Population Change and Lifecourse

Posters


Anaïs Simard-Gendron, Université de Montréal
Contextual and Individual Effects Behind Fertility Change in the West Bank and Gaza Strip

Contextual and Individual Effects Behind Fertility Change in the West Bank and Gaza Strip
Anaïs Simard-Gendron and Simona Bignami, Université de Montréal

Even though the fertility transition in the West Bank and Gaza Strip is well under way, it is clear that the classical theory of the demographic transition alone cannot explain the ongoing high demand for children in the modern yet conflicting context of the Palestinian territories. Individual-level variables have always been the main focus of studies on Palestinian fertility. However, the role of contextual variables is of central importance to best capture the mechanisms of fertility change in the region. To better understand the fertility behaviour of Palestinian women over time, we use the most recent retrospective data available from the Demographic and Health survey conducted by the Palestinian Central Bureau of Statistics in 2004 by modeling a multilevel discrete-time logistic regression on the birth histories of ever-married women aged 15-49 at the time of the survey. Regional characteristics representing the proportion of Jewish settlers, the status of women, and infant mortality are the three main contextual dimensions considered in this study. We argue that the status of women, especially through higher education, is the main factor behind the decline of Palestinian fertility, especially among older women. The decline in infant mortality only has a slightly negative impact on fertility. Finally, the presence of Jewish settlers contributes to decrease Palestinian fertility as regions with a higher proportion of settlers had a significantly lower fertility.

Angela Daley, Dalhousie University
The Well-Being of Adolescents in Northern Canada

The Well-Being of Adolescents in Northern Canada
Angela Daley, with Peter Burton and Shelley Phipps, Dalhousie University

The Territorial North (i.e. Yukon, Northwest Territories, Nunavut) is markedly different from the rest of Canada; yet there is little statistically reliable information about adolescent well-being in the region. The objective of this paper is to create a portrait of adolescent well-being in the Territorial North relative to Southern Canada. We do so using the Canadian Community Health Survey, a nationally representative dataset. We examine seven domains of well-being with 23 indicators by region and Aboriginal identity for youth aged 12 to 17. We include objective and subjective measures, reflecting the importance of adolescents’ perspectives in studies of their own well-being.

We find negligible differences among the non-Aboriginal population; while most indicators are substantially worse for Aboriginal youth, especially in Northern Canada (e.g. income, poverty, household education, family structure, crowding, food insecurity, exposure to second-hand smoke, school enrolment, smoking, sexual activity, obesity and overweight, oral and mental health). However, there are exceptions (e.g. physical activity, fruit and vegetable consumption, stress, body image, belonging). Nevertheless, Aboriginal youth in the North are generally less satisfied with life. This is not surprising since they fare worse in most well-being indicators considered in this study.

Keywords: well-being, adolescents, Aboriginal, Northern Canada

Astrid Flénon, Université de Montréal
The effect of acculturation on the health of new immigrants to Canada between 2001 and 2005

The effect of acculturation on the health of new immigrants to Canada between 2001 and 2005
Astrid Flénon and Alain Gagnon, Université de Montréal; Jennifer Sigouin and Zoua Vang, McGill University

When comparing the health of immigrants to the native-born, studies have found what is called a “healthy migrant effect” where immigrants are likely to have a health advantage compared to native-born individuals. In Canada, effect could partially be explained by the strict immigration criteria that select immigrants on their health status (Akresh and Frank, 2008). However, immigrants lose this advantage over time so that their level of health often deteriorates below the one of natives. This deterioration is an important issue for the health of populations in Canada and a challenge to adapt the health system to the needs of immigrants.

The Longitudinal Survey of Immigration to Canada (LSIC) provides an original way to assess the effects of acculturation, a process of adopting new cultural norms and practices, which has been often cited as one of the leading causes of immigrant’s health deterioration. The LSIC contains a cohort of 7716 landed immigrants in Canada between October 1st 2000 and September 30th 2001.

