFP, 1990: 44; Allman et al., 1991).That no women reported receiving an abortion outside of these public facilities is not surprising: The survey was taken before the private provision of abortion was legalized in 1989, and those seeking such services would be especially unlikely to report having done so.

Data In Vietnam, as in many other developing countries, questions are often raised about the quality of available data (see, for example, McCarty et al., 1992).Statistical figures

Volume 25 Number 6 November/December 1994 343

often find their way into publication without any source attribution, and even if sources are listed, doubt may arise about their reliability or completeness. An example is a series of annual abortion figures for the years between 1976 and 1987 that appeared in a compilation of social statistics, the publishers of which listed no source and suggested caution in interpreting some of the data (Center for Women's Studies, 1990). These figures were evidently based on unpublished statistics from the Institute for the Protection of Mothers and Newborns, because they match data reported from that office in Jones (1982). Abortion statistics are funneled upward to the National Ministry of Health in hierarchical fashion. They are gathered by means of a set of standard forms filled out by the provincial ministries of health, which, in turn, rely on information from district ministries of health, which assemble data from the public abortion providers within their jurisdictions. The reliability of the data thus depends on the accuracy and completeness of reporting at several different administrative levels. Some provinces have special forms distributed to clinics that call for the recording of the respondent's name, address, marital status, age, and prior number of abortions. Across provinces and districts, however, the collection of such subsidiary data is not uniform. Ideally, provincial ministries should submit monthly or quarterly reports to the National Ministry of Health, but some provinces have only enough human and financial resources to permit annual or biannual reporting. Figure 1 indicates the number of all reported pregnancy terminations in Vietnam between 1976 and 1993.

Figure 1 Total reported pregnancy terminations in Vietnam, 1976-93

-

-

-

2.0 ln

C

0 .-.-

E

.- C

One- or two-child policy officially established

1.5

- -=-. 0,

c

.-0 .-E

1.0

C

0

la

5 0.5 0

~olitical reunification

market reforms under Doi Moi (renovation)

o

z

1976 1978 1980

1982 1984 Year

1986

1992

Note: Estimate for 1993 based on data from first six months. Figures for

1988-90 are unavailable.

344

Studies in Family Planning

The Ministry of Health only considers the data for 1991 and beyond to be reasonably reliable and complete, but the national figures for 1987and before are included here for comparative perspective (figures for 1988-90 are unavailable).The figures depict all pregnancy terminations, which consist of two procedures for which separate records are normally kept: menstrual regulation (hut thai, literally "to suck out the fetus") is performed by a suction procedure within five weeks of pregnancy, while abortion (nao thai or pha thai, "to destroy the fetus") refers to all other pregnancy-termination procedures performed after five weeks. For 1991 and later, separate provincial figures are available for each of these procedures. In order to ensure comparability with 1992, the 1991 and 1993 totals had to be adjusted in various ways. In 1991, for instance, several provinces did not submit a report on pregnancy terminations, so the corresponding provincial values for 1992 were substituted, adjusted downward in accordance with the growth rate of abortions between 1991 and 1992 for provinces reporting in each year. The provinces for which such substitutions were made accounted for about 5 percent of all terminations in 1992. For 1993, a similar minority of provinces has yet to report any figures, and among those that have, only the first six months' data are currently available. Therefore, to approximate 1993 pregnancy terminations for provinces with missing values, half of the corresponding 1992 figures were substituted, this time adjusted upward in accordance with the growth rate of abortions between 1992 and 1993 for reporting provinces, and then the total results for the six-month interval were multiplied by two. This procedure probably underestimates the actual number of abortions in 1993, because such numbers are likely to have increased further in the second half of the year. Figure 1 indicates incremental increases in reported abortions over the interval. Between 1976 and 1987, pregnancy terminations increased from 70,281 to 811,176, a rise greater than tenfold. This increase occurred at a time when Vietnam's public health-care system was in the midst of a severe crisis, stemming from a variety of internal and external forces3As noted above, abortion data for the years between 1988 and 1990 are unavailable, so the short-term impact of the one- or two-child policy of late 1988 cannot be determined. However, the slope between the figures for 1987 and 1991 actually suggests a slight deceleration of annual abortion increases compared to the situation during the mid-1980s. For the years between 1991 and 1993, abortion totals increased from 1.13, to 1.34, to 1.37 million. The extent to which abortion numbers have increased as a result of more complete reporting at the commune and district levels, as opposed to actual rises in the number of pregnancy

