The ethics of research involving Canada's Aboriginal populations

• Published in: Canadian Medical Association journal (CMAJ) Vol. 172; no. 8; pp. 977; author reply 977 - 977
• Main Author: Smylie, Janet
• Format: Journal Article
• Language: English
• Published: Canada CMA Impact Inc 12-04-2005; CMA Impact, Inc; Canadian Medical Association
• Subjects: Bioethics; Canada; Disease - ethnology; Ethics, Research; European Continental Ancestry Group; Female; Health Services, Indigenous; Humans; Indians, North American; Inuits; Kidney Failure, Chronic - ethnology; Kidney Failure, Chronic - mortality; Kidney Transplantation - ethnology; Letters; Patient Participation; Pregnancy; Pregnancy Outcome - ethnology; Proportional Hazards Models; Renal Dialysis; Risk Factors; Canada; Biomedical and Behavioral Research
• ISSN: 0820-3946; 1488-2329
• DOI: 10.1503/cmaj.1041676

"Aboriginality" is a social construct with little grounding in the day-to-day realities of the heterogeneous groups to which it refers. Tremendous cultural, historical, socioeconomic and political diversity exists between and within these groups. What is shared is the experience of colonization and the resultant legacy of poverty and social stressors. Use of this pan-ethnic term as one of several "risk" variables, while perhaps necessary to achieve adequate study power, devalues the unique experiences of First Nations, Métis and Inuit communities and perpetuates colonial processes of marginalizing or "pathologizing" Aboriginal peoples. Methodologically, the use of the "Aboriginal" variable in the study by [Marcello Tonelli] and associates' is problematic. Cox regression analysis assumes that the effects of different variables on survival are constant over time and are additive, but I am not sure that these assumptions are met with respect to the "Aboriginality" factor. Rather, there may be an interaction or multiplicative effect between "Aboriginality" and socioeconomic status, for example. In the study by Wenman and colleagues2 there are methodologic problems with the way in which ethnicity was determined. In addition, multiple-ethnicity responses, such as Caucasian and First Nations, were excluded, and the study was underpowered because the sample of First Nations and Métis women was insufficient. [Janet Smylie] also raises some methodologic concerns. We are unsure what is meant by her reference to "methodologic problems with the way in which ethnicity was determined," since no details are given about where we might have erred. We used standard questions and classified ethnicity according to current standards. Smylie is concerned about the exclusion of "multipleethnicity responses, such as Caucasian and First Nations." Only 3 First Nations and 3 Métis women also checked off Caucasian ethnicity, and these people were not coded as Caucasian because the categories were mutually exclusive for analytical purposes. In our initial analysis we separated out various ethnic groups, but during the peer review and revision process we were asked to present pooled results for our Table 1 and for the final logistic regression. Nevertheless, we did specifically discuss differences between ethnic groups in our Results section. With regard to potential underpowering, we acknowledged small numbers as a limitation of the study and understand that there may have been a lack of power to detect other potential differences.