The objective of this paper is to analyze the effects of acculturation on immigrants’ general health and self-perceived mental health. The analysis is based on multivariate logistic regressions that control for pre-migration and post-migration factors which may potentially confound the relationship between acculturation and health. Our results show that acculturation outcomes proposed by Berry (1997) - integration, assimilation, separation, marginalization- influence the health of immigrants through socioeconomic variables such as education and financial status.

Key words: Immigrants, selection, acculturation, health, Canada.

Benoît Laplante, Institut national de la recherche scientifique
Pathways to adulthood in Uruguay

Pathways to adulthood in Uruguay
Ana Fostik, McGill University; Daniel Ciganda, Universitat Pompeu Fabra; and Benoît Laplante, Institut national de la recherche scientifique

This paper aims to analyse the pathways to adulthood in Uruguay by means of biographical data collected from the National Youth Surveys (NYS) carried out in 1990 and 2008. Two complementary strategies are used to accomplish this objective. Firstly, the evolution of typical sequences in the pathway to adulthood of Uruguayan young women from two cohorts (1961-1965 and 1979-1983) is analysed through sequence analysis. These analyses address four events indicating the pathway to adulthood: home leaving, school leaving, first job and first birth. Secondly, the variations in the timing of the first birth are analysed using the combination of the other three events by order of occurrence as the main independent variable. Such combination is introduced in the survival analysis as a time-varying variable and it enables identifying states that are more likely to delay or induce the first birth. Thus, this paper provides some keys to the changes in the pathway to adulthood, understood as a single process rather than a sum of events. Also, it allows understanding the relations between the types of pathways to adulthood and the first birth at a younger age. We find a high degree of heterogeneity in the pathways to adulthood, which is rooted in the unequal social structure. While some women experience fast routes to adult roles, others delay even the take off of the transition to adulthood. School insertion plays a key role in explaining these differences.

Bonnie McIntosh, University of Ottawa
Using the Cancer Risk Management Model to Predict the Long-term Impact of Colorectal Cancer Screening: A Study Protocol

Using the Cancer Risk Management Model to Predict the Long-­term Impact of Colorectal Cancer Screening: A Study Protocol
Bonnie McIntosh and Michael Wolfson, University of Ottawa

Background: Colorectal cancer (CRC) is the second leading cause of cancer death among adults in Ontario. Evidence suggests that CRC mortality rates can be reduced through early detection with screening. The Fecal Occult Blood Test (FOBT) and the Fecal Immunochemical Test (FIT) are screening tests that may be used in the early detection of CRC. There is evidence suggesting the FIT is a more effective screening test to implement in CRC screening interventions than the FOBT. Interventions are known to be most effective when founded on evidence-­‐based research. Model simulation allows researchers to implement evidence-­‐based research using ‘what if” scenarios’ by exploring the broad impact of CRC interventions within populations.

Objective: The purpose of this study is to document the process of using a microsimulation model known as the Cancer Risk Management Model (CRMM), while, also using this model to estimate the long-­term impact of CRC screening.

Methods: The CRMM is a comprehensive web-­based, dynamic micro-­simulation tool helping to guide cancer control decision-­making, planning and budgeting for those involved in public health. A case study approach will be implemented to document the process of using the CRMM to explore the long-term impact of CRC screening among those ages 50 to 74 and living in Ontario for the FIT, the FOBT, and no screening. The CRMM is a validated micro-­simulation model that is transparent and modular, and combines data from a wide range of sources. Therefore, model inputs will be altered at the user’s request to explore ‘what if scenarios.’ Projections will be generated by creating new scenarios (reflecting national data) from a base case scenario; results will then be extracted for the province of Ontario.

Results: Preliminary results will be presented.

Conclusion: Findings from this study will serve to directly inform the CRMM developers with feedback relating to the use of the model to explore an applied case, while, also informing cancer control decision-­making, planning and budgeting for government and public sector policy leaders, decision-­makers, and researchers.

