Introduction
Is it too late to write a will when a person is dying? A deathbed will is one which is created and executed when the testator is facing imminent death (USLegal, 2012), although what “imminent” means is unclear. Man has been afforded the privilege of disposing of his estate and bestowing it upon whom he chooses at least since the time of the ancient Greeks. If ever this right was championed, it has always been for the dying man, even if he could only give a verbal indication of his wishes (Unknown author, 1834). In this way, will-making has been traditionally considered part of the “ritual preparation for death,” the consequences of not doing so leading to the “horror of dying intestate” (Coppel, Reference Coppel1988). Yet, the obvious association of significant physical and psychological morbidity with dying and the vulnerability of the dying person make deathbed wills often problematic. As early as the 16th century, clerical writers and moralists advised against writing wills “amidst the pains and distractions of a sickbed” rather than setting one's affairs in order while in good health (Coppel, Reference Coppel1988).
Historically, in some jurisdictions, a will produced on the deathbed was automatically dismissible. For example, until 1871 in Scotland a will could be reduced “ex capite lecti” (on the ground of deathbed) if it was written within 60 days of a person's death (M'Laren, Reference M'Laren1894). Yet, there is nothing intrinsic to a deathbed will that makes it invalid, but the circumstances associated with the production of a will in a rushed manner, and the likelihood of a delirious and vulnerable testator may render the will susceptible to challenge. Moreover, there may be inadequate time to get a capacity determination that would help a will withstand such challenge. Furthermore, mistakes can occur in the witnessing or otherwise meeting the statutory requirements for producing a valid will in that jurisdiction. For example, the circumstances of imminent death may force a will to be handwritten rather than printed, which may be invalid in some jurisdictions (Myjourneytomillions, 2012), although most jurisdictions do accept the validity of a holograph (handwritten) will. In the absence of time to reflect and consider upon a “just and careful disposition of lands” (Coppel, Reference Coppel1988), a hastily drawn will may also not provide the intended distribution of property or provide the best protection against taxes even if it is considered valid. Anecdotally, from the experience of taskforce members, these factors make a deathbed will more likely to be successfully contested than a will produced while in good health, although there is no empirical data available to support this observation.
The issue of capacity in these circumstances is clearly crucial. In order to make a valid will, a person must understand the nature of a will and the nature and general extent of any assets; they must appreciate the claims of those who might expect to benefit from the will and understand the impact of the distribution described in the will; and be free of any disorder of mind or delusions that influenced the disposition (Banks v Goodfellow, 1870). While this is testamentary capacity in common law countries, the specific legal criteria for capacity vary in other jurisdictions where testamentary capacity may be defined otherwise or using only part of the Banks v Goodfellow criteria (Shulman et al., Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009).
Equally important in those jurisdictions that recognize the concept, a person writing a will must also be free from undue influence. These tests are particularly relevant if the deathbed will differs substantially from previous wills. Large estates with considerable assets are also more likely to be contested.
To what extent does dying interfere with testamentary capacity and/or render a person vulnerable to undue influence? Moreover, in this context, when does dying begin and how do we define “imminent” death? This paper will address these issues by exploring some of the physical and psychological morbidities associated with dying, psychological understanding of the dying state, and the concept of undue influence as it relates to death and dying.
Methods
Search procedure
We conducted a search in the Medline database (PubMed http:/pubmed.gov/) and Medline (1950–May 2013) with the following terms: “deathbed and wills,” “deathbed and testamentary capacity,” and “dying and testamentary capacity.” The search yielded two papers, one written by our group (Peisah et al., Reference Peisah2009) and one in Norwegian (Anonymous, 1969), which we were unable to translate. A search of the individual terms “testamentary capacity” and “deathbed” yielded 52 and 60 papers, respectively. The abstracts of these papers were reviewed and a single additional paper (Coppel, Reference Coppel1988) was identified as relevant to the present study. A focused selective review was therefore conducted using these papers and related search terms such as “delirium and palliative care.” We present two cases to illustrate the key issues in assessing testamentary capacity of the dying patient.
