Published on Mar 03, 2021

Video Transcript

[MUSIC PLAYING] JOHN WHYTE: Welcome, everyone. I'm Dr. John Whyte, chief medical officer at WebMD. And you're watching Coronavirus in Context. There was a very important report released recently on life expectancy. It showed a decrease from 78.8 years to 77.8 years, the biggest drop since the Second World War and for Black Americans three times what it was for white Americans.

So to help unpack the data and provide some insight, I've asked Dr. Elizabeth Arias, one of the lead researchers of the report and the director of the US Life Tables Program at the National Center for Health Statistics. Dr. Arias, thanks for joining.

ELIZABETH ARIAS: OK, well, thank you for having me. It's my pleasure.

JOHN WHYTE: It's overall a decrease in one year on average but greater decrease in minority populations. Can you walk us through the data?

ELIZABETH ARIAS: Yes. What we saw was a decrease of one year in life expectancy at birth for the total US population and a decrease of 2.7 years for the non-Hispanic Black population and 1.9 years for the Hispanic population. And we haven't seen decreases like that in decades. The last time we saw a decrease of greater than one year was during the Second World War when life expectancy decreased by 2.9 years between 1942 and 1943.

JOHN WHYTE: And this is only for the first half of the year. What do you expect to see when you average it all together for the entire year?

ELIZABETH ARIAS: Well, that depends. So what this measure was basically the increase in what we call excess deaths during the first six months of the year. So we have the expected number of deaths-- I mean, mortality is pretty stable from year to year-- And the excess number of deaths added to the total number of deaths, which is what we use to produce the life tables, which is what gives us the life expectancy estimate.

If we have higher numbers of excess deaths during the second part of the year that would be July through December, everything else remaining unchanged from previous years, then we should expect to have an even higher decrease in or greater decrease in life expectancy at birth for the full year than for the first half. If the excess numbers of deaths remain similar to what we saw during the first half of the year, then we may see the same numbers. Or we may see if the number of excessive deaths are lower than the first half of the year, everything else remaining unchanged, then we should see a lower decline in life expectancy.

JOHN WHYTE: Do we think this decrease in life expectancy is a direct result of the COVID pandemic? Is it because people did not get care? Is it because the total number of deaths? Is it because increase in overdoses? What do you think explains this decrease?

ELIZABETH ARIAS: Well, we do know we did a study that we published recently that looked at the number of excess deaths in the US from January through October. And there were approximately 300,000 excess deaths. And of those, 2/3 were due to the COVID pandemic. So therefore, there was a third of those excess deaths that were due to other causes.

So those included increases in drug-involved mortality and increases in our standard-- I should say standard with our chronic disease causes of deaths that we see every year increases in heart disease deaths, cancer deaths. So overall, the COVID-19 deaths were the leading factor in the increases in excess deaths. And so when we estimate the life tables, we estimate them based on all deaths. And so it includes deaths from COVID and from all the other causes.

JOHN WHYTE: Does this surprise you, this decrease in life expectancy?

ELIZABETH ARIAS: Well, I expected to see a decrease in life expectancy given this increase in excess deaths. And I expected to see disparities between the groups because they already exist. But I was, I would say, surprised at how large the declines were particularly for the African-American population and for the Hispanic population.

JOHN WHYTE: We're actually increasing the difference in life expectancy for people of color versus Caucasians. Isn't that right? There's greater difference in life expectancy than there is-- now than there was several years ago.

ELIZABETH ARIAS: Yes, that's right. So if you go back to when we started producing life expectancy estimates, that would be 1900. And at that time, life expectancy-- the disparities in life expectancy between the white and Black population was over 14 years, close to 15. So the white population had an advantage in life expectancy of over 14 years. And that declined gradually all the way down to 4.1 years of advantage for the white population in 2019.

And then in 2020, the disparity went from 4.1 year to six years. So now the advantage of the white population relative to the Black population is six years. We haven't seen such a large disparity in a long time in many years.

JOHN WHYTE: It shows COVID in many ways is making the situation worse, not better, and a greater urgency to address disparities. I want to ask you, though, about this concept of the Hispanic paradox. On average, Hispanic population actually lives longer than any other minority group population in the United States. Yet, there's concern about the traditional access to health care for that population. Can you help explain why you might see the longest life expectancy in the Hispanic community compared to Caucasians and the Black community?

ELIZABETH ARIAS: Yes, that's basically termed the Hispanic mortality paradox. And it was identified in the late 1980s by some researchers that were comparing mortality between Mexican origin population and the non-Hispanic white population in Texas. And they found that despite having lower socioeconomic status, the Mexican population has significantly lower mortality.

We concurred in those findings when we began producing life tables for the Hispanic population with 2006 data. And what we found at the time was that the Hispanic population indeed had lower mortality and therefore higher life expectancy than the non-Hispanic white population. It was around 2.4 years or 2.1 years in 2006. And that increased to three years in 2019.

Now, there's been a lot of research to try to identify what are the causes behind this Hispanic mortality advantage. And so there's three main hypotheses. One is that it's due to selection. So among immigrants or people who immigrate tend to be selected for better health, hardier constitutions.

The other explanation is cultural factors such as behaviors related to diet, smoking. And then the other one is something called the salmon bias effect where you have people who return to their country of origin when they're sick. So then they are missed in our statistics. And then another one is what we call data artifact or data problems, which basically says the data is bad. And that's why you get this advantage.

JOHN WHYTE: But I want to ask you about that because there's been criticism at times on the collection of race and ethnicity data. We see that in other areas of collection of data by the government. We see that in clinical trials. A lot of data in terms of race and ethnicity are missing. And you talk about in the report how you address those missing data points, which could be very important in terms of the collection of life expectancy. Can you explain how you do that?

ELIZABETH ARIAS: The misclassification on death certificates is a result of the incongruence between what individuals self identify with with what is put down on their death certificate. So say in a person who's self identifies as American Indian is then when they die identified as white on the death certificate. Now, this section of the death certificate, which is the demographic portion, is usually filled out by the funeral director. And many times, the funeral director does not ask the family informants and decides what the person is through observation. Or even if they ask, the family members may not be right in what they think the person self identified as.

So for example, that's a big issue for the American Indian population. So over 30% of American Indians who self identify as American Indians are classified as something else on the death certificate. So then what that leads to is an underestimation of mortality.

Now, for the Hispanic population, that underestimation is around 3%. For the non-Hispanic Black population, it's very, very small, less than like half a percent. And for the non-Hispanic white population, there's barely any misclassification at all. So we took into consideration those levels of misclassification when we estimated the life tables in order to get the life expectancy estimate. So all the life tables for the Hispanic, non-Hispanic Black, and non-Hispanic white populations were adjusted to correct for the misclassification.

JOHN WHYTE: And, finally, I want to ask you about one other interesting data point. And it says here that males, their average life expectancy now is 75.1. Females is 80.5, the biggest gap that we've seen in many years. Dr. Arias, are women the stronger sex?

ELIZABETH ARIAS: Well, we have seen that throughout our history and in most populations, women have higher life expectancy than men and except for earlier in the 20th century and before that when maternal mortality was higher than it is now. But consistently, women have had higher life expectancy than men. And then, in this case of the COVID pandemic, we have seen that mortality is higher for men than for women as a result of the pandemic. So we saw the gap between men and women in life expectancy, which had been declining mainly due to improvements in mortality for men, actually increased. So it's now larger than it was last year or 2019, for example.

JOHN WHYTE: Dr. Arias, I want to thank you for sharing these important insights. It has exposed again the disparities that exist here in the United States and really is an urgent call to action to address it. So thank you for this important data.

ELIZABETH ARIAS: You're welcome. It was my pleasure.

JOHN WHYTE: If you have questions, you can send them my way to [email protected] as well as post them on Facebook, Instagram, and Twitter. Thanks for watching.

[MUSIC PLAYING]