Cassandra Cotton, McGill University
Does Mother's Migrant Status Affect Child Fostering in sub-Saharan Africa?: Evidence from Two Informal Settlements in Nairobi, Kenya

Does Mother's Migrant Status Affect Child Fostering in sub-Saharan Africa?: Evidence from Two Informal Settlements in Nairobi, Kenya
Cassandra Cotton, McGill University; and Donatien Beguy, African Population & Health Research Center

Children across sub-Saharan Africa reside in a variety of different living arrangements. In slum communities with high rates of circular migration and urban poverty, parents may choose alternative living arrangements for young children other than co-residence. Despite the importance of residence for child well-being, we know relatively little about the number of children out-fostered from slums and with whom they reside. Using birth history data from the Nairobi Urban Health and Demographic Surveillance System collected between 2005 and 2009, we will determine percentages of children under 15 living away from their mothers by mothers’ migrant status and duration of stay. We use logistic regression to analyze characteristics of migrant and non-migrant mothers in order to determine what may influence child fosterage out of Nairobi’s slum settlements. We find approximately 15% of children under age 15 live apart from their mothers, with mothers’ socio-demographic characteristics, child’s age, and mother’s migrant status associated with child fostering.

David Pelletier, Université de Montréal
The Diffusion of Cohabitation and Children’s Risks of Family Dissolution

Families headed by a cohabiting couple are often found to be less stable than those headed by a married couple. There is however much transnational diversity regarding the size of the stability gap between the two union forms. We make use of the enormous internal heterogeneity of union behaviour in Canada to explore in detail a hypothesis stating that the more cohabitation is common within a society, the more cohabiting unions are stable. Taking the point of view of children, we analyze the hazards of parental separation for children born to cohabiting or married parents between 1983 and 2008. To better represent the normative environment in which families live, we define both geographically- and culturally-based contextual units. We use individual-level data from the National Longitudinal Survey of Children and Youth as well as contextual-level data from national censuses. The results of our multilevel models suggest that cohabiting families do indeed become more stable with increasing diffusion of cohabitation but that the evolution of the marriage/cohabitation stability gap might not be linear. Moreover, the diffusion process accounts for about half of the contextual-level variance of the hazards of parental separation and other explanatory processes must thus be investigated.

Gunjan Sondhi, York University
Encounters in the transnational: care, responsibility and ‘doing family’

This paper engages with the theme of transnational care and responsibility by examining the different ways an international student migrant undertakes ‘doing family’ with their ‘traditional family at home, their siblings and with their friends in ‘host’ country. ‘Doing family’ like ‘doing gender’ involves a construction that is reproduced through repeated performances. The process and performance of doing family is cast against the idea of the ‘naturally’ existing set of relations. The discussion draws on 14 months of fieldwork in Toronto, Canada and New Delhi, India. In-depth semi-structured interviews were conducted with 65 respondents: 22 Indian international students studying at universities and colleges in Toronto, 23 family members/parents and 20 return international students in New Delhi. I mounted an online survey to facilitate data collection on the socio-economic background and educational profiles of the past and present students and their families. This paper draws on the interviews and ethnographic research in Toronto and New Delhi. The findings reveal that ‘doing family’ in the transnational context involves the interaction of different groups and individuals – some who would be defined as family in traditional models (parents, siblings) and others who would not (friends). It is a reciprocal process to which all members have to contribute and also receive from the relationship through activities of ‘caring’, ‘supporting’, and feeling responsible for anothers mental and physical well-being. Technology plays an important role in enabling migrants to be able to perform of the activities from afar with the same ease as they did when they were closer to their parents/siblings ‘at home’.

Kim Deslandes, Université de Montréal
Effects of Women's Attitudes Towards Divorce on Marital Instability in Rural Malawi

Some women in sub-Saharan Africa have successfully managed to promote their own agenda using marital strategies and, in return, have gained greater control over their lives. While we know which factors are likely to predict a union dissolution, we know little about the effects of women's attitudes toward divorce on marital stability in sub-Saharan Africa. To bridge this gap, we use data from the Malawi Longitudinal Study on Family and Health and between 1998 and 2008 to assess the effects of women’s attitude toward divorce on a series of contexts carrying large social and economic implications for a woman in rural Malawi. We use event history analyses to model the risk of divorce over union duration across three attitudinal levels, regions, and by birth cohort. We also use logistic regression analyses to capture the effects of women’s attitudes and socio-demographic and marital characteristics known to have an effect on divorce. Our results show that, overall, a more permissive attitude towards divorce is associated with a greater likelihood of divorce. We also find a variation in the odds of divorcing by contexts, which are influenced by traditional practices, and social norms.