Results
Physical and medical comorbidities associated with the terminal state
Delirium
Delirium is a disturbance of consciousness and a change in cognition that develops over a short period of time as a consequence of a general medical condition, substance intoxication or withdrawal, use of medication or toxin exposure, not better accounted for by a pre-existing or evolving dementia (American Psychiatric Association, 2000). It is highly prevalent amongst dying patients, not only as a pre-terminal event but also in the last weeks of life, with prevalence rates from 25% to 85% (Massie et al., Reference Massie, Holland and Glass1983; Friedlander et al., Reference Friedlander, Brayman and Breitbart2004) and up to 90% in the hours to days preceding death (Lawlor and Bruera, Reference Lawlor and Bruera2002). It is particularly prevalent amongst elderly patients who are especially vulnerable to developing cognitive disturbance when they are unwell (Massie et al., Reference Massie, Holland and Glass1983).
Yet, delirium per se does not preclude capacity in all cases (Liptzin et al., Reference Liptzin, Peisah, Shulman and Finkel2010). Whether the person has testamentary capacity largely depends on whether the person can satisfy the legal tests for capacity as outlined by Banks v Goodfellow. So while this setting predisposes to compromised cognition, the ultimate question will not be whether or not the person has delirium, but rather, despite this delirium, can they make the particular will in question; the complexity of the task being extremely important to this question. This includes the complexity of the person's estate and their situation (i.e. whether they have family conflict and or multiple potential beneficiaries) (Shulman et al., Reference Shulman, Cohen, Kirsh, Hull and Champine2007; Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009). Any determination of testamentary capacity by a healthcare professional, regardless of the clinical setting, must be structured and systematic, taking into account the person's mental state examination, cognitive function, understanding of their estate and potential beneficiaries, rationale for distribution and reasons for deviation from any past pattern of disposition (Shulman et al., Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009).
The major issue with delirium is that it may be missed by lawyers and lay witnesses who usually rely on the assessments and opinions of healthcare professionals. Yet, delirium is frequently missed by these very healthcare professionals, both in hospital (Inouye et al., Reference Inouye, Foreman, Mion, Katz and Cooney2001; Friedlander et al., Reference Friedlander, Brayman and Breitbart2004) and nursing home environments (Voyer et al., Reference Voyer, Richard, Doucet, Danjou and Carmichael2008), where less then 20% of patients with delirium are detected. Older patients (Voyer et al., Reference Voyer, Richard, Doucet, Danjou and Carmichael2008) and those with pre-existing cognitive impairment or dementia (Fick and Foreman, Reference Fick and Foreman2000) are more likely to have undetected delirium. Hypoactive presentations of delirium, characterized by less activity and more withdrawal and decreased speech (Meagher et al., Reference Meagher2008), are prevalent and frequently missed. For example, in a study of 100 consecutive cases of delirium in a palliative care unit, Leonard et al. (Reference Leonard, Donnelly, Conroy, Trzepacz and Meagher2011) found that 33% of patients were classified as hypoactive and these patients had the same impairment in cognitive functioning as patients with other motor variants of delirium, with similar deficits in orientation, memory, and comprehension on cognitive test scores. Other explanations for the poor detection of delirium include lack of formal and sensitive cognitive testing, the characteristic fluctuation in symptoms over time, and lack of comparison with premorbid cognitive status (Meagher and Leonard, Reference Meagher and Leonard2008).
General symptomatology associated with dying
The physical and cognitive symptoms and signs commonly experienced by people who are dying, and the treatment of such, may have an impact on decision-making capacity. Consciousness and ability to communicate is often reduced in the pre-terminal state, clearly more so with closer proximity to death (Lynn et al., Reference Lynn1997). Patients who are dying often suffer fatigue, may be in severe pain (and on pain medication such as opioids that can cloud cognition), and/or suffering from other distressing symptoms such as shortness of breath, nausea, vomiting, and itch (Lynn et al., Reference Lynn1997; von Gunten, Reference von Gunten2005), depending on the nature of the terminal illness. They may also be on other psychoactive medication, including benzodiazepines, which cause cognitive impairment and changes in alertness. Benzodiazepines may be used in up to 58% of palliative care inpatients in the last three weeks of life (Henderson et al., Reference Henderson, MacGregor, Sykes and Hotopf2006).
