This article summarises the findings of studies relating to why women present for abortion at gestations of more than 12 weeks. Its primary focus is on British experience, but relevant studies from other countries are described. Key findings reveal that there are many different reasons. Much of the delay occurs prior to women requesting an abortion; other key issues include women’s concerns about what is involved in having the abortion and aspects of relationships with their partners and/or parents. Further, after requesting an abortion, delays are partly ‘service-related’ – for example, waiting for appointments – and partly ‘woman-related’ for example, missing or cancelling appointments. The relative contributions to the delay of these various factors are discussed. The implications of the research for abortion education and service provision are considered. Abortion for reasons linked to foetal abnormality is not covered in this article.
This article presents the findings of the relatively low number of studies that have considered why women present for termination of pregnancy procedures at a gestation of more than 12 weeks. Much of the discussion summarises the findings of the largest and most recent British study, which was conducted by the authors in 2006 and 2007. Previously published work from Britain and relevant studies from other countries are also discussed where appropriate. Finally, we consider the implications of this research for abortion education and service provision. Abortion for reasons relating to foetal abnormality is not covered in this article.
The discussion of second trimester abortion needs to be set in the context of its incidence. In Great Britain, whilst the overall abortion rate has been gradually increasing since the new legal framework took effect in 1968, and especially over the past decade, there have also been changes in the temporal patterning of the procedures. Table 1 shows that the large majority of procedures are provided in the first trimester and, further, that there has been a general shift towards earlier procedures in recent years.
| Year | Overall rates a | Percentages gestation length (in weeks) b | |||
|---|---|---|---|---|---|
| <10 | 10–12 | 13–19 | 20 + | ||
| 1985 | 12.5 | – | – | 12 | 2 |
| 1990 | 15.5 | – | – | 11 | 1 |
| 1995 | 14.5 | 56 | 34 | 10 | 1 |
| 2000 | 17.0 | 58 | 30 | 10 | 1 |
| 2005 | 17.8 | 67 | 23 | 9 | 1 |
a age-standardised rates per 1000 women aged between 15 and 44.
b data for some gestation lengths prior to 1995 are not available due to a change in the way data were presented.
The overall proportion of abortions carried out under 10 weeks has increased from 56% in 1995 to 67% in 2005. However, these developments have been accompanied by only a slight reduction in the demand for abortions in the second trimester, with the proportions at 13 or more weeks declining from 14% in 1985, 11% during 1995–2000 and 10% in 2005.
Second trimester abortions are less desirable from a public health policy perspective; policy in Great Britain is that abortion should be provided to women as early as possible. This is predicated on the recognition that procedures performed at early gestations are relatively safer medically than those performed at more advanced gestational stages. Other factors also exert influence; for example, financial costs increase with gestation and, while discussion of this area is beyond the scope of this article, the perception that moral objections to abortion are more powerful at later gestations may have influence. During the recent public debate in Britain, those who object to the provision of abortion have made so-called ‘late abortion’ a major focus of their arguments. Early abortion may therefore appear to some to be less ‘objectionable’. Stressing that the abortion service aims to provide early abortion where possible may be attractive to those wishing to avoid confrontation about the ethics of abortion.
Policy on abortion provision has developed in important ways. The Department of Health has increasingly ‘mainstreamed’ abortion services as part of the broader framework for provision of sexual and reproductive health services. National disparities in the proportions of abortions funded and provided by the National Health Service (NHS) have also been recognised as a concern. In addition, policy has sought to address problems of access to abortion by using the capacity of specialist independent service providers. The overall proportion of procedures carried out by such agencies has increased from 45% to 60% in the past 10 years, with a shift in the proportion of these abortions being carried out under contract to the NHS from around one-third to three-quarters over the same time period.
Second trimester abortion features in public health policy. Appropriate medical regimens are set out by the Royal College of Obstetricians and Gynaecologists (RCOG) of United Kingdom, and these, as well as the Department of Health, have stated that Primary Care Trusts (PCTs) should have systems in place to ensure that ‘no’ woman seeking abortion and meeting relevant legal criteria should have to wait more than 3 weeks from first referral to abortion procedure. The Medical Foundation for AIDS and Sexual Health Services (MedFASH) has stated that commissioners of services and service providers should ensure that abortion is available locally up to the maximum legal time limit. The Social Exclusion Unit’s report on teenage pregnancy noted the relatively higher proportion of abortions amongst teenagers that occur during the second trimester than is the case for women over 20 years of age.
The provision of abortion in the late second trimester, specifically, has become a policy concern. In response to the media debate about ‘late abortion’, the Chief Medical Officer (CMO) stated in 2005 that women are legally entitled to seek abortion on the basis of the terms of the law, under which abortion can be provided up to 24 weeks gestation, and that PCTs should ensure that services are available. The CMO made a number of important recommendations for the late abortion service (abortion at 20–23 completed weeks) which covered, amongst other issues, the development of a best practice protocol, commissioning of a review, staff training needs and improved identification of sources of delay. At the time of writing in Summer 2009, the Department of Health has indicated that the CMO’s recommendations will be included in good practice commissioning guidance for contraception and abortion services, and that streamlining of referral for late abortion (as yet undefined) will be pursued by establishing a central telephone booking service.
Attention has also been paid to encourage innovation in regard to increasing the proportion of early medical abortions (EMAs). Policy emphasises the importance of women accessing abortion at 9 weeks’ gestation or earlier whenever possible, because they can then have the choice of medical or surgical abortion, with the former avoiding the need for anaesthesia and surgical intervention. Data presentation changed in the mid-1990s but, in 1995, 10% of under 9-week abortions used EMA (this represented 4% of all abortions), in 2002, 18% used EMA (10% of all abortions) and, in 2005, 30% used EMA (20% of all abortions). PCTs defined by the Healthcare Commission as providing a ‘good’ service are those where 70% of abortions are performed by 9 completed weeks’ gestation. In 2005–06, funding was allocated to assist PCTs to increase the proportion of EMAs.
In sum, although policy enables abortion to be provided throughout the second trimester, emphasis is placed on taking measures to increase the proportion of abortions that occur at earlier stages of gestation. In this light, questions to consider are why do women present for abortion in the second trimester? Also, what, if anything, can be done to reduce the proportion of abortions provided after 12 weeks? In the remainder of this article, we outline some of the answers that research has provided to these questions.
British research findings prior to 2006
One of the issues that has been explored in the limited research hitherto is the extent to which delays created by services explain late procedures. One view is that some women present at an early point in gestation, but are delayed either at the point of referral or because the procedure is not provided soon enough after referral. Finnie et al. found that, for women attending hospital abortion clinics in the South Durham area, 44% were treated within 3 weeks from the point of referral, pointing to a large disparity with national standards. A survey of all English PCTs by the All Party Parliamentary Pro Choice Group found that more than one-quarter had waiting times of more than 3 weeks.
Access to second trimester abortion procedures has also been considered. The attitudes of NHS gynaecologists practising in Britain to providing abortion procedures at later gestations were assessed by Francome and Savage in a now somewhat dated study. They found that 89% stated their own personal upper limit for performing abortions was 20 weeks, although there was ‘considerable theoretical support for later abortions’, suggesting ‘there is a disparity between the fact that many were unwilling to perform an abortion after 20 weeks, except for malformation (of the foetus) or to save a woman’s life, and yet would support the legal availability of later abortions for wider implications if someone else were willing to perform them’. p154 Roe et al. considered attitudes and training of trainee doctors working as obstetrician–gynaecologists in the NHS. Whilst large majorities said they were prepared to do some work associated with abortion (e.g., assess or clerk patients), 28% of junior doctors reported being unwilling to perform first trimester abortions and 38% as being unwilling to perform second trimester procedures (although 88% said they would participate in abortion for foetal abnormality).
More recently, a study of under-18 pregnancies ending in abortion included a survey that collected the perceptions of, and information from, service commissioners, NHS hospital consultants in obstetrics and gynaecology and teenage pregnancy co-ordinators. The quality of the abortion service was generally perceived as being notably poorer at later gestational stages in almost all areas. There was a commonly reported problem of poor access to abortion after the first trimester. Despite the legal upper time limit of abortion of 23 completed weeks, the highest reported cut-off point was, in practice, 20 weeks. In most sites, it was reported that abortions are not provided in NHS hospitals from an earlier stage – between 13 and 16 weeks in most cases. Comments made by respondents indicated that, commonly, a ‘division of labour’ existed between NHS providers and the independent sector with the latter providing later terminations, albeit often funded by the NHS. It was noted that this often required women to travel some distance to access the procedure. In some sites, it also appeared that NHS units did not provide these procedures because of dislike amongst staff of involvement with abortion provision. These findings help to explain the very marked shift in provision from the NHS to the independent sector in recent years.
Other research, however, challenges the view that late procedures are ‘primarily’ a result of service factors (and thus amenable to improvement through better access). The relative contribution to late procedures of service-related delays, and delays for other reasons, was considered in a study by George and Randall published in the mid-1990s. It found that only 13% of second trimester abortions could have been managed earlier by service improvements, since most women requesting later abortions did not ‘seek’ abortion until a relatively late gestational stage. The study reported on findings of a retrospective analysis of records of all 111 women who had an appointment during the first year of a second trimester Unplanned Pregnancy Counselling Clinic (UPCC). The reasons women gave for late presentation were grouped into ‘unpreventable’ or ‘preventable’. Ninety women received counselling, of whom 71 had reasons recorded for late presentation. Twelve potentially preventable late presentations were found. Unpreventable reasons for late presentations for abortion were varied, and included concealed teenage pregnancies, peri-menopausal women or women with irregular menstrual cycles who did not associate amenorrhoea with pregnancy, as well as pregnancies that had been initially wanted.
More recently, abortion at 19–24 weeks gestation was investigated through a study of clients attending Marie Stopes International (MSI) clinics. Twenty-six women were interviewed face to face and 84 completed questionnaires. Although service factors emerged as part of the explanation for late presentation, it was found that for ‘most’ women ‘a combination of factors’ led to late abortion. These included significant delays in accessing services, failure to realise they were pregnant until relatively late, not recognising the ‘signs and symptoms’ of pregnancy until ‘an advanced stage’ and, for a small number who were aware they were pregnant at an early stage, denial. A few reported a change in their life circumstances since becoming pregnant.
In the light of the policy context, administrative data and insights of the limited research discussed above, a recent study by the present authors sought to find out more about the reasons for the incidence of second trimester abortion in England and Wales. The key results are summarised.
Second trimester abortions in England and Wales
The key concepts used in this study were ‘pathway to abortion’ and ‘delay’. Five stages on the pathway to abortion were identified, and the study sought to find out about experiences of delay at these stages. These stages were
- 1.
time to suspecting pregnancy;
- 2.
time between suspecting and taking test;
- 3.
time between test result and decision;
- 4.
time between decision and requesting abortion; and
- 5.
time between requesting abortion and procedure.
The methodology was a self-completion questionnaire, with a range of options – including 39 specific potential reasons for delay – from which respondents could select those that they felt had applied. The options were generated from a combination of close reading of the literature discussed above and pilot work involving detailed interviews with staff working for abortion providers. Questions also included space for further open comments by respondents.
The sample was 883 women recruited at eight British Pregnancy Advisory Service (BPAS) clinics and two additional independent sector clinics. The selected clinics carried out around 41% of the almost 20 000 second trimester abortions to England and Wales residents in 2005. The sample composition was compared against national data and, where necessary, data were weighted to take any disparities into account. The main findings were as follows ( Table 2 ).
| Reason | Percentage |
|---|---|
| I was not sure about having the abortion, and it took me a while to make my mind up and ask for one | 41 |
| I didn’t realise I was pregnant earlier because my periods are irregular | 38 |
| I thought the pregnancy was much less advanced than it was when I asked for the abortion | 36 |
| I wasn’t sure what I would do if I were pregnant | 32 |
| I didn’t realise I was pregnant earlier because I was using contraception | 31 |
| I suspected I was pregnant but I didn’t do anything about it until the weeks had gone by | 30 |
| I was worried how my parent(s) would react | 26 |
| I had to wait more than 5 days before I could get a consultation appointment to get the go-ahead for the abortion a | 24 |
| My relationship with my partner broke down/changed | 23 |
| I was worried about what was involved in having an abortion so it took me a while to ask for one | 22 |
| I didn’t realise I was pregnant earlier because I continued having periods | 20 |
| I had to wait more than 7 days between the consultation and the appointment for the abortion a | 20 |
| I had to wait over 48 hours for an appointment at my/a doctor’s surgery to ask for an abortion | 20 |
General reasons for second trimester abortion
Despite our own data being concerned with abortions occurring at 13 or more weeks’ gestation, the results closely mirrored those obtained in the smaller MSI study (mentioned above) on much later procedures. Multiple factors explained late procedures, with the vast majority (85%) reporting more than one reason for delay.
The five stages of the pathway and associated delays
The specific reasons reported were categorised into five stages in the pathway as noted previously. The proportions of women who reported at least one reason for delay during each of these five stages were
- §
suspecting pregnancy (71%);
- §
between suspecting and confirming pregnancy with a test (64%);
- §
between confirmation of result and deciding to have an abortion (79%);
- §
between deciding and first asking for an abortion (28%); and
- §
between asking and obtaining an abortion (60%).
In terms of time, women’s reports indicated that much of the delay occurred ‘prior to’ requesting an abortion. Fifty percent of the women were at 13 or more weeks’ gestation by the time they ‘first asked’ for an abortion.
Delay in suspecting pregnancy
Of the overall sample, 71% of respondents reported at least one reason for delay within this category; lack of early awareness of pregnancy is thus an important factor. Half of the sample were at over 52 days’ gestation when they first suspected they were pregnant, with one-quarter being over 79 days’ gestation. The most common factors reported by those who were above this group’s median gestation (52 days) at suspecting pregnancy were
- §
because my periods are irregular (49%);
- §
because I continued having periods (42%); and/or
- §
because I was using contraception (29%).
Delay in taking pregnancy test
Sixty-four percent of respondents overall selected at least one reason for delay between suspecting pregnancy and confirming it with a test. The median reported time between suspecting and taking a test was 14 days, with the most commonly reported reason for this (reported by 45% of women) being that, while they suspected they were pregnant, they ‘did not do anything about it until the weeks had gone by’. Other responses reported by fairly large minorities were that they were ‘not sure about what they would do if they were pregnant’ and/or fears over the reaction of their parents and/or partners.
Seventy-one percent of women confirmed their pregnancy at home; these women were less likely to wait over 14 days to do this – as opposed to those who were tested at a doctor’s surgery or clinic. Part of the delay amongst some of the latter group was due to their not wanting to see their regular doctor and having to spend time finding somewhere else to go for a pregnancy test.
Delay in deciding to have abortion
Seventy-nine percent of respondents reported a delay in deciding to have an abortion, with around half the respondents taking 1 week or more between their test result and the decision to have an abortion. Among respondents who took over the median time of 1 week, the reasons cited by more than 10% of them are shown in Table 3 .