Laura Wright, Western University
From Zero to Sixty in Four Decades but the Same Time to the Finish Line: Change and stability in the type and timing of first partnership across Canadian cohorts

Trends in the median age at marriage have been well documented, yet very little is known about median age at first cohabitation, especially among recent cohorts in Canada. Using the 2011 Canadian GSS, I document changes across birth cohorts in the type of first union Canadians form and assess whether increases in cohabitation have offset declines in marriage in Canada. I also examine regional and educational differences in the propensity of Canadians to marry or cohabit with their first partner and how these differences have changed over 50 years. Finally, I examine age at first union formation, at first marriage, and at first cohabitation to determine if the trend of delaying marriage extends to all types of partnerships in Canada. I find that the share of first unions that were formed through marriage has declined across cohorts in Canada, and that for Canadians born after 1970, cohabitation is the modal way to enter conjugal life. This trend towards cohabitation as first union has been more intense in Quebec than in the rest of Canada. Regional differences in prevalence of cohabitation as first union increased for Canadians born between 1930 and 1969, but the gap in cohabitation between Quebec and the rest of Canada has narrowed for the most recent cohort. Educational differences in the type of first union Canadians form are less dramatic than differences found in the U.S., and have changed across cohorts. Finally, I find that Canadians born in the 1970s have delayed their first marriage by nearly a decade compared to those born in the 1930s, but age at first union, whether marriage or cohabitation, has only increased by 2.5 years during the same time span. How Canadians form their first unions has changed dramatically, but when they form these unions has remained remarkably stable.

Marianne Caron, University of Montreal
Marriage patterns in historical perspective: What can we learn from three centuries of marriages in Québec?

The historical demographic literature on marriage has devoted a great deal of attention to understanding how and why age at marriage and the rate of celibacy were historically higher in North Western European countries. More recently, comparative work in a Eurasian perspective confirmed earlier and more universal marriages in Asia. On the other side, North America also had earlier ages at first marriages and lower rates of permanent celibacy than North Western Europe, but later marriage than Asia. This study will provide insights on marriage patterns in a region somewhere in the middle of North Western European and Asian marriage. We will explore how families and the context may have influenced jointly or separately the timing and probability of marriage in Quebec using three centuries of marriages and a large territory covering from sparsely populated and isolated parishes to densely populated cities like Québec and Montréal.

For this poster, we will use the richness of longitudinal data available in Quebec to study historical population from the early settlements up to 1912. For that period, there was a good registration of catholic marriages in the province of Quebec, a data source well exploited by the Programme de recherche en démographie historique (PRDH) at Université de Montréal for the years before 1800, and BALSAC project at the Université du Québec à Chicoutimi for the marriages after 1800.

Maxwell Hartt, University of Waterloo
Local Scale Population Projection Methods – Shrinking and Aging Communities

The emergence of a globalized economy has given rise to ‘global cities’ where knowledge, resource and human capital conglomerate – often at the cost of outmigration of resources in smaller cities. In the Canadian context, the growth of a few major centers is contrasted with many smaller and peripheral cities that may be coping with shrinking populations and economic decline. These effects are increasingly compounded by a second demographic transition, which is characterized by falling birth rates and an aging population. Continued loss of population, changing demographic structure, and economic decline can lead to a myriad of challenges, including underused infrastructure, high vacancy rates, and socio-economic inequality. As Statistics Canada’s population projections are limited to the provincial, territorial and national level, individual municipalities are left to calculate their own projections, which could be hindered by a lack of resources, the complexity of calculating local-scale migration rates, or simply may not be done. This paper reviews the methodological differences reflected in the approaches taken by various levels of government and concludes that more complex, time consuming and expensive models are used at higher levels of governance and in larger cities and are more likely to provide more accurate and precise results. Smaller and peripheral cities tend to use simpler, less time- and resource-intensive methods. An assessment framework of nine criteria concluded that the share capture method is the best methodological alternative for local scale population projection. The share capture model is applied to every municipality (with population above 10,000) in Ontario and projected dependency ratios are calculated to ascertain the future distribution of aging communities in Ontario.

