- Split View
-
Views
-
Cite
Cite
P H David, L Reichenbach, I Savelieva, N Vartapetova, R Potemkina, Women's reproductive health needs in Russia: what can we learn from an intervention to improve post-abortion care?, Health Policy and Planning, Volume 22, Issue 2, March 2007, Pages 83–94, https://doi.org/10.1093/heapol/czm003
- Share Icon Share
Abstract
It has been well documented that abortion is a common means of controlling fertility in Russia. Women undergo repeat abortions throughout their reproductive lives, but recent studies of abortion trends in the Russian Federation suggest that abortion rates are on the decline, use of modern contraceptives is increasing, and women dislike abortion as a method of fertility control. Using data collected during 1999–2003 in women's health facilities in three Russian cities, this paper reports the results of an evaluation of interventions to improve post-abortion care, which show an impressive increase in post-abortion contraceptive counselling but no reduction in the rate at which women present at clinics for repeat abortions. The findings indicate a discrepancy between women's stated preferences for modern medical contraceptive methods and their abortion-seeking behaviour. Further exploration of these data suggests that certain women resort to abortion with greater frequency than others, and points to the need for a more focused investigation of these women. These results indicate the complexities associated with changing what has been a relatively common and long-standing practice, and have implications for improving reproductive health services. Meeting the reproductive health needs of Russian women requires not only improved provider and client knowledge but may also demand a more focused delivery of client-centred care than may be the case in other settings.
An identifiable group of women appears to resort to frequent abortion procedures and can be characterized as disadvantaged. Compared with other repeat abortion clients, they are likely to be younger, living in informal unions, report socio-economic reasons for the abortion, and have given little thought to future contraceptive needs. These women who repeat the procedure rapidly may need more support than others to safely achieve their reproductive intentions.
Expanding access to a wider range of long-term and permanent contraceptive methods needs consideration by the health system. Providing the option of IUD insertion at time of abortion may be a safer alternative than repeated abortion for many women, especially those at risk of rapid repetition of the procedure. Causes of apparent IUD failures need to be investigated and addressed.
In light of the apparent disparity between Russian women's dislike of abortion as a method of fertility control and their actual practice, an in-depth investigation of the motivations of women who rely on frequent abortion could be useful for designing more focused and more effective reproductive health programmes.
Introduction
Since the International Conference on Population and Development (ICPD) held in Cairo in 1994, reproductive health programmes have aimed to ensure that health policies and programmes consider women's preferences with respect to fertility control and ‘client-centred approaches’ to elicit and respond to their needs. For the Russian Federation and other former states of the Soviet Union, where abortion is legal and provided on demand, implementing the ICPD approach poses a unique challenge. The ICPD Programme of Action issued a specific call for international support to meet the needs of women in such countries:
‘Those countries … must themselves give a higher priority to reproductive health services, including a comprehensive range of contraceptive means, and must address their current reliance on abortion for fertility regulation by meeting the needs of women in those countries for better information and more choices on an urgent basis.’ (UN 1995: 32: 7.10)
Throughout the former Soviet Union induced abortion has been a common means of preventing births for decades (Popov 1991; Entwisle and Kozyreva 1997; Hutter 2003). More recently, declining fertility and negative population growth have prompted nationalist, pro-natalist forces within Russian society to place increased pressure on health officials to reduce the number of abortions being performed, as well as to limit nascent family planning programmes (Deschner and Cohen 2003). Nevertheless, abortion continues to be a widely used method of birth control, even though use of modern contraceptives is believed to be rising and data from the few population-based surveys suggest that modern methods now account for more than half of all contraceptive use (CDC and ORC Macro 2003). A large proportion of Russian women still rely on traditional (and much less effective) methods of contraception (Entwistle and Kozyreva 1997; CDC/VCIOM 1998, 2000; David et al. 2000, 2003).
Abortion and unintended pregnancy
Absolute numbers from different data sources vary, but there is evidence that the abortion rate declined dramatically during the last decade, from 225 per 1000 women in 1992 to 128 per 1000 in 2002 (Ministry of Health, Russian Federation, 2002 statistical report).1 Nevertheless, the abortion rate and the prevalence of unintended pregnancies in Russia remain high relative to that of other countries in Eastern Europe (CDC and ORC Macro 2003). A combination of early marriage, desire for only one or two children, and rapid fulfillment of childbearing intentions soon after marriage means that most Russian women are exposed to the risk of an unwanted pregnancy for a large part of their reproductive years.
Approximately 75% of women aged 40–44 report at least one abortion (CDC/VCIOM 2000). Data from recent surveys in Russia show that as many as two of every three pregnancies are either unwanted or mistimed (Goldberg and Serbanescu 2001; CDC and ORC Macro 2003) and between 80 and 90% of these unintended pregnancies are aborted (CDC/VCIOM 1998, 2000; David et al. 2000, 2003).
Despite the frequency with which women resort to abortion, surveys also reveal a strong dislike for the procedure (CDC/VCIOM 1998, 2000; David et al. 2000, 2003). Women of reproductive age (15–44) were asked to rate methods of birth control on a scale from 1 to 10, with respect to safety, cost and overall like/dislike of the methods. In four surveys conducted in selected Russian cities between 1996 and 2003, women rate induced abortion and mini-abortion (vacuum aspiration up to 6 weeks gestation) most negatively of all methods of fertility control, 95% rating both methods 3 or lower overall. These surveys also show relatively high levels of dislike of hormonal contraceptives, 40% giving a rating of 3 or below to oral contraceptives and 57% to injectable methods.
Contraceptive choices
The divergence between women's attitudes toward abortion and their behaviour may be due at least in part to the limited range of contraceptive choices available to them and prevailing policies regarding provision of long-term methods. During Soviet times, only limited supplies of modern contraceptives (condoms, IUDs and imported high-estrogen pills) were available (Sherwood-Fabre et al. 2002). Today, choices are wider and foreign pharmaceutical companies advertise their products, including a range of expensive hormonal methods, but supplies of specific brands in pharmacies are often inconsistent (Chalmers et al. 1998). Few medical practitioners receive training in contraceptive technologies (other than pharmaceutical company information) or family planning counselling. Consequently, women are not well informed about side effects or the differences between formulations of oral contraceptives, which may be one reason for the high rates of method switching and discontinuation of oral contraceptives found among women during the first year of use (CDC/VCIOM 1998).
The IUD is popular and Russian IUDs are inexpensive, but providers prefer the higher quality IUDs imported from Europe. They complain that the cheaper, Russian-made devices are expelled more frequently, lead to excessive bleeding more often than imported IUDs, and pregnancies occur even when they are in place, problems that could be due to improper insertion as well as the device itself. Due to concern about infection and prior experience of complications following immediate IUD insertion, Ministry of Health guidelines governing provision of IUDs do not permit insertion at the time of abortion.
Other long-term methods, such as injectables and implants, are not widely available and providers have little training or experience carrying out female sterilization procedures, which were prohibited from the 1930s until the early 1990s; the current Ministry of Health policy directive permits sterilization only for women 35 years or older and those who have two or more children. In most regions, clients must pay for the procedure. There is almost no demand for male sterilization, and virtually no survey respondents reporting that their partner had a vasectomy (CDC/VCIOM 1998, 2000; David et al. 2000, 2003).
Modern contraceptive methods are also generally more costly relative to abortion. An operations research study in the city of Perm estimated the average amount paid for an abortion at US$16 (Savelieva et al. 2003), while the average cost of 1 year of contraceptive use was US$25 for contraceptive pills and US$33 for French condoms, preferred to the cheaper locally produced condoms, which are perceived to be of lower quality (Goldberg et al. 2001). The Russian IUD cost US$4 and an imported Copper T IUD cost US$11. (At the time of the study, one US dollar was the equivalent of about 30 rubles.)
Health risks associated with abortion
Outdated techniques and a high prevalence of complications make even legal abortion a procedure that unnecessarily threatens women's health. For example, dilation and curettage (D&C) procedures are still the norm in Russia, even for first-trimester abortions, although vacuum aspiration is a much safer procedure (Ipas 2003). High rates of complications following abortion also raise concerns about the health risks of repeated resort to abortion (CDC and ORC Macro 2003).
According to women's survey reports, post-abortion complications occur in 20% of all abortion procedures and around 14% occur within 6 months following the abortion (CDC/VCIOM 1998, 2000; David et al. 2000, 2003). Infection or severe bleeding account for between 30 and 40% of these early abortion complications (David et al. 2000; CDC and ORC Macro 2003). Although it is not possible to gauge the severity of self-reported complications, these rates are much higher than in the US and western Europe, and raise concern about the quality of abortion services (CDC and ORC Macro 2003), which may be affected by diminishing resources available in the health system (Remennick 1991).
Despite abortion being legal, some women obtain abortions outside of the health system, increasing the risk of medical complications. One study of abortion-related deaths indicates that most are due to complications of illegal abortion (performed outside the medical system), but at least 15% of the abortion-related deaths identified in that study occurred following a procedure performed in a medical institution (Volgina and Frolova 1991, cited in Savelyeva et al. 1997). Nearly half of these deaths were caused by peritonitis and sepsis. A more recent study found that 24% of the 113 cases of abortion-related deaths in 1999 for which records could be located were performed in a medical institution, of which 70% were second trimester abortions (Zhirova et al. 2004). In Western Europe and the USA, deaths associated with abortions performed in a medical facility are very rare (CDC and ORC Macro 2003).
