Student Number : 0516841M - MSc research report - School of Public Health - Faculty of Health Sciences / Objective This work investigates the hypothesis that individuals recently migrating into rural areas have a higher mortality than those always resident and that migrant deaths are more likely to be HIV/AIDS related than non migrant deaths. Methods Data from the Africa Centre Demographic Surveillance System (ACDIS), South Africa, was used for the analysis. A total of 41519 adults aged 18 to 60 years since their last visit dates were categorized into three groups; internal migrants, external in migrants and always resident individuals since 2001. Follow up period was from 1st January 2001 to 31st December 2005. Cox proportional hazard regression method was used to quantify the additional risk of dying for migrants who have recently migrated into the DSS area. Logistic regression was used to examine the relationship between migration status and dying from AIDS related complications for the members in the sample whose cause of death have been identified using verbal autopsy procedures. Results External in migrants into the DSS area were 1.52 times more likely to die than those always resident. After adjusting for the effects of sex, age group, socio-economic status and educational level an external in migrant has a relative risk of 1.19, [adjusted HR=1.19, P=0.001, 95% CI (1.08,1.32)] of dying compared to those always resident. Internal migrants were 18% less likely to die compared to always resident individuals, [adjusted HR=0.82, P=0.008, 95% CI (0.71, 0.95)] and males were 1.38 times more likely to die within the follow up period compared to females, [HR=1.38, P<0.001, 95% CI (1.28, 1.49)]. These results were statistically significant at 95% confidence level. Out of a total of 1119 deaths that occurred in 2001 and 2002 whose cause of death have been identified through verbal autopsy procedures, 763 (66%) died of AIDS. The odds of dying from AIDS are 2.09 if you are an external in migrant compared to an always resident member, [unadjusted OR = 2.09, P = 0.009 95% CI (1.38, 3.16)]. After controlling for other factors in the model, the odds of dying from AIDS as an external immigrant was 1.79 times, [adjusted OR = 1.79, P = 0.009, 95% CI (1.15, 2.77)] compared to those always resident. There was no significant difference in AIDS mortality between always resident individuals and internal migrants. The odds of a female dying of AIDS was 2.33 times, [OR = 2.33, P<0.001, 95% CI (1.78, 3.06)] compared to males after controlling for migration status, age, socioeconomic status and educational level. Conclusion External in migrants have an increased risk of death among adults aged 18 to 60 years compared to those always resident. External in migrants are also more at risk of dying from AIDS related illnesses than those always resident. Internal migrants are less likely to die than those always resident. Females are more at risk of dying from AIDS than males. In resource-poor settings, especially in many parts of Africa and other developing countries with very high prevalence of HIV/AIDS and over burdened health services in rural areas, it is important to identify and quantify some of these trends contributing to high disease burdens and mortality in rural areas in order to put in place effective interventions to better the health conditions of the people in these areas.

Background Mortality data are critical to understanding and monitoring changes in population health status over time. Nevertheless, the majority of people living in the world’s poorest countries, where the burden of disease is highest, remain outside any kind of systematic health surveillance. This lack of routine registration of vital events, such as births and deaths, constitutes a major and longstanding constraint on the understanding of patterns of health and disease and the effectiveness of interventions. Localised sentinel demographic and health surveillance strategies are a useful surrogate for more widespread surveillance in such settings, but rigorous, evidence-based methodologies for sample-based surveillance are weak and by no means standardised. This thesis aims to describe, evaluate and refine methodological approaches to mortality measurement in resource-poor settings. Methods Through close collaboration with existing community surveillance operations in a range of settings, this work uses existing data from demographic surveillance sites and community-based surveys using various innovative approaches in order to evaluate and refine methodological approaches to mortality measurement and cause-of-death determination. In doing so, this work explores the application of innovative techniques and procedures for mortality surveillance in relation to the differing needs of those who use mortality data, ranging from global health organisations to local health planners. Results Empirical modelling of sampling procedures in community-based surveys in rural Africa and of random errors in longitudinal data collection sheds light on the effects of various data-capture and quality-control procedures and demonstrates the representativeness and robustness of population surveillance datasets. The development, application and refinement of a probabilistic approach to determining causes of death at the population level in developing countries has shown promise in overcoming the longstanding limitations and issues of standardisation of existing methods. Further adaptation and application of this approach to measure maternal deaths has also been successful. Application of international guidelines on humanitarian crisis detection to mortality surveillance in Ethiopia demonstrates that simple procedures can and, from an ethical perspective, should be applied to sentinel surveillance methods for the prospective detection of important mortality changes in vulnerable populations. Conclusion Mortality surveillance in sentinel surveillance systems in resource-poor settings is a valuable and worthwhile task. This work contributes to the understanding of the effects of different methods of surveillance and demonstrates that, ultimately, the choice of methods for collecting data, assuring data quality and determining causes of death depends on the specific needs and requirements of end users. Surveillance systems have the potential to contribute substantially to developing health care systems in resource-poor countries and should not only be considered as research-oriented enterprises.

This thesis examines the impacts of adult HIV/AIDS related mortality on elderly women and their households in Agincourt, a rural area in the north-eastern part of South Africa. It focuses specifically on demographic, socio-economic and socio-cultural impacts of adult AIDS and non-AIDS illness/death on near-old women aged 50-59 and older women aged 60 and above. The study uses the Agincourt Health and Demographic Surveillance System (AHDSS) 2004 census data which contains some history about individuals and their households (e.g. household mortality experience between 1992 and 2004). The AHDSS dataset is used to examine elderly female household headship and its relationship with, firstly, pension status and secondly, adult AIDS/non-AIDS mortality, through statistical analyses. Also, the AHDSS census data is utilised as a sampling frame to select a random sample of 60 households in which 30 women aged 50-59 and 30 women aged 60-75 lived, for the qualitative part of this study. The sample was made up of 20 women who lived in households that had experienced an HIV/AIDS death between 2001 and 2003, 20 women that lived in households where another type of adult death had occurred, and 30 women in households with no adult death during the period. The findings from the quantitative analyses of the AHDSS dataset reveal that there is no significant relationship between adult AIDS/non-AIDS death and elderly female household headship. Further findings, however, show that elderly female household headship is strongly associated with pension status, thereby suggesting that pension grant is a determining factor in the household headship status of elderly women. One implication of this is that elderly female household heads who are pensioners may be able to cope better with HIV/AIDS impacts as findings from the qualitative data demonstrate that pensioners (older women) are more likely, than non-pensioners (near-old women), to have access to coping strategies, which enable them to deal with the financial crises of adult illness/death in their households. There is, therefore, the need for programmes to target near-old women, who experience the financial impact of adult morbidity/mortality like their older peers. iii Further findings from the qualitative data explicate secondary stigma as a socio-cultural impact of adult HIV/AIDS on elderly women who are caregivers to infected children. Findings also highlight different types of secondary stigma such as physical stigma in the form of separation from family members and social stigma in the form of social isolation. The study suggests that there is need for intervention programmes that address the issue of secondary stigma as it makes caregiving responsibilities more burdensome for elderly women.

Understanding mortality patterns is an essential pre-requisite for guiding public health action and for supporting development of evidence-based policy. However, such information is not sufficiently available in Vietnam. Mortality statistics and causes of death are solely collected from health facilities while most deaths occur at home without the presence of health professionals. Facility-based data cannot represent what happened in the wider community. This thesis studies the patterns and burdens of mortality as well as their relationships with socio-economic status in rural Vietnam. The overall aim is to contribute to the improvement of the current system of mortality data collection in the country for the purposes of public health planning and priority setting. The study was carried out within the framework of an ongoing Demographic Surveillance System (DSS) in Bavi district, Hatay province, northern rural Vietnam. This study used a verbal autopsy (VA) approach to identify cause of death in a cohort of approximately 250,000 person- years over a five-year period from 1999 to 2003. During the five year study, a total of 1,240 deaths were recorded and VA was successfully completed for 1,220 cases. Results revealed that VA was an appropriate and useful method for ascertaining cause of death in this rural Vietnamese community where specific data were otherwise scarce. The mortality pattern reflected a transitional pattern of disease in which the leading cause of death was cardiovascular diseases (CVD), followed by neoplasms, infectious and parasitic diseases, and external causes, accounting for 28.9%, 14.5%, 11.2%, and 9.8%, respectively. In terms of premature mortality, there were 85 and 55 Years of Life Lost (YLL) per 1,000 population for males and females respectively. The largest contributions to YLL were CVDs, malignant neoplasms, unintentional injuries, and perinatal and neonatal causes. In general, men had higher mortality rates than women for all mortality categories. In adults of 20 years and above, mortality rates increased substantially with age, and showed similar age effects for all mortality categories with the strongest association for non-communicable diseases (NCD). Education was an important factor for survival in general, and high economic status seemed to benefit men more than women. Compared with cancer and other NCD causes, higher CVD rates were observed among males, the elderly, and those without formal education, using a Cox proportional hazards model. This study is an initial effort to provide information on mortality patterns in a community using longitudinal follow-up of a dynamic cohort. Continuing the study using the VA approach as part of routine data collection in the setting will help to show trends in mortality patterns for the community over time, which may be useful for priority setting and health planning purposes, not only locally but also at the national level. Further analyses are needed to understand mortality inequality across all ages to have a comprehensive picture of mortality burdens in the setting. Validation studies and further standardization of VA methods should be carried out whenever possible to improve the performance and extension of the technique.

