Introduction
In 2007 at an expert meeting of clinicians and advocates in London, access to good-quality second trimester services was deemed an essential part of safe abortion services (Comendant & Berer, Reference Comendant and Berer2008). Given the higher morbidity and mortality associated with unsafe abortion later in pregnancy, it may be that significant reductions in complications of unsafe abortion cannot be achieved without addressing the need for safe second trimester abortion services (Harris & Grossman, Reference Harris and Grossman2011). A few studies have documented the complex and multiple reasons why women present for second trimester elective abortion, which include late recognition of pregnancy, difficulty finding a provider, delayed referrals, financial barriers, lack of awareness of abortion law and changing health or relationship status (Drey et al., Reference Drey, Foster, Jackson, Lee, Cardenas and Darney2006; Harries et al., Reference Harries, Orner, Gabriel and Mitchell2007).
The South African Choice on Termination of Pregnancy Act (CTOP) of 1996 replaced the previously restrictive Abortion and Sterilization Act of 1975. The CTOP promotes a woman's right to have an early, safe and legal abortion. As a result of this legislation, abortion-related mortality decreased by 91.1% between 1998 and 2001 (Jewkes & Rees, Reference Jewkes and Rees2005). However, despite the legislation there are still major barriers to women in South Africa accessing abortion services. These include provider opposition, stigma associated with abortion and a dearth of providers trained and/or willing to perform abortions, especially second trimester abortions (Harrison et al., Reference Harrison, Montgomery, Lurie and Wilkinson2000; Varkey, Reference Varkey2000; Jewkes et al., Reference Jewkes, Gumede, Westaway, Dickson, Brown and Rees2005). These barriers further contribute to delays in accessing services, resulting in a high proportion of abortions that are done later in pregnancy (Harries et al., Reference Harries, Orner, Gabriel and Mitchell2007; Grossman et al., Reference Grossman, Constant, Lince, Alblas, Blanchard and Harries2011).
Abortion is a time-restricted health service. The CTOP Act states that a pregnancy may be terminated at a woman's request during the first 12 weeks of gestation. Currently pregnancies of 12 weeks gestation or less can be performed not only by a registered medical practitioner, but also by a registered nurse or midwife who has completed the prescribed abortion training course.
Beyond 12 weeks and up to 20 weeks gestation, an abortion may be performed for any of the following reasons: if after consultation with a pregnant woman, a medical practitioner is of the opinion that the continued pregnancy would pose a risk to the woman's physical or mental health; there is a substantial risk that the fetus would suffer from severe physical or mental abnormality; the pregnancy resulted from rape or incest; or the continued pregnancy would significantly affect the social and economic circumstances of the woman. Abortions in the second trimester (13–20 weeks) can only be performed by a registered medical practitioner (i.e. a medical doctor).
Second trimester abortions account for over 25% of abortions performed in South Africa, which is greater than in other countries with legalized abortion such as the USA and England & Wales, where 12% or less of abortions occur in the second trimester (South African Department of Health, 2005; Department of Health Statistical Bulletin England and Wales, 2007; Pazol et al., Reference Pazol, Gamble, Parker, Cook, Zane and Hamdan2009). Although safe abortion has a very low risk of complications overall, abortions performed after 12 weeks of gestation pose greater risks of medical complications than abortions performed during the first trimester (Bartlett et al., Reference Bartlett, Berg, Shulman, Zane, Green, Whitebread and Atrash2004; WHO, 2004).
A dearth of trained second trimester abortion providers in South Africa, and more specifically in the Western Cape, has resulted in the outsourcing of surgical second trimester services to a small group of doctors employed by the Provincial Department of Health to provide abortion services in public sector facilities. Currently there are only three surgical second trimester abortion providers in the Western Cape public sector and it is the only province in South Africa that offers dilation and evacuation (D&E) in public health care facilities.
