Terminally Ill Adults (End of Life) Bill (Twenty-ninth sitting)

Sarah Olney Excerpts

Committee stage
Tuesday 25th March 2025 Public Bill Committees

On resuming—

[Sarah Olney (Richmond Park) (LD)]

It is a pleasure to serve under your chairmanship, Sir Roger. I rise to speak in support of new clause 33 and new schedule 3.

I am concerned that amendment 455 will be inadequate to maintain trust in the system. Collecting comprehensive data, such as the detailed statistics outlined, in an assisted dying regime serves several critical purposes that revolve around ensuring transparency, accountability, safety, equity and continuous improvement of the system. It is about ensuring from the very start that there is good data and learning from the paucity or the blanks in other regimes and from where they have been criticised.

The Secretary of State would have the power to amend the new schedule, but Parliament’s intent and our expectation that the system will be scrutinised would be crystal clear. We need to collect data for monitoring and oversight and gather detailed information about individuals who request assisted dying, and about the process itself. That would allow regulators to monitor how the system is functioning, which would help to ensure that the law is being applied consistently and fairly, preventing misuse or abuse.

We need to safeguard our vulnerable population. Data on characteristics such as age, ethnicity, disability and mental health conditions and on deprivation indices can reveal whether certain groups are disproportionately seeking or being denied assisted dying. That helps to identify potential coercion, discrimination or gaps in care that might drive requests, ensuring that the system protects vulnerable individuals.

For the purpose of improving healthcare and support services, information about the prior involvement of palliative care or psychiatry teams, and about referrals made after requests, highlights whether patients are receiving adequate support before opting for assisted dying. If data shows low referral rates or unmet needs, it could prompt improvements in healthcare, access or quality, potentially reducing the number of requests driven by untreated suffering.

For the purpose of evaluating the process, tracking timelines—that is to say, from initial discussion to death—and procedural steps such as second opinions and panel decisions ensures that the process is efficient yet thorough. It can reveal bottlenecks, delays or inconsistencies, allowing policymakers to refine the system for both patients and clinicians.

For the purpose of understanding motivations and outcomes, collecting the reasons for requests, refusals and complications such as drug efficacy or adverse reactions provides insight into why people seek assisted dying and how the process unfolds. That can inform public policy, clinician training and patient counselling, ensuring that decisions are well informed and voluntary.

For the purpose of ensuring accountability, data on clinicians, pharmacies and panel decisions creates a transparent record of who is involved and how often. That helps to prevent overreliance on a small group of professionals, ensures that ethical standards are upheld and allows for audits if concerns arise.

For the purpose of public trust and ethical debate, detailed statistics foster transparency, which is essential for maintaining public confidence in a sensitive and controversial practice. They also provide evidence for ongoing ethical and legal debates, enabling society to assess whether assisted dying aligns with its values and goals, and for the purpose of learning and adaptation. By analysing complications, drug effectiveness and reconsideration requests, authorities can refine protocols, such as by adjusting approved substances or dosages, to make the process safer and more humane. Long-term data collection supports evidence-based adjustments to the regime.

Almost all jurisdictions around the world have systems for reporting on people requesting and having assisted deaths. However, there is a high degree of variability in the data reported around the world; very few data items are published consistently in all jurisdictions. We have an opportunity to learn from monitoring and reporting systems elsewhere, and to put the most robust possible system in place.

Research published in BMJ Supportive & Palliative Care in 2022 examined in depth the data reported in jurisdictions around the world. It identified official data reports from 16 jurisdictions in which assisted suicide or euthanasia is legal. It found that although most jurisdictions report the number of patients who die by assisted dying, only three—New Zealand, Canada and the state of Victoria in Australia—document the number of patients who make a request to die by assisted dying. The research is from 2022, so it is possible that there are now a few more jurisdictions.

Unless we know about the applications that are assessed as ineligible, we will not have adequate insight into the functioning of patient safeguards and eligibility criteria. That is a point that Dr Annabel Price made in oral evidence and in an editorial in The BMJ in 2015. We need to understand not just who has an assisted death, but who is ineligible and why, to understand how the safeguards are functioning.

