Health, Wealth, and Happiness in Canada

Saskatoon, October 31 - November 1, 2016
Sheraton Cavalier Saskatoon Hotel

Health, wealth and happiness are states of being to which most Canadians aspire. How well are Canadians doing in achieving those aspirations? Is Canada becoming more or less equitable? Join us in Saskatoon tolearn more about these important questions from more than 35 researchers. For more details,see the conference program.

Presentations

Developments in Statistics Canada surveys

• The Survey of Household Spending (SHS)
  Julie Charest, Statistics Canada
  This presentation will provide an overview of the SHS and the data available in RDCs.
  
  
• Longitudinal Immigration Database (IMDB)
  Scott McLeish, Statistics Canada
  The IMDB is an administrative dataset that combines administrative immigration data with annual tax data. It provides comprehensive information on the socio-economic outcomes of the immigrant taxfiler population in Canada. A pilot project to support the introduction of the IMDB in the RDCs is under consideration. This presentation will provide an overview of the database, its individual components, and the analytical possibilities it affords.
  
  
• Redesign of the Canadian Community Health Survey (CCHS) – main component
  Andrew McKenzie, Statistics Canada
  The presentation will provide an overview of the significant changes brought to this survey in January 2015 (including changes to questionnaire content, a new sampling frame and allocation methodology, a new collection strategy and a revised dissemination plan), explain the implications for time series comparability and prepare researchers to use the new data once it is released (the first data will be released on November 15th, 2016).
  
  
• Education data in the RDCs and Building the Education Longitudinal Linkage Platform (ELLP)
  Christine Hinchley, Statistics Canada
  The Centre for Education Statistics of Statistics Canada is proposing to pilot putting the postsecondary information system (PSIS) annual and longitudinal files in the RDCs, as well as any additional linked files based on the PSIS data, such as files used for the graduate outcomes pilot project. Since Statistics Canada has been involved in linkage projects using the postsecondary information system data, there has been increasing pressure from stakeholders and researchers to have access to the PSIS and related linked file. The data will be discussed during the presentation.
  
  
• 2006 Census Linked to the Discharge Abstract Database: New linked data for hospital research
  Gisèle Carrière, Statistics Canada
  Statistics Canada, together with The Canadian Institute for Health Information (CIHI), will be conducting a pilot project to provide access to the 2006 Census linked to the Discharge Abstract Database (DAD) for 2006/07-2008/09. The pilot project will give researchers access to these linked data in RDCs and will aim to evaluate and improve confidentiality vetting rules and help to improve documentation for researchers. The data will be discussed during the presentation.
  
  
• Canadian Health Measures Survey (CHMS): an introduction and overview
  Joanne Boisjoli, Statistics Canada
  Information on the objectives and components of CHMS, including the CHMS Biobank, will be presented as well as on the data files, on analytical products and metadata that have already been released, and on what can be expected in the future.

Earnings and income

• Profit sharing and workplace productivity growth: Does team work play a role?
• Tony Fang, Memorial University and University of Toronto
• Using panel data, this paper examines whether there is any link between adoption of an employee profit sharing plan and subsequent productivity growth in Canadian establishments, and whether this relationship is affected by various contextual factors, particularly use of work teams.
• We utilize a longitudinal research design which compares within-firm productivity growth in the three-year and the five-year periods subsequent to profit sharing adoption to within-firm productivity growth during the same time periods in firms which did not adopt profit sharing. We find significant positive effects of profit sharing adoption on workplace productivity growth in firms that had work teams in place at the time of profit sharing adoption, but not in firms without work teams.
  
  
• Quantifying the effects of social structure on earnings trajectories in North America using PSID and SLID
  Charles Plante, MGill University
  Sociologists have long claimed that people occupy social positions within a wider social structure. These various positions are determined by lengthy processes of socialization that result in people fitting in and being capable of succeeding in certain social milieu and not in others. It follows that large and, especially, unexpected changes in social position can reduce a person’s life chances. Earnings are an important indicator of both social position and achievement in modern societies.If sociologists are right about social structure, then we should expect large changes in earnings, negative or positive, to undermine long run earnings prospects. In this paper I evaluate this proposition using dynamic panel data methods and estimation, and longitudinal earnings data in PSID and SLID. I find robust evidence to suggest that the sociologist may be onto something: large changes in earnings do in fact induce a reduction in earnings trajectories over time.
  
