Hum. Reprod. Advance Access originally published online on March 27, 2008
Human Reproduction 2008 23(6):1320-1323; doi:10.1093/humrep/den101
Induced abortions previous to IVF: an epidemiologic register-based study from Finland
1 National Research and Development Centre for Welfare and Health (STAKES), PO Box 220, 00531 Helsinki, Finland 2 School of Public Health, 33014 University of Tampere, Finland
3 Correspondence address. Tel: +358-9-3967-2307; Fax: +358-9-3967-2227; E-mail: elina.hemminki{at}stakes.fi
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
BACKGROUND: The purpose of this study was to identify how many women treated for infertility had an abortion history, as well as when those abortions were carried out, and for what reasons.
METHODS: Data on all women treated in Finland from 1996–1998 for infertility either with IVF (n = 9175) or ovulation induction (OI, n = 10 254) and the age-matched controls of IVF women were linked to the Abortion and Hospital Discharge Registers for the period 1969–2000.
RESULTS: A notable proportion of IVF women (12%) and OI women (11%) had previous induced abortion(s). Practically all abortions were for social or age reasons. Most IVF women (72% of n = 1099) had their most recent abortion more than 10 years previous to fertility treatment, but more recently among OI women (45% of n = 1123 of the most recent abortions were in the preceding 10 years). Many IVF- and OI women were young and single at the time of the most recent abortion. At the time of IVF treatment most women were aged over 30 and married; OI women were also frequently married, but 42% of them were aged younger than 30.
CONCLUSIONS: At different points in their life, women may rely on opposite fertility regulation strategies. Health care professionals providing IVF need to consider the possibility of a previous abortion. Young women need information on the possibility of future infertility in later age.
Key words: infertility/IVF/abortions/assisted fertilization/register linkages
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
In developed countries, access to effective contraception and induced abortions have given women control over the decision not to have a child. In contrast, a woman’s decision to have a child is no guarantee that she can have a child. The effectiveness of fertility treatments is still limited (Nyboe Andersen et al., 2007
), the partner’s co-operation is needed and, at least in Finland, many women lack information on the negative effects of age on fertility (Sannisto et al., 2001). Decisions not to continue with a pregnancy often occur at a young age or when the father of the child is not supportive to pregnancy or not suitable as the partner; the possibility of remaining unintentionally childless may seem remote. If a woman faces infertility later in life, the history of a previous induced abortion may be a psychological burden, because the opportunity to have a child was once missed. We found no published studies on how common this situation is.
Baseline data from our earlier study on the health consequences of IVF (Klemetti et al., 2004, 2005b) showed that the pregnancy history of some IVF-treated women in the birth register included induced abortions (unpublished data). In Finland, in the 1990s, about 5% of children were born either after IVF or ovulation induction (OI) (Klemetti et al., 2005a), and the rate of IVF treatments per fertile-aged woman was 7.3 cycles per 1000 women per year (Klemetti et al., 2004), one of the highest in Europe (Nyboe Andersen et al., 2007). Treatments are commonly made in the private sector, but partly reimbursed by public funds. Despite financial barriers and age limits in the public sector, IVF is well accessed (Klemetti et al., 2004).
Induced abortions (hereafter simply abortions) are not subject to a woman’s choice alone and a certificate from a physician indicating a legally founded reason is required. Nevertheless, since 1970, abortions have been readily available (Gissler, 1999; Knudsen et al., 2003). Abortions are carried out mainly in public hospitals and the costs to women are small. Compared with other European countries, Finland has low abortion rates: since 1990, there were less than 10 abortions per 1000 women aged 15–49 years (Induced Abortion Worldwide, 1999; WHO, 2006).
The aim of this study was to identify how many women who were treated for infertility in Finland had an abortion history, when those abortions were carried out, and for what reasons.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
The study links data on register-based cohorts of women who had received IVF or OI during the period 1996–1998 to nationwide health registers with the purpose of identifying any history of induced abortions. As part of a previous study, two exposed cohorts had already been formed of women who had received IVF (IVF women) or (only) OI (OI women) during the period 1996–1998 and a comparison group of the IVF women. The methods for creating the cohorts and the coverage of treated women have been described in detail elsewhere (Klemetti et al., 2005b
; Hemminki et al., 2003). The exposed cohorts were formed on the basis of reimbursement claims and included practically all Finnish women who had received fertility treatment in the three-year period. Women were identified from the reimbursement files of the Social Insurance Institution (SII) with a pre-designed algorithm based on specific drugs and investigations. The control women were a random sample of non-treated Finnish women (from the population register), matched by age to the IVF women. Women’s background characteristics were obtained from the SII and the Population Register. Births as a result of IVF (n = 4559) and OI treatment (n = 4467) and the perinatal health of the children were obtained from the Finnish Medical Birth Register using the woman’s unique identification (ID) number and the children’s date of birth as linkage keys. To identify a history of abortion, the cohort women were matched against the Finnish Hospital Discharge Register (HDR) for the period 1969–1982 and the Finnish Abortion Register for the period 1983–2000 using the women’s unique personal ID numbers. Abortions from the HDR were selected by diagnosis, with the HDR collecting information on inpatient care by diagnosis (codes 640–641, the 8th revision of the International Classification of Diseases). The diagnosis in the HDR included the main diagnosis and two secondary diagnoses. The abortion diagnoses were combined into medical (ICD8-codes 640.90, 640.93, 641.91, 641.92), social (640.91, 641.90) and other or not known (640.00, 640.94, 640.99, 641.00–641.69, 641.94–641.99).
