Women’s deaths are under-counted, may skew Covid impact estimates

4 months ago 28

The death of one in four women is never counted in India. But when it comes to men, the gap narrows to one in over six. The gender divide in death registration has been worsening — from a 6-percentage point difference in 2009 to 11 percentage points in 2018, says a new study by researchers from Unicef in Bihar and the University of Melbourne. And it could be distorting the visibility of the gendered impact of Covid.
For the study, published in the ‘International Journal for Equity in Health’ by BioMed Central on August 31, the researchers compiled death registration data from Civil Registration Reports between 2000 and 2018, and used statistical analysis — the empirical completeness method — to estimate how far deaths were registered and if inequalities between states and sexes have narrowed or widened.
They found that the estimated completeness of death registration went up from 58% in 2000 to 81% in 2018. Data categorised on the basis of sex was only available 2009 onwards which, the researchers said, was dependent on how well the states compiled data. Since then, completeness of male death records rose from 60% to 85% but that of female death records from 54% to only 74%.
“It is quite a significant difference and is important for two reasons. Firstly, it means that the official registration of female deaths is less common than for male deaths, despite registration of deaths being compulsory. This means the family is less likely to receive an official death certificate for a female death,” corresponding author Dr Timothy Adair told TOI. “Secondly, it means that a higher proportion of female deaths are unreported and so we know less about female mortality patterns compared to male mortality. Accurate measurement of mortality is important to monitor trends in population health, including measurement of excess mortality during the COVID-19 pandemic. If mortality of females during the pandemic is under-reported compared with male mortality, then the impact of the pandemic on female mortality would be under-estimated.”
The differences among states are stark. In Rajasthan, the latest estimates in the study show, male completeness was 87% while female was just 62% — a 25-percentage point difference, the highest in the country. Arunachal Pradesh came next (22-percentage point difference), followed by Madhya Pradesh (17), Assam (14), Mizoram (13) and Uttarakhand (12).
“Social and cultural factors can interact with characteristics of the national death registration system to create barriers to female death registration. A death needs to be registered so that a death certificate can be issued. A death certificate is required for property-related inheritance and social security/insurance issues, which in India are usually linked to males in the family,” said Adair. “Another issue is that a higher proportion of male than female deaths occur in hospitals, where deaths are more likely to be registered than if occurring at home. The attainment of more equitable levels of death registration between males and females requires the reduction in barriers to death registration.”
There were some outliers. In Sikkim, female completeness (95%) surpassed that of male records (87%), an 8-percentage point difference. Nagaland came next (4-point difference), then Jammu & Kashmir (2), Meghalaya (2) and Odisha (1).
Besides the gender divide, there were also wide differences between states in overall completeness of death records. That of Nagaland, for instance, was 19%, the lowest. Bihar followed, with just 26%, and then Manipur (36%), Jharkhand (42%) and Jammu & Kashmir (49%). At the top was Goa, with 99% completeness, followed by Delhi (99%), Kerala (97%), Tamil Nadu (94%) and Sikkim (91%).
It is a continuation of trends over the two-decade period of study, the paper found: “The level of increase in completeness was very low in some of the states, particularly from eastern (Bihar, Jharkhand) and north-eastern (Manipur, Nagaland) regions of the country. States from the southern region (Kerala, Karnataka, Tamil Nadu) had consistently higher completeness throughout the period.”
These differences, Adair said, were correlated with economic development and education levels. “Registration completeness is commonly higher in states with high socio--economic development and lower in states with lower socio--economic development. This relates to the supply side: the states with good governance, political will and infrastructure (including human resources) have a good death registration system. It also relates to the demand side: states with low literacy and relatively poor socio-demographic indicators have lower demand from the communities for death registration.”
That, too, would skew the understanding of Covid’s impact on states. The paper said, “During the COVID-19 pandemic, the lack of complete and timely death registration data in India have prevented timely measurement of excess mortality and potentially masks the true extent of its impact in some states more than others.”


Read Entire Article