  
• Income-Based Inequities in Access to Psychotherapy
  Mary Bartram
  To what extent does income determine access to psychotherapy? Access to psychotherapy in Canada is currently provided through a two-tier system, with public financing of a narrow range of services that are either provided by physicians, in hospitals, or in publicly-funded clinics, and private financing (either through employment-based insurance or out-of-pocket payments) of the broader range of services that are provided by psychologists, social workers, and other mental health professionals.
  This presentation will highlight the results of a large-N study of the distributive impacts of this exclusion, using data from 2012 Canadian Community Health Survey-Mental Health and Annual. Concentration indices are calculated to show the effect of income on access to psychotherapy, broken out by service providers that are and aren’t covered by public health insurance. This is complemented by a comparison of the effect of income on rates of unmet need for health care for physical vs mental health problems. In order to be comprehensive, both perceived and evaluated mental health needs are included in the model, as well as potential and realized access. To explore possible differences in policy impacts for mental health and addictions issues, the distribution of access to services for substance use problems and other mental health problems is also compared.
  
  
• The Survey of Household Spending (SHS)
  Julie Charest, Statistics Canada
  This presentation will provide an overview of the SHS and the data available in RDCs.

Education

• What role do skills and parents play in explaining the gender gap in university participation?
  Kelly Foley, University of Saskatchewan
  University participation among women has been increasing over the last three decades such that now in Canada more than half of all new degrees are awarded to women. Recent research has suggested that young men are also falling behind in their grades and educational aspirations during high school. Both grades and aspirations reflect many different individual characteristics and socio-economic circumstances.
  In this paper, I use a factor model to decompose the underlying sources of the gap in educational aspirations between boys and girls at age 15, and quantify the role those factors play in determining university participation by age 21. In particular, I focus on cognitive skills, personality characteristics and parental attitudes toward education. I find that all three of these factors play an important role in explaining both aspirations and university participation. Additionally, differences between boys and girls in the distribution and impact of these factors can fully account for the gap in university participation.
  
  
• Education data in the RDCs and Building the Education Longitudinal Linkage Platform (ELLP)
  Christine Hinchley, Statistics Canada
  The Centre for Education Statistics of Statistics Canada is proposing to pilot putting the postsecondary information system (PSIS) annual and longitudinal files in the RDCs, as well as any additional linked files based on the PSIS data, such as files used for the graduate outcomes pilot project. Since Statistics Canada has been involved in linkage projects using the postsecondary information system data, there has been increasing pressure from stakeholders and researchers to have access to the PSIS and related linked file. The data will be discussed during the presentation.

Health

• Uncovering inequities in access to health care services among adult Canadians with chronic back disorders
  Brenna Bath, University of Saskatchewan
  Chronic back disorders (CBD) are a prevalent and costly public health issue. Despite back disorders being the most common pain problem in the general population and one of the most common reasons for seeking health care, little is known regarding the comparative use of family physician, chiropractic, and physiotherapy services in Canada. Elucidating these differences may help to identify potential gaps in access to care and inform the development of strategies to optimize equitable access.
  The research objectives were to investigate patterns of use and to profile the sociodemographic and other factors associated with self-reported use of family physician, chiropractor and physiotherapy services among adult Canadians with CBD. The combined 2010 and 2011 Canadian Community Health surveys were used to investigate self-reported health care use among adults with CBD. This complex survey employs population weights and bootstrapping to be representative of the Canadian population.
  Descriptive analyses along with multiple logistic regression was used to control for possible confounding. The majority of respondents with CBD sought care only with a family physician (53.8%) with 20.9% and 16.2% seeking care with combined chiropractic/family physician or physiotherapy/family physician, respectively. Very few respondents sought care only with a chiropractor (2.5%) or physiotherapist (1.0%). After adjustment, differential patterns of utilization (p<0.05) among those with CBD are evident between provider groups with respect to age, gender, socioeconomic status, residence, functional limitations, and presence of co-morbidities. These differences highlight potential inequities in access to physiotherapy and chiropractic services among adult Canadians with CBD.
  
