# Tricky Numbers: Measuring Life Expectancy in the United States and Britain

From a lecture series presented by The Great Courses

## Life expectancy and infant mortality used to be much lower before the Industrial Revolution, in the range of 20 or 30 years. Starting in the 1800s, world life expectancy rates shot up to 48 years in 1950 and 71.5 years in 2014. That’s a lot of progress, but what is going on with life expectancy in the world today?

You might think we should be getting better and better at preventing what Shakespeare called “a necessary end,” with every generation living longer than the one preceding it. But that’s not happening right now in the United States. We’re slipping backwards. In the big picture, the important health story is the age at which our life expectancy and that of our descendents would be. How has the media in present-day times been reporting life expectancy? Is the media reporting what is needed to improve our health and longevity?

### What Is Life Expectancy?

The term “life expectancy” means the number of years a person can expect to live. Though it’s usually expressed from birth—that is, a person born today can expect to live, on average, 78 years—it can be expressed from a certain, current age. For example, if you are 20 years old now, you can expect to live another 64 years, or until the age of 84. For clarity in this lecture, life expectancy means expectancy from birth.

This is a transcript from the video seriesThe Skeptic’s Guide to Health, Medicine, and the Media. Watch it now, on Wondrium.

The best way to arrive at a life expectancy expected for a group of people is to choose a group of babies, born today, and see how long they live. Of course to do that, you would need to watch that cohort of babies for 100 years or more, until all of them died, and then compute the average number of years lived. That method of calculating is not very practical and doesn’t help us know what a life expectancy is currently. Instead, the most common way to calculate what’s formally known as a “period life expectancy” is to use a snapshot of time, looking today at the chance of dying during every year of life. You then take a hypothetical cohort of babies, not a real cohort, and mathematically run them through their years of life, looking at how the cohort decreases a little bit from year to year as some of them die.

This kind of computation does not take into account how trends in the chance of death are changing, or how they will change over the period of someone’s life. It assumes that mortality patterns present at the time of birth remain constant in the future, but it’s the method that’s most commonly used by both U.S. and international health authorities. But it’s not the only method.

### Boosting British Life Expectancy

In 2017, the UK’s Telegraph reported a headline “Statistical Re-Jig Boosts UK Life Expectancy by Three Years—But Increases Are Drying Up.” The British government had for the first time reported life expectancy as a median—that is, the age at which half of the population would expect to die—rather than the more-traditional average or mean age.

That simple change effectively increased the lifespan of British citizens by three years. Now, that’s just math. Of course, no one instantly started living longer after the report. And a case could be made that this median measure is, in a way, more realistic, because it dampens the effect of people dying at the extreme ends of the scale, either very young or very old. But the bottom line here is that it’s not an apples-to-apples comparison.

If you want to look at trends in life expectancy, or compare life expectancy or other vital statistics like infant mortality from country to country, the computations have to be done the same way. Hats off to Britain’s Telegraph for picking up on a bit of statistical jiggery-pokery that made their government statistics look a little better than they really were. The Telegraph report also talked about another trend, and it’s an important one. Over the last 100 years, we’ve gotten used to reports about an increasing lifespan—that is, every generation lives longer than the generation before it. But those increases are stalling.

As the article reports, between 2006 and 2011, life expectancy at birth increased by more than a year for men and by nine months for women. But over the most recent, five-year span, the life expectancy increase has been only three months for men, and half that for women. It’s still increasing, at least in the UK, but not by nearly as much as it used to. That change does not bode well for the continued health of future generations.

### Bad News on the American Side of the Pond

And the news is worse for those of us living in the United States. CNBC reported this headline in 2016: “U.S. Death Rate Increases for First Time in Two Decades, as Overdoses, Car Crashes, Gun Deaths Rise.” The headline is somewhat awkward. It refers to a death rate. But the death rate of the population, in terms of what percentage of us die each year, had risen, according the article, by 1.2 percent. It is unclear exactly what that percentage meant, but the article itself talked about the more familiar life expectancy statistic, pointing out that a child born in 2015 was expected to live 78.8 years, a drop from 78.9 years in 2014.

NPR reported the same story under the headline “Life Expectancy in U.S. Drops for First Time in Decades, Report Finds.” Life expectancy is a much more common way to express lifespan, more so than “death rate,” so NPR gets the nod this time for a clearer headline. Now, 78.9 to 78.8 years, that is not a big drop, of course, but it’s a trend that seems to be continuing. In December 2017, CNN reported on a further drop in the headline “U.S. Life Expectancy Drops for Second Year in a Row.”

The details provided in these articles have some important messages. In terms of quantities, or absolute numbers of deaths, the largest increases in deaths occurred among the usual causes of deaths, including heart disease and diabetes. But the causes that had the largest increase by percentage are striking. Drug overdoses surpassed 50,000 deaths a year—that includes both prescription drugs and drugs of abuse, like heroin. The Chicago Tribune focused on that sobering statistic in the story “America’s Opioid Crisis Has Cut into U.S. Life Expectancy.”

Those overdose deaths have now surpassed deaths due to car crashes, which had been for years the most common cause of accidental death in the United States. But the news about car accident deaths isn’t good either. After about 40 years of steady declines in death rates from car crashes, the rates have shot back up by 7 percent. These statistics can’t tell us the underlying reasons, but authorities suspect both substance abuse and distracted driving are playing a role. The cars continue to get safer to drive, but we drivers are getting more reckless.

### Like Real Estate, Life Expectancy Is Local

The Washington Post stressed another observation from these statistics in the article “U.S. Life Expectancy Varies by More Than 20 Years from County to County.” It’s a chilling story. Health disparities in terms of differences in life expectancy are widening, with the gap between different areas of the country varying by 20 years. The Huffington Post ran this memorable headline about the same issue: “Why Your Zip Code May Be More Important to Your Health Than Your Genetic Code.”

The first few sentences illustrates how the story is framed: “How you see a problem drives how you create the solution. We are not a healthy country. And while health reform focuses on coverage, cost, access, and care, this is simply triage to a system that fails to ask the question ‘Why aren’t we healthier in the first place?’” The Huffpo article also points out that, “evidence now suggests that medical care accounts for only 10 to 15 percent of preventable early deaths.”

A citation of that evidence is needed—those numbers cannot be accepted at face value—but, overall, demographics and neighborhoods play a huge part in health. Even differences in education have a health impact. The article points out that college grads live on average five years longer than those who didn’t complete high school, and that people who are poor are three times more likely to suffer physical limitations from chronic illness.

Important to point out, the Huffpo article isn’t well-referenced and it is fair to say that it has more editorial bent than a traditional newspaper article ought to have. Editorials traditionally are opinion pieces. They used to be found only on the editorial pages of a newspaper, while the remainder of the newspaper would be news, just reporting the facts. Today, the lines between opinion and fact have become much blurrier. Many websites have an especially pronounced slant to their news coverage. This trend seems to have rubbed back on traditional newspapers. You might think that health and medical coverage would be less prone to bias, but it’s still a good idea to beware of the source of any article you’re reading, and consider the editorial slant of the author. How a story is framed has a lot of do with how you digest and absorb the information, and how you remember it afterwards.