Prejudice in medicine
Our role in creating health care disparities
Can Fam Physician
Vol. 54, No. 11, November 2008, pp.1511 - 1513
Copyright © 2008 by The College of Family Physicians of Canada
John Guilfoyle, MD FCFP Family physician in Sioux Lookout, Ont, and an Associate Professor for the Northern Ontario School of Medicine and the University of Manitoba Len Kelly, MD MClinSc FCFP
Family physician and an Associate Professor of Family Medicine at the Northern Ontario School of Medicine and McMaster University in Sioux Lookout
Natalie St Pierre-Hansen
Research intern at the Northern Ontario School of Medicine in Sioux Lookout
Correspondence: Dr J. Guilfoyle, Northern Ontario School of Medicine, Box 489, Sioux Lookout, ON P8T 1A8; e-mailfjguilfoyle@mac.com
How welcome is the patient in our office and waiting room who is different from ourselves? Does the patient who is less educated, poor, or from a different culture feel comfortable? How do these differences affect their care, if in fact they do?
Current emphasis on evidence-based medicine and its application in the arenas of clinical treatment and health policy development have been noteworthy. Its antithesis, prejudice-based medicine, is not as well-studied and is not considered as carefully in current medical practice. Prejudice—making assumptions and decisions based on inaccurate or faulty information and assumptions—is the stuff of history. At various times we have shown ourselves capable, through a variety of psychological maneuvers, of parlaying spurious data about the objects of prejudice into discriminatory practices. This discrimination has run the gamut, from denial of various rights and services to genocide. No one is immune from prejudice; thus, it is hardly surprising that it has an effect on how we practise medicine.
Recent work, particularly by the Institute of Medicine—an American nonprofit organization that provides evidence-based information and advice on matters of medicine and health—has suggested that prejudice and discrimination directly affect the receipt of much needed health care services in certain groups. Commissioned by the US Congress to study racial and ethnic disparities in health care, they found that health care providers’ behaviour, assumptions, and attitudes can have a detrimental influence on the health of those who seek care.1
What is prejudice?
Prejudice is both an attitude and a cognitive process, the identifiable and measurable outcome of which is the practice of discrimination. A standardized definition of discrimination does not exist. Even within the framework of the law definitions vary: some jurisdictions focus on intent and others on effect.1 For the purposes of health care, the Institute of Medicine defines it as "differences in the quality of healthcare that are not due to access-related factors or clinical needs, preferences, and appropriateness of intervention." It identified discriminatory health care practices on 2 levels: the health care structure (systemic discrimination) and discrimination that results from "biases, prejudices, stereotyping, and uncertainty in clinical communication."1
Many American studies have demonstrated that minorities in the United States receive lower quality health services and are less likely to receive medical procedures than white Americans are.2–5 Disparities in access and quality of care exist even when income and sociodemographic factors are controlled.2 African Americans, for example, are less likely than whites to receive surgery for early stage lung, colon, or breast cancer.3 A study by Laditka et al of delivery outcomes in South Carolina (N = 26 866) revealed substantially higher rates of "potentially avoidable delivery complications" in African Americans and Hispanic Americans compared with whites, even when health care insurance was controlled.4 Similarly, a study by Chung et al of deliveries in Long Beach, Fla, (N = 37 688) found that blacks and Hispanics were 75% and 22% more likely to undergo cesarean deliveries, respectively.5
In Canada
Despite Canada’s universal health care system, health disparities on the basis of race and immigrant status persist. A handful of Canadian studies have found that aboriginal and foreign-born Canadians face barriers to access. There is ample discussion around these barriers, as well as the importance of effective communication strategies, but there are few studies that examine the measurable differences in health care delivery based on race or ethnicity.
