The Public Endorses Collection of Ethnicity Information in Hospital: Implications for Routine Data Capture in Canadian Health Systems
A telephone survey was conducted in Calgary, Alberta to assess public opinion on collection of ethnicity information in hospitals. Of the 2,799 respondents, 84.8% felt comfortable about recording their ethnicity in hospital charts. This rate held across respondents' age, marital status and ethnic origin. These findings suggest that Canadian health systems should explore the feasibility and ethical suitability of collecting ethnicity data, as this information could contribute to the evaluation and subsequent reduction of ethnic disparities in health and health services access.
[To view the French abstract, please scroll down]
About four million Canadians consider themselves visible minorities, and ethnic minority populations continue to grow in numbers due to immigration (Statistics Canada 2004a). Because of this continuing growth, understanding and reducing health disparities and promoting equality for ethnic minority populations have become top priorities in Canadian research and public policy (Canadian Institute of Health Research 2004). However, ethnic variation in population health, health services utilization and outcomes is little studied in Canada, largely because hospital charts and most routinely compiled health databases do not record information on ethnicity. We conducted a random survey of residents in a large Canadian city to assess how the general public feels about the routine collection and compilation of individual-level data on ethnicity in hospital records.
Data for this study were collected through a cross-sectional telephone survey using a structured questionnaire in the city of Calgary, Alberta between September and December 2003. We randomly selected only primary phone numbers from the 2003 Calgary telephone directory for the survey. Information on the type of telephone number - whether it is a primary or secondary residential number - is flagged in the directory. We interviewed one respondent, aged 18 or over, at each telephone number in either English, two major Chinese dialects (Mandarin and Cantonese) or four South Asian dialects (Hindi, Urdu, Punjabi and Gujarati). These languages were chosen because Chinese and South Asian Canadians are the two largest visible minority populations in Calgary (Statistics Canada 2004a).
Our survey questionnaire was developed in English. Forward and backward translations were performed to ensure that the meanings in the translated versions were consistent. A different translator performed each translation. The survey collected data on age, sex, marital status, birthplace, ethnicity and respondents' opinions regarding the routine collection (in hospital records) of information on ethnicity. Self-perception of ancestry of origin was determined by the following question: "People living in Canada come from many different cultural and racial backgrounds. Would you describe your ethnic origins as …" Following the question, various ethnicities were listed for selection. We also asked respondents to describe their ethnicity in their own words. The respondent's opinion on providing ethnicity information to hospitals was collected by asking: "If you were admitted to a hospital, would you feel comfortable having your ethnic or cultural background recorded in hospital charts?" Respondents were given a choice of "Yes," "No" or "Don't Know." Simple descriptive and chi-square statistics were used to describe our findings.
Of 6,585 telephone numbers dialled, 5,124 people were contacted and 3,021 were surveyed, among whom 2,799 had complete data amenable to analysis (55% of the 5,124 individuals contacted). Among these 2,799 respondents with complete data, 84.8% (95% confidence interval: 83.4% to 86.1%) felt comfortable recording their ethnicity in hospital records. This proportion did not vary by age, marital status or race/ethnicity (Table 1). Persons born in Canada (86.2%) were more supportive than immigrants (80.3%).
Our study demonstrates that a majority of Calgarians support the collection of ethnicity information in hospital records. This finding suggests that the collection of data on ethnicity in health system databases should be further explored, given the clear importance of this personal variable as a determinant of health and healthcare delivery.
Our study did find that about 15% of respondents reject the concept of recording ethnicity in medical records. Indeed, it must be acknowledged that some consider ethnicity a sensitive and private matter, and when that is the perception among some individuals, a case could be made against the widespread collection of such information. In Canada, however, the collection of private or sensitive information is not new. Statistics Canada started to collect ethnicity information as early as its first national census in 1871 (Statistics Canada 2005). Two recent national health surveys - the National Population Health Survey (Statistics Canada 2004b) and the Canadian Community Health Survey (Statistics Canada 2003) - also collect ethnicity data. Furthermore, hospital charts and their accompanying administrative databases routinely record patient names and, in some instances, religion, which some would argue is at least as sensitive and personal as ethnicity.
To overcome the lack of ethnicity information in many databases, researchers have recently proposed the use of patient surnames as a proxy for defining ethnicity (Quan et al. 2004; Lauderdale and Kestenbaum 2000). Paradoxically, however, this amounts to using names, arguably the most sensitive personal variable in health databases, as a proxy for the perhaps less sensitive variable of ethnicity. Indeed, we anticipate that many individuals would object to having their names used to define ethnicity because a truly personal variable is being used, and also because of the potential for misclassifying ethnicity (e.g., as could occur with the existing name-ethnicity algorithms when a person's last name is "Lee").
A case can indeed be made for the merit of collecting individual-level information on ethnicity in health records and accompanying health system databases. Such information would facilitate research into the health of ethnic minority populations and their access to health services. Health systems in the United States have routinely collected information on race in health records and administrative data, and this information has permitted the compilation of a considerable body of knowledge on racial disparities in health and health system access (Long et al. 2004). Such a body of knowledge provides a foundation for interventions to reduce disparities. Canada, meanwhile, has maintained a commitment to focusing on ethnicity rather than race, as ethnicity more comprehensively represents the cultural factors (beyond skin colour) that are passed through families from generation to generation and that may influence health and health services access and quality.
The validity of ethnicity information has been questioned because of ambiguity surrounding the definition and meaning of ethnicity to patients. The concept of ethnicity is complex. Definitions are multidimensional and may include language, culture, physical appearance, religion, nationality, self-perception and ancestors' place(s) of origin. In our survey, when we asked respondents to describe their ethnicity in their own words, a majority reported a single ethnic origin. However, some described their ethnicity from various perspectives, for example, "Canadian," "Catholic," "Islamic," "Christian," "Diverse," "Caucasian Black," "mixed Norwegian, English and Native American," "Spanish, Chinese and American" and "mixed nationalities."
In the literature, ethnicity is often defined by self-perceived ethnic identity, ancestral origin or both. The terms "race" and "ethnicity" are frequently used interchangeably in the classification of race/ethnicity although they are not synonymous. Race mainly refers to differences of biology, and ethnicity to differences of culture and geographic origin (Caldwell and Popenoe 1995). In hospital administrative data, the United States has employed a single category of race/ethnicity that includes six groups: American Indian, Asian, Black, Hispanic, White and Unknown or Missing (Moy et al. 2005). The United Kingdom has used the categories White, Black, Caribbean, Black African, Black Other, Indian, Pakistani, Bangladeshi, Chinese and Other (National Statistics 2001). New Zealand has used New Zealand European, Maori, Samoan, Cook Island Maori, Tongan, Niuean, Chinese, Indian and Other (such as Dutch, Japanese, Tokelauan) (Statistics New Zealand 2002).
Statistics Canada defines ethnicity in the Census as
the ethnic or cultural group(s) to which the respondent's ancestor belongs. An ancestor is someone from whom a person is descended, and is usually more distant than a grandparent. Ethnic origin pertains to the ancestral "roots" or background of the population and should not be confused with citizenship or nationality. (Statistics Canada 1998)
Even though the definition in the Census has not changed through the years, the classification of ethnicity has been updated several times. In the 1991 Census, 15 of the most frequent origins were listed under the question: "To which ethnic or cultural group(s) did this person's ancestors belong?" and respondents were asked to mark as many as were applicable (see Table 2). In addition, two blank spaces were provided for respondents to write other responses that might have been applicable. In the 1996 Census, four blank spaces were provided for respondents to write in their origins, and 24 categories of such origins were provided as illustrations. In addition, "Canadian" was included for the first time as one of the categories because it was the fifth most frequently reported origin in 1991. The 2001 Census provided four blank spaces for respondents to write in their origins and provided 25 categories, the first 21 of which were based on the frequency of single ethnic origin counts from the 1996 Census. The National Population Health Survey (Statistics Canada 2004b) and Canadian Community Health Survey (Statistics Canada 2003) collected ethnicity information on cultural/racial background using 13 categories and ethnic/cultural ancestry using 19 categories.