terminations, cannot be determined, but reported rates of pregnancy terminations appear to have yet to reach a peak. The total abortion rate (TAR)for Vietnam in 1992 was 2.5 abortions per woman's reproductive lifetime. This measure is analogous to a total fertility rate (TFR); it indicates the average number of abortions a Vietnamese woman would be expected to have during her reproductive lifetime, given current age-specific abortion rates4 From an international perspective, this rate of abortion is exceptionally high. In Table 1, a variety of abortion and fertility statistics are compared for Vietnam and other countries in Eastern Europe, Asia, and the United States. For some countries a TAR is not available, but they do report a general abortion rate (GAR),the number of abortions per 1,000 women aged 1544. Vietnam's GAR was 100.1 in 1992, lower than that in Romania in 19905and than that in the former Soviet Union. But these countries have the highest rates of abortion in the world (David, 1992):Vietnam's rate is more than two times higher than that of other countries of the former Eastern bloc, such as Czechoslovakia and Hungary. It is also far higher than that of neighboring China, where the GAR never exceeded 62, even at the inception of the one-child policy campaign (United Nations, DESD, 1992). Vietnam is also unusual for having such high abortion rates at levels of fertility that are still high; the TFR was about 3.7 children per woman in 1992 (the TFR was about 4.9 from 1975-79). As family-size preferences decline, the demand for abortion is likely to increase further, unless a concurrent rise in the number of unwant-

Table 1 Indicators of fertility and reported abortion in Vietnam and other countries, drawn from the latest annual data available

count^

Total fertility rate

Abortions per 1,000 women aaed 15-44

Ratio of abortions to oreanancies

Total abortion rate

.76 .55 .41 .43 .42 ,351 .31 .33 .30 .09

na na 2.5b 1.4 1.2 na na .8 .8 .2

~ - -

Romania (1990) USSR (1987) Vietnam (1992) Czechoslovakia (1990) Hungary (1987) Romania (1987) China (1987) Singapore (1987) United States (1987) Netherlands (1987)

1.8 2.5 3.7 1.9 1.8 2.2 2.4 2.2 1.9 1.6

199.3 111.9 100.1 46.4 40.8 39.2 38.8 30.1 28.0 5.3

The number of abortions a woman would expect to have in her lifetime if current age-specific abortion rates remained constant. Calculated by distributing total pregnancy terminations in 1993 (1.33 million) according to age distributionsof abortion in Hanoi and Thai Binh (Do et al., 1993; national figures unavailable). The number of pregnancy terminations were then divided by populationfigures from the 1989 census (aged forward). na = Not available. Sources: Abortion: Romania (Blayo, 1991); USSR, Czechoslovakia (FSO, 1991); Hungary (CSO, 1992); Vietnam (Ministry of Health, unpublished registration); others (Henshaw, 1990). Fertility: Vietnam (estimated on the basis of unpublished data); others (United Nations, 1992). a

ed pregnancies can be avoided through contraceptive means. Vietnamese officials are aware of the high rates of abortion and consider lowering them a priority. Some unpublished provincial population planning models indicate target levels of abortion over the next few years that are 2540 percent below the current levels.

Provincial Figures Of course, the national abortion estimates for Vietnam are only as reliable as the provincial data on which they are based. In this section, provincial data are examined in order to check the general reliability of reporting as well as to look for other clues that may explain regional patterns of pregnancy termination. Table 2 indicates pregnancy terminations by province during 1992, according to the provincial boundaries that existed in 1989. Although some districts have been subdivided since then, and records are currently kept by the new provincial authorities, the older boundaries are used here because the population figures from the census that must be used to calculate abortion rates are available only according to the 1989 boundaries. In order to determine the abortion rate in each province, each figure is divided by the number of women aged 1 1 4 1 in that province according to the 1989 census. The census was taken in April, so the denominator approximates the number of 15-44-yearolds in 1992. The table lists the three major cities first, followed by the provinces in the order listed in census publications, which is basically north to south. The highest rates of abortion, as might be expected, were found in the three major urban provinces, Hanoi, Ho Chi MinhSaigon, and Hai Phong, as well as the densely populated provinces in and around the northern Red River delta and southern Mekong delta (see Figure 2). The lowest rates were in the North central, Central, and South central highlands. Perhaps the best way to assess the reliability of the provincial abortion figures is to compare them with fertility data, which were estimated independently on the basis of age distributions from the 1989 census (Central Census Steering Committee, 1991).An exact correspondence between the two figures should not be expected, because fertility is affected by other factors, such as age at marriage, contraceptive use, and breastfeeding practices, but some correspondence between lower fertility and greater incidence of abortion may be anticipated. Figure 3 suggests such a relationship, albeit not a perfect one.6 Some might conjecture that abortion registration data could be overreported because of pressures to reduce birth rates since 1988; if so, Figure 3 could simply imply that overreporting was most prevalent in areas with

