No description available.

No description available.

No description available.

In studying natural history of a disease, incident studies provide the best quality estimates; in contrast, prevalent studies introduce a sampling bias, which, if the onset time of the disease follows a stationary Poisson process, is called length bias. When both types of data are available, combining the samples under the assumption that failure times in incident and prevalent cohorts come from the same distribution function, could improve the estimation process from a revalent sample. We verify this assumption using a Smirnov type of test and construct a likelihood function from a combined sample to parametrically estimate the survival through maximum likelihood approach. Finally, we use Accelerated Failure Time models to compare the effect of covariates on survival in incident, prevalent, and combined populations. Properties of the proposed test and the combined estimator are assessed using simulations, and illustrated with data from the Canadian Study of Health and Aging.

No description available.

Abstract Background: Studies into risk factors and causes of childhood mortality present the opportunity to identify intervention programs appropriate in different populations in our attempt to reach the WHO Millennium Development Goals. Objectives: To determine whether there is an association between parental labour migration and child mortality in the Agincourt Health and Demographic Surveillance Site (HDSS) in 2002. Methods: Secondary data extracted from the longitudinal database from the Agincourt Health and Demographic Surveillance System were used to study the association between father’s migration and child mortality in 2002 using logistic regression and survival analysis. The analysis included 10050 children born between 01 January 1998 and 31 December 2002. Results: The child mortality in 2002 was 12.9 deaths per 1000 person years. There was a statistically significant difference in death rate in infants (50.9 deaths per 1000 person years), and in children aged 1-4 years (9.6 deaths per 1000 person years). There was no association observed between migration and child mortality (OR: 0.97, 95% CI 0.59-1.60). The factors associated with mortality were the age of the child, the number of siblings that a child had, the refugee status of the mother, age of the mother at birth, breast feeding and whether the mother was deceased or not. The chances of dying were lower in older children compared to younger ones (OR: 0.58, 95% CI 0.50-0.68). Children who had one or more siblings were less likely to die (OR: 0.62, 95% CI 0.51-0.93) compared to those with none. Child mortality risk was higher in children born to refugees than to local residents (OR: 1.56, 95% CI 1.05-2.33). Those children who were not breast fed had increased chances of dying than those breast fed (OR: 5.33, 95% CI 2.60-10.95). The death of a mother increased the risk of the child dying (OR: 9.35, 95% CI 5.02-17.40). About 84.3% of migrants were sending remittances to members of the households remaining behind but there was no significant difference in child mortality among remitters and nonremitters. The leading causes of death among households with migrant father and those with a non-migrant father were infectious diseases (mainly HIV/AIDS related illness) with 47% and 50% respectively. Conclusion: The results from this study suggest that on average children of migrants in a rural area in South Africa do not experience increased mortality compared to children of non-migrants. The findings from this study where no association between fathers’ migration and child mortality was observed were rather inconclusive. So far, this area of research has not been adequately addressed and much remains to be learned about the lives of children left behind by fathers migrating to seek employment elsewhere. For future studies it would be advisable to study in great depth the long term effects of migration on child mortality particularly in Africa.

Egypt is a geographically large, populous and internationally-recognised state with an organised bureaucracy and armed forces. In contrast, Hamas is an armed social movement, which, after its electoral victory in January 2006 and, more importantly, after acquiring full control over the Gaza Strip in June 2007, emerged as a quasi-state with internal sovereignty. Egypt enjoys a peace treaty with Israel and a strategic alliance with the US, whereas Hamas is in conflict with Israel, and is designated by the US as a terrorist group. This thesis traces the interactions between Hamas and Egypt during the 2006-2014 period, with a focus on the Mubarak era. The dissertation’s main aim is not only to present how and when asymmetry and strategic differences between Hamas and Egypt were reflected in their relations, but also to explain why and how on certain occasions their interactions took on the form of tactical cooperation. Hence, I show that small or quasi-states in the contemporary Middle East are in position to extract political gains from larger neighbouring state actors even in the presence of strategic differences. This thesis contextualises the situations it discusses through Omni Balancing Theory (OBT), which understands an actor’s foreign policy as the outcome of the efforts of its leader to survive politically by balancing between external and internal threats. In this regard, Egypt’s approach towards Hamas and vice versa at a given time is seen as the result of a cost-benefit calculation that has assessed the value of simultaneous foreign and domestic threats. Accordingly, the dissertation looks at Hamas-Egypt relations through three lenses: firstly, through the impact of international and regional pressures; secondly, through cross-border interactions; and thirdly, through the effect of domestic pressures. Finally, the thesis separately discusses the course of Hamas-Egypt relations between February 2011 and August 2014. This is due to the density of the political developments during this period. To be precise, the three weak post-Mubarak Egyptian governments faced quite diverse threats the dealing of which generated considerable fluctuations in Cairo’s approach towards Hamas.

Thesis (Ph. D.)--University of Hong Kong, 2005. / Title proper from title frame. Also available in printed format.

The scientific debate on longevity and its determinants is lively and involves researchers from different disciplines. Finding a clear cut explanation for why some people live longer than others is not an easy task as longevity proves to be connected to more than one factor. Moreover, variables that are shown to be significant for longer survival in certain populations are not always relevant to individuals of other populations. Recently, researchers have identified in Sardinia the Blue Zone (BZ), thus denominated due to the significant number of cases of centenarians recorded as well as a low sex ratio value. This points to an exceptional favourable longevity of males when compared with the expected values in developed countries. A village located in this area was selected for the carrying out of a more detailed analysis at individual level and socio-demographic determinants of longevity were examined. A family reconstructed database was created using a multi-source approach, as this data collection strategy allowed for careful cross-checking of the data, thus guaranteeing data of high quality as well as an exceptionally high level of coverage of known survival. Historical data was therefore used to study today’s population, establishing a unique bridge between historical demography and the current level of longevity. The study focused on survival of newborns in the selected village during a period of 50 years – from 1866 to 1915 – and also took into consideration information on related family members. Each individual included in the database was followed from birth to death, as he/she went through marriage and family formation. Complete survival of all newborns was observed until the present. Based on careful review of the existing literature, the empirical model considered factors such as sex differential, the contribution of familial transmission both in terms of genes, biological, and shared environment, and the role played by environment operationalised considering early-life conditions. Concepts were organized into a life-course approach for survival analysis. The analysis of the data enabled to the estimation of mortality trajectories, which in turn confirmed the exceptional longevity of the population understudy and in particular of males. The investigation of differential survival among members of the same community brought to light the relevance of certain familial variables on survival, which are not exclusively genetic. The results achieved open the way to further studies.

In studying natural history of a disease, incident studies provide the best quality estimates; in contrast, prevalent studies introduce a sampling bias, which, if the onset time of the disease follows a stationary Poisson process, is called length bias. When both types of data are available, combining the samples under the assumption that failure times in incident and prevalent cohorts come from the same distribution function, could improve the estimation process from a revalent sample. We verify this assumption using a Smirnov type of test and construct a likelihood function from a combined sample to parametrically estimate the survival through maximum likelihood approach. Finally, we use Accelerated Failure Time models to compare the effect of covariates on survival in incident, prevalent, and combined populations. Properties of the proposed test and the combined estimator are assessed using simulations, and illustrated with data from the Canadian Study of Health and Aging.