Validity of ethnicity information is also related to the process of collection. The information should be recorded based on self-report. However, hospital clerks or clinicians may hesitate to ask for information on ethnicity because they may feel that such questions are discriminatory and offensive to patients (Caldwell and Popenoe 1995; Moscou et al. 2003). This scenario could result in some staff's guessing a patient's ethnicity from name or appearance, or copying the ethnicity from previous records without confirming or updating it. Nevertheless, several studies have shown that the agreement between self-report and administrative data is high for Asian, Black or African American and White race/ethnicity, and relatively low for American Indian and Hispanic race/ethnicity (Moscou et al. 2003; Gomez et al. 2005; Kressin et al. 2003; Boehmer et al. 2002; Swallen et al. 1997).
This study has limitations. Because the survey was conducted in one Canadian city (Calgary), generalizability of the study findings to other regions is unknown. However, we believe the rate of agreeing to record ethnicity in hospital charts is likely to be high in Canada, as nearly all respondents in the CCHS survey answer ethnicity questions. Our non-response rate was 45%, which may bias our findings. In the introduction to the telephone survey in the community, we stated that the survey was being conducted for research purposes. However, recording information on ethnicity in hospitals or clinics would have many purposes beyond research, such as administration, healthcare quality improvement and provision of culturally sensitive services. It is possible that respondents would have viewed the collection of information on ethnicity even more favourably had they been explicitly informed of such potential uses of ethnicity data. Future research, perhaps using qualitative methods, would be required to clarify respondents' perspectives on these nuances.
We hope that this simple study will encourage hospitals, health regions, the Canadian Institute for Health Information and the general public to engage in dialogue regarding the feasibility and ethics of beginning to collect ethnicity data at an individual level. The collection of such information has considerable potential to enhance our understanding of disparities in health and health system access and quality, so that interventions can be designed and implemented to contribute to the health of ethnic minority populations. A consistent format of gathering race/ethnicity information should be considered to achieve comparability among Canadian databases.
Le public approuve la collecte de renseignements sur l'origine ethnique dans les hôpitaux : répercussions sur la saisie systématique de données dans les systèmes de soins de santé canadiens
Un sondage par téléphone a été mené à Calgary, en Alberta, afin d'évaluer l'opinion publique relativement à la collecte de renseignements sur l'origine ethnique dans les hôpitaux. Des 2 799 répondants, 84,8 p. cent ont dit ne pas avoir d'objection à ce que leur origine ethnique soit inscrite dans leur dossier de patient. Ce pourcentage s'est maintenu chez tous les répondants, quel que soit leur âge, leur état civil et leur origine ethnique. Ces résultats suggèrent que les responsables des systèmes de santé canadiens devraient explorer la faisabilité et le bien-fondé éthique de la collecte de renseignements sur l'origine ethnique, puisque de tels renseignements pourraient aider à évaluer et, subséquemment, à réduire les disparités ethniques dans l'accès aux soins et aux services de santé.
About the Author(s)
Hude Quan, MD, PhD Department of Community Health Sciences University of Calgary, Calgary, AB
Alison Wong, MSc Centre for Health and Policy Studies University of Calgary, Calgary, AB
Delaine Johnson, MA Calgary Health Region, Calgary, AB
William A. Ghali, MD, MPH Department of Community Health Sciences Centre for Health and Policy Studies Department of Medicine University of Calgary, Calgary, AB
AcknowledgmentThe conduct of this study was supported by operating grants from the Calgary Health Region and the Institute of Health Economics, Alberta, Canada. Dr. Quan is supported by a Population Health Investigator Award from the Alberta Heritage Foundation for Medical Research, Edmonton, Alberta and by a New Investigator Award from the Canadian Institutes of Health Research. Dr. Ghali is supported by a Health Scholar Award from the Alberta Heritage Foundation for Medical Research, Edmonton, Alberta and by a Government of Canada Chair in Health Services Research.
Correspondence may be directed to: Dr. Hude Quan, Department of Community Health Sciences, University of Calgary, 3330 Hospital Dr. NW, Calgary, AB T2N 4N1; tel.: 403-944-8912; fax: 403-944-8950; e-mail: email@example.com.
