Life expectancy (LE) at birth is an important indicator to assess social health status [1]. China sets a goal of increasing the LE at birth to 79 years old by 2030, and one of Russia’s national development goals is to raise the LE to 80 years old by 2030. However, improving the LE remains a worldwide problem, as the LE growth has slowed in many countries in recent years [2, 3].

Many factors can impact LE, such as advances in public health, modern medical technology [4], health expenditure, genes, and personal behavior. However, from the perspective of an abbreviated lifetable calculation, the mortality rate of each age group is likely to be one of the most significant factors [5]. Therefore, specifically determining the age group that may have the most severe impact on the LE and relevant causes of death would provide references for pragmatic and targeted measures to improve LE.

As one of the developed countries, the total health expenditure in the United States (the US) accounted for 17.7% of GDP in 2018 [6],which was much higher than in other countries. While a pending question is whether such volumed input echoed the country’s health output. Many studies with cross-national comparison methods have demonstrated that the LE of the US population has fallen below other developed European countries [7–9]. Analyzing the reasons behind this phenomenon would be helpful to improve the LE globally.

The U.S. is a multi-ethnic immigrant country, among which, Non-Hispanic White (NHW) people account for 60.5% of the total US population [10]. Therefore, there is no doubt that the LE of NHW Americans would have a great impact on the LE of the US total population. Considering the NHW population has been more comparable to white Europeans in terms of genes and lifestyle, it may be more appropriate to take the Europeans as the comparator in this study.

This study aims to compare the overall LE trend of the NHW Americans with that of the United Kingdom (the UK) and some selected European countries, and to explore the negative influencing factors as well as causes of mortality. The result is expected to provide references for the US and other countries with similar demographic profiles to improve LE in the future.

In this study, we found that the LE of NHW Americans was lower than that of some European countries. The result demonstrated that NHW aged 30–34 years had the most negative contribution to the increment of LE. Besides, unintentional accidents and intentional self-harm were the top two causes of death in this age group.

Some currently available cross-national studies revealed similarly sluggish growth in the LE of the Americans [17], but these studies usually did not take the demographic composition into account. In contrast, by following the US Census Bureau’s definition of NHW, the NHW Americans includes persons of European, Middle Eastern and North African origin, we selected European countries in the OECD to control the ethnicity differences and increase comparability. However, our result showed the LE of NHW Americans was lower than that of the UK and the 25 European countries, furthermore, the gap has been increasing. Given the NHW population accounts for the largest proportion of the US total population, the LE trend of NHW can partly explain why the US is lagging behind the European countries.

From 2006 to 2018, per capita health cost in the US increased from $6,821 to $10,624, while it just increased from $3,035 to $4,620 in the United Kingdom [18], but the LE in the US was not as optimistic as that in the UK, and did not rank high in contrast with the European countries. So, it seems that the large volume of health expenditure input of NHW Americans did not echo its health output. However, the LE at birth just one of the health output indicators, it would be irresponsible to judge the US health input-output ratio just by one metric. The output should be evaluated by many aspects [19], for example, the mortality rate of malignant neoplasms for NHW Americans in 30–34 age group was well controlled, which may reveal the clinical medical advances in the United States.

The finding that the 30–34 age group had the most negative influence on the increment of LE concurs with the result done by Kenneth D. Kochanek’s group who pointed out that NHW aged 25–34 contributed a great loss of 0.125 years to the changes in LE from 2000 to 2014 [20]. Besides, our findings on the direct cause of this phenomenon were consistent with those studies on “deaths of despair”—death by drugs, alcohol and suicide, which was brought up by Case and Deaton in 2015 [21]. Actually, both the 45–54 age group, the population studied by Case and Deaton, and the 30–34 age group were also struggling with death of despair.