The 2022 paper identified Colorado, Hawaii and Maine as reporting the number of patients who received assisted suicide prescriptions, but not the number of persons who actually die by their ingestion. In these jurisdictions, there is no record of how many of the patients ingested the prescribed drugs. This is stipulated in the 2021 data summary for the Colorado End-of-life Options Act, which states that the Act

“does not authorize or require the Colorado Department of Public Health and Environment to follow up with physicians who prescribe aid-in-dying medication, patients, or their families to obtain information about the use of aid-in-dying medication.”

The official statistics reported therefore

“reflect all deaths identified among individuals by prescribed aid-in-dying medication…irrespective of whether their death was caused by ingestion of medication, the underlying terminal illness or condition, or some other cause.”

These states do not appear to have a mechanism to ascertain the amount of unused assisted suicide drugs circulating within the community. The fate of many therapeutic drugs with the potential for abuse or use for fatal overdose, such as diazepam or morphine, is not monitored, but the doses and combinations of lethal drugs used in assisted suicide and euthanasia may present a heightened risk, so monitoring the fate of those drugs should be considered.

Concern about pressure on vulnerable groups warrants the detailed reporting of patient demographics. It is needed to ensure patient safety because it allows researchers and monitoring organisations to monitor trends to determine whether there is disproportionate participation among vulnerable or minority groups and, if so, why.

Even basic patient characteristics, including gender, age and ethnicity, are not universally reported in other jurisdictions. Although marital status could be an indicator of at-home support, fewer than half of reports include it. The level of education can be an indicator of socioeconomic position, but is not routinely reported by many jurisdictions.

Although we have been told that studies that have analysed routine statistics did not indicate a disproportionate use by lower socioeconomic groups, socioeconomic trends have changed over time and new evidence has emerged. For example, in Oregon, the proportion of assisted dying patients on state health aid has doubled in the past decade; it reached 79% in 2021. That indicates an increase in use over time by those in a lower socioeconomic position.

Similarly, an in-depth review by Ontario’s coroner last year showed that those dying by their assisted death legislation, both track 1 and track 2, were more likely to be those with longer prognoses. Given the impossibility of prognostication and the pressure being brought to bear by some campaign groups for the Bill to be wider, socioeconomic data will be very important to ascertain whether there is equity of access to assisted dying and whether financial pressures, including access to benefits, are involved in areas of higher or lower uptake.

Knowing whether patients have received care from hospice or palliative care services provides an understanding of whether patients have been provided with alternative end of life care options. That information is included in fewer than one half of the reports from jurisdictions around the world, with none providing information on the duration or scope of hospice or palliative care involvement. For example, while official reports indicate that 80% to 90% of Canadian patients who have assisted deaths receive palliative care, other studies have shown that fewer than half of such patients had seen a specialist palliative care team.

It is also essential to have information about the clinicians who participate in assisted dying, because the clinical demand of assisted suicide and euthanasia is highly time-consuming, with evidence from other jurisdictions, such as an Australian study of those participating clinicians, showing that it takes 60-plus hours of work per assisted death. Understanding the number of clinicians participating in assisted dying and their speciality is important in understanding the willingness of clinicians to participate, and how the clinical demand for assisted dying is distributed across specialities and practitioners.

Information on the drugs used in assisted deaths is critical to understanding the safety and efficacy of different drugs and drug combinations. The 2022 paper found that just six jurisdictions report the drugs used to bring about patient death. There is no one drug or drug combination that has been shown to be most effective and safe for ending patients’ lives. If that research existed, we would not see such variation in the drugs used. Data on complications following drug ingestion or administration is needed to understand the safety of different drug combinations, but it is rarely reported.

Even in jurisdictions where information on safety is included, missing data can be high. For example, in Oregon, data on complications is unknown in up to 70% of cases annually. Reporting the drugs used to bring about patient death, as well as information about the assisted dying process, including the time from drug ingestion or administration to patient unconsciousness and death, and the presence of complications, would enable the safety and efficacy of these drugs to be assessed.

As in many jurisdictions around the world, what I am proposing relies on retrospective reporting of data, after the patient has died. I suggest that this is the minimum standard that must be achieved. Ideally, we would also build in processes for prospective scrutiny, before the person has died, as well as formal review processes that scrutinise individual cases in more detail. Analyses from Belgium and the Netherlands, where review processes are established, have shown that 48% of assisted deaths in Belgium, and one in five of such deaths in the Netherlands, are not reported via the official reporting system. In some cases, legal requirements are not followed.