Abstract Background: Maternal mortality is one of the most sensitive indicators of the health disparities between poorer and richer nations. It is also one of the most difficult health outcomes to measure reliably. In many settings, major challenges remain in terms of both measuring and reducing maternal mortality effectively. This thesis aims to quantify overall levels, identify specific causes, and evaluate local interventions in relation to efforts to reduce maternal mortality in Tigray Region, Northern Ethiopia, thereby providing a strong empirical basis for decision making by the Tigray Regional Health Bureau using methods that can be scaled at national level.   Methods: This study employed a combination of community-based study designs to investigate the level and determinants of maternal mortality in six randomly selected rural districts of Tigray Region. A census of all households in the six districts was conducted to identify all live births and all deaths to women of reproductive age occurring between May 2012 and September 2013. Pregnancy-related deaths were screened through verbal autopsy with the data processed using the InterVA-4 model, which was used to estimate Maternal Mortality Ratio. To identify independent determinants of maternal mortality, a case-control study using multiple logistic regression analysis was done, taking all pregnancy-related deaths as cases and a random sample of geographical and age matched mothers as controls. Uptake of ambulance services in the six districts was determined retrospectively from ambulance logbooks, and the trends in pregnancy-related death were analyzed against ambulance utilization, distance from nearest health center, and mobile network coverage at local area level. Lastly, implementation of the Family Folder paper health register, and its potential for accurately capturing demographic and health events, were evaluated using a capture-recapture assessment.   Results: A total of 181 deaths to women of reproductive age and 19,179 live births were documented from May 2012 to April2013. Of the deaths, 51 were pregnancy-related. The maternal mortality ratio for Tigray region was calculated at 266 deaths per 100,000 live births (95% CI 198-350), which is consistently lower than previous “top down” MMR estimates. District–level MMRs showed strong inverse correlation with population density (r2 = 0.86). Direct obstetric causes accounted for 61% of all pregnancy–related deaths, with hemorrhage accounting for 34%. Non-membership in the voluntary Women’s Development Army (AOR 2.07, 95% CI 1.04-4.11), low husband or partner involvement during pregnancy (AOR 2.19, 95% CI 1.14-4.18), pre-existing history of other illness (AOR 5.58, 95% CI 2.17-14.30), and never having used contraceptives (AOR 2.58, 95% CI 1.37-4.85) were associated with increased risk of maternal death in a multivariable regression model. In addition, utilization of free ambulance transportation service was strongly associated with reduced MMR at district level. Districts with above-average ambulance utilization had an MMR of 149 per 100,000 LB (95% CI: 77-260) compared with 350 per 100,000 (95% CI: 249-479) in districts with below average utilization. The Family Folder implementation assessment revealed some inconsistencies in the way Health Extension Workers utilize the Family Folders to record demographic and health events.   Conclusion: This work contributes to understanding the status of and factors affecting maternal mortality in Tigray Region. It introduces a locally feasible approach to MMR estimation and gives important insights in to the effectiveness of various interventions that have been targeted at reducing maternal mortality in recent years.