  
• Can stress process explain immigrants' mental health deterioration? Results from the Canadian Community Health Survey
  Fei-Ju Yang, McMaster University
  Previous research shows that recent Canadian immigrants' mental health is better than their long-term counterparts, but this mental health advantage does not last, an empirical
  phenomenon called 'healthy immigrant effect (Ali, 2002)'. Despite numerous studies investigating the existence of healthy immigrant effect, potential factors leading to immigrants' mental health deterioration remain underexplored.
  Guided by the stress process theory, this study investigates two major pathways, social support and interpersonal strain, which potentially affect the association between years of migration and immigrants' mental health. The measures of mental health of this study include three distinctive psychological constructs, including psychological distress, positive mental health, and self-rated mental health. Based on 4,282 immigrant samples extracted from the Canadian Community Health Survey-Mental Health 2012, results suggest that health immigrant effect exists for psychological distress, but its existence for positive mental health and self-rated mental health is weak.
  Results also show that, compared to recent immigrants, long-term immigrants experience 'better support but higher interpersonal strain.' Though long-term immigrants are blessed with higher social support, the beneficial effects of social support do not counteract the deleterious effects of interpersonal strain on psychological distress. Future research should focus on identifying sources of interpersonal strain and social support in order to have more effective intervention on reducing immigrants'post-settlement stress.
  
  
• Comorbid substance use and major depression: Disability and risk of suicide in a nationally representative sample
  Vivian Onaemo, University of Saskatchewan
  Studies have shown that major depressive disorder (MDD) is associated with a wide range of negative outcomes amongst individuals with substance use disorders (SUDs). This study aims to examine the degree of disability and suicide risk associated with comorbid diagnosis of major depressive episode (MDE) and SUDs (alcohol use disorders (AUD), cannabis use disorders (CUD), other drugs excluding cannabis use disorders (DUD)).
  We used data from the Canadian Community Health Survey (CCHS) – Mental Health, 2012, a national cross sectional survey of individuals (n = 25,113) aged 15 years and older living in the ten Canadian provinces between January and December, 2012. Diagnoses were based on DSM-IV criteria and 12-month disorders were considered for SUDs and MDE. Analysis was done using multilevel mixed effects logistics regression. Survey design effects were accounted for using survey weights. STATA 14 was used for the analysis.
  Compared to single diagnosis, individuals with comorbid MDE and DUD had significantly higher disability OR(95%CI): 10.39(3.35-32.21). However, disability resulting from comorbid AUD and MDE 2.99(1.51-5.91) was significantly lower; than that of MDE only 8.28(6.38-10.74). Individuals with a single MDE diagnosis only were twice as likely to report suicide while comorbid diagnoses were 2 to 5 times more likely compared to those without a psychiatric diagnoses. SUDs only were not associated with increased risk for suicide. There seem to be increased disability and risk for suicide in comorbid SUD and MDE. Effective integration between addiction and mental health services would improve client outcomes.
  
  
• Sex moderates the relationship between childhood abuse and internalizing and substance use disorders later in life
  Xiangfei Meng, McGill University and the Douglas Mental Health Research Institute
  Although some studies examined the moderating role of sex in the relationship between childhood maltreatment and mental disorders later in life, a number of them examined only one or two types of maltreatment on an individual mental disorder, for instance, depression, substance use. It is of considerable clinical and theoretical importance to have in-depth understanding what roles of different types of childhood abuse play out in a wide range of mental disorders among women and men using well accepted instruments on abuse and mental disorders.
  The present study aimed to meet the need by using a large nationally representative population sample to explore the sex effect of childhood abuse in mental disorders, and assess the moderating role of sex in the abuse-mental disorder relationship. Using data from the Canadian Community Health Survey 2012: Mental Health we sought to answer this question. Respondents with information on childhood maltreatment prior to age 16 were selected (N=23, 395).
  We found: i) strong associations between childhood abuse frequency and sex; ii) significant differences between men and women in terms of mental disorders; iii) strong associations between childhood abuse and mental disorders; and, iv) sex moderated the role of childhood abuse history on adulthood mental disorders. Females with a history of sexual abuse and/ or exposure to interpersonal violence were at a greater risk of alcohol abuse or dependence later in life.
  Intervention should occur as early as possible, and should help female victims of childhood sexual abuse and/ or exposure to interpersonal violence, and their families to build more constructive ways to effectively reduce the negative effects of these experiences. Recognition of the moderating role of sex on the relationship between childhood abuse history and mental disorders later in life may aid clinicians and researchers in providing optimal health services.
  
