Identifying patterns of main causes of death in the young EU population Simona Korenjak- Cerne1and Nata sa Kej zar2

2025-05-08 0 0 352.36KB 13 页 10玖币
侵权投诉
Identifying patterns of main causes of death
in the young EU population
Simona Korenjak-ˇ
Cerne1and Nataˇsa Kejˇzar2
1University of Ljubljana, School of Economics and Business,
and Institute of Mathematics, Physics and Mechanics, Slovenia
simona.cerne@ef.uni-lj.si,
2University of Ljubljana, Faculty of Medicine,
Institute for Biostatistics and Medical Informatics, Slovenia
natasa.kejzar@mf.uni-lj.si
Abstract. The study of mortality patterns is a popular research topic in
many areas. We are particularly interested in mortality patterns among
main causes of death associated with age-gender combinations. We use
symbolic data analysis (SDA) and include three dimensions: age, gen-
der, and patterns across main causes of death. In this study, we present
an alternative method to identify clusters of EU countries with similar
mortality patterns in the young population, while considering compre-
hensive information on the distribution of deaths among the main causes
of death by different age-gender groups. We explore possible relationships
between mortality patterns in the identified clusters and some other so-
ciodemographic indicators. We use EU data of crude mortality rates from
2016, as the most recent complete data available.
Keywords: mortality pattern, the main cause of death, young popula-
tion, symbolic data analysis, adapted clustering methods
1 Introduction
Mortality data on causes of death provide important information about the pub-
lic health in the observed geographic area. Consequently, the study of mortality
patterns is a very popular research topic in many areas (e.g., demography, health
economics, public policy, and actuarial science), and therefore several approaches
to the analysis of these data have been developed depending on specific needs.
For an overview of these approaches with many references, see for example the
paper of Van Raalte [20]. As highlighted in this paper, most life course studies
are still conducted at the individual level, while the main challenge is to link life
course studies more closely to mortality patterns at the aggregate level [20] (p.
S107). In this paper, we present a possible approach to the analysis of mortality
data at the aggregate level.
Our exploratory study focuses on the leading causes of death in the young
population. Our study was motivated by the Centres for Disease Control and
Prevention (CDC) publication summarising 10 leading causes of death by age
arXiv:2210.04469v2 [stat.AP] 11 Oct 2022
2 Korenjak-ˇ
Cerne et al.
group in the United States. The three leading causes of death among the young
population in the United States are unintentional injuries, suicide, and homicide.
In the fourth place are malignant neoplasms. Since the most common causes of
death in the young population aged 20-39 are external (i.e., accidents, especially
traffic accidents, suicides, and assaults), which are preventable, we focus on
mortality patterns in the young population in EU countries.
It is well-known that the number of deaths is strongly related to age, gender,
and cause of death. All three factors can be captured by selecting all possible
combinations of these factors as variables and then applying classical multivari-
ate data analysis techniques. However, in this case internal relationships among
these variables are usually not taken into account. These relationships may be
described with an empirical distribution of deaths among different causes of
death. To include them simultaneously in the analysis, more advanced analysis
methods are needed.
The data that we deal with are aggregated deaths, and we analyze them using
symbolic data analysis (SDA) [4–6]. SDA methods consider internal variability;
in our case, variability among deaths by cause of death. Following this path,
we use symbolic data descriptions [17] and consider data as symbolic objects,
where the units of interest are EU countries and the symbolic variables are five-
year age-gender groups described with distributions of the (expected ) number of
deaths across selected causes of death. We name these distributions mortality
patterns across causes of death. To find groups of EU countries with similar
mortality patterns, we adapt the hierarchical and clustering method presented
in [16] and implemented in the R package clamix, which we modify to include
the requirements of our specific problem.
In addition to preserving the relationship structure, another question that
we want to tackle is how to sufficiently transform mortality rates to make them
directly comparable between countries.
The objectives of this study are twofold: (1) to present an alternative method
