Attitudes towards the regulation and provision of abortion among healthcare professionals in Britain; cross sectional survey data from the SACHA study

Participants and procedures

We carried out a survey of knowledge, attitudes and experience of abortion provision between November 2021 and August 2022 in England, Scotland and Wales¹⁹. Healthcare professionals including nurses, midwives, doctors and pharmacists in a range of settings were eligible for inclusion. We used a stratified cluster sampling strategy, selecting a random sample of services as ‘clusters’. Service types were drawn from separate sampling frames. A list of general practices was compiled from data from the Care Quality Commission in England (CQC); Health Inspectorate Wales; and NHS Inform Scotland. A list of NHS and independent sector abortion services was compiled from abortion statistics in England and Wales reported to the Chief Medical Officer in 2020 and clinic lists available from the independent sector providers; and from communication with those involved in abortion provision in Scotland. Registered pharmacies in England, Scotland and Wales were identified via the General Pharmaceutical Council. For sexual and reproductive health, and midwifery services, our sampling frame consisted of a complete list of all six-digit postcodes in England, Wales, and Scotland. Randomly selected postcodes were entered into the ‘find a service’ function on the NHS website to identify the nearest service for selection. All eligible staff within each selected service were invited to take part. Practitioners within each service were identified from website staff profiles or, where not publicly available, by contacting service managers.

Prevalence estimates of pro-choice attitudes among healthcare professionals from earlier research in Britain¹⁷ informed sample size calculations. Following piloting for comprehension and time to completion, questionnaires, together with information sheets, consent forms, unconditional tokens of gratitude (vouchers) and free-post return envelopes, were mailed to individual practitioners within each service. The option of completing the survey online was provided. Non-responders were followed up with two reminder phone calls or emails. Unique ID numbers were provided to enable response rates to be calculated. The fully standardised survey questionnaire probed knowledge of and attitudes towards the regulation and provision of abortion care and support. Space was provided for free text comments to be added.

Measures

Knowledge of the law was measured by probing awareness of the legal requirement for abortion to be signed off by a doctor. Views on regulatory aspects of abortion were measured by seeking agreement with statements relating to abortion as a woman’s choice; abortion as a health rather than a legal issue; gestational age limits; and the probity of abortion. Attitudes towards abortion provision were measured by seeking respondents’ views on whether abortion care should be standard practice in their specialty and on the ability of different practitioners to provide aspects of abortion care. Verbatim wording of statements is provided in Tables 1 and 2. Response options were three-point Likert scales (True, False and Don’t Know; and Agree; Neither Agree nor Disagree; Disagree). Respondent characteristics included in the analysis were gender, age, time since qualification, current involvement in abortion provision; service type, profession, constituent country of Britain and importance of religion and political persuasion.

Analysis

Survey data were analysed in Stata 17,²⁰ taking account of clustering within health service sites. We present descriptive analyses (estimated percentages and 95% confidence intervals) of the knowledge and attitude measures by respondent characteristics. Analysis of data on views on the regulation of abortion included the total sample of healthcare professionals. Analysis of data on views on whether it should be provided in respondents’ own healthcare services excluded providers in dedicated abortion services.

Free text comments relevant to the survey responses were selected for thematic analysis with the aim of better understanding survey responses. Textual data were independently analysed thematically by KW and RF using the Framework method^21,.

Patient and public involvement

A Patient and Public Involvement panel was established for the study to help inform study design and recommendations.

Knowledge

Of all healthcare professionals surveyed, 78.6% selected ‘True’ to the statement: ‘An abortion is a criminal offence unless it has been signed off by a doctor’; 9.4% selected ‘False’. The proportion providing the correct answer was higher among women (80.3%) than men (65.2%) and increased with years since qualification and with age (Table 1). A third of health care professionals aged under 30 were unaware of this aspect of the law. Correct understanding of the law was more common among those currently providing abortion in any service (86.9%) compared to those not doing so (65.3%) and was near universal among those working in a specialist abortion service (96.3%). It was less common among those in other service types; almost half of those in pharmacies were unaware of this legal requirement.

Attitudes towards the regulation of abortion

Agreement with the view that abortion was a woman’s choice was high; 90.7% overall saw it as such and only 2.4% disagreed (Table 2). Agreement was higher among women (92.3%) than men (79.4%), among respondents currently providing abortion (95.9%) compared to those not doing so (82.1%) and among those seeing religion as not important in their lives (93.6%) compared with those for whom it was very important (73.3%). Agreement was also higher among healthcare professionals in maternity services (95.4%) and specialist abortion services (98%) compared with those in general practice (77.4%) and pharmacies (80.8%).