  
• Spatial distribution of physiotherapy use across Canadian health regions in relation to the distribution of physiotherapists
  Tayyab Ikram Shah, University of SaskatchewanEnsuring equitable access to non-physician care providers, such as physiotherapists, has traditionally received less research and policy attention in comparison to medical services. This is a potentially vulnerable scenario where the growing and diverse health service needs of an aging population and those with chronic conditions may not be adequately met, particularly in relation to mobility and pain management needs.The objective of this study was to examine the spatial distribution of self-reported physiotherapy use across Canadian health regions in relation to the distribution of physiotherapists. This cross-sectional study is based on the physiotherapy use question from the 2014 Canadian Community Health Survey (CCHS). The CCHS data is collected from persons aged 12 and over living in the 117 health regions; however, to avoid data suppression, a few health regions have been grouped with neighbouring regions and as a result, 97 health regions and region group are available for analysis. Regional proportions of self-reported physiotherapy utilization and socio-demographic factors (such as sex, highest level of education, unmet health care needs) were estimated using population weights. Distribution of physiotherapists (number of physiotherapists per 1000 people) was obtained from the Canadian Institute of Health Information’s Physiotherapy in Canada report (2010). A set of spatial mapping techniques was applied to explore the association between self-reported physiotherapy use and physiotherapy distribution at the health region level.Our findings indicate that variation in the distribution of providers, self-report use, and population health needs exists across health regions indicating potential access inequities to physiotherapy services.
  
  
• Income-Based Inequities in Access to Psychotherapy
  Mary Bartram
  To what extent does income determine access to psychotherapy? Access to psychotherapy in Canada is currently provided through a two-tier system, with public financing of a narrow range of services that are either provided by physicians, in hospitals, or in publicly-funded clinics, and private financing (either through employment-based insurance or out-of-pocket payments) of the broader range of services that are provided by psychologists, social workers, and other mental health professionals.
  This presentation will highlight the results of a large-N study of the distributive impacts of this exclusion, using data from 2012 Canadian Community Health Survey-Mental Health and Annual. Concentration indices are calculated to show the effect of income on access to psychotherapy, broken out by service providers that are and aren’t covered by public health insurance. This is complemented by a comparison of the effect of income on rates of unmet need for health care for physical vs mental health problems. In order to be comprehensive, both perceived and evaluated mental health needs are included in the model, as well as potential and realized access. To explore possible differences in policy impacts for mental health and addictions issues, the distribution of access to services for substance use problems and other mental health problems is also compared.
  
  
• Redesign of the Canadian Community Health Survey (CCHS) – main component
  Andrew McKenzie, Statistics Canada
  The presentation will provide an overview of the significant changes brought to this survey in January 2015 (including changes to questionnaire content, a new sampling frame and allocation methodology, a new collection strategy and a revised dissemination plan), explain the implications for time series comparability and prepare researchers to use the new data once it is released (the first data will be released on November 15th, 2016).
  
  
• 2006 Census Linked to the Discharge Abstract Database: New linked data for hospital research
  Gisèle Carrière, Statistics Canada
  Statistics Canada, together with The Canadian Institute for Health Information (CIHI), will be conducting a pilot project to provide access to the 2006 Census linked to the Discharge Abstract Database (DAD) for 2006/07-2008/09. The pilot project will give researchers access to these linked data in RDCs and will aim to evaluate and improve confidentiality vetting rules and help to improve documentation for researchers. The data will be discussed during the presentation.
  
  
• Canadian Health Measures Survey (CHMS): an introduction and overview
  Joanne Boisjoli, Statistics Canada
  Information on the objectives and components of CHMS, including the CHMS Biobank, will be presented as well as on the data files, on analytical products and metadata that have already been released, and on what can be expected in the future.

Immigration

• Immigrant post-migration human capital investment and income profile: A family perspective
  Amanda Lu, Statistics Canada
  While the majority of Canadian immigrants are married at landing, most previous studies concentrate on post-migration post-secondary education investment at an individual level. However, family level analysis is essential to understanding education investment decisions among married immigrants. Whom the immigrant couple chooses to invest in and when to invest are important to their lifetime family income maximization outcomes.
  This research utilizes the Longitudinal Administrative Databank to explore the patterns of investment in post- secondary education among married immigrant couples in Canada.
  The empirical results reveal that there is an apparent gender difference when only one person from the immigrant couple participates in university at any given tax year among the first ten years after landing. Female immigrants are less likely to be invested in, in the first five years after landing, compared to their partner. However, after five years since landing, this situation is reversed. The effect of immigrant couple education investment patterns on their family income are also discussed. Immigrant couples who ever participated in Canadian universities during the same tax year have the fastest income growth from 2000 to 2012. They also have the highest family income in 2012, thirteen years after landing. Immigrant couples who never invest in Canadian post-secondary education in the first ten years after landing have the lowest average family income in 2012, compared to immigrant families where either only one spouse registers in a Canadian university, or both of the spouses participate in Canadian education in any given tax year.
  