Setting up a post-event review panel, such as happens in the Netherlands and Ontario, would improve understanding and safety. In the Netherlands, every case is reviewed by a panel, and many cases, especially the controversial ones, are published to promote education and debate. Controversial or worrying cases will occur in England and Wales if this Bill passes, and it is essential for ongoing patient safety that these are not swept under the carpet, but that there are transparent processes to understand and learn from them.

We have heard frequently in Committee that there is no evidence of harm from other jurisdictions, but we have also heard it powerfully argued that other jurisdictions do not collect the data we would need to determine that. In Oregon, information on complications following ingestion of lethal substances is missing in around 70% of cases. Robert Clark, the former Attorney-General for Victoria, has written:

“The Victorian oversight and accountability structure can best be described as one of ‘hear no evil, see no evil, speak no evil’. In other words, it appears designed for the regulator to find out nothing, investigate nothing and report nothing that could suggest that assisted dying has been anything other than an unblemished success.”

Let us not repeat those mistakes. We can lead the way in rigorous, comprehensive data collection and transparent monitoring.

Amendments tabled by the hon. Member for Spen Valley change responsibility for oversight from the Secretary of State and chief medical officers to the commissioner. This brings a danger of bias in reporting, as there could be an element of marking one’s own homework. This means that there is even more reason for complete data reporting and transparency. Ensuring that data collected is comprehensive and reporting is transparent may go some way to assuage that concern. Importantly, strengthening data collection reporting and monitoring will not make the process harder or more lengthy for the person requesting an assisted death. It just makes the system safer and stronger.

In this amendment, I am not proposing doing anything new or radical. We are simply taking the strongest aspects of what is done in other jurisdictions and bringing them together to enable the best and most comprehensive reporting in the world. All new laws have risks as well as benefit. We must identify and be open about those risks in order to learn from harms and improve onward safety. New schedule 3 will help us to achieve that.

[Naz Shah]

I rise to speak to amendments 455 and 456, tabled by my hon. Friend the Member for Spen Valley. Under the Bill as drafted, an annual report would be issued by the chief medical officers for England and Wales. If the other amendments pass, that report will be issued by the voluntary assisted dying commissioner.

Amendment 455 will do two things, broadly speaking. Proposed new subsection (2A) states that the commissioner’s report

“must include information about the application of the Act in relation to…persons who have protected characteristics”.

Amendment 456 clarifies that the definition of “protected characteristics” is the same as that used in the Equality Act 2010. Proposed new subsection (2A) also provides that the commissioner must report on any other description of persons specified in regulations made by the Secretary of State.

--- Later in debate ---

[Sarah Sackman]

I suggest that those are ultimately policy choices, first for the promoter, my hon. Friend the Member for Spen Valley, and then—if the regime is voted on and passed—for the commissioner. In reality, a lot of the data that one will collect when one has due regard to the differential impact on these protected groups will in practice have a large degree of overlap with other aspects of socioeconomic disadvantage, as is so often the case in equality impact assessments.

[Sarah Olney]

I had an interesting exchange with the hon. Member for Penistone and Stocksbridge last week about socioeconomic indicators. They are in the Equality Act but have not yet been implemented. I understand that it was in the Labour Government’s manifesto to bring forward the implementation of the socioeconomic factors in the Equality Act. Given that, as I understand it, the implementation of the Bill will probably take us up to the next general election, does the Minister not think that, at some point before it is fully implemented, there will be a way of collecting the relevant socioeconomic statistics?

[Sarah Sackman]

The hon. Member is right. Section 1 of the Equality Act 2010 refers to taking into account the impact of public administration decisions on socioeconomic disadvantage. The previous Labour Government brought that provision in under that landmark Act, but it was never brought into force. She is right that the Government are looking to do that. The question is: what is appropriate for the Bill that we are considering right here, right now? From the Government’s point of view, the voluntary assisted dying commission will be subject to the Equality Act, as it stands, and it is right that there is no need to specify the carrying out of an equality impact assessment in the Bill. That assessment will, at the appropriate time, contain whatever the Equality Act requires of that public decision maker. I do not think that we need to pre-empt, through this vehicle, things that have not yet happened in other areas of legislation.

The other point is that, leaving over for the regulations and guidance that the voluntary assisted dying commission is anticipated to produce, to take into account what needs to be taken into account in order to fully understand the regime and how it is operating, can all be developed by what will ultimately be an expert body in this field, should Parliament decide to vote for it. The question in the context of the amendments is: what do the Committee and Parliament say needs to go in the Bill right now? The Government’s position is that the amendments are neither necessary nor workable in our current legal regime.