The abortion register has been operating since 1977 (Gissler and Haukka, 2004), and has been computerized since 1983 (Gissler et al., 1996). Legislation demands obligatory notifications from physicians to the register. The reasons for abortions given in the register are those listed in the legislation. We combined them into medical (‘medical’, ‘potential foetal defect’, ‘detected foetal defect’), social, age or parity (‘less than 17 years’, ‘over 40 years’, ‘4+ children’), other (‘criminal’, ‘limited ability to take care of children’ or ‘unknown’).
Differences between groups were compared using the t-test, a test for relative proportions and the chi-square test.
The whole project was successfully reviewed by the National Research and Development Centre for Welfare and Health research ethics committee (January 2000), while the Data Protection Authority was also consulted and the specific permissions obtained from each register controller.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
A notable proportion of women who received infertility treatment had experienced one or more induced abortions in their history, even though that was less common than among the women in the comparison group (Table I). Most abortions previous to treatment were for social reasons and some for age reasons. Table I gives the reasons for the most recent abortions, and similar percentages were found for first abortions (59% for social reasons among IVF women, 69% among OI women and 70% among comparison women, data not shown).
|
IVF women with an abortion history were somewhat older, from a lower social class and had less often had children previous to IVF treatment compared with other IVF women (Supplementary material, Table SI). All differences were small, though statistically significant, with similar differences found among OI women. The proportions of women who had a live birth after treatment were similar in IVF women with or without an abortion history, but somewhat more OI women without an abortion history had a live birth than women with an abortion history.
Many women treated for infertility, though not all, were young when they had their (most recent) abortion previous to treatment (Table II). For detailed data, see Supplementary material, Table SII.
|
Most IVF women had had the most recent abortion more than 10 years previous to treatment, and only a few had had it less than 5 years previously (Table II). Among OI women there were more women with a short interval between the most recent abortion and first treatment (Table II) than among IVF women (16 versus 7%, P < 0.001), and their timing of abortions resembled more that of women in the comparison group.
Most women were single at the time of the most recent abortion (Table III), and at the time of infertility treatment over half were married. Information on whether they were in a steady relationship or were cohabiting is not available from our data source.
|
Abortion after treatment for infertility was found in 1.2% of IVF women and in 1.9% of OI women. Most of these post-treatment abortions were carried out for social reasons, even while the proportion of abortions for medical reasons (21% of all abortions among IVF and 18% among OI women) was notably higher than among the women in the comparison group (6%, P < 0.001 in both groups). Most of the abortions for social reasons (83% among IVF and 79% among OI women) occurred later than 48 weeks after the last IVF treatment, suggesting that they were not carried out for pregnancies resulting from the treatment.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
This record-linkage study showed that a notable proportion of women treated with IVF (12%) or OI (11%) had one or more induced abortions in their history. We found no previous studies with which to compare our findings. Taking into account the relatively low rate of induced abortions (abortions) in Finland, the history of abortions among infertility patients may be even more common in other countries. Our analysis produced a secondary finding that some women had abortions in the period after infertility treatments, though typically not of pregnancies resulting directly from treatment. This phenomenon of post-treatment abortion has been reported on earlier in England (Edozien, 1998
; Marcus and Brinsden, 2000).
This study was based on administrative registers and has the merits and drawbacks of such a study design. The merits include good representativeness (all women rather than a sample), large numbers and availability of the personal ID number, which helps ensure accuracy in record linkage. Drawbacks include possibly incorrect information in the available register data and its narrow scope. Because the abortion register is founded on legal requirements, its coverage and accuracy are good (Gissler et al., 1996). Information prior to 1983 was provided by the HDR. In that register some women may have been wrongly identified as having had abortions, and some may have been missed due to wrong diagnosis, wrong ID code or the event being missed. No recent validation study exists for the HDR; however, in the 1980s, 95% of hospitalizations and 97% of the main diagnoses in regard to pregnancy and birth were correctly recorded (Keskimäki and Aro, 1991). In the period 1969–1982, the HDR included information only on inpatient care, though at that time women who were having an induced abortion were usually inpatients (Gissler et al., 1996). Illegal abortions and abortions done outside Finland are believed to be exceptional. In both registers the classifications of the reason for abortion were crude.
The identification of the infertility treatments was based on treatment reimbursements, a method very likely to correctly identify the treated women (Hemminki et al., 2003). In all the registers used, background data for the women were scant and no subjective information was available.