  
• Can stress process explain immigrants' mental health deterioration? Results from the Canadian Community Health Survey
  Fei-Ju Yang, McMaster University
  Previous research shows that recent Canadian immigrants' mental health is better than their long-term counterparts, but this mental health advantage does not last, an empirical
  phenomenon called 'healthy immigrant effect (Ali, 2002)'. Despite numerous studies investigating the existence of healthy immigrant effect, potential factors leading to immigrants' mental health deterioration remain underexplored.
  Guided by the stress process theory, this study investigates two major pathways, social support and interpersonal strain, which potentially affect the association between years of migration and immigrants' mental health. The measures of mental health of this study include three distinctive psychological constructs, including psychological distress, positive mental health, and self-rated mental health. Based on 4,282 immigrant samples extracted from the Canadian Community Health Survey-Mental Health 2012, results suggest that health immigrant effect exists for psychological distress, but its existence for positive mental health and self-rated mental health is weak.
  Results also show that, compared to recent immigrants, long-term immigrants experience 'better support but higher interpersonal strain.' Though long-term immigrants are blessed with higher social support, the beneficial effects of social support do not counteract the deleterious effects of interpersonal strain on psychological distress. Future research should focus on identifying sources of interpersonal strain and social support in order to have more effective intervention on reducing immigrants'post-settlement stress.
  
  
• Longitudinal Immigration Database (IMDB)
  Scott McLeish, Statistics Canada
  The IMDB is an administrative dataset that combines administrative immigration data with annual tax data. It provides comprehensive information on the socio-economic outcomes of the immigrant taxfiler population in Canada. A pilot project to support the introduction of the IMDB in the RDCs is under consideration. This presentation will provide an overview of the database, its individual components, and the analytical possibilities it affords.

Indigenous issues and data

• Cultural connectedness and self-rated health among First Nations people living off-reserve
  Jeffrey S. Denis, McMaster University
  Indigenous scholars often tout the benefits of cultural connectedness (Snowshoe et al., 2015) and traditional knowledges and practices (e.g., Alfred, 2005) for Indigenous health and well-being. Indeed, Chandler and Lalonde’s (1998, 2008) ground-breaking research showed that cultural continuity at the community level is associated with lower suicide rates on First Nations reserves. However, statistical analyses of culture-health associations among Indigenous people living off reserve have been mixed and seem to depend on the specific measures of culture and health, the datasets, and the populations of interest (e.g., Currie et al., 2013, 2015; Ryan et al., 2015, 2016a, 2016b; Wilson & Cardwell, 2012).
  Our paper contributes to this debate by using data from the 2012 Aboriginal Peoples Survey to investigate the impact of cultural connectedness – as measured by language ability, participation in traditional activities, time spent with elders, social support, and traditional spirituality – on the self-rated health of a representative sample of First Nations people 15 years and older living off reserve. Self-rated health is a widely used indicator of overall health and a valid predictor of mortality and morbidity (Bailis, Segall, & Chipperfield, 2003; Chandola & Jenkins, 2000).
  Employing ordinal logistic regression, we hypothesize that, controlling for well-known determinants of Indigenous health (such as income, education, and residential school attendance), greater cultural connectedness is associated with better self-rated health. While the impact of any single cultural variable may be limited, deep connections – as indicated by engagement in multiple aspects of Indigenous culture – likely play a health-promoting role.
  
  
• Correlates of participation in organized sports and physical activity among indigenous youth
  Piotr Wilk, Western University
  Although previous research has shown that social cohesion may promote physical activity, social cohesion at the individual level was not always differentiated from social cohesion at the community level, and studies were often limited to specific population subgroups or geographical areas.
  We addressed the above limitations through the use of a multilevel modelling approach and nationally-representative data from the 2009-2014 Canadian Community Health Survey. Physical activity level was operationalized as average daily energy expenditure; social cohesion was assessed by self-rated sense of belonging to the local community; and communities were represented by Canada’s Forward Sortation Areas. The sample included 245,150 respondents from 1,570 communities.
  Geographical location was found to explain a significant proportion (4.1%) of the overall variance in physical activity level. After adjusting for age, sex, household income, education and urban-rural status, both individual- and community-level social cohesion were found to be positively associated with physical activity (p<0.001 for both). Thus, efforts to promote social cohesion and integration within communities may also promote physical activity and overall health.
  
  
• Indigenous data landscape in Canada: An overview
  Tim Leonard, Indigenous and Northern Affairs Canada
  This presentation will summarize the government's priorities, the drivers of policy research, and a general scan of Indigenous data source.
  
  
• Aboriginal data at Statistics Canada
  Vivian O’Donnell, Statistics Canada
  The presentation will provide an overview of key Statistics Canada data sources on Aboriginal people in Canada, with specific focus on the Aboriginal Peoples Survey and the Census. Content of these sources, their strengths and weaknesses, as well as their research potential will be highlighted. Finally, as Statistics Canada is gearing up for the 2017 Aboriginal Peoples Survey, an update of the 2017 cycle and other new initiatives will be provided.

Social cohesion

• The association between social cohesion and physical activity in Canada: a multilevel analysis
  Piotr Wilk, Western University
  Habitual physical activity is associated with numerous health benefits, and yet approximately 80% of adolescents worldwide are insufficiently active. Several correlates of Indigenous youths’ participation in physical activity and sport have been recognized, however, there is a paucity of research examining cultural and socio-economic factors, especially those related to the context in which youth’s physical activities take place.
  The purpose of this cross-sectional study is to explore correlates of participation in organized sports and/or physical activity among off-reserve Indigenous youth. The analysis was limited to those who were between the age of 12 and 17, who attended elementary or high school and identified as having single Indigenous identity: First Nations, Métis and Inuit (N=4,910).
  Using logistic regression, we first assessed unadjusted effects of each of the correlates on participation in organized sports. Then, we examined the magnitude of the independent effects of these factors, controlling for the other effects. Sampling weights and bootstrap weights were used to account for the multi-stage sampling design employed in the 2012 APS. The results of the multivariate analysis suggest that, controlling for the effects of other variables, only youth’s sex, age, health status, time spend with Elders, his or her participation in extracurricular activities, as well as parental involvement in school activities and strength of family ties had statistically significant effects on participation in sport and physical activities.
  
  
• Internet and e-government use among Canadians
  Jeff Moon, Queen's University
  The Canadian Internet Use Survey (CIUS) was used to explore factors influencing Canadians’ use of the Internet, extending current research on the digital divide by studying this over time (2005 – 2012) and including analysis of Canadian’s interaction with government websites. Findings confirm the influence of demographic factors such as age, gender, marital status, income, education, urban/rural, and labour force status in determining likelihood of Internet access. Likelihood of accessing the Internet increased with higher education and income levels over all years.
  The likelihood of Internet use trended upward for university graduates and current students over the study period, while remaining relatively stable for other education levels. Likelihood of Internet use trended downward over time for the highest income quintile.
  The CIUS 2010 and 2012 were used to investigate Canadians' interaction with government websites. Our results showed that respondents with more education, higher household income, and those who were unemployed, were more likely to have interacted with government websites. These findings highlight a continued digital and democratic divide in Canada, demonstrated by inequality in levels of Internet access and use of E-government along a variety of demographic lines.
  
  
• Neighborhoods matter: Variation in food insecurity not explained by household characteristics
  Olayemi Olabiyi, Southern Alberta Institute of Technology
  This study investigated the relationship between the social and economic contexts of neighborhoods and household food insecurity. Four years of data (2007 through 2010) drawn from the nationally representative Canadian Community Health Survey, which measures food insecurity using the Household Food Security Survey Module, were matched with the 2006 Census. Economic and social indicators from Census tracts were aggregated to the level of provincial and territorial health regions.
  Using random intercept logistic multi-level modeling in a Bayesian environment, with household characteristics and health region characteristics as level 1 and level 2, respectively, it was found that 14% of the variations in food insecurity prevalence lies between neighborhoods. After controlling for relevant household-level predictors, the prevalence of female-lone parent led households in a neighborhood raised the population prevalence of food insecurity by 2% as did low average household income. Therefore, the social and economic contexts in which households reside contribute independently to increased food insecurity among their residents. They reveal important differences in quality of life across Canadian provinces and territories.