New approaches to the collection and analysis of data on mortality and cause-of-death

The level of infant and child mortality is taken as indicator of health condition and also be regarded as general indicator of development of society. Decline in mortality has been quite rapid in India, but the rate of infant and child mortality is still very high. Thus, primary concern is to further reduce the level of infant as well as child mortality by exploring the high risk group. In this paper, death under age five year is termed as child death i.e. the sum of infant and child mortality. In this paper we have tried to simulate the distribution of risk of child death (p) for all parity, which can not be observed directly. Since p is affected by many factors, so it seems more logical to consider the p as a random variable and have a distribution. Beta-binomial probability distribution is considered as a model for number of child death to women for fixed parity to estimate the risk of child death. The suitability of model is checked by using National Family Health Survey (NFHS-III, 2005-06) for the two states of north India i.e. Uttar Pradesh and Bihar.

Urban air pollution is an important environmental factor that is effecting a large portion of urban dwellers in developing economies. Due to the limitations of dose and impact nature, direct measurement of negative implications of urban air pollution is difficult and WHO has developed indirect methods such as Environmental Burden of Disease. However, in this study to measure the economic costs of urban air pollution, we find that WHO methodology needs modifications to suit local conditions.

Objective To determine patterns and risk factors for cause-specific adult mortality in rural southern Tanzania. Method The study was a longitudinal open cohort and focused on adults aged 15–59 yrs from 2003 to 2007. Causes of deaths were ascertained by VA. Cox proportion hazards regression model was used to determine factors associated with cause-specific mortality over the 5-year. Results 1352 of 65548 adults died, representing a crude AMR of 7.3 per 1000PYO. HIV⁄AIDS [231(20.4%)] was the leading cause of death followed by malaria [150(13.2%)]. AMR for CD causes was 2.49 per 1000PYO, 1.21 per 1000PYO for NCD and 0.53 per 1000PYO for accidents⁄injury causes. NCD deaths increased from 16% in 2003 to 24% in 2007. High level of education was associated with a reduction in the risk of dying from NCDs. Those with primary education (HR=0.67, 95% CI:0.49, 0.92) and with education beyond primary school (HR=0.11, 95% CI:0.02, 0.40) had lower mortality than those who had no education. Compared with local residents, in-migrants were 1.7(95% CI:1.37, 2.11) times more likely to die from CD causes. Conclusion NCDs are increasing as a result of demographic and epidemiological transitions taking place in most African countries including Tanzania and require attention to prevent increased triple disease burden of CD, NCD and accident ⁄ injuries.

Mortality among adults should be addressed as a priority in Brazil because it has become relatively more important than mortality in other groups. Reliable mortality data are important in particular for monitoring homicide and road traffic accidents, which are the top causes of death among adult males in Brazil. However, deaths coded to ill-defined diseases and ill-defined injuries still represent a substantial proportion of all deaths. The aim of this proposal is to evaluate the mortality from ill-defined causes of death in 2000-2010 and the impact of the health services’ investigation into measures of adult mortality among men in 2010 in Brazil. The proportion of ill-defined causes of death among male adult deaths in Brazil was 10.5% in 2000 and in 2010 the proportion fell significantly to 6.6% after investigation and to 1.9% after correction. Ill-defined injuries fell from 9.0% in 2000 to 5.2% in 2010 after investigation and 0.6% after data correction. After adjustment, homicide and road traffic accidents were estimated to have caused 30% of all deaths in men. The findings indicate the importance of adjusting the data based on the improved strengthening of expertise in death investigation of ill-defined causes.

Polio cases worldwide have declined from 60,000 per year in 1980 to <200 today, following repeated mass vaccination campaigns in affected countries. In 2010, 400 Million children worldwide participated in a polio vaccination campaign. Beyond specific effects on polio incidence, such campaigns may reduce under-5 mortality because 1) they can stimulate the demand for other health services, 2) there are potential non-specific effects of the oral Polio vaccine (OPV) on mortality and 3) investments in polio eradication may strengthen health systems. On the other hand, mass vaccination campaigns may also divert scarce human/financial resources from routine services to OPV provision, or foster passive health-seeking behaviors. Despite the scale of the polio eradication initiative, its impact on mortality is unclear because program evaluations have ignored issues of reverse causality. Polio campaigns are indeed more likely to be conducted in areas with limited health seeking behaviors. We devise a new instrumental variable strategy to account for this endogeneity. We use DHS data collected in 8 African countries to construct a time-series dataset on mortality outcomes and campaign exposure spanning 1996-2011. Initial results from 2-stage probit models suggest that polio campaigns reduced the utilization of health services.

In terms of competing risks Mixture Long-term Survival Models are widely used for the analysis of individuals may never suffer the considered cause of failure. Under condition of a cured fraction, some individuals will be treated as immune to a specific cause of failure or be defined as long-term survivors. In case of multi- or bivariate cause-specific survival data different dependence structures between variables can be suited with different copula functions. There are two main methodical aspects for the marginal distributions need to account for: first the maximum of flexibility and second the application in case of masked causes. We proposed a bivariate mixture long-term model based on the Farlie-Gumbel-Morgenstern (FGM)copula. Data simulations will be provided with SEER Breast Cancer Data, and comparing the model with different types of copulas e.g. FGM, Positive Stable, Frank and Clayton Copula. Otherwise we will discuss optional ideas for this approach in a semi-competing risk setting.

Abstract
Reducing child mortality rates in Tanzania is a challenging issue and a public health emergency in the country. This study, based on Tanzanian Demographic and Health Survey (DHS) 2008 data, presents a method for estimating specific causes of child mortality. Tanzania’s child mortality rate was ranked 27th globally the rate was above the world average. Study findings shows wealth index was the major contributor in the odds of child mortality. The preceding birth interval approximately 76% spaced their births at 24 months or more. Hazard rates findings focus on specific causes of child mortality based on the mother’s fertility behaviors, such as birth intervals, age at childbirth and multiplicity of birth, rather than on other determinants of socioeconomic and health care services. Children belonging to the group of higher birth order faced the odds for dying of 0.17. Age at first birth became significant with the odds for survival increasing by 1.68 times, with the increase in age of the mother at first childbirth. There are two important variables, health care facilities and health care service providers, not showing any statistical effect on child mortality. This means health infrastructure and service providers need to work hard. Therefore, further in-depth study is strongly recommended.

The study proposes a new way to compute confidence intervals for life expectancy estimates, based on Monte Carlo simulations. Taking into account the stochastic variability is particularly relevant for geographical areas with small population size, as the limited number - if any - of death events and/or population at risk, especially at younger ages, may cause fluctuations in the estimates of life expectancies which make it difficult to identify trends. To test its robustness, the method is applied to about 300 complete life tables of European regions computed by Eurostat, representing a variety of mortality patterns and possible issues of data quality/availability. A comparison is also made with other two stochastic methods proposed in the literature (Chiang 1984 and Silcock et al. 2001), as well as sensitivity analyses to assess the impact of different age grouping in the life table.

Human life length tripled in historical times. To examine whether an end to increases in life length is imminent, we computed the median life length and the remaining life expectancy for 6,560 life tables from 1800 to 2008. Within almost all intervals of 20 years, for almost all countries, as median life length increased, the remaining life expectancy at the median life length decreased linearly. This empirical pattern in life tables has not been previously observed. The slope and intercept of this linear relation differed among countries in a given time interval and among time intervals for a given country. This linear relation suggested a hypothetical, easily calculated index of maximal median life length (MMLL) for a given country and time interval, namely, the median life length, according to this decreasing linear relation, at which the remaining life expectancy is zero. In successive 20-year time intervals starting from 1800-1819, the largest value (over all countries, within a given 20-year interval of observation) of this MMLL increased by more than one year of life per year of calendar time. There was no sign of a decreasing rate of increase, suggesting that a limit on median life length is not imminent.

In Mexico, given the high mortality ratio, maternal deaths are notified and analyzed carefully every year by the Ministry of Health (MoH) who has developed tools aiming to provide recommendations for the health systems.
We analyzed the national 2010 abortion related mortality files. We used the International Classification of Diseases (ICD-10) that includes all pregnancies with abortive outcome under codes O00- O06. Overall, abortion related mortality accounted for 7.2% of Maternal Mortality (MM) during 1990-2008. During 2010, 9.27% of maternal deaths were due to abortion. Almost 75% of these deaths were in women 15-34 years old, younger than other causes; 26% were single compared with 15% for other causes. 90% of women received health care before the death.
Quality of the information of abortion cases was poorer than other MM causes; regularly a MM case includes 11 elements for the comprehensive analysis into their file. In the case of abortion deaths, only 66% of the files had verbal autopsies, and only 23% necropsy reports.
Despite the interest in MM, first trimester deaths are not being adequately addressed in the current framework. Abortion related deaths enquiries demand a renovated effort from the health information systems, to provide more adequate data and help guide recommendations.