Public health, populations, and lethal ingestion

Hypotheses 

Methods 

Discursive analysis of

Introduction 

Public health's purpose is to advance population health. When a policy is adopted in its name, it is reasonable to query links between policy and population perspectives. This applies also to the APHA's 2008 endorsement of physicians writing lethal prescriptions to qualified patients as a general public health policy [1].

Population perspectives are generally not emphasized in controversial issues such as induced death or abortion (excepting aggregate maternal health, eugenics, population control), while advocacy accounts often frame issues through personal narrative invoking autonomy [6]. While not insensitive to individual sagas, public health examines specific cases primarily to illuminate population implications. Its ethics frequently subordinate individual autonomy to population interest. What population-health interests are implied by availing life-ending means to specific populations? Are there externalities for other population segments or impacts on public health practice standards? Will population discourse increasingly shape such issues?

Terms 

Naming is power. Terminology concerning induced death is contested. “Assisted suicide” has been interrogated regarding whether it is logically coherent as a term and for medicine as a practice [7]. In Switzerland since 1937 assisted suicide need not be mediated through medical personnel to avoid prosecution; anyone unmotivated by “selfish interest” may facilitate (Swiss Penal Code, SR 311, Article 110, Vorsätzliche Tötung, Intentional Killing) [8], [9].

“Assisted suicide” yet predominates a recent PubMed search of related terms1 notwithstanding that “assisted suicide” and “physician-assisted suicide” have been statutorily, if not logically, defined out of existence in Oregon and Washington [“Actions taken in accordance with ORS 127.800 to 127.897 shall not, for any purpose, constitute suicide, assisted suicide, mercy killing or homicide, under the law,” Oregon Revised Statutes (ORS) 127.880 §3.14; similarly Revised Code of Washington (RCW) 70.245.180] [10]. Oregon's Public Health Division continued using the term over a decade until prevailed upon by the proponent organization Compassion and Choices in 2006. Precision of scope and etymology notwithstanding, market research indicated the term did not inspire public support of the policy [11], [12]. Garrow notes the presence or absence of “suicide” or “assisted suicide” can generate a 15% to 20% difference in poll responses [13].

Instead of direct references to life-ending acts, the Oregon Public Health Division now uses “death with dignity.” Hemlock Society co-founder Derek Humphreys objected: “The department's cop-out choice of words, ‘death with dignity,’ is wildly ambiguous and means anything you want. Let's stick to the English language and in this matter call a spade a spade” [14]. Still, “PAS” remains embedded like a palimpsest in the Death With Dignity Act's URL (egov.oregon.gov/DHS/ph/pas/index.shtml).

Excepting death certificates, Washington's Measure I-1000 (codified as RCW 70.245) prescribes relatively clear descriptive language for ingesting a prescribed, lethal pharmakon2 [15]: “state reports shall refer to practice under this chapter as obtaining and self-administering life-ending medication” [16].

“Physician-assisted Death” (PAD) is used by the American Academy of Hospice and Palliative Medicine (AAHPM) “with the belief that it captures the essence of the process in a more accurately descriptive fashion than the more emotionally charged designation physician-assisted suicide” [17], while its URL references suicide (www.aahpm.org/positions/suicide.html). “Physician-assisted death,” however, semantically shifts focus to the physician as (if sole) agent. Less focused than “physician-assisted suicide,” it would cover direct euthanasia, voluntary or nonvoluntary, without revision.

“Death with dignity” has been promoted by proponents to embed lethal ingestion into vaguer semantic and broader value fields. ODWDA reifies it to a terminus technicus [2]. Persons dying under it are “persons who use the Death with Dignity Act” [12].

The Oxford English Dictionary's (OED's) first definition of dignity is “the quality of being worthy or honourable; worthiness, worth, nobleness, excellence” [18]. Asa terminus technicus, “death with dignity” commandeers a broad value range even as it narrows it. Other deaths are not “death with dignity”—dedignified [19]. Ironically, a fractured sense of dignity is predictive of a desire for hastened death in terminally ill persons (given appearance concerns, sense of burden to others, negative self-perception/perception or projection of others' perception of oneself, institutional setting, incontinence, or pain intensity, absent significant differences in median survival) [20].

Population health and early exits 

Public health's business is “Preventing Illness/Promoting Health/One Population at a Time” (increasing mean health, narrowing disparities) [21]. Policies are to prudentially maximize a conception of population health. What counts as health is key given public health's commitment to promote it.

The World Health Organization (WHO)'s definition is asymptotic: “Health is a complete state of physical, mental and social well-being, and not merely the absence of disease or infirmity” [22]. It has been criticized as missing dimensions, unattainable, and nonaccommodating regarding functional adaptation [23]. The International Covenant on Economic, Social and Cultural Rights states everyone has a right to “the highest attainable standard of physical and mental health” [24]. The Universal Declaration of Human Rights Article 25 mediates health through a family-economic/safety-net perspective with right of security for conditions beyond one's control [25]. These articulations denominate health primarily through individual and family perspectives (with a social augment) rather than by population. Could induced deaths be conceived to advance population health, drawing in public health through its own logic?

Population health comprises mortality and morbidity. One disease burden measure is the DALY (disability adjusted life year), combining mortality and morbidity into a single loss function against a maximum life expectancy (Japan's) [26], [27]. Losses are considered “bads” generally to be minimized [28]. Lost life years seem straightforward, but years count unequally in DALY calculations: early and late years are discounted due to social dependency discounting and role-value judgments [26]. Mid-years are weighted greater than 1.0. For population tradeoffs between late and mid-years, mid-years are judged more valuable. Resource transfers that support fewer deaths or lower morbidity for years weighted greater than 1.0, with earlier “exits” in late dependency (years weighted less than 1.0), particularly given high morbidity components (disability class weight × time), would generate aggregate positive population health offsets measured by DALYs. Substantial equity issues arise: at every age, extending the life of person with a durative impairment is worth less under DALYs than extending the life of a healthier same-aged individual, ceteris paribus [28].