Nicole Denier, McGill University
Leaving work, leaving home: job loss and socio-geographic mobility in Canada

The recent economic downturn magnified a routine occurrence in the Canadian labor market: involuntary job loss resulting from an employer downsizing, moving, or going out of business. Yet, in recent decades, even in times of economic expansion, rates of involuntary job loss have persisted across a wide-range of demographic and labor market groups. Moving is one way individuals may respond to job loss, either to relocate to cheaper housing or in search of work. Drawing on data from the 1996-2010 Survey of Labour and Income Dynamics, this article examines the relationship between involuntary job loss and geographic mobility in Canada, and further provides evidence on the types of neighborhoods to which individuals move. I find that involuntary job loss is associated with short- and long-distance mobility and increased risk of selection into materially deprived neighborhoods. Together, the findings establish job loss as both a key life course transition motivating residential mobility and long-distance migration in Canada, and as a trigger event that initiates entry into high deprivation areas.

Renée Carter, McGill University
Effect of Family Medicine Groups on visits to the emergency department among diabetics in Quebec between 2000 and 2011: A population-based segmented regression analysis of an interrupted time series

Effect of Family Medicine Groups on visits to the emergency department among diabetics in Quebec between 2000 and 2011: A population-based segmented regression analysis of an interrupted time series
Renée Carter, Amélie Quesnel-Vallée, McGill University; Céline Plante, Philippe Gamache, Institut national de santé publique du Québec; Jean-Frédéric Lévesque, Bureau of Health Information (New South Wales, Australia)

Background: Family Medicine Groups (FMG) were introduced in Quebec in 2002 to reorganize primary care practices and encourage team-based and inter-professional approaches to service delivery. We measured the effect of this reform on the rate of emergency department (ED) visits among patients diagnosed with diabetes.

Methods: Administrative databases were used to derive the weekly rate of ED visits between April 1, 2000 and March 31, 2012. We performed an interrupted segmented regression analysis to derive the estimated and extrapolated rates of visits in the years following the initial reform implementation. We employed an outcome control series of diabetics visiting the ED to treat appendicitis to strengthen the study’s internal validity.

Results: A gradual decline in the rate of visits was observed for short term diabetes related complications and total ED visits. After 9 years of reform implementation, we observed a reduction of 1.42 and 1.70 ED visits per 10,000 diabetics to treat short term complications in urban and rural areas, respectively. A steady decrease was also observed in the total rate of ED visits in urban areas where we observed a reduction of 6.72 visits per 10,000 diabetics 9 years following the reform. Visits coded for appendicitis showed no clinically relevant changes over the study period.

Interpretation: Our results suggest that the decreases in the rate of ED visits are attributed to the implementation of the FMG model across the province. The steady decline in the rate of total ED visits in urban areas is of particular relevance where overutilization of the ED is a problem. Evidence of these decreases despite the low-intensity nature of the FMG reform suggests the potential for this model to act as a future platform for implementing comprehensive care models for chronic disease management.

Stacey Hallman, Western University
Probabilistic Records Linkage using Historical Canadian Censuses and Vital Statistics

This poster presents the results of a records linkage project to analyze potential mortality risks for young adults in Ontario during the 1918 influenza epidemic. With a team of research assistants at Western University, McMaster University, and the Université de Montréal, death records were linked to birth records in order to determine exact date of birth for the calculation of exact age at death. This date of birth is then compared to age as listed on other historical documents, such as the 1901 and 1911 Canadian censuses. The family environment (or living situation) was transcribed for each individual at each census to examine the impact of socioeconomic conditions throughout the lifecourse on mortality in 1918.