Mood symptoms are also commonly present in dying patients and may also affect decision-making capacity due to negative thoughts related to apathy, guilt, worthiness, and poverty. At the very least, patients may have symptoms of dysphoria and a sense of isolation, if not frank depression and anxiety (Lynn et al., Reference Lynn1997; O'Connor et al., Reference O'Connor, White, Kristjanson, Cousins and Wilkes2010). To complicate matters, neurovegetative symptoms (e.g anergia, anorexia, weight loss, sleep or psychomotor disturbance) that are common during the dying process, overlap with the symptoms of depression and delirium making the diagnosis of these psychiatric syndromes more difficult. Hence, the need for a psychiatric interview and/or alternative criteria for diagnosing depression in dying patients. In the case of medically ill patients, it has been suggested that the somatic symptoms of depression be replaced by particular psychological and mental state features such as tearfulness, depressed appearance, brooding, self-pity, social withdrawal and reduced talkativeness, and lack of reactivity in order to make the diagnosis (Endicott, Reference Endicott1984; O'Connor et al., Reference O'Connor, White, Kristjanson, Cousins and Wilkes2010).
Undue influence
Traditionally, in common law, the concept of undue influence in will-making has relied upon a severity threshold such that an element of coercion by others has had to be present (Peisah et al., Reference Peisah2009). However, over a century ago, the English judiciary recognized that in a deathbed setting, the testator's vulnerability may be so great that they require “very little pressure” to be coerced. In Wingrove v Wingrove (1885) LR11PD 81at 82–83, Sir James Hannen elaborated:
The coercion may of course be of different kinds, it may be in the grossest form, such as actual confinement or violence, or a person in the last days or hours of life may have become so weak and feeble that very little pressure will be sufficient to bring about the desired result. . ..
We have previously identified the deathbed setting as an opportunity for undue influence to be exerted due to the high prevalence of delirium, and the highly medicalized, acute-care setting which encourages regression and dependency (Peisah et al., Reference Peisah2009). The more frail and ill the testator the less influence it would take to be considered ‘undue’ (Shulman et al., Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009). Coppel (Reference Coppel1988) has emphasized the difficulties encountered by the terminally ill testator subjected to “scenes of importunate relatives pressing around the deathbed.” Nursing and medical staff, who are often the gatekeepers of access to terminally ill testators may unwittingly facilitate or collude with the exploitation of an impaired testator merely by virtue of their failure to identify delirium. Suffice it to say that the dying patient is very susceptible to undue influence. However, it is solely for the court to determine if such influence was exerted.
Conversely, the failure to write a will at all may also render one, or at least one's estate, vulnerable to “ill-usage” by others, as suggested by Jeremy Bentham, an early 19th century philosopher and legal reformer (cited in Bentham, Reference Bentham1843). At the very least, in the absence of making one's wishes known, we abdicate choices about disposition to formulaic prescriptions determined by the rules of the court or family provision determinations of the relevant jurisdiction. As such, different “voices” – other than the testator – will be heard in determining what is fair and just in terms of disposition, which may not necessarily accord with how the testator viewed these issues in life.
Psychological understanding of the dying state
Dying often prompts the individual to seek closure or to resolve unfinished business or unresolved hurts. Dying testators may seek restitution or forgiveness or alternatively, revenge and justice. Projection and displacement of fear and anger may be expressed in hostility and resentment toward others such as family members, or “substitute” fears about money (Culkin, Reference Culkin2002). These fears may find their expression in the way property is disposed in the final testament. Should a person's disposition be tainted by such distortions? Conversely, the act of writing a will has the potential to resolve these feelings and offer the testator peace of mind.
Why do people wait until their deathbed to write their will?
