Contribution
My Lords, I introduced an identically worded Bill in the previous Session and when it was debated at Second Reading, almost exactly two years ago in December 2024, there was pushback from certain noble Lords who accused me of being in the hands of dark American money. I hope we will hear no more along these lines today. It is untrue as well as insulting, and it trivialises an important topic.
This Bill is not about altering in any way access to abortion. It is about the welfare of women, who are entitled to accurate information about the risks and procedure involved. I hope noble Lords who speak against the principle of the Bill will have good arguments as to why women and clinicians should not have accurate information. Nor is it my purpose to claim that abortion is unsafe; I am quite happy to agree that abortion is generally a safe procedure for women. Rather, the question this Bill raises is how safe it is. To know that, we need statistics that reflect real experience.
The starting-point is that the Abortion Act 1967 requires each termination to be reported to the Chief Medical Officer and specifies the information that has to be provided. The information must include any complications arising from the procedure. This is known as the abortion notification system. The great majority of abortions, 80%, are provided by independent clinics, such as the charities BPAS—the British Pregnancy Advisory Service—and Marie Stopes, though there are other, smaller providers as well. The abortion notification system covers these independent providers and the 20% of terminations carried out in an NHS setting. The form they return is known in DHSC jargon as an HSA4 but, as I say, its contents are set out in a section of the Abortion Act. Completing and returning it is a statutory obligation for each termination.
I have no reason to think that providers are not completing and returning the forms. I am sure they are complying with the statutory requirements put in place by Parliament in 1967, but that is where the weakness in statistics arises. In 1967, all abortions would have involved a surgical procedure and very likely an overnight stay, whether in an independent clinic or an NHS hospital. Complications would have become apparent quickly and so could have been included in the HSA4 form.
The position today is very different. A surgical abortion is nowadays a rarity. Currently, about 85% of abortions are medically induced—that is, by taking a sequence of pills. As recently as 2012, only 48% of abortions were medically induced, but, by 2023 it was 85%. These terminations frequently take place at home, and indeed since the Covid pandemic it is possible for a woman to obtain the medication from an abortion provider without an initial in-person interview or examination.
This shows that the abortion landscape is shifting very rapidly and clinicians need to have available the most robust data about complications, as indeed women do, for the purpose of informed consent. What does this changing pattern of provision mean for our understanding of statistics? If a complication is to arise from an abortion, it is likely that it will do so quickly. Haemorrhage, sepsis and the like are soon apparent, but a woman who has undergone a medical abortion at home is not, by definition, in a clinic or medical setting. What is she to do? The overwhelming likelihood is that she will report to A&E. The complication in that case will not be communicated to the abortion provider and will not therefore be reported on the HSA4 form. It will, however, be recorded in hospital statistics.
I add at this point that there is no legal obligation on the Department of Health and Social Care to publish the statistics on abortion that it collects through the abortion notification system. It has, however, published them unfailingly until two years ago. When we debated my Bill in the last Session, in December 2024, we were making use of the 2023 abortion statistics. Noble Lords will be disturbed to learn that they are, even today, the most recently published statistics. Something seems to have gone grievously wrong at the DHSC, and no explanation has been provided.
I hope that, today, the Minister can tell your Lordships why exactly the statistics have not been published in the last two years. What has gone wrong? I understand that, in the course of next month, it is likely that the 2024 statistics will be published, and the Minister will be able to say whether that understanding is correct. However, we are still way behind. When are the 2025 statistics expected? What has gone wrong and why are the figures not being made available?
To some extent, that is a side issue to my main argument. Returning to that, do the official abortion statistics properly capture the true rate of complications? To know that, we do need to proceed on the basis of conjecture. If one wants to know the rate of complications arising from abortions, one needs to look, fairly obviously, not merely at forms returned by abortion providers but at what is going on in hospitals.