to identify clusters of EU countries with similar mortality patterns in the young
population that takes into account more comprehensive information on the three
dimensions of the data: age, gender, and distribution across causes of death; and
(2) to explore possible relationships between mortality patterns in the identi-
fied clusters and some other sociodemographic indicators, which can serve as a
starting point for further detailed quantitative and qualitative investigations of
possible associations. We use EU data on crude mortality rates from 2016, which
are the most recent complete data available.
The rest of the paper is organized as follows. In the second section, we present
an additional rationale for our data selection and its preparation. In the third
section, we explain the adaptations of the clustering method to our symbolic
descriptions of the data. In the fourth section, we present and comment on some
of the results. The final section will give the concluding remarks of this study.
Mortality cause patterns in young people in the EU 3
2 Data
The young population is generally considered to be very healthy, and therefore
the leading causes of death in this population are often associated with risky
behaviors. The leading causes of death in the general population in EU countries
in 2016 and 2017 were circulatory diseases and various cancers; followed by
respiratory diseases; while external causes of death, including accidents, suicides,
homicides, and other violent causes of death, ranked fourth [9, 18]. The leading
causes of death in the young population are very different [10], with external
causes being by far the most common causes of death (the percentage varies by
age group but usually comprises more than half of the deaths).
The boundaries of the age groups that comprise the young population are not
clearly defined. In this work, we focus on the ages from 20 to 39 years because this
is the time when most people face major changes in their lives (usually starting
a new job, becoming independent from parents, finding a new place to live,
planning own family, raising children, etc.). This selection also coincides with
the psychological developmental stage of young adulthood that was introduced
by Erikson [8]. An additional reason for this particular data selection is that
the generation of ages from 20 to 39 years is often overlooked in specific health
studies because their causes of death are associated with risky behaviors that are
often considered to be personal choices. Meanwhile, risky behaviors are closely
related to mental status, with stress being one of the most important factors.
Most of these causes of death are preventable, and consequently we should not
overlook these important issues in our society. It is therefore important to develop
appropriate prevention programs for this target group because they face very
specific challenges in life and represent about a quarter of the total population
of the EU.
2.1 Symbolic representation of mortality data
For easier understanding and interpretation, mortality data are usually presented
with crude death rate (CDR), which is calculated as the ratio between the (to-
tal) number of deaths and the (total) population size, and is usually expressed
per 100,000. Thus, it represents the number of deaths per 100,000 persons at
risk in the observed region. Because mortality data are affected by age and gen-
der, age- and gender-adjusted death rates are more appropriate for comparisons
between regions with different population age and gender structures. Another
factor strongly associated with age and gender, in addition to the number of
deaths, is the cause of death.
Given the key recommendations in the work of Anderson and Rosental [1] to
obtain as much of the age- and gender-cause-specific information and to make the
data comparable across countries, we decided to convert the observed number
of deaths to the expected number of deaths. To do this, we consider age- and
gender-specific death rates for a common (standard) population, which we also
recalculated on a two-dimensional (age-gender) structure for this purpose.
摘要：

IdentifyingpatternsofmaincausesofdeathintheyoungEUpopulationSimonaKorenjak-Cerne1andNatasaKejzar21UniversityofLjubljana,SchoolofEconomicsandBusiness,andInstituteofMathematics,PhysicsandMechanics,Sloveniasimona.cerne@ef.uni-lj.si,2UniversityofLjubljana,FacultyofMedicine,InstituteforBiostatisticsan...

展开>> 收起<<
Identifying patterns of main causes of death in the young EU population Simona Korenjak- Cerne1and Nata sa Kej zar2.pdf

共13页,预览3页

还剩页未读， 继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

相关推荐

更多
立即下载
分类：图书资源 价格：10玖币 属性：13 页 大小：352.36KB 格式：PDF 时间：2025-05-08

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务

作者详情

相关内容

更多

热门标签

人际关系 配电装置 动力学连接体 力的合成 高考理综 全宋诗作者索引 公务员考试
/ 13
评分 收藏
分享
立即下载
客服
关注