Accommodating conscientious objection in the midwifery workforce: a ratio-data analysis of midwives, birth and late abortions in 18 European countries in 2016

In the United Kingdom (UK), two senior midwives lost a Supreme Court case which ruled that conscientious objection must be restricted to “hands on” activities [13]. Similar cases affecting midwives have also been reported in other European countries notably Croatia and Sweden, the latter of which does not have a legal position on conscientious objection. One of the midwives who lost her case in Sweden has lodged it with the European Court of Human Rights [14].

The potential workforce issues resulting from midwives who make conscientious objections have never been addressed with a quantitative focus on the associated workload. Having evidence-based information however is essential for a better understanding of and contribution to the debate.

This article therefore presents an analysis of relevant statistics in 18 European countries in relation to potential exposure to late abortions by midwives in order to throw new light on the controversial topic in relation to the midwifery workforce.

The literature is vague as to the extent to which conscientious objection should be permitted. Taking note of the ambiguities, the authors of one article propose that “European countries should critically assess the laws governing conscientious objection and its effects on women’s rights” ([15], p. 231).

This position has been supported by writers who suggest that the various treaties permitting conscientious objection on the grounds of human rights have compromised women’s right to abortions [16]. However, in a case study report on four European countries, the conclusion drawn is that, although complex, it is possible to accommodate individuals who object to providing abortion-related care, while still ensuring that women have access to legal health care services in the countries concerned [17].

The academic commentaries on conscientious objection also are divided. The seminal work of Wicclair provided a comprehensive link between conscience and integrity in medicine, concluding that carte blanche rights of conscientious objection should not be given but rather that respect for moral integrity of the physician, even in practices endorsed by the medical profession, is the best way forward [18]. Claims of conscientious objection thus should derive from the importance attributed to the integrity underpinning them. Weinstock (p. 12) comments that when a health professional’s right to conscientious objection is observed, “respect [is afforded to] the moral agency of those who hold reasonable dissenting views” [19]. In the same vein, Curlin et al. ([20], p. 1891) reflected that “acting conscientiously is the heart of the ethical life” hypothesising that if medical practitioners give this up they no longer have the capacity to make moral judgments or act in accordance with them.

Other writers challenge such positions proposing that the rights of health care professionals to allow their private values should not interfere with their work [21]. Conscientious objection to abortion-related care has also been labelled as “dishonourable disobedience” ([22], p .12). Conversely, Pellegrino avers that a health professional’s conscience or religious values must never be placed in a position secondary to the health service’s requirements [23]. Taking a nuanced approach, Neal [24] suggests that the apparent expansion of conscientious objection claims is based on poorly defined or even contradictory professional guidelines and there is a need for sound research establishing working definitions.

Various professional bodies or regulatory authorities have established such guidelines, the International Federation of Obstetricians and Gynaecologists’ (FIGO) criteria for conscientious objection [25] being to provide notice, refer patients timeously and provide emergency care. While brief and practical, the standards, like those of other such bodies, are not based on research with practitioners. A “White Paper” drawing on international, multidisciplinary literature sums up the issue and develops a road map for the future [26]. The authors point out the lack of well carried out empirical research on the topic but conclude from reviewing the available evidence that there is a growing trend towards refusal to provide certain reproductive health services, especially abortion. Acknowledging the difficulty of the situation, they recommend that a standard definition of conscientious objection be developed together with accompanying obligations.

What is evident from the literature is that its key focus is on medical practitioners, and if midwives are mentioned, it is in a secondary position [9] despite the WHO’s emphasis on them as key providers of abortion services [27]. This has most recently been discussed in two articles, one specifically focusing on abortion [9] and the other on end of life care [28], each of which urged nurses and midwives respectively to be more proactive in contributing to the debate.

With the overwhelming change from surgical to medical abortions in both first and second trimesters of pregnancy, there is an increasing role for midwives and the above arguments are clearly relevant to them. In the past, abortions were carried out surgically exclusively by doctors who may have been assisted by nurses or midwives in an operating theatre. Both in the theatre and pre-and post-operatively, those who expressed a conscientious objection to the procedure were not expected to participate. With medical abortions, the prescription is generally written by a medical practitioner, but the drug is often administered by a nurse, midwife or the woman herself. While in the first trimester in some countries the woman now may labour at home, in the second trimester in most countries, she is cared for throughout the subsequent labour in an inpatient setting by midwives, or occasionally, nurses, thereby increasing their workload.

Until recently, only one article acknowledged this, commenting that many more health professionals are now involved over a much longer period of time [29]. However, in the last 2 years, the topic of midwives or nurses in relation to conscientious objection has come to the fore with two articles emphasising their invisibility [9, 28].