In a 2004 study, Tonelli et al (N = 4333) discovered that aboriginal renal dialysis patients were significantly less likely to receive renal transplantations than nonaboriginal patients, even after adjusting for potential confounders (hazard ratio 0.43, 95% confidence interval 0.35 to 0.53).6 A later study by Tonelli and colleagues (N = 835) found that aboriginal patients were half as likely to be activated on the transplant waiting list, owing to difficulties in the process of completing the workup rather than because they were medically unsuitable (P < .01).7
Heaman and colleagues’ 2005 study8 compared survey responses about prenatal care received by aboriginal and nonaboriginal women in Manitoba (N = 652). Aboriginal women were 4 times more likely to receive inadequate prenatal care than nonaboriginal women were (15.7% vs 3.6%). After controlling for relevant confounders a difference still persisted. When socioeconomic status was factored into the analysis, the authors discovered that the most important predictor of inadequate care was poverty.8 In 2007, a province-wide study of prenatal care in Manitoba (N = 149 291) examined disparities among geographic districts: the highest rates of inadequate prenatal care were in the lowest income neighbourhoods.9 These areas also had the highest proportion of recent immigrants and aboriginal populations.10
On the other hand, some Canadian studies found no differences in care based on race or ethnicity. Reime et al, for example, found no significant association between ethnicity and treatment of patients (N = 20 488) in neonatal intensive care units.10 Wenman and colleagues’ comparison of aboriginal women and nonaboriginal women (N = 2047) revealed that aboriginal women were more likely to have low-birth-weight infants; however, after controlling for socioeconomic status, smoking, and poor nutrition, the differences were not statistically significant.9,11
Poverty, culture, and access to care
These studies highlight the interconnectedness of ethnic or racial disparities and socioeconomic status as they relate to quality of health care and health care outcomes. Confounders aside, these disadvantaged groups face barriers to health care. New federal initiatives are beginning to address access-related barriers for aboriginal people in Canada. For example, the federal government is responding to First Nations and Inuit access barriers in remote and isolated areas by establishing patient wait times guarantees.12 In 2006, the diabetes care pilot project was announced, followed by the prenatal care pilot project to increase early interventions.12
Janet Smylie, past chair of the Society of Obstetricians and Gynaecologist of Canada’s Aboriginal Health Issues Committee, underscores the importance of examining the surrounding sociopolitical factors that affect health, rather than simply attributing explanations to the "aboriginal" variable.13,14 Similar barriers to access were identified for foreign-born Canadians who face language and cultural barriers to accessing care.11,15 One study examined the sociodemographic factors associated with low rates of cervical cancer screening (N = 24 584). The lowest rates of screening were seen in areas with the highest immigrant population.16 The length of time living in Canada was also an important contributing variable: Papanicolaou smear rates for recent immigrants were 36.9% compared with 60.9% for other immigrants.16,17
The disparities in the health of minorities in Canada—particularly with regard to aboriginal health—are well documented.17,18 In 2000, the life expectancy of First Nations men was recorded as 7 years shorter than the overall national life expectancy for men, and in 1997, the prevalence of diabetes among First Nations women 65 years of age and older was documented as being more than 3 times the national figure.19 In fact, the rate of chronic illness overall among aboriginal people is 3 times higher than the national average.20
The complexity of these reduced health outcomes requires the implications of sociopolitical and historical factors be considered. Analysis of the marginalization that emerges from inequities in health care systems examines the interconnectedness of these factors.
Caregiver attitudes
Canadian studies of caregiver attitudes and measurable differences in care based on ethnicity are rare. No Canadian studies were found when using the search term prejudice on major research databases. Some scholars argue that by focusing on attitudinal and behavioural levels of discrimination for explanations, we are overlooking the contribution of systemic discrimination in creating persistent inequities.13 Nevertheless, inquiries isolating the role of prejudice on the part of the care-giver might provide us with valuable insight into ourselves and our institutions.
Scott Plous, author of Understanding Prejudice and Discrimination, suggests that prejudice is amenable to intervention strategies.21 These include education and other efforts to expose those who practise prejudice to information and role play that allow exploration of issues and the development of attitudes to prevent discrimination.21 The developing concept of cultural safety has emerged in continuing health education and institutional practices. The concept, which developed out of the nursing education context in New Zealand, is predicated on the understanding that a caregiver’s own culture, and the assumptions that follow, affect how a clinical encounter plays out, subsequently affecting the patient’s care. This approach acknowledges that all attitudes and behaviour—extending beyond blatant displays of prejudice and measurable discriminatory practices—can have serious implications for minority care and health.
Naming and blaming strategies toward caregivers are not constructive; rather, the development of a climate where prejudice is recognized as possible is a priority so that efforts directed toward understanding and empathy can be emphasized and reinforced. Concomitant measures within health care systems to remove barriers that can create discrimination would also be helpful.
Defining and measuring attitudes and behaviour is difficult.
Understanding the complex determinants of minority health is
perhaps even more challenging; blatant acts of prejudice or
racism are only a small part of the problem. Studies that isolate
these aspects of care (such as those seen in the American body
of literature) provide measurable outcomes, which serve as valuable
reference points in the Canadian context. When confounders such
as socioeconomic status and education levels are controlled,
we might be left with a stark mirror image of caregiver attitudes.
The determination of relevant indicators of discrimination,
the measurements of those indicators, and regular reporting
of progress will assist efforts to reverse the deleterious effects
of discrimination.
