Ethnicity_ Spatial Equity and Ut

Document Sample
Ethnicity_ Spatial Equity and Ut Powered By Docstoc
					          Ethnicity, Spatial Equity and Utilization of Primary Care Physicians
         - A Case Study of Mainland Chinese Immigrants in the Toronto CMA

                                          Final Report

                                           Submitted to

                             CERIS - The Ontario Metropolis Centre

                                            Lu Wang
                                    Department of Geography
                                       Ryerson University
                               350 Victoria Street, Toronto, Canada
                                         December, 2007


The study concerned ethnicity, spatial equity and healthcare access in the context of diversity and
integration. It examined gaps and barriers, both spatial and socio-cultural, facing Mainland
Chinese (MLC) immigrants with regard to accessing primary care physicians, who are often
heterogeneous in their cultural background and language of practice. The MLC immigrants are the
most important source of immigrants to Canada since 1998. A random telephone questionnaire
survey and seven focus groups were conducted to explore the experiences of MLC in utilizing
primary care physicians. Geo-referenced Census data and physician data were utilized in a spatial
analysis of accessibility to family physicians. The primary data revealed a distinct preference
among MLC immigrants for linguistically-matched physicians and their unique health
management strategies in the host country. The spatial and quantitative analysis suggests a certain
degree of spatial inequality among Chinese immigrants in accessing culturally sensitive care. The
study yields important methodological and policy implications.


The research examined the spatial and socio-cultural barriers facing MLC immigrants in utilizing
family physician resources. MLC immigrants are the most important source of immigrants to
Canada since 1998. They account for 2.9% of the Toronto CMA’s total 2001 population. The
majority of MLC immigrants came after the Tiananmen Square incident in 1989. Because these
MLC immigrants have not been in Canada long, and because they come from a different culture
with a different healthcare system, one of the greatest challenges they face in adapting to Canadian
society is access to primary care services including those provided by family physicians. Toronto’s
primary care physicians are a particularly diverse group, speaking 100 different languages and
dialects other than French and English (College of Physicians and Surgeons in Ontario (CPSO),
2005). Among the primary care physicians in the Toronto CMA, over 6% self reported to speak
Chinese languages such as Cantonese, Mandarin and Hakka. The characteristics of MLC
immigrants and the diversity of the physicians provide an excellent opportunity to explore issues
of spatial equity in immigrant healthcare access.

The study employed mixed methods to examine the experiences of Mainland Chinese (MLC)
immigrants in the Toronto Census Metropolitan Area (CMA) in utilizing primary care physicians.
Attention was focused on family physicians as they are the gate keepers to specialists in Canada.
The research has three specific objectives:

1) to investigate the choices among MLC immigrants between Chinese-speaking family
   physicians and other family physicians,
2) to identify factors that impact access to family physicians,
3) to analyze geographic accessibility to family physicians and identify under-serviced areas, and
4) to examine how MLC immigrants’ health management strategies impact physician utilization
   and vice versa.

Both primary data and second data were collected and analyzed in order to achieve the research
objectives. A random telephone questionnaire survey was administered to MLC immigrants in
both the city and the suburbs. The survey explored the physician-seeking behaviour such as
frequency of visiting a family physician, physician location, communicating language with the
physician, average wait time, and alternative medicine used. The survey revealed a distinct
preference among MLC immigrants for linguistically-matched family physicians. Apart from the
quantitative survey data collected, seven focus groups were conducted to provide qualitative
information on the experiences of MLC participants in utilizing primary care services including
family physicians, barriers to access, and the role of culture and health belief in health
management strategies.

The study also utilized two secondary datasets: (i). the geo-referenced Census data on the
settlement patterns of MLC immigrants, and (ii) the geo-referenced physician data obtained from
the College of Physicians and Surgeons in Ontario (CPSO). The physician data provides
information of the location of physicians in the Toronto CMA, language spoken by a physician,
graduating university and other attributes. Combining the survey data on MLC immigrants’
physician seeking behaviour, the geo-referenced physician data and Census data, the study
performed quantitative analysis of geographic accessibility to family physicians and identified
spatial inequality among MLC immigrants in access linguistically-matched family physicians.
Qualitative analysis was conducted based on focus groups and field observation to reveal barriers
for MLC immigrant to utilizing physician resources and their unique health management
strategies. By employing mixed methods, the study provided a detailed portrait on how MLC
immigrants integrate in the domain of health and yielded important policy implications, which will
be addressed in this report.