Fortunately, we know about this. The Office for Health Improvement and Disparities, which is a branch of the NHS, turned its hand to the task of interrogating hospital data, and in November 2023 produced its report. The task was more challenging than one might have thought. There were, of course, many of the usual statistical conundra: what to count, what not to count, and whether definitions in different datasets were the same and so forth. In fact, almost the first half of its report is a careful and thorough essay on the methodology it used.
It focused entirely on incidents arising in a hospital setting, using hospital episode statistics, and, importantly, separated out incomplete abortions that were not accompanied by a further complication—that is an important point I will come back to. It broke the complication rate down by age, showing, as is generally accepted, that it is somewhat higher for older women, and by type—for example, between haemorrhage, which is the most common complication, sepsis, cervical tear and so forth.
Its headline finding, so to speak, was that the complication rate was between 3.5 and 4.4 per 1,000 in the period 2017 to 2021, which was the period it looked at, depending on the year. This compares with data published by the DHSC up to 2023 based on the abortion notification system alone, which shows a complication rate of about 1.2 to 1.4 per thousand in recent years. I understand this is statistically significant. However, the complication rate established by the OHID rose significantly, to about 17, 18 or 19 per thousand, if incomplete abortions without further complications were included. It is unfortunate, therefore, that it appears that there are no plans for the OHID to continue to collect the valuable data contained in its report. My Bill would place an obligation on the Government to do so.
Clause 1 requires the Secretary of State to publish an annual report on complications, states that the purpose is
“to inform policy and safe practice”,
requires the inclusion of data from both the ANS and the hospital episode statistics, specifies that the report must cover the same areas of information as the 2023 report, gives the Secretary of State power to add further information he or she sees fit to include and states that the first report must appear within a year. Clause 2 covers territorial scope and commencement. In effect, the Bill applies to England. There is no other clause.
It is important to note that the Bill mandates the collection of both ANS and HES data. Those who worry—especially professional statisticians—that there would be a rupture in the long-established ANS statistical sequence need have no concern. The ANS figures would still be collected, as required by law, before being aggregated with separately sourced HES data.
There are real stories at the bottom of this. This morning, I received an email from a paediatrician practising in the Midlands. I do not feel I can name the person without their permission, but they said, “It is perfectly possible within current practice and guidance for a woman to be prescribed and take abortion medication remotely when she is more than 12 weeks pregnant, but only to realise this once the abortion has taken place. In this circumstance, it is entirely possible, and indeed has happened, that the woman delivers a baby who is of viable gestation and may be born with signs of life”—this would be at home—“and then subsequently dies. I am aware from my current practice and through liaising with peer networks that cases such as this have occurred. However, there is no formal reporting mechanism that exists to capture and monitor the frequency with which this is occurring”. Women have a right to know these things, and that email demonstrates that their clinicians want to know as well. We need better statistics.
I will deal with some other possible objections. First, I refer to the comment from BPAS, quoted in the Library brief, that the Bill “exceptionalises” abortion. This is not so. What exceptionalises abortion is the original Abortion Act, with its unusual statutory requirement for mandatory reporting. BPAS is entitled to lobby for legislation to remove that exceptional requirement, but that is not this Bill, which simply accepts the legal status quo on this point. BPAS’s other comments seem to me to be very weak. The key point is that even it agrees that
“further work needs to be done on collection and analysis of large datasets relating to women’s reproductive health”.
The fact is that nobody, not even the British Pregnancy Advisory Service, defends the existing ANS statistical sets as giving a realistic picture. It also claims that there are better ways of dealing with the problem than this Bill. That may be true, but it does not suggest any.
I have one further objection and then I shall sit down. It may be said that an Act of Parliament is a bit of a sledgehammer to crack this particular nut. I certainly see the force of that—it is true. But it is a Private Members’ Bill, intended to some extent to provoke debate. Indeed, if the Minister in responding were to give a firm undertaking to do what the Bill requires without the bother of statutory obligation, I would be perfectly content with that. I beg to move.