Intended for healthcare professionals

Views & Reviews Personal View

Mental capacity as a safeguard in assisted dying: clarity is needed

BMJ 2015; 351 doi: (Published 19 August 2015) Cite this as: BMJ 2015;351:h4461
  1. Annabel Price, consultant psychiatrist and visiting researcher, Department of Psychiatry, University of Cambridge, Cambridge CB2 0SP
  1. ap806{at}

The Marris-Falconer bill includes provisions for assessment of mental capacity in some cases, but Annabel Price calls for more detail on how this would work in practice

The Assisted Dying Bill shortly to be debated in the UK parliament, as with previous bills to legalise physician assisted suicide for terminally ill adults, proposes safeguards to help ensure that vulnerable people do not access it inappropriately.1

A key proposed safeguard is the determination of mental capacity: the bill requires that the person “has the capacity to make the decision to end their own life,” using the framework of the Mental Capacity Act 2005. Compared with some other aspects of the bill this safeguard has received relatively little scrutiny, and the bill as it stands lacks the clarity to guide assessing clinicians on what should be expected of them.

The challenges include knowing where to set the standard to determine capacity (just how much capacity does one need for the decision to request physician assisted suicide?) and what elements should be brought into the decision making process.2 The deceptively simple legal concept of mental capacity does not always map neatly to complex clinical scenarios: determining whether a person (whose capacity is a priori presumed) “does not lack capacity” under the terms of the Mental Capacity Act is very different from determining that a person “has capacity” as proposed in the wording of the bill.

And the difference is not simply semantic; the current wording invites wide clinical interpretation as to where, and with whom, the burden of proof lies. Is it—at one end of the spectrum—for the clinician, having presumed capacity, to need clear evidence that capacity is lacking before judging the person as such; or is it—at the other—for the people making the request to demonstrate to the clinician after extensive assessment that they fully appreciate what they are asking for?

Variation in approach

Linked to the issues above is the further problem of likely variation among individual clinicians’ approaches to assessment. A study in the US state of Oregon (a jurisdiction where physician assisted suicide for terminally ill people has been legal for almost two decades) has shown that standards of competence are likely to be influenced by the individual values of the assessing clinician,3 potentially affecting the assessment’s outcome.

In the study, doctors who considered suicide ethically unacceptable advocated higher standards of competence for assisted suicide than those who found it acceptable in specific (or all) circumstances. In practice, this variation based on individual doctors’ values would be difficult to mitigate unless the assessment were reduced to a standardised “tick box” exercise—one that would be unlikely to exclude any but the most severely depressed or cognitively impaired patients, or those otherwise obviously lacking capacity. It is in the grey areas that capacity is the most difficult to determine, and these grey areas are where clinicians’ individual values are most likely to affect the process.

Depression is strongly associated with a desire for hastened death in terminally ill people, as are the severity of symptoms and a perceived loss of dignity, but the strength of this desire can change over time.4 Psychiatrists have previously been called on to engage in the assisted dying debate, as a source of experience and knowledge in the assessment of mental capacity (including the potential effect of depression on decision making), how patients adapt to adversity (including terminal disease), and how desires can change over time, particularly in relation to suicide.5 6 All of these factors are potentially relevant in assessing a person who requests assisted dying.

Do safeguards work?

How do those jurisdictions that have legalised assisted dying aim to ensure that their safeguards fulfil their intended purpose? Oregon, for example, requires that assessing clinicians report specific data on all patients who have the lethal drug prescribed (whether or not they eventually choose to take it) to a central body, with data collated and reported yearly. These data are often cited as evidence that the process is properly safeguarded.

Weaknesses exist in the dataset, however, because only data relating to patients who have been prescribed the lethal drug are reported; no data are included on patients who requested assisted suicide but were not deemed eligible and why. Without this information, how can we begin to assess the effectiveness of determining capacity as a safeguard?

Few decisions about capacity could be so grave in their consequences as in assisted dying, so greater clarity on this aspect of the bill is imperative, as well as detailed consideration of how its effectiveness as a safeguard will be determined if the bill becomes law.


Cite this as: BMJ 2015;351:h4461


  • Competing interests: I have read and understood the BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; not externally peer reviewed.


View Abstract