2.1 Qualitative research
Seven focus groups were conducted in Mandarin Chinese involving 42 MLC immigrants. The
discussions were centred around physician-seeking experiences, barriers with regard to access to
primary care services, experiences in adapting to Canadian health care system and Western
medical practices, and levels of satisfaction in using primary care services. The participants were
drawn from both the city and the suburbs to encompass a range of social and income classes.
Among the seven focus groups, three group consisted of young and middle-aged newcomers (20-
29 years old) with an average of length of residency in Canada of 1.4 years, three groups were
composed of middle-aged long-standing immigrants (30-59 years old) who had stayed in Canada
for 5.5 years on average, and one group included exclusively senior immigrants who are over 60.
The participants were identified and chosen partly through a snowball sampling method. The
research assistants also sought help from community organizations in participant recruitment, for
example, COSTI Language & Skills Training Services in Markham and Rolia Chinese Cultural
Association that runs an Internet forum that has over 40,000 registered MLC users in the GTA.
The focus groups were audiotaped and transcribed.

2.2 Quantitative research.

2.2.1 Questionnaire survey
A random survey was conducted over the telephone in the summer of 2006, through a 15-20
minute telephone interview in Mandarin Chinese. Respondents were identified and recruited from
the Canada Phone CD-ROM 2006, for a total of 104 useful responses. The Canada Phone CD-
ROM 2006 lists the names, phone numbers, and addresses of subscribers to Bell Canada. Chinese
subscribers were identified based on their last names. The total number of phone calls made was
4360. However, in most of the cases the phone call was not answered and in some cases the
number was invalid. Among those answered phone calls, the survey achieved a response rate of
approximately 22.5%. Two screening questions were asked during the survey to confirm whether
the respondent was indeed a first-generation immigrant from Mainland China, and that the
respondent had a regular family physician. If the answer to both questions was yes, the interviewer
went on to ask the remaining questions. The questions included, but were not limited to, the
respondent’s residential and work location, information about his/her family physician such as
language at practice. The survey also provided information on the respondent’s socio-economic
and demographic background such length of stay in Canada, educational background, average
family income, age, approximate travel time to physician’s office, and car ownership.

Using the survey data, the latest geo-referenced 2001 Canadian Census on MLC settlement
patterns and geo-referenced physician data, the research performed a series of quantitative and
spatial analysis in modeling geographic accessibility to family physicians.

2.2.2 GIS-based accessibility modeling
This part of research project on GIS-based accessibility modeling was published in a refereed
journal Health and Place in 2007. The accessibility models used in this research originated from
the classic gravity-type spatial interaction model that considers two aspects of spatial behaviour: (i)
attractiveness of destination such as number of opportunities, and (ii) disincentives to reach a
destination (e.g. journey distance, driving time) that are typically incorporated in a distance
impedance function. A series of modifications and improvements were made to the classic model
in the research to better capture the nuisance of the spatial behaviour among MLC immigrants in
seeking family physician resources. Recent literature on modeling accessibility also provided
useful insights on the improvement of existing accessibility models (Kwan 1997; Wang and Luo
2003). The models used in this research include: (1) the modified gravity-model (Joseph and
Bantock 1982; Luo and Wang 2003; Shen 1998), (2) the FCA (Floating Catchment Area) method
(Luo and Wang 2003; Wang and Luo 2005); (3) Sub-accessibility measure I on accessibility of
Chinese immigrants to ethnic Chinese physicians; and (4) Sub-accessibility measure II on
accessibility of Chinese immigrant subgroups to Mandarin- and Cantonese-speaking physicians.
Findings of accessibility analysis will be discussed in the next section.

