An analysis of nearly three decades of mortality data in the United States of America (USA) published in Nature Health has found that cigarette smoking accounted for nearly two-thirds of excess deaths among White Americans with lower levels of education.
Led by researchers at Oxford Population Health and the Centre for Global Health Research at the University of Toronto, Canada, the study examined 33 million death records to quantify the contribution of smoking to widening disparities in death rates among White adult Americans by their level of education.
Life expectancy gains among adults in the USA have stalled in recent decades, driven mainly by worsening life expectancy among people with low education (11 years or less), and stagnation in the middle-education group (12 years). By contrast, life expectancy continues to improve in the high-education groups (13 years or more). These trends have been most pronounced among non-Hispanic White Americans – the focus of this study.
The researchers found that for the lower-educated, the risk of a 30-year-old White man dying before reaching age 80 increased from 65% to 79% over three decades. Among lower educated women, it rose from 44% to 67%. By contrast, the risk of premature death for those with higher levels of education fell from 44% to 34% among men and from 31% to 25% among women. Over the 30-year period studied, the gap in the risk of premature death between low- and high-educated men more than doubled, while the gap among women tripled.
The researchers used this divergence in mortality rates to estimate that between 1992 and 2019, there were 3.2 million excess deaths among White Americans in the two lower levels of education (12 years or less). They determined that nearly two thirds (63%) of these deaths were due to smoking-attributable diseases.
The findings challenge the narrative that recent increases in mortality among lower educated White Americans are mainly due to opioid overdoses, suicide and alcohol-related diseases. Only 3% of excess deaths over the three decades were the result of opioid poisoning. These deaths mainly occurred before age 65 and were greater in the decade starting in 2010. Cirrhosis (often caused by heavy alcohol use) accounted for 1% over the full study period.
Professor Prabhat Jha, Head of Department at Oxford Population Health and senior author of the study, said:
Much of the public discussion about rising mortality in the USA has focused on opioids and “deaths of despair”, but our findings show that smoking is the main driver of premature death.
Although smoking prevalence has declined across all socioeconomic groups in the USA, the decline has been slower among less-educated adults and those over the age of 50. According to the study, highly educated Americans are increasingly likely to have never smoked.
The researchers estimated smoking-attributable deaths by using lung cancer mortality as a proxy for long-term tobacco use. They then estimated what proportion of deaths from smoking-related diseases such as cancer, heart disease and respiratory illness were likely to have been caused by tobacco use. They separately examined trends in mortality from opioids, accidental poisoning, cirrhosis, suicide, injuries and other external causes.
They acknowledged that changes in the makeup of education groups over time, differences in healthcare access and possible errors in death certificate data may have affected the results but were unlikely to have changed the study’s main findings.
The study concluded ‘Absolute increases in smoking cessation among low-education groups would yield the largest gains in population health and offer the best intervention to reverse the remarkable stagnation of survival among white people in the USA.’