Economics, Ethics and Medical Aid in Dying: A Social Work Perspective

By Carolyn Smith, University of Windsor, MSW Program

I began this assignment with a strongly pro-assisted suicide stance. I’d seen lovely depictions in popular media of baby boomers and young people with terminal cancer throwing end-of-life parties and slipping quietly off to the “great beyond” surrounded by friends and family, aided by a lethal prescription. In Oregon, where it is is better-established, assisted suicide seems to be the domain of older, white, male, well-educated patients, whose main motives are loss of autonomy, dignity and no longer being able to enjoy life’s activities (Emanuel, Bregje, Onwuteaka-Philipsen, Urwin & Cohen, 2016). It seemed to me to be a kinder alternative to the classic acute care death setting -- the pain, fear and trauma to which we are more accustomed. And perhaps the end-of-life paradigm under baby boomers would follow a similar path to the beginning-of-life, with more consumer choice and less of an emphasis on technology.

What I hadn’t bargained on was learning how economic pressures and culture change could skew a society toward cultural acceptance of assisted suicide. By chance, I drew the task of researching the economics of medical assistance in dying (MAID) for our group presentation. MAID (the terminology of Bill C-14 and used by OCSWSSW) is defined by the Canadian Medical Association as “... a situation whereby a physician intentionally participates in the death of a patient by directly administering the substance themselves, or by providing the means whereby a patient can self-administer a substance leading to their death” (CMA, 2014). Initially I thought of it as a small piece of the assisted suicide critical analysis, particularly as it pertains to social work. However, studying the economics of assisted suicide in particular, and end-of-life more broadly, unsettled my initial pro-MAID stance.

In this paper I will examine the landscape in which this debate is unfolding, and the ethics of MAID vs formal and informal end-of-life care through a structuralist lense. I will briefly review the literature on the topic, and I will contrast the insights of a deontological and a relativist perspective to discover the way in which economics and ethics intersect with a philosophical evolution around the taking of a human life. I will examine the relevant principles of the CASW Code of Ethics as they relate to MAID and will identify the need for social workers to be on guard for economically-driven decision-making at the level of the individual, family, community and government.  I will show that, self-determination notwithstanding, palliative care, defined as “care that strives to meet the physical, psychosocial, social and spiritual needs of people who are dying and their family members” (CHPCA, 2010), best reflects the values of social work. It’s only when that fails, when quality of life is close to nil, that we can legitimately affirm a client’s right to self-determination and support them in their bid for assisted suicide. Finally, during my own self-reflection, it has become evident to me that our “...profession’s unique perspective that considers a client holistically: as an individual and as part of a family, community, a system” (CASW, 2016) lends a lot of value to this ethical debate and points to a new way forward.

Reviewing the literature, we see that assisted suicide is a rapidly evolving ethical issue in Canada, due to the passing of Bill C-14, the assisted dying bill, in June 2016. As social workers working in palliative or end-of-life care, we are at a historic confluence of trends: 1. Death has gone from being ubiquitous and unremarkable at the outset of the 1900s, to being a medicalized process behind closed doors (Gawande, 2014). 2. Our population is aging dramatically, because we remain healthier, longer (Carstairs, 2010). Brown & Walter (2014), writing about the parallel situation in the UK, state: “With unprecedented numbers of people living longer and with higher expectations of how they will live out their last years, the management of end-of-life (EOL) services is being brought into sharper focus” (p. 2375). 3. The third trend is primarily cultural, and I won’t focus on it here. However it is critical to note that in Western Anglophone communities and countries, the values of individual autonomy, choice and control are very much influencing the approach to end-of-life care (Brown et al 2013, Boudreau et al 2014) in a way that does not position the natural process of death within a person’s social environment and ecosystem, and may not resonate with values of other cultural backgrounds.

Let’s remember that the chances of a human life ending in death is 100%. By 2056, with our aging population, 1.1%, of the population, will die each year. That’s 1 out of every 90 Canadians,  or 480,000 people annually. Since each death in Canada affects five other people, by 2056, 2.4 million Canadians will be having some tough conversations, caregiving or making decisions about the care of family and loved ones each year. (Carstairs, 2010).  

When asked, most Canadians would prefer to die swiftly. However, only 10% of deaths in one Western Canadian study happened quickly, and it’s fair to assume those numbers hold across Canada (as quoted in Carstairs, 2010).