Coppel (Reference Coppel1988) explored historical accounts of why testators postponed the act of will-making. He noted that some testators deferred the act in the hope of a longer life or accumulating further riches, others waited for their wives to die first to give them more freedom in disposition, while others still were fearful that the very act of writing a will would hasten their death. Will-making has traditionally been associated with “black foreboding” or superstition and in some cultures this is still the case. In depressed suicidal patients, the act of writing a will is suggestive of suicidal intent and thus proximal to death for that reason, warranting careful scrutiny of testamentary capacity in such cases.
Most often, will-making on the deathbed is instigated by someone other than the testator seeking to gain from the disposition. The very act of a beneficiary being involved in the procurement of the will is a classic “indicia” of undue influence (Spar and Garb, Reference Spar and Garb1992; Frolik, Reference Frolik2001).
Solutions
The identification and management of delirium – empowering the dying
Maintaining mental awareness for as long as possible – ostensibly to interact with and enjoy the company of loved ones, but possibly also to execute important decisions such as the disposition of property – is a priority for dying patients (Steinhauser et al., Reference Steinhauser, Christakis, Clipp, McNeilly, McIntyre and Tulsky2000; Gawande, Reference Gawande2010). In order to achieve this goal, delirium needs to be identified using simple, non-intrusive routine screening such as that adopted in settings such as the Milford Hospice, Ireland (Leonard et al., Reference Leonard, Donnelly, Conroy, Trzepacz and Meagher2011), where it has been recognized that many delirium episodes may be at least partially reversible with relatively low-burden interventions (Leonard et al., Reference Leonard, Agar, Mason and Lawlor2008; Reference Leonard, Donnelly, Conroy, Trzepacz and Meagher2011). Treating delirium might not only improve cognition, and potentially, the capacity to make legal decisions such as executing a will, but also improve quality of life and reduce distress in patients and their family in the terminal phase of illness (Steinhauser et al., Reference Steinhauser, Christakis, Clipp, McNeilly, McIntyre and Tulsky2000).
Proper procedures for the assessment of testamentary capacity
While in common law there is presumption of capacity, the very existence of the dying state may be sufficient to reverse or rebut the presumption and act as a “trigger” to justify an assessment of capacity, to prove it in the affirmative. There may be sufficient doubt about capacity in this setting to recommend that all dying patients require a formal assessment of testamentary capacity prior to making a will, hence a formal consultation between healthcare professionals and lawyer, preferably initiated by the lawyer. Healthcare professionals who become involved in such assessments should demonstrate an understanding of the principles underlying mental capacity and knowledge of the specific test for testamentary capacity.
There is great variability and little guidance as to appropriate actions in situations where healthcare staff become aware of lawyers taking instructions from obviously impaired patients. At one extreme, some healthcare systems advise “minimal involvement” in will-making. For example, Australian guidelines have suggested that the healthcare professionals should determine whether there is already a will in existence if a patient requests to make one and to refer the matter back to the solicitor holding the will or, in the absence of such, ask the patient to contact a solicitor of their choice. Assessment of testamentary capacity by public hospital staff is discouraged unless specifically requested by the solicitor handling the matter (New South Wales Health, 2005). In contrast, in other jurisdictions, a process such as a lawyer taking instructions from an obviously impaired patient might be construed as fiduciary or financial abuse, and thus compel staff to report under mandatory elderly people abuse reporting laws, failing to do so potentially incurring penalty (e.g. California Welfare and Institutions Code Section 15630).
Assessment of testamentary capacity
Healthcare professionals may be consulted to give expert opinion regarding Testamentary Capacity either contemporaneously, in the deathbed setting, or retrospectively, when there is challenge to the will after the testator is deceased. Regardless, we recommend the same structured methodology of assessing both the testator's mental status and their ability to meet the task-specific aspects of testamentary capacity, guided by the Banks v Goodfellow criteria. This methodology involves an assessment of (i) the testator's understanding of the nature and extent of their property; (ii) awareness of potential beneficiaries and the testator's ability to evaluate and discriminate between the claims of such beneficiaries; (iii) the testator's rationale for deviating from any pattern of disposition identified in previous wills or wishes regarding testamentary intent; (iv) the presence of any disorder of mind such as delusions or hallucinations which might be influencing the testator's disposition, and (v) ensuring the will-making is a free and voluntary act (Shulman et al., Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009).