Volume 25 Number 6 November/December 1994 345

Table 2 Preanancv terminations in Vietnam, 1992 Total number of pregnancy terminations

Cities1 arovinces Entire countrv -

Proportion of Regional Ratesof terminations pregnancyterminations that were termination per 1,000 menstrual rates per 1,000 women 15-44 regulationsa women 15-44

1.336.017

100

.45

100

-

Major cities Hanoi Ho Chi Minh-Saigon Hai Phong Provinces Cao Bang Ha Tuyen Lang Son Lai Chau Hoang Lien Son Bac Thai Son La Vinh Phu Ha Bac Quang Ning

140 (Big city provinces)

61 (Northern mountain)

104 (Foothills)

Ha Son Binh Hai Hung Thai Binh Ha Nam Ninh

102

(Red River delta)

Thanh Hoa Nghe Thin Quang Binh

36,083 24,383 6,058

54 31 44

Quang Tri Tranh-Tien Hue QuangNam DaNang

3,872 11,913 20,028

38 58 48

.55

Quang Ngai Binh Dinh Phu Yen Khanh Hoa Thuan Hai Gia Lai Dac Lac

3,690 4,513 3,350 7,087 12,631 4,890 7,955

16 16 23 36 48 26 37

.I9 .05 1 .OO

Lam Dong Song Be Tay Ninh Dong Nai Long An

9,278 14,364 24,985 31,753 27,722

61 64 134 67 103

.95 .84 .48 .OO .00

Dong Thai An Giang Tien Giang Ben Tre Cuu Long Hau Giang Kien Giang Minh Hai Vung Tau

29,527 63,780 40,749 16,307 38,950 95,719 26,365 51,194 11,498

93 155 113 55 90 149 96 145 329

.03 .OO .83 .74 .37 .96 .00 .OO .99

.46

,

A fL

'0° (North central) .68 .86 .67

49

(Central)

Qn JL

.OO (South central) .78

,751

.41

(86 Upper south)

120 (Mekong delta)

Source: Ministry of Health, Vietnam.

"Menstrual regulations are performed for pregnancies of less than five weeks'

duration.

lower fertility. However, other survey data corroborate the patterns revealed in the registration statistics. In the summer of 1993, the United Nations Population Fund (UNFPA) undertook a survey of knowledge, attitudes, and practice and information, education, and communication in the seven provinces in which it currently administers special programs. Although yearly abortion figures cannot be estimated from this survey, ever-use figures for pregnancy termination can be determined. In the top section of Table 3, the numbers of those saying

346 Studies in Family Planning

that they had ever experienced an abortion or menstrual regulation procedure are added together, then divided by the total number of married women surveyed in each province. Although the survey-based and registration figures have been calculated on a different basis: the results suggest some general provincial correspondence between the figures. The registration and survey data exhibit an even stronger correspondence regarding the proportion of abortions (as opposed to menstrual regulations) among all pregnancy terminations. In fact, the bottom of Table 3 indicates a startling correspondence between the two figures in six of the seven provinces. In the seventh province, Khanh Hoa, the registration-based proportion of menstrual regulations among all pregnancy terminations was 1.00in 1992 and 0.85 for the first six months of 1993; the latter figure was also similar to the 1993survey-based figure of 0.78. Evidently, in Khanh Hoa, either nonmenstrual-regulation procedures have just recently been introduced or pregnancy-termination registration was simply not disaggregated by medical procedure prior to 1993. Thus, clear evidence exists that the provincial variation in the proportion of menstrual regulations performed among all pregnancy terminations (Table 2) was real. Nationwide, this proportion was .45, but among the three major cities, for instance, it varied from Hanoi (.29), to Ho Chi Minh-Saigon (.42), to Hai Phong (0.88).Such wide discrepancies, in fact, occurred in all regions of Vietnam. Table 2 had shown that in eleven provinces (four in the North and seven in the South), 95 percent or more of all pregnancy terminations were abortions (as opposed to menstrual regulations). Conversely, five provinces (all in the South) reported that 95 percent or more of pregnancy terminations consisted of menstrual regulations. This finding does not surprise Vietnamese population specialists. Medical health-delivery systems are often organized at the provincial level, where the professional experiences of a small core of doctors and practitioners may influence the kinds of services and training available. If they have been trained to perform just one kind of procedure, they may be unlikely to attempt anything else. In addition, province-specific shortages of special supplies, such as the suction equipment needed for menstrual regulation, may also dictate which kinds of pregnancy-termination services will be available. The increased responsibilities of provincial and local governments in the era of decollectivization since the mid1980s (see, for example, Werner, 1988; Allen, 1993) have led to further variation in the timing of service availability. A wall chart at an office of population and family planning in one Red River delta province shows a sharp leap upward in the number of menstrual regulations per-