The majority of Canadians, 75%, would also prefer to die in the comfort of their own home, according to a 2014 Harris/Decima survey. However, research also shows this outcome to be extremely unlikely, with 70% of Canadians dying in hospital (as quoted in Canadian Hospice Palliative Care Association Fact Sheet, 2014). Dying at home in Ontario, even if partial home care services are provided, is most likely if a family member is prepared to act as a caregiver. Wolff, Dy, Frick & Kasper (2007) report that at the time of their national study in the U.S., 72.3% of those in their last year of life were receiving help from an informal caregiver. When a loved one is dying, it takes about 54 hours per week to care for them (CHPCA, 2014), and only an estimated 16-30% of Canadians have access to palliative care depending on where they live in Canada (CHPCA, 2014). We will discuss this issue more when we turn to the impact of end-of-life on the community.

The cost of dying to individuals, families and informal caregivers is steep, and the public funding to support people’s wishes is insufficient. It’s only logical that this influences an individual’s choices about their end-of-life care, including seeking MAID. Or as Fowler & Hammer (2013) state,

“Although there is not widespread acknowledgement of the need to ration specific resources at the end-of-life in Canada, there is universal acceptance that there are insufficient resources to match a patient’s stated desires for end-of-life care (most commonly, out-of-hospital, symptom-focused, non-technologically-laden care) and the care that we are able to deliver (predominantly hospital-based dying, frequently in the ICU [in] a patient’s final interaction with the health care system).”

These facts collide head on with the deontological perspective, best expressed by Boudreau and Somerville (2014), who reason that “there exists a universal morality and that, at the very least, there is significant intercultural agreement on core concepts of ethics… medicine is necessarily constrained by moral absolutes… [and] some things must never be done to patients by their physicians.” Yet patients do not want the end of life they’ve witnessed watching loved ones die within the medical system. We see time and again the use of “futile care” measures which can actually increase suffering and decrease quality of life, and seem contrary to the “do no harm” vow physicians take (Hendry et al., 2013).  A qualitative synthesis of international studies highlights the following concerns when people are asked about the pros and cons of MAID: concerns about quality of life, including being a burden; the desire for a good quality of death (autonomy and control); concerns of abuse if assisted dying were legalized (need for safeguards, financial pressure, vulnerable groups, discrimination and the role of others in decision-making); the importance of individual stance (moral or religious views, personal experience of death or suffering, being for or against) (Hendry et al., 2013).

As we look to the future, an excellent question for future research would be to unpack just what is meant by ‘being a burden.’ Certainly in Ontario, between vulnerability, loss of hope, selfhood, function and suffering, empathizing with the work of their informal caregivers, and trying to navigate the byzantine, underfunded and non-transparent public-private system to receive help in the home (Grant & Church, 2015), it would be understandable to feel burdensome. It can be hard for the individual to see that despite the significant emotional, physical and financial strains reported by end-of-life caregivers, more than two-thirds felt personal rewards related to their helping role (Wolff et al, 2007).

Stepping back to look through a relativist lense, we can see how MAID could appeal to the individual: the steep economic and personal cost for individuals and caregivers, coupled with the fact of present and future suffering and loss of self. Yet, Dolgoff, Harrington & Loewenberg (2012), in discussing the NASW’s policy statement on euthanasia, [defined as when a physician administers the lethal injection (Boudreau & Somerville, 2014)], ask social workers to consider whose quality of life is supported by assisted suicide. Theoretically, that’s the person opting for it. Yet, they ask, whose life is harmed? There’s evidence that “death work” is extremely difficult for those who engage in it as well as those who are left behind, but more work needs to be done to understand what supports they require (Gamandi, Pott & Payne, 2013).

While affirming a client’s right to self-determination, as outlined in Section 1.3.1 of the CASW code of ethics, when we are confronted with a request for MAID working in gerontology, hospice or palliative care, we need to work through our roles as described by the ecological perspective to ensure the best decision for the client. Through working as a conferee, we can understand the end-of-life crisis as experienced by the client, we can enable the client to explore all options open to them and to hear their wishes, we can connect the client with services including palliative and a doctor willing to discuss MAID, we can mediate between the client and family who may have mixed feelings about MAID, we can advocate for improved palliative strategies and services that can enhance quality of life, and we can make sure that clients are approaching MAID fully-informed, and are not just operating out of a place of fear.

According to Linda Hochstetler, MSW, who runs some of Toronto’s Death Cafés and teaches at Palcare Network, there are not enough social workers contributing at the end of life. Says Hochstetler, who specializes in end-of-life care, “I have had one client ever who considered [MAID]. She was only 42, estranged from family, lived alone with no husband or kid, and had progressively bad MS. She eventually said if she could be guaranteed good palliative care, she wouldn't consider it…”

But what if this client did have family? Hochstetler says “The biggest questions I have are how to work with individuals and families when the individual wants to die and the family wants to keep them alive. Ethically, does autonomy always win?” (Hochstetler, personal correspondence, 2016). The deontological perspective, thoroughly elaborated by Boudreau and Somerville (2014), says no, there are a universal set of moral ethics that preclude the taking of a human life. Relativism suggests the benefit of ending her suffering might outweigh the harm done to her family. But as ecological, person-in-environment social workers, we should perhaps consider the client in the case of MAID as both the terminally ill person and his or her immediate family. Family work would likely be effective in this case, and would require a lot of transparency and self-reflection on the part of the social worker. The OCSWSSW (2016) offers little guidance to date on such real world applications, short of listening and referring, and reminding workers they may opt out if they are uncomfortable with the process.          

As social workers with a structuralist and anti-oppressive practice, we must also ask how access to end-of-life care and MAID affect different protected groups in society. Interestingly, database analysis of PAS Oregon and the Netherlands shows “no increased incidence of physician assisted death in elderly people, women, people with low socioeconomic status, minors, people in racial and ethnic minorities, and people with physical disabilities or mental illness” (Quill, 2007). This can give us some reassurance that MAID is not and will not be foisted on protected classes of people, though one must always remember that those at the end-of-life are exquisitely vulnerable persons, and there is ample evidence the slippery slope exists, in that in countries with euthanasia, it can be offered to ever broader groups of people (Bilefsky, 2016). It is also important to note that the economics of end-of-life choices are likely to be different in a more for-profit healthcare system, where the burden of protracted illness on individuals and families are bound to be much greater.

Research evidence indicates that people often ask about MAID because of intractable pain, and change their minds when pain is controlled (Wilson et al, 2007, Hendry et al, 2012). So if we can mostly control the pain, if we can alleviate some of the burdens of suffering through palliative and hospice care, is MAID truly the choice our clients would make if they had better end-of-life care options that didn’t involve financial/service scarcity? Are our clients adapting their ethics to accommodate economic scarcity? These are some of the questions social workers need to ask in a clinical setting, while performing the roles of an ecological social worker.

The CASW code of ethics also reminds us that while we must maintain the best interests of our clients as a priority, we must pay due regard to the respective interests of others (CASW, 2005, section 1.1.1). Again, this points to the notion that our clients may be the patient, the caregiver, or, if we’re working from a person-in-environment model, perhaps both. Caregivers are very much impacted by end-of-life and MAID. Social workers must not forget the economic impact of end-of-life on caregivers. Estimates suggest that Canadian family caregivers provide $25 to $26 billion worth of care each year, and incur $80 million in out-of-pocket expenses each year (quoted in Stajduhar, 2013). In one study quoted in Quality End-of-life Care Coalition of Canada’s 2012 literature review, a caregiver’s mean monthly lost wages and leisure totalled $17,453 per month -- and that was with palliative care support, which as we’ve seen is by no means sufficiently available using public funds. As structuralist or ecological social workers, it behooves us to encourage caregivers to get support as well, and group work helps those caregivers feel both less alone and part of a larger class of individuals with shared concerns.

We turn our attention now toward the cost to the government and taxpayers around end of life. Not only is dying quickly the preference of most people, it’s the least expensive way to go, costing the system an average of $10,000 per person (QELCCC, 2012). However, if one doesn’t die quickly, the costs climb drastically.

In Ontario alone, the current cost of care in the last year of life accounts for $4.7 billion, or 10% of Ontario’s total healthcare budget. Average healthcare cost per patient in the last year of life is $53,661 (Tanuseputro et al, 2015). According to the Economist Intelligence Unit’s 2015 Quality of Death Index, palliative care around the world needs investment, but offers significant cost savings (Murray, 2015). It also can “improve patient and caregiver satisfaction, [provide] better symptom control and greater likelihood of the person dying in the setting he or she prefers… while saving the system $7-8,000 per patient” (Hodgson, 2012).