Two major factors were found to be potentially responsible for the exceptionally outstanding death rate increase of NHW Americans aged 30–34 in the accidents. The first one would be poisoning, which mainly refers to drug overdose. In 2018, there were 67,367 deaths due to a drug overdose in the United States, and NHW accounted for 73.7% (49,637 deaths) of those cases [22]. The overuse of opioids was relatively severe, among all the drug deaths. Additionally, white Americans were more sensitive to opioid deaths than other races, based on the centralized media coverage [23]. The second one is Motor Vehicle (MV) accidents. According to the Association for Safe International Road Travel (ASIRT), the US suffered the most road crash deaths of any high-income country, about 50% higher than Western European countries, Canada, Australia and Japan [24]. This is highly associated with poor driving awareness [25] and growing number of drug users and drunk drivers [26].

Moreover, this study demonstrated that the unstable growth of intentional self-harm or suicidal mortality should not be ignored since the majority of suicides are linked to mental illness. A higher suicide death rate in NHW males than females was probably related to job/financial problems and criminal/legal problems [27]. The number of suicidal deaths among the NHW population in rural areas was higher than that in large or medium-sized metropolitan cities [28]. Therefore, special attention needs to be paid to the rural NHW males aged 30–34 and further analyses on this population’s mental health are necessary.

Age is a factor that is always overlooked in the study of health inequalities and inequities. There have been very few studies which examine health inequities across age groups. Yukiko Asada [29] proposed the idea that health inequality and inequity analysis results vary by choice, as age increases, the ability of individuals to make choices also increases, therefore, this may lead to a reduction in health inequities. Damien Bricard [30] proved that inequalities of opportunity in health and mortality deepen with age, at least until middle age, because of differences in the environment of a person’s upbringing, by a cohort of 17,500 people.

Apart from that, it appears that health inequalities and inequities on different age groups should be based on the background of the era. From 2006 to 2018, all age groups experienced the same development background of the times, especially the same financial crisis and economic recovery. During this period, all age groups of NHW Americans mistakenly believed that their economic and social status was declining, which brought various forms of psychological and physical stress, and finally manifested in the impact on health [31]. But it seems that 30–34-year-olds have more serious health problems than any other age groups because of the bad experiences accumulated by this crisis. This may be because of the social tag on this group—30–34 years old is the decisive point of destiny. A report from 2018 [32] indicated that 30-year-old Americans were exposed to more various types of financial stress, including home loans, car loans, and student loans than any other age groups, which was even higher than their parents’ generation as 30-year-olds. Exposure to so much stress was more likely to lead to mental health problems and risk-taking behaviors [33].

In addition, the total effect calculated by Arriaga’s method is additive. Apart from the group aged 30–34, NHW women aged 20–59 also contribute negatively to LE growth, which indicated that NHW women in the US have an increasing risk of death during their working age [34]. This is consistent with Michelle Anne Parsons’ analysis of NHW mortality in Arizona, which found that all 5-year age groups from 25 to 64 experienced an increase in “deaths of despair”. Both genders have been affected but females had increased by a greater percentage [35]. A long series of negative total effect values, from 20 to 60 years old, demonstrated that the impact of social determinants of health on LE could not be ignored. Social determinants of health and health equity are the most concerned issues in the global health. People’s social status and resources, including income, education, sanitation facilities and living conditions, can have great impact on their health. So, with the age span becoming larger, the influencing factors involved would be much more complicated. It was indicated that corresponding interventions for deaths of despair should be comprehensive, and the measures should be considered to cover a wide range of populations, especially to the working-age ones.

Finding the age group contributing mostly to the loss of LE and the leading causes of death in this age group will help figure out action plans to increase LE and improve the health status of the people. However, complete and continuous data in most developing countries are not available.

In this study, there are three unavoidable data biases: First, since LE data by ethnicity were not available in the United Kingdom, we used the LE of British population included all the ethnicities, among them, British white accounted for about 84% of the total population in 2018. Therefore, we adopted the multi-national median value to enable data comparability. Second, the LE tables for the UK and the US were from corresponding national statistical agencies. Systematic errors may arise due to different sources of the statistical database, nevertheless, the quality of the data was somehow assured as the sources of the data were authoritative. Third, the data on causes of death for NHW in 2018 were unpublished and therefore discarded in this study.