Amendment 451 seeks to further specify what is included in the commissioner’s annual report under the promoter’s amendments to clause 34. Under amendment 451, the report must contain information on where there have been complications in the procedure, and we assume that to mean the procedure relating to the self-administration of the approved substance, as I think was clarified earlier by my hon. Friend the Member for Bradford West, who referred to clause 9 in that regard. The commissioner’s report would also be required to include analysis of a specified list of matters, including the diagnosis and prognosis of persons seeking assistance, concerns raised by family members or the multidisciplinary panel, and demographic data, including socioeconomic data and information on protected characteristics.

The Government have worked with the promoter on her amendments 455 and 456. Under amendment 455, the commissioner’s annual report would be required to include information about the application of the Bill to persons with protected characteristics and any other descriptions of persons specified in regulations made by the Secretary of State. That is the point: the regime can evolve as appropriate when we know what it looks like and how it is operating. It can even go further than some of the categories that Members have described. That may include persons from different demographics and socioeconomic backgrounds, as proposed by the amendment. The amendment would also require the commissioner to consult, when preparing an annual report, with the chief medical officers for England and Wales and others who appear to the commissioner to represent the interests of persons who have protected characteristics and who the commissioner deems appropriate.

--- Later in debate ---

[Kim Leadbeater]

This is a standard power for regulations to include the power to make incidental, consequential, transitional or saving provision. This is a minor amendment, and it is a standard but vital measure to ensure the legislation works effectively and cohesively.

[Sarah Olney]

The clause, when amended by amendment 545, which we will vote on shortly, brings into stark relief how much of the Bill is to be left to regulations and how little say MPs will eventually get over it. When it comes to voting on Third Reading, MPs will have little certainty on how the Bill will actually operate—so much is to be taken on trust. If, through the gaps and grey areas, people come to harm, that will be on Parliament for failing to address these issues while we had the chance.

Let us consider all the powers that the Bill creates. Clause 5 creates a power to determine what goes into the first declaration, subject to the requirements of amendment 418. Clauses 5, 8 and 19 create powers to determine the training, qualifications and experience of the medical practitioners, with no minimum floor, and we do not even know who will ultimately decide the training—that, too, is to be left to regulations under amendment 186. Clause 6 creates powers to determine the forms of proof of identity, with no requirement for those to be photographic or to prove residency.

Clauses 7, 13 and 21 create powers to determine the statements, declarations and reports required from the doctors and the applicant. Clause 11 creates powers to make regulations for the replacement of the co-ordinating doctor if they have died, are ill or are otherwise unable or unwilling. Clause 15 creates a new power to determine who can be a proxy. Clause 20 creates a power for the approval of drugs and substances to end people’s lives. Clause 28 creates a power to determine the prescribing, dispensing and transporting of approved lethal substances —the right hon. Member for North West Hampshire has called this purely administrative. Clause 29 creates a power to make regulations relating to the registration, certification and recording of deaths. Clause 30 creates a power to issue codes of practice that practitioners must simply “have regard to”.

Clause 32 will create an extraordinarily broad Henry VIII power, as we are completely in the dark as to how the assistance will actually be provided. Clause 33 creates a power to determine which events should be notified to the commissioner and how to enable the exchange of information. That is all to be decided by the Secretary of State, with the detail removed from the Bill.

All those things are to be subject to the negative procedure, except the training and qualification requirement for the first and second doctor, the codes of conduct, and the power for the Secretary of State to arrange for delivery of assisted dying. When a statutory instrument is made through the negative procedure, there is no need for a vote. The only exception, by convention, is when the Leader of the Opposition asks for one, but they understandably would not do so on a conscience issue. In practice, Parliament will never get a say on these matters. Is the Committee content with that?

Even on affirmative statutory instruments, Parliament’s say is limited to a 90-minute debate on a motion that cannot be amended. Suppose that the Secretary of State published a code of practice on the assessment of capacity that MPs felt did not take sufficient account of mental health conditions. They would be faced with the option to vote for it despite its inadequacy, or to vote against it and risk creating a position where there are no codes of practice at all. That is an impossible position to place parliamentarians in, yet that is what the Bill does. It is a massive blank cheque to the Executive.