In addition to the mandatory review of past wills, medical records, depositions of fact witnesses, financial records, and relevant correspondence (Shulman et al., Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009), much of the retrospective assessment of testamentary capacity in this setting relies on the accuracy of staff's descriptions of mental state and the lawyer's documentation of the task-specific aspects of capacity at the time instructions were taken. In interpreting this evidence, one must be mindful that there is a different agenda for detecting delirium in the palliative or intensive care setting, where the focus is on noting changes in clinical state using tools which grossly describe consciousness and mental state such as Glasgow Coma scale, whereas for the purposes of assessing testamentary capacity, we are interested in identifying subtle perturbations of mental state which might be incompatible with testamentary capacity, which is often reliant on complex decision-making and higher levels of cognition.
Thus, descriptions of patients as “lucid” in progress notes may say very little about the presence or absence of any disorder of mind relevant to testamentary capacity, other than the absence of any gross abnormality. We have previously discussed the legal concept of “lucid intervals” as they pertain to delirium, where, theoretically, a person may be capable of signing a will during such an interval if their cognition has substantially improved and there is a clear rationale and consistency over time (hours, days, weeks, or months) in the person's expressed wishes (Liptzin et al., Reference Liptzin, Peisah, Shulman and Finkel2010). However, “lucidity” is a relative concept relating to severity of the delirium at the time the will was made, and the issue is whether the testator, with their particular cognitive function and their particular situational complexity, can or could make this particular will, at the relevant time (Shulman et al., Reference Shulman, Peisah, Jacoby, Heinik and Finkel2009).
Finally, reports of “alert and orientated” do not preclude delirium. First, such entries are often based on the person's general demeanor or awareness that they are in hospital, rather than on any specific testing or screening. Such interactions so documented between staff and patients may be brief and superficial and insufficient to detect thought disorder, problems with comprehension, and attention indicative of delirium (McCartney and Palmateer, Reference McCartney and Palmateer1985). Second, while disorientation may be commonly a feature of delirium, it is not invariably so (Meagher et al., Reference Meagher2007). Indeed, disorientation is often used to screen for cognitive impairment in clinical settings, but may be unhelpful due to both the fluctuating nature of delirium and its lack of sensitivity for the syndrome (Meagher et al., Reference Meagher2007; Gupta et al., Reference Gupta, de Jonghe, Schieveld, Leonard and Meagher2008). Disorientation has been noted in only two-thirds of delirious patients in palliative care settings, compared to inattention, which is almost universal (present in 96%, Leonard et al., Reference Leonard, Donnelly, Conroy, Trzepacz and Meagher2011). The assessment of attention is fundamental to diagnosing delirium, yet doctors and nurses have difficulty correctly identifying inattention (Lemiengre et al., Reference Lemiengre2006) including in palliative care settings (Ryan et al., Reference Ryan2008).
In view of the high likelihood of the presence of a disorder of mind such as delirium in this setting, the “mother lode” of evidence pointing to testamentary capacity is the testator's contemporaneous understanding and reasoning articulated in their own words, ideally in response to open-ended questions about the task-specific aspects of capacity. Affirmative responses or accession to the closed questioning of “Is this what you want?” “Do you understand?” do not necessarily reflect understanding. This is particularly important with a complex estate or distribution or when a testator deviates from previous wills or expressed wishes, when a higher threshold might be used for assessing capacity than, say for an elderly man leaving his one asset, his house, to his wife of 50 years, with whom there has been no conflict. Structured questions offering choices may be necessary in a very frail individual but should still ensure that the testator's responses reflect a consistent wish and clear rationale.