Due to the relatively recent significant changes to abortion legislation in South Africa, little is known formally about the personal and professional attitudes and experiences of individuals who are currently working in abortion service provision, including second trimester abortion provision (Alblas, Reference Alblas2008; Turner et al., Reference Turner, Hyman and Gabriel2008; Dearham et al., 2009). Furthermore, no current comprehensive data are available on how second trimester abortion services are organized in South Africa despite these high numbers (Alblas, Reference Alblas2008). Investigating the complex factors that determine health care providers' involvement or disengagement in second trimester services could potentially provide important insights that could inform not only South African policy and service provision, but would also be valuable internationally, especially in developing country contexts.
This study set out to explore health service providers' and health managers' perceptions and experiences of second trimester abortion services in the Western Cape Province, South Africa, with a special emphasis on service quality and capacity and the potential for service improvement. This qualitative study formed part of a larger cross-sectional study that examined women's experiences seeking and undergoing second trimester abortion, as well as the efficacy and safety of two second trimester abortion techniques: medical induction with misoprostol alone and dilation and evacuation (D&E) (Grossman et al., Reference Grossman, Constant, Lince, Alblas, Blanchard and Harries2011).
Methods
Study setting
The study was conducted between July and October 2008 at five public sector hospitals providing second trimester abortions. Research sites were based within the greater Cape Town area and two outlying peri-urban areas within the Western Cape Province, South Africa. These sites were five of the nine public hospitals in the province that provided second trimester abortion services. The remaining four hospitals providing second trimester abortions were not included in the study due to logistical reasons, infrequent provision of services and low patient volume.
The D&E method was performed at three of the five hospitals by three private doctors who were employed by the Provincial Department of Health to provide abortions in designated facilities where there were no public sector providers willing to provide abortions. Medical induction with misoprostol alone was used in two of the study hospitals. In other provinces within the public sector, medical induction using misoprostol alone is the preferred method for second trimester abortions, if later abortions are performed at all.
Study population
A total of nineteen qualitative in-depth interviews were conducted with health care providers and hospital managers who were involved in a range of aspects of second trimester abortion provision at the facilities sampled. Respondents were selected through purposive sampling to include all health care providers (nurses and doctors) and health managers who were involved in second trimester abortion services at all five study sites. Participants included two obstetrician–gynaecologists who were heads of the Obstetrics and Gynaecology Departments at two of the hospitals, three doctors directly involved in abortion provision and nine nurses involved in pre- and post-abortion care. Five senior hospital managers not directly involved in abortion services were included as they played a key role in abortion service provision and oversight. The small sample size is reflective of the shortage of second trimester abortion providers, whether directly involved in the abortion procedure or involved in other aspects of the abortion service.
Interview guides were semi-structured and open-ended. Key issues explored included: second trimester work experiences and perceptions of second trimester abortion services including second trimester abortion methods and impact of gestational age limits on abortion practice, perceptions of women who seek second trimester abortions and barriers to changing or improving second trimester services. Data collectors trained in qualitative methods conducted the interviews in English, except for one interview, which was conducted partially in Afrikaans. Interviews lasted approximately one hour and were held in a private setting.
Interviews were digitally recorded and transcribed verbatim. All participants provided written informed consent, and confidentiality and anonymity were ensured. Ethical approval was obtained from the Research Ethics Committee, University of Cape Town and University of Stellenbosch and the Allendale Institutional Review Board, USA. Approval to conduct the study in the five hospitals was obtained from the Western Cape Provincial Department of Health.
Data analysis
Data collection and analysis were inter-related, and an iterative process was used to allow questions to be refined and new avenues of inquiry to develop. Data were analysed using a thematic analysis approach, in which main themes and categories were identified and analysed within and across data (Boyatzis, Reference Boyatzis1998; Braun & Clarke, Reference Braun and Clarke2006).
The computer software package ATLAS ti 5.2 was used to facilitate sorting and data management (Scientific Software Developments, Berlin, Germany). Members of the research team developed and refined the codes using the key issues explored. Transcripts were reviewed by members of the research team, and a preliminary list of codes and code definitions was developed and subsequently refined through discussion. Data were collaboratively coded and reviewed; major trends and cross-cutting themes were identified. Inter-coder reliability was assessed among members of the research team and found to be consistently high.