I say to Government Members, who I know have great faith in the Health Secretary, that according to the Interpretation Act 1978, “Secretary of State” means any Secretary of State. The powers will not necessarily be exercised by the Health Secretary; they could be exercised by another Secretary of State. Indeed, under the doctrine of ministerial delegation, they could be delegated to another Minister who is not a Secretary of State.

I make these points to illustrate that there is an issue of principle that cannot be waved away on the basis that we think someone sensible will exercise the powers. Parliamentary scrutiny matters, and I very much regret that the Bill avoids it through these wide regulation-making powers.

[Stephen Kinnock]

The Government have worked with my hon. Friend the Member for Spen Valley on amendment 223, which would broaden the regulation-making power using standard legal wording so that any regulations made under the Bill can ensure that it works effectively with other legislation.

Amendment 223 agreed to.

Amendment proposed: 545, in clause 39, page 23, line 6, leave out subsections (3) to (5) and insert—

“(3) The Secretary of State may not make a statutory instrument containing (whether alone or with other provision) regulations under section 5(3A), 8(6A),30(3), (Voluntary assisted dying services: England) or (Voluntary assisted dying services: Wales) unless a draft of the instrument has been laid before, and approved by a resolution of, each House of Parliament.

(4) Any other statutory instrument made by the Secretary of State containing regulations under this Act is subject to annulment in pursuance of a resolution of either House of Parliament.

(5) The Welsh Ministers may not make a statutory instrument containing regulations under section (Voluntary assisted dying services: Wales) unless a draft of the instrument has been laid before, and approved by a resolution of, Senedd Cymru.”—(Kim Leadbeater.)

This amendment brings together the various provisions about the procedure for regulations. It also makes regulations under clauses 5 and 8 about training, qualifications and experience, and regulations under NC36 and NC37, subject to the draft affirmative procedure.

Question put, That the amendment be made.

--- Later in debate ---

[Sarah Olney]

I rise to speak to my amendments 531 and 535 and to amendment 490 in the name of my constituency neighbour, my hon. Friend the Member for Twickenham (Munira Wilson).

I welcome what the hon. Member for Spen Valley has just said about needing more time to get this right. As I said earlier, there is still so much that has been left to secondary legislation. I regard that as unsatisfactory in itself, but the fact that we are going to take time to get it right will surely go some way to mitigating some concerns.

I have heard raised around the estate on a number of occasions today the concern that the four-year gap between passing the legislation and implementing it will take us right up to the next general election. I have to say that I share those reservations, even though I am not a member of the party of government, about how the general election campaign will intersect with the implementation of this legislation. I share people’s fears that this whole process will get enmeshed in party politics, which we have hitherto been able to avoid in the Committee, but nevertheless could create a real danger for successful implementation.

I did think that the auto-commencement of this Bill after two years was a little bit reckless. The extension to four years at least permits some potential improvement in palliative care—to re-ignite the spirit of consensus, as that is something that we all want to see—but an NHS already grappling with long waiting lists and staffing shortages could buckle under the added burden of implementing a complex new framework leading to rushed or inconsistent application. An automatic commencement of the new laws, regardless of the situation in the NHS, could leave patients and healthcare workers unsupported.

It is important is that, before the moment comes—whenever that is—Parliament should be fully informed as to how the service will be provided and the strength of other provisions, so that MPs can be confident that those choosing to end their lives through this service are given a fair choice and are not being pushed down a single pathway. We must also have confidence that the medical professionals and panels involved have received adequate training.

Amendment 531 would ensure that at least we go into this with our eyes open and that shortcomings are flagged if not addressed. It would provide that the critical provisions of the Bill, without which assisted dying cannot lawfully operate, do not come into force until a report is made setting out the readiness of services to be provided under the legislation and that the training has been provided. Similarly, amendment 488, in the name of the hon. Member for York Central, mandates a statement from both Houses that all necessary regulations are in place. This would prevent a premature roll-out without clear rules. These steps would not halt progress; they would ensure we are ready to proceed.

Amendment 490, in the name of my hon. Friend the Member for Twickenham, does two important things. First, it requires that if the Secretary of State wants to commence before the auto-commencement deadline, they must publish a report covering the same matters covered by the report under clause 35. This would not be a new requirement, but would instead bring it forward. Crucially, that report would include details about the provision and availability of palliative care services in England and Wales.

