PODCAST: Suicide Trends in the U.S. and Weekly NCHS Updates

February 26, 2021



HOST:  Last week NCHS released the latest trend report on suicide rates in the nation.  Joining us today is Holly Hedegaard, the lead author of this new report.

Holly, so what do the latest final numbers tell us?

HOLLY HEDEGAARD:  Well the report that was just released from the National Center for Health Statistics looked at suicide rates over the last 20 years and what we saw was that from 1999 through 2018 there’s been a steady increase in the suicide rate – it increased about 35% over that time period. But what’s interesting is that in 2019 the rate is lower than it was in 2018 and that’s the first significant drop in suicide rates we’ve seen in the past 20 years.  While that’s an encouraging sign, I think it’s important to remember that a single year drop doesn’t necessarily say that’s a meaningful change in the overall trend is just that within a single year we saw a decrease in the suicide rates in 2019 compared to 2018

HOST:  Youth suicide in particular is a major concern.  What do the trends show among young people?

HOLLY HEDEGAARD:  So for young people suicide rates are actually lower than for other age groups – so that’s a good thing that the rates are lower – but what’s concerning is that these are the age groups where we’ve seen quite a bit of an increase in the suicide rates in recent years.  And so for example for girls who are age 10 to 14, their rates have increased about four-fold in the past 20 years, but their rates are still among the lowest of all the age and sex groups.  Rates have also increased for boys and for young men but not to the same extent as for girls.  And so again, for both boys and girls and for age 10 to 14 and ages 15 to 24, the rates are low but they are increasing – and I think that’s the reason of concern about suicide rates in young people.

HOST:  What groups have the highest suicide rates in the country?

HOLLY HEDEGAARD:  This report focuses on rates by sex and by age group, so the report looks at those particular characteristics, and the suicide rates are highest for men age 75 and older and that’s been true for a long period of time so the highest rates among men aged 75 and older.  For females the highest rates are for women ages 45 to 64 so it’s more of the middle-aged female when you look for high suicide rates among females.

HOST:  There aren’t full-year data available yet for 2020, but mental health professionals worry that the stress and isolation from the pandemic will result in a spike in suicide rates.  Do you have any insight at all about 2020 at this point?

HOLLY HEDEGAARD:  As you mentioned, we don’t have any of the final data for 2020 yet so we can’t give a definitive answer but NCHS has been generating from provisional estimates to try to get a sense of what has been happening during 2020.  And NCHS has posted some provisional estimates for the first quarter of 2020 – which it goes through March of 2020 – and as of the beginning of last year the rate, the suicide rate, was slightly higher than the rate during the comparable time period in 2019.  So a slight increase in the first quarter.  NCHS has been developing some additional modeling techniques to look at the trends in a variety of different types of deaths including drug overdose, suicide, and transportation related deaths during the early months of 2020, and based on that modeling technique the predicted weekly numbers of suicide deaths early 2020 were similar to historic levels, and then declined a little bit between March and June, and then again was pretty much no different than historic levels from July through October.  So based on these model estimates, that suggested there hasn’t really been a spike in suicide mortality, at least in the first half of 2020.  But it’s important to recognize that these are modeled estimates – these are not final numbers, they aren’t the final rates – and we’ll continue to be refining and confirming these estimates as NCHS receives more data for the deaths that occurred in 2020.  So as of now, we don’t have anything that looks like there’s been a huge increase in suicide during 2020 but that’s again based on modeled estimates.

HOST:  Your report looks at the different mechanisms used in suicides in the U.S.  What do those numbers tell us?

HOLLY HEDEGAARD:  The means of suicide varies by males compared to females, and for males about little over half of the suicides involve use of a firearm and about 28% involve hanging or suffocation… A much smaller proportion involved poisoning or other means. We’ve seen a slight increase in the rates for firearm-related suicides among men over the past 20 years but where there’s been a rather large increase has been in the rate for suicide by hanging or suffocation.  That rate among men has doubled over the last 20 years.  The picture for women is a little bit different.  From about 2001 through 2015, poisoning was the leading means of suicide among women.  But Interestingly in the last few years, since about 2016, we’ve actually seen a decline in the rate of suicide by poisoning among women and an increase in the rate of suicides that involve firearms or suffocation.  And so in the most recent years, the rates of suicide by firearm and by suffocation are slightly higher than the rate of suicide by poisoning.  The rate of suicide by suffocation among females has actually tripled in the past 20 years.

HOST:  Now by poisoning are you referring to drug overdoses?

HOLLY HEDEGAARD:  No, poisoning is actually a broader terminology that includes drug poisoning, but it also includes other types of poisons like carbon monoxide or chemicals or a variety of other things that sometimes people ingest or take. But they aren’t drugs there are used for other purposes.

HOST:  So your data then show that drug overdoses are really not a significant method used in suicides?

HOLLY HEDEGAARD:  It’s different – again, as I mentioned – for men or for women.  For men, only about 5% of suicide actually involve a drug overdose.  For women, it’s about 27% of their suicides involve a drug overdose.  So they’re not the, drug overdoses are not the leading means of suicide for either men or women.  For both men and women, rates of firearm-related suicide or suicide by hanging and suffocation are higher than the rates of suicide by drug overdose.

HOST:  This report doesn’t look at geographical differences but what areas of the country are having a tougher time with this problem?

HOLLY HEDEGAARD:  So the higher suicide rates are found in the Rocky Mountain states such as Wyoming, Montana, New Mexico, Colorado, Utah, as well as Alaska.  So these are states that have historically been high and they continue to remain high.  In the most current years or recent years, we’ve seen increase in the rates in some of the other states in the Midwest and in the New England states, up in Maine and Vermont and New Hampshire.  They aren’t the highest rates but they are increasing, so it’s important to sort of recognize that there are states in addition to the Rocky Mountain stage that also are seeing higher suicide rates.

HOST:   The National Health Interview Survey issued two new reports, on Tuesday and Wednesday of this week.  On Tuesday, NCHS teamed with the VA on a report that examined multiple chronic conditions among veterans and non-veterans.  Based on data from the 2015-2018 NHIS, the study authors found that about one-half of male veterans and over one-third of female veterans had two or more chronic conditions, compared with less than one-fourth of male nonveterans and less than one-fifth of female nonveterans.  Hypertension and arthritis were the most prevalent chronic conditions among all veterans age 25 and over.  Diabetes was also prevalent among male veterans ages 25 to 64 and asthma was also prevalent among female veterans in this age group.  Cancer was also prevalent among all veterans age 65 and older.

On Wednesday, NCHS released another study looking at health care utilization among those afflicted with inflammatory bowel disease, or IBD.  The study used NHIS data and found that adults with IBD were more likely than those without IBD to have visited any doctor or mental health provider in the past year, and were also more likely to have been prescribed medication or to have received acute care services such as ER visits, overnight hospital stays, or surgeries.

On Thursday, NCHS released a third study – on dietary supplement use among American adults age 20 and over.  The report used data from the 2017-2018 National Health and Nutrition Examination Survey, and found that over half of adults used a dietary supplement in the past month – nearly two-thirds /3 of women and just over half of men.  Eight out of ten women age 60 and over used dietary supplements, and older Americans are more likely to use more than one dietary supplement.   The most common dietary supplement used was multivitamin-mineral supplements.  Vitamin D and omega-3 fatty acid supplements were also commonly used.

Finally, today NCHS is releasing the latest quarterly provisional data on birth rates in the United States, through the third quarter of 2020, showing that fertility rates in the country continued to drop compared to the same point in 2019.  Teen birth rates and pre-term rates also declined in Quarter 3 of 2020 compared with Quarter 3 of 2019, while cesarean delivery rates increased over this period.


QuickStats: Motor-Vehicle–Traffic Death Rates Among Persons Aged 15–24 Years and ≥25 Years — United States, 2000–2019

February 26, 2021

From 2000 to 2006, rates of death caused by motor-vehicle–traffic injuries among persons aged 15–24 years and ≥25 years did not change significantly.

From 2006 to 2010, motor-vehicle–traffic death rates per 100,000 population declined among those aged 15–24 years, from 25.1 (2006) to 16.1 (2010), and among those aged ≥25 years, from 15.9 (2006) to 12.5 (2010).

Throughout most of the period, motor-vehicle–traffic death rates were higher among persons aged 15–24 years; however, motor-vehicle–traffic death rates began to converge in more recent years, and by 2019, the difference in the rate among those aged 15–24 years (13.7) and those aged ≥25 years (13.6) was not statistically significant.

Source: National Center for Health Statistics, National Vital Statistics System, Mortality Data, 2000–2019. https://www.cdc.gov/nchs/nvss/deaths.htm


Dietary Supplement Use Among Adults: United States, 2017–2018

February 25, 2021

NCHS releases a new report that describes recent prevalence estimates for dietary supplement use among U.S. adults, the distribution of the number of dietary supplements used, and the most common types of dietary supplements used.

Trends in dietary supplement use from 2007–2008 through 2017–2018 are also reported.


  • Among U.S. adults aged 20 and over, 57.6% used any dietary supplement in the past 30 days, and use was higher among women (63.8%) than men (50.8%).
  • Dietary supplement use increased with age, overall and in both sexes, and was highest among women aged 60 and over (80.2%).
  • The use of two, three, and four or more dietary supplements increased with age, while the percentage of adults not using any dietary supplement decreased with age.
  • The most common types of dietary supplements used by all age groups were multivitamin-mineral supplements, followed by vitamin D and omega-3 fatty acid supplements.
  • From 2007–2008 through 2017–2018, the prevalence of dietary supplement use increased in all age groups among U.S. adults.

Health Care Utilization Among U.S. Adults With Inflammatory Bowel Disease, 2015–2016

February 24, 2021

NCHS releases new report that measures health care utilization among adults with inflammatory bowel disease (IBD) and compare with adults without IBD.


  • Compared with adults without IBD, adults with IBD were more likely to have visited any doctor or mental health provider in the past 12 months.
  • IBD was also associated with higher prevalence of being prescribed medication, and having received
    acute care services, such as emergency room visits, overnight hospitalizations, or surgeries.
  • Differences by IBD status were greatest for visiting a specialist and home visits in the past 12 months.

Multiple Chronic Conditions Among Veterans and Nonveterans: United States, 2015–2018

February 23, 2021

A new NCHS report describes the prevalence of multiple (two or more) chronic conditions (MCC) among veterans and nonveterans and examines whether differences by veteran status may be explained by differences in sociodemographic composition, smoking behavior, and weight status based on body mass index.

Key Findings:

  • Among adults aged 25 and over, age-adjusted prevalence of MCC was higher among veterans compared with nonveterans for both men and women (22.2% compared with 17.0% for men aged 25–64, 66.9% compared with 61.9% for men aged 65 and over, 25.4% compared with 19.6% among women aged 25–64, and 74.1% compared with 61.8% among women aged 65 and over).
  • Following stratification by age and adjustment for selected sociodemographic characteristics, the prevalence of MCC remained higher among veterans compared with nonveterans for both men and
  • After further adjustment for smoking status and weight status, differences in the prevalence of MCC by veteran status were reduced

QuickStats: Death Rates Attributed to Excessive Cold or Hypothermia† Among Persons Aged15 Years or older, by Urban-Rural Status and Age Group

February 19, 2021

In 2019, among persons aged 15 years or older, death rates attributed to excessive cold or hypothermia were higher in rural areas than in urban areas across every age group.

Crude rates were lowest among those aged 15–34 years at 0.2 and 0.5 per 100,000 population in urban and rural areas, respectively.

Rates increased with age, with the highest rates among those aged 85 years or older at 4.6 in urban areas and 8.6 in rural areas. Differences between urban and rural rates also increased with age.

Source: National Center for Health Statistics, National Vital Statistics System, Mortality Data 2019. https://wonder.cdc.gov/mcd-icd10.html


PODCAST: Decline in Life Expectancy

February 19, 2021



HOST:  NCHS for the first time is releasing provisional, mid-year 2020 estimates on life expectancy in the United States.  With the arrival of COVID-19 as one of the major causes of death in 2020, NCHS has been releasing mortality and other data related to the pandemic in record time.  This new release of data documents that life expectancy at birth for the U.S. population dropped a full year in the first half of 2020 alone.  This one-year drop in life expectancy, from 78.8 years in 2019 to 77.8 years in mid-year 2020, is the largest drop in life expectancy since 1943, during World War II.

Joining us today is the lead author of this new study, Elizabeth Arias, with NCHS’s Division of Vital Statistics.

HOST:  So this is the first time NCHS has released a mid-year estimate on life expectancy.  How were you able to do this for the first time?

ELIZABETH ARIAS:  We were able to produce provisional life expectancy estimates for the first half of 2020 because of steady improvements to the timeliness and quality of death records over the past few years.  We will be updating the half year provisional life expectancy estimates with estimates based on deaths for the entire year with the goal of publishing them by May or June of 2021.  And then we will also publish life expectancy estimates for 2020 based on final mortality data as part of the U.S. annual life tables.

HOST:  Given the surge in COVID-19 deaths around the holidays, should we expect that the full-year life expectancy estimate for 2020 will be a much larger drop?

ELIZABETH ARIAS:  Well if the number of excess deaths occurring in the months of July through December are greater than the number of excess deaths that occurred in the months January through June, then the full year life expectancy estimates may be lower.  Excess deaths are typically defined as the difference between the observed number of deaths in specific time periods and expected number of deaths in the same time period.

HOST:  How does this drop in life expectancy compare with the declines in life expectancy during the 1918 flu pandemic?

ELIZABETH ARIAS:  The decline in life expectancy that we saw for 2020 is the highest decline that we have seen, not since the pandemic, but since the second world war when life expectancy declined 2.9 years between 1942 and 1943.

HOST: In general, men have shorter life expectancy than women.  Does this mid-year 2020 analysis show reflect the same disparity?

ELIZABETH ARIAS:  Yes it does.  The disparity in life expectancy between men and women in fact increased in 2020 to 5.4 years from 5.1 years in 2019.

HOST:  The new data shows bigger declines in life expectancy among minority populations.  Could you discuss that?

ELIZABETH ARIAS:  The COVID-19 pandemic has disproportionately affected minority populations, including African Americans, Hispanics, and Native Americans.  The significantly larger declines in life expectancy in the African American population, which was 2.7 years, and in the Hispanic population, which was 1.9 years, reflect the racial and ethnic disparities in the effects of the pandemic.  The decline in life expectancy for the African American population was 3.4 times greater than the decline for the non-Hispanic white population.  And the decline in life expectancy for the Hispanic population was 2.4 times greater than the decline for the non-Hispanic white population.

HOST:  Past studies have shown that the Hispanic population has longer life expectancy than other groups.  But it appears the pandemic has cut into that advantage significantly?

ELIZABETH ARIAS:  Yes it has.  For over 30 years, studies have consistently shown the Hispanic population has lower mortality than the non-Hispanic white population.  Since we began estimating life expectancy by Hispanic origin with data year 2006, this Hispanic mortality advantage relative to the non-Hispanic white population increased from 2.1 year to 3.0 years between 2006 and 2019.  As a result of the pandemic, the Hispanic life expectancy advantage declined 37% between 2019 and 2020 to an advantage of just 1. 9 years.

HOST:  Will you at some point have life expectancy estimates for Asians or other populations?

ELIZABETH ARIAS:  Yes – we are currently working to expand the U.S. life table program to include life tables for the Asian and Native American populations.  Problems with data quality has been the main reason we have not produced official U.S. life tables for these groups in the past.  However, through a combination of improvement in data quality and the development of methods to address the data quality problems, we will be able to publish life tables for these populations very soon.

HOST:  What about geographic differences in life expectancy during the pandemic?  Any insight on that?

ELIZABETH ARIAS:  Well, we did not produce life expectancy estimates by geography, including state or County estimates, so we cannot speak directly to geographic differences in life expectancy during the pandemic.  We can speculate though that the national life expectancy estimates we produced for the first half of 2020 may underestimate life expectancy due to the over-representation of mortality in urban areas, where the pandemic was more prevalent during the first half of 2020 and where groups who experienced the worst effects are concentrated, such as the Hispanic and African American populations.

HOST: We’ve also seen very substantial increases in drug overdose deaths in the first part of 2020.  Could that also be driving this decline in life expectancy?

ELIZABETH ARIAS:  The life expectancy estimates for the first half of 2020 were estimated based on the total number of deaths that occurred during that period. The difference in life expectancy between the first half of 2020 and 2019 is due to the increase in the number of excess deaths during the former period. So excess deaths during that period may include deaths from causes other than COVID-19, such as drug overdose deaths.

HOST: Thanks to Elizabeth Arias for joining us to discuss her new study on the drop in life expectancy during the 2020 pandemic.

Today, NCHS is also releasing an updated trend report on suicide in the U.S., covering two decades from 1999 to 2019.  The report shows that the suicide rate in the U.S. declined in 2019 for the first time since 2005.  Holly Hedegaard, the author of the study, will be joining us on next week’s edition of “Statcast” to discuss.

Provisional Monthly Drug Overdose Deaths from July 2019 to July 2020

February 17, 2021

On Wednesday (February 17) NCHS released the next set of monthly provisional drug overdose death counts.

Provisional data show that the reported number of drug overdose deaths occurring in the United States increased by 22.8% from the 12 months ending in July 2019 to the 12 months ending in July 2020, from  68,023 to  83,544.  After adjustments for delayed reporting, the predicted number of drug overdose deaths showed an increase of 24.2% from the 12 months ending in July 2019 to the 12 months ending in July 2020, from 69,266 to  86,001.

The reported number of opioid-involved drug overdose deaths in the United States for the 12-month period ending in July 2020 (61,297) increased from 47,474 in the previous year. The predicted number of opioid-involved drug overdose deaths in the United States for the 12-month period ending in July 2020 (63,129) increased from 48,429 in the previous year. Recent trends may still be partially due to incomplete data.

The reported and predicted number of drug overdose deaths involving synthetic opioids (excluding methadone; T40.4) and psychostimulants with abuse potential (T43.6) continued to increase compared to the previous year. Both reported and predicted overdose deaths involving cocaine increased compared to the previous year. The reported and predicted number of natural and semi-synthetic opioid deaths also increased compared to the previous year

PODCAST: Quarterly NHIS Early Release Key Health Indicator Estimates

February 12, 2021


HOST:  This week NCHS released its latest quarterly estimates on a number of key health indicators from the National Health Interview Survey, one of the oldest health surveys in US history dating back to 1957. This latest quarterly release covers the period up to the midpoint of 2020.  Health data from NHIS have always been driven by the types of questions asked in this traditionally in-person survey and in order to improve the quality of data the survey has been redesigned on occasion over the years, most recently in 2019. Tainya Clarke, an epidemiologist with the survey, elaborates:

TAINYA CLARKE:   The NHIS underwent a survey redesign to better meet the needs of data users.  Some questions were dropped from the survey, new questions were added, and some question text or the order they appear changed.  All these changes mean that the NHIS survey for 2019 going forward is quite distinct from the past survey.  In addition to the questionnaire design, changes made to the weighting approach have the potential to impact direct comparisons between the estimates for 2019 to June 2020 and earlier years.  Because of this we have not examined trends prior to 2019 in this release.

HOST:  The last time the NHIS was redesigned was in 1997.  That redesign laid the foundation for the creation of the early release program, which features quarterly preliminary estimates on a number of high profile health topics.  The 2019 redesign introduced several new topics to the survey.  However, the arrival of the pandemic in 2020 forced more changes to the way the survey operates, which in turn has had an impact on response rates:

TAINYA CLARKE:    Due to the current pandemic and the need for physical distancing, in quarter two we switched to a telephone-only approach, and in quarter three and four a telephone-first approach and followed up in person for households with no response or without a listed telephone contact.   We may have some new questions on COVID-19 and related health behaviors added in the future early releases.  So we’re not quite sure what those questions will be – only time will tell.

HOST:  In recent years this release has switched from a publication-based format to an interactive web-based data visualization format.  But there remains a gold mine of important data topics in this quarterly release.  For example, hypertension among adults is now being tracked, which is important particularly now during the pandemic with high blood pressure being a major risk factor for people with COVID-19.

TAINYA CLARKE:   The early release data in 2019 and the first half of 2020 showed that about 1/4 of US adults have been diagnosed with hypertension in the past 12 months, and the prevalence is highest among non-Hispanic Blacks, with more than one in three having hypertension in the past 12 months.

HOST:  The NHIS has tracked cigarette smoking among adults since the 1960s, and has documented along running decline in the percentage of adults who smoke.  Cigarette smoking is also one of the topics featured in this week’s new quarterly release showing that an all-time low of 12.2% of American adults were current cigarette smokers in Quarter Two of 2020.  And recently, the NHIS has added E-cigarette use to this quarterly release.

TAINYA CLARKE:   In 2019 we had approximately 4.4% of adults using E-cigarettes.  The percentage is even lower in January to June 2020 at 3.6%.

HOST:  Some of these second quarter estimates from 2020 may indeed reflect the impact of the pandemic on the country.  However, many of the survey questions are based on the past 12 months, so any direct connection to the pandemic is inconclusive.  This includes the second quarter 2020 finding that a higher percentage of adults are seeking mental health counseling, as well as the fact that a lower percentage of adults visited the emergency Department during this period. Another important measure featured in this quarterly release is flu vaccination.  Nearly half of adults reported they received a flu vaccine, according to data from the second quarter of 2020.  But Tainya Clarke says some context is needed when interpreting those immunization numbers.

TAINYA CLARKE:   I think I want to point out that even though this is almost 50% – and to some people that may seem like a large percentage – the target for vaccinating adults against influenza is much higher.  It’s closer to 70% for HP 2020 – that’s a Healthy People 2020 initiative.  And although that initiative is targeted at seasonal flu, we really hope to see a larger percentage of the U.S. adult population receiving more flu vaccines going forward.

HOST:  The latest quarterly release of data from the NHIS also includes new data on health insurance coverage in America.  During the first half of 2020, over 30 million Americans – or 9.4% – were uninsured at the time they were interviewed as part of the survey.  This proportion of the population who had no insurance includes over 13% of those ages 18 to 64.  In this age group, a little more than one in five had public health insurance and a little more than 2/3 had private insurance.  The poor or near poor in this age group were more likely to be uninsured than those who are not poor. Hispanic adults in this 18 to 64 age group were twice as likely to be uninsured as non-Hispanic black adults and nearly three times as likely as non-Hispanic white or non-Hispanic Asian adults.  Among children under age 18, less than 5% were uninsured, and among those who did have insurance over 41% had public coverage and over 56% had private coverage.

HOST:  Our thanks to Tainya Clarke of the National Health Interview Survey for joining us on this edition of “Statcast.”  Join us next week for a special segment on data related to the pandemic’s impact on life expectancy.

QuickStats: Trends in Secondhand Smoke Exposure Among Nonsmoking Adults, by Race† and Hispanic Origin — National Health and Nutrition Examination Survey, United States, 2009–2018

February 12, 2021

The percentage of nonsmoking adults exposed to secondhand smoke (SHS) declined from 27.7% in 2009–2010 to 20.7% in 2017–2018.

During this period, decreasing trends in the percentage of persons with SHS exposure also were observed for nonsmoking non-Hispanic White, non-Hispanic Black, and Hispanic adults.

There was no significant decline in the percentage of persons with exposure for nonsmoking non-Hispanic Asian adults from 2011–2012 to 2017–2018.

The percentage of persons with SHS exposure was consistently higher for nonsmoking non-Hispanic Black adults throughout the period.

During 2017–2018, 41.5% of nonsmoking non-Hispanic Black adults were exposed to SHS compared with 22.7% non-Hispanic Asian, 17.8% non-Hispanic White, and 16.2% nonsmoking Hispanic adults.

Source: https://www.cdc.gov/mmwr/volumes/70/wr/mm7006a6.htm