Results
The majority of respondents were female (78.9%) and the median number of years working in abortion services was six (range 1–13). Out of the nineteen providers interviewed, six doctors (32%) and two nurses (11%) had undergone formal abortion training. Doctors who were performing abortion using the medical induction method had not undergone formal training but were taught on site.
Respondents expressed a range of views that illustrated the difficulties and often the reluctance on the part of providers and managers to participate in second trimester services. Predominant themes included: gestational age as an important factor influencing decisions around abortion provision; second trimester abortion methods and their differing impact on providers; health service-related barriers; perceptions of why women seek later abortions and suggestions for improving services.
Gestational age
Gestational age was a key indicator of acceptability, and providers mentioned that they found it more traumatic to deal with a termination performed around 17–20 weeks than a termination at 12 weeks, because with the latter one was dealing with a ‘blood clot’ rather than a ‘formed fetus’. Providers drew distinctions between early and later fetal development, ascribing more personhood and resemblance to an infant as the pregnancy progressed.
A nurse provider described the physiological distinctions between early and later fetal development, whereby the abortion process was made more real and hence more troubling with increasing gestational age:
With first trimesters it is not so difficult or real, as the terminated pregnancy is more like a small blood clot or tissue but now with second trimesters, there are all these fetal parts … they have a human shape, something that we recognize and that suddenly makes it an awful lot more real.
Providers actively sought ways to remove themselves from the abortion process, especially for procedures performed later in pregnancy. For example, some nurses restricted their practice to less invasive measures, such as preparing surgical trays, monitoring the patients' vital signs, administering medications, and providing pre- and post-abortion counselling. A senior hospital manager who was responsible for second trimester services at a district level explored the difficulties associated with second trimester abortion provision in his health district:
The nursing staff have a problem physically being in the theatre and having to sort out products afterwards, because they don't like it, so we try just to get them to talk to the patient and other routine activities such as preparing surgical instruments and the doctor does the rest. I think the moment you start doing things that they're not happy with, then the whole thing might collapse and we will have no service.
Because physicians must perform (or prescribe the medications for) abortions after 12 weeks gestation, respondents also highlighted the challenges of recruiting doctors to work in the abortion services – especially those that provide D&E.
It's always difficult to get medical officers to work with us in the first place and be involved in the whole process from beginning to end. We don't want to alienate them and then they don't apply for the jobs. So we try and work around it … if you do counselling, someone else writes the prescription, the patient takes the medication, she aborts and then you just do evacuation afterwards, and people are happy with that, but physically removing a live baby with the suction curettage or dilatation and evacuation, they're not prepared to get involved in this.?
Abortion methods
Providers' perceptions of the two different second trimester abortion methods – namely D&E, a surgical procedure, and medical induction using misoprostol alone – were explored to assess whether providers felt differently about the two methods, and whether it would impact on their decisions around abortion provision. Overall providers were more uncomfortable with the D&E procedure, as it elicited more physical and emotional responses to the abortion process.
The following vignette from a doctor who coordinated the hospital's medical induction abortion service highlights the complex issues of providing second trimester abortion services and the difficulties in finding staff to assist with the procedure. Opposition to abortion, underscored by exposure to the aborted fetus, was particularly distasteful for the staff and distinguished D&E from first trimester abortion and from medical induction:
They are currently busy building an extra theatre, so hopefully within the next year or two, we'll be able to [offer D&E], if we can find the staff that will be able to, or that are willing to help because most of the staff don't want to get involved with the D&E procedure. They're fine with the medical induction and with the first trimester MVA procedures, but the D&E procedure they don't want to get involved with. I think the reason is because they see the fetus and they see the products and they don't want to get involved with that because they can see what they are doing.
Similar sentiments were echoed by a doctor at a tertiary hospital providing medical abortion, who intimated that medical staff were not prepared to provide D&E services, even though they recognized their current difficulties with respect to shortages of bed space and duration of hospitalization associated with medical induction. At this particular hospital only six hospital beds were allocated a week for second trimester abortions and as women often took more than 24 hours to abort they required longer hospital stays. He spoke candidly about his discomfort with the D&E procedure:
I don't think that D&Es are appropriate for this facility as we don't want to have the wrong aura and be associated with D&Es … as we do not want to get a bad reputation.
Consequently reluctance towards second trimester abortion provision resulted in the services being supported and driven by a small cadre of providers dedicated to providing D&E and predominantly outsourced to the private sector as previously discussed.
Health service barriers
Health service barriers to second trimester abortion provision related to barriers within the health care system and were also related to women seeking abortions.
Infrastructure barriers
Providers alluded to numerous difficulties associated with the provision of second trimester abortion services. These included problems with an increasing demand for services and a concomitant lack of infrastructure, physical space and health personnel to respond to these demands. Staffing the service seemed to be the biggest challenge at all of the facilities, with a roving team of doctors providing the service, as previously discussed. As a result, they acknowledged that levels of service provision were often fragmented and quality of care was compromised. A doctor illustrated the extent to which medical care was compromised while paradoxically stating that they were doing a ‘fantastic job’; the general sentiment was that in the face of high demand, providers rendered the service with little consideration for overall quality of care:
I think we're doing a fantastic job, but unfortunately, it's a bit of Wild West every now and then. I think there are far too many patients,… and they're all sitting in one room and the Sister that's doing the first trimesters can't look after the second trimesters … there's never blood pressure taken … it's terrible and there's of course no privacy, there's nothing, it's awful. It can improve a lot.
Despite the recognition of compromised care, providers maintained that women were grateful and relieved to be able to obtain an abortion regardless of the type of care received.
Stigma
The contentious nature of abortion provision resulted in many providers being stigmatized in their workplace. However, stigma was often heightened with second trimester abortion services compared with first trimester services. A nurse provider commented on her ‘exaggerated experience’ of stigma in relation to second trimester abortion provision:
Now with second trimester abortions it becomes worse and other professionals call you ‘murderers and baby killers’ and you get known as someone who is involved in that type of abortion. Somehow it is just much more exaggerated with second trimester abortions and somehow first trimesters are not such a big issue.
Related to this abortion services were often located in hidden or difficult to access places where providers who worked there were viewed as performing the ‘devil's work’, as one nursing manager recounted. This was illustrated by a nursing manager's comments, referring to lack of signage indicating where second trimester abortion services were located:
I would have liked the second trimester services more in the open. I actually asked them also for the signage … I wanted a sign that says, ‘This way is TOP’ but that wasn't really met very well, so as a result we still don't have a sign.
Perceptions of why women delay seeking an abortion
Providers were asked why they thought women accessed abortion services in their second trimester, so as to explore how this influenced attitudes towards abortion provision. Some providers recognized health systems limitations, which in turn contributed to delays, whereas others held women responsible for late presentation, largely attributed to poor contraceptive uptake and usage. Perceptions of why women sought later abortions were informed by underlying negative attitudes towards abortion in general, and were heightened with second trimester abortion services.
Health systems limitations
Health systems limitations mentioned by participants included cumbersome booking systems, too few facilities with limited capacity to accommodate large patient numbers, and a shortage of willing, trained providers, which resulted in restrictions on the number of second trimester abortions performed per week in designated second trimester abortion facilities. These delays meant that some women presented at a facility in their first trimester, but due to the facility's limited capacity, they were not able to obtain an abortion until the second trimester.
Contraceptive uptake and usage
The notion that women were using abortion as a contraceptive method appeared to be a major concern for second trimester providers, underscored by perceptions that abortion was often a substitute for responsible family planning. Furthermore, a woman who came back to the services for a second or third time was identified as coming for a ‘repeat abortion’ and seeking an abortion more than once was indicative of using abortion as a contraceptive method and was key to providers' frustrations towards women seeking later abortions.
By way of illustration, a senior medical superintendent expressed an underlying discomfort with abortion, particularly second trimester procedures, suggesting an aggressive promotion of family planning, including more permanent methods such as female sterilization, as a feasible alternative.
My other concern with the second trimester TOPs is that I think a lot of them are repeat people coming in … and it's a problem, I think that's not my idea of the TOP, it's not a family planning method, and it's something else. So I feel that there should be aggressive programmes to look at family planning and other sterilizations and that type of thing, rather than just going for a TOP. I'm not comfortable with second trimester TOPs.
Other providers suggested that there should be more focus on broader reproductive health services, especially contraceptive services, rather than expanding abortion services as improving family planning services was preferable to abortion and ultimately ‘women needed to take more responsibility in using contraception’.
Beyond the abortion service, respondents said it was critical to reduce demand for later abortion and abortion generally by improving family planning services. However, participants perceived multiple barriers to improving family planning services in the public sector, including limited contraceptive choice, overemphasis on condom promotion, little or no pre- and post-abortion counselling, and problems with access to family planning services such as restricted clinic opening times and contraceptive services not always being available at abortion sites.
Difficulties in accessing contraceptives were further illustrated by a nursing manager who indicated that contraceptive methods were not available at one of the sites where abortions were performed:
Where I do get complaints from the Sister, is that because the clinic [where abortions performed] is here,… but the family planning is 200 metres away from there,… and sometimes they finish late and then the family planning clinic is closed, now she's stayed out of work today, so now she must stay out of work tomorrow again, or maybe 2, 3 days after that – that's ridiculous, so with the result they don't come for the family planning.
Suggestions for improving services
A few respondents thought there was little that could be done to improve the services, explaining that they were doing the best they could with limited resources. Most, however, said that there were specific improvements that could be made. Some suggestions related to how services were organized, with providers mentioning the possibility of outsourcing services to the private or NGO sector to deal with the large demand and limited capacity of the public services. It was noted, however, that this was already happening at many designated abortion facilities. For example, in some regions of the province, the Department of Health has contracted with Marie Stopes South Africa to provide second trimester services (Alblas, Reference Alblas2008).
Opportunities to attend abortion and values clarification training workshops were also cited as a critical component of service improvement, but training was often sporadic or not possible due to staff shortages.
Providers involved with medical induction services recognized that the capacity of services was limited due to the lengthier abortion process and a small number of allotted inpatient beds, and that more clients could be served by offering D&E. However, many acknowledged the difficulties in recruiting doctors and nurses willing to participate in D&E services. In the medical induction service, doctors were generally aware of the superior efficacy of the combined mifepristone–misoprostol regimen compared with misoprostol used alone (Kapp et al., Reference Kapp, Borgatta, Stubblefield, Vragovic and Moreno2007) – the standard at these hospitals – but they were concerned about the cost of mifepristone. However, several pointed out that using mifepristone might be cost-effective if it reduced the duration of hospitalization. Mifepristone is registered in South Africa only for use in the first trimester, and therefore it cannot be used routinely in the second trimester at public hospitals.
Discussion and recommendations
Although second trimester abortion in South Africa must be provided by doctors, nurses play a critical role in service delivery, assisting at all levels for both procedure types and playing an even greater role with the medical induction procedure. Shortages of trained doctors and nurses contribute to poor service quality and limited access to services and have implications for long-term sustainability of second trimester abortion services. Training in abortion is not a required component of nursing or medical education in South Africa, limiting the number of doctors and nurses with the skills necessary to provide abortion care. In addition, in many facilities, the personal views of doctors and nurses contribute to the shortage of abortion providers.
There is a need to build support for abortion services among hospital staff generally, including those who are not directly involved in the service. This would not only improve the pool of willing providers, but also improve the working conditions for existing providers who report feeling both stigmatized and isolated.
Values clarification and other training that is designed to convey the abortion client's perspective could improve health care providers' perceptions of the women who access abortion services and is a strategy that has been used for garnering support for, or helping, health care providers shift or reflect on their attitudes towards abortion and has been successful in South Africa (Mitchell et al., Reference Mitchell, Trueman, Gabriel, Fine and Manentsa2005; Turner et al., Reference Turner, Hyman and Gabriel2008).
Values clarification and client-centred training might also improve providers' likeliness to provide pain management and emotional support to clients. In the cross-sectional study that was conducted in parallel with this study, around 50% of clients undergoing both abortion methods reported high or extreme pain, and 28% reported high or extreme emotional discomfort. Only about 20% of clients received any kind of pain medication (Grossman et al., Reference Grossman, Constant, Lince, Alblas, Blanchard and Harries2011). The low use of pain medication is particularly concerning given the findings of this study indicating that some providers feel that the abortion experience must be unpleasant in order to deter women from seeking a repeat abortion.
A concern that contraceptive use was being replaced by abortion was underscored by perceptions that abortion was often a substitute for responsible contraceptive practices, although participants also highlighted significant shortcomings in public sector contraceptive services. The potential for improvements in contraceptive service provision to positively impact various aspects of health and welfare – both from society's and the health system's perspective – cannot be overstated. It is particularly important that contraceptive counselling and method provision in the context of abortion care be strengthened. However, it is critical to recognize that even with complete family planning coverage, abortion, including second trimester abortion, will always be a necessary health care service.
Comments about repeat abortions seemed to indicate respondents' disproportionate concern with this phenomenon. Although women may under-report previous abortions, in the larger cross-sectional study only 4% of over 300 women participants reported having had a previous abortion, suggesting that providers thought that a much higher proportion of clients had had a previous abortion than was actually the case (Grossman et al., Reference Grossman, Constant, Lince, Alblas, Blanchard and Harries2011).
When commenting on ways to potentially expand the overall capacity of second trimester services in the Western Cape Province, there were mixed feelings. Although hospital staff recognized that D&E services might better meet the large demand for second trimester abortion than current medical induction services, there was reluctance by respondents at medical induction sites to consider introducing a D&E service. They indicated that although a D&E service might be easier for women, it would probably be more difficult for providers who experience a physical and emotional response to removing fetal parts. This perception of D&E services as difficult for the provider has been reported previously in other contexts (Kaltreider et al., Reference Kaltreider, Goldsmith and Margolis1979; Harris, Reference Harris2008). Introducing the mifepristone regimen at public sector facilities that do not have trained and willing D&E providers might significantly improve the capacity of these services to meet the needs of women seeking second trimester abortion, especially if a low-cost mifepristone product becomes available in South Africa. It might also better meet the needs of women by offering an acceptable abortion method that can largely be completed as a day procedure (Ashok et al., Reference Ashok, Templeton, Wagaarachchi and Flett2004). It remains to be seen if a pharmaceutical company will attempt to register mifepristone for use in the second trimester in South Africa or if the Department of Health will allow more widespread off-label use of this regimen.
This study has several limitations. The research was conducted in one province in South Africa, and the findings may not be generalizeable to other parts of the country or to other countries. The small number of participants interviewed is also a limitation, although as noted above this relates to the limited number of providers involved in second trimester abortion care. In addition, it was felt that saturation of themes was achieved.
Little attention has been paid to the everyday experiences of abortion providers and the often disquieting aspects of second trimester abortion provision (Harris, Reference Harris2008). Whilst providers were open about their experiences, opportunities for support and a safe space in which to engage with these issues were largely absent. Exploring ways of dealing with these difficulties such as providing ongoing team support could strengthen and support service provision. Debriefing sessions or other non-threatening and supportive spaces could contribute to preventing possible burn-out and fatigue among providers in South Africa.
Ongoing health service-based support and recognition is crucial, as currently most abortion services are provided by an older group of committed abortion providers, which has implications for long-term sustainability of the services. The chronic shortage of abortion providers in South Africa has resonance elsewhere, where concerns have been raised about the ‘thinning ranks’ of abortion care providers, with few younger abortion providers replacing a larger, older cohort (Grimes, Reference Grimes1992; Lazarus, Reference Lazarus1997). By mainstreaming abortion care, including later procedures, into standard physician and nursing training and creating a supportive environment, it may be possible to encourage younger practitioners to play a more active role in these services.
The CTOP Act was a ground-breaking piece of legislation for women's rights and health, and for overall public health in South Africa. However, the momentum for realizing the full extent of the legislation regarding service implementation needs to be maintained. In order for this to continue there is a pressing need to address the provider shortage, and abortion education and training needs to be formalized and initiated in medical and nursing schools. This should include ongoing training and support for those health care providers who become involved in abortion provision and care.
Acknowledgments
This work was funded by the Safe Abortion Action Fund. The authors would like to acknowledge contributions by field researchers, Dr Marijke Alblas and all study participants.