The high rates of secondary sterility in Russia are thought to be another consequence of the heavy reliance on abortion and high rates of complications. While implications of a causal relationship need closer investigation, some authors suggest that in countries with consistently high levels of abortion, permanent fertility impairment is far greater than in Western Europe. One study in Western Siberia estimated that the prevalence of infertility exceeded the 15% level considered critical by the World Health Organization, and nearly a quarter of women with secondary infertility experienced complications (12% due to infection) either during birth or abortion (Philippov et al. 1998).
The intervention
In this context, the Women and Infant Health Project (WIN), designed to improve the overall quality and effectiveness of maternal and infant health services, including post-abortion care, began in 1999. In cooperation with the Ministry of Health and the Health Care Departments of Perm and Novgorod Oblasts, this pilot project promoted evidence-based medical practices, quality assurance methods, and client-centred approaches to service delivery among women's health care providers and their clients. In three cities, Perm and Berezniki in Perm Oblast, near the Ural Mountains, and Veliky Novgorod, near St Petersburg in the northwest,2 the project worked in 20 health care sites (five maternity hospitals, six women's consultation centres linked to those hospitals, three family planning clinics and six children's polyclinics) with catchment area populations totalling about 1.1 million residents.
Project interventions included training courses in family-centred maternity care, essential care of the newborn, exclusive breast-feeding, contraceptive technologies and client-centred family planning counselling. Physicians, midwives and paediatric nurses participated in these courses. One of the project's specific objectives was to reduce unwanted pregnancies and repeat abortions at participating sites. Here, we focus on only the family planning interventions.
Provider training began in September 2000 with Contraceptive Technology Update (CTU) seminars in each city for obstetrician-gynaecologists and midwives who provide postpartum and post-abortion care and family planning services, selected by the Oblast Health Administration on the basis of their ability to influence and guide other staff in their facilities. Curricula were updated, training materials produced and a cadre of local ‘master trainers’ were trained.3
Courses incorporating general principles of family planning counselling, interpersonal communication skills and essential elements of quality post-abortion family planning services, including discussing a range of contraceptive options, management of complications and side effects, began in Perm in November 2000 and continued in the other cities throughout the following year. Three special competency-based courses on IUD insertion and removal were given between November 2001 and January 2002, when the main training activities were completed. Some providers attended more than one type of training course. As part of the final year's work to establish a training resource centre in Perm, additional training of trainers continued through September 2002. Post-training follow-up visits to review progress at facilities implementing new practices continued throughout 2002.
In addition to training health care providers, the project also supplied informational brochures to participating facilities, and a national media campaign took place between December 2001 and February 2002 to inform women about contraceptive choices.
Data and methods
To assess the project's effectiveness and impact, pre- and post-intervention surveys of providers and clients were conducted in the 20 participating facilities. The anonymous surveys were designed to measure changes in key indicators of service quality, as reflected in reported provider practices and client experiences. Pre- and post-intervention household surveys were also conducted (see detailed reports in David et al. 2000, 2003). In this report, we focus on data from abortion clients obtained in the three facility-based surveys conducted in 2000, 2002 and 2003.4 The first survey took place in January and February 2000 before training activities began; the project was initially scheduled to end in mid-2002, and the second survey was to collect end-of-project data. The project was later extended for a fourth year, and a third survey was conducted in March 2003, which provided data for the final report.
Changes in the following key indicators were used to assess the effectiveness of project family planning interventions. The percentage of abortion clients who:
report that they conceived while using a contraceptive method (pregnancy resulted from contraceptive failure);
report receiving contraceptive counselling from their provider on the day of the abortion;
intend to use a modern contraceptive following the abortion;
chose a specific contraceptive method prior to discharge;
report that they discussed their chosen method with a member of medical staff;
report receiving an informational brochure on contraception to take home;
report that the counsellor explained when to make a follow-up visit to a health provider;
retained key information imparted through counselling (timing of return to fertile state);
of those with a previous abortion, the proportion who experience a repeat abortion within the previous calendar year. (By early 2002, when the first follow-up survey was conducted, more than one calendar year would have elapsed from the start of training interventions.)
Questionnaire design
Survey questionnaires were drawn from instruments developed by the Population Council and World Health Organization and adapted to the WIN Project interventions. Abortion clients were asked unprompted questions about their fertility history, previous and desired future methods of contraception, and their experiences during counselling and the abortion procedure. A comprehensive list of available contraceptive methods, including traditional methods, as well as provision for recording ‘other’ open-ended responses, was included in the structured questionnaire. The Russian translation was pre-tested in non-participating facilities in a city near Moscow, and adjustments were made prior to back-translation of the baseline questionnaire.5
Quality of counselling was assessed through a series of questions asking women to recall information received during counselling. While observations of actual provider-client interactions might enrich our data, neither a sufficient number of knowledgeable observers nor the time to conduct such observations was available before the WIN Project training activities began.
Sample
Sample size was calculated based on the minimum expected change in prevalence of key indicators we wanted to detect at the end of the project. We aimed to obtain a sample of at least 450 abortion clients, which we estimated could be achieved during three weeks of field work. All abortion clients coming to each facility during this period were invited to participate in the survey, and all consenting women were interviewed. A total of 489 abortion clients in 2000, 559 in 2002 and 527 in 2003 were successfully interviewed.
During each survey, interviewers were asked to record refusals as well as interview those who consented. Despite these instructions, no information regarding the number of clients who refused to participate was obtained during the baseline survey. To assess the magnitude of the refusal rate, we compared the numbers of abortion and delivery clients interviewed with attendance at the facilities in each city. Average client loads were almost identical to numbers interviewed in Veliky Novgorod and Berezniki, and the numbers interviewed in Perm (where overall client loads are higher) were slightly below the facility reported client averages. This suggests that almost all clients were interviewed, but it is not possible to identify how many women actually refused interview during the first survey. In the 2002 survey, there were five refusals recorded among abortion clients, and seven refusals among abortion clients in 2003, a refusal rate of less than 1%.
Informed consent and confidentiality
Survey implementation proceeded with permission of the project's Technical Advisory Group, including the directors of each participating facility, representatives of the oblast Health Administration and the Federal Ministry of Health, and USAID, and informed verbal consent was obtained from all respondents. Interviewers were assigned specific times to cover client interviews in facilities and instructed to approach each client during her recovery period, just before she was discharged from the unit. A statement explaining the purpose of the survey, ensuring confidentiality and requesting consent to interview was read to each woman. Interviewers were instructed to proceed only if verbal consent was given, and were assigned a private area in which to conduct the interviews. The client's name was not recorded on the questionnaire.
Data collection
Nineteen medical students and interns were recruited to conduct the fieldwork, with three senior medical administrators, one in each city, serving as supervisors. In early February 2000, a 5-day training course for field staff was held in Moscow, which included one day of field practice in facilities in a nearby city. Interviews took place between 14 February and 18 March 2000; 19 December 2001 and 7 February 2002 (holidays interrupted fieldwork for about 2 weeks); and again from mid-January to mid-February 2003.
Completed questionnaires were shipped to project headquarters in Moscow, where office editing was carried out. These data were entered into computer data files at the All-Russian Centre for Public Opinion and Market Research (VCIOM) and data files were produced in an English version ready for analysis with the SPSS statistical analysis package.
Analysis
All results reported here are based on reports from abortion clients about their experiences and satisfaction with services. The analyses are based on aggregated reports of individual respondents and provide estimates reflecting the experience of the average client in the entire network of participating facilities. Except in a few cases, the sample size precludes analysis at city or facility level.
Results
Description of women coming for abortion and changes over time
As the data in Table 1 indicate, abortion clients are in general younger and better educated than the population of women as a whole, as can be seen by the comparison with the 2003 household survey sample (column 4). Women in a current union (either formal or unregistered marriage) are over-represented among abortion clients compared with the general population, probably reflecting their greater exposure to sexual activity and thus risk of pregnancy. Abortion clients are more likely to be living in unregistered (common law) marriages than women in the general population. The hospitals in Perm perform more abortions in the course of a year than those in Veliky Novgorod and Berezniki and, as intended, the distribution of the facility sample reflects the disproportion in these client loads.
. | Percentage of clients . | Percentage of sample . | ||
---|---|---|---|---|
. | 2000 . | 2002 . | 2003 . | 2003 HH survey . |
City | 3 cities combinedc | |||
Veliky Novgorod | 29.9 | 28.3 | 29.4 | |
Perm | 43.8 | 49.6 | 46.1 | |
Berezniki | 26.4 | 22.2 | 24.5 | |
Age distribution | ||||
Mean age (years) | 26.4 | 26.1 | 26.3 | 29.2 |
15–24a | 47.0 | 46.7 | 42.9 | 34.0 |
25–34 | 39.3 | 37.6 | 41.0 | 35.5 |
35–44b | 13.7 | 15.7 | 16.1 | 30.5 |
Education | ||||
Less than complete secondary | 7.2 | 6.1 | 5.7 | 11.7 |
Completed secondary | 35.2 | 41.0 | 34.2 | 39.1 |
Any higher post- secondary | 57.7 | 51.1 | 60.0 | 49.2 |
Missing | 0 | 1.8 | 0.2 | 0 |
Marital status | ||||
Married | 49.1 | 45.6 | 51.2 | 43.4 |
In unregistered marriage | 21.9 | 25.8 | 24.1 | 14.8 |
Single, never married | 22.5 | 23.6 | 19.7 | 26.3 |
Divorced/separated/ widowed | 6.5 | 5.0 | 4.9 | 15.6 |
Number of respondents | 489 | 559 | 527 | 3900 |
. | Percentage of clients . | Percentage of sample . | ||
---|---|---|---|---|
. | 2000 . | 2002 . | 2003 . | 2003 HH survey . |
City | 3 cities combinedc | |||
Veliky Novgorod | 29.9 | 28.3 | 29.4 | |
Perm | 43.8 | 49.6 | 46.1 | |
Berezniki | 26.4 | 22.2 | 24.5 | |
Age distribution | ||||
Mean age (years) | 26.4 | 26.1 | 26.3 | 29.2 |
15–24a | 47.0 | 46.7 | 42.9 | 34.0 |
25–34 | 39.3 | 37.6 | 41.0 | 35.5 |
35–44b | 13.7 | 15.7 | 16.1 | 30.5 |
Education | ||||
Less than complete secondary | 7.2 | 6.1 | 5.7 | 11.7 |
Completed secondary | 35.2 | 41.0 | 34.2 | 39.1 |
Any higher post- secondary | 57.7 | 51.1 | 60.0 | 49.2 |
Missing | 0 | 1.8 | 0.2 | 0 |
Marital status | ||||
Married | 49.1 | 45.6 | 51.2 | 43.4 |
In unregistered marriage | 21.9 | 25.8 | 24.1 | 14.8 |
Single, never married | 22.5 | 23.6 | 19.7 | 26.3 |
Divorced/separated/ widowed | 6.5 | 5.0 | 4.9 | 15.6 |
Number of respondents | 489 | 559 | 527 | 3900 |
aIncludes four 14-year-old abortion clients.
bFor abortion clients, this age group includes seven abortion clients aged 45–49.
cData shown are unweighted proportions.
. | Percentage of clients . | Percentage of sample . | ||
---|---|---|---|---|
. | 2000 . | 2002 . | 2003 . | 2003 HH survey . |
City | 3 cities combinedc | |||
Veliky Novgorod | 29.9 | 28.3 | 29.4 | |
Perm | 43.8 | 49.6 | 46.1 | |
Berezniki | 26.4 | 22.2 | 24.5 | |
Age distribution | ||||
Mean age (years) | 26.4 | 26.1 | 26.3 | 29.2 |
15–24a | 47.0 | 46.7 | 42.9 | 34.0 |
25–34 | 39.3 | 37.6 | 41.0 | 35.5 |
35–44b | 13.7 | 15.7 | 16.1 | 30.5 |
Education | ||||
Less than complete secondary | 7.2 | 6.1 | 5.7 | 11.7 |
Completed secondary | 35.2 | 41.0 | 34.2 | 39.1 |
Any higher post- secondary | 57.7 | 51.1 | 60.0 | 49.2 |
Missing | 0 | 1.8 | 0.2 | 0 |
Marital status | ||||
Married | 49.1 | 45.6 | 51.2 | 43.4 |
In unregistered marriage | 21.9 | 25.8 | 24.1 | 14.8 |
Single, never married | 22.5 | 23.6 | 19.7 | 26.3 |
Divorced/separated/ widowed | 6.5 | 5.0 | 4.9 | 15.6 |
Number of respondents | 489 | 559 | 527 | 3900 |
. | Percentage of clients . | Percentage of sample . | ||
---|---|---|---|---|
. | 2000 . | 2002 . | 2003 . | 2003 HH survey . |
City | 3 cities combinedc | |||
Veliky Novgorod | 29.9 | 28.3 | 29.4 | |
Perm | 43.8 | 49.6 | 46.1 | |
Berezniki | 26.4 | 22.2 | 24.5 | |
Age distribution | ||||
Mean age (years) | 26.4 | 26.1 | 26.3 | 29.2 |
15–24a | 47.0 | 46.7 | 42.9 | 34.0 |
25–34 | 39.3 | 37.6 | 41.0 | 35.5 |
35–44b | 13.7 | 15.7 | 16.1 | 30.5 |
Education | ||||
Less than complete secondary | 7.2 | 6.1 | 5.7 | 11.7 |
Completed secondary | 35.2 | 41.0 | 34.2 | 39.1 |
Any higher post- secondary | 57.7 | 51.1 | 60.0 | 49.2 |
Missing | 0 | 1.8 | 0.2 | 0 |
Marital status | ||||
Married | 49.1 | 45.6 | 51.2 | 43.4 |
In unregistered marriage | 21.9 | 25.8 | 24.1 | 14.8 |
Single, never married | 22.5 | 23.6 | 19.7 | 26.3 |
Divorced/separated/ widowed | 6.5 | 5.0 | 4.9 | 15.6 |
Number of respondents | 489 | 559 | 527 | 3900 |
aIncludes four 14-year-old abortion clients.
bFor abortion clients, this age group includes seven abortion clients aged 45–49.
cData shown are unweighted proportions.
In all three surveys, abortion clients interviewed at the time of the procedure averaged just over two abortions, including the abortion just completed, and one living child. Among clients aged 35 and older, women averaged more than three previous abortion procedures (Table 2). More than 20% of clients were coming for abortion of their first pregnancy.
. | . | Mean of clients . | ||
---|---|---|---|---|
. | . | 2000 . | 2002 . | 2003 . |
Mean number of pregnancies | ||||
Age: | 15–24 | 2.0 | 1.8 | 1.9 |
25–34 | 4.2 | 4.6 | 4.0 | |
35–45 | 5.6 | 5.8 | 5.4 | |
Total | 3.3 | 3.5 | 3.3 | |
% aborting first pregnancy | 21.7 | 25.8 | 20.1 | |
Mean number of abortionsa | ||||
Age: | 15–24 | 1.6 | 1.4 | 1.4 |
25–34 | 2.7 | 3.0 | 2.6 | |
35–45 | 3.6 | 3.9 | 3.3 | |
Total | 2.3 | 2.4 | 2.2 | |
Number of respondents | 489 | 559 | 527 |
. | . | Mean of clients . | ||
---|---|---|---|---|
. | . | 2000 . | 2002 . | 2003 . |
Mean number of pregnancies | ||||
Age: | 15–24 | 2.0 | 1.8 | 1.9 |
25–34 | 4.2 | 4.6 | 4.0 | |
35–45 | 5.6 | 5.8 | 5.4 | |
Total | 3.3 | 3.5 | 3.3 | |
% aborting first pregnancy | 21.7 | 25.8 | 20.1 | |
Mean number of abortionsa | ||||
Age: | 15–24 | 1.6 | 1.4 | 1.4 |
25–34 | 2.7 | 3.0 | 2.6 | |
35–45 | 3.6 | 3.9 | 3.3 | |
Total | 2.3 | 2.4 | 2.2 | |
Number of respondents | 489 | 559 | 527 |
aIncludes abortion procedure on day of interview.
. | . | Mean of clients . | ||
---|---|---|---|---|
. | . | 2000 . | 2002 . | 2003 . |
Mean number of pregnancies | ||||
Age: | 15–24 | 2.0 | 1.8 | 1.9 |
25–34 | 4.2 | 4.6 | 4.0 | |
35–45 | 5.6 | 5.8 | 5.4 | |
Total | 3.3 | 3.5 | 3.3 | |
% aborting first pregnancy | 21.7 | 25.8 | 20.1 | |
Mean number of abortionsa | ||||
Age: | 15–24 | 1.6 | 1.4 | 1.4 |
25–34 | 2.7 | 3.0 | 2.6 | |
35–45 | 3.6 | 3.9 | 3.3 | |
Total | 2.3 | 2.4 | 2.2 | |
Number of respondents | 489 | 559 | 527 |
. | . | Mean of clients . | ||
---|---|---|---|---|
. | . | 2000 . | 2002 . | 2003 . |
Mean number of pregnancies | ||||
Age: | 15–24 | 2.0 | 1.8 | 1.9 |
25–34 | 4.2 | 4.6 | 4.0 | |
35–45 | 5.6 | 5.8 | 5.4 | |
Total | 3.3 | 3.5 | 3.3 | |
% aborting first pregnancy | 21.7 | 25.8 | 20.1 | |
Mean number of abortionsa | ||||
Age: | 15–24 | 1.6 | 1.4 | 1.4 |
25–34 | 2.7 | 3.0 | 2.6 | |
35–45 | 3.6 | 3.9 | 3.3 | |
Total | 2.3 | 2.4 | 2.2 | |
Number of respondents | 489 | 559 | 527 |
aIncludes abortion procedure on day of interview.
Over the 3 years of project implementation, our data show an increase in the percentage of women seeking abortion at participating facilities who report that they want no more children, from around 25% of all clients to nearly 40% (Table 3).
. | Percentage of abortion clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
I. Want no more children: | 26.6 | 31.8* | 38.0** |
Previously used a contraceptive method | 70.6 | 69.1 | 61.3** |
Got pregnant while using a method (of all clients) | 49.1 | 42.4 * | 38.0 ** |
n of respondents | 489 | 559 | 527 |
II. Of ever-users, type of method last used: | |||
Medical reversible | 24.3 | 21.2 | 19.5 |
Barrier methods | 48.7 | 54.4 | 53.9 |
Lactational amenorrhoea method (LAM) | 0.6 | 2.3 | 4.3 |
Traditional methods | 26.1 | 21.8 | 21.1 |
Other | 0.3 | 0.3 | 1.2 |
Of ever-users, got pregnant while using a method | 69.6 | 61.4** | 61.9** |
n of cases (ever-users of contraceptives) | 345 | 386 | 323 |
III. Of those who conceived while using a method, % using: | |||
Medical reversible | 11.7 | 9.7 | 10.0 |
LAM | 0.8 | 2.1 | 7.0 |
Barrier | 51.3 | 53.6 | 47.5 |
Traditional | 35.8 | 34.2 | 33.5 |
Other | 0.04 | 0.04 | 2.0 |
n of cases (conceptions while using contraceptive method) | 240 | 237 | 200 |
IV. Percentage of ever-users of each method who conceived while using: | |||
Medical reversible (n) | 33.3 (84) | 28.0 (82) | 31.7 (63) |
LAM (n) | (2) | (9) | (14) |
Barrier (n) | 73.2 (168) | 60.5** (210) | 54.6** (174) |
Traditional (n) | 95.6 (90) | 96.4 (84) | 98.5 (68) |
. | Percentage of abortion clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
I. Want no more children: | 26.6 | 31.8* | 38.0** |
Previously used a contraceptive method | 70.6 | 69.1 | 61.3** |
Got pregnant while using a method (of all clients) | 49.1 | 42.4 * | 38.0 ** |
n of respondents | 489 | 559 | 527 |
II. Of ever-users, type of method last used: | |||
Medical reversible | 24.3 | 21.2 | 19.5 |
Barrier methods | 48.7 | 54.4 | 53.9 |
Lactational amenorrhoea method (LAM) | 0.6 | 2.3 | 4.3 |
Traditional methods | 26.1 | 21.8 | 21.1 |
Other | 0.3 | 0.3 | 1.2 |
Of ever-users, got pregnant while using a method | 69.6 | 61.4** | 61.9** |
n of cases (ever-users of contraceptives) | 345 | 386 | 323 |
III. Of those who conceived while using a method, % using: | |||
Medical reversible | 11.7 | 9.7 | 10.0 |
LAM | 0.8 | 2.1 | 7.0 |
Barrier | 51.3 | 53.6 | 47.5 |
Traditional | 35.8 | 34.2 | 33.5 |
Other | 0.04 | 0.04 | 2.0 |
n of cases (conceptions while using contraceptive method) | 240 | 237 | 200 |
IV. Percentage of ever-users of each method who conceived while using: | |||
Medical reversible (n) | 33.3 (84) | 28.0 (82) | 31.7 (63) |
LAM (n) | (2) | (9) | (14) |
Barrier (n) | 73.2 (168) | 60.5** (210) | 54.6** (174) |
Traditional (n) | 95.6 (90) | 96.4 (84) | 98.5 (68) |
Differences from baseline proportions significant at: *P < 0.05; **P < 0.01. Estimates based on less than 25 cases omitted.
. | Percentage of abortion clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
I. Want no more children: | 26.6 | 31.8* | 38.0** |
Previously used a contraceptive method | 70.6 | 69.1 | 61.3** |
Got pregnant while using a method (of all clients) | 49.1 | 42.4 * | 38.0 ** |
n of respondents | 489 | 559 | 527 |
II. Of ever-users, type of method last used: | |||
Medical reversible | 24.3 | 21.2 | 19.5 |
Barrier methods | 48.7 | 54.4 | 53.9 |
Lactational amenorrhoea method (LAM) | 0.6 | 2.3 | 4.3 |
Traditional methods | 26.1 | 21.8 | 21.1 |
Other | 0.3 | 0.3 | 1.2 |
Of ever-users, got pregnant while using a method | 69.6 | 61.4** | 61.9** |
n of cases (ever-users of contraceptives) | 345 | 386 | 323 |
III. Of those who conceived while using a method, % using: | |||
Medical reversible | 11.7 | 9.7 | 10.0 |
LAM | 0.8 | 2.1 | 7.0 |
Barrier | 51.3 | 53.6 | 47.5 |
Traditional | 35.8 | 34.2 | 33.5 |
Other | 0.04 | 0.04 | 2.0 |
n of cases (conceptions while using contraceptive method) | 240 | 237 | 200 |
IV. Percentage of ever-users of each method who conceived while using: | |||
Medical reversible (n) | 33.3 (84) | 28.0 (82) | 31.7 (63) |
LAM (n) | (2) | (9) | (14) |
Barrier (n) | 73.2 (168) | 60.5** (210) | 54.6** (174) |
Traditional (n) | 95.6 (90) | 96.4 (84) | 98.5 (68) |
. | Percentage of abortion clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
I. Want no more children: | 26.6 | 31.8* | 38.0** |
Previously used a contraceptive method | 70.6 | 69.1 | 61.3** |
Got pregnant while using a method (of all clients) | 49.1 | 42.4 * | 38.0 ** |
n of respondents | 489 | 559 | 527 |
II. Of ever-users, type of method last used: | |||
Medical reversible | 24.3 | 21.2 | 19.5 |
Barrier methods | 48.7 | 54.4 | 53.9 |
Lactational amenorrhoea method (LAM) | 0.6 | 2.3 | 4.3 |
Traditional methods | 26.1 | 21.8 | 21.1 |
Other | 0.3 | 0.3 | 1.2 |
Of ever-users, got pregnant while using a method | 69.6 | 61.4** | 61.9** |
n of cases (ever-users of contraceptives) | 345 | 386 | 323 |
III. Of those who conceived while using a method, % using: | |||
Medical reversible | 11.7 | 9.7 | 10.0 |
LAM | 0.8 | 2.1 | 7.0 |
Barrier | 51.3 | 53.6 | 47.5 |
Traditional | 35.8 | 34.2 | 33.5 |
Other | 0.04 | 0.04 | 2.0 |
n of cases (conceptions while using contraceptive method) | 240 | 237 | 200 |
IV. Percentage of ever-users of each method who conceived while using: | |||
Medical reversible (n) | 33.3 (84) | 28.0 (82) | 31.7 (63) |
LAM (n) | (2) | (9) | (14) |
Barrier (n) | 73.2 (168) | 60.5** (210) | 54.6** (174) |
Traditional (n) | 95.6 (90) | 96.4 (84) | 98.5 (68) |
Differences from baseline proportions significant at: *P < 0.05; **P < 0.01. Estimates based on less than 25 cases omitted.
At the same time, the data show a decline in women seeking abortion who had ever used a contraceptive method. In 2000, 71% of all abortion clients had previously used some type of contraceptive, but this fell to 61% in 2003. Although it says nothing about future behaviour, the baseline figure is strikingly low, and implies that three in every 10 abortion clients interviewed relied on abortion as their chief means of birth prevention. In 2003, an even larger proportion of abortion clients reports no prior experience with contraceptives (Table 3, Panel 1).
About half of prior users report that the last contraceptive used was a barrier method (condom, spermicide, diaphragm), with the rest almost equally distributed between reversible medical methods (oral contraceptives, IUD, injectables, implants and the post-coital pill) and traditional methods (withdrawal, calendar method). Note the increase over time in clients who report having used lactational amenorrhoea (LAM) as a family planning method—a method the project promoted to postpartum women in project sites.6 Together, women who had last used LAM or traditional methods accounted for a quarter of all abortion clients in each survey.
Pregnancies resulting from contraceptive failure
The proportion of abortion clients who report conceiving while using a contraceptive method (due to incorrect use or method failure) is an indication of the extent to which abortions might be reduced by improved family planning services. A decline in the proportion reporting a contraceptive failure is an indirect indication that family planning services are improving access to effective methods or more effective use. Information from this indicator can also be used to improve counselling if high failure rates are associated with specific methods. Almost 50% of all clients interviewed in the baseline survey report that the pregnancy occurred while using a contraceptive method. This proportion declined significantly in the two post-intervention surveys, to 42% and then 38%, respectively (Table 3, Panel 1). Among ever-users of contraceptives, the proportion reporting a contraceptive failure (‘got pregnant while using the method’) also declined significantly over the course of the project, from almost 70% to 62% (Table 3, Panel 2).
As we can see from data shown in the third panel of Table 3, about 10% of the contraceptive failures reportedly occurred while using reversible medical methods. About half occurred while using barrier methods, and another third while using traditional methods. Although there were no significant changes in this pattern over the course of the project, pregnancies occurring while using LAM accounted for a small but increasing proportion of all reported contraceptive failures during the course of the three years, from less than 1% at baseline to 7% in 2003. The differences are not statistically significant. Reported conceptions while using LAM are omitted from the last panel of Table 3 because altogether only 25 women report ever using LAM. Of these 25 women, all but four clients reported that the pregnancy occurred while using LAM.
The reduction in pregnancies reported to occur while using a contraceptive method appears to be due primarily to a reduction among users of barrier methods (Table 3, Panel IV). While the proportion of clients who had last used a barrier method remained fairly stable—about half of all ever-users—there was a significant decline in reports of conception while using a barrier method. Seventy-three per cent of respondents in the pre-intervention survey reported using a barrier method at the time of conception compared with only 55% in 2003 (Table 3, Panel IV).
Reported contraceptive failures among users of other methods remained virtually the same as before the intervention. More than 95% of traditional method users report that the pregnancy occurred while using one of these methods. In 2003, 36% of pill users said they conceived while on the pill, indicating a high rate of user failure. More than 25% of IUD users reported that they were using this method when they became pregnant.
Now that we have examined clients’ prior reproductive behaviour, we next examine the effects of the project on the experiences of abortion clients.
Effects of counselling training on provider services
One aim of the project was to change the access clients had to information about contraception before leaving the abortion facility. Over the 3-year period, our data show a marked increase in post-abortion clients who report receiving contraceptive counselling from their providers on the day of the abortion. When interviewed at baseline, only 41% of abortion clients report being counselled about how to prevent another unplanned pregnancy (Table 4). The proportion that report receiving counselling more than doubled by the time of the second round survey and by the 2003 survey had increased to more than 90% of all abortion clients.
Indicator . | Percentage reporting ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Planning to use a contraceptive method and know which method | 85.3 | 83.5 | 83.3 |
Counselled about contraception before discharge | 41.1 | 82.1*** | 91.5*** |
Discussed a specific method with doctor or midwife | 47.2 | 56.4 | 54.5 |
Received a brochure on pregnancy prevention to take home | 5.9 | 74.8*** | 60.9*** |
Told when to make a follow-up visit | 85.3 | 88.2 | 86.0 |
Say fertility can return immediately or within 2 weeks | 61.6 | 75.8** | 76.5** |
n of respondents | 489 | 559 | 527 |
Of those who had chosen a method before discharge: | |||
Intend to use a modern method | 96.9 | 97.8 | 98.3 |
Intend to use a medical method | 78.6 | 79.4 | 77.6 |
Intend to use sterilization | 3.1 | 3.6 | 5.7 |
Discussed use of this method with doctor or midwife | 55.4 | 63.3 * | 65.4** |
n of cases | 417 | 498 | 439 |
Indicator . | Percentage reporting ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Planning to use a contraceptive method and know which method | 85.3 | 83.5 | 83.3 |
Counselled about contraception before discharge | 41.1 | 82.1*** | 91.5*** |
Discussed a specific method with doctor or midwife | 47.2 | 56.4 | 54.5 |
Received a brochure on pregnancy prevention to take home | 5.9 | 74.8*** | 60.9*** |
Told when to make a follow-up visit | 85.3 | 88.2 | 86.0 |
Say fertility can return immediately or within 2 weeks | 61.6 | 75.8** | 76.5** |
n of respondents | 489 | 559 | 527 |
Of those who had chosen a method before discharge: | |||
Intend to use a modern method | 96.9 | 97.8 | 98.3 |
Intend to use a medical method | 78.6 | 79.4 | 77.6 |
Intend to use sterilization | 3.1 | 3.6 | 5.7 |
Discussed use of this method with doctor or midwife | 55.4 | 63.3 * | 65.4** |
n of cases | 417 | 498 | 439 |
Source: WIN Project Facility Survey 2000, 2002, 2003 data files.
All differences in proportions from baseline, significant at: *P < 0.05, **P < 0.01, ***P < 0.001.
Indicator . | Percentage reporting ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Planning to use a contraceptive method and know which method | 85.3 | 83.5 | 83.3 |
Counselled about contraception before discharge | 41.1 | 82.1*** | 91.5*** |
Discussed a specific method with doctor or midwife | 47.2 | 56.4 | 54.5 |
Received a brochure on pregnancy prevention to take home | 5.9 | 74.8*** | 60.9*** |
Told when to make a follow-up visit | 85.3 | 88.2 | 86.0 |
Say fertility can return immediately or within 2 weeks | 61.6 | 75.8** | 76.5** |
n of respondents | 489 | 559 | 527 |
Of those who had chosen a method before discharge: | |||
Intend to use a modern method | 96.9 | 97.8 | 98.3 |
Intend to use a medical method | 78.6 | 79.4 | 77.6 |
Intend to use sterilization | 3.1 | 3.6 | 5.7 |
Discussed use of this method with doctor or midwife | 55.4 | 63.3 * | 65.4** |
n of cases | 417 | 498 | 439 |
Indicator . | Percentage reporting ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Planning to use a contraceptive method and know which method | 85.3 | 83.5 | 83.3 |
Counselled about contraception before discharge | 41.1 | 82.1*** | 91.5*** |
Discussed a specific method with doctor or midwife | 47.2 | 56.4 | 54.5 |
Received a brochure on pregnancy prevention to take home | 5.9 | 74.8*** | 60.9*** |
Told when to make a follow-up visit | 85.3 | 88.2 | 86.0 |
Say fertility can return immediately or within 2 weeks | 61.6 | 75.8** | 76.5** |
n of respondents | 489 | 559 | 527 |
Of those who had chosen a method before discharge: | |||
Intend to use a modern method | 96.9 | 97.8 | 98.3 |
Intend to use a medical method | 78.6 | 79.4 | 77.6 |
Intend to use sterilization | 3.1 | 3.6 | 5.7 |
Discussed use of this method with doctor or midwife | 55.4 | 63.3 * | 65.4** |
n of cases | 417 | 498 | 439 |
Source: WIN Project Facility Survey 2000, 2002, 2003 data files.
All differences in proportions from baseline, significant at: *P < 0.05, **P < 0.01, ***P < 0.001.
Stated contraceptive preferences post-abortion
Post-abortion contraceptive intentions were believed to be a good indicator of the effectiveness of family planning counselling. However, abortion client intentions to use contraception in the future were high at baseline and remained high in post-intervention surveys. In all three surveys, four out of every five abortion clients said they were planning to use a contraceptive and knew what method they would use (Table 4). Nearly all of the women who named a method choice planned to use a modern contraceptive method, and in each survey almost 80% indicated that they planned to use one of the most effective methods available—medical methods (e.g. IUD, oral contraceptives, implants or post-coital pills) (Table 4, Panel 2).
Indicators of information provision
Among those who chose a specific method prior to discharge, there was a significant increase in method-specific counselling by a health provider over the course of the project (Table 4, Panel 2), which should improve women's knowledge and thus more effective use of the chosen method. The data also show a significant increase in clients who could correctly report when return to the fertile state occurred: in both the second and third round surveys, three-quarters of clients said they could become pregnant again immediately, or within 2 weeks of the abortion, up from 62% at baseline (Table 4). Another significant change is the 10-fold increase in women who report receiving an informational brochure at the abortion facility (Table 4).
Some of the intended changes seem to have taken hold. Nevertheless, our data show a continuing pattern of repeated abortion among women coming to these facilities.
Repeat abortion—a key outcome indicator
Despite the improvements in post-abortion contraceptive counselling, the cross-section of clients interviewed at participating facilities were just as likely at baseline as at the time of both 2002 and 2003 surveys to be having a repeat abortion. Of clients with more than one previous pregnancy, three-quarters reported at least one previous abortion (Table 5). Seventeen per cent of repeat abortion clients were returning for an abortion within a year of a previous one. This amounts to almost one in every 10 abortion clients, a figure that remained constant over the course of the project.
. | Percentage of clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Aborting first pregnancy | 21.7 | 25.8 | 20.1 |
n of respondents | 489 | 559 | 527 |
Of clients with more than 1 pregnancy: | |||
Had a previous abortion | 76.2 | 80.0 | 74.6 |
n of cases | 383 | 415 | 421 |
Of clients with more than 1 abortion: | |||
Last abortion occurred within 12 months | 17.2 | 17.5 | 17.2 |
n of cases | 91 | 332 | 314 |
. | Percentage of clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Aborting first pregnancy | 21.7 | 25.8 | 20.1 |
n of respondents | 489 | 559 | 527 |
Of clients with more than 1 pregnancy: | |||
Had a previous abortion | 76.2 | 80.0 | 74.6 |
n of cases | 383 | 415 | 421 |
Of clients with more than 1 abortion: | |||
Last abortion occurred within 12 months | 17.2 | 17.5 | 17.2 |
n of cases | 91 | 332 | 314 |
. | Percentage of clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Aborting first pregnancy | 21.7 | 25.8 | 20.1 |
n of respondents | 489 | 559 | 527 |
Of clients with more than 1 pregnancy: | |||
Had a previous abortion | 76.2 | 80.0 | 74.6 |
n of cases | 383 | 415 | 421 |
Of clients with more than 1 abortion: | |||
Last abortion occurred within 12 months | 17.2 | 17.5 | 17.2 |
n of cases | 91 | 332 | 314 |
. | Percentage of clients . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
Aborting first pregnancy | 21.7 | 25.8 | 20.1 |
n of respondents | 489 | 559 | 527 |
Of clients with more than 1 pregnancy: | |||
Had a previous abortion | 76.2 | 80.0 | 74.6 |
n of cases | 383 | 415 | 421 |
Of clients with more than 1 abortion: | |||
Last abortion occurred within 12 months | 17.2 | 17.5 | 17.2 |
n of cases | 91 | 332 | 314 |
The surveys were originally designed only to measure this key indicator, prevalence of repeat abortions within the project timeframe, and information pertaining to their previous abortion experience was not obtained. However, despite the small numbers, we take the opportunity provided by these data to look more closely at the characteristics of this group of women who become pregnant and rapidly repeat the abortion procedure.
Profile of frequent repeat abortion clients
Although only 50 to 60 women in each sample fall into this group, we observe a significant decrease in the proportion of these women reporting a contraceptive failure (e.g. using a method when conception occurred), from 83% at baseline to just over half in 2003 (Table 6). To try to characterize these women more clearly, we used multivariate logistic regression models to assess the relative contribution of a number of characteristics to the odds that a woman would repeat an abortion within a year.
. | Percentage responding ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
One year repeat abortion clients who: | |||
Were in a current union | 70.6 | 77.6 | 81.5 |
Ever used contraception | 70.6 | 63.8 | 59.3 |
Were using contraception when conception occurred | 83.3 | 64.9* | 53.1** |
Want no more children | 23.5 | 34.5 | 31.5 |
Plan to use a contraceptive method in future | 70.0 | 74.0 | 72.2 |
Plan to use medical method in future | 89.6 | 81.7 | 81.3 |
n of cases | 50 | 58 | 54 |
. | Percentage responding ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
One year repeat abortion clients who: | |||
Were in a current union | 70.6 | 77.6 | 81.5 |
Ever used contraception | 70.6 | 63.8 | 59.3 |
Were using contraception when conception occurred | 83.3 | 64.9* | 53.1** |
Want no more children | 23.5 | 34.5 | 31.5 |
Plan to use a contraceptive method in future | 70.0 | 74.0 | 72.2 |
Plan to use medical method in future | 89.6 | 81.7 | 81.3 |
n of cases | 50 | 58 | 54 |
Source: WIN Project Facility Survey 2000, 2002, 2003 data files.
*P < 0.05, **P < 0.01.
. | Percentage responding ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
One year repeat abortion clients who: | |||
Were in a current union | 70.6 | 77.6 | 81.5 |
Ever used contraception | 70.6 | 63.8 | 59.3 |
Were using contraception when conception occurred | 83.3 | 64.9* | 53.1** |
Want no more children | 23.5 | 34.5 | 31.5 |
Plan to use a contraceptive method in future | 70.0 | 74.0 | 72.2 |
Plan to use medical method in future | 89.6 | 81.7 | 81.3 |
n of cases | 50 | 58 | 54 |
. | Percentage responding ‘yes’ . | ||
---|---|---|---|
. | 2000 . | 2002 . | 2003 . |
One year repeat abortion clients who: | |||
Were in a current union | 70.6 | 77.6 | 81.5 |
Ever used contraception | 70.6 | 63.8 | 59.3 |
Were using contraception when conception occurred | 83.3 | 64.9* | 53.1** |
Want no more children | 23.5 | 34.5 | 31.5 |
Plan to use a contraceptive method in future | 70.0 | 74.0 | 72.2 |
Plan to use medical method in future | 89.6 | 81.7 | 81.3 |
n of cases | 50 | 58 | 54 |
Source: WIN Project Facility Survey 2000, 2002, 2003 data files.
*P < 0.05, **P < 0.01.
Table 7 displays the distribution of clients interviewed in 2003 who had at least one previous abortion, by the timing of their last abortion and predictive factors we considered: demographic characteristics, fertility intentions, prior contraceptive use, stated reasons for the abortion and future contraceptive intentions. One other factor was also considered: the woman's knowledge of when she was again at risk of conceiving.
. | Abortion within 1 year . | |
---|---|---|
. | No . | Yes . |
Marital status | ||
Married | 61.9 | 40.7 |
In unregistered marriage | 21.5 | 40.7 |
Single (never married) | 8.8 | 14.8 |
Divorced/separated/widowed | 7.7 | 3.7 |
Age | ||
15–24 | 16.5 | 53.7 |
25–34 | 56.5 | 37.0 |
35–49 | 26.9 | 9.3 |
Education | ||
Less than complete secondary | 1.5 | 1.9 |
Completed secondary | 32.8 | 35.2 |
Any higher post-secondary | 65.6 | 63.0 |
Planning to have a child in future | ||
Yes | 38.1 | 61.1 |
No | 51.5 | 31.5 |
Don’t know | 10.4 | 7.4 |
Ever used a contraceptive method | 64.2 | 59.3 |
Of users, type last used: (n) | (167) | (32) |
Medical | 26.3 | 15.6 |
Barrier | 44.3 | 78.1 |
Traditional | 28.1 | 3.1 |
Other | 1.2 | 3.1 |
Using contraception when conception occurred | 39.2 | 31.5 |
Not using | 25.0 | 27.8 |
Never used | 35.8 | 40.7 |
Stated reason for abortion | ||
‘Not a good time’ to have a baby (68) | 23.8 | 27.8 |
Socio-economic reasons (109) | 33.5 | 55.6 |
Wants no more children (93) | 33.1 | 13.0 |
Other (no partner, health reasons) (44) | 14.6 | 11.1 |
Plan to use a contraceptive method in future | ||
Yes | 81.5 | 72.2 |
No | 13.8 | 13.0 |
Have not yet thought about it | 4.6 | 14.8 |
Correct knowledge of when fertility returns | ||
Yes | 79.6 | 88.9 |
No | 20.4 | 11.1 |
Total | 100.0 | 100.0 |
n of cases | 260 | 54 |
. | Abortion within 1 year . | |
---|---|---|
. | No . | Yes . |
Marital status | ||
Married | 61.9 | 40.7 |
In unregistered marriage | 21.5 | 40.7 |
Single (never married) | 8.8 | 14.8 |
Divorced/separated/widowed | 7.7 | 3.7 |
Age | ||
15–24 | 16.5 | 53.7 |
25–34 | 56.5 | 37.0 |
35–49 | 26.9 | 9.3 |
Education | ||
Less than complete secondary | 1.5 | 1.9 |
Completed secondary | 32.8 | 35.2 |
Any higher post-secondary | 65.6 | 63.0 |
Planning to have a child in future | ||
Yes | 38.1 | 61.1 |
No | 51.5 | 31.5 |
Don’t know | 10.4 | 7.4 |
Ever used a contraceptive method | 64.2 | 59.3 |
Of users, type last used: (n) | (167) | (32) |
Medical | 26.3 | 15.6 |
Barrier | 44.3 | 78.1 |
Traditional | 28.1 | 3.1 |
Other | 1.2 | 3.1 |
Using contraception when conception occurred | 39.2 | 31.5 |
Not using | 25.0 | 27.8 |
Never used | 35.8 | 40.7 |
Stated reason for abortion | ||
‘Not a good time’ to have a baby (68) | 23.8 | 27.8 |
Socio-economic reasons (109) | 33.5 | 55.6 |
Wants no more children (93) | 33.1 | 13.0 |
Other (no partner, health reasons) (44) | 14.6 | 11.1 |
Plan to use a contraceptive method in future | ||
Yes | 81.5 | 72.2 |
No | 13.8 | 13.0 |
Have not yet thought about it | 4.6 | 14.8 |
Correct knowledge of when fertility returns | ||
Yes | 79.6 | 88.9 |
No | 20.4 | 11.1 |
Total | 100.0 | 100.0 |
n of cases | 260 | 54 |
Source: WIN Project Facility Survey data, 2003.
. | Abortion within 1 year . | |
---|---|---|
. | No . | Yes . |
Marital status | ||
Married | 61.9 | 40.7 |
In unregistered marriage | 21.5 | 40.7 |
Single (never married) | 8.8 | 14.8 |
Divorced/separated/widowed | 7.7 | 3.7 |
Age | ||
15–24 | 16.5 | 53.7 |
25–34 | 56.5 | 37.0 |
35–49 | 26.9 | 9.3 |
Education | ||
Less than complete secondary | 1.5 | 1.9 |
Completed secondary | 32.8 | 35.2 |
Any higher post-secondary | 65.6 | 63.0 |
Planning to have a child in future | ||
Yes | 38.1 | 61.1 |
No | 51.5 | 31.5 |
Don’t know | 10.4 | 7.4 |
Ever used a contraceptive method | 64.2 | 59.3 |
Of users, type last used: (n) | (167) | (32) |
Medical | 26.3 | 15.6 |
Barrier | 44.3 | 78.1 |
Traditional | 28.1 | 3.1 |
Other | 1.2 | 3.1 |
Using contraception when conception occurred | 39.2 | 31.5 |
Not using | 25.0 | 27.8 |
Never used | 35.8 | 40.7 |
Stated reason for abortion | ||
‘Not a good time’ to have a baby (68) | 23.8 | 27.8 |
Socio-economic reasons (109) | 33.5 | 55.6 |
Wants no more children (93) | 33.1 | 13.0 |
Other (no partner, health reasons) (44) | 14.6 | 11.1 |
Plan to use a contraceptive method in future | ||
Yes | 81.5 | 72.2 |
No | 13.8 | 13.0 |
Have not yet thought about it | 4.6 | 14.8 |
Correct knowledge of when fertility returns | ||
Yes | 79.6 | 88.9 |
No | 20.4 | 11.1 |
Total | 100.0 | 100.0 |
n of cases | 260 | 54 |
. | Abortion within 1 year . | |
---|---|---|
. | No . | Yes . |
Marital status | ||
Married | 61.9 | 40.7 |
In unregistered marriage | 21.5 | 40.7 |
Single (never married) | 8.8 | 14.8 |
Divorced/separated/widowed | 7.7 | 3.7 |
Age | ||
15–24 | 16.5 | 53.7 |
25–34 | 56.5 | 37.0 |
35–49 | 26.9 | 9.3 |
Education | ||
Less than complete secondary | 1.5 | 1.9 |
Completed secondary | 32.8 | 35.2 |
Any higher post-secondary | 65.6 | 63.0 |
Planning to have a child in future | ||
Yes | 38.1 | 61.1 |
No | 51.5 | 31.5 |
Don’t know | 10.4 | 7.4 |
Ever used a contraceptive method | 64.2 | 59.3 |
Of users, type last used: (n) | (167) | (32) |
Medical | 26.3 | 15.6 |
Barrier | 44.3 | 78.1 |
Traditional | 28.1 | 3.1 |
Other | 1.2 | 3.1 |
Using contraception when conception occurred | 39.2 | 31.5 |
Not using | 25.0 | 27.8 |
Never used | 35.8 | 40.7 |
Stated reason for abortion | ||
‘Not a good time’ to have a baby (68) | 23.8 | 27.8 |
Socio-economic reasons (109) | 33.5 | 55.6 |
Wants no more children (93) | 33.1 | 13.0 |
Other (no partner, health reasons) (44) | 14.6 | 11.1 |
Plan to use a contraceptive method in future | ||
Yes | 81.5 | 72.2 |
No | 13.8 | 13.0 |
Have not yet thought about it | 4.6 | 14.8 |
Correct knowledge of when fertility returns | ||
Yes | 79.6 | 88.9 |
No | 20.4 | 11.1 |
Total | 100.0 | 100.0 |
n of cases | 260 | 54 |
Source: WIN Project Facility Survey data, 2003.
More than 60% of women repeating the procedure rapidly plan to have a child in future, compared with only about 40% of other repeat clients. Nearly 80% had used a barrier method of contraception in the past, two times more than other repeat clients (Table 7). Only 3% had used traditional methods compared with nearly 30% of other repeat abortion clients. When asked their reason for obtaining an abortion, 56% of women repeating the procedure within a year mentioned socio-economic reasons, compared with only 34% of women whose previous abortion was not as recent.
We entered all the above factors into the regression model simultaneously; the results are displayed in Table 8. Given the small sample size and the likelihood that a number of the variables are inter-correlated, we experimented with other models, using backward elimination based on the likelihood ratio test to obtain a more parsimonious model. The parameter estimates for both the full and parsimonious models were similar and results are consistent, and so although we present both models, we discuss only the parameter estimates from the full model.
. | Log odds of abortion within 1 year of previous abortion . | |
---|---|---|
. | FULL model . | Parsimonious model . |
Marital status | ||
Married | 1 | |
In unregistered marriage | 2.19* | |
Single (never married) | 1.08 | |
Divorced/separated/widowed | 1.32 | |
Age | ||
15–24 | 5.88*** | 5.44*** |
25–34 | 1 | 1 |
35–49 | 0.68 | 0.65 |
Level of education | ||
Less than complete secondary | 1.06 | |
Completed secondary | 0.63 | |
Any higher post-secondary | 1 | |
Planning to have child in future | ||
Yes | 1 | |
No | 1.15 | |
Don’t know | 1.09 | |
Contraceptive status prior to conception | ||
Conceived while using method | 0.93 | |
Not using at time of conception | 1.03 | |
Never used contraception | 1 | |
Reason for abortion | ||
Socio-economic | 1.84 | 1.90 |
Other | 1 | 1 |
Know what method will use | ||
Yes | 1 | 1 |
No | 1.10 | 0.90 |
Have not yet thought about it | 4.14* | 3.57* |
Know when fertility returns | ||
Yes | 1 | 1 |
No | 2.91* | 3.30* |
n of cases | 313 | 313 |
. | Log odds of abortion within 1 year of previous abortion . | |
---|---|---|
. | FULL model . | Parsimonious model . |
Marital status | ||
Married | 1 | |
In unregistered marriage | 2.19* | |
Single (never married) | 1.08 | |
Divorced/separated/widowed | 1.32 | |
Age | ||
15–24 | 5.88*** | 5.44*** |
25–34 | 1 | 1 |
35–49 | 0.68 | 0.65 |
Level of education | ||
Less than complete secondary | 1.06 | |
Completed secondary | 0.63 | |
Any higher post-secondary | 1 | |
Planning to have child in future | ||
Yes | 1 | |
No | 1.15 | |
Don’t know | 1.09 | |
Contraceptive status prior to conception | ||
Conceived while using method | 0.93 | |
Not using at time of conception | 1.03 | |
Never used contraception | 1 | |
Reason for abortion | ||
Socio-economic | 1.84 | 1.90 |
Other | 1 | 1 |
Know what method will use | ||
Yes | 1 | 1 |
No | 1.10 | 0.90 |
Have not yet thought about it | 4.14* | 3.57* |
Know when fertility returns | ||
Yes | 1 | 1 |
No | 2.91* | 3.30* |
n of cases | 313 | 313 |
*P < 0.05, **P < 0.01, ***P < 0.001.
. | Log odds of abortion within 1 year of previous abortion . | |
---|---|---|
. | FULL model . | Parsimonious model . |
Marital status | ||
Married | 1 | |
In unregistered marriage | 2.19* | |
Single (never married) | 1.08 | |
Divorced/separated/widowed | 1.32 | |
Age | ||
15–24 | 5.88*** | 5.44*** |
25–34 | 1 | 1 |
35–49 | 0.68 | 0.65 |
Level of education | ||
Less than complete secondary | 1.06 | |
Completed secondary | 0.63 | |
Any higher post-secondary | 1 | |
Planning to have child in future | ||
Yes | 1 | |
No | 1.15 | |
Don’t know | 1.09 | |
Contraceptive status prior to conception | ||
Conceived while using method | 0.93 | |
Not using at time of conception | 1.03 | |
Never used contraception | 1 | |
Reason for abortion | ||
Socio-economic | 1.84 | 1.90 |
Other | 1 | 1 |
Know what method will use | ||
Yes | 1 | 1 |
No | 1.10 | 0.90 |
Have not yet thought about it | 4.14* | 3.57* |
Know when fertility returns | ||
Yes | 1 | 1 |
No | 2.91* | 3.30* |
n of cases | 313 | 313 |
. | Log odds of abortion within 1 year of previous abortion . | |
---|---|---|
. | FULL model . | Parsimonious model . |
Marital status | ||
Married | 1 | |
In unregistered marriage | 2.19* | |
Single (never married) | 1.08 | |
Divorced/separated/widowed | 1.32 | |
Age | ||
15–24 | 5.88*** | 5.44*** |
25–34 | 1 | 1 |
35–49 | 0.68 | 0.65 |
Level of education | ||
Less than complete secondary | 1.06 | |
Completed secondary | 0.63 | |
Any higher post-secondary | 1 | |
Planning to have child in future | ||
Yes | 1 | |
No | 1.15 | |
Don’t know | 1.09 | |
Contraceptive status prior to conception | ||
Conceived while using method | 0.93 | |
Not using at time of conception | 1.03 | |
Never used contraception | 1 | |
Reason for abortion | ||
Socio-economic | 1.84 | 1.90 |
Other | 1 | 1 |
Know what method will use | ||
Yes | 1 | 1 |
No | 1.10 | 0.90 |
Have not yet thought about it | 4.14* | 3.57* |
Know when fertility returns | ||
Yes | 1 | 1 |
No | 2.91* | 3.30* |
n of cases | 313 | 313 |
*P < 0.05, **P < 0.01, ***P < 0.001.
The strongest predictor that a woman is returning for abortion within one year is her age (full model, Table 8). Women aged 15–24 were almost six times as likely as those aged 25–34 to repeat the procedure within a year. Women living in an unregistered marriage were twice as likely as women in formal unions to repeat the procedure within a year. Level of education, intention to have a future birth and contraceptive status prior to the last conception did not independently contribute to predicting the odds of repeating the procedure quickly. A woman repeating the procedure rapidly is almost twice as likely as other repeat abortion clients to state a reason related to her socio-economic situation. Although this factor does not quite attain statistical significance, it contributes an important dimension of information, and was retained by the parsimonious model.
When all other factors in the model are held constant, two other factors contribute significantly to predicting which women are likely to repeat the procedure rapidly: a woman's knowledge of when she becomes fecund again following an abortion, and her intention to use a contraceptive method after the current abortion, which might be considered an indicator of the woman's ability or desire to plan ahead. Women who report not having thought about what contraceptive method they would use in future were four times as likely as other women to repeat an abortion within one year. All else equal, not having correct information about when fertility would return following an abortion was also a strong predictor: women without correct knowledge were almost three times as likely as other women to repeat the procedure rapidly.
Discussion
The project interventions appear to have extended the coverage of contraceptive counselling to nearly all abortion clients. Although the quality of communication and the details of what is communicated between provider and client cannot be assessed from our data, women's reports suggest that they retained some of the key information imparted during counselling and in brochures. These data suggest that the training for providers resulted in increased, and possibly more effective, discussions about contraception with abortion clients.
In all three surveys, more than 80% of post-abortion clients expressed not only an intention to use medical contraception but had identified their method of choice. More than three-quarters intended to use a medical method, the most efficacious type of contraception, a finding that indicates that women who have just had an abortion are highly motivated to use an effective contraceptive.
Given the ineffective use of LAM among the postpartum women who appeared in the abortion clinics during the course of the project, this method may be a poor choice to promote in the Russian context.7 Our data show that understanding of LAM was low among both postpartum women and their doctors. In places such as Russia, where alternative modern contraceptives appropriate for breastfeeding mothers are available, it makes little sense to send a woman out of the delivery ward with a method that is effective for only six months, at most, and which requires following strict and very specific conditions that must all be met for the method to be effective.
The large proportion of users of medical methods who report that they were using the method when conception occurred is reason for concern. Copper IUDs are effective for only 4 to 10 years, and we have no information on the effectiveness of Russian-made IUDs. These failures could be caused if the device needed replacement, or if it was expelled without the woman's knowledge. These reported failures of what should be highly effective methods when used correctly needs to be addressed.
More women were provided more information after the interventions began, which should mean that they are better prepared to carry out their intention to use effective contraception. Method-specific information reached a larger proportion of clients, yet all three rounds of facility surveys revealed that a large proportion of abortion clients—between 30 and 40%—were not actually using any contraceptive method at the time the conception occurred. About 17% of repeat abortion clients reported a previous abortion in the past year, unchanged from the start of the project.
These findings suggest that certain women conceive again soon after an abortion procedure and are more likely to abort again rather than try to prevent the unwanted pregnancy. Their need to prevent an unwanted or mistimed birth appears to be stronger than their dislike of abortion as a means of birth control. Compared with other abortion clients, these women form a distinct group: they are younger and more likely than other clients to be living in an unregistered marriage. Socio-economic reasons for the abortion outweigh others. They have given less thought to future contraceptive needs, and have incomplete knowledge that could help prevent a future unplanned pregnancy. If they had tried a contraceptive method, it was likely to be a barrier method rather than a traditional method.
The culture does not proscribe abortion; on the contrary, it has been an acceptable, if undesirable, method of birth control for much of the 20th century. The procedure is widely available and easily accessible, reducing the need for women to practice consistent and effective contraception, but most women say that they would prefer to prevent unwanted pregnancies through the use of modern contraception.
Multiple factors probably contribute to this disparity between women's desire to use modern contraception and their practice of induced abortion in these Russian cities. In order to design programmes to better help women to achieve their reproductive goals using means they find acceptable, we need more information on what those factors are as well as more information from women themselves about what motivates their behaviour.
The evidence presented here suggests that the project interventions were implemented effectively, but the training, increased post-abortion counselling, and other information provision did not diminish repeat abortion rates at the project sites, and the gap between knowledge and practice of effective contraception remains.
One limitation of this evaluation stems from the cross-sectional nature of the three client surveys, and the fact that clients cannot be linked to the providers who cared for them. The findings are based on unvalidated reports of women's experiences, which may be biased in unknown ways. Further, some of the women who repeat the procedure rapidly may have obtained their previous abortion elsewhere, and were not exposed to the counselling intervention. Even if they attended a participating facility, some of these women may not have received contraceptive counselling at the time of their previous abortion. Some may have received counselling but for some reason were unable to use their chosen contraceptive method or used it inconsistently, and others may have chosen not to use a contraceptive method at all. Another limitation of this study is the lack of observations of provider–client interactions and the more direct insight into changes in the quality of the counselling these might provide. The structured questionnaire, with its limited opportunities for open-ended responses, provided little information about client motivations.
Conclusions
It seems clear that the women who resort to multiple procedures in quick succession are among the most disadvantaged women in our sample. They are young, living in informal unions and have insufficient knowledge to protect themselves from further unwanted conceptions. Although lack of planning and lack of knowledge characterize these women, our results demonstrate that in Russia, and perhaps other countries in the region, programmes designed to respond to women's needs and preferences must also address other barriers that prevent women from choosing safer ways to achieve their reproductive goals.
In light of our findings, client-centred counselling about a range of contraceptive methods, while probably effective in helping some women avoid repeated abortions, is apparently not sufficient to meet the needs of all potential clients, especially women at greatest risk of rapidly repeating the procedure. These women may need more support than others to achieve their reproductive goals safely.
One obstacle is the restricted range of long-term or permanent contraceptive methods. A small but growing proportion of abortion clients mentioned that they would prefer to use tubal ligation as their future contraceptive method, but less than 2% of women interviewed in household surveys reported use of this method (David et al. 2000, 2003). We have no information on the cost of this procedure, which is considerably restricted and in most oblasts requires client payment. The acceptability of expanding access to permanent methods of contraception, or to high quality IUDs, including progesterone-releasing IUDs,8 inserted correctly, may be a safer, more acceptable alternative to abortion and should be investigated.
The women most likely to repeat the abortion procedure may also be less able to afford imported devices and the oral contraceptives they are advised to use until the IUD can be inserted. These women are likely to benefit most from having the option of IUD insertion at the time of the abortion procedure. These institutional barriers to improving post-abortion care need to be addressed, as well as the causes of apparent IUD failure reported earlier.
Post-abortion counselling could usefully focus not only on the range of pregnancy prevention measures that are available alternatives to abortion, but also refer to what is known about the relative health risks of these methods compared with the abortion procedure. At present, this kind of information is not presented to women, and providers, while not unaware, may also have only an incomplete understanding of the actual morbidity risks associated with contraceptive methods relative to abortion. A review of hospital records to document the extent of post-abortion complications, which are not currently tracked through the health information system, would be a useful first step in providing more information about these risks in the Russian context.
Our findings also point to the need for a more focused investigation of the women most at risk of multiple repeat abortions, to identify their reasons for resorting to a procedure that most women do not like and that may put them at greater risk of subsequent morbidity. An in-depth examination of women who rely frequently on abortion for their birth control needs would be a next step toward understanding these women's motivations and the obstacles they face, and help to design more effective reproductive health programmes.
An important lesson learned in the Women and Infant Health Project is that Russian medical practitioners are eager to learn from developments in medicine elsewhere. Their enthusiastic reception of the evidence-based approach to medical care introduced in WIN Project training courses presents an opportunity. It should be possible to work with highly motivated physicians and with the health system to find ways to ensure that more contraceptive options are available, that providers are well equipped to offer the full array of contraceptive options in a high quality manner, and are recompensed appropriately for doing so. The evidence presented in this paper suggests that women's fertility control needs and preferences are not being met by the existing system. Addressing women's reproductive health needs in Russia, and in other countries where similar conditions prevail, requires a more focused approach to delivering client-centred care than may be the case in other settings.
Acknowledgements
The authors are indebted to Michael Reich, Alaka Basu, George Zeidenstein, Mary Lee Mantz, Leo Morris, Emma Ottolenghi and two anonymous reviewers for their valuable comments on an earlier version of this paper. We are also grateful for comments received during a presentation at the Harvard Center for Population and Development Studies and at the IUSSP Seminar on Reproductive Health Issues in Eastern Europe and the Former Soviet Union held in Bucharest in 2004. We also gratefully acknowledge the support provided by John Snow, Inc. during the preparation of this paper.
The US Agency for International Development provided funding for the Women and Infant Health Project, including collection of the data used in this paper, under the terms of Contract No. HRN-1–00–98–0032–00 Delivery Order No. 803. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID. Any omissions or shortcomings in the paper remain the responsibility of the authors.
Endnotes
Until recently at least part of this decline was thought to be an artefact of incomplete reports, but a recent study using data from three independent surveys appears to confirm the validity of official statistics and the sharp decline they record (Philipov et al. 2004).
In Veliky Novgorod and Berezniki, all women's health care facilities participated in the project; in the larger city of Perm, the project worked in 10 facilities, including four of the 19 facilities that provided abortion services in that city.
An introduction to the Comprehensive PAC Training Curriculum for four local ‘master’ trainers; a Training of Trainers (ToT) workshop in Family Planning Counselling for 12 family planning specialists; an Advanced Reproductive Health ToT (for 26 ‘best’ trainers who would work under the ‘master’ trainers).
Information collected from abortion providers was not as informative as client reports, in part because even before training began many providers understood some components of ‘ideal’ practice. Their reports of current practice diverged from the experiences reported by clients, and are not discussed in this report. A more complete description of the results of the provider surveys is reported elsewhere (David and Vartapetova 2003).
Adjustments made following the pretest included response categories, replacement of medical terms with colloquial ones for some items, and clarification of instructions for filter questions.
The proportion of postpartum clients who say they plan to use LAM as their first contraceptive rose from only two women at baseline (less than 1%) to 17% in 2003 (data not shown).
A critical appraisal of LAM as a contraceptive method is found in Bracher (1992).
Progesterone-releasing IUDs are currently a very expensive option, making them a less attractive alternative.
References
Author notes
This study was completed while the corresponding author was Senior Evaluation Advisor with John Snow, Inc., 44 Farnsworth Street, Boston, MA, USA.