Estimating Adult Death Rates From Sibling Histories: A Network Approach

Intergenerational Clustering of Under-Five Mortality: A Cohort Perspective in Low- and Middle-Income Countries

A burgeoning demographic literature documents the exceedingly high rates at which contemporary cohorts of women across the Global South experience the death of their children-even amid historic declines in child mortality. Yet, the patterning of maternal bereavement remains underinvestigated, as does the extent to which it replicates across generations of the same family. To that end, we ask: Are the surviving daughters of bereaved mothers more likely to eventually experience maternal bereavement? How does the intergenerational clustering of maternal bereavement vary across countries and cohorts? To answer these questions, we make use of Demographic and Health Survey Program data from 50 low- and middle-income countries, encompassing data on 1.05 million women and their mothers spanning three decadal birth cohorts. Descriptive results demonstrate that maternal bereavement is increasingly patterned intergenerationally across cohorts, with most women experiencing the same fate as their mothers. Multivariable hazard models further show that, on average, women whose mothers were maternally bereaved have significantly increased odds of losing a child themselves. In most countries, the association is stable across cohorts; however, in select countries, the risk associated with having a bereaved mother is shrinking among more recent birth cohorts.

Assessment of the national and subnational completeness of death registration in Nepal

Background

Reliable and timely mortality data from a civil registration and vital statistics (CRVS) system are of crucial importance for generating evidence for policy and monitoring the progress towards national and global development goals. In Nepal, however, the death registration system is not used to produce mortality statistics, because it does not providing data on age at death and only reporting deaths by year of registration. This study assesses the completeness of death registration in Nepal – both the existing offline system and the newer online system – as well as the completeness of death reporting from a CRVS Survey, and assesses differences by year, sex, ecological belt, and province.

Methods

The empirical completeness method is used to estimate completeness at all ages from the offline (paper-based) registration system (2013-17), the online registration system (2017-19) and the CRVS Survey (2014-15).

Results

Completeness of the offline death registration system was 69% in 2017, not increasing since 2013 and being higher for males (73%) than females (65%). Completeness of online registration was only 32% in 2019, but almost double the 2017 figure. Completeness of death reporting in the CRVS Survey was 75% in 2015. The largest subnational differentials in completeness exist for the offline registration system, ranging from 90% in Gandaki to just 39% in Karnali.

Conclusions

Improvement in the utility of the Nepalese death registration system for mortality statistics is dependent on continued roll-out of the online death registration system (which reports age at death and deaths by year of occurrence) throughout the country, focusing on areas with low registration, building a strong coordination mechanism among CRVS stakeholders and implementing public awareness programs about death registration.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-022-12767-z.

Assessing mortality registration in Kerala: the MARANAM study

Complete or improving civil registration systems in sub-national areas in low- and middle-income countries provide several opportunities to better understand population health and its determinants. In this article, we provide an assessment of vital statistics in Kerala, India. Kerala is home to more than 33 million people and is a comparatively low-mortality context. We use individual-level vital registration data on more than 2.8 million deaths between 2006 and 2017 from the Kerala MARANAM (Mortality and Registration Assessment and Monitoring) Study. Comparing age-specific mortality rates from the Civil Registration System (CRS) to those from the Sample Registration System (SRS), we do not find evidence that the CRS underestimates mortality. Instead, CRS rates are smoother across ages and less variable across periods. In particular, the CRS records higher death rates than the SRS for ages, where mortality is usually low and for women. Using these data, we provide the first set of annual sex-specific life tables for any state in India. We find that life expectancy at birth was 77.9 years for women in 2017 and 71.4 years for men. Although Kerala is unique in many ways, our findings strengthen the case for more careful attention to mortality records within low- and middle-income countries, and for their better dissemination by government agencies.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1186/s41118-021-00149-z.