QALYs (quality adjusted life years) equal time multiplied by a health-related quality-of-life weight (0 … 1; optimal=1). QALYs employ a restrictive health definition (contrasting with that of WHO) based on disease or infirmity [29]. States considered “worse than death” may receive a negative weight, lessening the aggregate. A marginal preference for induced death over dying from an underlying disease process implies a negative weight. Earlier “exits” increase population health measured by aggregated QALYs (albeit “better off dead” does not imply better off killed [30]).

Not every model of induced death would necessarily produce a net population health-status gain under such measures: An unfettered right to assisted suicide (Swiss “dignitas”) could eliminate relatively high health status individuals.

Disciplines, distance, and public health endorsements of lethal ingestion 

WSPHA and APHA lethal ingestion policies contradict more proximate health professional organizations, including the American Medical Association [31], American Geriatrics Society (AGS) [32], and American Academy of Hospice and Palliative Care [17]. Why, then, public health associations?

Support for lethal ingestion among health practitioners and associations tends to decrease as contact with individuals who are at life's end increases [33]. The AGS, whose population has increasing vulnerabilities, opposes intentionally providing lethal means (information, pharmaceutics, mechanical), arguing legalization undermines patient trust [32], [34]. The AGS notes supervening cultural messages framing the dependent aged as a disutility, which could “open the door to abuse of the frail, disabled, and economically disadvantaged of society, by encouraging them to accept death prematurely rather than to burden society and family” [32].3 Implied is structural violence against the aged-infirm: socially mediated values and structures internalized to their own detriment, deprecating their worthiness to continue living (internalized ageism and ableism similar to internalized racism [35]).

Acknowledging diverse views in its ranks, the AAHPM “takes a position of ‘studied neutrality’ on whether PAD should be legally regulated or prohibited. Its members should instead continue to strive to find the proper response to those patients whose suffering becomes intolerable despite the best possible palliative care” [17]. If requested to hasten death, the AAHPM “strongly recommends that medical practitioners carefully scrutinize the sources of fear and suffering leading to the request with the goal of addressing these sources without hastening death.”

The AMA Code of Ethics states “[p]hysician-assisted suicide is fundamentally incompatible with the physician's role as healer, would be difficult or impossible to control, and would pose serious societal risks” (June, 1996) [31]. Control will be a point of debate given Oregon (1996), Washington (2009–), possibly Montana, and “wild” practices (ever) pushing limits [17], [36].

Also noting diverse views, opposite the AAHPM, the nonspecialized American Medical Women's Association Board endorsed “aid in dying” (September 2007), its homepage now linking to Compassion and Choices [37]. The American College of Legal Medicine (physician lawyers) initially jettisoned “physician-assisted suicide” for “treatment intended to end life,” later endorsing “a process intended to hasten the end of life” (October 2008) [38]. The American Medical Students Association, least experienced with patients, endorsed “physician aid in dying” (January 2008) for “current standards” of terminal illness, subjective judgment of unbearable suffering (including emotional), but with steeper standards than Oregon: compulsory psychiatric evaluation in every case; a lasting patient-physician relationship; and publicly accessible disidentified data [39].

Critical care nurses spend more daily time with patients in extremis and their families than do physicians. A 2001 American Association of Critical-Care Nurses survey (906 of 3000) revealed 37% had been asked to hasten a patient's death. Nearly 100% supported palliative care, withholding or withdrawing life support in specific circumstances, but overwhelmingly rejected assisted suicide, voluntary and nonvoluntary euthanasia (83%; 95%; 89% to 98%) [40].

The British Medical Association first opposed PAS, went neutral (July 2005) [41], and then opposed again (2006 members' vote) [42], recently rejecting suicide tourism/exoneration (July 2009) [43]. A 2009 doctors' survey in Wales and England (random sample of 1000, 50% response) followed a defeated House of Lords Assisted Dying for the Terminally Ill bill (May 2006) [5]. Across specialties, 49% opposed changing the law; 39% were in favor. Variables predicting opposition included time caring for dying people, religious belief, and reading at all in the bill. Those never caring for a dying patient most supported PAS (66%)—5 times more than those daily providing care. And 52%, 58%, and 60% of those caring for the dying monthly, weekly, or daily opposed legalization, increasing with experiential intensity. The study noted 93% opposed in a 2003 Association for Palliative Medicine member survey.

While Washington State's medical association strongly opposed Measure I-1000 [44], its public health association board unanimously endorsed it [45], [46]. Proximate oppose; most distant endorse.

Edging toward future policy? The 2005 APHA Amicus Curiae brief in Gonzales v. Oregon 

In July, 2005, APHA submitted an amicus curiae brief supporting respondents in Gonzales v. Oregon [546 U.S. 243 (2006)] [47]. At issue: whether the Attorney General could invoke the Controlled Substances Act [CSA, 21 U.S.C. § 812(b)(2)] to prohibit lethal prescribing of Schedule II drugs as beyond legitimate medical purposes for CSA listed entities (not alternatives). The majority opinion rejected so using the CSA [48].

Under “Interests of Amicus Curiae,” APHA positions itself as the world's preeminent public health association:

Founded in 1872, the American Public Health Association (APHA) is the oldest, largest and most diverse organization of public health professionals in the world. The association aims to protect all Americans and their communities from preventable, serious health threats and strives to assure community-based health promotion and disease prevention activities and preventive health services are universally accessible in the United States.

Stated prominently at the outset: “APHA does not take a position on whether Oregon's law is wise policy.”

The brief cites APHA's 1981 policy “Death with Dignity” (for right to refuse treatment): “to the seriously ill and infirm, death is not only a distinct possibility, but sometimes preferable to any alternative” and that health care policy “should not blindly stress the continuation of life” [49].

The brief argues against the attorney general's directive, defending federalism: states' health powers (licensing, practice standards) and public health federalism (“Public health is served by legal principles allowing experimentation and diversity at the state level,” at ii); also arguing separation of powers (legislative, executive) and sphere distinctions within the executive (Health and Human Services, Department of Justice).