Figure 2 Vietnam, provincial boundaries, 1989

36. Cu'u Long 37. Hau Giang 38. Kien Giang 39. Minh Hai Note: Borders are shown as of late 1989 (see text).

Volume 25 Number 6 November/December 1994 347

Figure 3 Total fertility rate (1989) vs.abortion rate per 1,000 women aged 1 5 4 4 (1992) for Vietnamese provinces (excludes Vung Tau)

cal autonomy. However, distinctive regional patterns of contraceptive use may persist in the wake of free-market reforms and decentralization (Allman et al., 19911, and these may influence abortion patterns in turn.

Explanations

200

Despite some uncertainties noted above regarding the reporting of pregnancy terminations, enough evidence exists from independent sources to suggest that reported pregnancy-termination figures in Vietnam are reasonably close to those actually occurring. But what can explain the high prevalenceof reported pregnancy terminations? There are several overlapping answers to this question. The first concerns the general institutional setting. Since the 1950s,Vietnam has relied heavily on the aid, advice, and developmental strategiesof other centralized, socialist societies including China, the former Soviet Union,

formed in one particular year during the late 1980s,suggesting that such procedures suddenly became available. The national proportion of menstrual regulations among all pregnancy terminations has increased over time, from about 30 percent in 1991, to 45 percent in 1992, to just over 50 percent in the first six months of 1993 (not shown). Although no relationship can be seen between current provincial abortion levels and the proportion of menstrual regulations among all procedures (R""2 = .00, not shown), the strong variation in each of these measures among contiguous provinces suggests relative lo-

and the countries of Eastern Europe. Vietnam's family planning strategy has been congruent with those of many of these countries, which have been characterizedby relative poverty and a lack of resources to make available inexpensive, supply-based alternative contraceptive methods at a time when the desire for large families is declining (McIntyre, 1972; David, 1992).Their governments have also been characterized by a desire to maintain centralized control over the medical establishments that provide such services. In such contexts, governments tend to favor the promotion of the IUD and abortion. In Vietnam, the IUD, which has been imported from Czechoslovakia and China since the late 1960s (Vu, 1992),remains the most widely used contraceptive method in the North (Allman

0)

.L

C 5 .-&

2 -

a

.+ -

m +

4

-

3

-

a

I

2

o

50

a

I

I

100

150

Abortion rate

Table 3 Comparison of registration(1992) and survey (1993) data for abortion and menstrual regulationin seven Vietnamese provinces Northern provinces Rates/proportion of menstrual regulation Pregnancytermination rates Registration-based(per 1,000 women 15-44)0 Survey-based (married women < 45)C (N) Proportionof menstrual regulation procedures among all pregnancy terminations Registration-baseddata (N) Survey-based data (N)

Southern provinces

Yen Bai*

Ha Bac

Thai Binh

Ouangnam DaNang

51

121

126

48

Phu Yen

Khanh Hoa

Song Be

23

36

64

42 (1,013)

86 (1,097)

329 (1,100)

14 (1,000)

20 (998)

.I7 (11,381) .07 (28)

.70 (57,438) .63 (57)

.70 (49,395) .70 (219)

.67 (20,028) .67 (9)

1.OO (3,350) 1.OO (13)

27 (1,007)

102 (1,051)

1.OOd .84 (7,087) (14,364) .78 .82 (18) (67)

aFormerlypart of Hoang Lien Son province. b1993 registrationfigures from Table 2. cSurveyfigures from UNFPA KAP-IEC survey, June 1993. Survey-based estimates of pregnancy termination rates are calculated by adding totals of ever-users of abortion and menstrual regulation procedures and dividing them by the number of women surveyed. The resulting estimate is thus qualitatively different from that based on registration. dTheproportionfor the first six months of 1993 was .85.

348

Studies in F a m i l y Planning

et al., 1991). Occasional reports have been made, however, of insufficient IUD supplies, inappropriate fit of those devices that have been available, and the use of "homemade" alternatives that are ineffective and sometimes dangerous (Nguyen, 1992). Similarly, abortion seems to have become an accepted method of fertility control among some women belonging to ethnic minorities (Do et al., 1993). Tables 4 and 5 show the characteristics of women having abortions in Vietnam, the data for which are derived from a survey of 2,088 women having abortions in Hanoi (five selected hospitals) and Thai Binh (all nine district hospitals), a Red River delta province just south of Hanoi. Table 4 indicates that in Vietnam, fewer than 13 percent of all pregnancy terminations were performed among women under the age of 25, a far smaller proportion than that for any other country listed. Table 5 indicates that fewer than 10 percent of such women were currently unmarried and furthermore, that those with no children constituted only 13.6 percent of the total in Hanoi and less than one percent of the total in Thai Binh. Some abortion clinic directors believe that young women, when registering to have an abortion, may misstate their age and marital status (not to mention their names and addresses) because of a lingering atmosphere of shame and embarrassment. The very low proportion of such women in Tables 4 and 5 should, therefore, be interpreted with some skepticism. Many young women may obtain abortions outside the public system, and, thus, are never counted. The reported data suggest that the great majority of women undergoing abortion were married, were between the ages of 25 and 34, and had one or more children (for ethnographic and health perspectives on these issues, see Le et al., 1994). About 50 percent of the women surveyed had already had at least one pregnancy termination (see Table 5). Between 26 and 50 percent of women in Hanoi and Thai Binh, respectively, were not using any contracepTable 4 Percentage distribution of pregnancy terminations, by age group, according to country, for various years Aae arouD Country Vietnam (1991)8 Czechoslovakia (1987) Hungary (1987) Singapore (1987) United States (1987) Netherlands (1987)

Total

100.0 100.0 100.0 100.0 100.0

Table 5 Percentage distribution of Vietnamese women having pregnancy terminations at large hospitals in Hanoi and Thai Binh, by background characteristicsand type of procedure. 1992 Characteristics Marital status Married Not married Age c20 20-24 25-29 30-34 35-39 40-45 >45 Education None Elementary Secondary High school Tertiarya Number of live children 0 1 2 3 4+ Type of procedure Abortion Menstrual regulation Prior pregnancy terminations 0 1 2 3+

Hanoi

Thai Binh

91.9 8.1

99.5 .5

1.7 15.8 22.7 32.0 17.5 7.9 2.4

0.1 7.1 22.2 29.6 24.9 14.0 2.1

0.4 2.5 26.1 55.4 15.6

0.3 4.0 58.2 34.2 3.3

13.6 39.5 34.2 9.3 3.3

0.9 22.0 44.9 22.8

9.4

46.3 53.7

35.1 64.9

47.2 32.4 12.8 7.6

52.6 29.3 12.1

. 6.0

100.0

" Based on age distributions for Hanoi and rural Thai Binh only (Do et al., 1993; see also Table 5). Other sources from Henshaw (1990).

tive method just before becoming pregnant, as shown in Table 6, which suggests a latent need for contra~eption.~ A fairly high prevalence of users of less effective methods may also be contributing to high rates of unintended pregnancy. Although the number of methods available is improving slowly, those couples relying on rhythm or withdrawal constituted almost 30 percent of current contraceptive users in 1988 (NCPFP, 1990), and the prevalence of such use was still above 20 percent in late 1993, even in areas where alternative contraceptive methods were most accessible (Goodkind, 1994~). Second, there is the issue of population policy and how it interacts with this general context of inadequate availability and use of contraceptives. Vietnam's official one- or two-child policy is promoted through posters, ubiquitous in urban and rural areas, and prime-time television coverage several times a week on one or both of the two channels available. Article 6 of this policy stipulates: "The state will supply, free of charge, birth control devices . . . to eligible persons who are cadres, manual workers, civil servants or members of the armed forces . . . and poor persons who register to practice family

Source: Do et a1.(1993); see also Khong (1992). "Category may include trade school as well as university.