Boehmer, U., N.R. Kressin, D.R. Berlowitz, C.L. Christiansen, L.E. Kazis and J.A. Jones. 2002. "Self-Reported vs. Administrative Race/Ethnicity Data and Study Results." American Journal of Public Health 92: 1471-1473.
Caldwell, S.H. and R. Popenoe. 1995. "Perceptions and Misperceptions of Skin Color." Annals of Internal Medicine 122: 614-617.
Canadian Institutes of Health Research. 2004. Listening for Direction II: National Consultation on Health Services and Policy Issues for 2004-2007. Retrieved December 9, 2005. < http://www.cihr -irsc.gc.ca/e/20461.html >.
Gomez, S.L., J.L. Kelsey, S.L. Glaser, M.M. Lee and S. Sidney. 2005. "Inconsistencies between Self-Reported Ethnicity and Ethnicity Recorded in a Health Maintenance Organization." Annals of Epidemiology 15: 71-79.
Kressin, N.R., B.H. Chang, A. Hendricks and L.E. Kazis. 2003. "Agreement between Administrative Data and Patients' Self-Reports of Race/Ethnicity." American Journal of Public Health 93: 1734-1739.
Lauderdale, D.S. and B. Kestenbaum. 2000. "Asian American Ethnic Identification by Surname." Population Research and Policy Review 19: 283-300.
Long, J.A., V.W. Chang, S.A. Ibrahim and D.A. Asch. 2004. "Update on the Health Disparities Literature." Annals of Internal Medicine 141: 805-812.
Moscou, S., M.R. Anderson, J.B. Kaplan and L. Valencia. 2003. "Validity of Racial/Ethnic Classifications in Medical Records Data: An Exploratory Study." American Journal of Public Health 93: 1084-1086.
Moy, E., I.E. Arispe, J.S. Holmes and R.M. Andrews. 2005. "Preparing the National Healthcare Disparities Report: Gaps in Data for Assessing Racial, Ethnic, and Socioeconomic Disparities in Healthcare." Medical Care 43 (Supplement 3): I9-I16.
National Statistics. 2005. Census 2001. Retrieved December 9, 2005. < http://www.statistics.gov.uk/census2001/cn_143.asp >.
Quan, H., W.A. Ghali, S. Dean et al. 2004. "Validity of Using Surname to Define Chinese Ethnicity." Canadian Journal of Public Health 95: 314.
Statistics Canada. 1998. "1996 Census: Ethnic Origin, Visible Minorities." Retrieved December 9, 2005. < http://www.statcan .ca/Daily/English/980217/d980217.htm >.
Statistics Canada. 2003. "The Canadian Community Health Survey (CCHS): Extending the Wealth of Health Data in Canada." Retrieved December 9, 2005. < http://www.statcan .ca/english/ concepts/health/cchsinfo.htm >.
Statistics Canada. 2004a."Ethnocultural Portrait of Canada: Highlight Tables, 2001 Census." Retrieved December 9, 2005. < http://www12.statcan.ca/english/census01/products/ highlight/Ethnicity >.
Statistics Canada. 2004b. "National Population Health Survey (NPHS)." Retrieved December 9, 2005. < http://www.statcan.ca/english/concepts/nphs/index.htm >.
Statistics Canada. 2005. "History of the Census of Canada." Retrieved December 9, 2005. < http://www12.statcan.ca/ english/census01/ Info/history.cfm >.
Statistics New Zealand. 2005. "Change in Ethnicity Question - 2001 Census of Population and Dwellings." Retrieved December 9, 2005. http://www.stats.govt.nz/census/change-in-ethnicity-question.htm.
Swallen, K.C., D.W. West, S.L. Stewart, S.L. Glaser and P.L. Horn Ross. 1997. "Predictors of Misclassification of Hispanic Ethnicity in a Population-Based Cancer Registry." Annals of Epidemiology 7: 200-206.
Be the first to comment on this!
Personal Subscriber? Sign In
Note: Please enter a display name. Your email address will not be publically displayed