Accordingly, we offer an approach to the assessment of capacity in a delirious dying patient with different levels of questioning and response depending on the circumstances. Thus, the disposition of a simple estate consistent with previously expressed wishes or wills may only require a consistent passive response by the testator to various choices offered. This should ideally be done more than once over a time interval even if it is only a short period to demonstrate the requirement for stability (and consistency) in decision-making. Alternatively, the disposition of a more complex and substantial estate that deviates from previously expressed wishes requires a higher level of sophistication of understanding with consistent rationale. If there is a change in the pattern of disposition, then some rationale for this change should be provided. The vital question to ask the testator is “why?” It is not sufficient to simply document that the testator was emphatic or “clear” in their wishes to disinherit or favor a beneficiary – often assumed to be synonymous with capacity despite the fact that clarity or emphasis may reflect cognitive impairment or psychotic thinking.
Case example 1
In Wharton v Bancroft and Others [2011](United Kingdom), millionaire George Wharton (78) had lived with Maureen (63) for 32 years. When diagnosed with cancer, his tax advisers suggested marrying Maureen; he said he would not think of marrying again unless he knew he was close to death.
When he left hospital, to allow him to die at home, he made a will in expectation of his marriage to Maureen and, later that day, he married her at home. His will left everything to Maureen and excluded his three daughters who would have inherited everything had he died unmarried and intestate. The three daughters challenged the will on the basis of Maureen's “undue influence,” but their challenge failed. The court determined that their father was a man who knew what he was doing, why he was doing it, and what he wanted to do. The court felt unable to infer anything untoward in a man wishing solely to benefit his partner of 32 years, his wife in all but law. This case exemplifies the importance of examining the consistency of the person's testamentary act with their previous wishes or behavior.
Case example 2 (unheard)
Mr K was a 76-year-old widower with no children. He had previously written two wills leaving the bulk of his estate to his two brothers with nominal amounts to various friends. He had a history of heavy alcohol use escalating in recent years with associated self-neglect. An old friend, Mrs J, offered him assistance in cleaning and subsequently moved in to provide more intensive support in exchange for free board. Mr K was admitted precipitously to hospital on the 1st February 2008 following a fall and was noted to have hepato-renal failure with marked ascites. Over the next week, he deteriorated and was noted to be “drowsy” and “confused” at times, while at other times he was described as “lucid,” “alert,” or “no encephalopathy.” On the 7th February, Mrs J rang her solicitor and told him that her friend, Mr K, was very sick and wanted to make a will leaving her the bulk of his estate. The solicitor drew up a will accordingly and visited Mr K in the intensive care unit on the 8th February whereupon he asked Mr K if this is how he wanted his estate distributed. Mr K replied that Mrs J had been very good to him and he executed the will on that day. He died on the 16th February, 2008. Contrary to the previous case, this case exemplifies a scenario of a precipitous will change procured by a beneficiary, and associated with deviation from a previous will pattern without rationale or apparent awareness of the previous wills or consideration of the claims of other potential beneficiaries. This suggests that this testator probably lacked testamentary capacity and even if found to have testamentary capacity, was probably vulnerable to undue influence.
Conclusion
The “deathbed will” is but one example of precipitous decision-making and execution of legal documents that often occurs in a hospital environment where many factors collide to render the decision-maker, the testator, vulnerable to both impaired capacity and undue influence. Even a delirious dying patient could still have testamentary capacity if certain conditions apply, and these conditions are documented at the time the will is made. However, there has been no empirical investigation or gathering of data in relation to deathbed wills, and there is scant discourse in the scientific literature about this, which should be addressed. In the current environment of an internationally promulgated United Nations Convention of the Rights of People with Disabilities, where respect for individual autonomy and the freedom from undue influence and abuse are paramount, the trick is to ensure that testators are encouraged, and perhaps facilitated, to make the wills they are capable of making, while being protected from making the wills they are not capable of making, either because of lack of testamentary capacity or vulnerability to undue influence.
Conflict of interest
None.
Description of authors’ roles
All authors contributed to the writing of the paper.