Rationale: Vital registration is lacking in developing settings where health and development problems are most pressing. Policy-makers confront an “information paradox”: the critical need for information on which to base priorities and monitor progress, and the profound shortage of such information. Aims: To better understand the dynamics of mortality transition in rural South Africa over a decade of profound socio-political change coupled with emerging HIV/AIDS. Thereby to inform health and development programming, policy formulation, and the research agenda; and contribute to debate on the nature of the ‘health transition’. Methods: The Agincourt health and demographic surveillance system is based on continuous monitoring of the Agincourt sub-district population in rural north-east South Africa. This involves annual recording of all vital events, specifically deaths, births and migrations in 11,700 households comprising some 70,000 persons. A “verbal autopsy” is conducted on every death, and special modules provide additional data. Key findings: A major health transition has occurred over the past decade, with marked changes in population structure and rapidly escalating mortality particularly among children and younger adults. A quadruple burden of disease is evident with persisting infectious disease and malnutrition in children, emerging non-communicable disease in the middle-aged and older, high levels of violence in an apparently peaceful community, and rapidly escalating HIV/AIDS and tuberculosis. There is evidence of sex differences and socio-economic differentials in mortality; vulnerable sub-groups include the children of Mozambican immigrants and recently returned labour migrants. Implications: With respect to health transition, empirical data demonstrate a marked “counter transition” with mortality increasing in children and young adults; “epidemiologic polarization” is evident with the most vulnerable experiencing a higher mortality burden; and a “protracted transition” is reflected in the co-existence of persisting infectious disease and malnutrition, emerging HIV/AIDS, and increasing chronic non-communicable disease. With respect to health policy and practice there is urgent need to: strengthen HIV/AIDS prevention, treatment and care; offer effective long-term care to control the rising burden of chronic illness and related risk; maintain and improve maternal and child health services; and address differential access to care. This poses a substantial challenge to a severely stretched health system.

Introduction: The burden of road traffic injuries (RTI) is worsening globally, particularly in low- and middle-income countries (LMIC) and among the young and economically productive populations. A major barrier to improving road safety in India and other LMIC is that existing RTI data sources are severely limited by poor population coverage and data quality. This dissertation explores the reliability and feasibility of using a novel data source with verbal autopsy (VA) methods for the purposes of RTI surveillance in India. Methods: The reliability of the VA methods was assessed using physician agreement on the specific categories of injury death as the metric. Next, a nationally representative household mortality survey with VA methods was used to directly estimate the age- and gender-specific RTI death rates and to identify context-specific RTI risk factors in India. Finally, a national spatial database was constructed to quantify potential access to trauma care in relation to the spatial distribution of RTI deaths in India. Results: Across a broad array of application settings in India, the level of physician agreement was high indicating that the VA methods were reliable in distinguishing RTI deaths among other specific categories of injury deaths. The estimated 183,600 RTI deaths in 2005 from the mortality survey were over 50% more than the national police statistics. Of these RTI deaths, 65% were males between ages 15-59 years, 68% were pedestrians and other vulnerable road users, and over 55% occurred at the scene of collision, within minutes of collision, and/or involved a head injury. The existing community health centres and district hospitals in the Indian public health system had inadequate trauma care capacity but were suitably located to allow broad spatial access to timely trauma care for the majority of RTI deaths in India, which were most problematic in the northern states of Punjab, Haryana, Himachal Pradesh and in Tamil Nadu. Conclusions: Properly designed VA studies can provide accurate and reliable RTI surveillance data and assist in identifying context-specific road safety interventions.

Introduction: The burden of road traffic injuries (RTI) is worsening globally, particularly in low- and middle-income countries (LMIC) and among the young and economically productive populations. A major barrier to improving road safety in India and other LMIC is that existing RTI data sources are severely limited by poor population coverage and data quality. This dissertation explores the reliability and feasibility of using a novel data source with verbal autopsy (VA) methods for the purposes of RTI surveillance in India. Methods: The reliability of the VA methods was assessed using physician agreement on the specific categories of injury death as the metric. Next, a nationally representative household mortality survey with VA methods was used to directly estimate the age- and gender-specific RTI death rates and to identify context-specific RTI risk factors in India. Finally, a national spatial database was constructed to quantify potential access to trauma care in relation to the spatial distribution of RTI deaths in India. Results: Across a broad array of application settings in India, the level of physician agreement was high indicating that the VA methods were reliable in distinguishing RTI deaths among other specific categories of injury deaths. The estimated 183,600 RTI deaths in 2005 from the mortality survey were over 50% more than the national police statistics. Of these RTI deaths, 65% were males between ages 15-59 years, 68% were pedestrians and other vulnerable road users, and over 55% occurred at the scene of collision, within minutes of collision, and/or involved a head injury. The existing community health centres and district hospitals in the Indian public health system had inadequate trauma care capacity but were suitably located to allow broad spatial access to timely trauma care for the majority of RTI deaths in India, which were most problematic in the northern states of Punjab, Haryana, Himachal Pradesh and in Tamil Nadu. Conclusions: Properly designed VA studies can provide accurate and reliable RTI surveillance data and assist in identifying context-specific road safety interventions.

Bangladesh has made encouraging progress in reducing maternal and neonatal mortality over the past two decades. However, deaths are much higher than in many other countries. The death reporting system to address maternal, neonatal deaths and stillbirths is still poor. Moreover, cause identification for each of the community and facility deaths is not functional. The overall objective of this thesis is to develop, implement and evaluate the Maternal and Neonatal Death Review (MNDR) system in Bangladesh. The study has been conducted in two districts of Bangladesh. A mixed method is used in studies I and II, whereas a qualitative method is used in studies III-V, and cost of MNDR is calculated in study VI. In-depth interviews, focus group discussions, group discussions, participant observations and document reviews are used as data collection techniques. Quantitative data are collected from the MNDR database. In study I, community death notification in the MNDR system was found to be achievable and acceptable at district level in the existing government health system. A simple death notification process is used to capture community-level maternal and neonatal deaths and stillbirths. It was useful for local-level planning by health managers. In study II, death-notification findings explored dense pocket areas in the district. The health system took local initiatives based on the findings. This resulted in visible and tangible changes in care-seeking and client satisfaction. Death numbers in 2012 were reduced in comparison with 2010 in the specific area. In study III, verbal autopsies at community level enabled the identification of medical and social causes of death, including community delays. Deceased family members cordially provided information on deaths to field-level government health workers. The health managers used the findings for a remedial action plan, which was implemented as per causal findings. In study IV, social autopsy highlights social errors in the community, and promotes discussion based on a maternal or neonatal death, or stillbirth. This was aneffective means to  deliver some important messages and to sensitize the community. Importantly, the community itself plans and decides on what should be done in future to avert such deaths. In study V, facility death review of maternal and neonatal deaths was found to be possible and useful in upazila and district facilities. It not only identified medical causes of death, but also explored gaps and challenges in facilities that can be resolved. The findings of facility death reviews were helpful to local health mangers and planners in order to develop appropriate action plans and improve quality of care at facility level. Finally, in study VI, the initial piloting costs required for MNDR implementation were estimated, including large capacity development and other developmental costs. However, in the following year, costs were reduced. Unit cost per activity was 3070 BDT in 2010, but, in the following years, 1887 BDT and 2207 BDT, in 2011 and 2012 respectively.

Ethiopia is one of the five nations that bear the global burden of nearly 50% maternal mortalities and newborn deaths. Cause-specific maternal mortality and newborn death information are vitally important for prevention, but little is known about the causes of deaths. Many maternal mortalities and newborn deaths occur at home, outside the formal health sector, and few are attended by qualified medical professionals. Despite the fact that, non-medical factors are often more important in determining whether a woman/newborn lives or dies than the medical cause of death itself. This study determines and explores factors contributing to maternal mortalities and newborn deaths in Ethiopia with the aim of developing a community-based model for averting maternal mortalities and newborn deaths in Ethiopia. The study was organised in three phases. In Phase 1, a community-based-retrospective approach using explorative, descriptive and contextual study design, combining both qualitative and quantitative methods (mixed methods) were used to make an in-depth investigation and analysis of the circumstances and events surrounding individual cases of maternal mortality and newborn deaths. The result of the study revealed various direct and indirect as well as possible contributing factors to maternal mortalities and newborn deaths which outlined bases for forwarding Phase 2 of the study called concept analysis. In Phase 3, a prototype model was developed according to Chinn and Kramer’s approach to theory generation: initially, based on the empirical perspectives of the study, concept analysis was conducted. The structure and process of a model to avert maternal mortality and newborn death were described; and, six survey list; namely, agent, recipient, context, procedure, dynamic and terminus of Dickoff, James and Wiedenbach (1968) form the basis for development and description of a model for averting maternal mortality and newborn deaths in Ethiopia. Impediment in receiving prompt, adequate and appropriate care were common problems encountered even after reaching an appropriate medical facility. For any attempt to attain a significant reduction in maternal mortality and newborn death, the health care system in Ethiopia must assume its tasks to institute critical changes in both the structure and process of health care delivery services. / Health Studies / D. Litt. et Phil. (Health Studies)