The BIAS FREE Framework

Objective: To test the applicability of the BIAS FREE Framework in African settings. Participants: Researchers from the Tanzanian National Institute for Medical Research, university and community-based researchers from Tanzania, the Gambia and South Africa. Setting: National Institute for Medical Research, Dar es Salaam Tanzania. Intervention: An intensive two-day workshop to examine the applicability of the BIAS FREE Framework within an African setting. This involved clarification of the following concepts: construction of knowledge, objectivity, logic of domination, hierarchy, power, sex and gender, disability, and race/ethnicity. The Framework identifies three types of bias problems that derive from social hierarchies based on gender, race and disability: maintaining hierarchy, failing to examine differences, and using double standards. Participants used the 20 diagnostic questions at the heart of the Framework to analyze various research publications, including some authored by participants. Outcomes: Participants uniformly stated that the Framework is useful for uncovering bias in public health research, policy and programs; that it is immediately applicable in their work settings; and that doing so would improve equity in research and, ultimately, in health. One participant re-analyzed published data using the Framework and submitted a supplementary report with some new recommendations. Implications: The applicability of the BIAS FREE Framework has been demonstrated in diverse settings. It is now being offered for broader application as a tool for uncovering and eliminating biases in health research that derive from social hierarchies and for addressing the persistence of global health inequities. MeSH terms: Research design; research methodology; objectivity; logic of domination; prejudice; social discrimination; gender bias; sexism; ableism; racism In a recent issue of the CJPH, there was a call for health research to adopt “a human rights and health perspective” that “brings the language of health and medicine face to face with the language of power and social inequality”.1 It is well established that much research, including health research, suffers from biases that derive from sexism, racism, ableism and other social hierarchies.2-8 There have been a number of instruments – such as gender-based analysis – developed to avoid the presence of bias in research but these tools have focussed on only one type of social hierarchy – typically gender and, occasionally, race or disability.9-18 However, sexism and racism not only intersect with each other but with other social hierarchies like class and sexual orientation. The BIAS FREE Framework is an innovative tool designed to provide a unified approach to detect biases that derive from any and all social hierarchies. BIAS FREE is an acronym for Building an Integrative Analytical System For Recognizing and Eliminating InEquities. The Framework identifies three major forms of biases – maintaining hierarchy, failing to recognize differences and using double standards – and employs a set of 20 analytical questions to alert users to their presence in research. It also points the way to preventive or counterbalancing solutions (see Tables Ia-c). The Framework is anchored in 25 years of published research by the two authors18-29 and is informed by the voluminous research on gender, racist and ableist biases, some of which is cited here.8,18,30-37 The Framework synthesizes insights from several bodies of research literature that are usually not considered together and presents a pioneering universal matrix for detecting biases in research derived from social hierarchies. The Framework has been applied primarily in search of bias stemming from gender, race and dis/ability hierarchies. While it is applicable to biases deriving from other social hierarchies (e.g., sexual orientation, social class, area of residence), such applications have not yet been fully explored. The Framework has been tested and applied in a number of countries and settings. It has been included in undergraduate and graduate coursework at the La traduction du résumé se trouve à la fin de l’article. 1. Ontario Institute for Studies in Education, University of Toronto, Toronto, Ontario, Canada 2. Global Forum for Health Research, Geneva, Switzerland Correspondence and reprint requests: Mary Anne Burke, Health Analyst/Statistician, Global Forum for Health Research, 1-5 route des Morillons, PO Box 2100, 1211 Geneva 2, Switzerland, Tel: + 41 22 791 3808, Fax: + 41 22 791 4394, E-mail: [email protected] Acknowledgements: The support of the Global Forum for Health Research is gratefully acknowledged. We also thank Jane E. Foy for her editorial commitment and support. JANUARY – FEBRUARY 2006 CANADIAN JOURNAL OF PUBLIC HEALTH 63 University of Toronto and Calgary in the period 2002-2005, and in Germany in 2005. Workshops have been conducted in a number of countries, including Canada, Costa Rica, Germany, India, Mexico, Senegal, Switzerland and Tanzania, and at the United Nations as part of an Ad Hoc Committee Meeting on the UN Convention on the rights of disabled persons, where it was well received.38 More workshops are planned. This article reports on the outcome of one intervention only: a workshop conducted in Dar Es Salaam, Tanzania, in May 2005. The workshop was conducted in Africa to test whether the Framework was as applicable in Africa as in highincome countries, where it was developed. PARTICIPANTS, SETTING