Taking into account the divergence of opinions shown by both academic writers and policy makers and the lack of visibility of midwives, we believe that the whole debate on conscientious objection would benefit from giving consideration to numerical data, not simply considering the numbers of abortions but their relationship to other variables.

We have chosen to focus on midwives in relation to abortions conducted after the first trimester of pregnancy, because it is mainly midwives who are expected to participate in drug administration and the subsequent care for women in labour, including delivery of the foetus and placenta and the provision of immediate postnatal care [27].

This background has shown that a controversial debate around conscientious abortion exists and that both philosophical arguments and laws exist for the provision and the opportunity to conscientiously object to the provision of abortion services. There is, however, a lack of quantitative research showing the dimensions of the topic, and so accurate workload indices cannot be produced. This study seeks to address the issue.

The aim of this study was to provide a picture of the potential exposure midwives in Europe have to abortions conducted after the first trimester of pregnancy (hereafter “late abortions”), an important factor in the integration of accommodation of conscientious objection to abortion by midwives into workload planning. This has been investigated by collecting and analysing data on late abortions, live births and numbers of midwives by European country to provide a ratio of late abortions to midwife.

We selected 32 European countries due to 28 of them being Member States of the European Union (EU) during the period of analysis and the remaining four being signatories to the Shengan accord, which provides for free movement without border controls between them. Of the 28 EU Member States, Croatia, Ireland and the United Kingdom (UK) were not part of the Shengan area.

We decided only to include and analyse robust statistical data provided by Ministries of Health or government statistics departments, with data provided by “third parties” excluded. Only countries that could provide full data sets on all variables were included in the analysis, with countries that had only partial or no data excluded.

A rigorous process of data collection was conducted between January 2018 and March 2019.

Data were collected for the four variables: “number of live births”, “number of abortions”, “number of late abortions” and “number of midwives according to the OECD [30]” for the year 2016, as this year offered the most complete data. This was not a straightforward process and a number of “stages” were undertaken (detailed below) in order to collect data that were considered robust and that met the inclusion criteria.

We identified differences related to the reporting of abortions within the sample. Several countries’ abortion data, most noticeably those in Eastern Europe, contained data on “spontaneous abortions” as well as “induced” and “other” abortions. On inquiring with the MoHs concerned, a comprehensive response was obtained from Latvia which stated “other” meant “abortion with unknown origin or termination of the pregnancy where the factors that lead to this termination are unknown” [personal communication with the Centre for Disease Prevention and Control of Latvia to the authors, 28 September 2018]. Similar confirmations were then received from the other countries. As it was thus clear that these figures were active terminations rather than miscarriages, they have been included in our analysis among the number of abortions. Abortions classified as “spontaneous” conversely have been omitted from our analyses.

All our data were inputted into Excel by one author and double-checked for discrepancies by another author and a statistician against the raw data. (The sources of all data are available on request.) We conducted a ratio data analysis among those countries that met the inclusion criteria. Ratio distributions were calculated between midwives to live births, late abortions to live births and late abortions to number of midwives for each country using Excel. We created box plots using Excel in relation to live births and late abortions/midwife, calculating the median, range and the highest and lowest ratios. We also created a line comparison graph using Excel to show live births and late abortions per midwife by country.

As our study is based on published or freely available statistical data and no data were sought from individuals, no approval from an ethics committee was required.

This study has several strengths and limitations. A major strength is that a robust and comprehensive process of data collection was undertaken, with the data provided by the MoHs throughout Europe enabling us to analyse a considerable amount of data which has not been undertaken previously. However, despite the transparency throughout Europe regarding the reporting of birth rates, the same could not be said for the reporting of abortions. The reporting of abortion figures and especially the reporting of late abortions differed with some countries providing no data on the gestational age when the abortion was conducted. It may be that the actual number of abortions carried out differs from what is reported in the official statistics of some countries with some abortions being reported as “curettage” or similar. It is further accepted that not every midwife included in the numbers in this study will be practising. Of those who are in current practice, some will not encounter women having late abortions as these often take place in specialised foetal medicine units. These limitations also apply to the ratio of midwives to births as again many midwives do not work in labour and delivery units. Thus, we would conclude that our findings concerning the ratio of midwives to late abortions are general in nature rather than context specific. However, this limitation is acknowledged and influences our conclusions. Additionally, we do not know if midwives might object in some cases but not all, and we have been unable to identify a database in which such information could be found. The method of data collection that we used also exhibits limitations, due to potential under-reporting of abortions by some countries. In addition, it might be that some health-care systems offer abortion services in private clinics that do not provide data on governmental databases, and as such, data on late abortions and midwives who work in such settings will not have been included in this study.