Volume 25 Number 6 November/December 1994 349

Table 6 Percentage distribution of Vietnamese women having pregnancy terminations at large hospitals in Hanoi and , contrace~tivemethod used beforehand. 1992 Thai ~ i n hbv Method IUD Pill Condom Withdrawal Rhythm Withdrawal and rhythm Breastfeeding None Total

(N)

Hanoi

Thai Binh

9.2 0.6 8.4 21.8 13.8 19.1 1.4 25.7

20.4 2.0 5.8 13.8 2.5 3.6 2.0 49.9

100.0 (1,012)

100.0 (1,076)

Source: Khong (1992); Do et al. (1993).

planning. . . . The widespread sale of birth control devices will be permitted to facilitate their use by everybody that needs them." (JPRS, 1989)It also calls for penalties for third and higher parity births or for violations of a three-to-five-year spacing rule. Although penalties are not universally enforced, fines for noncompliance are currently being levied in certain areas. A survey conducted by the author in two villages in the rural North indicated that such fines were commonplace.The size of the fines varied between the two villages and depended on the order of the birth and other extenuating circumstances. The minimum fine was usually the equivalent of about one month's salary (Goodkind, 1994~). A strong association was revealed between the number of respondents who knew someone who had received a fine and the number of people they knew who had undergone an abortion. A comparison with studies of other Red River delta areas also suggests a possible relationship between the incidence of abortion and the implementation of the policies (Le et al., 1994).A strong possibility exists that fear of penalties and of more complicated abortion procedures have encouraged women to undergo menstrual regulation as a pre-emptive measure, even when they are unsure if they are pregnant. As mentioned earlier, the proportion of menstrual regulations among all pregnancy terminations has increased in each of the last three years. However, although those areas implementing the fines have raised the marginal cost of childbearing somewhat, most rural women undergoing abortions evidently have not otherwise been coerced to do so. Only five percent of women in two Thai Binh communes reported feeling "pushed" to have abortions, with most of them reporting that they had been pressured by mass organizations (such as the Women's Union) or family members (Le et al., 1994). These reported levels are not much higher than what might be expected in more developed countries with less stringent population policies, although some underreporting of such pressure might be expected in the Vietnamese political context. Further350 Studies in Family Planning

more, at the provincial level, no definite relationship between the fines and aggregate abortion rates appears to exist. Abortion rates are about as high, if not higher, in most Mekong delta provinces as those in the Red River delta provinces, even though the former show less evidence of implementing fines, let alone having the political authority to enforce them. A rigorous study of Vietnam's regional fertility decline and the relative causal roles of economic incentives, disincentives, general development, and other factors has yet to be undertaken and faces many methodological challenges (Goodkind, 1994~). Finally, the issues of modernization and development must be considered. Recently, Vietnam has attempted to shift toward a free-market system with limited glasnost, but the opening of the mass media to Western images has fomented rapid social change. An explosion of Western videos and entertainment, as well as a boom in learning English among teenagers, particularly in the past three years, has contributed to a growing familiarity with Western cultural and dating patterns and is probably contributing to a rise in sexual contacts, both premarital and otherwise. Unfamiliarity with contraception, and, in fact, certain local groups' resistance to acquainting young, unmarried couples with such methods are undoubtedly contributing to increased premarital pregnancy. Continuing social disapproval pressures teenagers to hide their age or marital status from abortion providers, making gauging the extent to which young people are contributing to the rise in pregnancy terminations difficult. However, the consensus among doctors and medical personnel is that demand for such services is growing faster for this age group than for any other. In Ho Chi Minh-Saigon, an area with better-quality and more detailed data collection, the proportion of people seeking abortion services who report that they are younger than 20 has reportedly increased markedly in the past year. A variety of research projects attempting to shed further light on the causes and consequences of adolescent and premarital pregnancy are currently under way.

Conclusion Although abortion registration data may contain some flaws, enough independent sources of corroboration exist to suggest that they are reasonably accurate. The 1992 total abortion rate in Vietnam was at least 2.5 per woman, unusually high by international standards, especially for a country with a total fertility rate of 3.7. Regional patterns of abortion are related to levels of fertility and urbanization, although provincial limitations of resources and medical expertise result in the provision of abortion

services that vary widely, even between contiguous provinces. Unlike the situation in some other countries, abortion appears to be concentrated among married women in their peak childbearing years. The prevalence of repeat abortions reported in some provinces, and survey responses of those reporting having had abortions, suggest that some women are using abortion as a means of fertility control. At least three reasons can be advanced to account for Vietnam's high rate of abortion. First, like other poor socialist nations, Vietnam has heretofore viewed the IUD and abortion as methods allowing maximum state control and monitoring of family planning. More recently, although the number of contraceptive methods available has improved marginally, Vietman still lacks the resources to provide alternative choices. Second, the country's population policies, although not directly compelling abortion, raise the marginal costs of childbearing in some areas to such an extent that pregnancy termination becomes more acceptable. Third, the current trend toward modernization and development in the wake of the free-market reforms of the mid-1980s contributes to premarital and unwanted pregnancy (and hence the demand for abortion services), as it has elsewhere in the world. The appropriate policy prescription would provide effective contraceptive alternatives that are acceptable, safe, and affordable, a goal that Vietnamese specialists hope to achieve. Unless these efforts continue, future declines in Vietnamese family-size preferences may result in even higher rates of abortion.

population figures from Vietnam's 1989 census, aged forward by three years.