Support for the view that abortion was a health, and not a legal issue was less widespread, but was nevertheless a majority opinion with 68.2% overall agreeing and 8.9% disagreeing.

Agreement was lower among those considering religion to be very important in their lives (45%) compared with others and decreased with time lapse since qualification. Important differences by service type were seen only for healthcare professionals working in general practice, fewer than half of whom (46.5%) endorsed this view.

A small minority, 6.2%, of respondents agreed that abortion at any gestational age was against their personal beliefs, 83.1% disagreed. Levels of agreement were marginally higher among men, respondents with right of centre political views, and those employed in general practice and pharmacies (rising slightly above 10% in each case). Among those for whom religion was very important in their lives, it was considerably higher, at 28.3%.

A similarly small minority, 6.7% agreed with the statement that abortion should not be carried out after 12 weeks gestation; 76.9% disagreed. Endorsement of the statement was more common among respondents aged 40 and over and among those for whom religion was very important.

Again, marked differences were seen by service type. The proportion of healthcare professionals who held this view was higher among staff in general practice and reached one in five in pharmacies (18.9%).

Attitudes towards provision of abortion in different service settings

Of the 517 respondents employed outside of dedicated abortion services, 37% agreed that abortion care should be standard practice in their specialty; 28.5% disagreed. (Table 1) Differences were seen by service type and professional role. Fewer than one in five healthcare professionals in general practice (18.7%) and one in four of those in pharmacies (25%) supported the idea, compared with the majority of those in sexual and reproductive health services (58.4%) and a sizeable minority (42.9%) of those in maternity services. Receptivity to the idea of abortion provision in their service was higher among nurses and midwives than among doctors, across all services.

Views on provision of abortion care by different practitioners

All respondents were asked for their views on which practitioners, other than designated abortion providers, should be able with training to provide aspects of abortion care. Support for prescription and provision of abortion medication by such practitioners was higher than for surgical methods being carried out by them. (Table 2) Endorsement of the notion that doctors should be able to prescribe and dispense was near unanimous; 96.5% and 85.8% respectively signified approval. For nurses it was slightly lower but still a majority view; nearly two thirds of respondents (65.3%) felt they should be able to prescribe, and 85.8% that they should be able to dispense. Support for midwives and pharmacists prescribing abortion medication was indicated by roughly a third (31.6% and 34.7% respectively) but more than half felt they should be able to dispense abortion medication.

With regard to surgical methods, again there was strong support for doctors carrying out surgical abortion, with 94.8% indicating that they should be able to perform surgery before 14 weeks and 92.9% after this time. Affirmation of surgical methods carried out by nurses was lower, but a sizeable minority of respondents (41%) felt that nurses should be able to carry out surgical abortion before 14 weeks. Support for surgical methods being carried out by midwives was lower, and by pharmacists was negligible.

Free text comments

26 free text comments added to the questionnaire related to the regulation of abortion and 56 to the appropriateness of abortion provision in different healthcare settings.

Comments relating to the regulation of abortion were universally in favour of relaxing the law. (Table 3) None stated a preference for retaining the current legal restrictions which were seen as outdated and as having adverse consequences for the efficiency and quality of abortion provision. Where this view was qualified, it was with reference to the need for a prior medical consultation. Strong opposition was expressed to the continued need for two doctors to authorise an abortion. It was held that, should the legal requirements remain in place, since abortion was increasingly led by nurses their role should include responsibility for certifying that the grounds for abortion were met.

There was a consensus that abortion should be nurse-led with the proviso that this should be a provider’s individual choice. Mandating responsibility for abortion provision was considered likely to negatively impact both patients and healthcare professionals. Comments suggested that the lower levels of endorsement for incorporating abortion into standard heath care seen in the survey data may be attributable to recognition of the challenges rather than to disinclination to provide abortion. While benefits of integration were identified, notably reducing stigma surrounding abortion, increasing access and availability of provision, and providing more holistic care, so too were potential costs. These included a possibly detrimental effect on other services, a poorer quality abortion service where skill sets were not adequate and possibly less sensitive treatment of patients. Greater involvement of primary care in abortion provision was seen as currently unfeasible given perceptions that the service was already understaffed and overburdened. Similar, though more muted, reservations were expressed in relation to midwifery with the additional concern that midwives were unlikely to see enough abortion patients to maintain essential skills. Pharmacists were considered useful sources of information but reservations focussed on the availability of private space. By contrast, and amplifying survey responses, clear benefits were identified for abortion provision in community sexual and reproductive health services in terms of continuity of care, provision of a more holistic service and attention from knowledgeable staff who were sympathetic to patients and comfortable with the service they provided. Absorption of the function was seen as conditional on a change to commissioning patterns and addressing resource limitations.

