Page 162 - Contributed Paper Session (CPS) - Volume 3

CPS1982 Dmitri J. et al.
All data series in the HMD are updated on a rolling basis. One of the main
principles of the HMD is to include countries with reliable population statistics,
especially requiring a full coverage of registration of vital events. The countries
and areas included thus are relatively wealthy and for the most part highly
industrialized. The Human Mortality Database contains original calculations of
death rates and life tables for national populations (countries or areas), as well
as the original input data used for constructing those tables and an extensive
documentation. More details about the HMD project can be found in (Barbieri
et al., 2015). The HMD has more than 45,000 registered users and has been
cited in more than 1500 scientific publications. The average number of
citations during the last two years is 200 per year.
One of the main advantages of the HMD is quality of provided data and
their comparability across time and space. The constructed and updated series
are carefully checked and reviewed for internal and external consistency
before publication. Every data series in the HMD have to meet strong
requirements. One of the most important criteria is vital registration system
which cover more than 90 percent of population.
Principles, recommendations and guidelines for evaluation of quality of
vital registration systems and vital data were discussed many times (Setel et
al., 2007; UN, 2014; WHO, 2010). The HMD relies on external researches by
preliminary selection of acceptable countries. Unfortunately, for most of the
world population, complete and accurate data on mortality are not available.
To produce such data an expensive and well-organized system for registration
of vital events and also censuses or population registers to count population
is needed. This is something that majority of developing nations have been
unable to achieve. Death registration does not exist or is very fragmentary in
most of the developing world including its most populated parts (China, India,
Indonesia) and also in countries that are facing the greatest health challenges
(Sub-Saharan Africa).
Vital registration that can be used to calculate life tables over the whole
range of ages exist in about 60 countries. But existing of nearly complete vital
registration system does not guarantee that quality of population statistics is
sufficient. In about 15 to 20 of these countries, quality of these data is a serious
concern. For other 40-45 countries, data quality can still be problematic during
some time periods or at some ages. Below we summarize most important
aspects of the work on evaluation of data quality within the HMD project. We
do not touch problems related to some of historical populations’ experience,
quality of population censuses or population statistics in developing countries.
We focus on the newly emerging data quality problems. These problems are
largely related to growing uncertainty about the population denominator due
to unregistered migration.

151 | I S I W S C 2 0 1 9