Of the 23,183 deaths registered in Ontario between September and December 1918 and transcribed by the International Infectious Disease Data Archive at McMaster University, 3,316 individuals met the inclusion criteria. Both birth and death must have occurred in Ontario to establish exact date of birth and the individual must have died between the ages of 23 and 35 (born between 1883 and 1895). Of the 3,316 included death records, 2,965 were linked to at least one other record, giving a linkage success rate of 89.4%. This poster analyzes the linkage rates of the death records to birth records from 1883-1895 and the 1901 and 1911 Canadian censuses and uses logistic regression to investigate the important factors that precluded linkage. It evaluates declared age at all three time periods to discuss the suitability of these records for historical demographic analyses of past epidemic disease.

Valérie Jarry, Université de Montréal
Exceptional Longevity: Are Socioeconomic Conditions in Childhood Important?

Exceptional Longevity: Are Socioeconomic Conditions in Childhood Important?
Valérie Jarry, Alain Gagnon, and Robert Bourbeau, Université de Montréal

The close relationship between early childhood conditions and health and mortality outcomes in old age has been extensively studied in both the epidemiological and demographic fields. The channels through which early life is hypothesized to influence mortality are diverse and could be direct or indirect via adult characteristics. Despite the ample evidence on the influence of childhood conditions on overall longevity, less established in the literature is whether this association holds true within long lived families and whether this is a direct effect or an indirect effect. In this paper we investigate the association between socioeconomic and biodemographic factors in early life and mortality after age 40 and through which pathways this effect may operate. An event-history database that links individuals to their childhood characteristics gathered from the 1901 and 1911 Canadian census records and to their adult characteristics is used. Gender-specific parametric models with a Gompertz specification of the risk of mortality are used to estimate the effect of early life and adult variables on mortality risks after age 40. The link between early life conditions and later life outcomes is examined both with and without intermediary characteristics, aiming for a better understanding regarding to what extent the effect of exposure to an early life insult can be mediated. Overall, it is found that adverse childhood conditions leave an indelible direct and indirect imprint on adult mortality among long-lived men, while childhood conditions only exert a significant direct effect among men from the general population. The results also suggest substantial gender differences, with men being more sensitive to early life socioeconomic conditions and adult occupation than women.

Yann Décarie, INRS
L'impact de l'immigration au Québec sur l'effet "The pig in the Python" du baby-boom / Impact of immigration in Quebec on “The Pig in the Python” effect of the baby boom

L'impact de l'immigration au Québec sur l'effet "The pig in the Python" du baby-boom
Yann Décarie, Institut national de la recherche scientifique & Université de Montréal, Jean-Francois Picard, Yves Carrière, Jacques Légaré, Université de Montréal

On observe deux grandes phases de l’évolution démographique de l’après-guerre au Québec. La première étant le baby-boom qui concerne les générations très nombreuses nées entre 1946 et 1966 et qui fut plus important au Québec que dans la majorité des autres pays touchés par ce phénomène. La seconde étant le baby-bust, qui lui, concerne les générations nées après 1966. L’écho du baby-boom fut relativement peu important par la dimension des cohortes de naissances concernées, et nettement insuffisant pour rétablir un certain équilibre dans la dimension des cohortes de naissances, d’où l’analogie avec « The Pig and The Python ». Cependant, une politique d’ouverture sur l’immigration internationale a fait en sorte que la disproportion entre les effectifs des baby-boomers et celles des baby-busters s’atténue graduellement au cours du temps. L’objectif de cette communication est de montrer:

- L’écart important entre les effectifs dans leur jeunesse
- Un écart de moins en moins important aux âges adultes du à l’immigration nette
- Une quasi-disparition de l’écart dans les âges de la retraite du à cette immigration cumulative.

Les effets négatifs de « The Pig and the Python » seront donc éliminés au moment où les baby-boomers entreront dans les âges de retraite. Cependant comme l’accommodement sera réalisé via d’importantes projections d’immigrants internationaux, la composition ethnoculturelle des futures personnes âgées au Québec sera nettement plus hétérogène. Les régimes futurs de soutien à la retraite et à la vieillesse devront en tenir compte.