The CASW has it right when it declares that we need “a palliative-first approach to care, including the development of a more robust and accessible model of palliative care… [this is] crucial in ensuring that Canadians do not pursue MAID due to a lack of adequate or accessible palliative services” (CASW, 2016). Palliative care aligns much more closely than MAID to the values of the social work profession, as it does not require us to remove the “protection of life” principle from the top of the ethical principles screen (Dolgoff, Harrington & Loewenberg, 2012).

The palliative system is itself bogged down by public perceptions that palliative and hospice care constitute ‘giving up.’ A quick Google search demonstrates that much like the terminology of MAID, palliative care is starting to be rebranded as “supportive care” and being offered earlier in people’s illness trajectories, to positive and cost-effective effect (Sayal, 2016). However as we’ve seen, “comprehensive hospice palliative care… strives to meet the physical, psychosocial, social and spiritual needs of people who are dying and their family members” (CHPCA, 2010). The CHPCA posits that many of those who are pro-MAID are simply unaware of the options to hasten death already available, such as refusing treatment, food, or water. Withdrawing treatment is an exercise in self-determination, and furthermore, when combined with the option for total sedation, both more ethical and fully legal (CHPCA, 2010).

Regardless of what we call it, palliative and hospice care has the potential to address both the issue of social worker values around MAID, and patients’ fears around painful and futile medical care at end of life. It also goes some way toward ameliorating the scarcity of resources in the system as the population ages. In fact, this represents the most ethical way forward, particularly because the low cost of MAID represents a Faustian bargain, with a rather unbeatable price tag: the cost of drugs to end a life vary from $250 (doctor-administered, covered by OHIP) to $23k (out-of-pocket for self-administered drugs due to a drug scarcity issue) (Ubelaker, Globe and Mail, 2016).

With the cost of assisted suicide so low, the conundrum for the system plagued by scarcity is whether we put our resources toward saving curable young people with a future of contributions to society, or do we continue to pay for “futile” or “desperation” care for those with terminal illnesses?

Linda Hochstetler says for her, “...the real issue is that most people who consider MAID don't actually get access to a social worker. Where would they? Most hospitals only use social workers to do discharge planning. The CCAC's (Community Care Access Centres) use only .17% of their home visits to go to social workers… Where would social workers come in with families to help out? In what settings? Can social workers help by addressing the total pain of someone who wants to end their life?”

As Canadians, and as social workers, we must address these structural issues and adopt the goal of The Way Forward Initiative, where living well until death is the goal of care (Hodgson, 2012). Once we get there, there are opportunities for the Canadian social work profession, in accordance with our values and ethics, and from an ecological perspective, help end-of-life care move toward an even better system than the palliative care one we have on the table.

Brown and Walter (2013), academics in the UK social work system, challenge us to consider “whether the hospice and palliative care model needs to be rethought.” They believe social work could help our end-of-life care systems by building on the resources and networks surrounding individuals through network mobilization. Described as building a social model of end-of-life care (we could call it a person-in-environment model, perhaps), they critique the palliative care model as replacing the medicalization of dying with the professionalization of death through the use of multidisciplinary teams. They would prefer to see, and cite examples of, end-of-life professional services relocated within the range of all an individual’s care networks, informal and formal. They see that social workers have an essential role to play in helping mobilize a dying person’s social networks, in the sense championed by thinker and political scientist Robert Putnam, who named it “social capital.” This is the essence of the ecological perspective; and would help to identify needs and wishes, and direct formal, professional resources more efficiently to where they are needed. In Ontario, this would address the constant concern that our system of accessing resources is confusing, spotty and insufficient (Grant & Church, 2015). As quoted in The Way Forward literature review, even shifting 10% of Ontario’s end-of-life patients from acute to hospice palliative care could save the health care system $9 million per year. And if patients were able to access home-based palliative care, the estimated cost savings would be 33%-68% of the cost of traditional home care.