Surprisingly, it states CSA concerns “apply with less force in [O]DWDA prescriptions than in other instances where physicians prescribe Schedule II drugs in professional practice” (at 8) despite probable consequences of unintentional exposure or abuse (minors in the medicine cabinet).

The relationship to public health emerges as the brief proceeds: CSA authority obtains “without compromising States' power to regulate medical practice, or terminating debate and experimentation in the field of public health” (at 10); “Public Health Is Served by Legal Principles Allowing Experimentation and Diversity at the State Level” (heading II.A at 14); Oregon's law constitutes “A Legitimate and Important Policy ‘Experiment'” (heading II.B at 17); for plural approaches “where disagreement has significant moral as well as empirical dimensions” (at 17); invoking “novel social ∗∗∗ experiments” from Justice Brandeis' dissent in New State Ice Co. v. Liebmann (text: “[i]t is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country” [285 U.S. 262 (1932) at 311]).

This laboratory “has catalyzed deliberation and action elsewhere,” “informed the public discussion of the subject and promoted the reasoned development of public health policy” (at 17), but without citing 7 states proscribing PAS with 25 others rejecting 80+ legalization bills or referenda [50]. The Attorney General's Directive may have a chilling effect on legalization “denying patients the formal safeguards and public scrutiny that formal regimes like Oregon's provide” (note 14 at 26).

It notes Oregon's recordkeeping requirements approvingly, “issuing detailed annual reports,” without disproportionate impact on vulnerable groups, while overlooking Oregon's practice of destroying (not disidentifying) primary data after aggregation for “confidentiality” [51], blocking secondary review, detailed longitudinal research, or forensic reconstruction.

The brief identifies neglect of palliative training and undertreatment of pain as pervasive public health problems (at 27), with Oregon's implementation not adversely affecting other care for the dying (increased palliative training; depression recognition; higher hospice enrollment, from 2001). Unnoticed were important 2004 Oregon studies finding increased family-based reports of moderate to severe pain in the final 2 weeks of life following ODWDA, up from 38% to 54% (1996–1997, 2000–2002), with severe pain nearly doubling (11% to 20%) [52]; and increased hospital-based opioid prescribing not accruing to dying patients (1997–1999) [53].

Regarding safeguards and depression, absent were ODWDA's precipitous drop in psychiatric referrals (31.3%, 43.5%, 20.0%, 14.3%, 13.2%, 4.8%, 5.4%; trend: p < .001) [54] and a 1996 study of Oregon psychiatrists indicating 6% were “very confident” of adequately assessing in 1 evaluation whether a possible psychiatric disorder impaired judgment in patients requesting a lethal prescription; 51% were “not at all confident” and 43% only “somewhat”[55].

Finally, the attorney general's directive could potentially impact any nontherapuetic or experimental scheduled drug use, skewing the “process for resolving difficult problems of policy and public morality” by administrative “stay of experimentation” beyond state or Congressional debate and was thus “inimical to the public health” (at 30–31).

While not taking a position on Oregon's wisdom, the brief is open to and identifies such policy as “public health experimentation.”

2007–2008 APHA hearings and policy adoption 

In 2007 and 2008, the Compassion and Choices Director of Legal Affairs submitted resolutions [56], [57] proposing APHA endorse lethal prescription for terminal, competent, qualified patients, citing ODWDA and its implementation as warrant.

Hearings were held at the APHA annual meetings before Joint Policy Committee Group D, Public Health Science, and Infrastructure. Proponents and opponents presented on the texts and context for themselves or representing a Section, Caucus, or Special Primary Interest Group (the author presented opposing [58], [59]).

The 2007 resolution, “Death with Dignity” (D2/D8) [56], was rejected. Recast in 2008 as “Patient's Rights to Self-Determination at End of Life” (D1) [57], it was substantially adopted by the APHA Governing Council. Selecting evidence, both included elements of import for public health infrastructure: framing policy for systems affecting large populations.

The 2007 proposal framed recommendations under a claim: “A national long-term care policy would be grossly inadequate without recognizing a patient's right to die” (D2-7). Long-term care is a “[r]ange of medical and/or social services designed to help people who have disabilities or chronic care needs” (U.S. Department of Health and Human Services) [60] not necessarily implying terminal [42 U.S.C. § 1395x(dd)(3)(A)]. Such a right to die would reframe assumptions regarding continued living for nonterminal individuals.

The “national long-term care policy” sentence derived from APHA's 1981 “Death with Dignity” policy [49] citing Barnard's Good Life, Good Death: A Doctor's Case for Euthanasia and Suicide (1980) [61]. Without using the term “long-term care” or discussing “long-term care policy” per se, Barnard's right-to-die horizon spanned treatment refusal and cessation, active euthanasia, and voluntary suicide at any time. On its face the 1981 policy advocates a right to refuse treatment; advanced directives; and ethics committees for enforcement. Its final recommendation allowed for expansion: “Supports further study to better delineate the ethical, legal, and medical issues involved in the concept of the right to die” [49].

The 2007 resolution preemptively claimed “APHA recognizes that Oregon's experience with aid in dying is a model for other states” (D2-1), citing California Public Health Association, North's endorsing a defeated Oregon-like bill (2007, AB 374). Unmentioned, the larger Southern California Public Health Association did not endorse the bill, while the APHA Gonzales v. Oregon position brief (2005) stated APHA did not take a position on Oregon's policy [47].

Like the brief, it cites increased hospice referral and physician attempts to improve palliative care knowledge, omitting the 2004 studies on more opioid prescribing not accruing to dying patients [53]; significantly higher reported pain after ODWDA [52]; and decreased hospice nursing visits (by 2001 12.7 from 16.0) [62].

Recommendations support implementing advanced directives laws; (again) “further study to better delineate the ethical, legal and medical issues involved in the concept of the right to die;” strong informed consent protections; culturally competent end-of-life option communication; nondiscriminatory pain management/palliative care; and passing laws “to allow control over the time, place and manner of his or her impending death” [56]. In 2008, the 2007 resolution substantially returned as D1, “Patients' Rights to Self-Determination at End of Life.” This APHA adopted almost entirely, archiving the 1981 “Death with Dignity” policy, which referred to treatment refusal, but in D1 it refers solely to lethal ingestion—abandoning the former scope as it extends.

The term's new semantics and syntax strain and limit: “Many Americans believe that the option of death with dignity should be open to those facing a terminal illness marked by extreme suffering.” ODWDA is “The Dignity Act.” A good death “may include dying quietly and with dignity, also being pain free, and without distress,” denying the predicate “dignity” to other contexts (dignity despite distress noted by medicus Freidrich Schiller (1793) [63]).

The adopted policy strikingly mischaracterizes APHA's own Gonzales v. Oregon brief, uncritically repeating language from the Compassion and Choices resolution (D1-8, note 4) [57]: “APHA filed an amicus brief in the case, supporting the [Oregon] law,” contradicting APHA's stated neutrality [47], citing misinformation as precedent.

The new policy cites 54 sources, 51 from D1 [omitting one questioning Oregon's data (D1-12, note 22 [64])]. Absent in 2007 but alluded to in the 2008 resolution but without title or journal (D1-13, note 23), the policy fully cites the study on increased reported pain [52] as suggesting “that high-quality of end-of-life palliative care has actually been reduced” [1].

The text derives almost entirely from the draft with few exceptions: briefly discussing subjective dimensions of a “good death” (citing the sole non-D1 sources, 2–4); noting “no evidence” of disproportionate, vulnerable-population impacts may be due to lacking data collection; language recommending rejecting terms such as suicide and assisted suicide derived from an unreferenced, now archived, 2006 APHA policy [65]; and recommending a moratorium should vulnerable populations be disproportionately impacted. A restrictive D1 amendment (June 2008) excluding persons with pre-existing disabilities prior to terminal illness (D1–19) was rejected, while calling for data collection on such hastening their deaths.

Neither D1 nor the new policy note there being no controls on access after filling the prescription; on who administers it; competence at ingestion; or independent witnesses. One request witness may be an heir, the other a health care facility designee (ORS 127.810 s.2.02; also Washington RCW 70.245.030) [66].

Neither cite evidence of abuse documented outside annual reports [67] [68, distributed], nor the well-publicized 2005 “assisted suicide” case where nurses directly administered morphine and phenobarbitol (i.e., active euthanasia) without informing the physician, suppressed over 2 years by the Oregon State Board of Nursing [69], [70]. The Oregon Department of Administrative Services Report concluded:

In our view, its deliberations and practices appear to lack an acute sense of urgency to protect the public, as months, sometimes years, may pass before action is taken against a licensee, potential criminal behavior is not always quickly reported […] [71].

References discuss distinctions between a nonterminal person's suicide and a terminal patient's with the caveat “whose judgments are not impaired by mental disorders.” Omitted again are plummeting psychiatric/psychological referrals [53]; depression being predictive of strong desire for hastened death among the ill [72] and strongest group identifier in Oregon cases [73]; or specifically that Oregon regulations “may fail to protect some patients whose choices are influenced by depression from receiving a prescription for a lethal drug” (2007) [74].

In a 2005 Dutch study, clinically depressed cancer patients (less than 3 months expected survival) requested lethal interventions at four times the rate of nondepressed patients having a like prognosis (95% CI, 2.0–8.5) [72]. A priori defining a desire for hastened death at life's end as unlike other suicidal ideation impedes addressing underlying factors, including prior risk factors for suicide.

No notice is made of Oregon's biased psychiatric/psychological evaluation instrument (allowing only a finding of competency [Figure 1]), nor objection made to destruction of primary data [51].

View Large Image Download to PowerPoint

Figure 1 Signature block of Psychiatric/Psychologist Consultant's Compliance Form, Oregon State Public Health Division Center for Health Statistics. Revised 11/06.

The new policy “[s]upports allowing a mentally competent, terminally ill adult to obtain a prescription for medication that the person could self-administer to control the time, place, and manner of his or her impending death, where safeguards equivalent to those in the Oregon DDA are in place.” It also supports “measures to ensure that patients eligible to receive information about death with dignity and are able to choose alternatives” (aggressive symptom/pain management, palliative care, hospice, care to maximize independence and quality of life, voluntarily stopping eating and drinking, palliative sedation).

While APHA's Gonzales v. Oregon amicus curiae brief lauded diversity in state approaches, the new policy declares “experiment over”: all jurisdictions should adopt Oregon's model as general public health policy, notwithstanding deviations [68], data destruction [51], or caution regarding generalizability [52].

ODWDA public health reporting 

In 1998, Oregon Health Services4 expounded “Regarding Oregon's Death with Dignity Act and Oregon State Public Health's Role”: “The Health Services must balance the public's need-to-know with our responsibility to protect the privacy of Oregonians who choose PAS. Without this commitment to patient privacy, we jeopardize our ability to collect complete and useful information about the Death with Dignity Act” [75]. While data privacy could be protected by disidentification, OHS instituted a unique policy: destroy ODWDA primary data on a rolling basis [51], undercutting oversight; agency accountability; time-series modeling; forensic review; and historical reconstruction.

Public reports from 1999–2006 spanned 17–25 pages [76], reduced thereafter to a 2-page summary plus data table. The ODWDA website posts only summaries for 2006–2007 plus the latest year's data vector with prior data aggregated, making it difficult for the public to track trends.

Attempts to increase transparency failed. A 1999 House bill to amend ORS 127.865 sought a comprehensive census of cases and to define disidentified ODWD data as a public record open to external research [77]. If (ever) passed, however, prior destroyed data are nonrecoverable.

The Oregonean editorial board stated: “Oregon's physician-assisted suicide program has not been sufficiently transparent. Essentially a coterie of insiders run the program, with a handful of doctors and others deciding what the public may know” (September 20, 2008) [78]. Callahan called it Petomkin village–styled regulatory obfuscation: “protected from public scrutiny, but with the ring of authority and oversight” [19], noting that Dutch claims of compliance were defended until confidential surveys demonstrated otherwise, which Oregon has yet to undertake. Officially published data flow through prescribing physicians [79] without media-reported deviations. The Office of Disease Prevention and Epidemiology (formerly DHS) has neither mandate, authority, nor budget to investigate possible violations (e.g., Matheny and Cheney cases) [67], [68].

From 1998 through 2007, 546 lethal prescriptions brought 341 ingestion deaths; 109 physicians wrote 271 ingested pharmaka from 2000 through 2007, of which 3 physicians wrote 22.8% [80]. Eighteen reporting irregularities were filed with the Oregon Medical Board. None was declared to violate “good faith compliance” [81], a standard immunizing physicians from liability more than other contexts (trumping negligence claims).

Oregon Health Plan (OHP) coverages for the poor are notably narrower than other insurance, proscribing life-extending treatments without a 5% expectation of 5-year survival. As APHA endorsed lethal ingestion, its American Journal of Public Health lionized a public health leader who survived pancreatic cancer 2 years with treatments that the OHP denies [82]. Several Oregoneons denied physician-requested treatments but offered palliative care or lethal ingestion reported a profound sense of societal abandonment [83], [84]: ethical (but not economic) indifference as to whether their lives ended sooner or later.

Oregon patients with HIV/AIDS or ALS were proportionally 57 and 67 times more represented than those with heart disease. An Oregon/Netherlands comparison identified HIV/AIDS patients to be at heightened risk (30 times over chronic respiratory disorders) although not other vulnerable populations [85] (bracketing that at life's end all are vulnerable [86]). Persons over 85 were proportionately fewer, possibly due to cognitive impairment [80].

A 10-year ODWDA summary (1998–2007) charts induced deaths tripling (5.3 to 15.6 per 10,000 deaths; average 11.3) [54]. Filtered through annual physician reporting, primary motivations concerned autonomy (75.0% to 100%); enjoyable activities (86.6% average); bodily functions (58.2% average); and pain (27.3% average; trending higher: 47.8% and 32.7% in 2006 and 2007). Sense of burden to family/friends/caregivers more than tripled [12.5% (1998), 44.9% (2007)], peaking in 2000 (63.0%, new millennium consciousness?) [54]. “Burden” was not primarily economic (average 2.5%; 3-year mode 0.0%) [54], but psychosocial: “a burden to others”—a relational/social evaluation/perception [87].

Physician presence at ingestion stepped markedly downward (1998–2001 range: 32.8% to 59.3%; 2002–2007: 16.2% to 34.2%). Only the 1998 report discussed impacts on physicians. A significant toll appears associated with presence at induced death [81]—perhaps informing the decline. Half of ODWDA patients knew the prescribing physician less than 3 months [81]. After 2001, Oregon DHS destroyed traceable data links to participating physicians, making longitudinal follow-up impossible [81].

Forty-six percent of requesters changed their minds with physician intervention [88]. While Oregon is legally indifferent to outcome, physicians are not.

Certifying fiction: Public health vital records in Washington State 

Beyond Oregon's language and forms, Washington State's Initiative Measure I-1000, The Washington Death with Dignity Act (2008) [89], [90], explicitly requires medical certifiers to falsify death certificates: “The attending physician may sign the patient's death certificate which shall list the underlying terminal disease as the cause of death” [91].

During the initiative run-up, the Washington State Medical Association (WSMA) opposed “providing means for patients to hasten their own deaths” [44], underscoring that falsifying vital records violated professional integrity [92]. WSMA president Brian P. Wicks: “To my knowledge, there's no other situation in medicine in which the death certificate is deliberately falsified—and in which this falsification is mandated by law” [92]. The WSMA Policy Compendium (2008) exhorts physicians to “provide accurate, timely certification of cause of death for patients who have been under their care” [93].

Death certification is a core public health medicine role [94]. “Accurate cause of death information is important: To the public health community in evaluating and improving the health of all citizens, often to the family, now or in the future, and to the person who settles the decedent's estate” [95].

Death registration is a state function whose errors and misrepresentations propagate into the CDC National Death Index research database (Available at http://www.cdc.gov/nchs/ndi.htm) [96], [97].

CDC directives for physicians [96], medical examiners/coroners [98], and funeral directors [99] address the U.S. Standard Certificate of Death. Its Physicians' Handbook on Medical Certification of Death notes:

Because statistical data derived from death certificates can be no more accurate than the information on the certificate, it is very important that all persons concerned with the registration of deaths strive not only for complete registration, but also for accuracy and promptness in reporting these events. Furthermore, the potential usefulness of detailed specific information is greater than more general information [96].

The National Association of Medical Examiner's Cause of Death standard is instructive: “Truthfulness, completeness, and reasonable accuracy should be the goal—convenience and expedience should not play a role when deciding upon a cause-of-death statement” [100]. Falsification for appearance is (culpable) expedience.

Modeled on WHO guidelines, Cause of Death in the certificate provides “an etiologic explanation of the order, type, and association of events resulting in death” [96], identifying efficient causes of mortality [101]. Part I records events “leading directly to the death.” Line A indicates “immediate cause of death (the final disease, injury or complication directly causing death)” grounded in “YOUR best medical OPINION” (original emphasis) [96]. (The associated “Due to” statement listing onset time to death must likewise be falsified under the new mandate.) Specifically, “any disease, abnormality, injury, or poisoning, if believed to have adversely affected the decedent, should be reported.” Listed for automatic medical examiner or coroner notification are drug or alcohol overdose/abuse, accident, suicide, homicide or undetermined.

“Manner of death” follows. Natural deaths are “not due to external causes” (CDC) [96]. Lethal ingestion linked to a dispensing record (DOH 422.067) is an external cause, yet “[t]he manner of death must be marked as ‘Natural'” [102].