Strengths and limitations

Our study provides much needed data on the views of healthcare professionals on abortion regulation. A strength lies in the range of healthcare professionals surveyed, unique in British studies to date, providing opportunities for comparisons to be made between specialities and settings. The study also benefits from use of a stratified cluster sampling strategy, increasing the generalisability of findings. The addition of qualitative data from free text comments provides insights into survey responses. The main limitation stems from the timing of the study. Scheduling fieldwork at the height of the COVID-19 pandemic in Britain and its immediate aftermath may have introduced both participation and reporting bias. Healthcare professionals most actively involved in coping with the pandemic may have declined to participate because of time pressures. The responses of those who did may reflect heightened awareness of the constraints of their workload which may have negatively influenced their propensity to be involved in abortion care and support.

Interpretation and contextualisation

Comparisons of the findings of this study and those of others are difficult because of differences in questions asked, populations under study, and recency of investigation. We found no other studies exploring knowledge of regulations governing abortion provision. Other studies in Britain have found similarly liberal attitudes among healthcare professionals towards the regulation of abortion. In 2000, a survey showed that four in five GPs considered themselves broadly pro-choice¹⁷ and in an investigation into the attitudes of medical students in Britain, 73% of medical students did so¹⁶. Strong associations between attitudes towards abortion and the importance of religion are also seen in other studies^22,16,23,24. We found no other research which compared attitudes towards abortion by service type or profession. The views of healthcare professionals appear to be generally in line with those of the general public. The most recent British Social Attitudes ^Survey25 shows 76% support for allowing abortions if the woman does not want the child.

Regarding provision of abortion, our findings are less consistent with those of others. The strong support amongst healthcare professionals in community sexual and reproductive health clinics for abortion provision being standard practice in their service is consistent with evidence that such settings would be sensitive to the needs of abortion patients and better able to meet other sexual and reproductive health needs including ongoing contraception ^{26 27}. However, the markedly lower level of enthusiasm for routinely incorporating abortion care and support into general practice sits in tension with recent international evidence suggesting the merits of doing so ^{28, 29,30,31 32}. Thus, the recommendation that insights from LMIC countries might translate to the UK setting²⁸ needs to be treated with caution. Contextual differences in Britain – notably, the pressures under which NHS is operating and the consequent burden on primary care and other health services – may adjust the cost-benefit ratio and limit knowledge translation from other settings.

Implications for policy and practice

We anticipate that our findings will be of interest to commissioners and policy makers. First, the data may inform changes to practice within existing legislation. The recognition among healthcare professionals that aspects of the current law are out of step with best practice echoes earlier findings of the RCOG⁷ and the House of Commons Science and Technology Committee³³ that the requirement for two doctors’ signatures no longer serves a useful purpose. With regard to which practitioners are permitted to provide abortion, the support shown in our study for nurses and midwives to be permitted to authorise abortions and to prescribe abortion medication is consistent with the WHO guideline urging consideration of the wider range of cadres which can safely perform medical and surgical abortion³⁴.

At the time of writing, abortions must be performed within NHS hospitals, licensed abortion clinics or other classes of places licensed by the Secretary of State for Health, including the patient’s home. Our findings suggest that consideration might also be given to routinely licensing sexual and reproductive health services to offer abortion care. They suggest the need to proceed with more caution in considering any wholesale extension of abortion care into General Practice. Although a sizeable minority of staff in this specialty would support such a move, offering the potential to increase access - especially in underserved rural areas - the majority would not. While examples can be provided from elsewhere in the world of the potential advantages of integration into general practice, high quality services are unlikely to result where practitioners currently have neither the capacity nor the resources to offer them.

Finally, should there be political will for a legal reform that removes specific criminal prohibitions against abortion, our study suggests that it would encounter strong support amongst healthcare professionals. Our respondents expressed a clear preference for regulating abortion as a matter of health and the individual’s choice, rather than as an issue properly addressed through criminal law.