My research has left me with a particular sense of urgency that a paradigm shift of this magnitude will be necessary, as demographics mean our healthcare system will become ever-more strained. Health care dollars are plagued by scarcity, and there are “no good grounds for ignoring the connection between euthanasia and the urge to save medical resources” (Häyry & Häyry, 1990). Examining this issue through the lenses of structuralism and person-in-environment focus, I came to see that, while being non-judgemental and supporting self-determination, social workers must also be proponents of good lives and dignified deaths, or as Atul Gawande (2014) writes, “...the fact that, by 2012, one in thirty-five Dutch people sought assisted suicide at their death is not a measure of success. It is a measure of failure. Our ultimate goal, after all, is not a good death, but a good life to the very end.”

In conclusion, the pendulum has swung from death being an omnipresent fact of life, to death being a hidden, measurable and controllable medical process. It makes sense that, as social workers, we begin to grapple with how to find the balance, and help our clients facing end-of-life to do the same. Perhaps we need to revisit the value of death, and consider end-of-life and death as a natural process and a teacher of connection. We risk making it so convenient to die that we lose the lessons it can teach us: about caring, connection and the value of our days together. I leave you with this caution from Boudreau and Somerville (2014).

““Self-willed death may be merciful to oneself and simultaneously cruel to others. There is an essential reciprocity in human life.””

References

Bilefsky, Dan, and Christopher F. Schuetze. (2016).  “Dutch Law Would Allow Assisted Suicide for Healthy Older People.” The New York Times, October 13, 2016. Retrieved from http://www.nytimes.com/2016/10/14/world/europe/dutch-law-would-allow-euthanasia-for-healthy-elderly-people.html.

Boudreau, J. Donald, and Margaret Somerville. (2014). “Euthanasia and assisted suicide: A physician’s and ethicist’s perspectives.” Medicolegal and Bioethics. Accessed October 7, 2016. https://www.dovepress.com/euthanasia-and-assisted-suicide-a-physicianrsquos-and-ethicistrsquos-p-peer-reviewed-article-MB.

Brown, Louise and Walter, Tony. (2014). “Towards a social model of end-of-life care.” British Journal of Social Work, 44, 2375-2390. doi:10.1093/bjsw/bct087

Canadian Association of Social Workers. (2005). Code of Ethics.

Canadian Association of Social Workers. (2016) “Physician-assisted death: CASW discussion paper. Accessed October 7, 2016. http://www.casw-acts.ca/en/physician-assisted- death-casw-discussion-paper. Published Friday, Jul. 10, 2015

Canadian Hospice and Palliative Care Association. (2014). “CHPCA fact sheet - Hospice palliative care in Canada. Accessed October 11, 2016.  http://www.chpca.net/media/330558/Fact_Sheet_HPC_in_Canada%20Spring%202014%20Final.pdf (LINK NO LONGER ACTIVE)

Canadian Hospice and Palliative Care Association. (2010). “CHPCA issues paper on euthanasia, assisted suicide and quality end-of-life care.”

Canadian Hospice Palliative Care Association. (2013). “What Canadians Say: The Way Forward Survey Report, for The Way Forward initiative. Harris/Decima Poll.

Carstairs, Sharon, The Honourable Senator, P.C. (2010). “Raising the Bar: A Roadmap for the Future of Palliative Care in Canada.” The Senate of Canada. Accessed October 7, 2016. http://www.virtualhospice.ca/Assets/Raising%20the%20Bar%20June%202010_Senator%20Sharon%20Carstairs_20100608160433.pdf

“CMA POLICY SUMMARY - CMA_Policy_Euthanasia_Assisted Death_PD15-02-E.pdf.” Accessed October 7, 2016. https://www.cma.ca/Assets/assets-library/document/en/advocacy/EOL /CMA_Policy_Euthanasia_Assisted%20Death_PD15-02-e.pdf.

Dolgoff, R., Harrington, D., & Loewenberg, F. (2012). Ethical decisions for social work practice (9th ed.). Belmont CA: Brooks/Cole.

Emanuel, Ezekiel J., Bregje D. Onwuteaka-Philipsen, John W. Urwin, and Joachim Cohen. (2016). “Attitudes and Practices of Euthanasia and Physician-Assisted Suicide in the United States, Canada, and Europe.” JAMA 316, no. 1: 79–90. Accessed October 7, 2016. http://jamanetwork.com.ezproxy.uwindsor.ca/journals/jama/fullarticle/2532018

Fowler, Robert & Hammer, Michael. (2013). “End-of-life care in Canada.” Clin Inest Med, 36 (3): E127-E132. Accessed October 13, 2016. http://cahs-acss.ca/wp-content/uploads/2015/07/End-of-Life-Care-in-Canada.pdf.