Secondly, amendment 490—this is it goes further than my amendment 531 and amendment 488—requires that regulations bringing the Act into force prior to the auto-commencement deadline use the affirmative procedure: in other words, that there will be a vote in Parliament. It stands to reason that if the Secretary of State wants to go quicker than the four-year deadline, then Parliament ought to be able to consider whether all necessary preparations have been done and to refuse early commencement if that is not the case.

Amendments 488, 490 and 531 would not interfere with the auto-commencement point in four years’ time: when four years is reached the whole scheme will come into force. Those amendments would not prevent that, but they would achieve greater oversight and accountability if the Secretary of State decided to commence early. I hope that even if the hon. Member for Spen Valley is not minded to accept striking out the auto-commencement provisions, she will still accept these amendments.

[Kim Leadbeater]

The hon. Member makes some very fair points, but I think they will be covered by my new clause 40.

[Sarah Olney]

I am grateful for that intervention, but the crucial point is the vote in Parliament, so that MPs get not only an opportunity to consider whether sufficient progress has been made—as per new clause 40—but a vote to confirm that they are content for the legislation to proceed.

[Kit Malthouse]

I am sure the hon. Lady understands that a vote of that nature would not necessarily be binding, or compel the Government to do anything, but there would be other opportunities for the House to consider the legislation in debate and vote on it. She knows that there are a number of ways to do that, including via a Backbench motion or an Opposition day.

[Sarah Olney]

I refer to my earlier comments: there are few enough opportunities for MPs to have any further control over how this legislation is going to be implemented. I do not think that it is asking too much to want a further opportunity to be mandated, and not left—dare I say it—to the vagaries of the Backbench Business Committee. There are Opposition days, but this is cross-party, conscience issue. I can speak only for my own party, but I think there are a number of other issues that the Opposition would want to use those debates to address. Taken together, these amendments would ensure that Parliament is properly informed if the Secretary of State decided to commence the scheme before the four-year auto-commencement deadline.

I want to address how the Bill will be implemented in Wales. On 23 October 2024, in anticipation of the Bill being published, Senedd Cymru debated a motion calling on the Welsh Government to

“a) support the principles of assisted dying; and

b) support Westminster parliament to introduce a compassionate assisted dying law in England and Wales.”

The motion was defeated 19 to 26. Among those who voted against were a number of Welsh Government Ministers, including First Minister Eluned Morgan, and I believe that in three political parties—Labour, Plaid and the Conservatives—there was a majority against. In that context—as Professor Emyr Lewis, who gave evidence to the Committee, has stated—it would be constitutionally wrong to pass the Bill without the consent of the Welsh Senedd. He believes that this is different from other instances where legislation has been passed by Westminster against the will of Wales:

“This is not the equivalent, for instance, of trying to re-base the UK’s internal markets following the chaos of Brexit. When the UK Parliament passed Acts ignoring refusal of consent from Cardiff or Edinburgh, it did so in what were considered to be the interests of the economic coherence of the UK. This Bill, however, is a different sort of creature. It is a Bill which would bring about a profound change in the law relating to life and death, where there are strongly held (and expressed) moral, religious and cultural views against as well as in favour. How can it be constitutionally appropriate for the UK Parliament to pass such a Bill for Wales, when the Senedd has indicated its opposition by democratic vote after a thorough debate?”

[Kim Leadbeater]

I hope the hon. Member can acknowledge that the motion voted on in the Senedd was very different from the legislation before this Parliament. I hope that I can reassure her, as I have previously, that I am in contact with the Senedd and will be meeting Welsh Ministers. New clause 37 will hopefully also provide some reassurance on issues around devolution.

[Sarah Olney]

I again refer to what I said earlier: I fear this is another example where, instead of debating and agreeing these points in Committee or elsewhere in Parliament, we are delegating these matters to be decided elsewhere by some other people, not in the legislation that we are debating and voting on.

I reiterate that the Senedd has already made its view clear. It may not have been a binding vote. It may not have been voting on this specific piece of legislation. However, I think the will of the Senedd is clear: they do not want an assisted dying Bill imposed upon them, and that is what the evidence of Professor Emyr Lewis told us.

[Kim Leadbeater]

I think that is a very inaccurate representation of the vote that took place in the Senedd.

[Sarah Olney]

I am sorry that the hon. Lady feels that.

