In their critique of our paper "Legal physician-assisted dying in Oregon and the Netherlands: evidence concerning the impact on patients in "vulnerable" groups," I.G. Finlay and R. George claim to challenge our underlying assumptions and methodology with "another perspective on Oregon's data." In our view, however, they miss the point of our paper and address a quite different issue. While we welcome their attempt to further explore issues about assisted dying, we do not believe they have in any way undercut our argument that where assisted dying is already legal (at the time of our study, Oregon and the Netherlands), there is no current evidence for the claim that legalized physician-assisted suicide or euthanasia will have disproportionate impact on patients in vulnerable groups.

Our paper was developed in response to the "slippery slope" concern widely prevalent in statements and position papers from variety of groups, including many professional medical groups. We cited among others this sample:

"Both society in general and the medical profession in particular have important duties to safeguard the value of human life. This duty applies especially to the most vulnerable members of society--the sick, the elderly, the poor, ethnic minorities, and other vulnerable persons. In the long run, such persons might come to be further discounted by society, or even to view themselves as unproductive and burdensome, and on that basis, "appropriate" candidates for assistance with suicide." "...the ramifications [of legalization] are too disturbing for the...value our society places on life, especially on the lives of disabled, incompetent, and vulnerable persons." American College of Physicians--American Society of Internal Medicine (ACP -ASIM), 2001

Drawing on this and many similar statements in the medical, policy, and bioethics literatures, we examined 10 groups variously identified here and in inequality studies generally as "vulnerable." We explicitly refrained from assuming that people in these groups are actually vulnerable and that people not in these groups are not, and we did not assert that issues about vulnerability could not be examined in other ways --as Finlay and George have undertaken to do. We did seek, however, to examine objectively this particularly widespread form of slippery-slope argument because it has played such a major role in the public, political, and professional debates over physician assistance in dying. We insisted that the careful examination of objective evidence should be of concern to those who oppose physician-assisted dying on moral grounds, to those who support aid-in-dying but are uneasy about the possible social consequences of legalization, to proponents of legalization who assume that the risks for vulnerable patients are heightened if these practices remain underground, and to those who favor legalization but fear that vulnerable patients will be denied a privilege reserved for better situated patients and that health care inequities already affecting vulnerable persons will be exacerbated--that is, to both those who do and those who do not find physician-assisted dying objectionable on moral grounds.

We certainly do not claim that people not in the 10 groups identified as vulnerable might not seek physician assistance in dying for the "wrong" reasons--disturbed emotional states, reactions to loss, personality types, and other factors Finlay and George mention. That is not the focus of our paper. Finlay and George provide as it were a supplemental look at the same issue, but with an entirely different focus. However, our paper does indirectly address some of their concerns. Not only do we examine rates of assisted dying in depression (rates of depression are elevated in people seeking assistance in dying, but may not be elevated in people receiving it), but in conditions like physical disability, stigmatized illnesses such as AIDS, chronic nonterminal illness, and psychiatric illness (all conditions in which physician-assisted suicide and euthanasia may be legal in the Netherlands). All may all be associated with distressed emotional states and loss, but, with the exception of AIDS (largely prior, it can be noted, to the development of highly active antiretroviral therapy), in none of these conditions are rates of assisted dying elevated. We did not assume that the categories we examined identified were the only respects in which individuals could be "vulnerable" and we did not attempt to distinguish between relevant and, as Finlay and George charge, "irrelevant" vulnerable groups; rather, we examined the categories we did because they had played such a prominent role in public and professional argumentation over the risks of legalization.

Although our paper examined data from both Oregon and the Netherlands, Finlay and George address only that of Oregon. Thus they do not recognize the relevance of data about chronic (nonterminal) illness or disability, conditions in which a patient may legally seek the assistance of a physician in dying in the Netherlands but not in Oregon. They misread our analysis of data about old age, claiming that we omit deaths among Oregonians aged between 65 and 84 years; this data does indeed not appear in the discursive text but is clearly displayed in Table 1, and supports our claim that there is no evidence of heightened risk of assistance in dying among the elderly (construed as age 65 and over). We do not see a "systematic error" here, though the fact that very elderly people die less frequently of cancer (the most frequent condition associated with assisted dying in both Oregon and the Netherlands, approximately 80%) may explain the some of the finding.

Finlay and George also refer to a "four-fold rise" in physician assistance in dying in Oregon between the initial years of the Death With Dignity Act and the present. Patients ingesting lethal medications represented 6/10,000 deaths in Oregon in 1998, the first full year of legalization, 9/10,000 deaths in Oregon in both 1999 and 2000, and 21/10,000 deaths in Oregon in 2010. Some increase subsequent to the first year of legalization represents persons who received prescriptions from the previous year dying in the subsequent year. It is a misleading representation of the trend to call this a four-fold increase.

Most of Finlay and George's comments are not really about our data and analyses; rather, they mainly concern the limitations of our study. A careful reading of our study will show that most of these concerns have already been addressed, and that we have been particularly careful to refrain from conjectural claims about the motivations of people who died with physician assistance, claims that are, in contrast, central to the concerns of Finlay and George. We welcome further attempts to examine what really happens where physician assistance in dying is legal, both concerning individuals who are members of groups identified in the literature as vulnerable and those who are not, including people with high SES and other indicators of comparative privilege. However, we do not welcome the seemingly ideologically biased assumption evident in the Finlay and George critique that requesting or receiving such assistance is itself a symptom of vulnerability; such an assumption would make it impossible to examine the facts of the matter in either Oregon or the Netherlands in any objective way.

Conflict of Interest:

None declared