Gamondi, C., Pott, M., & Payne, S. (2013). “Families' experiences with patients who died after assisted suicide: A retrospective interview study in southern switzerland.” Annals of Oncology, 24(6), 1639-1644. doi:10.1093/annonc/mdt033

Gawande, Atul. (2014). Being mortal: Medicine and what matters in the end. Canada: Doubleday Canada.

Grant, Kelly & Church, Elizabeth. (2015). “No Place like Home? Investigating Ontario’s Home-Care Shortcomings.” The Globe and Mail. Accessed October 13, 2016. http://www.theglobeandmail.com/news/national/no-place-like-home-investigating-ontarios-home-care-shortcomings/article25409974/.

Hendry, M., Pasterfield, D., Lewis, R., Carter, B., Hodgson, D., & Wilkinson, C. (2013). “Why do we want the right to die? A systematic review of the international literature on the views of patients, carers and the public on assisted dying.” Palliative medicine, 27(1), 13-26.

Hodgson, Corinne. (2012). “Cost-effectiveness of palliative care: A review of the literature. The Way Forward Initiative. Accessed October 2, 2016. http://www.hpcintegration.ca/resources/discussion-papers/economic-review.aspx

Hochstetler, Linda. (2016). Unpublished personal correspondence, via email dated October 11, 2016. Hochstetler is an MSW, RSW working in education at PalCare Network in Peel Region.

Livne, Roi. (2014). “Economies of Dying The Moralization of Economic Scarcity in U.S. Hospice Care.” American Sociological Review 79, no. 5 (October 1, 2014): 888–911. doi:10.1177/0003122414547756.

Murray, S. (2015) “The 2015 Quality of Death Index. Ranking palliative care across the world.” A report by The Economist Intelligence Unit. Line D, ed [accessed on December 8, 2015].

Ontario College of Social Workers and Social Service Workers (OCSWSSW). (2016). "Medical assistance in dying: What are my professional obligations?” Discussion paper.

“Quality End-of-life Care Coalition of Canada (QUELCCC). (2012). “Cost Effectiveness of Palliative Care: A Review of the Literature.” http://hpcintegration.ca/resources/discussion-papers/ economic-review.aspx. Accessed October 9, 2016.

Quill, Timothy. (2007). “Physician assisted death in vulnerable populations.” BMJ, Sept 29, 2007; 335(7621: 625-626.

Sayal, Priya. (2016). “First Step to Improving Palliative Care: Change Its Name | Toronto Star.” Thestar.com. Accessed October 5, 2016. https://www.thestar.com/opinion/commentary/2016/08/30/first-step-to-improving-palliative-care-change-its-name.html.

Stajduhar, Kelli I. (2013) “Burdens of family Caregiving at the End of Life.” Clin Invest Med. June 2013. Accessed October 10, 2016. EOL2 - Burdens-of-Family-Caregiving-at-the-End-of-Life.pdf

Tanuseputro, P., Wodchis, W. P., Fowler, R., Walker, P., Bai, Y. Q., Bronskill, S. E., & Manuel, D. (2015). The Health Care Cost of Dying: A Population-Based Retrospective Cohort Study of the Last Year of Life in Ontario, Canada. PLoS ONE, 10(3), e0121759. http://doi.org/10.1371/journal.pone.0121759

Ubelaker, Sheryl. (2016). “Drugs for physician-assisted death: What will they cost and who will pay?” Canadian Press, as published in The Globe and Mail. June 13, 2016.

Wilson, Keith G., Harvey Max Chochinov, Christine J. McPherson, Merika Graham Skirko, Pierre Allard, Srini Chary, Pierre R. Gagnon, et al. (2007). “Desire for Euthanasia or Physician-Assisted Suicide in Palliative Cancer Care.” In Health Psychology, 26:314. American Psychological Association. doi:10.1037/0278-6133.26.3.314.

Wolff, JL, Dy, SM, Frick, KD, Kasper, JD. (2007). “End-of-life-care: findings from a national survey of informal caregivers.” Arch Intern Med. Jan 8, 2007; 167(1): 40-6.