Should the pharmakon be slipped before its hour, the reporting rubric excludes medical examiner review, despite risk of undue influence: Parties barred from witnessing wills due to presumption of undue influence may witness requests for lethal prescription [67]. Once filled there is no control over its disposition; no requirement for independent observers nor that the patient still be competent. On close reading, self-administer=act of ingesting: “‘Self-administer' means a qualified patient's act of ingesting medication to end his or her life in a humane and dignified manner” [RCW 70.245.010 (12)] not excluding “help” [66]. Meeting families to guard against coercion (frequently occurring in elder abuse [103]) is not required.

After I-1000 passed, the Washington State Department of Health drafted Death Certificate Instructions [104], operationalizing the imposture at four levels: “Physicians and Other Medical Certifiers for Death Certificates” [102], “Funeral Directors” [105], “Local Registrars” [106], and “Medical Examiners, Coroners, and Prosecuting Attorneys” [107]. Each forces practitioners into a double bind fitting moral distress: a state of internal conflict (1) when one knows the ethically appropriate action but is unable to act upon it; (2) acting contrary to personal and professional values which undermines integrity and authenticity [108], notwithstanding that “[t]he State also has an interest in protecting the integrity and ethics of the medical profession …” (Washington v. Glucksberg [521 U.S. 702 (1997)].

Each instruction cites RCW 43.70.160 “Duties of registrar” whose task is “to secure the uniform observance of its provisions and the maintenance of a perfect system of registration” [102], [105], [106], [107]. Specifically, “[i]f any are incomplete or unsatisfactory, the State Registrar shall require such further information to be furnished as may be necessary to make the record complete and satisfactory”—expunging entries complying with CDC certification canons and redacting them to statutorily mandated misrepresentations.

Physicians/other medical certifiers and medical examiners/coroners/prosecuting attorneys [102], [107] must list the “underlying disease” as cause of death and manner “natural.” Suicide, secobarbital or Seconal, pentobarbital or Nembutal are proscribed terms; ODWDA intimations likewise. Prosecutors are denied language even to articulate a case of abuse.

J. L. Austin strikingly notes in How to Do Things with Words that legal discourse need not comport with conventional usage or facts (factum est: it is done) [109]. Legal definitions are performative, declaring beyond describing, and inhabit a realm apart: “Of all people, jurists should be best aware of the true state of affairs. Perhaps some now are. Yet they will succumb to their own timorous fiction, that a statement of “the law” is a statement of fact” [109].

Instructions to registrars and funeral directors like wise proscribe “assisted suicide,” “physician-assisted suicide,” “mercy killing,” “euthanasia,” or reference to the initiative banning the terms. Should secobarbital/Seconal or pentobarbital/Numbutal slip through possibly indicating “use of the Death with Dignity Act,” one must call the Registrar for guidance (maintaining the “perfect system of registration” [110]). Where the actual cause appears, the certificate is returned for “correction” and resubmission; only then is a permit to proceed with the body's disposition issued.

Certificate contortions are enough to raise the question: Were “death with dignity” so dignified, why camouflage it? Why force medical certifiers, under protest [92], to sacrifice integrity, transgress canons of sound public health documentation, invert semantics (“natural”), and undermine the public health rationale of accurate cause of death assessment? Possibly to avoid stigma and legal challenge, yet many accurately recorded life-ending circumstances involve potential stigma while Oregon and Washington statutes explicitly obviate legal challenges for wills, contracts [94], life insurance, and annuities [111].

Despite requirements to falsify public health records and medical society opposition, after review by its Legislative and Policy Committee, the Washington State Public Health Association Board unanimously endorsed I-1000 by email without discussion or communicating to the membership [45], [46]: “An urgent request” for endorsement by proponents fell between bimonthly meetings [46]. Member notice was a bullet point in the annual joint business meeting's President's Report to the Members (“Endorsed Initiative 1000, the Death with Dignity Act” [112]) dated 27 days before the election [45], but posted to the website only after.

Bias re psychiatric referral? 

The Oregon Death with Dignity Act: A Guidebook for Health Care Providers (Oregon Center for Ethics) advised psychological/psychiatric referral of all patients [113]. As noted in practice referrals plummeted [highest in 1999 (43%), 4.3% to 5.4% (2003–2006), 0% (2007), 3.3% (2008)] [54]. Evaluations were thrice as often for those older than 75 compared with younger than 65. Like Baile et al.'s prior concerns [114], the Oregon Psychological Association joined an amicus curiae brief in Gonzales v. Oregon (2006) “to underscore the scientific support for determining competence, impaired judgment, and clinical depression for patients who are terminally ill and for the subgroup of patients who may contemplate using the ODWDA” [115]. Scant referral and evaluation, and a biased reporting regimen fail this intent.

Structural bias open to inspection is found in Oregon's (and Washington's) Psychiatric/Psychological Consultant's Compliance Form [116] that ostensibly safeguards persons with impaired judgment. Following identification, medical-diagnostic, and psychiatric/psychological-evaluation blocks, a signature block declares, “I have determined through evaluation that the above-named patient is not suffering from a psychiatric or psychological disorder, or depression causing impaired judgment, in conformance with ORS 127.825” (see Figure 1). There is no alternative language, or space, for a signed finding of impaired judgment, and if found nothing prevents seeking another confirmatory evaluation or independently assures its filing. The form routes solely through the prescribing physician who forwards it (post-mortem) to Oregon's Public Health Division Center for Health Statistics.

In a case presented by Compassion in Dying (CID) [in 67], identified as “Helen,” a primary physician declined her request while a second diagnosed depression. In this regard, CID's director indicated “[i]f I get rebuffed by one doctor, I can go to another …,” one who did not consult the primary physician, only cursorily the second, but then subsequently stated he would not have prescribed had he known their reservations.