5 Note that Table 1 shows two years for Romania, which experienced a huge increase in reported abortions after the coup in 1989,when abortion services became easily obtainable. 6 The R-squared value for the linear relationship was .20. 7 These measures are noncongruent because the survey data depict those ever experiencing abortion, while the registration data depict annual abortion rates. Although survey data often suffer from abortion underreporting, the question of ever-use of abortion covers each woman's entire reproductive history rather than the single year for the registration figures. In addition, reliable comparisons between the two measures are potentially complicated by random survey errors, as well as distinct provincial patterns of repeat abortions, underreporting, and shifting abortion patterns over time. 8 Although the figure for Hanoi indicates that only 26 percent of women in Hanoi had not used any contraceptive method prior to abortion, the table indicates that 55 percent had reported using rhythm, withdrawal, or a combination of the two. Women in Hanoi may have overstated their use of these traditional methods prior to pregnancy termination because of region-specific courtesy biases resulting from the antinatal policies.

-

Acknowledgments

Notes

This report has benefited from conversations with and suggestions from James Allman, Vu Quy Nhan, Linda Demers, Do Tron Hieu, and others over the course of 1993. Do Tron Hieu kindly provided access to the Vietnamese provincial data used here, as did Libor Stloukal for the latest data from Eastern Europe. Additional suggestions were made during a presentation at The Population Council. Thanks are due to Tuong Lai and Pham Bich San of the Institute of Sociology (National Academy of Social Sciences and Humanities, Hanoi) for facilitating this research, and the Andrew W. Mellon Foundation for its financial support. The viewpoints expressed herein are solely those of the author.

1 These figures come from Do et al. (1993), Le et al. (1994), and Nguyen (1992),respectively.

References

2 Area studies specialists emphasize that free-market exchanges had been generated spontaneously in Vietnam during the late 1970s and early 1980s, in spite of considerable state efforts at collectivization. To some extent, the Doi Moi reforms of the late 1980s simply gave official legitimacy to these pre-existing activities, even if they further contributed to marketization (de Vylder and Fforde, 1988). 3 Vietnam's military involvement in Kampuchea had contributed to the United States-led economic embargo and an invasion by China in 1979, which resulted in severe shortages of civilian medical supplies and services. Subsequent emigration of a beleaguered Chinese business community disrupted the flow of commerce. Similarly, the postwar attempt to collectivize agricultural production in the South resulted in a severe food crisis (de Vylder and Fforde, 1988), which, in turn, resulted in rates of childhood malnutrition of 50 percent or higher throughout the 1980s (Allen, 1993). 4 The figure was calculated by apportioning the national number of reported abortions for Hanoi and Thai Binh (Do et al., 1993) and then by dividing the resulting age-specific abortion totals by the

Allen, Suki. 1993. "Health and the current situation and recent changes." In Vietnam and the Rule of Law. Eds. Carlyle A. Thayer and David G. Marr. Canberra: The Australian National University Department of Political and Social Change Monograph No. 19.Pp. 38-49. Allman, James, Vu Qui Nhan, Nguyen Minh Thang, Pham Bich San, and Vu Duy Man. 1991. "Fertility and family planning in Vietnam." Studies in Family Planning 22,5: 308-317. Blayo, C. 1991. "Les modes de prevention des naissances en Europe de l'est." Population 46,3: 527-546. Central Census Steering Committee. 1991. Vietnam Population Census, 1989, Completed Census Results, Volume I . Hanoi: Central Census Steering Committee. Central Statistical Office (CSO). 1992. Demographic Yearbook. (In Hungarian.) Budapest: CSO. Center for Women's Studies (National Centre for Social Sciences in Vietnam). 1990. Selected Indicators of Women's Status in Vietnam, 1975-

Volume 25 Number 6 November/December 1994 351

1989.Hanoi: Statistical Publishing House. Cimit, Ronald J. (ed.). 1989. Vietnam: A Country Study. Washington, DC: Library of Congress, Federal Research Division. David, Henry. 1992. "Abortion in Europe, 1920-91: A public health perspective." Studies in Family Plannitrg 23,l: 1-22. de Vylder, Stefan and Adam Fforde. 1988. Vietnam:An Economy in Transitiotr. Stockholm: Swedish International Development Authority. Do, Tron Hieu, John Stoeckel, and Nguyen van Tien. 1993. "Pregnancy termination and contraceptive failure in Vietnam." Asia-Pacific Population Journal 8 (December):3-18.