The complicating factor here is that the criminal law is not devolved to Wales, but health is, as is safeguarding. One possibility is to seek to respect devolution strictly, by either changing the law or giving the power to the Secretary of State to change the law in respect of what is reserved, while giving to the Welsh Ministers the powers in respect of what is devolved. The problem is that the Bill does not quite do that, even with the amendments that have been tabled by the hon. Member for Spen Valley. Those amendments fail to properly identify what is and what is not devolved; instead, they kick the can down the road.

Future litigation might be required to figure out the precise boundaries. For example, regulation of health professionals is a reserved matter, but adult safeguarding is a devolved matter, so is the training of those health professionals on spotting coercive control a reserved or a devolved matter? The Bill does not answer that question. The solution therefore carries a real risk. If Wales decides that it does not want assisted dying and therefore does not provide it on its NHS, and the Welsh Ministers decide not to set up proper training and regulatory programmes in respect of the safeguarding issues raised, the position in Wales will be that it is decriminalised, so the service can be provided by private providers charging a fee and not necessarily properly trained in safeguarding. That is not a sustainable middle ground, so it would naturally pressure Wales to move towards commissioning such services itself.

In other words, the Bill either respects devolution at the risk of creating a legal limbo, or it puts pressure on Wales to catch up with England, neither of which are desirable. Another alternative is to take away any choice that Wales might have on the matter entirely, and have the law apply in Wales in exactly the same way it does in England, including when it comes to the actual delivery of services. That would avoid the legal limbo problem, but at the cost of acting in a way that is constitutionally quite wrong.

That brings us to the solution that Professor Lewis proposed in his blog:

“How then might the Bill be amended to respect the Senedd’s vote on the matter of principle? One straightforward way would be to provide for different commencement provisions in the Bill. As things stand, under clause 42 of the Bill, most of the Bill will not come into force until the Secretary of State has brought it into force, with the approval of the UK Parliament. Why not provide that the Bill will only come into force in Wales when and if the Welsh Ministers bring it into force with the approval of the Senedd?”

That is what I seek to do through my amendment.

I can foresee three objections. First, it might be said that the criminal law is a reserved matter and therefore Parliament ought not to refrain from legislating for Wales or have different commencement arrangements for the criminal law in England compared with Wales, but Parliament is sovereign and can do what it wants. By the same reasoning, Parliament should never grant a temporary power to a devolved legislature to legislate on matters that are normally reserved, yet Parliament does do that.

The second objection might be that England and Wales are a single jurisdiction, and that there should not be a difference in the criminal law between them, but this misunderstands that, as a result of the devolution settlement, there are already differences in the criminal law of England and Wales. For example, in our law there is a defence of reasonable chastisement, which means that parents smacking their children would not be committing a criminal offence. Although the criminal law is not devolved to Wales, child welfare is and, using those powers on child welfare, the Senedd voted to abolish the defence of reasonable chastisement in Wales. That means that there is a difference in the criminal law of England and Wales. A position whereby assisting suicide was partially decriminalised in England but fully criminalised in Wales would not be an oddity. There is no principle of our constitution that says that the criminal law must be the same on both sides of the border.

The key issue is that the decriminalisation of assisted suicide, which the Bill will bring about, is not a stand-alone measure. What we are introducing is not something like the Swiss criminal code, for example, which does not criminalise assisted suicide when the person acted for non-selfish motives. Rather, it is a decriminalisation of assisted suicide that is contingent with complying with a detailed healthcare regulatory scheme, which is an issue that is devolved to Wales.

Finally, it might be objected that this might create difficult cross-border issues. As Professor Lewis put it in his blog,

“that would be true also if, for instance, the Scottish Bill did not become law. Far more difficult cross-border issues were successfully accommodated in the context of legislation, also involving profound ethical considerations, when the National Assembly for Wales (as the Senedd was then called) changed the law on organ donations.”

I accept that such a situation might require constitutional amendments at subsequent stages, but these could easily be made by the Government on Report.

Ultimately, when one looks at it in detail, having a different commencement provision for Wales is the only practical way of respecting the vote of the Senedd while not running the risk of creating a legal limbo. Finally, I draw the Committee’s attention to the supplementary written evidence of Professor Lewis, in which he commends my amendment 535.

Regardless of where we stand on assisted dying, we should respect democracy. The people of Wales, through their elected representatives—acting as they should, as representatives—have voted against assisted dying. We should respect that choice and not impose it on them, in whole or in part, without their consent.