Ganzini et al. (1996) queried Oregon psychiatrists whether, absent a long-term relationship, they could identify in one evaluation a psychiatric disorder impairing judgment in patients seeking lethal prescription—6% were “very confident,” 43% “somewhat confident,” and 51% “not confident at all” [55]. Of psychiatrists opposing lethal ingestion, 72% would not participate, whereas 64% of those favoring would; home evaluations were difficult to come by [113].

Public and private domination of autonomy and limited protective obligations 

Life-ending acts impact social life-worlds while being influenced through formal or informal social sanction. Similar to social determinants of disease, one can speak of social determinants of life-ending acts: supervening influences including social values and structure, media, institutionalization, and naming of context and act.

Spindelman has analyzed structural forms of domination regarding autonomy discourses concerning assisted suicide and euthanasia [117]. “Public domination” involves state structures including regulations and bureaucratic instruments; “private domination” is mediated through nongovernment structures, culture, and family relationships. Forms span from strong coercion to attenuated influence. Psychiatric/psychological reporting above involves four levels of structural domination: limiting scope of finding for the consultant; limiting possible findings regarding the patient; lack of reporting independence; and, finally, that ODHS “has no authority to investigate individual Death with Dignity cases” [118].

Under the U.S. 10th Amendment, public health, safety, and morals fall discretionarily within state police powers. Underscoring the federal constitutionally permissible scope of private domination, Spindelman cites Chief Justice Rehnquist (DeShaney v. Winnebago County Department of Social Services, 1989 [119]): “nothing in the language of the Due Process Clause itself requires the State to protect the life, liberty, [or] property of its citizens against invasion by private actors. The Clause is phrased as a limitation on the State's power to act, not as guarantee of certain minimal levels of safety and security.” Spindelman adds “[n]o prominent set of safeguards—proposed or enacted, including Oregon's—is truly designed to deal with private structural domination” (at 1646, note 13) [117]. Some disqualified persons also angrily report a sense of being dominated [88].

Californian Elizabeth Bouvia (as example) was subject to several forms of private structural domination: class, sex, and disability [117]. Expecting a (non-eventuating) fatal outcome, the court honored her “autonomous choice” to remove a state-ordered nasogastric tube, however it was conditioned: cerebral palsy; quadriplegia; abandonment by spouse, parents, friends; and homelessness. The court concluded: “If a right exists, it matters not what ‘motivates’ its exercise” [120]. The first person to die under Washington's Death with Dignity Act, Linda Fleming, filed bankruptcy, lived in low-income housing, and was disabled [121].

The New York State Task Force on Life and the Law (1994) noted forces supervening on vulnerable populations: poorly diagnosed depression, poverty, lack of access to proper care, stigmatization, disability, and a new onus to affirmatively justify one's continued existence given a legally indifferent environment embedded within a social and economic milieu that is not [122].

Are structural forms of domination, public and private, likely to increase in import at life's end?

Population and future vision: Warnock and Macdonald's not so easeful death 

Citing “half in love with easeful death” (Keat's Ode to a Nightengale) [123], Warnock and Macdonald consider life-ending across contexts and the demographic horizon [124]: neonates' parental choice, the elderly absent social function, raw autonomy, subjectively assessed quality of life, chronic psychiatric suffering, and suffering in extremis. Each balance tilts toward a life-ending solution.

Neonates' parents internalize an ethic of structural limits while alternatives for conditions incompatible with life (e.g., perinatal hospice) are not raised [125]. (Yet: planned scarcity entails worthy treatment hardly more than scarcity in disaster.)

Subjective measures of tolerability would expand the population eligible for a lethal course without classifiable physical or mental conditions (like the Royal Dutch Society for the Promotion of the Healing Art's lijden van het leven discussion, suffering from life [126]).

For persons with chronic mental illness, a negative absolute right to refuse treatment is “only a small step” toward the positive right “to be helped to end a life that is intolerable to the person who lives it”—“even if legislation is unable, yet, to alleviate it” (p. 34). The articulation is primarily individually focused: aggregate chronic mental illness costs, including from pharmacotherapeutic noncompliance [127], are unmentioned although such were staples in 1920s European [128] and 1940s U.S. discussions [129]. Economics-related entries are conspicuously absent from the index, as if not in play.

Elsewhere, Baroness Warnock, who chaired significant policy committees on special education (1978) and on assisted reproduction and research using human embryos (1982–1984), places a duty to more quickly die on persons with dementia squarely in terms of cost to family and the National Health Service: “If you're demented, you're wasting people's lives—your family's lives—and you're wasting the resources of the National Health Service” [130]. £35B per year by 2026 for the primary medical funder and employer in England and Scotland for . . .? Responding, Neil Hunt of the Alzheimer's Society's notes: “With the right care, a person can have good quality of life very late into dementia” [130].

In the burden-and-elimination discourse, an unraveling self should internalize that her continued existence is an essential indignity (ad. L. dignitat-em, “merit, worth”; dignus, “worthy”)—lacking in worth and unworthy of care [18]. The infirm are to loose familial commitments through self-sacrifice: social exclusion internalized as altruism, combining Spindelman's public and private forms of domination. Those bearing a fractured sense of dignity (worth) report being regarded by others as less worthy of respect or esteem, and where perceiving themselves to be a “burden to others” (a social reflection) are much more likely to desire hastened death [20].

In Warnock and Macdonald's closing chapter, “Looking Further Ahead,” population, philosophical, and social considerations converge: the life value of one with dementia shifts down a scale of the human post-Darwin on an organismic continuum: the corporeal subject with emergent states of consciousness is required to have continuity (biographical self-narrative) for value, without which “society” withdraws support.

For 2076, more than 1M British centenarians are projected while families now are no longer places “within which the old play an important and useful part,” nor a haven of care at life's end. Most spend last months or years in hospitals and institutions where “for many, their death is the result of a decision, not their own, that”—per the authors—“the effort to keep them alive may properly be abandoned, and they are deliberately allowed to die” (pp. 126–127). Promethean autonomy has disappeared.