Red River Delta of Vietman." Occasional Papers of the Program on Environment. No. 15.Honolulu: East-West Center. McCarty, Adam, Merlyne Paunlagui, and Vu Quoc-Huy. 1992. Vietnam Data Bank 1976-1991. Canberra: The Australian National University National Center for Development Studies. McIntyre, Robert J. 1972. "The effects of liberalized abortion laws in Eastern Europe." In Research in the Politics of Population. Eds. Richard L. Clinton and, R. Kenneth Godwin. Lexington, MA: Lexington Books. Pp. 183-216. Ministry of Health, Vietnam. 1992.Unpublished data.

Federal Statistical Office (FSO). 1991. Population Change in the Czech and Slovak Federal Republics in 1990. (In Czech.) Prague: FSO.

National Committee for Population and Family Planning (NCPFP). 1990. Vietnam Demographic and Health Survey, 1988.Hanoi: NCPFP.

Forbes, Dean K., Terence H. Hull, David G. Marr, and Bryan Brogan (eds.). 1991. Doi Moi: Vietnam's Economic Renovation, Policy and Performance. Canberra: The Australian National University Department of Political and Social Change.

Nguyen, The-Lap. 1992. "Abortion in Vietnam-Perspectives. Committee for Population and Family Planning of Thai Binh Province." Unpublished.

Goodkind, Daniel M. 1994a. "Vietnam's north-south fertility trends amidst free-market renovation." Unpublished.

Turley, William S. and Mark Seldon (eds.). 1993. Reinventing Viettmmese Socialism: Doi Moi in Comparative Perspective. Boulder, CO: Westview Press.

-.

1994b. "Rising gender inequality amidst Vietnam's postwar transition from socialism." Unpublished.

United Nations. 1992.Patterns ofFertility in Low-fertility Settings. Publication number ST/ESA/SER a131. New York: United Nations.

-.

1994~."Vietnam's one- or two-child policy in action." Unpublished.

United Nations, Department of Economic and Social Development (UNDESD). 1992. Abortion Policies: A Global Review, Volume I , Afghanistan to France. New York: UNDESD.

Henshaw, Stanley K. 1990. "Induced abortion: A world review, 1990." Family Planning Perspectives 22: 76-89. Joint Publications Research Service UPRS), East and Southeast Asia. 1989. Publication number SEA-89-007 (February 8). Reprinted in "Vietnam's new fertility policy." Population and Development Review 15,l: 169-172. Jones, Gavin W. 1982. "Population trends and policies in Vietnam." Population and Development Review 8,4: 783-810. Khong, Ngoc-Am. 1992. "Tinh hinh nao hut thai va viec su dung cac bien phap tranh thai que khao sat thuc te" (The situation of abortion and contraceptive methods observed) Thong Tin Dan So (Population Information) 4: 16-20. Le, Thi Nham Tuyet, Annika Johannson, and Nguyen The-Lap. 1994. "Abortions in two rural communes in Thai Binh province, Vietnam." Vietnam Social Sciences 39,l: 73-90. Le, Trong Cuc and A. Terry Rambo. 1993. "Too Many People, Too Little Land: The Human Ecology of a Wet Rice-growing Village in the

352 Studies in Family Planning

United States Agency for International Development (USAID), Asia Bureau. 31 December 1975. United States Economic Assistance to Vietnam, 1954-1975: Public Health-Vietnam Terminal Report. Washington, DC: USAID. Vu, Quy Nhan. 1992. "Population policies and development in Vietnam." In The Challenges of Vietnam's Recotrstruction. Eds. Neil L. Jamieson, Nguyen Manh Hung, and A. Terry Rambo. Fairfax, VA: The Indochina Institute of George Mason University. Pp. 40-54. Vu, Quy Nhan and Robert Hanenberg. 1989. "The 1988 Demographic

Survey of Vietnam." Asia and Pacific Population Journal 4,4: 3-14.

Werner, Jayne. 1988. "The problem of the district in Vietnam's devel-

opment policy." In Postwar Vietnam: Dilemmas in Socialist Development. Eds. David G. Marr and Christine P. White. Ithaca, NY: Cornell University, Southeast Asia Program. Pp. 147-162.