Why did women first terminate a pregnancy and then try to have another pregnancy through an assisted method? According to the few background characteristics available, the infertility treated women with abortions were relatively similar to women treated for infertility without an abortion history, pointing to the importance of the women’s life situation rather than their personal characteristics as such. In most cases the women’s marital status (and this may concern the partner, too) was different at the time of abortion and infertility treatment. Secondly, women’s fertility declines with age and previously fertile women may have become subfertile or infertile over time. The woman’s partner may have changed, and the reason for IVF may have been the current partner’s infertility rather than the woman’s own.
Two practical implications follow from our study: one for health care and the other for health education. First, when a woman is faced with infertility, the knowledge of a past abortion may be an extra emotional strain. Currently this is only speculation, because no research data exist on abortion experience among women seeking infertility treatment. Whether induced abortion increases the risk of secondary infertility is unclear (Thorp et al., 2005). The available literature suggests that at least in rich countries with legal abortions done by skilled health professionals, it is unlikely to be a strong risk factor. Thus, any regret may not be based on self-causation of infertility, but on the missed opportunity.
Secondly, many young women and men may not be aware of the possibility of later infertility. The fact that children cannot always be conceived even though desired, as well as information on the main determinants of such childbearing problems, could be made part of general health education. Whether such information should be given also at the time of considering an induced abortion is more complex. It may be appropriate where abortion is being considered because the pregnancy timing is not quite right. But it may be inappropriate in the case of a (young) woman not able to contemplate forming a family with the child’s father.
In summary, our study showed that women/couples adopt contrasting reproductive strategies at different stages in their lives. The consequences and interrelations of such decisions should be studied more closely, so as to better advise health practitioners and health educators.
|
Supplementary material |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
Supplementary material is available at HUMREP Journal online
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
A grant was provided by the Academy of Finland (grant number 73159) and by our employer STAKES (National Research and Development Centre for Welfare and Health). The funding sources had no involvement in the study design, analysis or interpretation of the data.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary materialFundingReferences |
|---|
Edozien L. Why do some women undergo termination of pregnancy after successful IVF treatment? Hum Reprod (1998) 13:2377–2378.
Gissler M. Induced Abortions in the Nordic Countries. STAKES: Statistical Report 10. (1999) STAKES, Helsinki.
Gissler M, Haukka J. Finnish health and social welfare registers in epidemiological research. Norsk Epidemiologi (2004) 14:113–120.
Gissler M, Ulander V-M, Hemminki E, Rasimus A. Declining induced abortion rate in Finland: data-quality of the Abortion Register. Int J Epidemiol (1996) 25:376–380.
Hemminki E, Klemetti R, Rinta-Paavola M, Martikainen J. Identifying exposures of in vitro fertilization from drug reimbursement files: a case study from Finland. Med Inf Internet Med (2003) 28:279–289.[CrossRef]
Induced Abortion Worldwide. Fact sheet. The Alan Guttmacher Institute (AGI). (1999) http://www.guttmacher.org/pubs/fb_0599.html.
Keskimäki I, Aro S. Accuracy of data on diagnoses, procedures and accidents in the Finnish Hospital Discharge Register. Int J Health Sci (1991) 2:15–21.
Klemetti R, Gissler M, Hemminki E. Equity in the use of IVF in Finland in the late 1990s. Scand J Publ Health (2004) 32:203–209.[CrossRef]
Klemetti R, Gissler M, Sevón T, Koivurova S, Ritvanen A, Hemminki E. Children born after assisted fertilization have an increased rate of major congenital malformation. Fertil Steril (2005) a 84:1300–1307.[CrossRef][Web of Science][Medline]
Klemetti R, Sevón T, Gissler M, Hemminki E. Complications of IVF and ovulation induction. Hum Reprod (2005) b 20:3293–3300.
Knudsen LB, Gissler M, Bender S, Hedberg C, Ollendorff U, Sundström K, Totlandsdal K, Vilhjálmsdottír S. Induced abortion among young women in the Nordic Countries. Acta Obstet Gynecol Scand (2003) 82:257–268.[CrossRef][Web of Science][Medline]
Marcus SF, Brinsden PR. Termination of pregnancy after conception with donor oocytes and donor spermatozoa. Hum Reprod (2000) 15:719–722.
Nyboe Andersen A, Goossens V, Gianaroli L, Felberbaum R, de Mouzon J, Nygren KG. Assisted reproductive technology in Europe, 2002. Results generated from European registers by ESHRE. Hum Reprod (2007) 22:1513–1525.
Sannisto T, Kosunen E, Hemminki E. Contraception and child bearing intentions among older Finnish women. Scand J Sexol (2001) 4:161–171.
Thorp JM, Hartmann KE, Shadigan E. Long-term physical and psychological health consequences of induced abortion: a review of the evidence. Linacre Quart (2005) 72:44–69.
WHO. WHO: Health for all—statistical database. (2006) World Health Organization, Regional Office for Europe, Copenhagen. http://www.who.dk/hfadb (14 December 2006, date last accessed).
Submitted on August 21, 2007; resubmitted on January 25, 2008; accepted on March 5, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||