Future palliative care is significance in its lack: “improvement of palliative care, an agreed goal for the present, will in future, with our rapidly aging population become economically unsustainable burden” (pp. 136–137). Demographic pressures, planned scarcity, intergenerational isolation, subquality pain management: all dominating factors confronting the infirm. The new physician is a thanatologist, an easer of death (p.138).

As with dignity above, semantics turn imperial and ironic: “In this way we may, as time goes on, introduce compassion into the laws that govern the end of life.” Ostensibly, (then) current law, and those opposing lethal solutions or indicting structural abandonment and isolation that drive increasing vulnerability, lack compassion. (The term's initial definition per the Oxford English Dictionary is Obs., obsolete: “[S]uffering together with another, participation in suffering; fellow-feeling, sympathy” [131].)

Population aging and juvenation are trailing trends of relative proportions involving (in)fertility, spacing (births delayed or reduced thin lower population pyramid level density), and health determinants [132]. Total fertility rate (TFR) for replacement is a key trending indicator (varying by country beyond baseline 2.1 up to 3.43 for Sierra Leone given mortality distributions [132], [133], [134]). System costs concentrating on younger cohorts are a consequence of priors. Also at life's end, many hands make light(er) work. “Not being a burden” recasts itself as intergenerational inequity (for progeny born or eliminated).

The Netherlands, Switzerland, Belgium, Luxembourg, and United States—each sanctioning forms of induced death—are all among the top 20 per capita economies (2008 PPP GDP per World Bank, IMF, and CIA Factbook). Except for the United States (2.05), each falls significantly below replacement TFR (below the 30th percentile worldwide, 1.45–1.78) [134].

“Assisted death,” however, is not a regulative sought in the developing world as expression of international equity—notwithstanding greater suffering and health infrastructure deficits. Would shipments of secobarbital or pentobarbital to AIDS stations short on antiretrovirals or to an Indian or Brazilian slum be received as a gesture of compassion (elevating resource-poor contexts, dignifying sufferers through a fuller range of end-of-life [life-ending] options), or seen as a new type of imperialism?

And the reverse?

A duty to die for international health equity? 

A prominent assisted suicide/euthanasia advocate reflects on whether developing world health deficits generate a developed-world duty to die [135]: sacrifice for the distant stranger. Summarizing: “This duty would be the duty to conserve health-care resources by fore-going treatment or directly ending one's life in the interest of justice in health care, and it would be reflected in more nearly equal health prospects and life expectancies around the globe” (p. 297)—from Sierra Leone (40) to Japan (83) (WHO, 2006). Wealthy countries with 16% of global population generate 89% of health care expenditures (U.S.: 5% population, 50% health spending). Wealthy nations suffer only 7% of 1.4 trillion lost DALYs. Elements of Warnock and MacDonald's future vision connect with an international horizon.

This vision considers resources as entities for which an asset entails a debit addressable by redistributive transfer,5 not withstanding that simple financing gap models fail in development even under relative stability [136]. Admittedly, nations with the bleakest prospects frequently have multiple morbidity and mortality overlays: AIDS, malaria, complex humanitarian emergencies, and high corruption indexes [137]. The thought experiment places great weight on differential health care as a causal path to global health disparities.

The argument invokes stepping stones: Jon Hardwig (being an excessive burden to family and resources); Norman Daniels (age-based rationing); rejecting medicide, Daniel Callahan considers as distortive maximizing medical means in every circumstance [135]. (Not cited, later Engelhardt counsels foregoing medical means should they become all-consuming, impediments to one's spiritual vocation [138].)

Allan Buchanan's concern for insufficient human rights allocation structures translates to insufficient international health transfer structures (bracketing NGOs and IGOs). The idealized model is a “closed redistribution system.” Savings from foregone medical services for condition C in developed land X are transferred solely for health purposes to developing land Y6 [135].

In passing, this structure defeats Hardwig's intent of decreasing local financial burden: foregone costs (savings) accrue to strangers, not to family. Used or transferred, health care GDP in wealthier lands would remain unchanged—only with the difference exported. Preventive savings are losses for the transfer equation: only treatment for unprevented conditions can be foregone. Similarly, more expensive procedures increase transfers; cost-saving innovations would lower them. Notwithstanding pragmatic challenges, as international health homeostasis is achieved, the emergent duty to die progressively self-extinguishes.

Where, then, does induced death fit in? Similar to the UK population vision above, foregoing physiologically necessary treatment would expose one to “the potentially cruel consequences of living without adequate care” (p. 283). Those in a Rawlsean Original Position (drawing up regulative parameters before knowing what allocation one has within the world) might opt for assisted suicide or euthanasia availability to avoid suffering. In nuce, “policies permitting or even encouraging voluntary choices of an earlier, more humane but also more resource-conserving death, if supported in the Original position, would then be just” (p. 283). Still, such an emergent duty is distant, lacking the closed transfer system.

Carbon credits, carbon deaths? 

Another structure would follow a similar logic: bounded (if not closed) anthropogenic carbon production fractionally influencing global warming/climate change (with accounts ranging from apocalyptic [139] to cautionary [140]). The operative assumption: everyone affects all. Those at end of life, the argument would go, have large carbon footprint potential given resources (that would be) attached to them. Excepting stars and politicos, few in the developed world are able to offset their existence as carbon neutral via indulgences (carbon credits). Resource-intensive extensions of life unjustifiably source externalities upon the planet per this view… . (Would similar logic advance against persons having resource-intensive, durative nonterminal impairments?)

Conclusion 

Public health's endorsement of lethal ingestion highlights a paradox of disciplinary proximity but can be related to population perspectives. Yet related statutes, regulations, and public practices demand sacrifices of professional integrity (e.g., falsification of public health records) as rhetoric has sacrificed semantic integrity. Future-oriented scenarios suggest autonomy-based justifications will increasingly yield to population-inflected discourse, increasing risk of private and public forms of domination and vulnerabilities toward life's end. However, public health should address institutionalized violations of data integrity and patient vulnerabilities, while rescinding policy positively supporting life-ending means.