2.2.3 Calibrating accessibility models using survey data
This part of research project utilized survey findings to further improve the gravity-type and FCA
accessibility models, and was the focus of Ms. Deborah Roisman’s Masters Major Research Paper
supervised by the PI and defended in September, 2007. In the literature, the distance friction
parameter in gravity models and accessibility modes has always been treated as fixed. Yet the
distance friction parameter is the most important element of a distance impedance function. It
reflects the most prominent aspect of spatial behaviour – travel threshold. The survey conducted in
the project provided valuable information on how far a MLC immigrant was willing to travel to
access a family physician. This information allowed to relax the travel time threshold by applying
varying beta (i.e. distance friction parameter) for city and suburban residents. This depicts a more
realistic picture of spatial travel and is a significant improvement in modeling geographic
accessibility. In both 2.2.2 and 2.2.3, travel time along street network was calculated in ArcGIS
and O-D (origin-destination) matrix was constructed to compute accessibility index.


3.1 Utilization of primary care physicians: patterns and experiences
The focus groups and survey that explored the experiences among MLC immigrants in utilizing
and accessing primary care services are still being analyzed. Some preliminary findings are
included in this report. In analyzing the transcripts, some patterns and themes with regard to
physician utilization and health experiences emerged:
(1)     Language and preference for Chinese-speaking physicians. In the survey,
approximately 88% of the respondents reported choosing Chinese-speaking family physicians, and
about half of them communicated with their physicians in Mandarin (the main Chinese dialect
spoken in Mainland China). Two thirds had stayed in Canada for 6 years or less, about half have at
least a university degree, and over 70% drove to the physician’s office. In the focus groups, the
vast majority of the participants who had a family physician were seeing a Chinese-speaking
doctor. Language was mentioned in the focus groups as the most important factor for choosing a
Chinese-speaking physician. Most participants preferred to communicate with their physicians in
Chinese as English “medical terminologies” were deemed “difficult” to comprehend and they
could communicate more “effectively” with Chinese-speaking physician and better understand
his/her diagnosis and instructions. New comers and seniors in the focus groups were particularly
challenged by this language barrier.
(2)     Culturally specific health beliefs. In Chinese culture, a balance of yin and yang (or cold
and hot) results in harmony in the body and good health; an imbalance leads to poor health or
disease. When describing symptoms, vocabularies such as yin/cold and yang/hot were often used
and understood by some Chinese-speaking physicians. Medicinal food and herbal medicines
provided by local traditional Chinese herbalist doctors were widely consumed.
(3)     Health management strategy. The focus groups revealed that self-diagnosis and self-
treatment using medicines brought from China and hot-cold balanced homemade meals are
especially common in managing health for MLC immigrants especially those who arrived to
Canada more recently.
(4)     Transnationalism and health care. The transnational aspect of MLC immigrants’ health
beliefs and health management strategy is particular interesting. Many survey respondents and
focus group participants mentioned bringing medicines, both Western and traditional, from country
of origin to Canada. The trans-border flow of health care products points to an important
dimension of transnational movement in the domain of health. The constant reflection on the
traditional health belief in terms of yin and yang balance even when seeking Western health care
can be viewed as another important element of transnationalism - transnational consciousness that
is created and re-created through the mind, and the awareness of multi-locality that stimulates the
desire to connect oneself with others and to engage in particular transnational activities (Ghosh and
Wang 2003; Vertorec 1999).

3.2 Findings from GIS-based accessibility modeling
The research applied and modified the widely used gravity-type accessibility measures, of which a
special type is the so-called floating catchment area (FCA) method, in evaluating three types of
geographical accessibilities in a GIS environment: (1) the general accessibility of Toronto
residents to family physicians (Figure 1); (2) the accessibility of Chinese immigrants to ethnic
Chinese physicians (sub-accessibility I) (Figure 2); and (3) the accessibility of MLC immigrants to
Mandarin- speaking Chinese physicians and the accessibility of HKC immigrants to Cantonese-
speaking Chinese physicians (sub-accessibility II).

Figure 1: FCA general accessibility to all family physicians   Figure 2 FCA accessibility of Chinese immigrants to
                                                               ethnic Chinese family physicians (sub-accessibility I)

While general accessibility decreases from city core to the outer suburbs, due to the concentration
of physicians and major hospitals in the City, sub-accessibility I increases when moving toward the
suburbs. The competition among co-ethnic residents in the City reduces the likelihood of accessing
Chinese physicians, even though Chinese physicians have a fairly heavy presence in the City. In
the Chinese immigrant community, HKC immigrants enjoy the highest accessibility to Cantonese-
speaking physicians and MLC immigrants have relatively low accessibility to Mandarin-speaking
physicians. This is due to the relative abundance of Cantonese-speaking Chinese physicians in

3.3 Findings from calibrating accessibility models using survey data
Accessibility scores for MLC immigrants to all family physicians and Chinese-speaking family
physicians were calculated for each census tract in the Toronto CMA using two different models:
the improved two-step floating catchment area (2SFCA) model and the gravity-type model. In the
2SFCA model, two different travel time thresholds were used – 9 minutes for urban areas and 13
minutes for suburban areas. The survey revealed two distinct two travel time thresholds for the city
and suburban respondents. In the enhanced gravity model, two values for the impedance parameter
β were used – a β of 18 for urban areas and a β of 37 for the suburbs, corresponding to the two
travel time thresholds in the 2SFCA model. When comparing scores between the two study areas
within each model, it is clear that even though urban scores are lower than the suburban ones on
average, there is far less variability than in the suburbs (Figure 3). The suburban area has pockets
of higher accessibility to the west (Mississauga, Brampton) and north (King, West Gwillimbury)
along with regions of low accessibility (Markham, Richmond Hill), whereas the distribution of
scores in the city is far more uniform, perhaps due to the concentrated pattern of both Chinese-
speaking family physicians and MLC immigrants. The patterns more or less confirm to the
accessibility patterns discussed in 3.2.

                            Figure 3 2SFCA accessibility of MLC immigrants

The research will yield important policy implications for identifying health professional shortage
areas for culturally-diverse populations, developing health programs, addressing issues related to
foreign-trained physicians, formulating health promotion strategies and enhancing primary-care
delivery relevant for MLC immigrants. The research findings will provide knowledge concerning
locational, cultural and linguistic aspects of the pool of physicians available in Toronto, thus
facilitating the search of MLC immigrants for appropriate physicians. The findings may also be
used by primary care physicians and clinicians to enhance their cultural understanding of
immigrant patients, a critical factor in addressing the health needs of immigrant populations.

5.1 Refereed contributions
The research has produced one published contribution in an important refereed health journal:
   Wang Lu, 2007, “Immigration, Ethnicity, and Accessibility to Culturally Diverse Family Physicians”
   Health and Place 13: 656-671
A manuscript that draws materials from this research project is currently under review by another
flagship health and social science journal.

5.2 Graduate supervision and Masters Major Research Papers (MRPs)
A Masters Major Research Paper was supervised by the PI focusing on modeling accessibility
using primary survey data (see 2.2.3 and 3.2). The MRP was successfully defended.
   Roisman, Deborah, 2007, “Modelling spatial accessibility to primary health care among mainland
   Chinese immigrants in the Toronto CMA”. Masters Major Research Paper, MSA program (Masters in
   Spatial Analysis), Ryerson University

The PI presented the some of the findings of the research projected in two conferences in 2007:
 Wang L, 2007, “A GIS approach to analyzing health care utilization” Paper presented at the 2007
 OAGEE (Ontario Association of Geographers and Environmental Educators) Conference. Toronto,
 Canada. October
 Wang L, 2007, “Ethnicity, Immigration and Access to Family Physicians” Paper presented at the Annual
 Meetings of the Association of American Geographers, San Francisco, USA, March

7.1 Masters MRPs
As mentioned in 5.2, Ms Deborah Roisman completed her Masters in Spatial Analysis (MSA) by
writing a major research paper on GIS-based accessibility modeling.
7.2 Research assistants
Two PhD students were employed as research assistants for the project. Ms Hongxia Shan (OISE,
University of Toronto) helped tremendously with primary data collection. She conducted the
telephone survey and several focus groups among Mainland Chinese immigrants. Ms Yuanyin
Huan (Geography, York University) conducted spatial analysis using geo-referenced Census data
and physician data. Ms. Wafa Raza (Ryerson University) helped collect literature on immigrant
health and mental health. Two community research assistants (Ms. Melody Meng and Ms. Teresa
Wang) also contributed to the project by recruiting participants for two focus groups, conducting
the focus groups and transcribing focus group discussions.


Shared By: