The temporary increase in the risk of death at the end of adolescence is a phenomenon that was first identified over a century ago. Although this abnormally high mortality rate has been extensively documented and recognised in demography, it has never been clearly defined, measured or explained.
Adrien Remund's doctoral study fills those gaps. Undertaken within the Swiss National Centre of Competence in Research LIVES at the University of Geneva, his dissertation shows that the presence of a very small vulnerable sub-population is enough to generate a mortality bump without any individual actually experiencing an increase in their personal risk of death.
Contrary to the vision shared until now by certain demographers and a number of (neuro)psychologists, the high death rate among young adults is not, therefore, primarily linked to a spread of dangerous behaviours during this phase of life, but rather to the presence of particularly pronounced social, economic and biological inequalities within this age group. Once the most exposed individuals have been taken out of the equation, the mortality curve resumes a more even trajectory.
To obtain these results, the researcher tested different theoretical hypotheses applying several methodological tools, some of which already existed and some of which were newly developed. He analysed the mortality statistics of more than 10,000 different population groups using the Human Mortality Database, which covers four centuries and four continents.
Remund's findings reveal that the high mortality rate among young adults is neither universal, nor limited to adolescents, nor caused solely by accidents and suicides. Prior to World War II, the phenomenon could be mainly attributed to tuberculosis and maternal mortality.
The young demographer also used the Swiss death records compiled by the Swiss National Cohort to study inequalities among young people in relation to death at a more local level, by observing survival rates between the ages of 10 and 34 in a cohort of approximately 375,000 residents born between 1975 and 1979.
Sub-populations at risk The Swiss data revealed the presence of inequalities "exceeding all expectations," particularly with regard to gender, level of education, type of household and socio-professional status. As the vulnerability factors combine, risk ratios ranging from 1 to 100 are found between the most privileged and most vulnerable profiles. For the researcher, this proves that it is not an inevitable phenomenon.
"No, the abnormally high mortality rate among young adults is not an inevitability, as numerous historical populations and a large proportion of the young people growing up in Switzerland avoid it. While road accidents and suicides do currently represent the main challenge in terms of public health policy, history teaches us that, in the past, non-violent causes of death greatly contributed to the high death rate among young adults. Indeed, the socio-economic context of the transition to adulthood brings about huge inequalities in the risk of death, which can explain the high mortality rate among young adults much better than theories based purely on the neuropsychological development of the adolescent," Remund concludes in his thesis.
The demographer hopes that his conclusions will have an impact on future public health policies aimed at young adults. He thinks that research should however go further: "I could definitely dedicate my whole career to this subject," he said during his viva on 21 May.
"A fundamental point of reference" Carlo Giovani Camarda, a member of the jury and researcher at the Institute for Demographic Studies (INED) in Paris, stated that this thesis will be a "fundamental point of reference."
The other members of the jury also noted the "boldness," "innovation" and "interdisciplinarity" of the thesis. "You have the ability to make complicated things simpler," said France Meslé, Director of Research at INED.
Materials provided by Swiss National Centre of Competence in Research LIVES. Note: Content may be edited for style and length.
Cite This Page: